[Federal Register: June 6, 2005 (Volume 70, Number 107)]
[Rules and Regulations]               
[Page 32867-32968]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr06jn05-6]                         
 

[[Page 32867]]

-----------------------------------------------------------------------

Part II





Department of Labor





-----------------------------------------------------------------------



Mine Safety and Health Administration



-----------------------------------------------------------------------



30 CFR Part 57



Diesel Particulate Matter Exposure of Underground Metal and Nonmetal 
Miners; Final Rule


[[Page 32868]]


-----------------------------------------------------------------------

DEPARTMENT OF LABOR

Mine Safety and Health Administration

30 CFR Part 57

RIN 1219-AB29

 
Diesel Particulate Matter Exposure of Underground Metal and 
Nonmetal Miners

AGENCY: Mine Safety and Health Administration (MSHA), Labor.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: This final rule revises MSHA's existing standards addressing 
diesel particulate matter (DPM) exposure in underground metal and 
nonmetal (M/NM) mines. In this final rule, MSHA changes the interim 
concentration limit measured by total carbon (TC) to a comparable 
permissible exposure limit (PEL) measured by elemental carbon (EC), 
which renders a more accurate DPM exposure measurement. Also, this 
final rule increases flexibility of compliance for mine operators by 
requiring MSHA's longstanding hierarchy of controls for its other 
exposure-based health standards at M/NM mines, but retains the 
prohibition on rotation of miners for compliance. Furthermore, this 
final rule: Requires MSHA to consider economic as well as technological 
feasibility in determining if operators qualify for an extension of 
time in which to meet the final DPM limit; deletes the requirement for 
a control plan; and makes conforming changes to existing provisions 
concerning compliance determinations, environmental monitoring and 
recordkeeping.

DATES: Effective Date: The final rule is effective on July 6, 2005.

FOR FURTHER INFORMATION CONTACT: Office of Standards, Regulations, and 
Variances, MSHA, 1100 Wilson Blvd., Room 2350, Arlington, Virginia 
22209-3939; 202-693-9440 (telephone); or 202-693-9441 (facsimile).
    You may obtain copies of this final rule and the Regulatory 
Economic Analysis (REA) in alternative formats by calling 202-693-9440. 
The alternative formats available are either a large print version of 
these documents or electronic files that can be sent to you either on a 
computer disk or as an attachment to an e-mail. The documents also are 
available on the Internet at http://www.msha.gov/REGSINFO.HTM.


SUPPLEMENTARY INFORMATION:

Outline of Preamble

    This outline will assist the mining community in finding 
information in this preamble.

I. List of Common Terms
II. Rulemaking Background
    A. First Partial Settlement Agreement
    B. Second Partial Settlement Agreement
III. The Final PEL
IV. The 31-Mine Study
    A. Summary
    B. Subsequent Activities
V. Compliance Assistance
    A. Baseline Sampling
    B. DPM Control Technology
VI. DPM Exposures and Risk Assessment
    A. Introduction
    B. DPM Exposures in Underground M/NM Mines
    C. Health Effects
    D. Significance of Risk
VII. Feasibility
    A. Background
    B. Technological Feasibility
    C. Economic Feasibility
VIII. Summary of Costs and Benefits
IX. Section-by-Section Analysis
X. Distribution Table
XI. Regulatory Impact Analysis
XII. References Cited

I. List of Common Terms

    Listed below are the common terms used in the preamble.

Commission........................  Federal Mine Safety and Health
                                     Review Commission.
CV................................  coefficient of variation.
DE................................  diesel exhaust.
DOCs..............................  diesel oxidation catalysts.
DPF...............................  diesel particulate filter.
DPM...............................  diesel particulate matter.
EC................................  elemental carbon.
ETS...............................  environmental tobacco smoke.
Filter Selection Guide............  Diesel Particulate Filter. Selection
                                     Guide for Diesel-powered Equipment
                                     in Metal and Nonmetal Mines.
First Partial Settlement Agreement  66 FR 35518 (2001) & 66 FR 35521
                                     (2001): basis for July 5, 2001
                                     NPRM.
HEI...............................  Health Effects Institute.
HWE...............................  healthy worker effect.
MARG..............................  Methane Awareness Resource Group.
M/NM..............................  metal/non-metal.
MSHA..............................  Mine Safety and Health
                                     Administration.
NIOSH.............................  National Institute for Occupational
                                     Safety and Health.
NTP...............................  National Toxicology Program.
OC................................  organic carbon.
PAPR..............................  powered air-purifying respirator.
PEL...............................  permissible exposure limit.
PPM...............................  parts per million.
QRA...............................  quantitative risk assessment.
REA...............................  Regulatory Economic Analysis.
Second Partial Settlement           67 FR 47296 (2002): basis for August
 Agreement.                          14, 2003 NPRM.
SD................................  standard deviation.
SKC...............................  SKC, Inc.
TC................................  total carbon.
USWA..............................  United Steelworkers of America.
[mu]g/cm \2\......................  micrograms per square centimeter.
[mu]g/m \3\.......................  micrograms per cubic meter.
2001 final rule...................  January 19, 2001 DPM final rule.
Amended 2001 final rule...........  2001 final rule amended on February
                                     27, 2002.
2002 final rule...................  February 27, 2002 final rule.
2002 ANPRM........................  Advance Notice of Proposed
                                     Rulemaking published on September
                                     25, 2002.
2003 NPRM.........................  Notice of Proposed Rulemaking
                                     published on August 14, 2003.



[[Page 32869]]

II. Rulemaking Background

    On January 19, 2001, MSHA published a final rule (2001 final rule) 
addressing DPM exposure in underground M/NM mines (66 FR 5706), amended 
on February 27, 2002 at 67 FR 9180 (2002 final rule). The 2001 final 
rule established new health standards for underground M/NM mines that 
use equipment powered by diesel engines. The effective date of the 2001 
final rule was listed as March 20, 2001. On January 29, 2001, AngloGold 
(Jerritt Canyon) Corp. and Kennecott Greens Creek Mining Company filed 
a petition for review of the 2001 final rule in the District of 
Columbia Circuit Court of Appeals. On February 7, 2001, the Georgia 
Mining Association, the National Mining Association (NMA), the Salt 
Institute, and the Methane Awareness Resource Group (MARG) Diesel 
Coalition filed a similar petition in the Eleventh Circuit. On March 
14, 2001, Getchell Gold Corporation petitioned for review of the rule 
in the District of Columbia Circuit. The three petitions were 
consolidated, and are pending in the District of Columbia Circuit. The 
United Steelworkers of America (USWA) intervened in the litigation.
    While these challenges were pending, the AngloGold petitioners 
filed with MSHA an application for reconsideration and amendment of the 
2001 final rule and for postponement of the effective date of the 2001 
final rule pending judicial review. The Georgia Mining Association 
petitioners similarly filed with MSHA a request for an administrative 
stay or postponement of the effective date of the 2001 final rule. On 
March 15, 2001, MSHA delayed the effective date of the 2001 final rule 
until May 21, 2001, in accordance with a January 20, 2001 memorandum 
from the President's Chief of Staff (66 FR 15032). The delay was 
necessary to give Department of Labor officials the opportunity for 
further review and consideration of new regulations. On May 21, 2001 
(66 FR 27863), MSHA published a document in the Federal Register 
delaying the effective date of the 2001 final rule until July 5, 2001. 
The purpose of this delay was to allow the Department of Labor the 
opportunity to engage in further negotiations to settle the legal 
challenges to the 2001 final rule.

A. First Partial DPM Settlement Agreement

    As a result of a partial settlement agreement with the litigants, 
MSHA published two documents in the Federal Register on July 5, 2001 
addressing the 2001 final rule. One document (66 FR 35518) delayed the 
effective date of Sec.  57.5066(b) regarding the tagging provision of 
the maintenance standard; clarified the effective dates of certain 
provisions of the 2001 final rule; and included correcting amendments.
    The second document (66 FR 35521) proposed a rule to clarify Sec.  
57.5066(b)(1) and (b)(2) regarding maintenance and to add a new 
paragraph (b)(3) to Sec.  57.5067 regarding the transfer of existing 
equipment between underground mines. MSHA published these changes as a 
final rule on February 27, 2002 (67 FR 9180) (2002 final rule), with an 
effective date of March 29, 2002.
    Under the first partial settlement agreement, MSHA also conducted 
joint sampling with industry and labor at 31 underground M/NM mines to 
determine existing concentration levels of DPM; to assess the 
performance of the SKC, Inc., Eighty Four, PA (SKC) submicron dust 
sampler with the NIOSH Method 5040; to assess the feasibility of 
achieving compliance with the standard's concentration limits at the 31 
mines; and to assess the impact of interferences on samples collected 
in the M/NM underground mining environment before the limits 
established in the final rule became effective. The final report was 
issued on January 6, 2003.

B. Second Partial Settlement Agreement

    Settlement negotiations continued on the remaining unresolved 
issues in the litigation. On July 15, 2002, the parties signed an 
agreement (second partial settlement agreement) that formed the basis 
for MSHA's August 14, 2003 proposed rule (68 FR 48668) (2003 NPRM). On 
July 18, 2002, MSHA published a document in the Federal Register (67 FR 
47296) announcing, among other things, that the following provisions of 
the 2001 final rule would become effective on July 20, 2002:
     Sec.  57.5060(a), Addressing the interim concentration 
limit of 400 micrograms of TC per cubic meter of air;
     Sec.  57.5061, Compliance determinations; and
     Sec.  57.5071, Environmental monitoring.
    The document also announced that the following provisions of the 
rule would continue in effect:
     Sec.  57.5065, Fueling practices;
     Sec.  57.5066, Maintenance standards;
     Sec.  57.5067, Engines;
     Sec.  57.5070, Miner training; and
     Sec.  57.5075, Diesel particulate records, as they relate 
to the requirements of the rule that went into effect on July 20, 2002.
    The document also stayed the effectiveness of the following 
provisions pending completion of this final rule:
     Sec.  57.5060(d), Permitting miners to work in areas where 
the level of DPM exceeds the applicable concentration limit with 
advance approval from the Secretary;
     Sec.  57.5060(e), Prohibiting the use of personal 
protective equipment (PPE) to comply with the concentration limits;
     Sec.  57.5060(f) Prohibiting the use of administrative 
controls to comply with the concentration limits; and
     Sec.  57.5062, DPM control plan.
    Finally, the July 18, 2002, document outlined the terms of the DPM 
settlement agreement and announced MSHA's intent to propose specific 
changes to the rule, as discussed below.
    On September 25, 2002, MSHA published an Advance Notice of Proposed 
Rulemaking (2002 ANPRM) (67 FR 60199) to amend certain provisions of 
the 2001 DPM rule.
    The comment period closed on November 25, 2002. MSHA received 
comments from underground M/NM mine operators, trade associations, 
organized labor, public interest groups and individuals. On August 14, 
2003, MSHA published the 2003 NPRM in the Federal Register (68 FR 
48668) recommending certain revisions to the DPM rule as part of a 
settlement agreement reached in response to a legal challenge to the 
DPM standard. Public hearings were held in Salt Lake City, Utah; St. 
Louis, Missouri; Pittsburgh, Pennsylvania; and Arlington, Virginia in 
September and October 2003. The comment period closed on October 14, 
2003. On February 20, 2004, MSHA published a document in the Federal 
Register announcing a limited reopening of the comment period on the 
2003 NPRM. This document reopened the comment period to obtain public 
input on three new documents related to the August 14, 2003 rulemaking 
(69 FR 7881). The three documents were as follows:
    (1) United States (U.S.) Department of Health and Human Services, 
Center for Disease Control, National Institute of Occupational Safety 
and Health, ``The Effectiveness of Selected Technologies in Controlling 
Diesel Emissions in an Underground Mine--Isolated Zone Study at 
Stillwater Mining Company's Nye Mine,'' January 5, 2004.
    (2) U.S. Department of Labor, Bureau of Labor Statistics, and U.S. 
Department of Health and Human Services, Center for Disease Control, 
National Institute of Occupational Safety and Health, ``Respirator 
Usage in Private Sector Firms, 2001,'' September, 2003.

[[Page 32870]]

    (3) Chase, Gerald, ``Characterizations of Lung Cancer in Cohort 
Studies and a NIOSH Study on Health Effects of Diesel Exhaust in 
Miners,'' undated, received January 5, 2004.
    The subsequent comment period closed on April 5, 2004. MSHA 
received and reviewed written and oral statements on the 2003 NPRM from 
all segments of the mining community.
    MSHA informed the mining community in both its 2002 ANPRM and its 
2003 NPRM of its intentions to incorporate into the record of the 
current rulemaking the existing rulemaking record, including the risk 
assessment to the 2001 final rule. Commenters were encouraged to submit 
additional evidence of new scientific data related to health risks to 
underground M/NM miners from exposure to DPM.
    This final rule for DPM exposure at M/NM mines is based on 
consideration of the entire rulemaking record, including all written 
comments and exhibits received related to the 2001 final rule as well 
as all related data received to the close of this rulemaking record. To 
serve the interest of the mining community, MSHA is revising Sec. Sec.  
57.5060, 57.5061, 57.5071, and 57.5075 and republishing Sec. Sec.  
57.5065, 57.5066, 57.5067, and 57.5070 of the DPM standards at 30 CFR 
part 57 in order to present all sections in their entirety in this 
document. What follows is a discussion of the specific revisions to the 
2001 DPM standard:
     Sec.  57.5060(a) addressing the interim limit on 
concentration of DPM. MSHA has changed the 2001 final rule's interim 
concentration limit of 400 micrograms of TC per cubic meter of air 
(400TC [mu]g/m3) to a comparable permissible 
exposure limit of 308 micrograms of EC per cubic meter of air 
(308EC [mu]/m3);
     Sec.  57.5060(c) addressing application and approval 
requirements for an extension of time in which to reduce the final DPM 
limit. MSHA has changed the 2001 final rule by requiring MSHA to 
consider economic feasibility along with technological feasibility 
factors in weighing whether to grant special extensions; has deleted 
the limit on the number of special extensions that may be granted to 
each mine; has limited each extension to a period of one year; has 
allowed for annual renewals of special extensions; and has allowed the 
MSHA District Manager, rather than the Secretary, to grant extensions. 
This final rule retains the scope of the 2001 provision for operators 
to apply for extensions to the final DPM limit;
     Sec.  57.5060(d) addressing certain exceptions to the 
concentration limits;
     Sec.  57.5060(e) prohibiting use of PPE to comply with the 
concentration limits;
     Sec.  57.5060(f) prohibiting use of administrative 
controls to comply with the concentration limits. MSHA has changed the 
2001 final rule by implementing the current hierarchy of controls as 
adopted in MSHA's other exposure-based health standards for M/NM mines. 
MSHA's hierarchy includes primacy of engineering and administrative 
controls to the extent feasible to reduce a miner's exposure to the 
PEL, but MSHA continues to prohibit rotation of miners for compliance 
purposes. If a miner's exposure cannot be reduced to the PEL with use 
of feasible controls, controls are infeasible, or do not produce 
significant reductions in DPM exposures, the new final rule requires 
mine operators to supplement a miner's protection with respirators and 
implement a respiratory protection program. This respiratory protection 
program must meet the requirements in existing 30 CFR 57.5005, but 
miners may only use the respirator filters specified by MSHA for DPM in 
this section. Therefore, MSHA removes the 2001 prohibition against use 
of respiratory protection without approval by the Secretary and 
clarifies that use of administrative controls other than rotation of 
miners is allowed;
     Sec.  57.5062, addressing the diesel particulate control 
plan. This final rule removes the existing requirement for a DPM 
control plan; and
     conforming changes to the following existing standards 
that were proposed on August 14, 2003:
    [cir] Sec.  57.5061, addressing compliance determinations;
    [cir] Sec.  57.5071, addressing exposure monitoring; and,
    [cir] Sec.  57.5075, addressing recordkeeping requirements.
    This final rule does not include provisions for written procedures 
for administrative controls, a written respiratory protection program, 
medical examination of miners before they are required to wear 
respiratory protection, and medical transfer of miners who are unable 
to wear respiratory protection for medical and psychological reasons.

III. The Final Concentration Limit

    In the 2002 ANPRM, MSHA notified the mining community that this 
rulemaking would revise both the interim concentration limit of 400 
micrograms per cubic meter of air and the final concentration limit of 
160 micrograms per cubic meter of air under Sec.  57.5060(a) and (b) of 
the 2001 final rule. Some commenters to the ANPRM recommended that MSHA 
propose separate rulemakings for revising the interim and final DPM 
limits to give MSHA an opportunity to gather further information to 
establish a final DPM limit. In the 2003 NPRM, MSHA agreed with these 
commenters and solicited other information from the mining community 
that would lead to an appropriate final DPM standard. Moreover, MSHA 
announced its intentions to publish a separate rulemaking to amend the 
existing final concentration limit in Sec.  57.5060(b). To assist MSHA 
in achieving this purpose, MSHA requested comments on an appropriate 
final permissible exposure limit rather than a concentration limit; and 
asked for information on an appropriate surrogate for measuring miners' 
DPM exposures. MSHA concluded its request for information by clarifying 
that revisions to the final DPM concentration limit would not be a part 
of this rulemaking.
    In their comments to the 2003 NPRM, organized labor requested that 
MSHA lower the final DPM limit below 160 micrograms based on 
feasibility data and the significance of the health risks from exposure 
to DPM. Industry trade associations and individual mine operators 
recommended that MSHA repeal the final limit based on issues related to 
health effects, inability of the mining industry to meet a lower limit 
than 400 micrograms per cubic meter of air, and the need for MSHA to 
have the results from the National Institute for Occupational Safety 
and Health/National Cancer Institute (NIOSH/NCI) study and exposure-
response data.
    MSHA believes that evidence in the current DPM rulemaking record is 
inadequate for MSHA to make determinations regarding revision to the 
final DPM limit.

IV. The 31-Mine Study

A. Summary

    On January 19, 2001, MSHA published a final standard addressing 
exposure of underground metal and nonmetal miners to diesel particulate 
matter (DPM). The standard contained staggered effective dates for 
interim and final concentration limits. The standard was challenged by 
industry trade associations and several mining companies, and the 
United Steelworkers of America (USWA) intervened in the litigation. The 
parties agreed to resolve their differences through settlement 
negotiations with MSHA. Thereafter, MSHA delayed the effective date of 
certain provisions of the standard. As part of the settlement 
negotiations, MSHA agreed to conduct joint sampling with the litigants 
at 31 metal and

[[Page 32871]]

nonmetal underground mines covered by the standard to determine 
existing concentration levels of DPM in operating mines and to measure 
DPM levels in the presence of known or suspected interferences.

    The goals of the study were to use the sampling results and 
related information to assess:
--The validity, precision and feasibility of the sampling and 
analysis method specified by the diesel standard (NIOSH Method 
5040);
--The magnitude of interferences that occur when conducting 
enforcement sampling for total carbon as a surrogate for diesel 
particulate matter (DPM) in mining environments; and,
--The technological and economic feasibility of the underground 
metal and nonmetal (MNM) mine operators to achieve compliance with 
the interim and final DPM concentration limits.
--The parties developed a joint MSHA/Industry study protocol to 
guide sampling and analysis of DPM levels in 31 mines. The parties 
also developed four subprotocols to guide investigations of the 
known or suspected interferences, which included mineral dust, drill 
oil mist, oil mist generated during ammonium nitrate/fuel oil (ANFO) 
loading operations, and environmental tobacco smoke (ETS). The 
parties also agreed to study other potential sampling problems, 
including any manufacturing defects of the DPM sampling cassette. 
(Executive Summary, Report on the 31-Mine Study)

    MSHA requested that NIOSH peer review the draft Report on the 31-
Mine Study, and NIOSH's conclusions were as follows:

    1. Most mines have DPM concentrations higher than 
400TC [mu]g/m\3\.
    2. The impactor was effective in eliminating mineral dust from 
collecting onto the filter analyzed for carbon by NIOSH Method 5040.
    3. The ANFO data was inconclusive.
    4. Oil mist from the stoper drill is a sub-micron aerosol and a 
potential interference. Oil mist contamination from the driller can 
be avoided by sampling upstream of stope or far enough downstream 
that the oil mist has been diluted enough to give minimal TC 
concentrations (if this type of sampling is possible).
    5. No information about the interference of environmental 
tobacco smoke is present in this report.
    6. The inter-laboratory comparison of the NIOSH method 5040 of 
paired punches from the same filter showed reasonable agreement 
between MSHA results and commercial laboratory results and excellent 
agreement between MSHA and NIOSH laboratory results. (Summary of 
Findings of this Report in ``NIOSH Comments and recommendations on 
the MSHA DRAFT report: Report on the Joint MSHA/Industry Study: 
Determination of DPM Levels in Underground Metal and Nonmetal 
Mines,'' dated June 3, 2002)

    On January 6, 2003, MSHA issued its final report entitled, ``MSHA's 
Report on Data Collected During a Joint MSHA/Industry Study of DPM 
Levels in Underground Metal And Nonmetal Mines'' (Report on the 31-Mine 
Study). MSHA's major conclusions drawn from the study are as follows:

--The analytical method specified by the diesel standard gives an 
accurate measure of the TC content of a filter sample and the 
analytical method is appropriate for making compliance 
determinations of DPM exposures of underground metal and nonmetal 
miners.
--SKC satisfactorily addressed concerns over defects in the DPM 
sampling cassettes and availability of cassettes to both MSHA and 
mine operators.
--Compliance with both the interim and final concentration limits 
may be both technologically and economically feasible for metal and 
nonmetal underground mines in the study. MSHA, however, has limited 
in-mine documentation on DPM control technology. As a result, MSHA's 
position on feasibility does not reflect consideration of current 
complications with respect to implementation of controls, such as 
retrofitting and regeneration of filters. MSHA acknowledges that 
these issues may influence the extent to which controls are 
feasible. The Agency is continuing to consult with the National 
Institute of Occupational Safety and Health, industry and labor 
representatives on the availability of practical mine worthy filter 
technology.
--The submicron impactor was effective in removing the mineral dust, 
and therefore its potential interference, from DPM samples. 
Remaining interference from carbonate interference is removed by 
subtracting the 4th organic peak from the analysis. No reasonable 
method of sampling was found to eliminate interferences from oil 
mist or that would effectively measure DPM levels in the presence of 
ETS with TC as the surrogate * * * (Executive Summary, Report on the 
31-Mine Study)

    MSHA's complete report on the 31-Mine Study is contained in the 
rulemaking record.
    MSHA and NIOSH have reviewed the performance characteristics of the 
SKC sampler, and are satisfied that it accurately measures exposures to 
DPM. NIOSH found in laboratory and field data that the SKC DPM cassette 
collected DPM efficiently. In a side protocol of the 31-Mine Study, 
MSHA tested the efficiency of the SKC DPM cassette to avoid mineral 
dust in four different mines and did not measure any mineral dust on 
the filter when the SKC DPM cassette was used. This was confirmed by 
laboratory results at NIOSH. (Noll, J. D., Timko, R. J., McWilliams, 
L., Hall, P., Haney, R., ``Sampling Results of the Improved SKC Diesel 
Particulate Matter Cassette,'' JOEH, 2005 Jan; 2(1):29-37.)
    Results of the 31-Mine Study and the MSHA baseline compliance 
assistance sampling demonstrated that the SKC submicron impactor 
removed potential interferences from mineral dust from the collected 
sample.
    Interference from drill oil mist was found on personal samples 
collected on the stoper and jackleg drillers and on area samples 
collected in the stope where drilling was being performed. Use of a 
dynamic blank did not eliminate drill oil mist interference. Tests to 
confirm whether oil mist from ANFO loading operations could be an 
interference were not conclusive. Blasting did not interfere with 
diesel particulate measurements. MSHA found no reasonable method of 
sampling to eliminate interferences from oil mist when TC is used as 
the surrogate.
    No reliable marker was identified for confirming the presence of 
ETS in an atmosphere containing DPM. Use of the impactor does not 
remove the ETS as an interferent. No reasonable method of sampling was 
found that would effectively measure DPM levels in the presence of ETS 
with TC as the surrogate.
    MSHA has found that the use of EC eliminates potential sampling 
interference from drill oil mist, tobacco smoke, and organic solvents, 
and that EC consistently represents DPM. In comparison to using TC as 
the DPM surrogate, using EC would impose fewer restrictions or caveats 
on sampling strategy (locations and durations), would produce a 
measurement much less subject to questions, and inherently would be 
more precise. Furthermore, NIOSH, the scientific literature, and the 
MSHA laboratory tests indicate that DPM, on average, is approximately 
60 to 80% elemental carbon, firmly establishing EC as a valid surrogate 
for DPM.
    As part of the 31-Mine Study, representatives from MSHA, NIOSH, and 
SKC met to address the following issues:
     The quality of manufactured SKC DPM cassettes;
     The feasibility of adding a dynamic blank filter to the 
SKC DPM cassette; and
     The possibility of putting a number on each SKC DPM 
cassette.
    Also, in its October 16, 2001 letter, MSHA informed SKC about the 
problems that MSHA and the industry encountered using the SKC DPM 
sampling cassette with the submicron impactor. These problems included: 
dark flecks, alleged leaks, loose fitting nozzles and connectors, and 
difficulty in shipping the sampler. As discussed in the report on the 
31-Mine Study, SKC was responsive in addressing those concerns.

[[Page 32872]]

B. Subsequent Activities

    Some industry commenters continued to state that the sampling and 
analytical processes for DPM are too new for regulatory use. Other 
commenters questioned the availability and reliability of the SKC 
impactor.
    MSHA moved expeditiously to help resolve the back-order and 
manufacturing delays for samplers reported in the 31-Mine Study. 
However, operators who sample alongside MSHA continued to request ample 
notice to have enough samplers available. MSHA purchased many of the 
initial production runs of these samplers to conduct its compliance 
assistance baseline sampling. Once the initial orders were filled, the 
sampler became more widely available.
    Some commenters stated that SKC changed the impactor, and that 
NIOSH should test the new SKC sampler and evaluate its comparability to 
the model used in the 31-Mine Study. One of these commenters stated 
that the shelf life of the prior sampler affected TC measurements by 
adsorbing organic carbon (OC) from the polystyrene assembly onto the 
filter media and increasing TC measurement. These commenters questioned 
MSHA's changes to the SKC sampler following completion of the 31-Mine 
Study, and suggested that a defect to the sampler could have affected 
the results of the study. During the 31-Mine Study, MSHA observed that 
the deposit area of the SKC submicron impactor filter was not as 
consistent as those obtained for preliminary evaluation. This was 
attributed to inconsistent crimping of the aluminum foil cone on the 
filter capsule.
    Prior to the 31-Mine Study, MSHA had determined the deposit area of 
the sample filter to be 9.12 square centimeters (cm\2\) with a standard 
deviation of 3.1 percent (%). During the initial phases of the sampling 
analysis of the 31-Mine Study, it became apparent that the variability 
of the deposit area was greater than originally determined. The filter 
area is critical to the concentration calculation. The filter area 
(measured in cm\2\) is multiplied by the results of the analysis 
(micrograms per cm\2\) to get the total filter loading (micrograms). 
While individual filter areas could be measured, it is more practical 
to have a uniform deposit area for the calculations. As a result, NIOSH 
and MSHA consulted with SKC to develop an improved filter cassette 
design. With the cooperation of MSHA and the technical recommendations 
and extensive experimental verification by NIOSH, SKC was able to 
modify their cassette design to produce a consistent and regular DPM 
deposit area, satisfactorily resolving the problem. SKC, in cooperation 
with MSHA and NIOSH, then modified the DPM cassette following the 31-
Mine Study.
    The modification was limited to replacing the foil filter capsule 
with a 32 millimeter (32-mm) ring. This was done to give a more uniform 
deposit area (8.04 cm\2\) with negligible variability, and to 
accommodate two 38-mm quartz fiber filters in tandem (double filters). 
These double filters are assembled into a single cassette along with 
the impactor. The 38-mm filters also eliminate cassette leakage around 
the filters. These modifications were completed and incorporated into 
units manufactured after November 1, 2002.
    The results of this project were prepared into a scientific 
publication, ``Sampling Results of the Improved SKC Diesel Particulate 
Matter Cassette,'' referenced above. This paper has been peer reviewed 
and was published in January 2005. The following abstract was prepared 
for the study results:

    Diesel particulate matter (DPM) samples from underground metal/
non-metal mines are collected on quartz fiber filters and measured 
for carbon content using National Institute for Occupational Safety 
and Health Method 5040. If size selective samplers are not used to 
collect DPM in the presence of carbonaceous ore dust, both the ore 
dust and DPM will collect on the quartz filters, causing the carbon 
attributed to DPM to be artificially high. Because the DPM particle 
size is much smaller than that of mechanically generated mine dust 
aerosols, it can be separated from the larger mine dust aerosol by a 
single stage impactor. The SKC DPM cassette is a single stage 
impactor designed to collect only DPM aerosols in the presence of 
carbonaceous mine ore aerosols, which are commonly found in 
underground nonmetal mines. However, there is limited data on how 
efficiently the SKC DPM cassette can collect DPM in the presence of 
ore dust. In this study, we investigated the ability of the SKC DPM 
cassette to collect DPM while segregating ore dust from the sample. 
We found that the SKC DPM cassette accurately collected DPM. In the 
presence of carbon-based ore aerosols having an average 
concentration of 8 mg/m3, no ore dust was detected on SKC 
DPM cassette filters. We did discover a problem: the surface areas 
of the DPM deposits on SKC DPM cassettes, manufactured prior to 
August 2002, were inconsistent. To correct this problem, SKC 
modified the cassette. The new cassette produced, with 99% 
confidence, a range of DPM deposit areas between 8.05 and 8.28 
cm2, a difference of less than 3%.

    Because the design of the inlet cyclone, impaction nozzles, and the 
impaction plate and the flow rate did not change, the modifications to 
the filter assembly did not alter the collection or separation 
performance of the impactor. Throughout the compliance baseline 
sampling, the impactor has been a consistent and reliable sampling 
cassette.
    Tandem filters were used in the oil mist and ANFO interference 
evaluations during the 31-Mine Study. The top filter collects the 
sample and the bottom filter is a dynamic blank. The dynamic blank 
provides a unique field blank for each DPM cassette. The use of EC as a 
surrogate would resolve the commenter's concern about shelf life and OC 
out-gassing on the filter. Shelf life and OC out-gassing are issues 
relative to OC measurements. These two issues do not apply to an EC 
measurement. Once the cassettes have been preheated during 
manufacturing, there is no source, other than sampling, to add EC to 
the sealed cassette filters.
    MSHA discussed in the preamble to the 2003 NPRM issues related to 
interferences, field blanks and the error factor. Some comments on the 
2003 NPRM still expressed concerns on interferences and further stated 
that the MSHA industrial hygiene studies, conducted to verify the 
magnitude of the interference problem, were not published or peer 
reviewed and should be removed from the rulemaking record. However, 
MSHA, organized labor, and the mining industry, through the 
negotiations process, jointly developed the protocol for conducting the 
31-Mine Study. All of the parties agreed on the protocol following 
numerous discussions among industry, labor, and government experts, and 
had an opportunity to comment and make changes to the document. 
Thereafter, MSHA conducted the study, following the agreed upon 
protocol, and published its results. Before publication, the report was 
peer reviewed by NIOSH. Industry was given an opportunity to publish 
their separate results simultaneously with the government. During this 
rulemaking, industry submitted to MSHA through the notice and comment 
process their conclusions on the 31-Mine Study in a report titled, 
``Technical and Economic Feasibility of DPM Regulations.'' The industry 
report is contained in the rulemaking record, and was considered by 
MSHA in reaching determinations for this final rule.
(1) Interferences
    In response to the question on whether there are interferences when 
EC is used as the surrogate, some commenters stated that interferences 
were thoroughly discussed in the preamble to the 2001 final rule, and 
that reasonable practices to avoid them were stipulated in the rule 
itself. According

[[Page 32873]]

to these commenters, this problem should not be revisited in this 
rulemaking.
    Other commenters maintained that the 31-Mine Study did not contain 
the necessary protocols to address all potential interferences. Thus, 
in their view, MSHA does not have all the data required to answer this 
question. More specifically, some commenters stated that carbonaceous 
particulate in host rock has a smaller diameter than the impactor cut 
point and so, may contaminate EC samples. These commenters then 
concluded that MSHA should propose additional research and seek 
comments on the research before concluding that sampling EC with an 
impactor will eliminate all interference problems. However, no data 
were presented to support this claim or conclusion. Commenters 
submitted no new information relative to interferences in response to 
the 2003 NPRM.
(2) Field Blanks
    A field blank is an unexposed control filter meant to account for 
background interferences and systematic contamination in the field, 
spurious effects due to manufacturing and storage of the filter, and 
systematic analytical errors. The tandem filter arrangement in the 
sample cassette provides a primary filter for collecting an air sample 
and a second filter, behind (after) the primary filter, which provides 
a separate control filter for each sample. This is a much more flexible 
method of sampling for the mining industry, since it eliminates the 
need to send a separate control filter to the analytical lab. MSHA 
informed the public of its intentions to adjust the EC result obtained 
for each sample by the result obtained for the corresponding media 
blank when MSHA measures for compliance purposes. When MSHA conducts 
compliance measurements, MSHA will adjust the result obtained for each 
corresponding sample by the field blank (tandem filter) result. No 
comments or information related to field blanks were submitted to MSHA 
in response to the 2003 NPRM.
    In its comments on the 2002 ANPRM, NIOSH noted that two types of 
blanks, media and field, are normally used for quality assurance 
purposes. A media blank accounts for systematic contamination that may 
occur during manufacturing or storage. A field blank accounts for 
possible systematic contamination in the field. NIOSH does not 
recommend use of field blanks when EC is the surrogate. This is because 
EC measurements are not subject to sources of contamination in the 
field that would affect OC and TC results. Quartz-fiber filters are 
prone to OC vapor contamination in the field and to contamination by 
less volatile OC (such as oils) during handling. However, such 
contamination is irrelevant when EC is the surrogate.
(3) Error Factor
    MSHA intends to cite a violation of the DPMEC exposure 
limit only when MSHA has valid evidence that a violation actually 
occurred. As with all other measurement-based M/NM compliance 
determinations, MSHA will issue a citation only if a measurement 
demonstrates noncompliance with at least 95% confidence. MSHA will 
achieve this 95% confidence level by comparing each EC measurement to 
the EC exposure limit multiplied by an appropriate error factor. 
Generally, an error factor is used to compensate for certain known 
inaccuracies in the sampling and analytical process, including such 
things as the reliability of sampling equipment and precision of 
analytical instrumentation. MSHA will continue to determine that an 
overexposure has occurred when a sample exceeds the interim limit times 
the error factor.
    In this rulemaking, MSHA is discussing the procedure used to obtain 
the error factor. This procedure is further discussed on the MSHA web 
site at http://www.msha.gov under, ``Single Source Page for Metal and Nonmetal 

Diesel Particulate Matter Regulations.'' Error factors are based on 
sampling and analytic errors. The manufacturers of sampling devices 
thoroughly investigate and quantify the error factors for their 
devices. While MSHA does not frequently change an error factor, it 
retains that latitude should significant changes to either analytical 
or sampling technology occur.
    The formula for the error factor was based on three factors 
involved in making an eight-hour equivalent full-shift measurement of 
EC concentration using NIOSH Method 5040: (1) Variability in air volume 
(i.e., pump performance relative to the nominal airflow of 1.7 L/min); 
(2) variability of the deposit area of particles on the filter 
(cm2); and (3) accuracy of the laboratory analysis of EC 
density within the deposit ([mu]g/cm2). Modifications made 
to the sampler since the time of the 31-Mine Study have no bearing on 
the first and third of these factors. Variability of the filter deposit 
area was represented by a 3.1% coefficient of variation, based on an 
experiment carried out before the foil filter capsule in the sampling 
cassette was replaced by a 32-mm ring. Measurements subsequent to 
introduction of the ring show that variability of the filter deposit 
area is now less than 3.1% (Noll, J. D., et al, ``Sampling Results of 
the Improved SKC Diesel Particulate Matter Cassette''). This change 
slightly reduces the error factor stipulated for EC measurements, but 
not by enough to be of any practical significance.
    MSHA's error factor model accounts for the joint and related 
variability in laboratory analysis, and combines that variability with 
pump flow rate, sample collection size, and other sampling and analytic 
variables. MSHA was then able to determine the appropriate error factor 
for EC samples based on a statistically strong database.
    The analytical method (NIOSH 5040) relies on a punch taken from 
inside the deposit area on the sample filter. In effect, the punch is a 
sample of the dust sample. To account for uniformity in the 
distribution of DPM deposited on the filter, as reflected by different 
possible locations at which a punch might be extracted, MSHA compared 
two punches taken from different locations on the same filter to 
evaluate the accuracy of the analytical method. Therefore, variability 
between punch results due to their location on the filter is also 
included in the error factor as calculated by MSHA.
    Commenters to the 2003 NPRM further questioned whether the NIOSH 
Method 5040 has been commercially tested. As in the preamble to the 
2003 NPRM, MSHA has discussed in detail its findings regarding the 
NIOSH Method 5040 in this section. NIOSH's peer review of the 31-Mine 
Study also concludes that the analytical method specified by the diesel 
standard gives an accurate measure of the TC content of a filter 
sample. NIOSH confirmed this position by letter of February 8, 2002, in 
which NIOSH stated that,

MSHA is following the procedures of NIOSH Method 5040, based on our 
review of MSHA P13 (MSHA's protocol for sample analysis by NIOSH 
Method 5040) and a visit to the MSHA laboratory.

V. Compliance Assistance

A. Baseline Sampling Summary

    Under the second partial DPM settlement agreement, MSHA agreed to 
provide compliance assistance to the M/NM underground mining industry 
for a one-year period from July 20, 2002 through July 19, 2003. As part 
of its compliance assistance activities, MSHA agreed to conduct 
baseline sampling of miners' personal exposures at every underground 
mine covered by the 2001 final rule.

[[Page 32874]]

    Our baseline sampling began in October 2002 and continued through 
October 2003. During this period a total of 1,194 valid baseline 
samples were collected. A total of 183 underground M/NM mines are 
represented by this analysis. The number of samples per mine range from 
one to twenty. All 874 valid baseline sampling results in the analysis 
published in the preamble of the 2003 NPRM are included in this updated 
analysis. MSHA is including 320 additional valid samples because MSHA 
decided to continue to conduct baseline sampling after July 19, 2003 in 
response to mine operators' concerns. MSHA has analyzed all baseline 
samples, and updated its analysis. Some of these mines were either not 
in operation or were implementing major changes to ventilation systems 
during the original baseline period. MSHA is including supplementary 
samples from seasonal and intermittent mines, mines that were under-
represented, and mines that were not represented in the analysis 
published in the preamble to the 2003 NPRM. Sixty mines included in the 
former analysis had additional samples taken during the extended 
assistance period. There are 12 mines in this updated analysis that 
were not represented in the 2003 analysis. The results of this sampling 
were used by MSHA in this preamble to estimate current DPM exposure 
levels in underground M/NM mines using diesel equipment. These sampling 
results also assist mine operators in developing compliance strategies 
based on actual exposure levels.
    This section summarizes analytical results of personal sampling for 
DPM collected during compliance assistance. There are a total of 1,206 
samples. However, 12 samples are invalid due to abnormal sample 
deposits, broken cassettes or filters, contaminated backup pads, 
instrument failure or pump failure. Table V-1 lists the frequencies of 
invalid samples within each commodity.
    The mines that were sampled produce clay, sand, gypsum, copper, 
gold, platinum, silver, gem stones, dimension marble, granite, lead-
zinc, limestone, lime, potash, molybdenum, salt, trona, and other 
miscellaneous metal or nonmetal ores. These commodities were grouped 
into four general categories for calculating summary statistics: Metal, 
stone, trona, and other nonmetal (N/M) mines. These categories were 
selected to be consistent with the categories used for analysis of data 
for the 31-Mine Study. Most commodities are well represented in this 
analysis with the average number of valid samples per mine ranging from 
6.0 to 8.2 (average across all mines is 6.5 samples per mine). The 
average number of samples per mine classified as ``Gold Ore Mining, 
N.E.C.'' increased from an average of 2.0 samples per mine published in 
the 2003 NPRM preamble to an average of 4.6 samples in this data set. 
Approximately 79% of all mines sampled during the assistance period 
have four or more results from DPM sampling in this analysis. Table V-3 
lists the number of samples for each category of specific commodity. 
Average number of samples for more general commodity groups is listed 
in Table V-2.
    MSHA used the same sampling strategies for collecting baseline 
samples as it intends to use for collecting samples for enforcement 
purposes. These sampling procedures are described in the Metal and 
Nonmetal Health Inspection Procedures Handbook (PH90-IV-4), Chapter A, 
``Compliance Sampling Procedures'' and Draft Chapter T, ``Diesel 
Particulate Matter Sampling.'' Chapter A includes detailed guidelines 
for selecting and obtaining personal samples for various contaminants. 
All personal samples were collected in the miner's breathing zone and 
for the miner's full shift regardless of the number of hours worked. 
For the 1,194 valid personal samples, 85% were collected for at least 
eight hours. TC and EC levels, as well as DPM levels, are reported in 
units of micrograms per cubic meter for an 8-hour full shift 
equivalent.
    MSHA collected DPM samples with SKC submicron dust samplers that 
use Dorr-Oliver cyclones and submicron impactors. The samples were 
analyzed either at MSHA's Pittsburgh Safety and Health Technology 
Center, Dust Division Laboratory or at the Clayton Laboratory using 
MSHA Method P-13 (NIOSH Analytical Method 5040, NIOSH Manual of 
Analytical Methods (NMAM), Fourth Edition, September 30, 1999) for 
determining the TC content. Each sample was analyzed for organic, 
elemental, and carbonaceous carbon and calculated TC. Raw analytical 
results from both laboratories as well as administrative information 
about the sample were stored electronically in MSHA's Laboratory 
Information Management System.
    If a raw carbon result was greater than or equal to 30 [mu]g/
cm2 of EC or 40 [mu]g/cm2 of TC from the exposed 
filter loading, then the analysis was repeated using a separate punch 
of the same filter. The results of these two analyses were then 
averaged. The companion tandem blank was also tested for the same 
analyses. Otherwise, an unexposed filter from the same manufacturer's 
lot was used to correct for background levels. In the event the initial 
TC result was greater than 100TC [mu]g/cm2, a 
smaller punch of the same exposed filter (in duplicate and with the 
corresponding blank) was taken and used in the analysis. Blank-
corrected averaged results were used in the analysis when the sample 
was tested in duplicate.
    The equation used to calculate a 480-minute (8-hour) full shift 
equivalent (FSE) exposure of TC is Total Carbon Concentration =
[GRAPHIC] [TIFF OMITTED] TR06JN05.014

Where:

EC = The corrected elemental carbon concentration measured in the 
thermal/optical carbon analyzer, [mu]g/cm\2\,
OC = The corrected organic carbon concentration measured in the 
thermal/optical carbon analyzer, [mu]g/cm\2\,
A = The surface area of the deposit on the filter media used to collect 
the sample, cm\2\,
Flow Rate = Flow rate of the air pump used to collect the sample 
measured in Liters per minute, and
480 minutes = Standardized eight-hour work shift.

    All levels of carbon or DPM are reported in 8-hour full shift 
equivalent TC concentrations measured in [mu]g/m\3\.
    Because personal sampling was conducted and no attempt was made to 
avoid interference from cigarette smoke or other OC sources, TC was 
also calculated using the formula prescribed in the second partial DPM 
settlement agreement:

[[Page 32875]]

    Total Carbon Concentration = EC x 1.3.
    MSHA agreed to use the lower of the two values (EC x 1.3 or EC + 
OC) for enforcement until a final rule is published reflecting EC as 
the surrogate.
    The electronic records of the 1,194 samples available for analysis 
were reviewed for inconsistencies. Internally inconsistent or extreme 
values were questioned, researched, and verified. Although no samples 
were invalidated as a result of the administrative verification, 12 
samples (1.0%) were removed from the data set for reasons unrelated to 
the values obtained. The reasons for invalidating these samples are 
listed in Table V-1. These samples were subjected to the same 
laboratory quality assessments as samples collected for compliance 
purposes. Accordingly, MSHA has included 1,194 samples from miners in 
the analyses. Table V-2 is a list of the number of valid samples by 
commodity group.

                                   Table V-1.--Reasons for Excluding Samples.
----------------------------------------------------------------------------------------------------------------
       Reason for excluding from analysis           Metal        Stone        Trona      Other N/M      Total
----------------------------------------------------------------------------------------------------------------
Abnormal Sample Deposit........................            0            1            0            0            1
Cassette/Filter Broken.........................            0            2            0            1            3
Contaminated Backup Pad........................            1            0            0            0            1
Instrument Failure.............................            1            1            0            0            2
Pump Failed....................................            1            4            0            0            5
------------------------------------------------
    Total......................................            3            8            0            1           12
----------------------------------------------------------------------------------------------------------------


                       Table V-2.--Number of Mines and Valid Samples, by Commodity Group.
----------------------------------------------------------------------------------------------------------------
                                                                                               Average number of
                    Commodity group                       Number of mines    Number of valid    valid samples by
                                                                                 samples              mine
----------------------------------------------------------------------------------------------------------------
Metal..................................................                 40                284                7.1
Stone..................................................                115                689                6.0
Trona..................................................                  4                 25                6.3
Other N/M..............................................                 24                196                8.2
--------------------------------------------------------
    Total..............................................                183              1,194                6.5
----------------------------------------------------------------------------------------------------------------

    Table V-3 lists the number of samples collected by specific 
commodities and sorted by average number of samples per mine. Although 
MSHA made efforts to sample all underground M/NM mines covered by this 
rulemaking within the specified time frame, several mines have few or 
no samples for DPM in this analysis. Some M/NM mining operations are 
seasonal in that they are operated intermittently or operate at less 
than full production during certain times. These types of variable 
production schedules limited efforts to collect compliance assistance 
samples. MSHA extended its period of baseline sampling especially to 
incorporate into its analysis those mines with a low sampling frequency 
or where no samples were collected as of March 26, 2003.

                      Table V-3.--Number of Valid Samples per Mine for Specific Commodities
----------------------------------------------------------------------------------------------------------------
                                                                                                      Average
                       Specific commodity                          No. of mines       No. of        samples per
                                                                                      samples          mine
----------------------------------------------------------------------------------------------------------------
Gemstones Mining, N.E.C.........................................               2               5             2.5
Dimension Marble Mining.........................................               3               9             3.0
Limestone.......................................................               2               6             3.0
Talc Mining.....................................................               1               3             3.0
Uranium-Vanadium Ore Mining, N.E.C..............................               1               3             3.0
Gold Ore Mining, N.E.C..........................................              19              87             4.6
Construction Sand & Gravel Mining, N.E.C........................               1               5             5.0
Crushed & Broken Sandstone Mining...............................               1               5             5.0
Hydraulic Cement................................................               1               5             5.0
Lime, N.E.C.....................................................               4              20             5.0
Copper Ore Mining, N.E.C........................................               2              11             5.5
Dimension Limestone Mining......................................               3              18             6.0
Crushed & Broken Limestone Mining, N.E.C........................              90             550             6.1
Crushed & Broken Marble Mining..................................               4              25             6.3
Trona Mining....................................................               4              25             6.3
Crushed & Broken Stone Mining, N.E.C............................               4              28             7.0
Gypsum Mining...................................................               4              29             7.3
Salt Mining.....................................................              14             122             8.7
Clay, Ceramic & Refractory Minerals, N.E.C......................               1               9             9.0
Miscellaneous Metal Ore Mining, N.E.C...........................               1               9             9.0
Lead-Zinc Ore Mining, N.E.C.....................................              10              96             9.6
Platinum Group Ore Mining.......................................               2              20            10.0
Potash Mining...................................................               3              30            10.0
Molybdenum Ore Mining...........................................               2              22            11.0

[[Page 32876]]


Silver Ore Mining, N.E.C........................................               3              36            12.0
Miscellaneous Nonmetallic Minerals, N.E.C.......................               1              16            16.0
                                                                 -----------------
    Average of all samples......................................             183           1,194             6.5
----------------------------------------------------------------------------------------------------------------

    There are 63 different occupations in underground M/NM mines 
represented in this analysis. The most frequently sampled occupations 
are Blaster, Drill Operator, Front-end Loader Operator, Truck Driver, 
Scaling (Mechanical), and Mechanic. Table V-4 lists the number of valid 
samples by occupation and commodity group. Only occupations with 14 or 
more total samples are listed individually. Occupations with fewer 
samples were aggregated into a combined group for this table.

                           Table V-4.--Valid Samples, by Occupation and Mine Category.
----------------------------------------------------------------------------------------------------------------
                   Occupation                       Metal        Stone        Trona      Other N/M      Total
----------------------------------------------------------------------------------------------------------------
Truck Driver...................................           87          152            0           13          252
Front-end Loader Operator......................           40          149            6           19          214
Blaster, Powder Gang...........................           12           98            0           24          134
Scaling (mechanical)...........................            1           66            0           13           80
Drill Operator, Rotary.........................            3           63            0            9           75
Drill Operator, Jumbo Perc.....................           10           19            0            9           38
Mechanic.......................................            7           15            0           12           34
Complete Load-Haul-Dump........................            7            2            0           23           32
Utility Man....................................            6            4           15            4           29
Scaling (hand).................................            4           20            0            2           26
Mucking Mach. Operator.........................           19            1            0            3           23
Roof Bolter, Rock..............................            5            9            0            7           21
Drill Operator, Rotary Air.....................            1           19            0            1           21
Miner, Drift...................................           16            1            0            0           17
Crusher Oper/Worker............................            0           13            0            2           15
Miner, Stope...................................           14            0            0            0           14
All Others Combined............................           52           58            4           55          169
                                                --------------
    Totals.....................................          284          689           25          196        1,194
----------------------------------------------------------------------------------------------------------------

    TC levels calculated by EC x 1.3 were lower than TC levels 
calculated by OC + EC in 858 (72%) of the 1,194 baseline samples. Of 
the 336 samples where TC = OC + EC was the lower value, 68% of the TC = 
EC x 1.3 values were within 12% of the TC = OC + EC value. Table V-5 
summarizes the results of the baseline samples when determining the TC 
level using either EC x 1.3 or OC + EC. Approximately 6.4% of the 
paired results did not concur with respect to the 400TC 
[mu]g/m\3\ standard when measuring TC by the two calculations (OC + EC 
vs. EC x 1.3). Approximately 19.3% of the samples were above the 
400TC [mu]g/m\3\ interim concentration limit when using TC = 
EC x 1.3 and approximately 22.7% were above the concentration limit 
when using TC = OC + EC. There is 93.6% concurrence between the two 
methods of calculating TC and comparing the calculations to the 
400TC [mu]g/m\3\ interim concentration limit.

           Table V-5.--Comparison of Results With 400TC [mu]g/m3 Calculating TC by OC + EC or EC x 1.3
----------------------------------------------------------------------------------------------------------------
                                                                             EC x 1.3
                                                                 --------------------------------
                        All valid samples                         < 400TC [mu]g/  > 400TC [mu]g/       Total
                                                                       m\3\            m\3\
----------------------------------------------------------------------------------------------------------------
OC+EC...........................................................
    <  400TC [mu]g/m\3\..........................................             905              18             923
                                                                         (75.8%)          (1.5%)         (77.3%)
    > 400TC [mu]g/m\3\..........................................              59             212             271
                                                                          (4.9%)         (17.8%)         (22.7%)
                                                                 -----------------
    Total.......................................................             964             230           1,194
                                                                         (80.7%)         (19.3%)        (100.0%)
----------------------------------------------------------------------------------------------------------------

    Table V-6 lists the 26 occupations found to have at least one 
sample in which the level of TC was over the 400TC [mu]g/
m\3\ interim concentration limit (TC = EC x 1.3). Table V-6 is sorted 
by the median (middle) TC result. The median is reported because it is 
a more robust measure of the middle value. Changing a single value 
won't change the median very much. In contrast, the value of the mean 
can be strongly affected by a single value that

[[Page 32877]]

is very low or very high. The table also lists the minimum value, 
maximum value, and the total number of valid samples for these 
occupations. TC values varied widely among all miners' occupations.

     Table V-6.--Occupations With at Least One Sample Greater Than or Equal to 400TC [mu]g/m3 (TC = ECx 1.3)
----------------------------------------------------------------------------------------------------------------
                                                                                        TC, [mu]g/m\3\
                           Occupation                                Total   -----------------------------------
                                                                    samples     Minimum     Median      Maximum
----------------------------------------------------------------------------------------------------------------
Diamond Drill Operator..........................................           1       2,030       2,030       2,030
Ground Control/Timberman........................................           2         368         545         722
Washer Operator.................................................           4         353         438         808
Engineer........................................................           1         438         438         438
Roof Bolter, Mounted............................................          12          98         335       1,063
Mucking Mach. Operator..........................................          23          15         334         872
Miner, Stope....................................................          14         100         283         622
Cleanup Man.....................................................           2          66         283         499
Scoop-Tram Operator.............................................           7          14         272         583
Drill Operator, Rotary Air......................................          21           0         240       1,353
Miner, Drift....................................................          17          16         228       1,459
Blaster, Powder Gang............................................         134           6         227       1,340
Belt Crew.......................................................           8          26         225         502
Roof Bolter, Rock...............................................          21          63         223       1,310
Truck Driver....................................................         252           0         211       1,581
Shuttle Car Operator (diesel)...................................           3          95         201         419
Complete Load-Haul-Dump.........................................          32          19         189         824
Drill Operator, Jumbo Perc......................................          38           5         179       1,098
Drill Operator, Rotary..........................................          75           3         171       1,109
Motorman........................................................           8          59         168         419
Front-end Loader Operator.......................................         214           0         158       2,979
Scaling (mechanical)............................................          80           0         139       1,246
Supervisor, Co. Official........................................          13           1         130         856
Utility Man.....................................................          29          29          94         991
Scaling (hand)..................................................          26          18          87       2,013
Mechanic........................................................          34           0          84         420
----------------------------------------------------------------------------------------------------------------

    Table V-7 and Chart V-1 provide the percent of overexposures among 
the four commodity groups. Chart V-2 provides the number of 
overexposures among the four commodity groups. The metal mines have the 
highest percent of overexposures followed by stone, then other non-
metal mines. For all samples combined, 19.3% were above 
400TC [mu]g/m\3\.

                            Table V-7.--Baseline Samples by Commodity (TC = EC x 1.3)
----------------------------------------------------------------------------------------------------------------
                                                                                                     Percent >
                    Commodity                     Number < 400TC  Number > 400TC   Total Samples   400TC [mu]g/
                                                    [mu]g/m\3\      [mu]g/m\3\                         m\3\
----------------------------------------------------------------------------------------------------------------
Metal...........................................             195              89             284            31.3
Stone...........................................             571             118             689            17.1
Other N/M.......................................             174              22             196            11.2
Trona...........................................              24               1              25             4.0
All Mines.......................................             964             230           1,194            19.3
----------------------------------------------------------------------------------------------------------------

BILLING CODE 4510-43-U

[[Page 32878]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.001

[GRAPHIC] [TIFF OMITTED] TR06JN05.002


[[Page 32879]]


    Chart V-3 shows the number of mines with a specific number of 
overexposures. Examination of the frequency of mines with one or more 
overexposures shows that 68 mines (37%) are in this category. There 
were no mines with more than 12 samples > 400TC [mu]g/m\3\ 
for that mine.
[GRAPHIC] [TIFF OMITTED] TR06JN05.003

    At four of the mines, all samples taken during the assistance 
period were above 400TC [mu]g/m\3\. Between one and ten 
samples were taken at each of these four mines. No overexposures were 
found in 115 (63%) of the mines sampled. (See Chart V-4.)
BILLING CODE 4510-43-C

[[Page 32880]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.004

    Tables V-8 and V-9 summarize sample statistics by commodity for TC 
calculated by TC = EC x 1.3 and TC = EC + OC respectively. Overall, the 
mean TC as calculated by EC x 1.3 is 255 [mu]g/m\3\. The median level 
is 174 [mu]g/m\3\. The mean TC level by OC + EC is 293 [mu]g/m\3\ and 
the median level is 226 [mu]g/m\3\. Individual exposure levels of TC 
vary widely within all commodities and most mines. The commodity 
groupings reported in Tables V-8 and V-9 were chosen to be consistent 
with those reported in the 31-Mine Study and the Quantitative Risk 
Assessment (QRA) for this rule.
    The mean and median TC values for each group, using EC x 1.3, are 
lower than the interim compliance limit of 400 [mu]g/m3. The 
mean (median) TC value for metal mines is 356(271) [mu]g/m3. 
The mean (median) for stone mines is 236(149), other non-metal mines is 
194(148), and trona mines is 105(82) [mu]g/m3. Table V-8 
lists additional statistics for TC values compiled by commodity.

                  Table V-8.--Average Levels of TC by Commodity Measured in [mu]g/m3 (EC x 1.3)
                      [Estimated 8-hour Full Shift Equivalent TC Concentration ([mu]g/m3)]
----------------------------------------------------------------------------------------------------------------
                 TC = EC x 1.3                      Metal        Stone      Other N/M      Trona      All Mines
----------------------------------------------------------------------------------------------------------------
No. of Samples.................................          284          689          196           25        1,194
Maximum........................................        2,026        2,979          960          407        2,979
Median.........................................          271          149          148           82          174
Mean...........................................          356          236          194          105          255
    Std. Error.................................           19           10           12           16            8
    95% CI Upper...............................          392          256          217          138          270
    95% CI Lower...............................          319          216          172           73          239
----------------------------------------------------------------------------------------------------------------

    The mean and median TC values for each group of mines as calculated 
by OC + EC are also lower than the interim compliance limit of 400 
[mu]g/m3. The mean (median) TC value for metal mines is 
370(313) [mu]g/m3. The mean for

[[Page 32881]]

stone mines is 282(209), other non-metal mines is 238(191) and for 
trona mines is 140(126) [mu]g/m3. Table V-9 lists additional 
statistics for TC values compiled by commodity group.

               Table V-9.--Average Levels of TC by Commodity Group Measured in [mu]g/m3 (OC + EC)
                      [Estimated 8-hour Full Shift Equivalent TC Concentration ([mu]g/m3)]
----------------------------------------------------------------------------------------------------------------
                  TC = OC + EC                      Metal        Stone      Other N/M      Trona      All Mines
----------------------------------------------------------------------------------------------------------------
No. of Samples.................................          284          689          196           25        1,194
Maximum........................................        2,045        2,796        1,230          344        2,796
Median.........................................          313          209          191          126          226
Mean...........................................          370          282          238          140          293
    Std. Error.................................           17           11           12           12            8
    95% CI Upper...............................          404          303          263          165          308
    95% CI Lower...............................          336          261          214          115          278
----------------------------------------------------------------------------------------------------------------

    Tables V-10, V-11, and V-12 show summary statistics for whole DPM 
exposures for the baseline sampling and the 31-Mine Study. For baseline 
sampling whole DPM was calculated by EC x 1.3 x 1.25 and by (OC + EC) x 
1.25. The 1.25 factor represents the assumption that TC comprises 80% 
of whole DPM. The other 20% includes the solid aerosols such as ash 
particulates, metallic abrasion particles, sulfates and silicates. The 
vast majority of these particulates are in the sub-micron range.
    Section VI-B discusses the relationship between EC and TC. For 
whole DPM concentrations, the mean (median) value is 444(339) [mu]g/
m3 for metal mines, 295(186) for stone mines, 243(185) for 
other non-metal mines, and 132(102) [mu]g/m3 for trona 
mines. The whole DPM exposures for Table V-11 were calculated as (OC + 
EC) x 1.25.

          Table V-10.--Baseline Whole DPM Concentrations (EC x 1.3 x 1.25, [mu]g/m3), by Mine Category
                   [Estimated 8-hour Full Shift Equivalent Whole DPM Concentration ([mu]g/m3)]
----------------------------------------------------------------------------------------------------------------
             DPM = EC x 1.3 x 1.25                  Metal        Stone      Other N/M      Trona      All Mines
----------------------------------------------------------------------------------------------------------------
Number of Samples..............................          284          689          196           25        1,194
Maximum........................................        2,532        3,724        1,200          509        3,724
Median.........................................          339          186          185          102          218
Mean...........................................          444          295          243          132          318
    Std. Error.................................           23           13           15           20           10
    95% CI Upper...............................          490          320          272          173          338
    95% CI Lower...............................          399          270          214           91          299
----------------------------------------------------------------------------------------------------------------


         Table V-11.--Baseline Whole DPM Concentrations ((EC + OC) x 1.25, [mu]g/m 3), by Mine Category
                   [Estimated 8-hour Full Shift Equivalent Whole DPM Concentration ([mu]g/m3)]
----------------------------------------------------------------------------------------------------------------
             DPM = (EC + OC) x 1.25                 Metal        Stone      Other N/M      Trona      All Mines
----------------------------------------------------------------------------------------------------------------
Number of Samples..............................          284          689          196           25        1,194
Maximum........................................        2,556        3,495        1,538          430        3,495
Median.........................................          392          262          238          158          283
Mean...........................................          463          353          298          175          366
    Std. Error.................................           21           13           16           15           10
    95% CI Upper...............................          505          379          329          206          385
    95% CI Lower...............................          421          327          267          144          347
----------------------------------------------------------------------------------------------------------------

    The mean whole DPM concentration for metal and stone mines (as 
measured by (EC + OC) x 1.25) was significantly lower during baseline 
compliance assistance sampling than the levels measured during the 31-
Mine Study.

                 Table V-12.--31-Mine Study Whole DPM Concentrations ([mu]g/m3) by Mine Category
                   [Estimated 8-hour Full Shift Equivalent Whole DPM Concentration ([mu]g/m3)]
----------------------------------------------------------------------------------------------------------------
                   DPM = (EC + OC) x 1.25                        Metal        Stone      Other N/M      Trona
----------------------------------------------------------------------------------------------------------------
Number of Samples...........................................          116          105           83           54
Maximum.....................................................        2,581        1,845        1,210          331
Median......................................................          491          331          341           82
Mean........................................................          610          466          359           94
    Std. Error..............................................           45           36           27            9
    95% CI Upper............................................          699          537          412          113
    95% CI Lower............................................          522          394          306           75
----------------------------------------------------------------------------------------------------------------


[[Page 32882]]

    Chart V-5 compares the means from Tables V-10, V-11 and V-12. The 
mines selected in the 31-Mine Study (Table V-12) were not randomly 
selected, and the study is, therefore, not considered representative of 
the underground M/NM mining industry. Additionally, the industry has 
continued to change the diesel-powered fleet to low emission engines 
that reduce DPM exposure. Workers inside equipment cabs were not 
sampled during the 31-Mine Study due to possible interference from 
cigarette smoke. During baseline compliance assistance sampling, 
however, personal samples were taken on miners inside cabs.
BILLING CODE 4510-43-U
[GRAPHIC] [TIFF OMITTED] TR06JN05.005

    MSHA received several comments on the baseline sampling. Some 
commenters stated that many mines were sampled in a manner that 
rendered results exceedingly low and not representative of operating 
conditions. Commenters also stated that the results of independent DPM 
sampling conducted by operators indicate MSHA's results underestimate 
DPM exposure. These commenters did not provide data or analyses from 
mine operators' sampling programs to substantiate their claim.
    MSHA compliance specialists collected baseline samples in the same 
manner they have been instructed to use for collecting samples for 
enforcement purposes. It is expected that personal exposure to DPM will 
fluctuate due to variations in day to day operations in a mine. 
Reported levels of DPM are representative of the exposures of the 
highest risk miners identified during compliance assistance. In an 
ideal situation, and with unlimited resources, every potentially 
exposed miner would be individually sampled. It is not necessary or 
practical, however, to sample all miners on a mine property in order to 
evaluate personal exposures. Suspected and potential health hazards may 
be reasonably and adequately evaluated by sampling the maximum risk 
miner in a work area. The maximum risk miner is the one expected to 
have the greatest exposure of all of the miners in the area. Other 
miners in the same work area or area of common exposure sources may 
reasonably be expected to experience lesser concentrations of 
occupational hazards than the maximum risk miner. There may be more 
than one maximum risk miner when activities, operations, and exposure 
sources vary throughout the day. MSHA acknowledges that some samples 
were not taken on the highest possible risk occupation at some mines. 
As previously stated, we continued baseline sampling past the date of 
July 19, 2003 in response to this concern.
    A miner experiences high risk because of the location and type of 
tasks performed relative to the source of the suspected hazard. The 
miner's predicted environment or duties may change during the course of 
the work shift. If the working conditions present during the exposure 
assessment are not typical of the regular mining operation, the sample 
results may not represent the typical exposure for that occupation. 
Compliance specialists strive to characterize the higher exposure 
levels during typical work shifts. The baseline samples are 
representative of the conditions experienced on work shifts during the 
defined compliance assistance period. MSHA has obtained the best 
available information for

[[Page 32883]]

characterizing recent activities at the relevant M/NM mines.

B. DPM Control Technology

    MSHA participated in a number of compliance assistance activities 
directed at improving sampling and assisting mine operators with 
selecting and implementing appropriate DPM control technology. Some of 
these activities were directed to either a segment of the mining 
industry, or to the entire industry, while others were conducted on a 
mine specific basis. In general, activities directed toward a large 
number of mines included outreach programs, workshops, website postings 
and publications, while activities directed at an individual mine 
included evaluation of a specific control technology, and review of the 
technology in use by or available to a specific mine.
    Regional DPM Seminars. During September and October, 2002, MSHA 
conducted regional DPM seminars at the following locations: Ebensburg, 
PA; Knoxville, TN; Lexington, KY; Des Moines, IA; Kansas City, MO; 
Albuquerque, NM; Coeur d'Alene, ID; Green River, WY; and Elko, NV. MSHA 
offered these full-day seminars free of charge in the major underground 
M/NM mining regions of the country to facilitate attendance by key 
mining industry personnel. The seminars covered the health effects of 
DPM exposure, the history and specific provisions of the regulation, 
DPM controls, DPM sampling, and the DPM Estimator, a computerized 
program that calculates DPM concentration reduction.
    NIOSH Diesel Emission and Control Technologies in Underground M/NM 
Mines Workshops. MSHA participated in these two workshops in February, 
2003 in Cincinnati, OH and March, 2003, in Salt Lake City, UT. The 
workshops served several purposes. They provided technical 
presentations and a forum for discussing control technology for 
reducing exposure to particulate matter and gaseous emissions from the 
exhaust of diesel-powered vehicles in underground mines. Additionally, 
they intended to help mine managers, maintenance personnel, safety and 
health professionals, and ventilation engineers select and apply 
control technologies in their mines. Speakers, representing MSHA, 
NIOSH, and several mining companies, provided ample time for questions 
and in-depth technical discussion of issues raised by participants.
    National Stone, Sand & Gravel Association (NSSGA)/MSHA DPM Sampling 
Workshop. This three day seminar, hosted by the Rogers Group, Inc.'s 
Jefferson County Stone and Underground in Louisville, Kentucky, was 
held on December 11 through 13, 2002. On the first day, MSHA reviewed 
DPM sampling procedures, and presented training on pump calibration, 
sample train assembly and note taking. On the second day, participants 
traveled to the Rogers Group Jefferson County Mine to conduct full 
shift sampling on underground miners. Our technical support staff took 
ventilation measurements and collected area samples to assess DPM 
emissions in the mine. On the third day, MSHA reviewed engine emission 
and ventilation measurements. Additionally, MSHA reviewed and discussed 
DPM outreach material. Approximately 10 industry participants attended 
the seminar.
    Nevada Mining Association Safety Committee. In April, 2003, MSHA 
discussed DPM control technologies at a meeting of the Nevada Mining 
Association Safety Committee in Elko, NV. Discussion topics included 
bio-diesel fuel blends, various fuel additives and fuel pre-treatment 
devices, mine ventilation, environmental cabs, clean engines, and 
diesel particulate filter (DPF) systems. Mining company representatives 
discussed their experiences with and perspectives on these 
technologies. MSHA discussed experiences and observations that it made 
at various mines, and results of its laboratory and field testing.
    MSHA South Central Joint Mine Safety and Health Conference. MSHA 
presented a DPM workshop at this conference in April 2003, in New 
Orleans, LA. The workshop included a detailed history and explanation 
of the provisions of the DPM regulation, and a technical presentation 
on feasible DPM engineering controls. At the April 2004 conference in 
Albuquerque, NM, MSHA presented a review of DPM control strategies that 
have generally been adopted in the underground M/NM mining industry.
    National Meeting of the Joseph A. Holmes Safety Association, 
National Association of State Mine Inspection and Training Agencies, 
Mine Safety Institute of America, and Western TRAM (Training Resources 
Applied to Mining). MSHA presented a DPM workshop at this conference in 
June 2003, in Reno, NV. The workshop included a detailed history and 
explanation of the provisions of the regulation, and a technical 
presentation on DPM sampling, analytical tools for identifying and 
evaluating DPM sources in mines, and feasible DPM engineering controls.
    DPM Sampling and Control Workshops. In March 2004, MSHA presented 
full one day workshops in Bloomington, IN and Des Moines, IA. In these 
workshops, MSHA reviewed the sampling procedures that MSHA inspectors 
would use for DPM, and MSHA provided hands on instruction to the 
participants in these procedures. MSHA also presented a review of DPM 
control strategies that have generally been adopted in the underground 
M/NM mining industry.
    Equipment Manufacturers Association (EMA) DPM Workshop. In August 
2003, MSHA conducted a DPM workshop for the EMA in Chicago, IL. At this 
workshop, MSHA reviewed the M/NM DPM regulations, discussed the need 
for clean engine technology, explained engine emission testing for 
mines, reviewed the importance of environmental cabs and discussed 
ventilation issues.
    Web site. Our Web site, http://www.msha.gov, contains a single source page 

for DPM rules for M/NM mines. The page has links to specific topics, 
including:
     Draft Metal and Nonmetal Health Inspection Procedures 
Handbook, Chapter T--Diesel Particulate Matter Sampling.
     DRAFT Diesel Particulate Matter Sampling Field Notes.
     Metal and Nonmetal Diesel Particulate Matter Standard 
Error Factor for TC Analysis.
     MSHA Metal and Nonmetal DPM Standard Compliance Guide of 
August 5, 2003, addressing the interim DPM limit.
     NIOSH Listserver.
     MSHA-NIOSH Diesel Particulate Filter Selection Guide for 
Diesel-powered Equipment in Metal and Nonmetal Mines (Filter Selection 
Guide), last updated February 20, 2003.
     Baseline DPM Sample Results, updated October 2003.
     Presentation from Compliance Assistance Workshop, October 
16, 2002.
     Summary of Requirements: MSHA Standard on Diesel 
Particulate Matter Exposure of Underground Metal and Nonmetal Miners 
that are in effect as of July 20, 2002.
     Link to SKC Web site: SKC Diesel Particulate Matter 
Cassette with Precision-jeweled Impactor.
     Diesel Particulate Matter Control Technologies, last 
updated January 14, 2004.

--Table I: Paper/Synthetic Filters.
--Table II: Non-Catalyzed Particulate Filters, Base Metal Particulate 
Filters, Specially Catalyzed Particulate Filters, and High Temperature 
Disposable Filters.

[[Page 32884]]

--Table III: Catalyzed (Platinum Based) Diesel Particulate Filters.

     Work Place Emissions Control Estimator.
     Federal Register documents concerning this and prior DPM 
rulemakings.
     Public comments on this rulemaking.
     Economic analyses for this rule and prior DPM rules.
     MSHA News Release: MSHA Rules Will Control Miners' 
Exposure to Diesel Particulate, January 18, 2001.
     Program Information Bulletins:

--PIB01-10 Diesel Particulate Matter Exposure of Underground Metal and 
Nonmetal Miners, August 28, 2001.
--PIB02-04 Potential Health Hazard Caused by Platinum-Based Catalyzed 
Diesel Particulate Matter Exhaust Filters, May 31, 2002.
--PIB02-08 Diesel Particulate Matter Exposure of Underground Metal and 
Nonmetal Miners---Summary of Settlement Agreement, August 12, 2002.

    Additionally, our diesel single source page for the coal industry 
contains topics that may also be of interest to the M/NM mining 
industry, particularly for those operations at gassy mines where 
permissible equipment is required.
    Specific control technology studies. Following the settlement 
agreement, MSHA was invited by various mining companies to evaluate the 
effectiveness of different control technologies for DPM, including 
ceramic filters, alternative fuels and a fuel oxygenator. Company 
participation was essential to the success of each test. MSHA evaluated 
ceramic filters in two mines, one where MSHA was the only investigator 
and one where NIOSH was the primary investigator. In our test, MSHA 
evaluated DPM on a production unit with and without ceramic filters 
installed on the loader and trucks. In the NIOSH study a variety of 
ceramic filters were tested in an isolated zone.
    MSHA evaluated bio-diesel fuel in two mines. In one, MSHA evaluated 
a 20% and a 50% recycled bio-diesel fuel and a 50% new bio-diesel. In 
the other, MSHA evaluated a 35% recycled bio-diesel fuel and a 35% new 
bio-diesel.
    MSHA evaluated the fuel catalyst system in one mine. MSHA sampled 
the mine exhaust with fuel catalyst systems installed on all production 
equipment, and also without the units installed.
    MSHA evaluated water emulsion diesel fuel in four mines.
    Following is a summary of the individual mine technology evaluation 
studies:
    Kennecott Greens Creek Mining Company: MSHA participated with 
Kennecott Greens Creek Mining Company in a collaborative test to verify 
the efficiency of catalyzed ceramic DPFs for reducing diesel 
particulate emissions. The goal of the testing was to identify site-
specific practical mine-worthy filter technology.
    This series of tests was designed to determine the reduction in 
emissions and personal exposure that can be achieved when ceramic 
filters are installed on a loader and associated haulage trucks 
operating in a production stope. MSHA also determined relative engine 
gaseous and DPM emissions for the equipment under specific load 
conditions.
    MSHA conducted the tests over a two-week period. MSHA sampled three 
shifts with ceramic after-filters installed; and three shifts without 
the after-filters. MSHA also collected personal samples to assess 
worker exposures, and area samples to assess engine emissions. MSHA 
took both gaseous and diesel particulate measurements.
    Sampling results indicate significant reductions in both personal 
exposures and engine emissions. These results also indicated that 
factors such as diesel particulate contamination of intake air, stope 
ventilation parameters, and isolated atmospheres in vehicle cabs as 
well as the ceramic DPFs may have a significant impact on personal 
exposures. The following findings and conclusions were obtained from 
the test:
    1. The results of the raw exhaust gas measurements conducted during 
the test indicate that the engines were operating properly.
    2. The ceramic filters installed on the machines used in this test 
do not adversely affect machine operation. Even with some apparent 
visual cracking from the rotation of the filter media, the ceramic 
filters removed more than 90% of the DPM. The filters passively 
regenerated during machine operation.
    3. The Bosch smoke test provides an indication of filter 
deterioration; however, the colorization method does not quantify the 
results.
    4. Personal DPM exposures were reduced by 60% to 68% when after-
filters were used.
    5. CO levels decreased by up to one-half while the catalyzed 
filters were used. There appeared to be an increase in NO2 
(Nitrous Dioxide) while catalyzed filters were being used; however, it 
is unclear whether this increase was due to data variability, changes 
in ventilation rate, or the use of the catalyzed filters.
    6. The use of cabs reduced DPM exposure by 75% when DPFs were in 
use and by 80% when DPFs were not in use.
    7. Ventilation airflow was provided to the stopes through fans with 
rigid and bag tubing. Airflow was the same or greater than the 
Particulate Index, but typically lower than the gaseous ventilation 
rate.
    8. The use of ceramic DPFs reduced average engine DPM emissions by 
96%.
    9. The reduction in personal exposure was not attributed solely to 
DPF performance because other factors such as ventilation, upwind 
equipment use, and cabs also influence personal exposure.
    Carmeuse North America, Inc., Maysville Mine: MSHA entered into a 
collaborative effort with NIOSH, industry, and the Kentucky Department 
of Energy to test DPM emissions and exposures when using various blends 
of bio-diesel fuels in an underground stone mine. As part of our 
compliance assistance program, MSHA provided support to mining 
operations to evaluate diesel particulate control technologies. The 
test was initiated by the industry partner, and, along with NIOSH, MSHA 
provided support for test design, data collection, and sample and data 
analysis. The project was funded by Carmeuse and Kentucky Department of 
Energy, through the Kentucky Clean Fuels Coalition.
    The initial test was conducted in two phases, using a 20% and a 50% 
bio-diesel blend of recycled vegetable oil (RVO), each mixed with low 
sulfur No. 2 standard diesel fuel. Baseline conditions were established 
using low sulfur No. 2 standard diesel fuel. In a third phase of the 
test, a 50% blend of new soy bio-diesel fuel was tested.
    Area samples were collected at shafts to assess equipment 
emissions. Personal samples were collected to assess worker exposure. 
These samples were analyzed by NIOSH using the NIOSH 5040 method to 
determine TC and EC concentrations. Results indicate that significant 
reductions in emissions and worker exposure were obtained for all bio-
diesel mixtures. These reductions were in terms of both elemental and 
TC. Results for the 20% and 50% RVO indicated 33% and 69% reductions in 
DPM emissions, respectively. Results for the tests on the 50% blend of 
new soy bio-diesel fuel, showed about a 37% reduction in DPM emissions.
    Carmeuse North America, Inc., Black River Mine: Following the 
success of the bio-diesel tests at Maysville Mine, Carmeuse requested 
our assistance in continuing the bio-diesel optimization testing at 
their Black River Mine. Two bio-diesel blends were tested, and a 
baseline test was made. In each test

[[Page 32885]]

personal exposures and the mine exhaust were tested for two shifts. The 
two bio-diesel blends included a 35% RVO and a 35% blend of new soy 
oil. Results for the 35% RVO showed a 32% reduction in DPM emissions. 
Results of the 35% blend of new soy bio-diesel fuel showed an 
approximate 16% reduction in DPM emissions.
    Stone Creek Brick Company, Water Emulsion Fuel Tests: During the 
Stone Creek Brick Company compliance assistance visit, MSHA identified 
several control strategies that would reduce DPM emissions and 
exposures. These strategies included: The installation of clean 
engines, the use of alternative fuels, and an increase in mine 
ventilation. The mine chose to implement alternative fuel use followed 
by an engine replacement program. MSHA provided in-mine testing to 
evaluate the impact of using an alternative fuel. The company chose to 
use a water emulsion fuel. This fuel is an EPA approved fuel, 
consisting of a 20% blend of water with No. 2 diesel fuel. A surfactant 
is added to keep the water and diesel fuel from separating. MSHA 
sampled at the mine before (using No. 2 diesel fuel) and after the 
implementation of the fuel. MSHA collected personal samples to evaluate 
the worker exposure and area samples to evaluate emissions.
    Results of the testing showed that the highest exposure was reduced 
from 823TC [mu]g/m3 to 321TC [mu]g/
m3 (61% reduction). EC emissions were reduced by 49% and TC 
emissions were reduced by 3%. The lack of a reduction in TC emissions 
was attributed to the lower combustion temperature resulting from the 
water emulsion fuel and the older engine technology in use. The older 
engines have larger injector nozzles which do not provide efficient 
fuel burning. The mine has been using the fuel for approximately one 
year, and continues to be satisfied with the results.
    Carmeuse North American, Inc., Maysville Mine, Water Emulsion Fuel 
Tests: MSHA provided assistance to Carmeuse North American, Inc., to 
evaluate summer and winter blends of a water emulsion fuel at their 
Maysville Mine. For the first test, emission reductions for a 10% blend 
(winter blend) of water with No. 2 diesel fuel was compared to a 35% 
blend of RVO. Emission reductions were compared to both a 35% blend of 
RVO and standard No. 2 diesel fuel. MSHA collected personal samples to 
evaluate the worker exposure and area samples to evaluate emissions.
    Results of the testing showed that the highest average exposure 
(high scaler working outside a cab) was reduced from 254TC 
[mu]g/m3 to 145TC [mu]g/m3 (43% 
reduction) when changing from RVO to the water emulsion. EC emissions 
were reduced by 52% and TC emissions were reduced by 49% for the water 
emulsion to 35% RVO fuel comparison. EC emissions were reduced by 77% 
and TC emissions were reduced by 74% for the water emulsion to standard 
diesel fuel comparison.
    For the second test, emission reductions for a 20% blend (summer 
blend) of water with No. 2 diesel fuel was compared to a 35% blend of 
RVO. Emission reductions were compared to both a 35% blend of RVO and 
standard No. 2 diesel fuel. The comparison to No. 2 diesel fuel was 
obtained by combining the water emulsion to the 35% RVO results and 
previously obtained 35% RVO to No. 2 diesel fuel results. MSHA 
collected personal samples to evaluate the worker exposure and area 
samples to evaluate emissions. For the summer blend, EC emissions were 
reduced by 60% and TC emissions were reduced by 59% for the water 
emulsion to 35% RVO fuel comparison. EC emissions were reduced by 81% 
and TC emissions were reduced by 79% for the water emulsion to standard 
diesel fuel comparison.
    Carmeuse North American, Inc., Black River Mine, Water Emulsion 
Fuel Tests: MSHA provided assistance to Carmeuse North American, Inc. 
to evaluate summer and winter blends of a water emulsion fuel at their 
Black River Mine. For these tests, emission reductions for 10% and 20% 
blends (winter blend) of water with No. 2 diesel fuel was compared to a 
35% blend of RVO. Emission reductions were compared to both a 35% blend 
of RVO and standard No. 2 diesel fuel. MSHA collected personal samples 
to evaluate the worker exposure and area samples to evaluate emissions.
    For the winter blend (10%), EC emissions were reduced by 46% and TC 
emissions were reduced by 45% for the water emulsion to 35% RVO fuel 
comparison. EC emissions were reduced by 63% and TC emissions were 
reduced by 62%, for the water emulsion to standard No. 2 diesel fuel 
comparison.
    For the summer blend (20%), EC emissions were reduced by 61% and TC 
emissions were reduced by 54% for the water emulsion to 35% RVO fuel 
comparison. EC emissions were reduced by 73% and TC emissions were 
reduced by 68% for the water emulsion to standard diesel fuel 
comparison.
    Martin Marietta, Durham Mine, Water Emulsion Fuel Tests: MSHA 
provided assistance to Martin Marietta to evaluate a summer blend of 
water emulsion fuel at their Durham Mine. This was a multi-level mine, 
with a 15% ramp between levels. For this test, emissions for a 20% 
blend of water with No. 2 diesel fuel was compared to standard No. 2 
diesel fuel. MSHA collected personal samples to evaluate the worker 
exposure and area samples to evaluate emissions. Even with the 15% 
ramps, the loss in horsepower due to the fuel did not adversely effect 
the mine operations.
    Results of the testing showed that the highest average exposure 
(powder crew working outside a cab) was reduced from 372TC 
[mu]g/m\3\ to 54TC [mu]g/m\3\ (85% reduction) when changing 
from No. 2 diesel fuel to the water emulsion. EC emissions were reduced 
by approximately 80% for the water emulsion compared to standard 
diesel.
    Rogers Group, Jefferson County Mine: MSHA was invited to this mine 
to evaluate a fuel catalyst system that was installed in the fuel line 
of the diesel equipment. The company had installed the units to 
increase fuel economy, and sought to determine the effects of the units 
on DPM. Prior to the units having been installed, MSHA had conducted 
baseline sampling and had collected personal samples on production 
workers and area samples in the mine exhaust airflow. After the units 
were installed on loaders and trucks and the units had accumulated 100 
hours of operation, sampling was repeated. Results indicated that the 
use of the fuel catalyst had no measurable effect on either DPM 
exposure or emissions.
    Summary of DPM control technology: In addition to conducting 
baseline sampling and providing assistance in developing DPM control 
strategies at specific mines, MSHA assessed the effectiveness of 
various DPM controls during and following the compliance assistance 
period. These controls included alternative fuels, fuel oxygenators, 
environmental cabs and ceramic DPFs. Alternative fuels evaluated 
included various blends of bio-diesel fuels (including both Virgin Soy 
Oil (VSO) and RVO), No. 1 diesel fuel, and water emulsion fuels.
    The resulting reduction in DPM emissions for each of these controls 
is given in Chart V-6. All reductions are compared to diesel emissions 
with low sulfur No. 2 diesel fuel. All bio-diesel tests were conducted 
at mines with relatively clean engines. The first water emulsion test 
was conducted at a mine utilizing older engines. Subsequent water 
emulsion tests were conducted at mines utilizing clean engines with 
oxidation catalytic converters.
BILLING CODE 4510-43-U

[[Page 32886]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.006

BILLING CODE 4510-43-C

Assistance for Developing Control Strategies

    Martin Marietta Aggregates: MSHA provided compliance assistance 
during full-day visits at the North Indianapolis Mine and the Parkville 
Mine in March, 2003, and at the Kaskaskia Mine and the Manheim Mine in 
May, 2003. MSHA

[[Page 32887]]

reviewed each mine's DPM sampling history, current operating and 
equipment maintenance practices, ventilation, diesel equipment 
inventory, and steps taken to date and future plans to reduce DPM 
exposures. MSHA discussed the full range of engineering controls, 
demonstrated an exhaust temperature measurement and data logging 
system, and presented a spreadsheet for using such data to select 
appropriate filter systems. MSHA presented a simple approach for 
measuring the effectiveness of cab air filtering and pressurization 
systems, identified the highest DPM-emitting equipment (so future 
equipment-specific DPM control efforts could be appropriately focused), 
and discussed the likely effect of various ventilation system upgrades.
    Rogers Group, Oldham County Mine: MSHA provided compliance 
assistance at this mine during a full-day visit in November 2002. MSHA 
conducted extensive DPM sampling at the mine, collecting both personal 
exposure samples and area samples. Further, MSHA collected DPM samples 
from both inside and outside of equipment cabs. No personal samples 
exceeded 160TC [mu]g/m\3\. MSHA reviewed current operating 
and equipment maintenance practices, ventilation, diesel equipment 
inventory, and steps taken to date and future plans to reduce DPM 
exposures. MSHA discussed the full range of engineering controls. 
Results from this survey indicate the environmental cabs significantly 
reduced the DPM exposure of equipment operators.
    Rogers Group, Jefferson County Mine: MSHA provided compliance 
assistance at this mine during a full-day visit in December 2002. MSHA 
collected both personal exposure samples and area samples. The highest 
personal sample, collected on the loader, was 468TC [mu]g/m 
\3\. This loader was operated with the window open. MSHA reviewed 
current operating and equipment maintenance practices, ventilation, 
diesel equipment inventory, and steps taken to date and future plans to 
reduce DPM exposures. Mechanical ventilation was provided for the mine. 
MSHA discussed the full range of engineering controls, demonstrated an 
exhaust temperature measurement and data logging system, and presented 
a spreadsheet for using such data to select appropriate filter systems. 
MSHA presented a simple approach for measuring the effectiveness of cab 
air filtering and pressurization systems, identified the highest DPM-
emitting equipment (so future equipment-specific control efforts could 
be appropriately focused), and discussed the likely effect of various 
ventilation system upgrades.
    Nalley and Gibson, Georgetown Mine: MSHA provided compliance 
assistance at this mine during a full-day visit in May 2003. MSHA 
reviewed current operating and equipment maintenance practices, 
ventilation, diesel equipment inventory, and steps taken to date and 
future plans to reduce DPM exposures. MSHA collected DPM samples to 
assess improvements since the baseline sampling. At that time, 
mechanical ventilation provided airflow to the mine. MSHA discussed the 
full range of engineering controls, demonstrated an exhaust temperature 
measurement and data logging system, and presented a spreadsheet for 
using such data to select appropriate filter systems. MSHA presented a 
simple approach for measuring the effectiveness of cab air filtering 
and pressurization systems, identified the highest DPM-emitting 
equipment (so future equipment-specific DPM control efforts could be 
appropriately focused), and discussed the likely effect of various 
ventilation system upgrades.
    Stone Creek Brick Company: MSHA provided compliance assistance at 
this mine during a full-day visit in May 2003. MSHA reviewed current 
operating and equipment maintenance practices, ventilation, diesel 
equipment inventory, and steps taken to date and future plans to reduce 
DPM exposures. MSHA collected DPM samples from underground miners. The 
mine was using mechanical ventilation. None of the equipment had 
environmental cabs. MSHA discussed the full range of engineering 
controls, presented a spreadsheet for using such data to select 
appropriate filter systems, identified the highest DPM-emitting 
equipment (so future equipment-specific DPM control efforts could be 
appropriately focused), and discussed the likely effect of various 
ventilation system upgrades.
    Wisconsin Industrial Sand Co., Maiden Rock Mine: MSHA provided 
compliance assistance at this mine during a full-day visit in May 2003. 
MSHA reviewed the mine's current operating and equipment maintenance 
practices, ventilation, diesel equipment inventory, and steps taken to 
date and future plans to reduce DPM exposures. MSHA discussed the full 
range of engineering controls, presented a spreadsheet for using such 
data to select appropriate filter systems, and identified the highest 
DPM-emitting equipment so future equipment-specific DPM control efforts 
could be appropriately focused.
    Gouverneur Talc Company, Inc., No. 4 Mine: MSHA provided compliance 
assistance at this mine during a full-day visit in May 2003. DPM 
samples were collected on underground workers. MSHA reviewed then 
current operating and equipment maintenance practices, ventilation, 
diesel equipment inventory, and steps taken to date and future plans to 
reduce DPM exposures. MSHA discussed the full range of engineering 
controls, demonstrated an exhaust temperature measurement and data 
logging system, and presented a spreadsheet for using such data to 
select appropriate filter systems. MSHA presented a simple approach for 
measuring the effectiveness of cab air filtering and pressurization 
systems, identified the highest DPM-emitting equipment (so future 
equipment-specific control efforts could be appropriately focused), and 
discussed the likely effect of various ventilation system upgrades.
    Additional specific mine compliance assistance: Following the 
initial baseline sampling period, MSHA compiled a list of mines having 
at least one DPM sample which exceeded the 400TC [mu]g/m\3\ 
limit. Of the 183 mines sampled, approximately 69 mines had at least 
one sample over the 400TC [mu]g/m\3\ interim TC limit. Of 
the 69 mines with one or more overexposures, 44 used room and pillar 
mining methods. These include stone mines, salt mines and a potash 
mine. Of the 44 room and pillar mines, MSHA provided specific 
compliance assistance to 36 of these mines (two mines were closed and 
two mines declined assistance). Although trona mines use room and 
pillar mining methods, they were not visited because they were in 
compliance with the 400TC [mu]g/m\3\ limit. The remaining 15 
mines with overexposures were multilevel metal mines using a variety of 
stoping mining methods. Industry seminars were provided to assist these 
mines.
    Typically, the high risk workers in the mines visited were the face 
workers that worked outside an environmental cab. Production loader and 
truck operators had elevated exposures when they either did not have an 
environmental cab or when the cab was not being properly maintained. 
Additional high risk workers include the blasting crew, drillers, and 
roof bolters.
    During each mine visit, DPM samples were collected unless the mine 
had been recently sampled or the mine reported no additional DPM 
controls had been implemented since MSHA's previous sampling was 
conducted. The DPM controls, including engines, ventilation, cabs, 
fuels and work practices, were reviewed with mine management. Specific 
engine emission rates, mine ventilation rates, cab pressures and

[[Page 32888]]

work practices were determined. At some mines, a temperature trace of 
an engine exhaust was made. The information was entered into a computer 
spreadsheet model to assess the effect of control changes on DPM levels 
and to assist the mine in developing a DPM control strategy.
    Laboratory Compliance Assistance: In addition to the compliance 
assistance field tests, our diesel testing laboratory has been working 
with manufacturers to evaluate various types of DPM control 
technologies. Certain of these technologies can be applied in either 
underground M/NM or coal mines.
    Evaluating paper/synthetic media as exhaust filters: MSHA has 
evaluated paper/synthetic media as exhaust filters. These filters have 
shown DPM removal efficiencies in excess of 90% in the laboratory when 
tested on our test engine using the test specified in subpart E of part 
7. The laboratory has tested approximately 20 different paper/synthetic 
media from 10 different filter manufacturers. Although much of this 
work is directed to underground coal mine applications for use on 
permissible equipment, this technology is available for use on 
permissible equipment that is used in underground gassy M/NM mines. In 
addition, some underground coal mine operators have considered adding 
exhaust heat exchanger systems to nonpermissible equipment in order to 
use the paper/synthetic filters in place of ceramic filters. The heat 
exchanger is needed to reduce the exhaust gas temperature to below 
302[deg] F for these types of filters. This could also be an option for 
equipment in M/NM mines, particularly gassy mines where permissible 
equipment is required.
    Evaluating Ceramic Filter Systems: MSHA worked with six ceramic 
filter manufacturers to evaluate the effects of their catalytic wash-
coats on NO2 production. As discussed under the 
``Effectiveness of the DPM Estimator'' portion of this preamble, 
catalytic wash-coats on the ceramic filters may cause increases in 
NO2 levels. MSHA used our test engine (Caterpillar 3306 
PCNA) and followed the test procedures in subpart E of 30 CFR part 7. 
The DPM single source webpage lists the ceramic filters that have 
significantly increased NO2 levels, as well as the ceramic 
filters that are not known to increase NO2 levels. MSHA 
tested the DPM removal efficiencies of these filters during the 
laboratory tests. The efficiency results agree with the efficiencies 
posted on our web site DPM Control Technologies with Percent Removal 
Efficiency page (85% for cordierite and 87% for silicon carbide). 
Finally, MSHA worked with NIOSH during these tests to collect DPM 
samples for EC analysis using the NIOSH 5040 method. The laboratory 
results showed that the filters removed EC at up to 99% efficiency.
    Evaluation of Fuel Oxygenator System: MSHA'S laboratory completed 
tests on the Rentar \TM\ in-line fuel catalyst. The Rentar \TM\ unit 
was installed on a Caterpillar\TM\ 3306 ATAAC, which was coupled to a 
generator. MSHA used an electrical load bank to load the engine under 
various operating conditions. To establish a baseline, MSHA tested the 
engine for gaseous and DPM emissions without the Rentar \TM\ unit. The 
unit was then installed, and MSHA operated the engine for a 100 hour 
break-in period. MSHA then repeated the gaseous and DPM emission 
measurements. The test results of the one laboratory evaluation for 
this control device to date showed no significant reductions in whole 
diesel particulate, however, the data did not show any adverse effects 
on the raw whole DPM exhaust emission. NIOSH's results were consistent 
with MSHA's results, and showed no significant EC reductions and no 
adverse effects on the engine's emissions. MSHA has discussed with 
Rentar \TM\ further laboratory tests.
    Evaluation of a Magnet System: MSHA performed laboratory tests for 
Ecomax, a manufacturer of a magnet system installed on the fuel line, 
oil filter, air intake and radiator. MSHA performed a preliminary field 
test of this product at a surface aggregate operation. The magnetic 
device demonstrated a 30% reduction in CO levels. The laboratory tests 
were performed with the Ecomax system installed and compared to our 
baseline engine data. The test results of the one laboratory evaluation 
for this control device to date showed no significant reductions in 
whole diesel particulate, however, the data did not show any adverse 
effects on the raw DPM exhaust emissions.
    Evaluation of the Fuel Preporator [reg] System: MSHA's 
laboratory tested a fuel preparator system. The system is designed to 
remove collected air from the fuel system for better fuel combustion. 
The results of the system installed were compared to the baseline 
engine. The test results of the one laboratory evaluation for this 
control device to date showed no significant reductions in whole diesel 
particulate, however, the data did not show any adverse effects on the 
raw DPM exhaust emissions. NIOSH also conducted tests in our lab on the 
Fuel Preporator [reg] and the results were consistent with 
MSHA's results. There were no significant EC reductions and no adverse 
effects on the engine's emissions.

VI. DPM Exposures and Risk Assessment

A. Introduction

    In support of the 2001 final rule, MSHA published a comprehensive 
risk assessment (66 FR at 5752-5855, with corrections at 35518-35520). 
In the following discussion, we will refer to the risk assessment 
published in conjunction with the 2001 final rule as the ``2001 risk 
assessment.''
    The 2001 risk assessment presented MSHA's evaluation of health 
risks associated with DPM exposure levels encountered in the mining 
industry. This was based on a review of the scientific literature 
available through March 31, 2000, along with consideration of all 
material submitted during the applicable public comment periods.
    The 2001 risk assessment was divided into three main sections. 
Section 1 (66 FR at 5753-5764) contained a discussion of U.S. miner 
exposures based on field data collected through mid-1998. An important 
conclusion of this section was that, prior to the 2001 final rule,

* * * median dpm concentrations observed in some underground mines 
are up to 200 times as high as mean environmental exposures in the 
most heavily polluted urban areas [footnote deleted] and up to 10 
times as high as median exposures estimated for the most heavily 
exposed workers in other occupational groups. [66 FR at 5764]

    Section 2 of the 2001 risk assessment (66 FR at 5764-5822) reviewed 
the available scientific literature on health effects associated with 
DPM exposures. This review covered effects of both acute and chronic 
exposures and also contained a discussion of potential mechanisms of 
toxicity. The review of acute effects included anecdotal reports of 
symptoms experienced by exposed miners, studies based on exposures to 
diesel emissions, and studies based on exposures to particulate matter 
in the ambient air. The review of chronic effects included studies 
based specifically on exposures to diesel emissions and studies based 
more generally on exposures to fine particulate matter in the ambient 
air. As part of this discussion, MSHA evaluated 47 epidemiologic 
studies examining the prevalence of lung cancer within groups of 
workers occupationally exposed to DPM and discussed the criteria used 
to evaluate and rank these studies (66 FR at 5774-5810). For both acute 
and chronic health effects, information from

[[Page 32889]]

genotoxicity studies and studies on laboratory animals was discussed in 
the separate subsection on mechanisms of toxicity. Section 2 of the 
2001 risk assessment also explained MSHA's rationale for utilizing 
certain types of information whose relevance had been questioned during 
the public comment periods: health effects observed in animals, health 
effects that are reversible, and health effects associated with fine 
particulate matter in the ambient air (66 FR at 5765-55767).
    In section 3 of the 2001 risk assessment (66 FR at 5822-5855), MSHA 
evaluated the best available evidence to ascertain whether exposure 
levels currently existing in mines warranted regulatory action pursuant 
to the Mine Act. To do this, MSHA addressed three questions: (a) 
Whether health effects associated with occupational DPM exposures 
constitute a ``material impairment'' to miner health or functional 
capacity; (b) whether exposed miners were at significant excess risk of 
incurring any of these material impairments; and (c) whether the 2001 
final rule would substantially reduce such risks. After careful 
consideration of all the submitted public comments, the 2001 risk 
assessment established three main conclusions:

    1. Exposure to dpm can materially impair miner health or 
functional capacity. These material impairments include acute 
sensory irritations and respiratory symptoms (including allergenic 
responses); premature death from cardiovascular, cardiopulmonary, or 
respiratory causes; and lung cancer.
    2. At dpm levels currently observed in underground mines, many 
miners are presently at significant risk of incurring these material 
impairments due to their occupational exposures to dpm over a 
working lifetime.
    3. By reducing dpm concentrations in underground mines, the rule 
will substantially reduce the risks of material impairment faced by 
underground miners exposed to dpm at current levels.

    The third of these conclusions was supported primarily by a 
quantitative risk assessment for lung cancer (66 FR at 5848-5854).
    Throughout the current rulemaking, MSHA advised the mining 
community of its intent to include the 2001 risk assessment in the 
current rulemaking record to support this final rule. In this preamble, 
MSHA supplements the 2001 risk assessment with new exposure data and 
health effects literature published after March 31, 2000. MSHA asked 
that public comment be focused on this supplemental information. 
Nevertheless, some commenters presented critiques challenging the 2001 
risk assessment and disputing scientific support for any DPM exposure 
limit, especially by means of an EC surrogate. Other commenters 
endorsed the 2001 risk assessment and stated that recent scientific 
publications support MSHA's conclusions.
    MSHA also received a number of comments from the mining industry 
suggesting that the risk assessment lacks an adequate scientific 
foundation and does not comply with present requirements under OMB and 
information quality guidelines to use the best available, peer reviewed 
science. The risk assessment sustaining this final rule uses the best 
available, peer-reviewed scientific studies. It supplements the risk 
assessment sustaining the 2001 final rule and the existing coal DPM 
final rule also promulgated on January 19, 2001 (66 FR 5526) (coal 
rule). The coal rule was unchallenged by the mining community.
    Before promulgating the 2001 final rule, MSHA provided a copy of 
its draft risk assessment supporting the 2001 rule for peer review to 
two experts in the field of epidemiology and risk assessment. These 
experts evaluated the overall methodology used by MSHA in the draft 
risk assessment, the appropriateness of the studies selected by MSHA, 
and MSHA's conclusions. MSHA had the draft independently peer-reviewed, 
published the evidence and tentative conclusions for public comment, 
and incorporated the reviewers' recommendations in the final version. 
In the 2001 risk assessment, MSHA carefully laid out the best available 
evidence, including shortcomings inherent in that evidence.
    Of particular note is that the two quantitative meta-analyses of 
lung cancer studies supporting the 2001 risk assessment were peer 
reviewed and published in scientific journals. (Bhatia, Rajiv, et al., 
``Diesel Exhaust Exposure and Lung Cancer,'' Journal of Epidemiology, 
9:84-91, January 1998, and Lipsett M., and Campleman, Susan, 
``Occupational Exposure to Diesel Exhaust and Lung Cancer: A Meta-
Analysis,'' American Journal of Public Health, (89) 1009-1017, July 
1999).
    MSHA informed the public as early as September 25, 2002, in the 
2002 ANPRM for this final rule, and again in the 2003 NPRM, that MSHA 
would incorporate the existing rulemaking record, including the 2001 
risk assessment, into the record of this rulemaking. MSHA was open to 
considering any new scientific evidence relating to its risk 
assessment. Commenters were encouraged in the instant rulemaking to 
submit additional evidence of new scientific information related to 
health risks associated with exposure to DPM. After considering both 
the more recent scientific literature and all of the submitted 
comments, MSHA has concluded that no change is warranted in the 2001 
risk assessment's conclusions with respect to health risks associated 
with DPM exposures.
    Section VI.B updates Section 1 of the 2001 risk assessment by 
summarizing the new exposure data that became available after 
publication of the 2001 final rule. This summary includes a description 
of the relationship between EC and TC observed in these exposure 
measurements, and addresses public comments on possible health 
implications of substituting EC for TC as a surrogate measure of DPM. 
In Section VI.C, MSHA reviews some of the more recent scientific 
literature (April 2000-March 2003) pertaining to adverse health effects 
of DPM and fine particulates in general. In addition, this section 
updates the 2001 risk assessment's discussion of scientific evidence on 
mechanisms of DPM toxicity. Thus, Section VI.C supplements Section 2 of 
the 2001 risk assessment. Section VI.C also discusses a document by Dr. 
Gerald Chase that purports to analyze preliminary data extracted from 
an ongoing NIOSH/NCI study. Finally, in Section VI.D, MSHA assesses 
current risk to underground M/NM miners in light of the most recent 
exposure and health effects information. Section VI.D also responds to 
a critique of the 2001 risk assessment submitted by Dr. Jonathan Borak 
on behalf of the MARG Diesel Coalition (MARG) and the NMA.

B. DPM Exposures in Underground M/NM Mines

    In Section 1 of the 2001 risk assessment, MSHA evaluated exposures 
based on 355 samples collected at 27 underground U.S. M/NM mines prior 
to promulgating the 2001 rule. Mean DPM concentrations found in the 
production areas and haulageways at those mines ranged from about 285 
[mu]g/m\3\ to about 2000 [mu]g/m\3\, with some individual measurements 
exceeding 3500 [mu]g/m\3\. The overall mean DPM concentration was 808 
[mu]g/m\3\. All of the samples considered in the 2001 risk assessment 
were collected prior to 1999, and some were collected as long ago as 
1989.
    Two new bodies of DPM exposure data, collected after promulgation 
of the 2001 final rule, have now been compiled for underground M/NM 
mines: (1) Data collected in 2001 and 2002 from 31 mines for purposes 
of the 31-Mine Study and (2) data collected between 10/30/2002 and 10/
29/2003 from 183 mines to establish a baseline

[[Page 32890]]

for future samples. Key results from these two datasets are summarized 
in the next two subsections below. Following these summaries, the 
relationship between EC and TC, including the ratio of EC to TC (EC:TC) 
is discussed. This discussion is based exclusively on samples taken for 
the 31-Mine Study, since those samples were controlled for potential TC 
interferences from tobacco smoking and oil mist, whereas the baseline 
samples were not. The subsection concludes with a response to comments 
on the potential health effects of substituting EC for TC as a 
surrogate measure of DPM.
    It should be noted that the new exposure data reflect conditions at 
least two years, and up to five years, later than the most recent 
miners' exposure data considered in the 2001 risk assessment. 
Furthermore, all of the new exposure data were obtained after 
promulgation of the 2001 rule. It is, therefore, reasonable to expect 
that the data discussed below would show generally different exposure 
levels than those presented in the 2001 risk assessment--both on 
account of normal technological changes over time and because of DPM 
controls that may have been implemented in response to the 2001 rule.
(1) Data from 31-Mine Study
    MSHA collected 464 DPM samples in 2001 and 2002 at 31 underground 
M/NM mines. (For a more detailed description, see MSHA's final report 
on the 31-Mine Study.) Of these 464 samples, 106 were voided--mostly 
because of potential interference by sources of OC other than DPM. 
Table VI-1 shows how the remaining 358 valid DPM samples were 
distributed across four broad mine categories. All samples at one of 
the metal mines were voided, leaving 30 mines with valid samples 
indicating DPM concentrations.

                              Table VI-1.--Number of DPM Samples, by Mine Category
----------------------------------------------------------------------------------------------------------------
                                                          Number of mines                        Avg. number of
                                                             with valid      Number of valid   valid samples per
                                                              samples            samples              mine
----------------------------------------------------------------------------------------------------------------
Metal..................................................                 11                116               10.5
Stone..................................................                  9                105               11.7
Trona..................................................                  3                 54               18.0
Other..................................................                  7                 83               11.9
                                                        --------------------
    Total..............................................                 30                358               12.5
----------------------------------------------------------------------------------------------------------------

    Table VI-2 summarizes the valid DPM concentrations observed in each 
mine category, assuming that submicrometer TC, as measured by the SKC 
sampler, comprises 80% of all DPM. The mean concentration across all 
358 valid samples was 432 [mu]g/m\3\ (Std. error = 21.0 [mu]g/m\3\). 
The mean concentration was greatest at metal mines, followed by stone 
and ``other.'' At the three trona mines sampled, both the mean and 
median DPM concentration were substantially lower than what was 
observed for the other categories. This was due to the increased 
ventilation used at these mines to control methane emissions.

                          Table VI-2.--DPM Concentrations ([mu]/m\3\), By Mine Category
                                         [DPM Is Estimated by TC / 0.8]
----------------------------------------------------------------------------------------------------------------
                                                   Metal            Stone            Trona            Other
----------------------------------------------------------------------------------------------------------------
No. of samples..............................            116              105               54               83
Minimum.....................................             46.              16.              20.              27.
Maximum.....................................           2581.            1845.             331.            1210.
Median......................................            491.             331.              82.             341.
Mean........................................            610.             465.              94.             359.
    Std. Error..............................             44.7             36.0              9.4             26.6
    95% UCL.................................            699.             537.             113.             412.
    95% LCL.................................            522.             394.              75.             306.
----------------------------------------------------------------------------------------------------------------

    After adjusting for differences in sample types and in occupations 
sampled, DPM concentrations at the non-trona mines were estimated to be 
about four to five times the concentrations found at the trona mines. 
Although there were significant differences between individual mines, 
the adjusted differences between the general categories of metal, 
stone, and other mines were not statistically significant.\1\ For the 
304 valid samples taken at mines other than trona, the mean DPM 
concentration was 492 [mu]g/m\3\ (Std. error = 23.0 [mu]g/m\3\).
---------------------------------------------------------------------------

    \1\ These conclusions derive from an analysis of variance, based 
on TC measurements, described in the Report on the 31-Mine Study. 
They depend on an assumption that the ratio of DPM to TC is 
uncorrelated with mine category, sample type (i.e., personal or 
area), and occupation.
---------------------------------------------------------------------------

    Again assuming that submicrometer TC as measured by the SKC sampler 
comprises 80% of DPM, the mean DPM concentration observed was 1019 
[mu]g/m\3\ at the single mine exhibiting greatest DPM levels. Four of 
the nine valid samples at this mine exceeded 1487 [mu]g/m\3\. In 
contrast, DPM concentrations never exceeded 500 [mu]g/m\3\ at 8 of the 
30 mines with valid samples (2 of the 11 metal mines, 1 of the 3 stone, 
all 3 trona, and 2 of the 7 others). (Note that 500 [mu]g/m\3\ is the 
whole particulate equivalent of the 400TC [mu]g/m\3\ interim 
limit.) Some individual measurements exceeded 200DPM [mu]g/
m\3\ at all but one of the mines sampled.
(2) Baseline Data
    MSHA s baseline sampling results are presented in Section III, 
Compliance Assistance. These results provide the basis for the present 
discussion. The baseline samples discussed here, in connection with the 
risk assessment, were collected and analyzed between

[[Page 32891]]

October 30, 2002 and October 29, 2003. They comprise a total of 1,194 
valid samples collected from 183 mines. MSHA is including 320 
additional valid samples because MSHA decided to continue to conduct 
baseline sampling after July 19, 2003 in response to mine operator's 
concerns. Some of these mines were either not in operation or were 
implementing major changes to ventilation systems during the original 
baseline period. MSHA is including supplementary samples from seasonal 
and intermittent mines, mines that were under-represented, and mines 
that were not represented in the analysis published in the proposed 
preamble in 2003.
    Table VI-3 summarizes, by general commodity, the EC levels measured 
during MSHA's baseline sampling through October 29, 2003. The overall 
mean eight-hour full shift equivalent EC concentration was 196 [mu]g/
m\3\, and the overall median was 134 [mu]g/m\3\. Table VI-4 provides a 
similar summary for estimated DPM levels, using DPM [ap] TC/0.8 and TC 
[ap] 1.3 x EC.\2\ Under these assumptions, the estimated mean DPM level 
was 318 [mu]g/m\3\, and the median was 218 [mu]g/m\3\. Since the 
baseline data and the 31-Mine Study both showed significantly lower 
levels at trona mines than at other underground M/NM mines, Tables VI-3 
and VI-4 present overall results both including and excluding the three 
underground trona mines sampled.\3\
---------------------------------------------------------------------------

    \2\ The relationship DPM [ap] TC/0.8 is the same as that assumed 
in the 2001 risk assessment. The relationship TC 1.3 x EC was 
formulated under the settlement agreement, based on TC:EC ratios 
observed in the joint 31-Mine Study, as described in the subsection 
VI.3 of this preamble.
    \3\ The distributions of EC values are skewed. Therefore, the 
standard errors and confidence intervals reported in Tables VI-3 and 
VI-4 should be interpreted with caution.

                                     Table VI-3.--Baseline EC Concentrations
----------------------------------------------------------------------------------------------------------------
                                                   8-hour Full Shift Equivalent EC Concentration ([mu]g/m\3\ )
                                               -----------------------------------------------------------------
                                                                                                         Total
                                                  Metal      Stone    Other  N/    Trona      Total    excluding
                                                                          M                              Trona
----------------------------------------------------------------------------------------------------------------
No. of Samples................................        284        689        196         25      1,194      1,169
Maximum.......................................      1,558      2,291        738        313      2,291      2,291
Median........................................        208        115        114         63        134        137
Mean..........................................        273        181        150         81        196        198
    Std. Error................................         14          8          9         12          6          6
    95% UCL...................................        302        197        167        106        208        210
    95% LCL...................................        245        166        132         56        184        186
----------------------------------------------------------------------------------------------------------------


                                    Table VI-4.--Baseline DPM Concentrations
                                     [DPM is estimated by (1.3 x EC) / 0.8]
----------------------------------------------------------------------------------------------------------------
                                                Estimated 8-hour Full Shift Equivalent DPM Concentration ([mu]g/
                                                                             m\3\ )
                                               -----------------------------------------------------------------
                                                                                                         Total
                                                  Metal      Stone    Other N/M    Trona      Total    excluding
                                                                                                         Trona
----------------------------------------------------------------------------------------------------------------
No. of Samples................................        284        689        196         25      1,194      1,169
Maximum.......................................      2,532      3,724      1,200        509      3,724      3,724
Median........................................        339        186        185        102        218        223
Mean..........................................        444        295        243        132        318        322
    Std. Error................................         23         13         15         20         10         10
    95% UCL...................................        490        320        272        173        338        342
    95% LCL...................................        399        270        214         91        299        303
----------------------------------------------------------------------------------------------------------------

    Baseline EC sample results varied widely between mines within 
commodities and also within most mines. Table VI-5 summarizes baseline 
EC results for the 26 occupations found to have at least one sample 
where the EC level exceeded the 308 [mu]g/m3 8-hour full 
shift equivalent interim EC limit. As indicated by the table, EC levels 
varied widely within each occupation.

   Table VI-5.--Baseline EC Concentrations for Occupations With at Least One Value Exceeding Interim EC Limit
----------------------------------------------------------------------------------------------------------------
                                                                              8-hour full shift equivalent EC
                                                               Number of        Concentration ([mu]g/m\3\ )
                         Occupation                              valid    --------------------------------------
                                                                samples      Minimum       Median      Maximum
----------------------------------------------------------------------------------------------------------------
Diamond Drill Operator......................................            1        1,561        1,561        1,561
Ground Control/Timberman....................................            2          283          419          555
Washer Operator.............................................            4          272          337          621
Engineer....................................................            1          337          337          337
Roof Bolter, Mounted........................................           12           76          258          818
Mucking Mach. Operator......................................           23           12          257          671
Miner, Stope................................................           14           77          218          479

[[Page 32892]]


Cleanup Man.................................................            2           51          217          384
Scoop-Tram Operator.........................................            7           10          210          449
Drill Operator, Rotary Air..................................           21            0          185        1,041
Miner, Drift................................................           17           12          175        1,122
Blaster, Powder Gang........................................          134            5          175        1,031
Belt Crew...................................................            8           20          173          386
Roof Bolter, Rock...........................................           21           48          172        1,007
Truck Driver................................................          252            0          162        1,216
Shuttle Car Operator (diesel)...............................            3           73          154          323
Complete Load-Haul-Dump.....................................           32           14          145          634
Drill Operator, Jumbo Perc..................................           38            4          137          845
Drill Operator, Rotary......................................           75            2          132          853
Motorman....................................................            8           46          129          322
Front-end Loader Operator...................................          214            0          121        2,291
Scaling (mechanical)........................................           80            0          107          958
Supervisor, Co. Official....................................           13            1          100          658
Utility Man.................................................           29           22           73          762
Scaling (hand)..............................................           26           14           67        1,548
Mechanic....................................................           34            0           64          323
----------------------------------------------------------------------------------------------------------------

    Figure VI-1 depicts, by mine category, the percentage of baseline 
samples that exceeded the interim EC limit of 308 [mu]g/m\3\. 
Underground metal mines exhibited the highest proportion of samples 
exceeding this limit, followed by stone and then other nonmetal mines. 
In the three trona mines sampled, 24 of the 25 samples were lower than 
the proposed limit. Across all commodities, 19.3% of the 1,194 valid 
baseline samples exceeded the interim EC limit.
BILLING CODE 4510-43-U

[[Page 32893]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.007

    Figure VI-2 shows how samples exceeding the interim EC limit were 
distributed over individual mines. One to 20 baseline samples were 
taken at each mine. In 115 of the 183 mines sampled (63%), none of the 
baseline EC measurements exceeded 308 [mu]g/m\3\. The remaining 68 
mines (37%) had at least one sample for which EC exceeded 308 [mu]g/
m\3\. All samples taken at 4 of the mines exceeded the interim limit.

[[Page 32894]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.008

BILLING CODE 4510-43-C
(3) Relationship Between EC and TC
    The 2001 final rule stipulated that TC (i.e., EC + OC) measurements 
would be used to monitor and limit DPM concentration levels. Although 
it was recognized that TC measurements were subject to various 
interferences from non-DPM sources, MSHA believed that, in underground 
metal and nonmetal mines, it could effectively eliminate such 
interferences by a combination of selective sampling procedures and 
careful analytical techniques. During the 31-Mine Study, however, MSHA 
found no reasonable sampling method that would adequately protect TC 
measurements from interference by such sources of organic carbon as oil 
mist and ammonium nitrate fuel oil (ANFO). Furthermore, MSHA found that 
it was cumbersome and impractical to restrict its TC sampling so as to 
avoid potential interference from environmental tobacco smoke (ETS). 
Indeed, as indicated earlier, nearly one fourth of the TC samples 
collected during the 31-Mine Study (106 out of 464) had to be voided on 
account of potential interferences from extraneous sources of OC. 
Therefore, in concert with the Second Partial Settlement Agreement, the 
2003 NPRM proposed to ``[r]evise the existing diesel particulate matter 
(DPM) interim concentration limit measured by total carbon (TC) to a 
comparable permissible exposure limit (PEL) measured by elemental 
carbon (EC) which renders a more accurate DPM exposure measurement.'' 
(68 FR 48668) Using EC as the surrogate permits direct sampling of 
miners (such as those who smoke, operate jackleg drills, or load ANFO) 
for whom accurate DPM monitoring would be difficult or impossible using 
TC measurements.
    Also in accordance with the Second Partial Settlement Agreement, 
the NPRM proposed to convert the existing interim exposure limit, 
expressed in terms of TC measurements, to a ``comparable'' EC limit by 
applying a specific conversion factor obtained from data gathered 
during the 31-Mine Study, as explained below. MSHA is adopting this 
proposal with the intention of providing at least the same degree of 
protection to miners as the existing interim limit. However, since it 
is unlikely that EC and OC have identical health effects, it is 
important to consider the extent to which the ratio of EC to OC (and 
hence of EC to TC) may vary in different underground mining 
environments.
    Unlike the 31-Mine Study, no special precautions were taken during 
MSHA's baseline sampling to avoid ETS or other substances that could 
potentially interfere with using TC as a surrogate measure of DPM. 
Therefore, the baseline data should not be used to evaluate the OC 
content of DPM or the ratio of EC to TC within DPM. In the 31-Mine 
Study, on the other hand, great care was taken to void all samples that 
may have been exposed to ETS or other extraneous sources of OC.
    Consequently, the analysis of the EC:TC ratio presented here relies 
entirely on data from the 31-Mine Study. It is important to note that 
nearly all of the samples in this study were taken in the absence of 
exhaust filters to

[[Page 32895]]

control DPM emissions. Since exhaust filters may have different effects 
on EC and OC emissions, the results described here apply only to mine 
areas where exhaust filters are not employed.
    Figure VI-3 plots the EC:TC ratios observed in the 31-Mine Study 
against the corresponding TC concentrations. The various symbols shown 
in the plot identify samples taken at the same mine. The EC:TC ratio 
ranged from 23% to 100%, with a mean of 75.7% and a median of 78.2%. 
Note that the reciprocal of 0.78, which is 1.3, equals the median of 
the TC:EC ratio observed in these samples.\4\ The 1.3 TC:EC ratio was 
the value accepted, under terms of the settlement agreement, for the 
purpose of temporarily converting EC measurements to TC measurements.
---------------------------------------------------------------------------

    \4\ The median of reciprocal values is always equal to the 
reciprocal of the median. This relationship does not hold for the 
mean.
---------------------------------------------------------------------------

BILLING CODE 4510-43-U
[GRAPHIC] [TIFF OMITTED] TR06JN05.009

BILLING CODE 4510-43-C
    The 2001 rule set a TC interim concentration limit of 400 [mu]g/
m3. Under the new rule, this TC interim limit is replaced 
with an EC interim limit of 400/1.3 = 308 [mu]g/m3. Table 
VI-6 indicates the impact of this change, based on the EC and TC data 
obtained from the 31-Mine Study. Both the original 400 [mu]g/
m3 TC limit and the new 308 [mu]g/m3 EC limit 
were exceeded by about 31% to 32% of the samples. The difference (one 
sample out of 358) is not statistically significant in the aggregate. 
Seven samples, however, exceeded the TC limit but not the EC limit, and 
six samples exceeded the EC limit but not the TC limit.

[[Page 32896]]



    Table VI-6.--Compliance With Original 400 [mu]g/m3 TC Limit and/or New 308 [mu]g/m3 EC Limit. Numbers in
                                           Parentheses Are Percentages
----------------------------------------------------------------------------------------------------------------
                                                                   TC > 400 [mu]g/m3
                   EC > 308 [mu]g/m3                    --------------------------------------       Total
                                                                 No                Yes
----------------------------------------------------------------------------------------------------------------
No.....................................................         239 (66.8)            7 (2.0)         246 (68.7)
Yes....................................................            6 (1.7)         106 (29.6)         112 (31.3)
                                                        --------------------
    Total..............................................         245 (68.4)         113 (31.6)        358 (100.0)
----------------------------------------------------------------------------------------------------------------

    Several commenters noted that the ratio of EC to TC in DPM can vary 
widely. One commenter pointed out that EC appeared to make up nearly 
all of the TC at the mine with which he was affiliated. This commenter 
stated that replacing a 400 [mu]g/m3 TC limit with a 308 
[mu]g/m3 EC limit would impose a much more stringent 
standard at that mine. Another commenter noted that a 308 [mu]g/
m3 EC limit would be less protective of miners than the 400 
[mu]g/m3 TC limit in cases where the ratio of EC comprised 
less than 78% of the TC. MARG submitted comments by a consultant, Dr. 
Jonathan Borak, who emphasized that the highly variable nature of the 
EC to OC ratio introduces ``large and important uncertainties in the 
exposure assessments needed to sustain QRA [i.e., quantitative risk 
assessment].''
    As indicated by Figure VI-3, the percentage of EC tended to 
increase with increasing TC concentration--except for cases showing a 
TC concentration of less than about 60 [mu]g/m3. In many of 
the samples for which TC < 60 [mu]g/m3, the recorded ratio 
of EC to TC was at or near 100%. Since TC concentrations less than 60 
[mu]g/m3 appear to deviate from the general pattern and are 
far below the interim limit, our response to commenters concerns about 
variability in the ratio of EC to TC will focus on those samples for 
which TC exceeds 60 [mu]g/m3.
    There were 319 samples with TC > 60 [mu]g/m3. For these 
samples, the mean and median EC:TC ratio were 76.3% and 78.4%, 
respectively. In accordance with standard statistical practice, an 
arcsine transformation was applied to these 319 EC:TC ratios in order 
to normalize them for further statistical analysis (Snedecor and 
Cochran, Statistical Methods, 7th Ed., pp 290-291). The transformed 
EC:TC ratios are plotted against corresponding TC concentrations in 
Figure VI-4. Various symbols are used to identify the mineral commodity 
corresponding to each sample.

[[Page 32897]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.010

BILLING CODE 4510-43-C
    It is clear from Figures VI-3 and VI-4 that individual samples in 
the 31-Mine Study exhibited considerable variation in their EC:TC 
ratios. What is not so clear from these plots, however, is whether 
different mines and/or working environments tended to experience 
different EC:TC ratios. To answer this question, an analysis of 
variance (ANOVA) was performed to determine whether there were 
statistically significant differences in the EC:TC ratios exhibited at 
different mines and on different days at the same mine. Table VI-7 
contains the results of this ANOVA. At a confidence level exceeding 
99.9%, the data show statistically significant differences in the mean 
EC:TC ratios between mines and between different sampling days within 
mines.

    Table VI-7.--Analysis of Variance for Arcsin of EC:TC Ratios, Restricted to Samples With TC > 60 [mu]g/m3
----------------------------------------------------------------------------------------------------------------
                                                                       Degrees
                          Source                             Sum of       of        Mean     F-ratio       P
                                                            squares    freedom     square
----------------------------------------------------------------------------------------------------------------
MINE.....................................................      3.360         29      0.116      6.960      0.000
DAY within MINE..........................................      1.643         30      0.055      3.290      0.000
Error....................................................      4.295        258      0.017  .........  .........
----------------------------------------------------------------------------------------------------------------


[[Page 32898]]

    Figure VI-5 illustrates the magnitude and extent of differences in 
the mean EC:TC ratio between mines. Note that values on the arcsin 
scale of 0.7, 0.9, and 1.1 correspond to EC:TC ratios of 64%, 78%, and 
89%, respectively.
    Since TC = EC + OC, variability in the EC:TC ratio corresponds to 
variability in the ratio of either EC or TC to OC. Dr. Borak stated 
that if DPM is carcinogenic, then the carcinogenic agents (for humans) 
are probably in the organic fraction (i.e., OC). Consequently, 
according to Dr. Borak, neither EC nor TC provides an appropriate 
surrogate for assessing or limiting health risks.
    MSHA believes that Dr. Borak's assumption that any carcinogenic 
effect of DPM is due entirely to the organic fraction is speculative. 
This assumption contradicts findings reported by Ichinose et al. 
(1997b) and does not take into account the contribution that 
inflammation and active oxygen radicals induced by the inorganic carbon 
core of DPM may have in promoting lung cancers. Indeed, identifying the 
toxic components of DPM, and particulate matter in general, is an 
important research focus of a variety of government agencies and 
scientific organizations (see, for example: Health Effects Institute, 
2003; Environmental Protection Agency, 2004b). The 2001 risk assessment 
discusses possible mechanisms of carcinogenesis for which both EC and 
OC would be relevant factors (66 FR at 5811-5822). Multiple routes of 
carcinogenesis may operate in human lungs--some requiring only the 
various organic mutagens in DPM and others involving induction of free 
radicals by the EC core, either alone or in combination with the 
organics.

[[Page 32899]]

[GRAPHIC] [TIFF OMITTED] TR06JN05.011

BILLING CODE 4510-43-C
    In focusing on the carcinogenic agents in OC, Dr. Borak has also 
ignored non-cancer health effects documented in the 2001 risk 
assessment--e.g., immunological, inflammatory, and allergenic responses 
in healthy human volunteers exposed to 300DPM [mu]g/
m3 (i.e.,  240TC [mu]g/m3) for as 
little as one hour (66 FR at 5769-70, 5816-17, 5820, 5823, 5837, 5841, 
5847).
    The 308 [mu]g/m3 interim EC PEL established by this rule 
is intended to be commensurate with the interim TC limit of 400 [mu]g/
m3 established under the 2001 rule--i.e., to be equally 
protective and equally feasible. Although, as shown by Table VI-7 and 
Figure VI-5, the EC:TC ratio can exhibit considerable variability in 
specific cases, MSHA has concluded that application of the 1.3 average 
conversion factor, as suggested in the second partial settlement 
agreement, generally achieves the goal of equal protection and 
feasibility.

C. Health Effects

    A key conclusion of the 2001 risk assessment was:

Exposure to DPM can materially impair miner health or functional 
capacity. These material impairments include acute sensory 
irritations and respiratory symptoms (including allergenic 
responses); premature death from cardiovascular, cardiopulmonary, or 
respiratory causes; and lung cancer. [66 FR at 5854-5855]


[[Page 32900]]


MSHA has reviewed the scientific literature pertaining to health 
effects of fine particulates in general and DPM in particular published 
later than what was considered in the 2001 risk assessment. As will be 
shown below, the more recent scientific evidence generally supports the 
conclusion above, and nothing in our review suggests that it should be 
altered. In fact, the U.S. Environmental Protection Agency (EPA) 
recently reached very similar conclusions after reviewing all of the 
evidence to date (EPA; 2002, 2004b).
    Some commenters endorsed the 2001 risk assessment, and suggested 
that the latest evidence strengthens its conclusions. For example, one 
group of commenters jointly stated:

The evidence presented in MSHA's 2001 risk assessment is 
overwhelming * * * The evidence linking exposure to particulate air 
pollution and/or diesel particulate matter with lung cancer, 
cardiovascular and cardiopulmonary and other adverse health effects 
continues to mount.

    Similarly, another pair of commenters jointly stated that ``[t]he 
scientific evidence for the [adverse] health effects of DPM is 
overwhelming'' and that ``evidence for the carcinogenicity and non-
cancer health effects of DPM has grown since 1998.''
    Other commenters contended that all of the evidence to date is 
insufficient to support limitation of occupational DPM exposures. 
Several of these commenters ignored evidence presented in the 2001 risk 
assessment and/or mischaracterized its conclusions. For example, the 
NMA, MARG, and the Nevada Mining Association (NVMA) all erroneously 
stated that promulgation of the 2001 rule was based on only ``two 
principal health concerns: (1) The transitory, reversible health 
effects of exposure to DPM; and, (2) the long-term impacts that may 
result in an excess risk of lung cancer for exposed workers.'' 
Actually, as shown in the conclusion cited above, the 2001 risk 
assessment identified three different kinds of material health 
impairment associated with DPM exposure: (1) Acute sensory irritations 
and respiratory symptoms (including allergenic responses); (2) 
premature death from cardiovascular, cardiopulmonary, or respiratory 
causes; and (3) lung cancer. Although the cardiovascular, 
cardiopulmonary, and respiratory effects leading to an increased risk 
of premature death were associated with acute DPM exposures, commenters 
presented no evidence that any such effects were ``transitory'' or 
``reversible.'' Nor did commenters present evidence that immunological 
responses associated with either short-term or long-term DPM exposure 
were ``transitory'' or ``reversible.''
    In addition, some commenters erroneously stated that ``no 
[quantitative] dose/response relationship related to the PELs could be 
demonstrated by MSHA.'' These commenters apparently ignored the 
discussion of exposure-response relationships in the 2001 risk 
assessment (66 FR at 5847-54) and failed, specifically, to note the 
quantitative exposure-response relationships shown for lung cancer in 
the two tables provided (66 FR at 5852-53). Relevant exposure-response 
relationships were also demonstrated in articles by Pope et al. cited 
in the 2003 NPRM, which will be discussed further below.
    Some commenters objected that the exposure-response relationships 
presented in the 2001 risk assessment did not justify adoption of the 
specific DPM exposure limits promulgated. These commenters mistakenly 
assumed the limits set forth in the 2001 final rule were derived from 
an exposure-response relationship. As explained in 66 FR at 5710-14, 
the choice of exposure limits, while justified by quantifiable adverse 
health effects, was actually driven by feasibility concerns. The 
exposure-response relationships provided clear evidence of adverse 
human health effects (both cancer and non-cancer) at levels far below 
those determined to be feasible for mining.
    In the 2003 NPRM, MSHA identified scientific literature pertaining 
to health effects of fine particulates in general and DPM in particular 
published subsequent to the 2001 final rule. The 2003 NPRM stated 
MSHA's intentions to continue its reliance on the 2001 risk assessment 
and cited the newer literature in a neutral manner, soliciting public 
comment on its implications for the 2001 risk assessment.
    Two commenters complained that MSHA had not described the recent 
scientific literature in sufficient detail to determine whether it 
supports the 2001 risk assessment. Most of the commenters who evaluated 
the recent literature found that it supported and/or strengthened the 
conclusions of the 2001 risk assessment. Some other commenters, 
however, disagreed. Accordingly MSHA will present the supplemental 
literature in more detail than in the 2003 NPRM and explain why MSHA 
believes that it continues to support the 2001 risk assessment. This 
discussion will include our review of an analysis by Dr. Gerald Chase 
of some preliminary data from an ongoing NIOSH/NCI study.
    The scientific literature cited in the 2003 NPRM was meant only to 
update and supplement the evidence of health effects cited in the 2001 
risk assessment. Although MSHA believes the 2001 risk assessment 
presented ample evidence to justify its conclusions, MSHA is adding 
this supplemental literature because it represents more recent 
scientific investigations related to DPM health effects. The following 
discussion of literature cited in the 2003 NPRM is organized into four 
categories, roughly corresponding to the three types of material health 
impairments identified in the 2001 risk assessment, followed by a 
category covering toxicology studies: (1) Respiratory and immunological 
effects, including asthma, (2) cardiovascular and cardiopulmonary 
effects, (3) cancer, and (4) mechanisms of toxicity. Although the 
discussion of cancer will focus on lung cancer, it will also take note 
of two recent meta-analyses of epidemiological studies investigating 
DPM in connection with bladder and pancreatic cancers.
(1) Respiratory and Immunological Effects, Including Allergenic 
Responses
    In the 2001 risk assessment, acute sensory irritations with 
respiratory symptoms, including immunological or allergenic effects 
such as asthmatic responses were grouped together, and all such effects 
as material health impairments likely to be caused or exacerbated by 
excessive DPM exposures were identified. This finding was based on 
human experimental and epidemiological studies and was supported by 
experimental toxicology. (For an explanation of why MSHA considers such 
effects to be material impairments, regardless of whether they are 
``reversible,'' See, 66 FR at 5766.)
    Table VI-8 summarizes six additional studies dealing with possible 
respiratory and immunological effects of DPM and/or fine particulates 
in general. Three of these studies (Frew et al., 2001; Holgate et al., 
2002; Salvi et al., 2000) involved experiments in which human subjects 
inhaled specified doses of DPM. These three studies all support the 
view that occupational DPM exposures are likely to promote or 
exacerbate adverse respiratory symptoms and immunological responses. A 
fourth study (Svartengren et al., 2000) exposed human subjects to high 
and low doses of an unspecified mix of diesel and gasoline engine 
exhausts. Although 30-minute PM2.5 exposures greater than 
100 [mu]g/m3 were found to increase asthmatic response, the authors of 
this study attributed the effects they observed primarily to 
NO2 exposure. The fifth study (Oliver et al., 2001) 
attempted to

[[Page 32901]]

relate pulmonary function test results and asthmatic conditions to 
estimated lifetime diesel exposure in a cohort of 359 ``heavy and 
highway'' (HH) construction workers. After adjustment for smoking and 
other potential confounders, the results indicated an elevated risk of 
asthma for exposed workers in enclosed spaces (tunnel workers), 
relative to other HH workers. The lack of additional statistically 
significant results may be attributable to the small cohort size. The 
sixth study (Fusco et al., 2001) examined the relationship between 
various markers of engine exhaust pollution levels and daily hospital 
admissions for acute respiratory infections, COPD, asthma, and total 
respiratory conditions in Rome, Italy. No direct measurements of fine 
particulate concentrations were available. However, having found a 
significant correlation between respiratory-related admissions and CO 
and NO2 levels, the authors noted that since CO and 
NO2 are good indicators of combustion products in vehicular 
exhaust, the detected effects may be due to unmeasured fine and 
ultrafine particles.

  Table VI-8.--Studies of Human Respiratory and Immunological Effects,
                                2000-2002
------------------------------------------------------------------------
        Authors, year              Description           Key results
------------------------------------------------------------------------
Frew et al., 2001...........  25 healthy subjects   Both the asthmatic
                               and 15 subjects       and healthy
                               with mild asthma      subjects developed
                               were exposed to       increased airway
                               diesel exhaust (108   resistance after
                               [mu]g/m3) or          exposure to diesel
                               filtered air for 2    emissions, but
                               hr, with              airway inflammatory
                               intermittent          responses were
                               exercise. Lung        different for the 2
                               function was          groups. The healthy
                               assessed using a      subjects showed
                               computerized whole    statistically
                               body                  significant BW
                               plethysmograph.       neutrophilia and
                               Airway responses      BAL lymphocytosis 6
                               were sampled by       hr after exposure.
                               bronchial wash        The neutrophilic
                               (BW),                 response of the
                               bronchoalveolar       healthy subjects
                               lavage (BAL), and     was less intense
                               mucosal biopsies 6    than that seen in a
                               hr after ceasing      previous study
                               exposures.            using a DPM
                                                     concentration of
                                                     300 [mu]g/m3.
Fusco et al., 2001..........  Analysis of daily     Respiratory
                               hospital admissions   admissions among
                               for acute             adults were
                               respiratory           significantly
                               infections, COPD,     correlated with CO
                               asthma, and total     and NO2 levels, but
                               respiratory           not with suspended
                               conditions in Rome,   particles. The
                               Italy.                authors noted that
                                                     since CO and NO2
                                                     are good indicators
                                                     of combustion
                                                     products in
                                                     vehicular exhaust,
                                                     the detected
                                                     effects may be due
                                                     to unmeasured fine
                                                     and ultrafine
                                                     particles.
Holgate et al. 2002.........  25 healthy and 15     Healthy and
                               asthmatic subjects    asthmatic subjects
                               were exposed for 2    exhibited evidence
                               hours to 100 [mu]g/   of
                               m3 of DPM and to      bronchioconstrictio
                               filtered air on       n immediately after
                               separate days.        exposure
                               Another 30 healthy   Biochemical tests of
                               subjects were         inflammation
                               exposed for 2 hours   yielded mixed
                               to DPM                results but showed
                               concentrations        small inflammatory
                               ranging from 25 to    changes in healthy
                               311 [mu]g/m3 and      subjects after DPM
                               compared to 12        inhalation.
                               different healthy
                               subjects exposed to
                               filtered air.
                               Exposure effects
                               were assessed using
                               lung function tests
                               and biochemical
                               tests of bronchial
                               tissue samples.
Oliver et al., 2001.........  Pulmonary function    After adjusting for
                               tests and             smoking and some
                               questionnaire data    other potential
                               were obtained for     confounders, HH
                               350 ``heavy and       workers showed
                               highway'' (HH)        elevated risk of
                               construction          asthma. One
                               workers. Intensity    subgroup (tunnel
                               of DPM exposure was   workers) also
                               estimated according   showed elevated
                               to job                risk of both
                               classification.       undiagnosed asthma
                               Duration of           and chronic
                               exposure was          bronchitis,
                               estimated based on    compared to other
                               length of union       HH workers.
                               membership.          Respiratory symptoms
                                                     appeared to
                                                     declined with
                                                     exposure duration
                                                     as measured length
                                                     of union
                                                     membership. The
                                                     authors interpreted
                                                     this as suggesting
                                                     that HH workers
                                                     tend to leave their
                                                     trade when they
                                                     experience adverse
                                                     respiratory
                                                     symptoms.
Salvi et al., 2000..........  15 healthy            Diesel exhaust
                               nonsmoking            exposure enhanced
                               volunteers were       gene transcription
                               exposed to 300        of IL-8 in the
                               [mu]g/m3 DPM and      bronchial tissue
                               clean air for one     and airway cells
                               hour at least three   and increased IL-8
                               weeks apart.          and GRO-[alpha]
                               Biochemical           protein expression
                               analyses were         in the bronchial
                               performed on          epithelium. This
                               bronchial tissue      was accompanied by
                               and bronchial wash    a trend toward
                               cells obtained six    increased IL-5 mRNA
                               hours after each      gene transcripts in
                               exposure.             the bronchial
                                                     tissue. Study
                                                     showed effects on
                                                     chemokine and
                                                     cytokine production
                                                     in the lower
                                                     airways of healthy
                                                     adults. These
                                                     substances attract
                                                     and activate
                                                     leukocytes. They
                                                     are associated with
                                                     the pathophysiology
                                                     of asthma and
                                                     allergic rhintisi.
Svartengren et al;. 2000....  Twenty nonsmoking     Subjects with PM2.5
                               subjects with mild    exposure >= 100
                               allergic asthma       [mu]g/m3 exhibited
                               were exposed for 30   slightly increased
                               minutes to high and   asthmatic
                               low levels of         responses.
                               engine exhaust air   Association with
                               pollution on two      adverse outcome
                               separate occasions    variables were
                               at least four weeks   weaker for
                               apart. Respiratory    particulates than
                               symptoms and          for NO2.
                               pulmonary function
                               were measured
                               immediately before,
                               during and after
                               both exposure
                               periods. Four hours
                               after each
                               exposure, the test
                               subjects were
                               challenged with a
                               low dose of inhaled
                               allergen. Lung
                               function and
                               asthmatic reactions
                               were monitored for
                               several hours after
                               exposure.
------------------------------------------------------------------------

    The 2003 NPRM also cited five new review articles that summarize 
the scientific literature pertaining to the respiratory and 
immunological effects of DPM and fine particulate matter in general. 
These review articles, published after the 2001 risk assessment, are 
identified and briefly described in Table VI-9. The three

[[Page 32902]]

articles most specifically dealing with DPM effects are Pandya et al. 
(2002), Peden at al. (2002), and Sydbom et al. (2001). In general, 
these reviews indicate that while DPM is likely to contribute to 
asthmatic and/or other immunological responses, the role of DPM in 
producing these health effects is complex. As noted by Pandya et al. 
(op cit.), DPM may have a far greater impact as an adjuvant with 
allergens than alone. Nevertheless, all three of these review articles 
support the view that there is significant evidence of adverse 
respiratory and immunological effects to warrant regulating DPM 
exposures. The remaining review articles (Gavett and Koren, 2001; 
Patton and Lopez, 2002) offer little new support for the 2001 risk 
assessment, but MSHA found no studies that either refute or challenge 
the 2001 risk assessment.

 Table VI-9.--Review Articles on Respiratory and Immunological Effects,
                                1999-2002
------------------------------------------------------------------------
        Authors, year              Description           Key results
------------------------------------------------------------------------
Gavett and Koren, 2001......  Summarizes results    Studies indicate
                               of EPA studies done   that PM enhances
                               to determine          allergic
                               whether PM can        sensitization in
                               enhance allergic      animal models of
                               sensitization or      allergy exacerbate
                               exacerbate existing   inflammation and
                               asthma or asthma-     airway hyper-
                               like responses in     responsiveness in
                               humans and animal     asthmatics and
                               models.               animal models of
                                                     asthma.
Pandya et al. 2002..........  Reviews human and     Evidence indicates
                               animal research       that DPM is
                               relevant to           associated with the
                               question of whether   inflammatory and
                               DPM is associated     immune responses
                               with asthma.          involved in asthma,
                                                     but DPM appears to
                                                     have far greater
                                                     impact as an
                                                     adjuvant with
                                                     allergens than
                                                     alone.
                                                    DPM appears to
                                                     augment IgE,
                                                     trigger eosinophil
                                                     degranulation, and
                                                     stimulate release
                                                     of numerous
                                                     cytokines and
                                                     chemokines. DPM may
                                                     also promote the
                                                     cytotoxic effects
                                                     of free radicals in
                                                     the airways.
Patton and Lopez, 2002......  Review of evidence    Evidence suggests
                               and mechanisms for    that air pollutants
                               the role of air       (including DPM)
                               pollutants in         ``affect allergic
                               allergic airways      response by
                               disease.              different
                                                     mechanisms.
                                                     Pollutants may
                                                     increase total IgE
                                                     levels and
                                                     potentiate the
                                                     initial
                                                     sensitization to
                                                     allergens and the
                                                     IgE response to a
                                                     subsequent allergen
                                                     exposure.
                                                     Pollutants also may
                                                     act by increasing
                                                     allergic airway
                                                     inflammation and by
                                                     directly
                                                     stimulating airway
                                                     inflammation. In
                                                     addition, it is
                                                     well known that
                                                     pollutants can be
                                                     direct irritants of
                                                     the airways,
                                                     increasing symptoms
                                                     in patients with
                                                     allergic
                                                     syndromes.''
Peden, 2002.................  Review of ``studies   DPM ``may play a
                               that exemplify the    significant role
                               impact of ozone,      not only in asthma
                               particulates, and     exacerbation but
                               toxic components of   also in TH2
                               particulates on       inflammation via
                               asthma.''.            the actions of
                                                     polyaromatic
                                                     hydrocarbons on B
                                                     lymphocytes.''
                                                    ``* * * PM in which
                                                     the active agents
                                                     are biologically
                                                     active metal ions
                                                     and organic
                                                     residues * * * may
                                                     have significant
                                                     effects on asthma,
                                                     especially
                                                     modulating immune
                                                     function, as
                                                     demonstrated by the
                                                     role of
                                                     polyaromatic
                                                     hydrocarbons from
                                                     diesel exhaust in
                                                     IgE production.''
Sydbom et al. 2001..........  Review of scientific  The epidemiological
                               literature on         support for
                               health effects of     particle effects on
                               disease exhaust,      asthma and
                               especially the DPM    respiratory health
                               components.           is very evident;
                                                     and respiratory,
                                                     immunological, and
                                                     systemic effects of
                                                     DPM have been
                                                     documented in a
                                                     wide variety of
                                                     experimental
                                                     studies.
                                                    Acute effects of DPM
                                                     exposure include
                                                     irritation of the
                                                     nose and eyes, lung
                                                     function changes,
                                                     and airway
                                                     inflammation.
                                                    Exposure studies in
                                                     healthy humans have
                                                     documented a number
                                                     of profound
                                                     inflammatory
                                                     changes in the
                                                     airways, notably,
                                                     before changes in
                                                     pulmonary function
                                                     can be detected.
                                                     Such effects may be
                                                     even more
                                                     detrimental in
                                                     subjects with
                                                     compromised
                                                     pulmonary function.
                                                    Ultrafine particles
                                                     are currently
                                                     suspected of being
                                                     the most aggressive
                                                     particulate
                                                     component of diesel
                                                     exhaust.
------------------------------------------------------------------------

    In its 2002 ``Health Assessment Document for Diesel Engine 
Exhaust,'' the Environmental Protection Agency (EPA) reached the 
following conclusion with respect to immunological effects of diesel 
exhaust:

Recent human and animal studies show that acute DE [diesel exhaust] 
exposure episodes can exacerbate immunological reactions to other 
allergens or initiate a DE-specific allergenic reaction. The effects 
seem to be associated with both the organic and carbon core fraction 
of DPM. In human subjects, intranasal administration of DPM has 
resulted in measurable increases of IgE antibody production and 
increased nasal mRNA for some proinflammatory cytokines. These types 
of responses also are markers typical of asthma, though for DE, 
evidence has not been produced in humans that DE exposure results in 
asthma. The ability of DPM to act as an adjuvant to other allergens 
also has been demonstrated in human subjects. (EPA, 2002)


[[Page 32903]]


(2) Cardiovascular and Cardiopulmonary Effects
    In the 2001 risk assessment, the evidence presented for DPM's 
adverse cardiovascular and cardiopulmonary effects relied on data from 
air pollution studies in the ambient air. This evidence identifies 
premature death from cardiovascular, cardiopulmonary, or respiratory 
causes as an endpoint significantly associated with exposures to fine 
particulates. The 2001 risk assessment found that ``[t]he mortality 
effects of acute exposures appear to be primarily attributable to 
combustion-related particles in PM2.5 [i.e., fine 
Particulate Matter] (such as DPM) * * *.''
    There are difficulties involved in utilizing the evidence from such 
studies in assessing risks to miners from occupational DPM exposures. 
As noted in the 2001 risk assessment,

First, although dpm is a fine particulate, ambient air also contains 
fine particulates other than dpm. Therefore, health effects 
associated with exposures to fine particulate matter in air 
pollution studies are not associated specifically with exposures to 
dpm or any other one kind of fine particulate matter. Second, 
observations of adverse health effects in segments of the general 
population do not necessarily apply to the population of miners. 
Since, due to age and selection factors, the health of miners 
differs from that of the public as a whole, it is possible that fine 
particles might not affect miners, as a group, to the same degree as 
the general population.

    However,

Since dpm is a type of respirable particle, information about health 
effects associated with exposures to respirable particles, and 
especially to fine particulate matter, is certainly relevant, even 
if difficult to apply directly to dpm exposures. [66 FR 5767]

    Pope (2000) reviewed the epidemiological evidence for adverse 
health effects of PM2.5 and characterized populations at 
increased risk due to PM2.5 exposure. He found that ``[t]he 
overall epidemiologic evidence indicates a probable link between fine 
particulate air pollution and adverse effects on cardiopulmonary 
health.'' The observed endpoints include ``death from cardiac and 
pulmonary disease, emergency and physician office visits for asthma and 
other cardiorespiratory disorders, hospital admissions for 
cardiopulmonary disease, increased reported respiratory symptoms, and 
decreased measured lung function.'' Moreover, according to Pope, recent 
research suggests that ``those who are susceptible to increased risk of 
mortality from acutely elevated PM may include more than just the most 
old and frail who are already very near death.'' Pope went on to state 
that, with respect to chronic exposure, ``[t]here is no evidence that 
increased mortality risk is confined to any well-defined susceptible 
subgroup.''
    Table VI-10 identifies five studies on cardiovascular and 
cardiopulmonary effects published since the 2001 risk assessment 
(Lippmann et al., 2000; Magari et al., 2001; Pope et al., 2002; Samet 
et al., 2000a, 2000b; Wichmann et al., 2000). Three of these studies 
(Pope et al., 2002; Samet et al., 2000a, 2000b; Wichmann et al., 2000) 
significantly strengthen MSHA's existing evidence implicating 
particulate exposures with premature mortality from cardiovascular and 
cardiopulmonary causes.\5\ The Samet and Pope (2002) articles both 
establish statistically significant exposure-response relationships.
---------------------------------------------------------------------------

    \5\ As discussed below, Pope et al. (2002) also provides strong 
evidence linking chronic PM2.5 exposure with an elevated 
risk of lung cancer.

  Table VI-10.--Studies Relating to Cardiovascular and Cardiopulmonary
                           Effects, 2000-2002
------------------------------------------------------------------------
       Authors, years              Description           Key results
------------------------------------------------------------------------
Lippmann et al. 2000........  Day-to-day            After adjustment for
                               fluctuations in       the presence of
                               particulate air       other pollutants,
                               pollution in the      significant
                               Detroit area were     associations were
                               compared with         found between
                               corresponding         particulate levels
                               fluctuations in       and an increased
                               daily deaths and      risk of death due
                               hospital admissions   to circulatory
                               for 1985-1990 and     causes. However,
                               1992-1994.            relative risks were
                                                     about the same for
                                                     PM2.5 and larger
                                                     particles.
Magari et al., 2001.........  Longitudinal study    After adjusting for
                               of a male             potential
                               occupational cohort   confounding factors
                               examined the          such as age, time
                               relationship          of day, and urinary
                               between PM2.5         nicotine level,
                               exposure and          PM2.5 exposure was
                               cardiac autonomic     significantly
                               function.             associated with
                                                     disturbances in
                                                     cardiac autonomic
                                                     function.
Pope et al., 2002...........  Prospective cohort    After adjustment for
                               mortality study,      other risk factors
                               based on data         and potential
                               collected for         confounders, using
                               Cancer Prevention     a variety of
                               II Study, which       statistical
                               began in 1982.        methods, fine
                               Questionnaires were   particulate (PM2.5)
                               used to obtain        exposures were
                               individual risk       significantly
                               factor data (age,     associated with
                               sex, race, weight,    cardiopulmonary
                               height, smoking       mortality (and also
                               history, education,   with lung cancer).
                               marital status,      Each 10-[mu]g/m\3\
                               diet, alcohol         increase in mean
                               consumption, and      level of ambient
                               occupational          fine particulate
                               exposures). For       air pollution was
                               about 500,000         associated with an
                               adults, these were    increase of
                               combined with air     approximately 6% in
                               pollution data for    the risk of
                               metropolitan areas    cardiopulmonary
                               throughout the U.S.   mortality.
                               and with vital
                               status and cause of
                               death data through
                               1998.
Samet et al., 2000a, 2000b..  Time series analyses  Results of both the
                               were conducted on     20-city and 90-city
                               data from the 20      mortality analyses
                               and 90 largest U.S.   are consistent with
                               cities to             an average increase
                               investigate           in cardiovascular
                               relationships         and cardiopulmonary
                               between PM10 and      deaths of more than
                               other pollutants      0.5% for every 10
                               and daily mortality.  [mu]g/m\3\ increase
                                                     in PM10 measured
                                                     the day before
                                                     death. (Estimated
                                                     effects are, in
                                                     general, slightly
                                                     lower using a more
                                                     stringent
                                                     statistical
                                                     analysis. See
                                                     Dominici et al.,
                                                     2002.)
Wichmann et al., 2000.......  Time series analyses  Higher levels of
                               were conducted on     both fine and
                               data from Erfurt,     ultrafine particle
                               Germany to            concentrations were
                               investigate           significantly
                               relationships         associated with
                               between the number    increased mortality
                               and mass              rate.
                               concentrations of
                               ultrafine and fine
                               particles and daily
                               mortality.
------------------------------------------------------------------------


[[Page 32904]]

    Pope et al. (2002) warrants special attention because this study 
addresses chronic effects of long-term PM2.5 exposures. 
(Other studies on PM2.5, described in the 2001 risk 
assessment, have almost all dealt with acute exposure effects.) The 
authors concluded that ``* * * the findings of this study provide the 
strongest evidence to date that long-term exposure to fine particulate 
air pollution * * * is an important risk factor for cardiopulmonary 
mortality.'' In the 2001 risk assessment, the conclusion related to 
cardiopulmonary effects was motivated mostly by evidence on short-term 
exposures from daily time series analyses. Therefore, in finding a 
significant increase in cardiopulmonary mortality attributable to 
chronic fine particulate exposures, this study provides important 
supplement evidence supporting this conclusion. The portion of the 
study related to lung cancer effects is summarized in the next section.
    The EPA's 2004 Air Quality Criteria Document for particulate matter 
(EPA, 2004b) describes a number of additional studies related to the 
cardiopulmonary and cardiovascular effects of PM2.5, 
including work published later than that cited in the 2003 NPRM. One of 
the summary conclusions presented in that document is:

Overall, there is strong epidemiological evidence linking (a) short-
term (hours, days) exposures to PM2.5 with cardiovascular 
and respiratory mortality and morbidity, and (b) long-term (years, 
decades) PM2.5 exposure with cardiovascular and lung 
cancer mortality and respiratory morbidity. The associations between 
PM2.5 and these various health endpoints are positive and 
often statistically significant. [EPA, 2004b, Sec. 9 p. 46]
1. Cancer Effects
    The 2001 risk assessment concluded that DPM exposure, at 
occupational levels encountered in mining, was likely to increase the 
risk of lung cancer. The assessment also found that there was 
insufficient evidence to establish a causal relationship between DPM 
and other forms of cancer. Both of these conclusions are supported by 
the most recent scientific literature. The first part of this update 
contains a description of three new human research studies and a 
literature review relating DPM and/or other fine particulate exposures 
to lung cancer. Since it relates specifically to lung cancer, this 
subsection also discusses Dr. Chase's analysis. New research on the 
relationship between DPM exposures and other forms of cancer are 
described immediately after the lung cancer discussion.

Lung Cancer

    Table VI-11 presents three human studies pertaining to the 
association between lung cancer and exposures to DPM or fine 
particulates in general completed after the 2001 risk assessment was 
done.

        Table VI-11.--Studies on Lung Cancer Effects, 2000-2002.
------------------------------------------------------------------------
        Authors, year              Description           Key results
------------------------------------------------------------------------
Boffetta et al., 2001.......  Cohort consisting of  Statistically
                               entire Swedish        significant
                               working population    elevations in
                               other than farmers.   relative risk (RR)
                               Exposure assessment   of lung cancer
                               based on job title    among men for job
                               and industry,         categories with
                               classified            medium, and high
                               according to          exposure to diesel
                               probability and       exhaust, compared
                               intensity of diesel   to workers in jobs
                               exhaust exposure.     classified as
                                                     having no
                                                     occupational
                                                     exposure
Gustavsson et al., 2000.....  Case-control study    Adjusted RR for the
                               involving all 1,042   highest quartile of
                               male cases of lung    estimated lifetime
                               cancer and 2,364      exposure was 1.63,
                               randomly selected     compared to the
                               controls (matched     group with no
                               by age and            exposure.
                               inclusion year) in
                               Stockholm County,
                               Sweden from 1985
                               through 1990. Semi-
                               quantitative
                               assessment of
                               exposure to diesel
                               exhaust. Relative
                               Risk (RR) estimates
                               adjusted for age,
                               selection year,
                               tobacco smoking,
                               residential radon,
                               occupational
                               exposures to
                               asbestos and
                               combustion
                               products, and
                               environmental
                               exposure to NO2.
Pope et al., 2002...........  Prospective cohort    After adjusting for
                               mortality study       other risk factors
                               using data            and potential
                               collected for the     cofounders, chronic
                               American Cancer       PM2.5 exposures
                               Society Cancer        found to be
                               Prevention II Study   significantly
                               (began 1982).         associated with
                               Questionnaires used   elevated lung
                               to obtain             cancer mortality.
                               individual risk       Each 10-[mu]g/m\3\
                               factor data           increase in mean
                               including age, sex,   level of ambient
                               race, weight,         fine particulate
                               height, smoking       air pollution
                               history, education,   (PM2.5) associated
                               marital status,       with statistically
                               diet, alcohol         significant
                               consumption, and      increase of
                               occupational          approximately 8% in
                               exposures. This       risk of lung cancer
                               risk factor data      mortality.
                               combined with air
                               pollution data for
                               metropolitan areas
                               throughout U.S. and
                               vital status and
                               cause of death data
                               through 1998 for
                               about 500,000
                               adults.
------------------------------------------------------------------------

    Boffetta et al. (2001) investigated a Swedish cohort comprised of 
the whole Swedish working population not employed as farmers. Job title 
and industry were classified according to probability and intensity of 
diesel exhaust exposure in 1960 and 1970 and also according to the 
authors' confidence in the assessment. Cohort members were followed up 
for mortality for the 19-year period from 1971 through 1989. Cause of 
death and specific cancer type, when applicable, were obtained from 
national registries.
    Compared to workers in jobs classified as having no occupational 
exposure to diesel emissions, relative risks (RR) of lung cancer among 
men were 0.95, 1.1, and 1.3 for job categories with low, medium, and 
high exposure intensity, respectively. The elevated risks for the 
medium and high exposure groups were statistically significant, and no 
similar pattern was observed for other cancer types. The authors 
concluded that these results ``provide evidence of a positive exposure-
response relationship between exposure to diesel emissions and lung 
cancer among men.''
    Although this study adds to the cumulative weight of evidence 
establishing a causal link between DPM exposure and lung cancer, it 
does not provide very strong evidence when viewed in isolation. One 
weakness of the study is that the exposure assessment was based on 
self-reported occupation and industry, with no information on duration 
of employment in various jobs. (This sort of uncertainty in the 
exposure assessment, however,

[[Page 32905]]

would not normally be expected to induce a false exposure-response 
relationship.) Another weakness is that there was no information on 
potential confounders, such as tobacco smoking and lifestyle factors 
that may be associated with certain jobs. While recognizing this 
limitation, the authors considered it unlikely that confounders could 
account for the increasing trend in relative risk observed according to 
intensity of diesel exposure.
    Gustavsson et al. (2000) performed a case-control study involving 
all 1,042 male cases of lung cancer and 2364 randomly selected controls 
(matched by age and inclusion year) in Stockholm County, Sweden from 
1985 through 1990. Occupational exposure, smoking habits, and other 
potential risk factors were assessed based on written questionnaires 
mailed to the subject or next of kin. Relative Risk (RR) estimates were 
adjusted for age, selection year, tobacco smoking, residential radon, 
occupational exposures to asbestos and combustion products, and 
environmental exposure to NO2. Compared to the group with no 
exposure, adjusted RR for the highest quartile of estimated lifetime 
exposure was 1.63 (95% CI = 1.14 to 2.33). The authors concluded that 
``[t]he present findings add further evidence for an association 
between diesel exhaust and lung cancer * * * ''
    Strengths of this study include a semi-quantitative exposure 
assessment and adjustment of the relative risk for several important 
potential confounders. The statistically significant result 
corroborates the finding of a link between DPM exposure and lung cancer 
in MSHA's 2001 risk assessment.
    Pope et al. (2002) used the cohort established by the American 
Cancer Society Cancer Prevention II Study to examine the relationship 
between lung cancer and PM2.5 air pollution. This 
prospective cohort mortality study, which began in 1982, used 
questionnaires to obtain individual risk factor data (age, sex, race, 
weight, height, smoking history, education, marital status, diet, 
alcohol consumption, and occupational exposures). For about 500,000 
adults, these risk factors were combined with air pollution data for 
metropolitan areas throughout the U.S. and with vital status and cause 
of death data through 1998.
    After adjusting for other risk factors and potential confounders, 
using a variety of statistical methods, chronic PM2.5 
exposures were found to be significantly associated with elevated lung 
cancer mortality.\6\ Each 10 [mu]g/m\3\ increase in the mean level of 
ambient fine particulate air pollution was associated with a 
statistically significant increase of approximately 8% in the risk of 
lung cancer mortality. Within the range of exposures found in the 
study, the exposure-response relationship between PM2.5 and 
lung cancer was monotonically increasing. The authors concluded that 
``[e]levated fine particulate exposures were associated with 
significant increases in lung cancer mortality * * * even after 
controlling for cigarette smoking, diet, occupational exposure, other 
individual risk factors, and after controlling for regional and other 
spatial differences.''
---------------------------------------------------------------------------

    \6\ As discussed earlier, Pope et al. (2002) also provides 
strong evidence that chronic PM2.5 exposure increases the 
risk of premature cardiopulmonary mortality.
---------------------------------------------------------------------------

    Szadkowska-Stanczyk and Ruszkowska (2000) performed a literature 
review of studies relating to the carcinogenic effects of diesel 
emissions. The authors concluded that long-term exposure (> 20 years) 
was associated with a 30% to 40% increase in lung cancer risk in 
workers in the transport industry. This article was written in Polish, 
and MSHA was unable to obtain a translation of it for this update. 
However, based on the English abstract, it appears to add no new 
information to the 2001 risk assessment.
    Several commenters expressed opinions on the unpublished document 
by Dr. Gerald Chase (2004) entitled Characterizations of Lung Cancer in 
Cohort Studies and a NIOSH Study on Health Effects of Diesel Exhaust in 
Miners, which was placed into the public record at MARG's request. This 
document presents an analysis of some preliminary data provided by 
NIOSH and NCI at a public stakeholder meeting held on Nov. 5, 2003. 
These data were taken from unpublished charts that NIOSH and NCI used 
to inform the public on the status and progress of their ongoing 
project, A Cohort Mortality Study with a Nested Case-Control Study of 
Lung Cancer and Diesel Exhaust Among Nonmetal Miners [NIOSH/NCI 1997]. 
Researchers involved in that project have thus far published no 
analyses or conclusions based on these data. Dr. Chase, however, 
concluded that ``based on the limited data available to date, the 
number and pattern of lung cancer deaths reported * * * are in 
agreement with lung cancer deaths from the general population for the 
age groups involved * * *'' and ``* * * are possible without 
attributing any excess cancers to the study subject matter: diesel 
exhaust'' [emphasis added]. He offered no opinion as to whether the 
preliminary data actually demonstrate that there were no excess lung 
cancers attributable to DPM exposures.
    Although Dr. Chase noted that his analyses and conclusions were 
limited and based on incomplete information, some commenters 
interpreted his report as casting serious doubt on any increased risk 
of lung cancer associated with occupational DPM exposures. For example, 
one commenter said the report ``suggests lung cancer is not a problem 
in this worker population.'' Another commenter interpreted Dr. Chase's 
findings as providing ``startling evidence rebutting MSHA's PELs and 
risk analysis.'' Other industry commenters asserted that Dr. Chase's 
analysis ``eliminates the rationale upon which the final 160 microgram 
standard was premised.'' Another commenter claimed that Dr. Chase's 
analysis shows MSHA's justification for limiting DPM exposures is 
``contradicted by the NIOSH/NCI data.''
    Commenters representing organized labor, on the other hand, focused 
on the preliminary and incomplete nature of the data Dr. Chase 
analyzed. One such commenter pointed out that these data had not been 
made directly available on MSHA's website and that the status of the 
NIOSH/NCI study was not discussed in the re-opening announcement. 
Another commenter argued that the Chase analysis does not meet minimal 
standards of ``real epidemiological research'' and that it ``is 
worthless for the purpose of [MSHA's DPM] rulemaking.'' This commenter 
also stated that ``the record already contains ample evidence of the 
carcinogenicity of DPM'' and that ``the NIOSH/NCI study will not shake 
those findings, even if it should prove to be inconclusive.''
    The Chase analysis ignores at least three factors that can 
reasonably be expected to heavily influence the findings of the NIOSH/
NCI study: (a) Differentiation between exposed and unexposed miners 
within the study, (b) quantification of exposure, and (c) possible 
``healthy worker effect.'' According to the 1997 NIOSH/NCI study 
protocol, these three factors will be taken fully into account before 
any conclusions are published. The remainder of this subsection will 
explain how ignoring them, as in the Chase report, can mask adverse 
health effects potentially associated with DPM exposures.

[[Page 32906]]

(a) Differentiation Between Exposed and Unexposed Miners
    Approximately 50% of the miners in the NIOSH/NCI study cohort are 
expected to be surface workers (NIOSH/NCI, 1997, Tables A.1 and B.2). 
These miners are likely to have experienced far lower levels of DPM 
exposure than underground miners in the cohort. The NIOSH/NCI study 
protocol specifies that such members of the cohort--i.e., those who 
have had little or no occupational DPM exposure `` will be used as the 
``unexposed'' control group for the study. In other words, the protocol 
calls for statistically comparing the health of these surface workers 
to the health of the much more highly exposed underground workers.
    Dr. Chase did not distinguish between surface and underground 
workers in the cohort. Consequently, his analysis may dilute the lung 
cancer rate for exposed miners by combining it with the rate for miners 
with relatively little exposure. As noted by Dr. Chase, the preliminary 
data presented indicate that 9.8% of the deaths in the overall cohort 
were from lung cancer. He also suggests that the normal or 
``background'' percentage is 8.0%, based on the national lung cancer 
mortality rate and that the excess of 9.8% over 8.0% is not 
statistically significant. Suppose, however, that the overall excess of 
lung cancer deaths arose entirely from that half of the cohort 
comprising exposed, underground workers. Then, for miners in the 
``exposed'' group, the percentage of deaths from lung cancer would 
actually be 11.6%. Since 8.0/2 + 11.6/2 = 9.8, the 8.0% rate for 
surface workers would have diluted the 11.6% rate for exposed 
underground workers to yield an average rate of 9.8%. In this case, the 
lung cancer rate for underground miners would be about 45% greater than 
the national background rate (i.e., 11.6/8.0).
    Dr. Chase also claims that the 8% ``background'' rate is too low, 
since it combines all ages and includes relatively low lung cancer 
death rates for ages below 55 years. Although it is true that age-
specific lung cancer mortality rates increase after age 55, this should 
be considered only in conjunction with the age at death for members of 
the specific study cohort. Approximately two-thirds of the cohort 
members were born after 1940, with a maximum age at death of 56 years. 
For this age group, less than 5% of all deaths are attributed to lung 
cancer. Therefore, for purposes of comparison with this particular 
study cohort, an 8% background rate may be too high rather than too 
low, and the excess for underground workers may be even greater than 
the 45% indicated above.
(b) Quantification of Exposure
    As explained in the 2001 risk assessment, quantification of 
exposure was an important element in MSHA's evaluation of epidemiologic 
studies on DPM and lung cancer (FR 66 at 5784-5785, 5795ff). Relatively 
little weight was placed on studies that took no account of duration 
and intensity of exposure. At the time of the NIOSH/NCI Joint Study 
Meeting to discuss information with stakeholders on the progress of the 
study, exposure data for individual miners still were being processed. 
Since such exposure data were not presented at the meeting, they could 
not be used in Dr. Chase's analysis.
    The lack of detailed exposure data in Dr. Chase's analysis could 
potentially cause major distortions in interpretation of the results. 
The study cohort includes a number of workers with relatively short 
exposure duration. This is demonstrated by a 1981 NIOSH study showing 
that the mean tenure of underground trona miners working in 1976 was 
only about 3 years for ages greater than 25 years. (Attfield et al. 
1981). The two largest trona mines included in that study were also 
included in the NIOSH/NCI study (identified as Numbers 6 and 8 in Table 
A.1 of the 1997 NIOSH/NCI study protocol). Therefore, a substantial 
portion of the NIOSH/NCI study cohort may have been occupationally 
exposed to DPM for three years or less. If such short exposures produce 
little or no excess in lung cancers, then this portion of the cohort 
could mask a significant excess among workers with longer exposures. 
Since Dr. Chase's analysis lumps miners together without regard to 
exposure duration, it provides no effective way to evaluate effects 
associated with long-term exposure.
(c) Internal Versus External Analysis
    Another important element in MSHA's evaluation of epidemiologic 
studies on DPM and lung cancer was equitable composition of the groups 
being compared (FR 66 at 5783-5784, 5795ff). As explained in the 
Federal Register, comparison of an exposed cohort to an external 
control group can give rise to various forms of selection bias. For 
example, the ``healthy worker effect,'' which is widely recognized in 
the occupational health literature, tends to reduce estimates of excess 
risk in a group of workers when that group is compared to a general 
population. Several of the lung cancer cohort studies reviewed in the 
2001 risk assessment cohorts showed no excess lung cancers among 
exposed workers compared to an external population. Nevertheless, those 
studies showed excess lung cancers among exposed workers compared to 
otherwise similar but unexposed workers.
    To avoid selection biases, the 2001 risk assessment favored 
comparisons against internal control groups or studies that compensated 
for the healthy worker effect by means of an appropriate adjustment. 
Dr. Chase's analysis, however, focuses entirely on external comparisons 
with no compensating adjustment--an approach that the 2001 risk 
assessment generally discounted. Although the NIOSH/NCI study protocol 
explicitly calls for internal comparisons, the detailed exposure data 
necessary for such comparisons were not available to Dr. Chase since 
they were not presented during the November 5, 2003 public meeting.
(d) Conclusions Regarding Dr. Chase's Analysis
    Dr. Chase has argued that some preliminary and incomplete data made 
available from the NIOSH/NCI study do not demonstrate any excess lung 
cancer associated with DPM exposure. Even if Dr. Chase is correct, 
however, this may merely reflect limitations of the preliminary and 
incomplete data upon which his analysis relies. Because necessary data 
were not yet available, the Chase analysis was unable to consider a 
possible healthy worker effect, occupationally unexposed workers within 
the cohort, or potentially important variations in exposure intensity 
and duration. When the NIOSH/NCI study is completed, we are confident 
that it will take all these factors into account in accordance with the 
protocol.
    MSHA concludes that the data on which Dr. Chase's analysis is based 
are inadequate for identifying or assessing the relationship between 
occupational DPM exposure and excess lung cancer mortality. These 
incomplete data provide little insight into what a comprehensive 
analysis of the NIOSH/NCI study results will ultimately show, when 
carried out in accordance with the study protocol.

Bladder Cancer

    Boffetta and Silverman (2001) performed a meta-analysis of 44 
independent results from 29 distinct studies of bladder cancer in 
occupational groups with varying exposure to diesel exhaust. Studies 
were included only if there were at least five

[[Page 32907]]

years between time of first exposure and development of bladder cancer.
    Separate quantitative meta-analyses were performed for heavy 
equipment operators, truck drivers, bus drivers, and studies with semi-
quantitative exposure assessments based on a job exposure matrix (JEM). 
The overall relative risk (RR) for heavy equipment operators was RR = 
1.37 (95% CI: 1.05-1.81); for truck drivers, RR = 1.17 (1.06-1.29); for 
bus drivers, RR = 1.33 (1.22-1.45); and for JEM, RR = 1.13 (1.0-1.27).
    A quantitative meta-analysis was also performed on 8 independent 
studies showing results for ``high'' diesel exposure. The combined 
results were RR = 1.23 (1.12-1.36) for ``any exposure'' and RR = 1.44 
(1.18-1.76) for ``high exposure.''
    The authors discovered a strong indication of publication bias for 
truck and bus driver studies, a tendency for studies to be published 
only when they showed a positive result. However, the summary RR for 
the seven largest truck or bus driver studies was 1.26 (1.18-1.34), 
which is very close to the RR based on all 27 truck or bus driver 
results. There was no indication of publication bias for studies with 
semi-quantitative exposure assessments.
    The results of this meta-analysis suggest a statistically 
significant association between diesel exposure and an elevated risk of 
bladder cancer not fully explained by publication bias. Nevertheless, 
potential confounding by vibration, dietary factors, and infrequency of 
urination among drivers preclude a causal interpretation of this 
association.
    Not included in this meta-analysis was a study by Zeegers et al. 
(2001). This was a prospective case-cohort study involving 98 cases of 
bladder cancer among men occupationally exposed to diesel exhaust. A 
cohort of 58,279 men who were 55 to 69 years old in 1986 was followed 
up through December 1992. Exposure was assessed by job history given on 
a self-administered questionnaire, combined with experts' assessment of 
the exposure probability for each job. A ``cumulative probability of 
exposure'' was determined by multiplying job duration by the 
corresponding exposure probability. Four categories of relative 
cumulative exposure probability were defined: none, lowest third, 
middle third, and highest third. Relative risks were adjusted for age, 
cigarette smoking, and exposure to other occupational risk factors.
    The relative risk for the category with highest cumulative 
probability of exposure was RR = 1.17 (95% CI: 0.74-1.84). In light of 
the meta-analysis results described above, the lack of statistical 
significance found in this study may be due to low statistical power 
for detecting diesel exhaust effects, combined with nondifferential 
errors in the exposure assessment.
    As with the epidemiological studies on diesel exposure and bladder 
cancer considered in the meta-analysis, no adjustment was made in this 
study for infrequency of urination or for dietary patterns possibly 
associated with occupations having diesel exposures. Therefore, this 
study, like the meta-analysis performed by Boffetta and Silverman, has 
no impact on the 2001 risk assessment.

Pancreatic Cancer

    Ojaj[auml]rvi et al. (2000) performed a meta-analysis of 161 
independent results from 92 studies on the relationship between diesel 
exhaust exposure and pancreatic cancer. No elevated risk was associated 
with diesel exposure. The combined relative risk was RR = 1.0 (95% CI: 
0.9-1.3). This result is consistent with the 2001 risk assessment, 
which identified lung cancer and bladder cancer as the only forms of 
cancer for which there was evidence of an association with DPM 
exposure.
4. Mechanisms of Toxicity
    Table VI-12 describes 15 DPM toxicity studies published after the 
2001 risk assessment and cited in the 2003 NPRM. Table VI-12 also 
describes a 16th toxicity study (Arlt et al., 2002), which was cited by 
Dr. Jonathan Borak in comments submitted by MARG. All of these studies 
lend some degree of support to the conclusions of the 2001 risk 
assessment. In addition to briefly describing each study and its key 
results, the table identifies the agent(s) of toxicity investigated and 
indicates how the results support the risk assessment by categorizing 
the toxic effects and/or markers of toxicity found. The categories used 
to classify toxic effects are: (A) Immunological and/or allergic 
reactions, (B) inflammation, (C) mutagenicity and/or DNA adduct 
formation, (D) induction of free oxygen radicals, (E) airflow 
obstruction; (F) impaired clearance; (G) reduced defense mechanisms; 
and (H) adverse cardiovascular effects.

                    Table VI-12.--Studies on Toxicological Effects of DPM Exposure, 2000-2002
----------------------------------------------------------------------------------------------------------------
                                                                    Agent(s) of        Toxic
        Authors, year             Description      Key results        toxicity     effect(s)  *    Limitations
----------------------------------------------------------------------------------------------------------------
Al-Humadi et al., 2002.......  IT instillation   Exposure to DPM  DPM and carbon   A             ...............
                                in rats of 5 mg/  or carbon        black
                                kg saline, DPM,   black augments   particles.
                                or carbon black.  OVA
                                                  sensitization;
                                                  particle
                                                  composition
                                                  (of DPM) may
                                                  not be
                                                  critical for
                                                  adjuvant
                                                  effect.
Arlt et al., 2002............  In Vitro and in   Increased DNA    3-NBA, a         C             No DPM used.
                                Vivo:             adduct           constituent of
                                investigation     formation due    the organic
                                of metabolic      to in the        fraction of
                                activation of 3-  presence of      DPM.
                                nitrobenzanthro   human N,O
                                ne (3-NBA) by     acetyltransfer
                                human enzymes.    ases and
                                                  sulfotransfera
                                                  ses.

[[Page 32908]]


B[uuml]nger et al., 2000.....  In Vitro:         Production of    DE generated     C             ...............
                                assessment of     black carbon     from diesel
                                content of        and              engine.
                                polynuclear       polynuclear     DPM collected
                                aromatic          aromatic         on filters and
                                compounds and     compounds that   soluble
                                mutagenicity of   are mutagenic;   organic
                                DPM generated     correlation      extracts
                                from four         with sulfur      prepared.
                                fuels, Ames       content of
                                assay used.       fuel and
                                                  engine speed.
Carero et al., 2001..........  In Vitro:         DNA Damage       DPM, urban       C             ...............
                                assessment of     produced, but    particulate
                                DPM, carbon       no               matter (UPM),
                                black, and        cytotoxicity     and carbon
                                urban             produced.        black (CB).
                                particulate                       DPM, UPM
                                matter                             purchased from
                                genotoxicity,                      NIST, CB
                                human alveolar                     purchased from
                                epithelial                         Cabot.
                                cells used.
Castranova et al., 2001......  In Vitro:         DPM depresses    No information   D, F, G       ...............
                                assessment of     antimicrobial    on generation
                                DPM on alveolar   potential of     of DPM.
                                macrophage        macrophages,    (details may be
                                functions and     thereby          found in
                                role of           increasing       previous
                                adsorbed          susceptibility   publications
                                chemicals; rat    of lung to       from this lab).
                                alveolar          infections,
                                macrophages       this
                                used.             inhibitory
                               In Vivo:           effect due to
                                assessment of     adsorbed
                                DPM on alveolar   chemicals
                                macrophage        rather than
                                functions and     carbon core of
                                role of           DPM.
                                adsorbed
                                chemicals, use
                                of IT
                                instillation in
                                rats.
Fujimaki et al., 2001........  In Vitro:         Adverse effects  DE generated     A             Sensitization
                                assessment of     of DE on         from diesel                    to OA via IP
                                cytokine          cytokine and     engine.                        injection.
                                production,       antibody        DPM, CO2, SO2,                 Changes in
                                spleen cells      production by    and NO/NO2/NOx                 pulmonary
                                used.             creating an      measured.                      function not
                               In Vivo:           imbalance of                                    assessed.
                                assessment of     helper T-cell
                                cytokine          functions.
                                production
                                profile
                                following IP
                                sensitization
                                to OA and
                                subsequent
                                exposure to 1.0
                                mg/mg \3\ DE
                                for 12 hr/day,
                                7 days/week
                                over 4 weeks,
                                mouse
                                inhalation
                                model used.
Gilmour et al., 2001.........  In Vivo:          Exposure to      Woodsmoke, oil   A, B          No DPM used.
                                assessment of     woodsmoke        furnace
                                infectivity and   increased        emissions, and
                                allergenicity     susceptibility   residual oil
                                following         to and           fly ash (ROFA)
                                exposure to       severity of      used.
                                woodsmoke, oil    streptococcal
                                furnace           infection,
                                emissions, or     exposure to
                                residual oil      residual oil
                                fly ash, mouse    fly ash
                                inhalation        increased
                                model used, IT    pulmonary
                                instillation      hypersensitivi
                                used in rats.     ty reactions.
Hsiao et al., 2000...........  In Vitro:         Seasonal         PM collected     C             No DPM used.
                                assessment of     variations in    Hong Kong area
                                cytotoxic         PM, in their     and solvent-
                                effects (cell     solubility,      extractable
                                proliferation,    and in their     organic
                                DNA damage) of    ability to       compounds used.
                                PM2.5 (fine PM    produce
                                and PM2.5-10      cytotoxicity.
                                (coarse PM),     Long-term
                                rat embryo        exposure to
                                fibroblast        non-killing
                                cells used.       doses of PM
                                                  may lead to
                                                  accumulation
                                                  of DNA lesions.
Kuljukka-Rabb et al., 2001...  In Vitro:         Temporal and     Some DPM         C             Use of only
                                assessment of     dose-dependent   purchased from                 soluble
                                adduct            DNA adduct       NIST, some DPM                 organic
                                formation         formation by     collected on                   fraction of
                                following         PAHs.            filters from                   DPM.
                                exposure to      Carcinogenci      diesel
                                DPM, DPM          PAHs from        vehicle, and
                                extracts,         diesel           solvent-
                                benzo[a]pyrene,   extracts lead    extractable
                                or 5-methyl-      to stable DNA    organic
                                chrysene,         adduct           compounds used.
                                mammary           formation.
                                carcinoma cells
                                used.

[[Page 32909]]


Moyer et al., 2002...........  In Vivo: 2-phase  Induction and/   Indium           B, H          Nine
                                retrospective     or               phosphide,                     particulate
                                study, review     exacerbation     cobalt sulfate                 compounds
                                of NTP data       of arteritis     heptahydrate,                  selected to
                                from 90-day and   following        vanadium                       represent al
                                2-yr exposures    chronic          pentoxide,                     PM.
                                to                exposure         gallium
                                particulates,     (beyond 90-      arsenide,
                                use of mouse      day) to          nickel oxide,
                                inhalation        particulates.    nickel
                                model.                             subsulfide,
                                                                   nickel sulfate
                                                                   hexahydrate,
                                                                   talc,
                                                                   molybdenum
                                                                   trioxide used.
Saito et al., 2002...........  In Vivo:          DE alters        DE generated     A             ...............
                                assessment of     immunological    from diesel
                                cytokine          responses in     engine.
                                expression        the lung and    DPM, CO, SO2
                                following         may increase     and NO2
                                exposure to DE    susceptibility   measured.
                                (100 [mu]g/m\3\   to pathogens,
                                or 3 mg/m\3\      low-dose DE
                                DPM) for 7-hrs/   may induce
                                day x 5 days/wk   allergic/
                                x 4 wks, mouse    asthmatic
                                inhalation        reactions.
                                model used.
Sato et al., 2000............  In Vivo:          DE produced      DE generated     C             ...............
                                assessment of     lesions in DNA   from light-
                                mutant            and was          duty diesel
                                frequency and     mutagenic in     engine.
                                mutation          rat lung.       Concentration
                                spectra in lung                    of suspended
                                following 4-wk                     particulate
                                exposure to 1                      matter (SPM)
                                or 6 mg/m\3\                       measured, 11
                                DE, transgenic                     PAHs and
                                rat ihalation                      nitrated PAHs
                                model used.                        identified and
                                                                   quantitated in
                                                                   SPM.
Van Zijverden et al., 2000...  In Vivo:          DPM skew immune  DPM, carbon      A             Questionable
                                assessment of     response         black                          relevance of
                                immuno-           toward T         particles                      exposure route
                                modulating        helper 2 (Th2)   (CBP) and                      (sc
                                capacity of       side, and may    silica                         injection).
                                DPM, carbon       facilitate       particles
                                black, and        initiation of    (SIP) used.
                                silica            allergy.        DPM donated by
                                particles,                         Nijmegen
                                mouse model                        University,
                                used (sc                           CBP and SIP
                                injection into                     purchased from
                                hind footpad).                     BrunschwichChe
                                                                   mie and Sigma.
Vincent et al., 2001.........  In Vivo:          Increases in     Diesel soot,     H             ...............
                                assessment of     endothelin-1     carbon black
                                cardiovascular    and -3 (two      and urban air
                                effects           vasoregulators   particulates
                                following 4-hr    ) following      used.
                                exposure to 4.2   ambient urban   Diesel soot
                                mg/m\3\ diesel    particulates     purchased from
                                soot, 4.6 mg/     and diesel       NIST, carbon
                                m\3\ carbon       soot exposure.   black donated
                                black, or 48 mg/ Small increases   by University
                                m\3\ ambient      in blood         of California,
                                urban             pressure         urban air
                                particulates,     following        particulates
                                rat inhalation    exposure to      collected in
                                model used.       ambient urban    Ottawa.
                                                  particualtes.
Walters et al., 2001.........  In Vivo:          Dose and time-   DPM, PM, and     A, B          ...............
                                assessment of     dependent        coal fly ash
                                airway            changes in       used.
                                reactivity/       airway          DPM purchased
                                responsiveness,   responsiveness   from NIST, PM
                                and BAL cells     and              collected in
                                and BAL           inflammation     Baltimore, and
                                cytokines         following        coal fly ash
                                following         exposure to PM.  obtained from
                                exposure to 0.5  Increase in BAL   Baltimore
                                mg/mouse          cellularity      power plant.
                                aspirated DPM,    following
                                ambient PM, or    exposure to
                                coal fly ash.     DPM, but
                                                  airway
                                                  reactivity/
                                                  responsiveness
                                                  unchanged.

[[Page 32910]]


Whitekus et al., 2002........  In Vitro:         Thiol            DE generated     A, D          Changes in
                                assessment of     antioxidants     from light-                    pulmonary
                                ability of six    (given as a      duty diesel                    function
                                antioxidants to   pretreatment)    engine, DPM                    associated
                                interfere in      inhibit          collected,                     with
                                DPM-mediated      adjuvant         dissolved in                   sensitization
                                oxidative         effects of DPM   saline, and                    not assessed.
                                stress, cell      in the           aerosolized.
                                cultures used.    induction of
                               In Vivo:           OA
                                assessment of     sensitization.
                                sensitization
                                to OA and/or
                                DPM and
                                possible
                                modulation by
                                thiol
                                antioxidants,
                                mouse
                                inhalation
                                model used.
----------------------------------------------------------------------------------------------------------------
* KEY:
(A) Immunological and/or allergic reactions.
(B) Inflammation.
(C) Mutagenicity/DNA adduct formation.
(D) Induction of free oxygen radicals cardiovascular effects.
(E) Airflow obstruction.
(F) Impaired clearance.
(G) Reduced defense mechanisms.
(H) Adverse.

    In addition to the new toxicity studies, four new reviews on 
various aspects of the scientific literature related to mechanisms of 
DPM toxicity were cited in the 2003 NPRM. These are listed in Table VI-
13. Two of these reviews (ILSI, 2000 and Oberdoerster, 2002) focus on 
the applicability of the DPM rat toxicity studies to low-dose 
extrapolation for humans and conclude that such extrapolation is not 
appropriate. Since the 2001 risk assessment does not attempt to make 
any such extrapolation, these reviews do not affect MSHA's conclusions. 
As noted in the 2001 risk assessment, evidence that the carcinogenic 
effects of DPM in rats are due to overload of the rats' lung clearance 
mechanism does not rule out a mutagenic mechanism of carcinogenesis at 
lower exposure levels in other species. The other two review articles 
generally support the discussion in the 2001 risk assessment of 
inflammation responses due to DPM exposures.

                Table VI-13.--Review Articles on Toxicological Effects of DPM Exposure, 2000-2002
----------------------------------------------------------------------------------------------------------------
                                                                                Agent(s) of      Toxic  effects
          Authors, year               Description          Conclusions           toxicity               *
----------------------------------------------------------------------------------------------------------------
ILSI Risk Science Institute       Review of rat        No overload of rat   Poorly soluble
 Workshop Participants, 2000.      inhalation studies   lungs at lower       particles
                                   on chronic           lung doses of DPM    nonfibrous
                                   exposures to DPM     and no lung cancer   particles of low
                                   and to other         hazard anticipated   acute toxicity
                                   poorly soluble       at lower doses.      and not directly
                                   nonfibrous                                genotoxic (PSPs).
                                   particles of low
                                   acute toxicity
                                   that are not
                                   directly genotoxic.
Nikula, 2000....................  Review of animal     Species differences  DE, carbon black,   B, F
                                   inhalation studies   in pulmonary         titanium dioxide,
                                   on chronic           retention patterns   talc and coal
                                   exposures to DE,     and lung tissue      dust.
                                   carbon black,        responses
                                   titanium dioxide,    following chronic
                                   talc and coal dust.  exposure to DE.
Oberdoerster, 2002..............  In Vivo: review of   High-dose rat lung   Fibrous particles,
                                   toxicokinetics and   tumors produced by   and nonfibrous
                                   effects of fibrous   poorly soluble       particles that
                                   and nonfibrous       particles of low     are poorly
                                   particles.           cytotoxicity         soluble and have
                                                        (e.g., DPM) not      low cytotoxicity
                                                        appropriate for      (PSP).
                                                        low-dose
                                                        extrapolation (to
                                                        humans); lung
                                                        overload occurs in
                                                        rodents at high
                                                        doses.
Veronesi and Oortigiesen, 2001..  In Vitro: review of  Pulmonary receptors  PM: residual oil    A, B
                                   nasal and            stimulated/          fly ash,
                                   pulmonary            activated by PM,     woodstove
                                   innervation          leading to           emissions,
                                   (receptors) and      inflammatory         volcanic dust,
                                   pulmonary            responses.           urban ambient
                                   responses to PM,                          particulates,
                                   mainly BEAS cells                         coal fly ash, and
                                   sensory neurons                           and oil fly ash.
                                   used.
----------------------------------------------------------------------------------------------------------------
* KEY:
(A) Immunological and/or allergic reactions
(B) Inflammation
(C) Mutagenicity/DNA adduct formation
(D) Induction of free oxygen radicals
(E) Airflow obstruction
(F) Impaired clearance
(G) Reduced defense mechanisms

[[Page 32911]]


(H) Adverse cardiovascular effects.

D. Significance of Risk

    The first principal conclusion of the 2001 risk assessment was:

Exposure to DPM can materially impair miner health or functional 
capacity. These material impairments include acute sensory 
irritations and respiratory symptoms (including allergenic 
responses); premature death from cardiovascular, cardiopulmonary, or 
respiratory causes; and lung cancer.

    MSHA agrees with those commenters who characterized the weight of 
evidence from the most recent scientific literature as supporting or 
even strengthening this conclusion. Furthermore, this conclusion has 
also been corroborated by comprehensive scientific literature reviews 
carried out by other institutions and government agencies.
    In 2002, for example, the U.S. EPA, with the concurrence of its 
Clean Air Scientific Advisory Committee (CASAC), published its Health 
Assessment Document for Diesel Engine Exhaust (EPA, 2002). With respect 
to sensory irritations, respiratory symptoms, and immunological 
effects, this document concluded that:

At relatively high acute exposures, DE [diesel exhaust] can cause 
acute irritation to the eye and upper respiratory airways and 
symptoms of respiratory irritation which may be temporarily 
debilitating. Evidence also shows that DE has immunological toxicity 
that can induce allergic responses (some of which are also typical 
of asthma) and/or exacerbate existing respiratory allergies. [EPA, 
2002]

    In 2003, the World Health Organization (WHO) issued a review report 
on particulate matter air pollution and health. WHO concluded that 
``fine particles (commonly measured as PM2.5) are strongly 
associated with mortality and other endpoints such as hospitalization 
for cardiopulmonary disease, so that it is recommended that air quality 
guidelines for PM2.5 be further developed.'' (WHO, 2003)
    In the 10th edition of its Report on Carcinogens, the National 
Toxicology Program (NTP) of the National Institutes of Health formally 
retained its designation of diesel exhaust particulates as ``reasonably 
anticipated to be a human carcinogen.'' (U.S. Dept. of Health and Human 
Services, 2002) The report noted that:

Diesel exhaust contains identified mutagens and carcinogens both in 
the vapor phase and associated with respirable particles. Diesel 
exhaust particles are considered likely to account for the human 
lung cancer findings because they are almost all of a size small 
enough to penetrate to the alveolar region.
* * * Because of their high surface area, diesel exhaust 
particulates are capable of adsorbing relatively large amounts of 
organic material * * * A variety of mutagens and carcinogens such as 
PAH and nitro-PAH * * * are adsorbed by the particulates. There is 
sufficient evidence for the carcinogenicity for 15 PAHs (a number of 
these PAHs are found in diesel exhaust particulate emissions) in 
experimental animals. The nitroarenes (five listed) meet the 
established criteria for listing as ``reasonably anticipated to be a 
human carcinogen'' based on carcinogenicity experiments with 
laboratory animals. [U.S. Dept. of Health and Human Services, 2002]

Similarly, EPA's 2002 Health Assessment Document for Diesel Engine 
Exhaust concluded that diesel exhaust (as measured by DPM) is ``likely 
to be a human carcinogen.'' Furthermore, the assessment concluded that 
``[s]trong evidence exists for a causal relationship between risk for 
lung cancer and occupational exposure to D[iesel]E[xhaust] in certain 
occupational workers.'' (EPA, 2002, Sec. 9, p. 20)
    Although most commenters agreed that the adverse health effects 
associated with miners' DPM exposures warranted an exposure limit, some 
commenters continued to challenge the scientific basis for linking DPM 
exposures with an increased risk of lung cancer. An industry trade 
group submitted a critique of the 2001 risk assessment by Dr. Jonathan 
Borak, and this critique was endorsed by several other commenters 
representing the mining industry. The following discussion addresses 
Dr. Borak's comments in the same order that he presented them.
    1. Dr. Borak suggested that MSHA should have classified studies 
into 3 categories: positive, negative, and inconclusive. He indicated 
that MSHA's classification was asymmetric in the way that it classified 
studies as ``positive'' or ``negative,'' thereby distorting the results 
of MSHA's tabulation and nonparametric sign test, as presented in the 
2001 risk assessment.
    This comment was apparently based on a misunderstanding of how MSHA 
classified a study as ``negative'' for purposes of the sign test. In 
describing MSHA's criterion for classifying a study as negative, Dr. 
Borak quoted a passage from the 2001 risk assessment that actually 
pertained to a statistically significant negative study. The 
tabulations to which Dr. Borak referred symmetrically counted 
epidemiologic results as positive or negative based on whether the 
reported relative risk or odds ratio fell above or below 1.0.
    2. Dr. Borak stated that ``MSHA approached the analysis as though 
any study failing to document a protective effect of diesel must 
perforce be evidence of a harmful effect.''
    This statement is false and stems from Dr. Borak's misunderstanding 
of the symmetric criteria for MSHA's tabulations, as explained above. 
Furthermore, Dr. Borak's discussion of statistical significance and 
hypothesis testing in connection with this comment is applicable to 
evaluating the results of a single study--not to risk assessment based 
on combining multiple results.
    To evaluate the statistical significance of the aggregated 
epidemiologic evidence, the 2001 risk assessment relied largely on two 
meta-analyses (Bhatia et al., 1998; Lipsett and Campleman, 1999). MSHA 
applied the nonparametric sign test to its tabulation of all 47 studies 
in order to roughly summarize the combined evidence.
    3. Dr. Borak quoted the 2001 risk assessment as stating that ``MSHA 
regards a real 10% increase in the risk of lung cancer (i.e., a 
relative risk of 1.1) as constituting a clearly significant health 
hazard.'' He then stated that the concept of a ``real 10-percent 
increase'' is ``actually undefined and subjective.''
    Dr. Borak paraphrased language in the 2001 risk assessment, 
substituting a ``reported'' 10% increase for a ``real'' 10% increase 
(top of his p. 5). The risk assessment's distinction between 
``reported'' and ``real'' relative risks is important and corresponds 
to the fundamental distinction between a statistical estimate and the 
quantity being estimated.
    Contrary to Dr. Borak's characterization, the risk assessment 
recognized that epidemiological results are often subject to a great 
deal of statistical uncertainty. Such uncertainty can be expressed by 
means of a confidence interval for the ``real'' value being estimated 
by a ``reported'' result. For example, a reported relative risk (RR) of 
1.5 estimates the real relative risk underlying a particular study, for 
which a 95% confidence interval might be 1.3 to 1.7. This interval is 
designed to circumscribe the real relative risk with 95% probability.
    A 95% confidence interval for the real relative risk may be so 
broad (e.g., 0.8 to 1.4) as to overlap 1.0 and thereby render the 
reported result statistically non-significant. Because of the 
statistical uncertainty associated with a reported RR, extremely large 
study

[[Page 32912]]

populations are required in order to obtain statistically significant 
results when the real relative risk is near 1.0. The point being made 
in the passage that Dr. Borak quoted and then incorrectly paraphrased 
is that notwithstanding this statistical uncertainty, a real (as 
opposed to merely reported) 10% increase in the risk of lung cancer 
would constitute a clearly significant health effect. Therefore, 
reported results whose associated confidence intervals overlap 1.1 are 
consistent with potential health effects that are sufficiently large to 
be of practical significance.
    4. Dr. Borak asserted that ``* * * Federal Courts have held that 
relative risks of less than 2.0 are not sufficient for showing 
causation * * * but MSHA has rejected that view.''
    MSHA has not rejected the view expressed in the court decisions to 
which Dr. Borak alluded. Daubert v. Merrell Dow Pharmaceuticals, 509 
U.S. 579 (1993); and Hall v. Baxter Healthcare Corp., 947 F Supp. 1387 
(1996). As explained in the 2001 risk assessment, these decisions 
pertain to establishing the specific cause of disease for a particular 
person and not to establishing the increased risk attributable to an 
exposure. (FR 66 at 5787-5789) This distinction was illustrated by two 
analogies in the 2001 risk assessment: (1) There is low probability 
that a particular death was caused by lighting, but exposure to 
lighting is nevertheless hazardous; and (2) a specific smoker may not 
be able to prove that his or her lung cancer was ``more likely than 
not'' caused by radon exposure, yet radon exposure significantly 
increases the risk--especially for smokers. (FR 66 at 5787) As stated 
in the 2001 risk assessment, the court decisions are inapplicable 
because ``[t]he excess risk of an outcome, given an excessive exposure, 
is not the same thing as the likelihood that an excessive exposure 
caused the outcome in a given case.'' (FR 66 5787)
    Dr. Borak ignored MSHA's explanation of why the federal court 
rulings do not apply to the 2001 risk assessment. Instead, he attempted 
to differentiate the available epidemiologic studies on diesel exposure 
and lung cancer from examples, presented in the risk assessment, of 
studies reporting RR less than 2.0 that were nevertheless instrumental 
in previous clinical and public health policy decisions. For example, 
Dr. Borak pointed out that all ten of the results cited on the 
relationship between smoking and cardiovascular-related deaths achieved 
statistical significance. The risk assessment presented these examples, 
however, only to support the position that there is ``ample precedent'' 
for utilizing studies with RR less than 2.0 in a risk assessment. This 
was in response to comments urging MSHA to ignore all such results, 
even the many results with RR less than 2.0 that were also 
statistically significant. Thus, the ten results linking smoking to 
cardiovascular deaths, eight of which involved RR less than 2.0, 
adequately serve their intended illustrative purpose. Similarly, Dr. 
Borak's discussion of radon studies is not germane to their use as 
examples of studies with RR less than 2.0 that have not been generally 
discounted. Although the residential radon studies cited may, as Dr. 
Borak suggests, have been more powerful and had better exposure 
assessments than those available for DPM, they nevertheless demonstrate 
that there has been no blanket rejection of epidemiologic results 
whenever RR is less than 2.0.
    5. Dr. Borak objected to what he termed MSHA's ``reliance on the 
`healthy worker effect' [HWE] to explain the finding of small or no 
differences in various studies.'' He argued that ``[a]s a result, MSHA 
has biased its own evaluation of this literature in a manner that 
exaggerates the alleged human cancer risks of DPM, while diminishing 
studies that are not directly supportive of the MSHA perspective.''
    The 2001 risk assessment expresses a clear preference for studies 
using internal comparisons or well-matched cases and controls--studies 
in which the question of whether an HWE adjustment is desirable does 
not even arise. In fact, internal comparisons or matched cases and 
controls were utilized in all eight of the epidemiological studies 
identified in the risk assessment as presenting ``the best currently 
available epidemiological evidence.'' In contrast, the risk assessment 
identified six negative (i.e., RR or OR <  1.0) studies (out of 47) and 
noted that all six relied on unmatched cases and controls or on 
external comparisons to general populations, with no allowance for any 
potential HWE. However, potential bias due to HWE was not the only 
weakness identified in these six studies. The assessment also noted 
that five of the six studies had low statistical power due to a small 
study population, insufficient allowance for latency, or both. 
Furthermore, the assessment noted that all six of these negative 
studies contained weak DPM exposure assessments and failed to adjust 
for potentially different patterns of tobacco smoking in the disparate 
groups being compared. Dr. Borak did not dispute MSHA's exclusion of 
these six studies from the rank of best available epidemiologic 
evidence.
    More specifically, Dr. Borak objected to a relatively simple method 
of adjusting for the HWE used in one part of a meta-analysis by Bhatia 
et al. (1998) and also in some of the individual studies cited in the 
risk assessment. Dr. Borak noted that ``most epidemiologists agree that 
the effects of selection bias are generally more important early in a 
person's work life and do not apply equally to all diseases and disease 
processes.'' Citing the adjustment formula from Bhatia et al. (1998), 
Dr. Borak claimed that it is ``implicit throughout the MSHA 
discussion'' that ``the effects of HWE on observed lung cancer 
mortality are essentially equivalent (i.e., proportional) to its 
effects on mortality from all causes.''
    Although most epidemiologists may agree selection biases do not 
apply equally to all diseases, this does not render consideration of 
HWE irrelevant to epidemiologic studies of lung cancer. Health Effects 
Institute (HEI) (1999) states that ``[w]orker mortality tends to be 
below average for all major causes of death.'' The 2001 risk assessment 
accepted a proportional adjustment only insofar as it was utilized in 
some of the published epidemiological studies. Although Dr. Borak may 
be correct that compensating for HWE is not really so simple, a 
proportional adjustment may nevertheless be better than no adjustment 
at all. MSHA did not itself make any such adjustments or otherwise 
attempt to quantify the impact of HWE in any of the studies. MSHA did, 
however, accept HWE adjustments as they appeared in published studies.
    Although he did not explicitly say so, Dr. Borak presumably shares 
what he says is ``the general view that studies of cancer, particularly 
lung cancers, are not much affected by HWE.'' This view, however, is 
not universal. It is not, for example, shared by HEI (1999) or U.S. EPA 
(2002). Dr. Borak dwelled on pre-employment interviews and health exams 
as a source of HWE that would probably not apply to lung cancer 
studies, but pre-employment health screenings are not, after all, the 
only potential source of bias leading to HWE. Dr. Borak did not dispute 
the proposition that HWE reflects a potential bias when a working 
population is compared to a more general control population, or that 
this may be one of several factors contributing to a lack of positive 
results or statistical significance in some studies. As he has 
suggested, the potential impact of HWE in lung cancer studies may be 
greatest among those

[[Page 32913]]

involving the shortest latency allowances and/or follow-up times.
    6. In the study published by S[auml]verin et al. (1999), exposure 
measurements were obtained in 1992, whereas ``the mines ceased 
production in 1991'' when ``most of the miners were dismissed and 
abandoned underground work and exposure.'' Based on this apparent 
discrepancy, Dr. Borak questioned the argument used by S[auml]verin et 
al., and accepted in the risk assessment, to justify their assumption 
that their exposure measurements were representative of exposures from 
1970 to 1991. Dr. Borak speculated that the 1992 exposure measurements 
were likely to have been made during a ``staged simulation'' and that 
these measurements may have underestimated DPM levels under conditions 
of routine production. 
    To resolve this issue, MSHA contacted Dr. S[auml]verin directly and 
asked him to explain the sequence of events relating to mine closures 
and exposure measurements. Dr. Saverin replied as follows:

* * * [t]he full potash production of millions of tons per year in 
the seventies and eighties declined in the years after 1989, the 
official closing date being in 1991. But until 1994, there was a lot 
of mining activity underground because a mine cannot be abandoned 
immediately. So, in 1992, we had no problems to find exposure 
conditions not merely similar to but exactly like the routine-
production situation before. Thus, we did not have to rely on any 
staged simulation but made our measurements as state of the art 
requires. [S[auml]verin, R. 2005]


Thus, despite any ambiguity in the published article, Dr. S[auml]verin 
maintains that the 1992 measurements were obtained under normal 
production conditions and were fully representative of exposures from 
1970 through 1991. MSHA accepts Dr. S[auml]verin's assessment.
    As stated in the 2001 risk assessment, NIOSH commented that 
``[d]espite the limitations discussed * * * the findings from the 
S[auml]verin et al. (1999) study should be used as an alternative 
source of data for quantifying the possible lung cancer risks 
associated with Dpm exposures.'' MSHA is not relying on any single 
study but, instead, is basing its evaluation on the weight of evidence 
from all available data.
    7. Dr. Borak identified a number of weaknesses and limitations in 
the epidemiologic studies by S[auml]verin et al. (1999) and Johnston et 
al. (1997). Despite their shortcomings, the 2001 risk assessment ranked 
these two studies among the eight ``that provide the best currently 
available epidemiologic evidence.''
    As Dr. Borak indicated, all of the weaknesses and limitations he 
identified were recognized and discussed in the 2001 risk assessment. 
The risk assessment consistently and repeatedly emphasized that the 
strength of evidence relating DPM exposure to an increased risk of lung 
cancer lies not in any individual study but in the cumulative weight of 
the research literature taken as a whole. As stated in the risk 
assessment,

* * * MSHA recognizes that no single one of the existing 
epidemiologic studies, viewed in isolation, provides conclusive 
evidence of a causal connection between DPM exposure and an elevated 
risk of lung cancer in humans. Consistency and coherency of results, 
however, do provide such evidence. An appropriate analogy for the 
collective epidemiologic evidence is a braided steel cable, which is 
far stronger than any of the individual strands of wire making it 
up. (66 FR at 5825)

    Both of the additional epidemiological studies cited in the 2003 
NPRM specifically relating DPM exposures to lung cancer (Gustavsson et 
al., 2000 and Boffetta et al., 2001) found statistically significant 
positive results. The 2002 EPA document, which was compiled too early 
to consider these two newest studies, concluded that even at the far 
lower levels typically encountered in ambient air, ``[t]he available 
evidence [from toxicity studies and occupational epidemiology] 
indicates that chronic inhalation of DE is likely to pose a lung cancer 
hazard to humans.''
    This conclusion has now received important additional confirmation 
from a large scale mortality study involving exposure to combustion-
related fine particulate air pollution (Pope et al., 2002). This study, 
which included estimates of lung cancer effects, was cited in the NPRM 
but not considered in either the 2001 risk assessment or the 2002 EPA 
document. As described earlier, a statistically significant exposure-
response relationship was discovered between chronic PM2.5 
exposure in the ambient air and an increased risk of lung cancer. This 
finding is especially significant for confirming causality because it 
represents an entirely new source of evidence not subject to unknown 
biases that might tend to distort occupational epidemiology results in 
the same direction.
    Dr. Borak also stated that presently available data are 
insufficient to establish an exposure-response relationship for lung 
cancer that would justify setting the PEL at any specific level. The 
2001 risk assessment recognizes uncertainty in lung cancer exposure-
response and presents a broad range of estimated exposure-response 
relationships (66 FR at 5852-53). Even the lowest estimate shows 
unacceptable risk at levels commonly encountered in underground mines. 
Lack of a definitive exposure-response relationship means MSHA cannot 
precisely distinguish differences in health effects--e.g., between 
50DPM [mu]g/m3 and 100DPM [mu]g/
m3. Nevertheless, as explained below, MSHA can confidently 
say that exposures above the interim PEL are significantly more 
hazardous than exposures below the interim PEL.
    The second principal conclusion of the 2001 risk assessment was:

At DPM levels currently observed in underground mines, many miners 
are presently at significant risk of incurring these material 
impairments due to their occupational exposures to DPM over a 
working lifetime.

As described in Section VI.B, two new bodies of exposure data have been 
compiled since promulgation of the 2001 rule. Comparison of these data 
is not straightforward, since they employed different methods for 
measuring DPM. Nevertheless, the data suggest that exposure levels in 
many underground M/NM mines have dropped significantly, as compared to 
the 1989-1999 period covered by the 2001 risk assessment.
    The 2001 risk assessment quantified excess lung cancer risk based 
on a mean DPM concentration of 808 [mu]g/m3. This was based 
on 355 MSHA area and personal samples collected in production areas and 
haulageways at 27 underground M/NM mines between 1989 and 1999. Nearly 
all of these samples were collected without an impactor and analyzed 
for DPM content using the RCD method. The new samples, on the other 
hand, were collected with an impactor and analyzed for TC or EC using 
NIOSH Method 5040. To see how more recent exposure levels tie into the 
quantitative exposure-response models used in the 2001 risk assessment, 
it is necessary to convert sample results from both new sources of 
exposure data to approximate DPM concentrations.
    Samples from the 31-Mine Study were collected in 2001 using an 
impactor and were analyzed by NIOSH Method 5040. These samples showed a 
mean DPM concentration of 432 [mu]g/m\3\--assuming, as in the 2001 risk 
assessment, that TC comprises 80 percent of total DPM. Excluding the 
samples from trona mines, which were found to have significantly lower 
DPM levels than the other 27 underground M/NM mines with valid samples, 
the mean DPM

[[Page 32914]]

concentration was approximately 492 [mu]g/m\3\.\7\
---------------------------------------------------------------------------

    \7\ These values may be somewhat inflated due to the old 
``crimped foil'' SKC sampler design used for many of the samples 
collected during the 31-Mine Study. As explained elsewhere in this 
preamble, this design resulted in lower-than-expected filter deposit 
areas in many cases, leading to overestimates of the corresponding 
TC concentrations. (The SKC sampler design was eventually modified 
by substituting a retainer ring for the crimped foil. However, the 
systematic errors in deposit area observed during the 31-Mine Study 
have no bearing on the ``paired punch comparison'' used in that 
study to evaluate analytical measurement precision.)
---------------------------------------------------------------------------

    The other, more recent and more extensive, body of DPM exposure 
data considered here consists of 1,194 baseline samples obtained at 183 
mines in 2002-2003. These samples were all collected using a 
submicrometer impactor and analyzed by NIOSH Method 5040. Assuming that 
TC [ap]1.3 x EC and, as before, that TC comprises about 80 percent of 
the DPM, the mean DPM concentration observed was approximately 320 
[mu]g/m\3\.\8\ MSHA considers the baseline sampling results to be more 
broadly representative of DPM concentrations currently experienced by 
underground M/NM miners than the generally higher DPM concentrations 
reported in the 31-Mine Study. Since the baseline samples were 
collected later, part of the apparent reduction in mean concentration 
levels may be due to improved DPM controls implemented in response to 
the 2001 rule.
---------------------------------------------------------------------------

    \8\ The laboratory analysis of the baseline samples yielded two 
measures of TC: TC = EC + OC and TC = 1.3 x EC. However, since the 
intention under baseline sampling was to rely always on the lesser 
of these two values from each sample, no precautions were taken to 
avoid sampling near tobacco smoke and other substances that 
potentially interfere with the use of TC = EC + OC as a surrogate 
measure of DPM. Therefore, in the present discussion, MSHA is using 
only the TC = 1.3 x EC value to estimate baseline DPM levels.
---------------------------------------------------------------------------

    The 2001 risk assessment used the best available data on DPM 
exposures at underground M/NM mines to quantify excess lung cancer 
risk. ``Excess risk'' refers to the lifetime probability of dying from 
lung cancer during or after a 45-year occupational DPM exposure. This 
probability is expressed as the expected excess number of lung cancer 
deaths per thousand miners occupationally exposed to DPM at a specified 
mean DPM concentration. The excess is calculated relative to baseline, 
age-specific lung cancer mortality rates taken from standard mortality 
tables. In order to properly estimate this excess, it is necessary to 
calculate, at each year of life after occupational exposure begins, the 
expected number of persons surviving to that age with and without DPM 
exposure at the specified level. At each age, standard actuarial 
adjustments must be made in the number of survivors to account for the 
risk of dying from causes other than lung cancer. Occupational exposure 
is assumed to begin at age 20 and to continue, for surviving miners, 
until retirement at age 65. The accumulation of lifetime excess risk 
continues after retirement through the age of 85 years.
    Table VI-14, taken from the 2001 risk assessment, shows a range of 
excess lung cancer estimates at mean exposures equal to the interim and 
final DPM limits. The eight exposure-response models employed were 
based on studies by Saverin et al. (1999), Johnston et al. (1997), and 
Steenland et al. (1998). Assuming that TC is 80 percent of whole DPM, 
and that the mean ratio of TC to EC is 1.3, the interim DPM limit of 
500 [mu]g/m\3\ shown in Table VI-14 corresponds to the 308 [mu]g/m\3\ 
EC surrogate limit adopted under the present rulemaking.

Table VI-14.--Excess Lung Cancer Risk Expected at Specified DPM Exposure
                  Levels Over an Occupational Lifetime
        [Extracted from Table III-7 of the 2001 risk assessment]
------------------------------------------------------------------------
                                           Excess lung cancer deaths per
                                           1,000 occupationally exposed
                                                 workers [dagger]
       Study and statistical model       -------------------------------
                                             Final DPM      Interim DPM
                                            limit  200      limit  500
                                            [mu]g/m\3\      [mu]g/m\3\
------------------------------------------------------------------------
Saverin et al. (1999):
    Poisson, full cohort................              15              44
    Cox, full cohort....................              70             280
    Poisson, subcohort..................              93             391
    Cox, subcohort......................             182             677
Steenland et al. (1998):
    5-year lag, log of cumulative                     67              89
     exposure...........................
    5-year lag, simple cumulative                    159             620
     exposure...........................
Johnston et al. (1997):
    15-year lag, mine-adjusted..........             313             724
    15-year lag, mine-unadjusted........             513            783
------------------------------------------------------------------------
[dagger] Assumes 45-year occupational exposure at 1,920 hours per year
  from age 20 to retirement at age 65. Lifetime risk of lung cancer
  adjusted for competing risk of death from other causes and calculated
  through age 85. Baseline lung cancer and overall mortality rates from
  NCHS (1996).

    The mean DPM concentration levels estimated from both the 31-Mine 
Study (432-492 [mu]g/m\3\, depending on whether trona mines are 
included) and the baseline samples ([ap]320 [mu]g/m\3\) fall between 
the interim and final DPM limits shown in Table VI-14. All of the 
exposure-response models shown are monotonic (i.e., increased exposure 
yields increased excess risk, though not proportionately so). 
Therefore, using the most current available estimates of mean exposure 
levels, they all predict excess lung cancer risks somewhere between 
those shown for the interim and final limits. Thus, despite substantial 
improvements apparently attained since the 1989-1999 sampling period 
addressed by the 2001 risk assessment, underground M/NM miners are 
still faced with an unacceptable risk of lung cancer due to their 
occupational DPM exposures.
    The third principal conclusion of the 2001 risk assessment was:

By reducing DPM concentrations in underground mines, the rule will

[[Page 32915]]

substantially reduce the risks of material impairment faced by 
underground miners exposed to DPM at current levels.

    Although DPM levels have apparently declined since 1989-1999, MSHA 
expects that further improvements will continue to significantly and 
substantially reduce the health risks identified for miners. There is 
clear evidence of DPM's adverse health effects, not only at pre-2001 
levels but also at the generally lower levels currently observed at 
many underground mines. These effects are material health impairments 
as specified under section 101(a)(6)(A) of the Mine Act. From the 
baseline sampling results, 68 out of the 183 mines (37%) had at least 
one sample exceeding the interim exposure limit. Because the exposure-
response relationships shown in Table VI-14 are monotonic, MSHA expects 
that industry-wide implementation of the interim limit will 
significantly reduce the risk of lung cancer among miners.

VII. Feasibility

A. Background

    Section 101(a)(6)(A) of the Mine Act requires the Secretary of 
Labor in establishing health standards, to most adequately assure, on 
the basis of the best available evidence, that no miner will suffer 
material impairment of health or functional capacity over his or her 
working life. Standards promulgated under this section must be based 
upon research, demonstrations, experiments, and such other information 
as may be appropriate. MSHA, in setting health standards, is required 
to achieve the highest degree of health and safety protection for the 
miner, and must consider the latest available scientific data in the 
field, the feasibility of the standards, and experience gained under 
this or other health and safety laws.
    The legislative history of the Mine Act states:

This section further provides that ``other considerations'' in the 
setting of health standards are ``the latest available scientific 
data in the field, the feasibility of the standards, and experience 
gained under this and other health and safety laws.'' While 
feasibility of the standard may be taken into consideration with 
respect to engineering controls, this factor should have a 
substantially less significant role. Thus, the Secretary may 
appropriately consider the state of the engineering art in industry 
at the time the standard is promulgated. However, as the circuit 
courts of appeals have recognized, occupational safety and health 
statutes should be viewed as ``technology-forcing'' legislation, and 
a proposed health standard should not be rejected as infeasible 
``when the necessary technology looms on today's horizon''. AFL-CIO 
v. Brennan, 530 F.2d 109 (3d Cir. 1975); Society of Plastics 
Industry v. OSHA, 509 F.2d 1301 (2d Cir. 1975), cert. denied 427 
U.S. 992 (1975).
    Similarly, information on the economic impact of a health 
standard, which is provided to the Secretary of Labor at a [public] 
hearing or during the public comment period, may be given weight by 
the Secretary. In adopting the language of [this section], the 
Committee wishes to emphasize that it rejects the view that cost 
benefit ratios alone may be the basis for depriving miners of the 
health protection which the law was intended to insure. Rep. No. 95-
181, 95th Cong. 1st Sess. 21 (1977).

    In promulgating standards, hard and precise predictions from 
agencies regarding feasibility are not required. The ``arbitrary and 
capricious test'' is usually applied to judicial review of rules issued 
in accordance with the Administrative Procedure Act. The legislative 
history of the Mine Act further indicates that Congress explicitly 
intended the ``arbitrary and capricious test'' be applied to judicial 
review of mandatory MSHA standards. ``This test would require the 
reviewing court to scrutinize the Secretary's action to determine 
whether it was rational in light of the evidence before him and 
reasonably related to the law's purposes.'' S. Rep. No. 95-181, 95th 
Cong., 1st Sess. 21 (1977). In achieving the Congressional intent of 
feasibility under the Mine Act, MSHA may also consider reasonable time 
periods of implementation. Ibid. at 21.
    Though the Mine Act and its legislative history are not specific in 
defining feasibility, the Supreme Court has clarified the meaning of 
feasibility in the context of OSHA health standards in American Textile 
Manufacturers' Institute v. Donovan (OSHA Cotton Dust), 452 U.S. 490, 
508-09 (1981), as ``capable of being done, executed, or effected,'' 
both technologically and economically.
    MSHA need only base its predictions on reasonable inferences drawn 
from existing facts. In order to establish the economic and 
technological feasibility of a new rule, an agency is required to 
produce a reasonable assessment of the likely range of costs that a new 
standard will have on an industry, and an agency must show that a 
reasonable probability exists that the typical firm in an industry will 
be able to develop and install controls that will meet the standard. 
United Steelworkers of America, AFL-CIO-CLC v. Marshall, (OSHA Lead) 
647 F.2d 1189, 1273.

B. Technological Feasibility

    Courts have ruled that in order for a standard to be 
technologically feasible an agency must show that modern technology has 
at least conceived some industrial strategies or devices that are 
likely to be capable of meeting the standard, and which industry is 
generally capable of adopting. Ibid. (citing American Iron and Steel 
Institute v. OSHA, (AISI-I) 577 F.2d 825 (3d Cir. 1978) at 832-35; and, 
Industrial Union Dep't., AFL-CIO v. Hodgson, 499 F.2d 467 (DC 
Cir.1974)); American Iron and Steel Institute v. OSHA, (AISI-II) 939 
F.2d 975, 980 (DC Cir. 1991). The existence of general technical 
knowledge relating to materials and methods which may be available and 
adaptable to a specific situation establishes technical feasibility. A 
control may be technologically feasible when ``if through reasonable 
application of existing products, devices or work methods with human 
skills and abilities, a workable engineering control can be applied'' 
to the source of the hazard. It need not be an ``off-the-shelf'' 
product, but ``it must have a realistic basis in present technical 
capabilities.'' (Secretary of Labor v. Callanan Industries, Inc. 
(Noise), 5 FMSHRC 1900, 1908 (1983)).
    The Secretary may also impose a standard that requires protective 
equipment, such as respirators, if technology does not exist to lower 
exposures to safe levels. See United Steelworkers of America, AFL-CIO-
CLC v. Marshall, (OSHA Lead) 647 F.2d 1189, 1269 (DC Cir. 1981).
    MSHA has established that it is technologically feasible to reduce 
underground miners' exposures to the DPM interim permissible exposure 
limit (PEL) of 308 micrograms of EC per cubic meter of air 
(308EC [mu]g/m3) by using available engineering 
control technology and various administrative control methods. However, 
MSHA acknowledges that compliance difficulties may be encountered at 
some mines due to implementation issues and the cost of purchasing and 
installing certain types of controls. Therefore, this final rule 
incorporates the industrial hygiene concept of a hierarchy of controls 
for implementing DPM controls. To attain the interim DPM limit, mine 
operators are required to install, use, and maintain engineering and 
administrative controls to the extent feasible. When such controls do 
not reduce a miner's exposure to the DPM limit, controls are 
infeasible, or controls do not produce significant reductions in DPM 
exposures, operators must continue to use all feasible engineering and 
administrative controls and supplement them with respiratory 
protection. When respiratory protection is required under the final 
standard, mine operators must establish a respiratory protection 
program that

[[Page 32916]]

meets the specified requirements. Thus, MSHA has provided a regulatory 
scheme that adequately accomplishes control of exposure under 
circumstances where a mine operator cannot reduce a miner's exposure to 
the interim PEL solely by use of engineering and administrative 
controls, including work practices.
    DPM control technology is not new to the mining industry. MSHA has 
afforded the mining industry a significant period of time to implement 
DPM controls. The existing DPM standard was first promulgated on 
January 19, 2001 (66 FR 5706) with an effective date of July 19, 2002 
for meeting the interim concentration limit of 400 micrograms of TC per 
cubic meter of air. The instant rulemaking provides for a comparable EC 
PEL of 308 EC [mu]g/m3. Under the settlement 
agreement, MSHA allowed mine operators an additional year in which to 
begin to install appropriate engineering and administrative controls to 
reduce DPM levels due to feasibility constraints at that time. 
Altogether, the mining industry has had over four years to institute 
controls required under this rulemaking. Any controls currently used to 
meet the existing concentration limit can be used to reduce miners' 
exposures to the interim PEL.
    MSHA acknowledges that the current DPM rulemaking record lacks 
sufficient feasibility documentation to justify lowering the DPM limit 
below 308 EC [mu]g/m3 at this time. Therefore, 
MSHA is not lowering the limit in this rulemaking. MSHA believes that 
this interim limit is reasonable, and that MSHA can document 
feasibility across the affected sector of underground M/NM mines. MSHA 
is continuing to gather information on the feasibility of the mining 
industry to comply with a final DPM PEL of less than 308 EC 
[mu]g/m\3\
    MSHA emphasizes that a DPM control may be deemed feasible, and 
therefore be required by MSHA even if a miner's exposure is not reduced 
to the DPM limit. Mine operators cited for DPM overexposures will 
continue to be required to implement feasible engineering and 
administrative controls even if these controls are not fully successful 
in attaining the DPM exposure limit. In the context of this rule, 
feasible DPM controls must be capable of achieving a significant 
reduction in DPM. MSHA considers a significant reduction in DPM to be 
at least a 25% reduction in the affected miners' exposures. Thus, for 
mines that are out of compliance with the DPM interim limit, controls 
would be required that attain compliance, or that achieve at least a 
25% reduction in DPM exposure if it is not possible to attain 
compliance by implementing feasible controls. If feasible engineering 
and administrative controls are not capable of attaining compliance, or 
at least of achieving a DPM exposure reduction of 25%, MSHA would not 
require the implementation of those controls. In such cases, which MSHA 
believes will be very limited, MSHA would require miners to be 
protected using appropriate respiratory protective equipment.
    Some commenters criticized the 25% threshold for a significant 
reduction because it lacks a scientific basis, and that controls should 
be evaluated individually in reference to site-specific conditions and 
DPM levels for significance or effectiveness. MSHA notes that the 25% 
threshold for DPM is lower than the 50% threshold adopted in MSHA's 
noise rule. However, DPM's classification as a carcinogen justifies the 
more protective 25% level for determining whether controls achieve a 
significant reduction for purposes of assessing feasibility.
    MSHA also notes that most of the practical and effective controls 
that are currently available, such as DPM filters, enclosed cabs with 
filtered breathing air, and low-emission engines will achieve at least 
a 25% reduction. Other controls such as ventilation upgrades or 
alternative fuel blends may achieve a 25% reduction, depending on 
exposure circumstances and the specific nature of the subject control. 
It should also be noted that reductions of less than 25% could be due 
to normal day-to-day variations in mining operations as opposed to 
reductions due to implementing a control technology. MSHA's Compliance 
Guide includes the 25% significant reduction for determining 
feasibility.
    If a particular DPM control were capable of achieving at least a 
25% reduction all by itself, MSHA would evaluate the costs of that 
individual control to determine its economic feasibility. If a number 
of controls could together achieve at least a 25% reduction, but no 
individual control, if implemented by itself, could achieve a 25% 
reduction, MSHA would evaluate the total costs of all controls added 
together to determine their economic feasibility as a group. In 
determining whether a combination of controls is economically feasible, 
MSHA would consider whether the total cost of the combination of 
controls is wholly out of proportion to the expected results. MSHA will 
not cost the controls individually, but will combine their expected 
results to determine if the 25% significant reduction criteria can be 
satisfied.
    MSHA's rulemaking record addressing feasibility includes: MSHA's 
final report on the 31-Mine Study; NIOSH's peer review of the 31-Mine 
Study; results from MSHA's baseline sampling at mines covered under the 
DPM standard; results of MSHA's comprehensive compliance assistance 
work at mining operations with implementation issues affecting 
feasibility; NIOSH's conclusions on the performance of the SKC sampler 
and the availability of technology for control of DPM; NIOSH's Diesel 
Emissions Workshops in 2003 in Cincinnati and Salt Lake City; the 
Filter Selection Guide posted on the MSHA and NIOSH Web sites; MSHA's 
final report on DPM filter efficiency; NIOSH's report titled, ``Review 
of Technology Available to the Underground Mining Industry for Control 
of Diesel Emissions''; and, the NIOSH Phase I Isozone study titled, 
``The Effectiveness of Selected Technologies in Controlling Diesel 
Emissions in an Underground Mine--Isolated Zone Study at Stillwater 
Mining Company's Nye Mine'' all of which were developed following 
promulgation of the 2001 DPM final rule.
    One other NIOSH document resulting from the DPM M/NM Partnership 
became available to MSHA in April 2004. That document is titled, ``An 
Evaluation of the Effects of Diesel Particulate Filter Systems on Air 
Quality and Personal Exposure of Miners at Stillwater Mining Case 
Study: Production Zone (Phase II Study).'' As stated in the final 
report:

The objective of Phase II of this study was to determine the effects 
of those DPF systems being used on production vehicles at Stillwater 
Mine on workplace concentrations of EC and regulated gases in an 
actual mining application where multiple diesel-powered vehicles 
operated simultaneously during full-shift mining activities.

    MSHA evaluated this evidence as it relates to feasibility and found 
that unlike the Phase I Isozone Study, the Phase II study does not 
contain any new significant information affecting the ability of the 
mining industry to comply with the requirements of this final rule. 
MSHA, therefore, finds this data to be cumulative in nature and has 
included it in the rulemaking record as supplemental information. MSHA 
discusses the Phase II study results in more detail in this section of 
the preamble. MSHA emphasizes that mine operators obtained access to 
this study on the date of publication since the study was generated by 
the DPM M/NM Partnership.
    MSHA committed to implementing several initiatives related to


[[Continued on page 32917]]


From the Federal Register Online via GPO Access [wais.access.gpo.gov]
]                         
 
[[pp. 32917-32966]] Diesel Particulate Matter Exposure of Underground Metal and 
Nonmetal Miners

[[Continued from page 32916]]

[[Page 32917]]

enforcement and enhancing the mining industry's ability to comply with 
the 2001 final rule. Among other things, MSHA agreed that it would not 
issue citations for potential violations of the interim concentration 
limit promulgated in the 2001 standard until after MSHA and NIOSH were 
satisfied with the performance characteristics of the SKC sampler and 
the availability of practical mine worthy DPM filter technology. MSHA 
also agreed to provide DPM sampling training for its inspectors, and to 
provide comprehensive compliance assistance to the industry through 
July 19, 2003. MSHA's compliance assistance activities included:
     Conducting compliance assistance meetings throughout the 
country to discuss how to comply with the DPM standard;
     Providing a compliance guide answering key questions;
     Conducting an inventory of existing underground diesel-
powered equipment;
     Providing information to mine operators on feasible DPM 
controls; and,
     Obtaining baseline sampling results at each underground 
mine covered under the standard solely for the purpose of compliance 
assistance rather than for enforcement purposes.

Additional compliance assistance activities also were conducted, and 
are discussed later in this section of the preamble.
    During the compliance assistance period, MSHA agreed that mine 
operators would not be cited for potential violations of the interim 
limit provided they took good-faith steps to develop and implement a 
written compliance strategy and cooperated with MSHA. Also, MSHA would 
issue a noncompliance citation for exceeding the interim concentration 
limit only if MSHA believed that an operator was not acting in good 
faith, or if an operator failed to cooperate in the compliance 
assistance. Per the agreement, after July 19, 2003, MSHA began to issue 
citations for violations associated with the interim limit. During the 
compliance assistance period (through July 19, 2003), MSHA did not 
identify any mines that failed to take good faith steps toward 
achieving compliance or cooperate with MSHA. Consequently, no citations 
for violations associated with the interim limit were issued prior to 
July 20, 2003.
    MSHA provided DPM training to its inspectors and to the extent 
possible, completed its compliance assistance activities in accordance 
with the settlement agreement. During September and October 2002, 
seminars covering the rule, MSHA's enforcement policy, DPM sampling, 
and DPM engineering control technologies were held in Ebensburg, PA, 
Knoxville, TN, Lexington, KY, Des Moines, IA, Kansas City, MO, 
Albuquerque, NM, Coeur d'Alene, ID, Elko, NV, and Green River, WY. The 
DPM Compliance Guide was posted on the MSHA DPM Single Source Page and 
also issued as an MSHA Program Policy Letter (PPL P03-IV-1, 
effective August 19, 2003). Extensive information on feasible controls 
for DPM was included in the Compliance Guide/Program Policy Letter and 
listed on MSHA's DPM Single Source Page for DPM. The inventory of 
diesel engines was completed September 30, 2002. Baseline DPM samples 
were not obtained at a remaining few mines until after July 20, 2003 
primarily to allow time to cover sampling at intermittent operations. 
However, enforcement sampling at these mines was delayed until after 
completion of baseline sampling to provide these mine operators with 
further opportunity to implement controls, if necessary.
    As discussed below in this section of the preamble, both MSHA and 
NIOSH are satisfied with the performance of the SKC sampler and on the 
availability of practical DPM filter technology.
    DPM Sampling Method. Though not under substantive review in this 
rulemaking, existing Sec.  57.5061(b) establishes that MSHA will 
continue to sample miners' personal exposures by using a respirable 
dust sampler equipped with a submicrometer impactor and analyze samples 
for the amount of EC using the NIOSH Analytical Method 5040, or any 
other method that NIOSH determines gives equal or improved accuracy in 
DPM sampling. The DPM sampling method is discussed in the section-by-
section portion of this preamble under Sec.  57.5060(a) addressing the 
permissible exposure limit. MSHA includes a more detailed discussion of 
its sampling method on its DPM Single Source Web page. Based on current 
information in the rulemaking record, MSHA concludes that it has a 
technologically feasible measurement method that operators and MSHA can 
use to accurately determine if miners' exposures exceed the interim 
PEL.
    Performance of the SKC Sampler. MSHA and NIOSH are satisfied with 
the performance of the SKC sampler. The 31-Mine Study includes a 
comprehensive discussion of MSHA's and NIOSH's work with SKC that 
improved the performance of the sampler. In MSHA's final report on the 
31-Mine Study, it concluded that SKC satisfactorily addressed concerns 
over earlier known defects in the DPM sampling cassettes and 
availability of cassettes to both MSHA and mine operators. Just prior 
to and during the 31-Mine Study, NIOSH and MSHA observed that the 
perimeter of the DPM deposit on the filter was not consistently 
circular and varied among the SKC samplers. This resulted in a variable 
and unpredictable deposit area. The cause of this was found and quite 
successfully remedied allowing NIOSH to express its satisfaction with 
the performance of the SKC sampler by letter of June 25, 2003, to MSHA 
that states, in part:

Concurrent with the work of the partnership were research tasks to 
ensure that diesel particulate matter can be accurately measured in 
these mines. The SKC DPM cassette is a size selective sampler 
designed to collect DPM samples that are characterized by an 
aerodynamic diameter less than 0.8[mu]m, while avoiding 
contamination with mineral dust. The use of the SKC sampler could 
not be recommended initially because of a problem relating to 
irregular deposition of DPM on the cassette sample. However, this 
problem has been solved, and we are now satisfied with the 
performance of the SKC sampler. The research regarding the 
performance of the SKC sampler has been documented, peer-reviewed, 
and is currently accepted for publication by Applied Occupational 
and Environmental Hygiene Journal.

    Baseline Sampling. For the 2001 standard, MSHA based its 
feasibility projections on an average DPM concentration level of over 
800TC [mu]g/m\3\. MSHA found in the 31-Mine Study that 
miners' average TC exposure was 345 [mu]g/m\3\. MSHA's baseline 
sampling revealed that miners average EC exposure was 196 [mu]g/m\3\. 
The average TC exposure measured as EC + OC was 293 [mu]g/m\3\, and as 
calculated by EC x 1.3 was 255 [mu]g/m\3\. MSHA believes that these 
lower averages probably result from the introduction of DPFs, clean 
engines, better maintenance, and the elimination of interferences as 
confirmed by MSHA's compliance assistance baseline sampling. The 
baseline sampling results are discussed in detail in Section V.
    DPM Enforcement. MSHA believes that final Sec. 57.5060(d) 
adequately addresses feasibility issues related to meeting the interim 
limit of 308EC [mu]g/m\3\ under Sec.  57.5060(a). Under 
these sections, MSHA has amended the type of exposure that will be 
regulated along with the methods of compliance with the interim PEL to 
provide mine operators with greater flexibility in reducing DPM 
exposures. This final DPM rule adopts MSHA's long-standing enforcement 
practice established for

[[Page 32918]]

other exposure-based standards applicable to M/NM mines. Also, MSHA 
underscores the fact that the enforcement scheme established in this 
final rule also is based on the DPM settlement agreement.
    In spite of the changes in this final rule that increase 
flexibility, MSHA realizes that some mine operators will continue to 
need on-site technical assistance. MSHA is committed to assisting these 
operators in special mining situations that could affect the successful 
use of DPFs or other engineering control systems. Mine operators can 
request this assistance from their respective MSHA District Manager.
    Additionally, MSHA concludes that the established hierarchy of 
controls for complying with the DPM interim limit adequately protects 
miners from exposure to DPM in those circumstances where MSHA found 
control methods to be infeasible under existing Sec.  57.5060(d)(2) for 
certain activities including inspection, maintenance and repair 
activities. MSHA has removed from this final rule the requirement for 
mine operators to apply to the Secretary of Labor for relief from 
applying control technology to comply with the final DPM limit. 
Instead, MSHA's hierarchy of controls strategy will result in quicker 
responses to supplementing protection for miners exposed to the health 
risks associated with DPM.
    MSHA believes that it has sufficiently accommodated the mining 
industry's needs with respect to complying with the DPM standard and 
has developed an appropriate and reasonable enforcement scheme under 
this rule. MSHA estimates that approximately 183 mines are covered 
under the standard. These mines produce commodities such as gold, 
limestone, trona, platinum, lead, silver, zinc, marble, gypsum, salt, 
and potash. Based on MSHA's baseline sampling results, over 70% of 
these underground mines were in compliance with the interim DPM limit.
    MSHA is confident that engineering and administrative controls 
(including work practice controls) exist that are capable of reducing 
DPM exposures to the interim PEL of 308EC [mu]g/m\3\ in all 
types of underground M/NM mines. MSHA believes that virtually all mine 
operators will successfully attain compliance with the interim limit by 
choosing from among various currently available feasible engineering 
and administrative DPM control options, including but not limited to 
DPF systems, ventilation upgrades, oxidation catalytic converters, 
alternative fuels, fuel additives, enclosures such as cabs and booths 
with filtered breathing air, improved diesel engine maintenance 
procedures and instrumentation, diesel engines with lower DPM 
emissions, various work practices and administrative controls. MSHA has 
given the mining industry flexibility under the final standard in 
selecting the individual or combination of DPM controls that best suit 
a mine operator's specific needs, conditions, and operating practices.
    MSHA received numerous comments concerning the technological 
feasibility of the 2003 NPRM. Some commenters opposed any changes in 
the 2001 DPM standard. A few of these commenters suggested that MSHA's 
current rulemaking record does not support revising the 2001 final 
rule. They believe that in order to justify a change that in their view 
reduces health protection, MSHA must first make a determination that 
the DPM limits established in the 2001 final rule are infeasible for 
the mining industry as a whole to attain. These commenters note that, 
to the contrary, MSHA fully substantiated its conclusions regarding 
feasibility in the 2001 final rule.
    According to these commenters, during the period from August 2001 
through January 2002, MSHA stated in the final report to the 31-Mine 
Study that the mean concentration of DPM was 345TC [mu]g/
m\3\, substantially below the required concentration limit of 
400TC [mu]g/m\3\. These commenters pointed out that these 
results were obtained at a time when MSHA believes few mining 
operations had begun to implement DPM controls, or where the 
implementation of such controls was in its early stages and had not yet 
achieved significant reductions in DPM exposure. Other supportive 
evidence noted by these commenters included the results of the baseline 
sampling indicating that only 30% of the mines tested were out of 
compliance.
    MSHA agrees that it should utilize data from its final report on 
the 31-Mine Study and the baseline sampling in assessing technological 
feasibility, but MSHA does not consider the mean concentration obtained 
in the 31-Mine Study or the number of mines with baseline samples 
exceeding the interim limit to be the definitive data sources in this 
assessment. For example, although the mean concentration of DPM 
reported in the final report to the 31-Mine Study was only 
345TC [mu]g/m\3\, the mean DPM concentration value does not 
reflect the wide range of sample results obtained between mines or 
within individual mines, some of which exceeded 1000TC 
[mu]g/m\3\. Likewise, although only 30% of the mines had baseline 
sampling results exceeding the interim limit, MSHA expects some of 
these mines may have encountered compliance difficulties due to 
implementation issues relating to such factors as DPF regeneration and 
retrofitting DPFs to existing pieces of equipment, and due to the costs 
of purchasing and installing DPM controls.
    Therefore, in assessing technological feasibility, MSHA believes it 
should also consider data obtained subsequently from other sources, 
including MSHA's comprehensive compliance assistance work at mining 
operations, current agency enforcement experience, the NIOSH Diesel 
Emissions Workshops in Cincinnati and Salt Lake City, and the NIOSH 
Phase I Isozone Study. MSHA agrees with commenters who take the 
position that the interim DPM limit can be attained by the industry as 
a whole through implementation of feasible engineering and/or 
administrative (including work practice) controls. However, MSHA does 
not agree with commenters who oppose any changes to the 2001 final 
rule.
    Some commenters suggested that the proposed modification to the 
2001 standard would reduce health protection for miners, a consequence 
that Sec.  101(a)(9) of the Mine Act prohibits. MSHA disagrees. Section 
101(a)(9) of the Mine Act provides that: ``No mandatory health or 
safety standard promulgated under this title shall reduce the 
protection afforded miners by an existing mandatory health or safety 
standard.'' MSHA interprets this provision of the Mine Act to require 
that all of the health or safety benefits resulting from a new standard 
be at least equivalent to all of the health or safety benefits 
resulting from the existing standard when the two sets of benefits are 
evaluated as a whole. Int'l Union v. Federal Mine Safety and Health 
Admin., 920 F.2d 960, 962-64 (DC Cir. 1990); Int'l Union v. Federal 
Mine Safety and Health Admin., 931 F.2d 908, 911 (DC Cir 1991).
    In fact, MSHA believes that the interim EC limit established in 
this rulemaking is comparable to the existing TC limit. Correcting the 
surrogate for identifying miners' exposures to DPM is critical for 
protection of miners and will result in a valid DPM sample that MSHA 
can adequately substantiate. MSHA's hierarchy of controls strategy in 
the final rule is based on long-standing industrial hygiene practice in 
both the mining industry and general industry. As implemented in this 
final rule, the hierarchy of controls ensures that the most protective 
means of compliance (engineering and administrative controls) are used 
first,

[[Page 32919]]

and that respiratory protection is permitted only where MSHA determines 
that: Engineering and administrative controls are infeasible; controls 
do not produce significant reductions in DPM exposures; or controls do 
not reduce exposures to the interim DPM limit.
    The DPM litigants raised their concerns to MSHA with implementation 
issues related to regeneration and retrofitting exhaust after-treatment 
controls on existing mining equipment. These, along with various other 
compliance concerns, eventually led to the 31-Mine Study. At that time, 
only a few mine operators in the U.S. had begun to implement after-
treatment control technology on their underground diesel-powered 
equipment. As is often the case when unfamiliar technologies are 
integrated into an industry sector, the process was slow, and at least 
initially, the results were less-than-fully satisfactory. As noted 
elsewhere in this section, many mine operators, for example, 
experimented with DPF installations on a few pieces of equipment on a 
trial basis, with mixed results at best. MSHA does not dispute these 
findings, but believes that DPF failures were the result of 
inappropriate DPF selection for a given application. However at the 
time, these operators were convinced that DPF technology was 
fundamentally deficient for application in underground mining. In an 
effort to resolve a variety of issues raised by the industry that were 
believed to present potential compliance problems, MSHA agreed to 
conduct the 31-Mine Study.
    Many commenters also claimed that MSHA's determination that the 
rule is technologically feasible assumed the widespread utilization of 
DPFs, which these commenters do not believe have proven mine worthy and 
which may be affected by the aforementioned implementation issues. In 
response, MSHA notes that while it continues to highly recommend use of 
DPFs, its technological feasibility determination was based on the 
application of a variety of engineering and administrative control 
approaches for obtaining compliance, and was not limited to DPFs. MSHA 
has determined that DPF systems are available and mine worthy for 
controlling miners' exposures to DPM. As discussed later in this 
section of the preamble, both MSHA and NIOSH are satisfied that DPF 
systems are currently available for most mining equipment, and that 
these systems can be successfully applied if mine operators make 
informed decisions regarding filter selection, retrofitting, engine and 
equipment deployment, operation, and maintenance, and specifically work 
through issues such as in-use efficiencies, secondary emissions, engine 
backpressure, DPF regeneration, DPF reliability and durability.
    Implementation issues, such as DPF regeneration and retrofitting 
DPFs to existing pieces of equipment, primarily affect a small number 
of mines. Mines affected are those that will need to utilize DPFs to 
attain compliance because other control options, such as ventilation 
upgrades, low-emission engines, alternative diesel fuels, and cabs with 
filtered breathing air are either infeasible at these particular mines, 
or because these mine operators have already utilized these other 
control options to the maximum extent feasible but have not yet 
attained compliance. Since a variety of feasible control options are 
available, and implementation issues relating to DPFs affect a 
relatively small number of mines, the industry as a whole will not be 
impeded from attaining compliance with the interim PEL.
    MSHA does not dispute this early experience with DPF installations 
in U.S. underground mines, and in fact, acknowledged these concerns in 
the final report of the 31-Mine Study. One of the major conclusions of 
the study states:

Compliance with both the interim and final concentration limits may 
be both technologically and economically feasible for metal and 
nonmetal underground mines in the study. MSHA, however, has limited 
in-mine documentation on DPM control technology. As a result, MSHA's 
position on feasibility does not reflect consideration of current 
complications with respect to implementation of controls, such as 
retrofitting and regeneration of filters. MSHA acknowledges that 
these issues may influence the extent to which controls are 
feasible. The Agency is continuing to consult with the National 
Institute of Occupational Safety and Health, industry and labor 
representatives on the availability of practical mine worthy filter 
technology.

    After completing the 31-Mine Study, however, MSHA obtained 
additional documentation on DPM control technology that it had 
previously lacked. This information includes data on both 
implementation issues and costs, and was obtained from such sources as 
MSHA's comprehensive compliance assistance activities, MSHA's 
enforcement experience, and NIOSH's Diesel Emission Workshops in 
Cincinnati and Salt Lake City. Also, MSHA now has in-mine data on the 
filter efficiency of DPFs in U.S. mines as a result of the NIOSH Phase 
I Isozone study (discussed in detail in this preamble).
    Effectiveness of the DPM Estimator. MSHA's DPM Estimator is a 
Microsoft[reg] Excel spreadsheet computer program that 
calculates the reduction in DPM concentration that can be obtained by 
implementing individual, or combinations of engineering controls in a 
given production area of a mine. MSHA has repeatedly advised the mining 
community throughout the DPM rulemakings that the Estimator is one of 
many tools that can be used to assist mine operators with assessing 
feasibility of compliance with the DPM limits. MSHA used the estimator 
to support its feasibility assessment for the 2001 final rule, as well 
as the feasibility section of the 31-Mine Study which is used to 
support this final rule.
    The analyses in the 31-Mine Study were based on the highest DPM 
sample result obtained at each mine. Using the Estimator, new DPM 
levels were computed for this ``worst case'' sample result based on the 
application of one, or a combination of the following control 
technologies: DPFs, low emission engines, and upgraded ventilation. To 
adequately protect all miners even if the mine operator changes 
equipment deployment schemes in the future, the methodology for the 
technological feasibility analysis required all major emission sources 
at a given mine, plus similar spare equipment, to be provided with the 
same DPM controls that were specified for the equipment associated with 
the ``worst case'' sample result.
    Likewise, the economic feasibility analysis for each mine was based 
on costing the same controls for all major DPM emission sources, and 
similar spare equipment, as were required to reduce the ``worst case'' 
sample result to the compliance level. The rationale for this approach 
is that if the same controls are applied to all major DPM sources and 
spare equipment as are required to attain compliance for the ``worst 
case'' exposures, all exposures in the mine will be reduced at least to 
the compliance level, if not lower, regardless of future equipment 
usage, equipment deployment, mine production levels, etc.
    In the 31-Mine Study, DPFs were assumed to be capable of achieving 
an 80% reduction in DPM emissions. This 80% filtration efficiency value 
was based on laboratory tests. Since the 2001 final rule was 
promulgated, MSHA has obtained the results of the NIOSH Phase I Isozone 
Study conducted under actual in-mine testing, and which concludes that 
filter efficiency is about 75% for total DPM and ranged over 88% to 90% 
for EC for ceramic monolith wall-flow type DPFs of either silicon 
carbide or

[[Page 32920]]

cordierite composition. DPM reductions obtained by replacing older 
existing engines with new, low-emission engines are based on the DPM 
emissions of the new engine relative to the DPM emissions of the 
existing engine. For instance, if a new engine emits 0.10 grams per 
brake horsepower-hour (g/bhp-hr) of DPM and the existing engine emits 
0.50 g/bhp-hr of DPM, the Estimator would compute a DPM reduction of 
80% when the new engine replaces the existing engine. DPM reductions 
obtained through ventilation upgrades are based on the new ventilation 
airflow rate compared to the existing ventilation airflow rate. For 
example, if the new ventilation airflow rate is 80,000 cfm and the 
existing airflow rate is 40,000 cfm, the Estimator would compute a 
reduction in the DPM concentration of 50%.
    The Estimator was peer-reviewed during the 2001 final rulemaking 
and was published both as an SME Preprint for the 1998 SME Annual 
Meeting (Preprint 98-146, March 1998) and in the April 2000 SME 
Journal. Its predictions have been compared to actual in-mine DPM 
measurements (before and after DPM controls were implemented) with good 
agreement. Indeed, one commenter who was critical of the Estimator, 
nonetheless, noted that, ``The math which forms the basis for the 
Estimator's calculations cannot be challenged `` total exhaust 
emissions from diesel equipment (in grams/hr) when diluted with mine 
ventilation air flows (in cubic feet per minute) yield an estimated DPM 
concentration (in micro-gram per cubic meter) if the emissions are 
perfectly mixed with the air flow.''
    Despite its sound mathematical basis, this and other commenters 
stated that the Estimator was flawed, and hence, the technological and 
economic feasibility assessments were likewise flawed. These commenters 
specifically stated that the Estimator was flawed because two inputs 
utilized by the Estimator, DPM emissions (both raw and reduced via 
DPFs) and air flows, are subject to interpretation and assumptions. 
Furthermore, they believe that the Estimator's computations of DPM 
concentrations are valid only if engine emissions are perfectly mixed 
with the air flow, which they suggest does not occur in an actual mine.
    MSHA disagrees with this conclusion. These commenters make an 
erroneous assumption with respect to MSHA's utilization of the 
Estimator. The Estimator actually incorporates two independent means of 
calculating DPM levels: one based on DPM sampling data for the subject 
mine, and one based on the absence of such sampling data. Where no 
sampling data exist, the Estimator calculates DPM levels based on a 
straightforward mathematical ratio of DPM emitted from the tailpipe (or 
DPF, in the case of filtered exhaust) per volume of ventilation air 
flow over that piece of equipment. This is referred to in the Estimator 
as the ``Column B'' option for calculating DPM concentrations. The 
commenters' observation that the Estimator fails to account for 
imperfect mixing between DPM emissions and ventilating air flows is a 
valid criticism of the ``Column B'' option. For this and other reasons, 
the Estimator's instructions urge users to utilize the ``Column A'' 
option whenever sampling data are available.
    In the ``Column A'' option, the Estimator's calculations are 
``calibrated'' to actual sampling data. Whatever complex mixing between 
DPM emissions and ventilating air flows existed when DPM samples were 
obtained, are assumed to prevail after implementation of a DPM control. 
This is an entirely reasonable assumption, and in fact, there is no 
engineering basis to assume otherwise. Indeed, comparisons of ``Column 
A'' Estimator calculations and actual DPM measurements taken in mines 
before and after implementation of DPM controls have shown good 
agreement, indicating that Estimator calculations do adequately 
incorporate consideration for complex mixing of DPM and air flows when 
the ``Column A'' option is used.
    The Estimator was originally developed with both the Column A and 
Column B options because at that time, the specialized equipment 
required for DPM sampling, such as the submicron impactor, was not 
widely available. Consequently, few mine operators were able to obtain 
the in-mine DPM sample data required for utilizing the Column A option. 
Now that the required sampling equipment is readily available, MSHA 
strongly recommends that the Column A option be used exclusively, as 
MSHA did in the 31-Mine Study. Since all Estimator analyses conducted 
during the 31-Mine Study utilized the Estimator's ``Column A'' option, 
the comment regarding imperfect mixing is not relevant.
    The Estimator utilizes raw (an unfiltered emission) tailpipe DPM 
emissions per se as an input data value only when a low-emission engine 
is specified as a DPM control. For most of the mines in the 31-Mine 
Study, unfiltered tailpipe DPM emissions were not factored into 
Estimator analysis because a change in engines was not specified. Where 
new engines were specified, MSHA based its estimate of unfiltered 
tailpipe emissions on laboratory dynamometer testing conducted 
according to the EPA 8-mode test duty cycle. This test is a common 
standard used by government and industry for diesel engine emissions 
analysis. Where actual test data were not available for a given engine, 
emissions were estimated based on the type of engine (make and model, 
model year, direct injection, pre-chamber, naturally aspirated, 
turbocharged, electronic controlled, etc.) and horsepower. Filtered 
emissions were assumed to be 20% of unfiltered tailpipe emissions, 
corresponding to 80% filter efficiency. As noted above, the 80% filter 
efficiency was a conservative assumption based on MSHA and other 
laboratory and NIOSH in-mine test data indicating DPM efficiencies of 
80% to 87% for both cordierite and silicon carbide filters. Note that 
these efficiencies relate to DPM filtration. Higher filtration 
efficiencies are obtained for TC and EC. Air flows, where relevant for 
estimator analysis, were based on the sampler's comments, and/or the 
accompanying mine ventilation plans or maps.
    A number of commenters suggested that MSHA's DPM sampling results 
in isolated sections of mines are assumed by MSHA to be representative 
of on-going exposure levels in those mines, despite the fact that 
results varied widely. In the 31-Mine Study, MSHA did not, in fact, 
assume a sample result from an isolated section of a mine was 
necessarily representative of on-going DPM exposure levels throughout 
that mine. The study methodology stipulated that the highest observed 
DPM level for a given mine would be the basis for specifying DPM 
controls for the entire mine. A key underlying assumption of this 
methodology is that DPM levels do vary, often significantly, from one 
part of a mine to another. However, to insure that study findings would 
be conservative, the study methodology required that the highest DPM 
level, not the average or lowest DPM level, was the basis for 
specifying controls.
    Some commenters asserted that when analyzing sampling data for the 
31-Mine Study, MSHA assumed that ventilation flows measured at the 
sampling location applied throughout the subject section of the mine. 
They also asserted that MSHA assumed effective ventilation for dilution 
existed throughout the mine, and that neither of these assumptions was 
necessarily valid. For most of the mines in the 31-Mine Study for which 
a DPM reduction was necessary, ventilation was not an issue, and 
consequently, MSHA did not specify any changes in ventilation. For 
these mines, DPM reductions were obtained

[[Page 32921]]

by utilizing DPFs and/or low-emission engines, and the only assumption 
regarding ventilation was that it would not be changed.
    In the few cases where ventilation upgrades were specified, the 
upgrades were limited to auxiliary systems that supplied air to the 
sampled area only. Initial air flows utilized by the Estimator for 
those areas prior to implementing the upgrades were based on the 
comments and/or any accompanying ventilation plans or maps accompanying 
the sample. Where upgraded auxiliary ventilation was specified, MSHA 
frequently noted deficiencies in existing auxiliary ventilation system 
components such as inappropriately placed fans and blast-damaged or 
otherwise deteriorated and compromised vent bags. In these cases, the 
specified ventilation changes involved simply correcting the obvious 
deficiencies in the existing systems and increasing fan capacity.
    MSHA recognizes that there has to be a sufficient air quantity 
present in the main ventilation system in order for an auxiliary system 
to function properly (i.e. without recirculation), and that DPM levels 
in the main ventilation system from which the auxiliary system draws 
its air must be sufficiently below the DPM limit to prevent miners' 
overexposures in the stopes.
    Some commenters stated that in the 31-Mine Study, MSHA assumed that 
the only equipment needing DPM controls was the equipment operating 
while sampling took place. As noted above, the study methodology 
insured a conservative result by applying the same controls required to 
attain compliance for the equipment associated with the ``worst case'' 
sample to all similar DPM sources (and spares) in the entire mine, even 
if the subject ``worst case'' sample concentration was substantially 
higher than the remaining samples for that mine, and regardless of 
whether a particular piece of equipment was operating during sampling 
or not. For most mines in the study requiring DPM reductions, controls 
were specified for all or most of the normal production contingent of 
equipment, along with an allowance for spare equipment, particularly 
loaders and trucks, which are typically the largest source of DPM.
    Some commenters stated that in the 31-Mine Study, MSHA assumed 80% 
DPF filtration efficiency, and gave no consideration to potential 
NO2 problems related to DPFs. As noted above, the assumption 
of 80% filtration efficiency is conservative, and is based on actual 
laboratory and in-mine test data. Regarding NO2 generation 
from DPFs and the associated health concerns, MSHA acknowledges that 
NO2 can be produced by passive DPFs that are wash-coated 
with platinum-based catalysts. However, when such filters are utilized 
under reasonable ventilation conditions, the NO2 increases 
should be manageable and should not constitute a serious health hazard 
or compliance problem for the mine operator. An example of successfully 
using highly platinum-catalyzed DPFs without creating hazardous 
NO2 concentrations is Greens Creek mine which has installed 
such filter systems on its large trucks and loaders. During MSHA 
compliance assistance sampling at this mine in January 2002, 
NO2 increases of around 1 ppm were observed downstream of 
stopes where 1 loader and 2 or 3 trucks were operating for 2 to 3 
hours.
    MSHA also notes that in situations where passive DPF regeneration 
is desired, but where ventilation may be insufficient to adequately 
dilute and carry away harmful NO2 concentrations, 
alternatives to highly platinum-catalyzed DPFs exist. Examples include 
base metal catalyzed DPFs and lightly platinum-catalyzed filters used 
in conjunction with a fuel-borne catalyst, which have a regeneration 
temperature somewhat higher than highly platinum-catalyzed filters. 
These passively regenerating DPFs do not increase NO2 
concentrations compared to unfiltered exhaust emissions.
    Even more importantly, however, in the 31-Mine Study, all DPFs were 
specified as active type regeneration systems, not passive type 
systems. Likewise, in the corresponding economic feasibility 
assessment, all costs for DPFs included an assumption that mine 
operators would opt for active regeneration. Without detailed on-site 
analysis and evaluation of the subject equipment and duty cycles, MSHA 
could not assume a DPF system would passively regenerate. Also, active 
filter systems are typically more costly than an equivalent passive 
system, so specifying an active system would be more conservative from 
a costing perspective. Since actively regenerated DPFs have no platinum 
wash-coatings applied to the filters (and in fact, have no wash-
coatings at all), they do not produce any increased NO2 
emissions compared to unfiltered engines. NO2 emissions and 
associated health concerns were not addressed in the 31-Mine Study 
because the DPM controls specified in the study did not affect 
NO2 emissions.
    Some commenters also stated that MSHA failed to specify any major 
ventilation upgrades (new main fans, new ventilation shafts, etc.) in 
the 31-Mine Study, and that by avoiding major ventilation upgrades, the 
resulting compliance cost estimates were unrealistically low. In 
responding, MSHA notes that it did not specify any major ventilation 
upgrades in the 31-Mine Study because, based on the study methodology, 
the analysis did not indicate the need for major ventilation upgrades 
in order to attain compliance with either the interim or final DPM 
limits at any of the 31 mines.
    This does not mean that major ventilation upgrades would have been 
ill-advised, ineffective, or unbeneficial for any of the mines in the 
study. MSHA did note in the final report that strategies other than 
those specified in the study could also be successful, and there may be 
valid reasons why a mine operator might choose a different mix of 
controls (such as a major ventilation upgrade) for a given mine based 
on mine-specific factors to which MSHA's analysts were not privy at the 
time of the study. It was explicitly stated in the final report that 
the DPM controls specified for a particular mine did not necessarily 
represent the only feasible control strategy, nor the optimal control 
strategy for that mine. The purpose of specifying controls for each 
mine was simply to demonstrate that feasible controls capable of 
attaining compliance existed, and to provide a framework for costing 
such controls on a mine-by-mine basis.
    Indeed, since the completion of the 31-Mine Study, MSHA has 
observed that mine operators in the stone industry, for example, have 
chosen to attain compliance without utilizing DPFs. These operators 
instead have opted to upgrade ventilation (usually by adding or re-
positioning booster fans and installing or repairing ventilation 
control structures such as air curtains and brattices), install low-
emission engines, utilize equipment cabs with filtered breathing air, 
initiate a variety of work practices that contribute to reducing 
personal exposures to DPM, and in a few cases, use alternative diesel 
fuels such as bio-diesel fuel blends and diesel/water emulsions.
    Some of these mine operators may have had reasons other than DPM 
compliance alone that helped justify their decisions. For example, 
ventilation upgrades can also improve gaseous emission levels, dust 
levels, visibility, clearance of blasting smoke and gases, and 
inefficient or even counterproductive deployment of booster fans. Mine 
operators that have opted to replace older, dirty engines with newer, 
low emission engines benefit from greater fuel economy and better 
maintenance diagnostics. Cabs

[[Page 32922]]

with filtered breathing air improve operator comfort and productivity, 
as well as reducing dust and noise exposures.
    DPF Systems.
    DPFs suitable for any duty cycle are currently commercially 
available for most engine sizes and types used in underground M/NM 
mining. DPF options include silicon carbide and cordierite ceramic 
monolith type wall flow filters designed for passive regeneration, 
active on-board or active off-board regeneration, or passive/active 
regeneration. For most filters requiring active regeneration, the time 
required for filter regeneration varies from less than 1 hour to 8 
hours, depending on system type. Another option that is suitable for 
smaller, light duty equipment is a high-temperature disposable pleated 
element filter.
    Although every mine is unique, and virtually every DPF application 
has unique features, the variety of DPF systems available make it 
feasible to apply a DPF to most types of equipment or engines, and 
application or duty cycle. The only exception known to MSHA would be 
applying a DPF to a very old (pre-1970s vintage technology) engine 
having very high DPM emissions and a medium or light duty cycle. In 
theory, such an application would collect DPM, but due to rapid soot 
build-up on the filter media and corresponding rapid increase in engine 
back-pressure, such a DPF application would probably be impractical. 
MSHA has observed very few such engines in the underground M/NM mining 
industry, but in the few instances where emissions from such engines 
need to be controlled, mine operators are advised to choose a control 
option other than a DPF.
    MSHA is aware of reports by mining companies and others that some 
DPFs have not performed satisfactorily in the field. These reports 
refer to problems such as short filter life (a matter of weeks in some 
cases), equipment that bogs down when filters are installed, and 
uncontrolled regenerations and similar problems resulting in damaged or 
destroyed filters. MSHA has determined that most DPF failures result 
from inappropriate filter selection due to the failure by mine 
operators to fully consider all filter selection criteria prior to 
ordering DPF systems. In a few cases, filter failures were traced to 
manufacturing defects that were later resolved, while in a few others, 
an unrelated component failure on the host equipment (such as a 
turbocharger failure) caused a failure in the downstream DPF.
    Most problems with filter selection relate to the installation of a 
passively regenerating type filter on a machine that does not produce 
sufficient exhaust temperature for a sufficient portion of the duty 
cycle to initiate passive regeneration. A passive type filter that 
doesn't regenerate continues to trap soot until the backpressure on the 
engine causes the engine to ``bog down,'' or an uncontrolled 
regeneration occurs. The system may function satisfactorily for a 
while, either regenerating as expected, or at least partially 
regenerating. But if the machine's duty cycle lessens in severity, even 
for a single shift (for example, a production loader that is normally 
worked very hard might be used for a shift to perform road maintenance 
or clean-up duty), the filter may become overloaded.
    MSHA's determination that DPFs are a technologically feasible DPM 
control option is based on two factors: Laboratory and in-mine testing 
which has documented their high filtration efficiency, and numerous 
successful applications in routine production mining situations where 
DPFs have been appropriately matched to machines and duty cycles. When 
DPFs are properly selected and maintained for an application, the 
result is optimal performance and maximum filter life.
    In order to achieve satisfactory filter performance, filter life, 
and filtration efficiency, it is critical that a DPF be appropriately 
matched both to the diesel engine, and to the duty cycle and intended 
application of the subject equipment. For example, two identical 
machines may need different types of filter systems based on the 
machines' respective duty cycles. One machine that works hard due to 
the road grades that the machine must transverse during a shift may 
generate sufficient exhaust gas temperatures to support a passive 
regeneration DPF system. However, the second machine may run 
continuously on flat roads in the mine and, therefore, may not be 
capable of generating sufficient exhaust gas temperatures to support 
passive regeneration. Consequently, the second machine must use an 
active regenerating DPF system, or change out a disposable filter on a 
regular basis. Importantly, if the first machine, due for example to a 
breakdown of the second machine, assumes the second machine's duties, 
even on a temporary basis, it would be very possible if not likely, 
that its passive DPF system would fail to regenerate. Hence, when 
specifying a DPF system for a particular piece of equipment, mine 
operators should consider not only the intended application and duty 
cycle of the machine, but also other applications and duty cycles to 
which that machine may be occasionally assigned on a nonroutine basis.
    In order to assist the mining industry in selecting an appropriate 
filter, the MSHA and NIOSH internet web sites include a comprehensive 
compliance assistance tool, the Filter Selection Guide. One of many 
MSHA DPM compliance assistance tools, the Filter Selection Guide 
provides mine operators with detailed step-by-step assistance in 
selecting appropriate DPF systems that are compatible with their 
specific equipment and duty cycles. Also, the Filter Selection Guide 
provides information on modifications and adjustments to diesel-powered 
equipment that mine operators may have to make to successfully apply 
DPF systems.
    Prior to initiating the DPF selection process, mine operators 
should make certain that they are properly maintaining their engines, 
and that the engines are not consuming excessive amounts of crankcase 
oil. Operators should then obtain exhaust temperature logs or traces 
for several shifts, and use these traces to help select the appropriate 
DPF system for that machine and application. Exhaust temperature traces 
can be analyzed by mine personnel or DPF suppliers to assist in 
selecting a workable DPF system. Exhaust gas temperatures are an 
important factor in selecting a DPF because passive filter regeneration 
is possible only if sufficient exhaust gas temperatures are attained 
for specified minimum time periods throughout the engine's duty cycle. 
The exhaust temperatures that must be attained, and the corresponding 
DPFs, are listed in Table VII-1.

[[Page 32923]]



                        Table VII-1.--Ceramic Wall-Flow Monolith DPF Regeneration Options
----------------------------------------------------------------------------------------------------------------
                                       Temperature that
                                     exhaust must exceed
       DPF regeneration type         at least 30% of the            DPF media                   Comments
                                       time for passive
                                    regeneration to occur
----------------------------------------------------------------------------------------------------------------
Passive...........................  >550[deg]C...........  Uncatalyzed media; can be   Exhaust temperatures
                                                            either cordierite or        >550[deg]C rarely if
                                                            silicon carbide.            ever occur; thus,
                                                                                        passive regeneration of
                                                                                        uncatalyzed DPFs is not
                                                                                        a practical option.
                                    >390[deg]C...........  Base metal catalyzed        No increase in NO2.
                                                            cordierite.
                                    >340[deg]C...........  Lightly platinum-catalyzed  Special provisions must
                                                            cordierite or silicon       be made to ensure
                                                            carbide with fuel           additive is always
                                                            additive.                   present in fuel and that
                                                                                        equipment w/o DPFs
                                                                                        cannot be fueled with
                                                                                        additive-containing
                                                                                        fuel. No increase in
                                                                                        NO2.
                                    >325[deg]C...........  Platinum-catalyzed          Lab results indicate
                                                            cordierite or silicon       significant NO to NO2
                                                            carbide.                    conversion; field
                                                                                        results are mixed;
                                                                                        successful application
                                                                                        depends on consistently
                                                                                        achieving required
                                                                                        exhaust temperatures and
                                                                                        adequate ventilation to
                                                                                        dilute and carry away
                                                                                        NO2.
Active............................  Not applicable.......  Uncatalyzed cordierite or   DPFs manually regenerated
                                                            silicon carbide.            on-board or off-board
                                                                                        depending on system
                                                                                        design.
                                    Not applicable.......  Uncatalyzed silicon         Active/passive\1\ type
                                                            carbide or cordierite.      system uses fuel burner
                                                                                        to assist regeneration
                                                                                        at any exhaust gas
                                                                                        temperature and duty
                                                                                        cycle; regeneration
                                                                                        initiated automatically
                                                                                        based on exhaust
                                                                                        backpressure.
----------------------------------------------------------------------------------------------------------------
\1\ MSHA is aware of another type of active/passive system utilizing an on-board electrical heating source to
  assist regeneration of sintered metal filter media, but is not aware of any underground mining applications of
  this system at this time.

    As Table VII-1 indicates, passive DPF systems will regenerate 
successfully at or above the exhaust gas temperature specified by the 
manufacturer. However, these exhaust gas temperatures must be 
maintained for at least 30% of the shift to be sufficient for passive 
regeneration. An active regenerating system will work at any exhaust 
temperatures.
    The tune of the engine will also be a factor for proper 
regeneration. If an engine goes out of tune and begins to emit higher 
DPM concentrations in the exhaust, the exhaust backpressure may 
increase too quickly. Therefore, MSHA and DPF manufacturers recommend 
that mine operators install backpressure monitoring devices on machines 
equipped with DPFs in order to properly monitor the condition and 
regeneration state of the filter.
    In the DPM settlement agreement, MSHA agreed to a compliance 
assistance period of one year beginning July 20, 2002 and ending July 
19, 2003. Among its many compliance assistance activities during this 
period, MSHA examined the mine worthiness of available DPF systems. In 
the preamble discussion to the 2003 NPRM, MSHA stated:

MSHA has found that most mine operators can successfully resolve 
their implementation issues if they make informed decisions 
regarding filter selection, retrofitting, engine and equipment 
deployment, operations, and maintenance. The Agency recognizes that 
practical mine-worthy DPF systems for retrofitting most existing 
diesel powered equipment in underground metal and nonmetal mines are 
commercially available and are mine worthy to effectively reduce 
miners' exposures to DPM. MSHA also recognizes that installation of 
DPF systems will require mine operators to work through technical 
and operational situations unique to their specific mining 
circumstances. In view of that, MSHA has provided comprehensive 
compliance assistance to the underground metal and nonmetal mining 
industry.

    NIOSH also stated its position on the DPF systems currently 
available for most mining equipment during this period. By letter of 
June 25, 2003, to MSHA, NIOSH stated:

With regard to the availability of filters and the interim standard, 
the experience to date has shown that while diesel particulate 
filter (DPF) systems for retrofitting most existing diesel-powered 
equipment in underground metal and nonmetal mines are commercially 
available, the successful application of these systems is predicated 
on solving technical and operational issues associated with the 
circumstances unique to each mine. Operators will need to make 
informed decisions regarding filter selection, retrofitting, engine 
and equipment deployment, operation, and maintenance, and 
specifically work through issues such as in-use efficiencies, 
secondary emissions, engine backpressure, DPF regeneration, DPF 
reliability and durability. NIOSH is of the opinion that these 
issues can be solved if the informed decisions mentioned above are 
made. This view is supported by comments made by mine operators at 
the NIOSH-sponsored workshops entitled ``Diesel Emissions and 
Control Technologies in Underground Metal and Nonmetal Mines.'' 
Analysis of the recently completed Stillwater Mine experiments and 
related in-mine tests will also provide information regarding in-
mine filter efficiency performance of these systems as compared to 
their performance in the laboratory.

Assuming that the results show comparable filter efficiency 
performance, metal/nonmetal mine operators in similar circumstances 
will be able to use the information with confidence to predict 
performance results in reducing DPM levels in particular 
applications.

    MSHA believes that this document confirms that DPF systems are 
available and mine-worthy to reduce miners' exposures to DPM.
    Some commenters stated that the intermittent duty cycles (bursts of 
heavy work, followed by idle time) common for large front-end loaders 
used in the stone mining industry are unlikely to produce sufficiently 
high exhaust temperatures for passive regenerating DPFs to be a 
feasible DPM control option. MSHA notes that during its 2003 compliance 
assistance visits, exhaust temperature monitoring conducted on a 
production loader indicated sufficient temperatures for a sufficient 
portion of the duty cycle to permit that loader to utilize a passively 
regenerating DPF system. Clearly, such limited testing was not 
definitive, and the mine operator would need to conduct additional 
temperature monitoring to

[[Page 32924]]

verify these results over the complete range of work activities 
performed by this loader. However, there was nothing particularly 
unusual about this loader or its duty cycle, so the commenter's 
suggestion that loaders in the stone industry, in general, cannot 
utilize passive regenerating DPFs, is inaccurate.
    Also, MSHA notes that there are feasible alternatives to passive 
regeneration for filtering the exhaust of any size engine used in the 
stone mining industry. Mine operators could choose on-board or off-
board active regeneration, including an on-board fuel burner type 
system that actively regenerates the filter during normal production 
operations without any intervention by the equipment operator, without 
shutting down the equipment, and without any increase in NO2 
generation.
    Industry commenters related the experiences of four mining 
companies to support the position that DPF systems are not a 
technologically feasible DPM control option for attaining compliance 
with the interim DPM limit in underground mining applications. The four 
companies were the Stillwater Mining Company (Stillwater mine in 
Montana), Newmont Gold (Carlin East and Deep Post mines in Nevada), 
Kennecott Minerals (Greens Creek mine in Alaska), and Cargill Salt 
(Avery Island mine in Louisiana).
    Commenters reported that platinum wash-coated passive DPFs have 
proven successful at the Stillwater mine. They indicated that the 
equipment best suited to utilizing passive systems includes 19 primary 
haulage trucks, eight locomotives, and two large LHDs which together, 
are estimated to account for about 35% of the mine's DPM emissions. 
This equipment tends to work in haulageways where there is frequently a 
good ventilation air flow. However, as noted elsewhere in this section 
of this preamble, the commenters noted problems with high 
NO2 emissions from equipment fitted with platinum wash-
coated passive DPFs. MSHA has determined that the NO2 
problems at this mine result from inadequate ventilation, and that high 
NO2 levels at this mine pre-dated the use of platinum wash-
coated passive DPFs.
    These commenters indicated that the remaining 321 machines at this 
mine do not have high enough duty cycles and exhaust temperatures to 
utilize passive DPFs, and that active DPF systems are not considered 
feasible by the mine operator. As discussed in detail below in this 
section, MSHA believes that the mine operator's determination of 
infeasibility of active filters is based on a proposed active 
filtration concept that is not optimal for this mine.
    These same commenters also discussed the technological and economic 
feasibility analyses for the Stillwater mine included in the 31-Mine 
Study. MSHA has acknowledged that the cost estimates contained in the 
31-Mine Study final report significantly underestimate the probable DPM 
compliance costs for this mine. At the time the 31-Mine Study was 
conducted, MSHA's analysts had been supplied with inaccurate 
information regarding this mine's diesel equipment inventory. MSHA 
subsequently revised its analysis based on updated equipment inventory 
data. The revised estimate of compliance cost for the Stillwater mine 
is considerably higher than the estimate included in the 31-Mine Study. 
However, as discussed later in this section, it is nonetheless 
consistent with the estimated compliance cost for a precious metals 
mine of this size as detailed in MSHA's REA for the 2001 final rule.
    The commenters indicated that Newmont has experimented with both 
passive and active DPFs in the Carlin East and Deep Post mines, and 
that a problem exists. The commenters state that engine backpressures 
range from 37 to 43 inches of mercury when DPFs are in use, and one of 
their engine suppliers, Caterpillar, will not warrant engines when 
backpressure exceeds 27 inches of mercury. In response, MSHA references 
the NIOSH/MSHA Filter Selection Guide, which states that DPF systems 
must be sized so that backpressure is within the engine manufacturer's 
specifications.
    The commenters go on to relate Newmont's successes with DPFs, 
including both platinum wash-coated passive filters on haulage trucks 
and base metal wash-coated passive/active filters on smaller LHDs and 
jammers. Although elevated NO2 emissions can be associated 
with platinum wash-coated DPFs, the trucks equipped with these filters 
are used to haul ore up well ventilated ramps to the surface, so the 
potential for NO2 overexposure is minimized. The smaller 
LHDs and jammers are typically used in production areas with lower 
ventilation rates, so base metal wash-coated filters are used which do 
not generate NO2. Because of the limited duty cycle of these 
smaller machines, total filter regeneration may not occur. However, the 
wash-coat promotes enough regeneration that the filters are able to 
function properly between set service intervals that coincide with the 
equipment's preventive maintenance schedule, at which time the filters 
are changed-out, and the ``dirty'' filters actively regenerated off-
board.
    The commenters also related Newmont's experience with ``failed'' 
DPFs, including a filter that was destroyed due to excess vibration and 
another that was destroyed when an upstream turbocharger failed and 
blew oil into the DPF. However, the commenter went on to describe the 
steps taken by Newmont to successfully correct the vibration problem 
(shock absorbing filter mounts), and the other destroyed DPF was 
clearly caused by the failed turbocharger, not an integral failure of 
the DPF. MSHA has repeatedly advised the mining community that a 
certain amount of applications engineering will be required to insure 
the successful deployment of DPFs on underground mining equipment. The 
vibration failure example illustrates that as mine operators obtain 
experience with DPFs, problems will inevitably be encountered, but they 
can be readily solved by applying reasonably simple hardware solutions.
    These commenters also questioned MSHA's assumptions regarding the 
feasibility of auxiliary ventilation system upgrades discussed in the 
31-Mine Study, however, the upgrades specified for Carlin East in the 
31-Mine Study related to achieving the final DPM limit. Compliance with 
the interim limit was projected without ventilation upgrades.
    These commenters concluded that overall DPM compliance costs are 
too high for Newmont Gold. Newmont estimates that the, ``purchase and 
installation of DPFs, including downtime on production vehicles, will 
be $1.9 million for its two mines--Deep Post and Carlin East.'' No 
further cost breakdown is provided, so MSHA could not assess the 
reasonableness of this estimate. However, accepting this estimate as 
submitted, and assuming a two-year DPF service life, Newmont's estimate 
of its DPF costs implies a yearly cost of $1.05 million for the two 
mines ($1.9 million annualized over two years at a 7% discount rate). 
MSHA notes in the REA for the 2001 final DPM rule that its estimated 
compliance cost for a medium-sized gold mine employing 20 to 500 miners 
is $171,900 per year based on a diesel equipment fleet size of 24 
pieces of diesel equipment. This estimate was based on analysis 
indicating about 78% of overall compliance costs would relate to DPFs. 
Adjusting MSHA's estimated annual cost to correspond to the combined 
166 pieces of equipment at Newmont's two mines yields an estimated 
annual DPF-related compliance cost of about

[[Page 32925]]

$927,000, which is only 12% less than Newmont's estimate of its annual 
DPF-related compliance cost.
    The same commenters described DPF installations on haulage trucks 
and loaders equipped with Detroit Diesel Series 60 engines rated at 450 
horsepower and 350 horsepower, respectively, at the Kennecott Greens 
Creek mine in Alaska. Regarding the trucks, the same commenters 
reported that, ``After initial problems, mainly caused by incorrect 
installation and sizing of filters, the mine has successfully equipped 
its fleet of six Toro trucks with DPFs.'' This experience confirms two 
important aspects of DPF utilization that MSHA has emphasized 
repeatedly in its compliance assistance communications with the 
industry, including (1) the likely need for a certain amount of 
applications engineering to resolve implementation and installation 
issues, and (2) the need to appropriately match the DPF to the machine 
and duty cycle.
    With respect to installations on two identical Toro 1250 loaders, 
it was noted that the platinum wash-coated DPF on one unit consistently 
passively regenerated, while the DPF on the other unit, which had a 
lesser duty cycle and exhaust temperatures that were 40 to 50[deg]C 
lower, did not. This experience does not illustrate the failure of DPF 
technology. Rather, it confirms MSHA's consistent advice that the 
successful deployment of passively regenerating DPFs requires careful 
determination of exhaust temperatures to assess whether passive 
regeneration is feasible for that particular machine and in that 
application. Indeed, in this example, the filter functioned precisely 
as designed. The failure of the filter to passively regenerate on the 
second machine could have been reliably predicted based on the exhaust 
temperature data.
    In their comments, industry also relates Greens Creek's successful 
application of an active DPF system on an Elphinstone R1300 3\1/2\-yd 
LHD with a Cat engine. This loader is used for relatively light duty 
clean up work, and is therefore not a suitable candidate for 
application of a passively regenerating DPF.
    It should be noted that industry also commented that, ``Those 
engines in the 250-350 horsepower, and greater-than 350 horsepower 
ranges are considered unsuitable for DPFs with present technology. This 
general conclusion of unsuitability for DPF usage for these large 
engines comes from use of DPFs in real mine situations.'' These 
statements are directly contradicted by Greens Creek's successful 
experience filtering the exhaust from 350 horsepower and 475 horsepower 
engines.
    Industry also presented the experience of Cargill Salt's Avery 
Island mine in Louisiana which installed two DCL Mine X DPF filters on 
a Cat 992G loader equipped with a Cat 3412 engine rated at 650 
horsepower. One 15 inch diameter by 15 inch long filter was connected 
to each bank of the V-12 engine. This model DPF is wash-coated with a 
platinum catalyst to facilitate passive regeneration. The mine reported 
that there are no problems with elevated NO2 levels, and 
visible emissions have been reduced. However, the mine also reported 
that the loader has lost almost all of its power, to such an extent 
that the loader is only used for clean-up duty.
    These symptoms--no elevated NO2 levels, visible 
emissions reduced, and loss of power--are all typical of a mismatch 
between the duty cycle of the application and the performance 
specifications of the DPF. In order to passively regenerate, this DPF 
requires exhaust temperatures of about 325[deg]C or higher for at least 
30% of its duty cycle. An insufficiently demanding duty cycle produces 
lower exhaust temperatures which are not sufficient to ignite and burn 
off accumulated DPM. Such a filter continues to collect DPM, resulting 
in lower visible emissions, but as the filter loads, even for a single 
work shift, backpressure on the engine increases, resulting in loss of 
power. Although these commenters report that mine mechanics worked 
closely with the local Caterpillar dealer in installing the system, it 
is very likely that this experience illustrates an inappropriate DPF 
application rather than a failed filter system.
    Normally, the local Caterpillar dealers and any other engine 
manufacturer's dealers work more with issues concerning the engine 
installation and repairs than with DPM filter applications. Since 
engine manufacturers at this time do not install a DPF to the engine at 
the time of engine production, the local engine dealers are not usually 
familiar with DPF systems that are installed as retrofits on the 
engine.
    However, even in the case of the Greens Creek experience, where the 
mine operator worked with the engine manufacturer, the vehicle 
manufacturer, and the filter manufacturer at the onset to incorporate a 
DPF on a new machine, the mine still initially had a failure of the DPF 
because of regeneration issues. As Greens Creek reported,

the unit (DPF) was used on a waste rock backhaul route, with loads 
being carried down the ramp or on relatively flat hauls. Had the 
unit been used for ore haulage uphill routes, it would have achieved 
the high exhaust temperatures for the designed passive regeneration.

This mine's experience continues to emphasize that the mine must 
understand the duty cycle of the machine to which the DPF is being 
equipped to see if the duty cycle can support the regeneration needed 
for the DPF. In the case of Greens Creek, the waste rock backhaul 
vehicle did not have a sufficiently demanding duty cycle to generate 
the exhaust gas temperature needed for regeneration for a passive 
regeneration system. In such instances, the mine operator needs to go 
to another method of regeneration for the vehicle's DPF as discussed 
elsewhere in this preamble. Mine operators should also refer to the M/
NM Filter Selection Guide on MSHA's Web site for assistance in choosing 
the appropriate DPF system for its particular circumstances.
    Industry also discussed various issues relating to compliance 
problems for stone mines, such as feasibility of filters for large 
engines, biodiesel fuel, and ventilation. These issues are addressed 
elsewhere in this preamble in sections that deal specifically with 
these topics. Some commenters stated that MSHA presumed that operators 
would retrofit DPFs on existing diesel-powered equipment as the primary 
method of compliance. These commenters questioned whether 
implementation issues with retrofitting and regeneration would make 
DPFs infeasible. In response, MSHA has determined on the basis of in-
mine tests conducted by NIOSH, MSHA, individual mining companies and 
others, and on the experiences of mining companies that have 
implemented DPM filtration on a routine production basis, that DPFs are 
a practical, mine-worthy, and effective means for reducing exposure to 
DPM in underground M/NM mines. Further, MSHA has determined that use of 
DPFs independently or in conjunction with other feasible and effective 
DPM engineering and administrative controls will enable most mine 
operators to attain compliance with the DPM interim limit. However, 
MSHA agrees with the commenters that implementation issues with 
retrofitting and regeneration may present compliance difficulties for 
some mines, and additional time may be required at some mines due to 
the cost of purchasing and installing controls.
    Many commenters have cited problems with DPFs which they believe 
support the contention that DPFs are neither technologically nor 
economically feasible. As noted above,

[[Page 32926]]

some commenters provided examples from several underground mines that 
experienced failed DPFs. Commenters indicated that a ceramic filter, 
using passive type regeneration would be the only type filter that 
would be acceptable to them. Commenters also stated that ceramic DPFs 
that require active regeneration, a fuel borne catalyst, a catalyst 
that could have the potential to increase NO2 emissions, and 
any kind of filter for engines less than 50 horsepower or greater than 
250 horsepower were infeasible for use in underground M/NM mines. Some 
commenters described installations that produced high exhaust 
backpressure on engines that could lead to voiding engine warranties or 
render a vehicle unusable. A commenter also stated that the number of 
regeneration stations that would be required to be built and maintained 
would make active regeneration infeasible.
    Other commenters stated that when DPFs are appropriately sized and 
fitted to equipment, and there is a good match between the equipment 
application/duty cycle and the DPF regeneration method, long filter 
life and significant DPM reductions will result. Several commenters 
indicated that, after an initial trial-and-error ``learning period,'' 
they had experienced success with passive type DPFs and were using them 
on a routine production basis.
    Some commenters stated that DPFs continue to be a feasible 
technology for significantly reducing DPM exposures. One commenter 
reported the successful application of an on-board active regeneration 
DPF. This system includes an exhaust backpressure monitor that warns 
the equipment operator when DPF regeneration is required. This is a 
feature MSHA recommends for all DPF installations.
    As noted above, MSHA acknowledges the numerous documented examples 
of failed DPF applications in the underground M/NM mining industry. 
However, MSHA believes such failures are the result of inappropriate 
filter selection, manufacturing defects, and unrelated failures of 
equipment components (such as turbochargers) that have caused damage to 
DPFs. MSHA is confident that proper filter selection will result in 
satisfactory long term DPF performance, and NIOSH agrees with MSHA that 
DPFs are technologically feasible for most mining equipment after some 
technical and operational problems are solved, and that these problems 
can be solved in most cases.
    To help mine operators avoid having to rely on costly and time 
consuming trial-and-error methods for DPF selection, the Filter 
Selection Guide was developed. It is the result of a joint effort of 
MSHA and the Diesel Team from the NIOSH Pittsburgh Research Laboratory. 
The Filter Guide provides mine operators with information on feasible 
and available DPFs. NIOSH will work with MSHA to maintain the Filter 
Guide on the internet.
    MSHA continues to urge mine operators to thoroughly evaluate each 
application to insure that the appropriate DPF and regeneration system 
is chosen. Such an evaluation is well within the technical capabilities 
of most mine operators to perform. For the few operators that would be 
unable to independently perform this evaluation, technical assistance 
can be obtained from mining equipment manufacturers, engine 
manufacturers, DPF manufacturers, and MSHA.
    As noted earlier, selection of an appropriate DPF for a given 
application requires consideration of such factors as engine type, 
model, and horsepower, as well as the intended usage of the equipment 
and related equipment duty cycles. Mine operators are fully capable of 
obtaining this information for every piece of equipment that is a 
candidate for DPF installation. In addition, the engine's DPM emission 
rate and exhaust temperatures must be obtained. For MSHA-approved 
engines, DPM emission rates are determined by MSHA and included with 
the engine approval. For non-approved engines, DPM emission information 
can be obtained from the engine manufacturer or estimated based on the 
characteristics of the engine (direct injection, pre-chamber, make and 
model, model year, naturally aspirated, turbocharged, electronically 
controlled, etc.). To obtain exhaust temperatures, various inexpensive 
(approximately $200) data logging thermocouple systems are commercially 
available that can be attached to the exhaust system to provide 
detailed exhaust temperature profiles over time periods ranging from 
several hours to several shifts. During its compliance assistance mine 
visits in the spring and summer of 2003, MSHA noted that several mine 
operators had acquired exhaust temperature data logging systems and 
were using them to systematically measure exhaust temperatures on 
equipment that might need to be equipped with a DPF in the future.
    DPFs collect significant amounts of DPM from the engine's exhaust, 
thus lowering DPM exposures. This fact was not disputed by the 
commenters. The results from MSHA's compliance assistance work with 
Kennecott at their Greens Creek Mine, NIOSH's isolated zone tests 
conducted at the Stillwater Mine, NIOSH's production zone tests at the 
Stillwater mine, MSHA's laboratory data, laboratory and in-mine test 
results from Canadian and European studies, and various other industry 
applications prove that DPFs provide high efficiency reductions in both 
DPM and EC. For EC, the data indicate filtration efficiencies as high 
as 90% to 99+%.
    MSHA disputes commenters' views that if passive regeneration cannot 
be successfully employed (due, for example, to an insufficient duty 
cycle and correspondingly low engine exhaust temperatures), then DPM 
filter technology is infeasible. Passive regeneration is only one of 
many regeneration schemes available to the mine operator. Clearly, not 
all machines or all applications are suitable for passive regeneration. 
One commenter stated that one of his firm's two loaders was able to use 
a passive regeneration DPF due to the exhaust gas temperatures reached 
during its duty cycle, while the other could not or was marginal. This 
experience demonstrates precisely what MSHA's consistent message to the 
industry has been--that successful application of passive regeneration 
DPFs depends on matching the filter to the application, and mine-worthy 
systems are commercially available for most any machine and any duty 
cycle.
    It is important to note that a sufficiently heavy duty cycle does 
not, by itself, guarantee that a passive regeneration DPF will function 
properly and provide satisfactory long-term performance. It is an 
essential prerequisite, but the other steps in the DPF selection 
process must also be followed rigorously. Without the necessary exhaust 
temperatures for the specified amount of time, passive regeneration is 
impossible, regardless of how carefully the other steps in the 
selection process are followed. However, once the necessary exhaust 
temperature profile has been verified through sufficient in-mine 
temperature monitoring, users are urged to carefully complete the 
remaining steps in the selection process.
    For whatever reason, if a particular machine requires a DPF, but is 
an unsuitable candidate for application of a passive regeneration 
system, the mine operator has the option of using a combination 
passive/active regeneration scheme or to use a purely active 
regeneration system. Because the option exists for utilizing either 
passive, active/passive, or active regeneration systems, MSHA maintains 
that a suitable DPF system is available for any size diesel engine and 
any application in the underground M/NM mining industry. The mine 
operator may need to address

[[Page 32927]]

various implementation issues regarding retrofitting and regeneration, 
but MSHA is confident these issues can be resolved.
    NIOSH's Phase I Isozone and Phase II Production Zone Studies 
Related to DPFs at the Stillwater Mine. NIOSH conducted a series of in-
mine tests on DPF systems at the Stillwater Mining Company's 
underground platinum mine at Nye, MT. The tests were conducted in two 
phases. The Phase I tests were conducted from May 19-30, 2003, and the 
Phase II tests were conducted from September 8-12, 2003. The purpose of 
Phase I was to assess the effectiveness of DPM control technologies in 
an isolated zone. The purpose of Phase II was to assess the capability 
of DPFs to effectively control the exposure of underground miners to 
DPM in actual in-mine production mining scenarios.
    NIOSH issued two final reports on these studies. The final report 
for Phase I was entitled ``Effectiveness of Selected Technologies in 
Controlling Diesel Emissions in an Underground Mine--Isolated Zone 
Study At Stillwater Mining Company's Nye Mine,'' and the report was 
released on January 5, 2004. NIOSH included the following in its 
discussion of the objective of the study:

    The objective of this study was to determine the in-situ 
effectiveness of the selected technologies available to the 
underground mining industry for reducing particulate matter and 
gaseous emissions from diesel-powered equipment. The protocol was 
established to determine the effectiveness of those technologies in 
an underground environment under operating conditions that closely 
resemble actual production scenarios.
    The study was designed to provide Stillwater, and the general 
mining community, with better insights into the performance of 
control technologies and enable them to identify the appropriate 
devices for reducing diesel emissions. The focus of the Stillwater 
research was on technologies that offer solutions for reducing DPM 
emissions. This report provides the results and assessment of the 
following control technologies: diesel particulate DPFs, disposable 
paper DPFs, diesel oxidation catalytic converter, and reformulated 
fuels.

    The Phase II final report was entitled, ``An Evaluation of the 
Effects of Diesel Particulate Filter Systems on Air Quality and 
Personal Exposures of Miners at Stillwater Mine Case Study: Production 
Zone,'' and the report was released April 1, 2004. The objective of 
Phase II was to determine the effects of DPF systems installed on 
production equipment at the Stillwater Mine on workplace concentrations 
of EC and regulated gases in an actual production mining application 
where multiple diesel-powered vehicles operated simultaneously during 
full shift mining activities. The effects of DPF systems were examined 
by comparing ambient concentrations of EC, CO, CO2, NO, and 
NO2 in a production area for two different test conditions. 
For the baseline condition, all vehicles that operated within the 
ventilation split were equipped with standard exhaust systems--a diesel 
oxidation catalyst (DOC) and muffler--but without DPFs. For the second 
condition, three of the vehicles, an LHD and two haulage trucks had 
their DOC and muffler systems replaced with DPF systems.
    The NIOSH Phase II study conducted at the Stillwater Mine is 
similar to the in-mine tests conducted by MSHA in January 2003 as a 
part of its compliance assistance program at the Kennecott Greens Creek 
Mine near Juneau, AK, which is discussed elsewhere in this preamble.
    NIOSH Phase I study. The majority of the control devices tested 
were DPFs. Phase I also tested biodiesel fuel and the differences 
between 1 diesel fuel (D1) and 2 diesel fuel (D2). 
DPFs included both ceramic and high temperature disposable (synthetic 
media) filters. NIOSH reported that some problems did occur during the 
tests, mainly dealing with ventilation issues in the isolated zone and 
an occasional vehicle passing nearby the intake to the isolated zone. 
However, these problems were minor and did not compromise most tests.
    As reported, NIOSH chose to normalize the data based on MSHA's 
nameplate gaseous ventilation rates. One commenter stated that he 
understood why NIOSH normalized the Phase I data to the MSHA nameplate, 
however, the commenter felt this was a disservice to the miners since 
M/NM mines do not have to comply with the ventilation rates on the 
approval plates. Indeed, engines in M/NM mines are not required to be 
MSHA approved and ventilation rates are not available for non-MSHA 
approved engines. MSHA agrees with the commenter that the Phase I 
report had the correct intent to normalize the data for reporting 
purposes. MSHA also agrees that the results may not be typical for 
operations in the M/NM sector because the ventilation schemes used by 
many M/NM mines do not comply with approval plate quantities for MSHA 
approved engines.
    The Phase I report shows that the EC reduction in the isolated zone 
with one system was 88%, and that two other systems gave greater than 
96% EC reductions when the measured concentrations were normalized by 
ventilation rate. NIOSH reported that several tests were discarded and 
not reported due to unexplainably low CO2 concentrations 
found at low ventilation rates.
    The filter media used in all the DPF systems during the Phase I 
test was either Cordierite, Silicon Carbide, or the disposable high 
temperature synthetic material. (An analysis conducted by an MSHA 
contracted laboratory indicated the synthetic material is fiberglass.) 
All the DPF media have very similar efficiencies for EC reductions. 
Even though NIOSH did not report the EC reduction efficiencies of all 
the DPF systems tested in Phase I, MSHA believes, based on its own 
evaluations, that the efficiencies for EC reductions of those DPFs not 
reported would have been approximately equal to the results obtained 
for DPF systems that were reported.
    Many commenters agreed that the Phase I study accomplished its 
objective by showing that DPM filters are viable for reducing DPM from 
diesel engines and that the filter systems performed as designed. 
However, some of these commenters stated that the elaborate test setup 
in the Phase I study was only a replication of a laboratory type 
environment that did not represent actual mine conditions. Commenters 
pointed out that some of the control technologies did not perform as 
well as expected during the study.
    MSHA agrees that the Phase I study demonstrated that DPM filters 
are an effective tool for reducing DPM emitted from diesel engines. The 
Phase I study did involve an elaborate test setup, but this test setup 
was primarily aimed at controlling the ventilation conditions so that 
extraneous DPM from upstream diesel traffic would be eliminated, 
thereby enabling a meaningful and accurate determination of the DPM 
reductions obtained by the various DPFs tested. In other respects, 
however, the test setup was quite realistic, in that the testing 
occurred underground and involved a realistic simulation of a 
production mining operation. For example, in testing of LHDs, the test 
protocol required a production LHD to repeatedly follow a proscribed 
duty cycle involving loading at a muckpile, tramming up a 9% grade 
along the main haulageway a distance of approximately 1,000 feet with a 
loaded bucket, various forward and reverse maneuvers over short travel 
distances at each end of the haulageway, and raising and lowering a 
loaded bucket to simulate loading a haulage truck. Other than the 
removal of existing exhaust system components (DOC and muffler) to 
accommodate installing the subject DPFs, and the installation of 
certain monitoring instrumentation, the equipment used in

[[Page 32928]]

the study was unmodified and in ``as is'' condition from the mine's 
equipment inventory. Although this testing was based on simulated 
mining operations, the suggestion that it replicates a laboratory 
environment is an inaccurate characterization.
    MSHA believes that the Phase I Isozone data is sound science, 
establishing with certainty that DPFs can be implemented on a broad 
scale in mines in the U.S. and that DPFs are capable of achieving 
significant reductions in miner's DPM exposures. MSHA notes that these 
data are consistent with the results of other similar tests, including 
both laboratory tests conducted by MSHA, NIOSH and others, and a 
Canadian in-mine isolated zone test in which NIOSH also participated. 
MSHA discussed the results of this Canadian test in the preamble to the 
2001 final rule.
    One commenter stated that the Phase I isolated zone test should 
have been completed long before the DPM rule was rushed to publication. 
MSHA does not agree with the commenter. In fact, MSHA used the results 
of the above mentioned Canadian isolated zone study in its original 
2001 DPM rule to show the effectiveness of DPFs. The recent NIOSH 
isolated zone testing confirmed the results obtained by the Canadians. 
As noted above, the pertinent data that were derived from the Canadian 
study on the efficiencies of DPFs were referenced in the preamble to 
the 2001 final rule.
    At the end of the Phase I report, NIOSH indicated that the 
Stillwater mine had at that time over one dozen DPFs in use for a 
combined total of over 22,000 operating hours. NIOSH reported that only 
one of these DPFs had failed (runaway regeneration), and that the other 
systems have been virtually maintenance free. Again, even though 
Stillwater's experiences with DPFs on a routine production mining basis 
have been with heavily platinum-catalyzed passive systems, the 
commercially available DPF media are the same for passive systems using 
other catalyst wash coats as well as for active regeneration systems 
that utilize uncatalyzed filter media. Moreover, all DPF media 
basically provide equivalent filtration efficiencies for DPM, TC, and 
EC.
    NIOSH Phase II study. The Phase II study confirmed and expanded on 
the results obtained in the Phase I study. In the final report, NIOSH 
indicated that greater EC reductions were observed in the field than 
were obtained in the laboratory for whole diesel particulate:

    * * * laboratory determination of DPF efficiencies, based on 
reductions in total DPM mass (fairly equivalent to TPM [Total 
Particulate Matter]), substantially underestimates the ability of 
DPF systems to reduce EC emissions, the metric used by MSHA for 
compliance,* * *

which highlights the high EC filtration efficiency for DPFs.
    MSHA believes that the Phase II study helped to confirm existing 
agency data that shows that it is technologically feasible to reduce 
miners' exposures to DPM to the 308EC [mu]g/m\3\ interim 
PEL. The Phase II study utilized three machines (1 LHD and 2 Haul 
Trucks) equipped for the first three days with highly platinum-
catalyzed Englehard DPX[supreg] DPFs, and the last day without the 
DPFs, but with DOCs. The equipment engaged in normal production 
activities in a typical production mining area of the Stillwater mine, 
as opposed to the simulated mining tasks that were conducted in an 
isolated zone in the Phase I study. Personal sampling on equipment 
operators was conducted, as well as area sampling upstream and 
downstream from the working area where the equipment was operating. 
Tests were conducted with and without DPFs installed so that the 
capability of the DPFs to reduce personal DPM exposures and DPM levels 
in the ambient mine air could be quantified.
    The results of the personal EC samples from the three machine 
operators equipped with filters were provided in the final report. 
NIOSH did not report Day 1 results due to inadequate sampling 
locations. The EC results for personal samples for Day 3 showed that 
the DPM exposures of all three miners were well below 308EC 
[mu]g/m\3\, and in fact, well below 160EC [mu]g/m\3\. Day 2 showed 
exposures also below 308EC [mu]g/m\3\, but almost double the 
results of Day 3. However, it appears that the ventilation air flow 
through the working area on Day 2 was about half the ventilation air 
flow for Day 3. Thus, the differences in measured DPM levels are not 
contradictory, but rather, demonstrate the effectiveness of increased 
ventilation flow as an engineering control to reduce DPM levels in the 
ambient air. The EC reduction efficiencies of the DPFs based on 
personal exposures comparing test days with and without the filters in 
place were approximately 71% for the LHD operator and 78% for the haul 
truck drivers. These reductions are very similar to the results 
obtained for personal exposures in the Greens Creek study conducted by 
MSHA in January 2003.
    NIOSH reported that some of the filters used during the Phase II 
testing at Stillwater may have been compromised. However, NIOSH 
indicated in the Phase II final report that, ``* * * even when the DPF 
systems are performing below expectations, they can significantly 
reduce the EC concentrations when compared to conditions when DPF 
systems were not used.'' Significantly, MSHA made a very similar 
observation in its report on Greens Creek. During testing at Greens 
Creek, there were obvious visible cracks in some of the ceramic media. 
But analysis of DPM concentrations in the equipment exhaust indicated 
that EC filtration efficiency was still quite high (>90%) despite the 
cracks. Clearly, even compromised DPM filters can reduce personal DPM 
exposures to levels below the interim PEL.
    NIOSH reported increased NO2 concentrations during the 
study when using DPFs, and suggested that the source of the increase 
was the platinum catalyst used as a wash coat for the Cordierite filter 
media. The platinum wash coat on the filter is used for regeneration 
purposes and does not affect filter efficiency for EC measurements. 
Therefore, the reduction observed in EC concentrations from the Phase 
II study should be expected when any filter is installed that has a 
Cordierite filter media. As discussed elsewhere in this preamble, a 
Silicon Carbide filter media is also used in many DPF systems and EC 
filtering efficiency for Silicon Carbide is very similar to Cordierite.
    As noted above, NIOSH reported increases in NO2 
concentrations when highly platinum-catalyzed DPFs were used. NIOSH 
stated in the Phase I final report that ``* * * if the required MSHA 
ventilation rates were maintained during the tests, the average 
concentration of NO2 over the test periods would have not 
exceeded 3 ppm, the long term exposure limit for NO2.'' The 
greatest increase in NO2 during the Phase I study came from 
the highly platinum-catalyzed DPF. When this filter was used, the 
ceiling limit of 5 ppm was briefly exceeded each time the equipment 
repeated the duty cycle. These NO2 peaks were noted at the 
downstream sampling location and at about the same levels at a sampling 
location on the equipment near the operator's position.
    NIOSH stated in the Phase II report that tests 2 and 3 (with DPF 
installed) were terminated when the multi-gas monitor carried by the 
equipment operator indicated that the 5 ppm NO2 ceiling 
limit had been exceeded. NIOSH

[[Page 32929]]

reported that they also believe the NO2 level may have been 
above 5 ppm for personal exposure on test 4 when the DPFs were not 
installed on the machines (DOCs were installed on test 4).
    Although tests 2 and 3 were terminated earlier than planned, these 
tests lasted between approximately 2\3/4\ hours and 4\3/4\ hours, 
respectively. MSHA believes that these tests were sufficient in 
duration to demonstrate the differences in EC exposures with and 
without DPFs. At most mines, mucking operations in an individual stope 
or development end are usually completed within 2-4 hours. In fact, the 
Greens Creek report results were based on approximately 2-3 hours of 
sample time, which was the total time required to muck out the subject 
stopes.
    From the intake side to the return side of the Phase II test zone, 
average NO2 increase as reported were 1.2 ppm for Day 2, and 
1.1 ppm for Day 3 with DPFs. The average NO2 increase was 
1.1ppm for Day 4 with DOCs. It is significant to note that these 
increases are consistent with the NO2 increases observed 
during the Greens Creek tests, and would not be expected to result in 
hazardous NO2 exposures in mines with adequate ventilation. 
It should also be noted that there was no significant difference 
between average NO2 increases with and without DPFs in the 
test area (the DPFs were replaced by DOCs on Day 4).
    As stated above, NIOSH noted that Phase II tests 2 and 3 were 
terminated early due to excessive NO2 levels measured in the 
cabs of the test equipment. Due to the layout of the area where Phase 
II tests were conducted, it is likely that the vehicles experiencing 
the highest NO2 levels were operated for part of the duty 
cycle in a lower quantity of ventilation air than was available in the 
main haulageway. The observed personal overexposures to NO2 
occurred when the haul trucks were in this poorly ventilated area where 
the intake air split at an orepass and a development section. MSHA 
believes that if the air flows to these locations had been maintained 
at levels near the nameplate value, the overexposure to NO2 
would very likely not have occurred.
    It should be noted that MSHA has documented very low ventilation 
air flows in several stopes at the mine where NIOSH's Phase II study 
was conducted. Ventilation measurements obtained by MSHA during a 
compliance assistance visit to the mine in June 2004 identified 
significant leakages from most of the auxiliary stope ventilation 
systems that were evaluated. In the six stopes for which ventilation 
air flow measurements could be obtained at both the auxiliary fan 
location and at the end of the vent bag, the average air flow at the 
fan location was 24,400 cfm and the average flow at the end of the vent 
bag was 5,100 cfm. In one stope, auxiliary ventilation system leakage 
was 89% and in another, leakage was 85%. Even in stopes where auxiliary 
system leakage was relatively low, significant recirculation was 
observed. With stope ventilation flow rates compromised to this extent 
due to auxiliary system leakage and recirculation, it is not surprising 
that both high gaseous emission levels and high DPM emissions have been 
measured at this mine.
    The NIOSH Phase II data show that gaseous contaminant levels and 
ventilation flows had stabilized in the test area a short time after 
the testing was initiated (within approximately the first 30 minutes), 
indicating that roughly steady-state conditions had been achieved. If 
tests 2 and 3 had not been terminated prematurely (i.e., if the poorly 
ventilated area had been sufficiently ventilated), it is therefore 
likely that the reported DPM and gaseous emission levels could have 
been maintained indefinitely, or at least until mining operations were 
completed in the test area.
    As stated earlier, MSHA advised mine operators through the issuance 
of a PIB that the use of highly platinum-catalyzed DPFs has the 
potential to increase concentrations of NO2. The increases 
in NO2 observed during the Stillwater Phase I and Phase II 
tests demonstrate that mine operators who choose to use highly 
platinum-catalyzed DPFs must maintain sufficient ventilation in areas 
where the machines operate, and must monitor for any increases in 
NO2. This advice is particularly important for mines that 
had experienced NO2 problems prior to the introduction of 
platinum wash-coated DPFs, as was the case at the Stillwater mine. 
Where NO2 levels cannot be adequately controlled by 
ventilation, alternatives to highly platinum-catalyzed passive filter 
systems are commercially available which do not increase ambient 
NO2 levels. An example that is particularly well suited to 
heavy duty applications is the fuel burner type active regenerating 
DPF. A system of this type is currently installed and under evaluation 
at the Stillwater mine.
    The results of these studies support MSHA's position that feasible 
control technology exists that is commercially available to effectively 
reduce miner exposures to DPM. As with any new mining machinery, mine 
operators will need to thoroughly evaluate their needs prior to 
ordering DPF systems to insure that each system is appropriate to the 
piece of equipment, engine, application, and duty cycle. Failure to 
appropriately consider these factors will likely result in poor filter 
performance, poor engine performance, possible engine and filter 
damage, or all of the above. Alluding to this issue, NIOSH states in 
the Phase II study final report that, ``Due to the nature of the study, 
Phase II did not address other and no less important matters relating 
to the application of control technologies in underground mines. These 
matters include selection of DPF regeneration strategies, economic, 
logistical, and technical feasibility of implementation of various DPF 
systems on mining vehicles, and the reliability and durability of the 
systems in mine settings.''
    MSHA has consistently stated that the application of commercially 
available DPF systems is a task that requires mines to evaluate machine 
installations on a case by case and application by application basis. 
NIOSH agrees. Consequently, NIOSH and MSHA jointly developed an on-line 
Internet-based Filter Selection Guide which is discussed elsewhere in 
this preamble. NIOSH's written response to MSHA in this rulemaking 
supports the use of DPFs as a control device that can significantly 
reduce DPM exposures, but also states that the mine operator must 
evaluate each machine prior to selection and installation of DPM filter 
systems to insure a successful match between filter and application. 
When properly selected and installed for an application, DPFs are both 
durable and mine worthy. Almost without exception, failed DPFs that 
have been reported to MSHA were the result of inappropriate filter 
selection, manufacturer defect, or the failure of an unrelated 
component (usually the turbocharger) that affected the DPF.
    Active Regeneration DPFs. The active regeneration systems discussed 
below are normally not catalyzed so they do not produce an increase in 
NO2.

[[Page 32930]]



                                Table VII-2.--Scenarios for Active Regeneration.
----------------------------------------------------------------------------------------------------------------
                                                                  Regenerating
            System name               Regenerating location   controller  location             Comments
----------------------------------------------------------------------------------------------------------------
On-board...........................  On Equipment..........  On Equipment..........  Requires on-board source of
                                                                                      electric power.
On-board...........................  On Equipment..........  Designated and fixed-   Requires equipment to come
                                                              location.               to a specific regeneration
                                                                                      site.
Off-board..........................  Off equipment.........  Fixed-location........  DPFs are exchanged and must
                                                                                      be small enough to be
                                                                                      handled by one person.
                                                                                      Increases number of DPFs
                                                                                      needed.
On-board...........................  On-equipment..........  On-equipment during     System is complex yet fuel
                                                              operation.              burner provides advantage
                                                                                      of regeneration during
                                                                                      equipment use.
----------------------------------------------------------------------------------------------------------------

    Scenarios for active regeneration systems are listed in Table VII-
2. The second system listed in Table VII-2 is an on-board active system 
that requires about one to two hours of machine down time for 
regeneration, which might be available between shifts at some mines. To 
regenerate these filters, the piece of equipment must be parked at a 
designated location during the regeneration period so that the filter 
can be connected to electrical power and compressed air. MSHA 
recognizes that presently in some mines, production equipment is not 
necessarily brought to a central location at the end of each shift. At 
such mines, operators may need to make operational changes to 
accommodate such DPF regeneration designs.
    Alternatively, mine operators may choose off-board active 
regeneration type filters, wherein, for example, the equipment operator 
removes the DPF at the end of the shift and brings it to a central 
station for regeneration. The next operator of that piece of equipment 
takes a regenerated DPF to the equipment at the start of the next 
shift. This system enables uninterrupted equipment operation, and does 
not require the equipment to travel to a central location for filter 
regeneration at the end of the shift. Where active off-board filters 
are used, the size and weight of the filter element is a significant 
factor in filter selection and overall system feasibility, as mine 
personnel need to be capable of removing the filter at the end of the 
shift and transporting it to a central regeneration station. Multiple 
DPFs may be installed on a machine in place of a single large filter in 
order to decrease the size and weight of individual DPFs.
    Engine malfunctions and effects on DPF. Normally in mining, engine 
malfunctions are indicated by excessively smoky exhaust. That indicator 
will not occur when a DPF system is installed. Malfunctions such as 
excessive soot emissions, intake air restriction, fouled injector, and 
over-fueling, may result in an abnormal rise in back pressure in 
systems that do not spontaneously regenerate. Also, these conditions 
could lead to abnormal changes in back pressure in passive systems 
because the malfunction may raise exhaust temperatures causing the 
excess soot to be burned off. These malfunctions may be detected during 
the usual 250-hour maintenance and emissions checks conducted upstream 
of the DPF using carbon monoxide (CO) as an indicator. The other major 
filter malfunction is excessive oil consumption that is sometimes 
associated with blue smoke that could be masked by the performance of 
the DPF. However, excessive oil consumption leads to a rapid increase 
in baseline backpressure due to ash accumulation. Excessive oil 
consumption can be detected if records are kept on oil usage.
    Detecting malfunctioning DPF. As noted above, the DPF can be 
damaged mainly by thermal events such as thermal runaway. Shock, 
vibration, or improper ``canning'' of the filter element in the DPF can 
also lead to leaks around the filter element. A Bacharach/Bosch smoke 
spot test can be used to verify the integrity of a DPF. Smoke spot 
numbers below ``1'' indicate a good filter; smoke numbers above ``2'' 
indicate that the DPF may be cracked or leaking. Smoke spot and CO 
tests during routine 250 hour preventative maintenance are good 
diagnostic practices. Note that although a smoke spot number above 
``2'' may indicate a cracked or leaking filter, such a result does not 
necessarily mean the filter has ``failed'' and is not functioning 
adequately. In MSHA evaluations of DPF performance at the Greens Creek 
mine, filters that tested with smoke numbers above ``2'' of 7 were 
still shown to be over 90% effective in capturing EC, based on 
subsequent NIOSH 5040 analysis of the smoke spot filters.
    Low DPM-Emitting Engines. Through its 2003 and 2004 compliance 
assistance mine visits and a review of its nation-wide inventory of 
diesel engines used in underground M/NM mines, MSHA has determined that 
hundreds of low DPM emission engines have been introduced into 
underground M/NM mines in recent years. MSHA notes that, for many mines 
in the stone sector, use of low emission engines has been one of the 
primary means of achieving compliance with the interim PEL.
    EPA and European on-highway and non-road engine emission standards 
have forced engine manufacturers to reduce both DPM and gaseous 
emissions from their engines. Mine operators can purchase newer design 
engines with low DPM emissions in their new diesel-powered equipment as 
well as retrofitting such engines in their older equipment.
    As noted earlier in this section of the preamble, the amount of DPM 
reduction that can be obtained by switching to low DPM emitting engines 
depends on the emission rate of the original engine compared to the 
emission rate of the replacement engine. For example, if the original 
engine emits 1.0 gram of DPM per horsepower per hour of operation, and 
the replacement engine emits 0.2 grams of DPM per horsepower per hour 
of operation, the engine replacement would achieve an 80% reduction in 
emitted DPM. Other benefits of newer technology engines include better 
fuel economy and more efficient maintenance diagnostics. The improved 
maintenance diagnostics associated with electronic engine monitoring 
systems enable lower overall equipment operating costs as well as 
allowing mine operators to better monitor their engines and provide the 
appropriate maintenance to keep exhaust emissions as low as possible.
    During the compliance assistance visits to mines that had at least 
one baseline DPM sample result exceeding the interim DPM limit, MSHA 
observed numerous new or nearly new pieces of equipment powered by 
Original Equipment Manufacturer (OEM)-installed MSHA-Approved engines 
that had very high DPM emissions. The operators at these mines 
indicated that they were unaware of the DPM

[[Page 32931]]

emissions of the engines that were supplied in the equipment they had 
just purchased. They believed that by specifying an MSHA-Approved 
engine, they would be in full compliance with the rule. While it is 
true that MSHA-Approved engines satisfy the requirements of Sec.  
57.5067, not all MSHA-Approved engines are necessarily low in DPM 
emissions. Non-Approved EPA Tier 1 (for engines less than 50 horsepower 
or 175 horsepower and greater) and Tier 2 (for engines of 50 horsepower 
or greater, but less than 175 horsepower) engines are also compliant 
with Sec.  56.6067, but they have lower DPM emissions. During the 
compliance assistance visits, and in subsequent discussions with the 
Equipment Manufacturer's Association (EMA), MSHA emphasized the need 
for modern low DPM emission engines to be installed in new machines 
earmarked for the underground mining industry.
    Ventilation Upgrades. Several commenters expressed the view that 
ventilation system upgrades, though potentially effective in principle, 
would be infeasible to implement for many mines. Specific problems that 
could prevent mines from increasing ventilation system capacity include 
inherent mine design geometry and configurations (drift size and 
shape), space limitations, and other external prohibitions, as well as 
economic considerations.
    MSHA acknowledges that ventilation system upgrades may not be the 
most cost effective DPM control for many mines, and for others, 
ventilation upgrades may be entirely impractical. However, at many 
other mines, perhaps the majority of mines affected by this rule, 
ventilation improvements would be an attractive DPM control option, 
either implemented by itself or in combination with other controls.
    Indeed, MSHA observed during its DPM compliance assistance visits 
that ventilation upgrades have been implemented at many mines in the 
stone sector for DPM control, directly contradicting the commenters' 
assertion that ventilation upgrades are infeasible. Nearly every stone 
mine visited by MSHA had completed, had begun, or was planning to 
implement ventilation system upgrades.
    At many high-back room-and-pillar stone mines, MSHA observed 
ventilation systems that were characterized by (1) inadequate main fan 
capacity (or no main fan at all), (2) ventilation control structures 
(air walls, stoppings, curtains, regulators, air doors, brattices, 
etc.) that are poorly positioned, in poor condition, or altogether 
absent, (3) free standing booster fans that are too few in number, too 
small in capacity, and located inappropriately, and (4) no auxiliary 
ventilation for development ends (working faces). At some mines, the 
``piston effect'' of trucks traveling along haul roads underground, 
along with natural ventilation pressure, provide the primary or only 
driving forces to move air.
    In naturally ventilated mines, temperature-induced differences in 
air density between the surface and underground result in natural air 
flows through mine openings at different elevations. Warmer and lighter 
mine air rises up out of a mine during the colder winter months, which 
draws in cooler and heavier air at lower elevation mine openings. In 
the summer, cooler and denser mine air flows out of lower elevation 
openings, which draws warmer less dense air into higher elevation 
openings. Under the right conditions, such air flows can be 
significant, but they are usually inadequate by themselves to dilute 
and carry away DPM sufficiently to reduce miners' exposures to the 
interim limit.
    The other principal shortcoming of natural ventilation is the 
inherent lack of a method of controlling air flow quantity and 
direction. Ventilation air flows can slow or stop when temperature 
differences between the surface and underground are small (common in 
the spring and fall), and the flow direction reverses between summer 
and winter, and sometimes even between morning and afternoon.
    Mine operators normally supplement natural ventilation with booster 
fans underground. However, if overall air flow is inadequate, as is 
usually the case with naturally ventilated mines, and when mine 
elevation differences or surface and underground temperature 
differences are small, booster fans are largely ineffective.
    The all too frequent result of these deficiencies is a ventilation 
system that is plagued by insufficient dilution of airborne 
contaminants, short circuiting, recirculation, and airflow direction 
and volume that are not controllable by the mine operator. These 
systems are barely adequate (and sometimes inadequate) for maintaining 
acceptable air quality with respect to gaseous pollutants (CO, 
CO2, NO, NO2, SO2, etc.), and are 
totally inadequate for maintaining acceptable concentrations of DPM. 
Mines experiencing these problems could benefit greatly from upgrading 
main, booster, and/or auxiliary fans, along with the construction and 
maintenance of effective ventilation control structures.
    MSHA believes that ventilation upgrades alone, along with the 
normal turnover of engines to newer, low-polluting models, may be 
sufficient for many stone mines to achieve compliance with the interim 
DPM limit. Consequently, MSHA has urged the mining industry to utilize 
mechanical ventilation to improve overall air flows and to enable 
better control of ventilating air.
    Ventilation fan upgrades for the stone mining sector are usually 
relatively inexpensive due to the low mine resistance associated with 
large openings. In many of these mines, a 250,000 cfm air flow can be 
obtained at less than 1 inch of water gage pressure. This air flow can 
be provided by a 50 horsepower motor. The major cost in these 
applications is usually distribution of the air flow underground to 
insure that adequate air quantities reach the working faces rather than 
short-circuiting to a return or recirculating around free-standing 
booster fans. Good air flow distribution requires such practices as 
installing or repairing ventilation control structures (brattice line, 
air curtains, etc.) or changes in mine design to incorporate unmined 
pillars as air walls.
    Deep multi-level metal mines have entirely different geometries and 
configurations from high-back room-and-pillar stone mines. They 
typically require highly complex ventilation systems to support mine 
development and production. These systems are professionally designed, 
they require large capital investments in shafts, raises, control 
structures, fans, and duct work, and they are costly to maintain and 
operate. At these mines, high ventilation system costs provide a major 
economic incentive to operators to optimize system design and 
performance, and therefore, there are typically few if any feasible 
upgrades to main ventilation system elements that these mines haven't 
already implemented, or would have implemented anyway, whether or not 
the DPM rule existed. Accordingly, and though it remains an option that 
might be attractive in new development, MSHA expects very few mines of 
this type to implement major ventilation system upgrades to achieve 
compliance with this rule.
    Despite the built-in incentives to design and operate efficient 
ventilation systems, however, MSHA has observed aspects of ventilation 
system operation at such mines that can be improved, usually relating 
to auxiliary ventilation in stopes. Auxiliary fans are sometimes sized 
inappropriately for a given application, being either too small (not

[[Page 32932]]

enough air flow) or too large (causing recirculation). Auxiliary fans 
are sometimes poorly positioned, so that they draw a mixture of fresh 
and recirculated air into a stope. Auxiliary fans are sometimes 
connected to multiple branching ventilation ducts, so that the air 
volume reaching a particular stope face may be considerably less than 
the fan is capable of delivering. Perhaps most often, the ventilation 
duct is in poor repair, was installed improperly, or has been damaged 
by blasting or passing equipment to the extent that the volume of air 
reaching the face is only a tiny fraction of that supplied by the fan. 
MSHA believes that these and similar problems exist at many mines, even 
if the main ventilation system is well designed and efficiently 
operated.
    An example is the mine where NIOSH conducted its Phase II 
Production Zone study of DPFs. As noted earlier, several auxiliary 
stope ventilation systems were evaluated by MSHA during an extended 
compliance assistance visit to this mine in June 2004. In the six 
stopes for which ventilation air flow measurements could be obtained at 
both the auxiliary fan location and at the end of the vent bag, the 
average air flow at the fan location was 24,400 cfm and the average 
flow at the end of the vent bag was 5,100 cfm. Auxiliary ventilation 
system leakage was 89% in one stope and 85% in another. Even in stopes 
where auxiliary system leakage was relatively low, significant 
recirculation was observed.
    Optimized auxiliary ventilation system performance alone, as one 
commenter noted, will not necessarily insure compliance with the DPM 
interim limit. Auxiliary ventilation systems simply direct air to a 
stope face so that the DPM generated within the stope can be diluted, 
transported back to, and carried away by the main ventilation air 
course. If this air is already heavily contaminated with DPM when it is 
directed into a stope, as could happen at mines employing series or 
cascading ventilation, its ability to dilute newly-generated DPM is 
diminished. In these situations, the intake to the auxiliary system 
must be sufficiently clean to achieve the desired amount of dilution, 
requiring implementation of effective DPM controls upstream of the 
auxiliary system intake. Such upstream controls might include a variety 
of approaches, such as DPM filters, low-polluting engines, alternate 
fuels or fuel blends, and various work practice controls, as well as 
main ventilation system upgrades at the few mines where they might be 
feasible. Toward the return end of a series or cascading ventilation 
system, if the DPM concentration of the auxiliary system intake is 
still excessive, other engineering control options would include 
enclosed cabs with filtered breathing air on the equipment that 
operates within the stope, or remote control operation of the equipment 
in the stope to remove the operator from the stope altogether.
    Environmental Cabs With Filtered Breathing Air. Cabs on mobile 
equipment and control rooms or booths for stationary installations, if 
provided with filtered breathing air, can be highly effective for 
reducing personal DPM exposures. MSHA has determined that environmental 
cabs can reduce operator exposures to DPM by 50% to 80%. In addition, 
such cabs and booths can significantly reduce exposures to harmful 
noise and dust, and they can also improve equipment operator comfort 
and productivity.
    The majority of equipment used in underground M/NM mining, 
especially in stone mines, have suitable cabs installed. However, MSHA 
has observed that many cabs, due to poor maintenance and operating 
practices, fail to provide effective control of DPM exposure. Typical 
problems are broken windows, ineffective door seals, inoperative AC 
systems and fans, plugged or missing air filters, openings into the cab 
where hoses or cables enter, and lack of company policies requiring 
doors and windows to be maintained in the closed position during 
operations.
    Some cab ventilation and filtration systems are undersized for the 
volume of air they should be moving. During MSHA's compliance 
assistance visits in 2003, MSHA observed numerous pieces of equipment, 
especially face drills, that were equipped with undersized cab air 
filtration systems. Research has shown that cab ventilation systems 
should be sized to achieve approximately one-half to one air change per 
minute in their respective cabs. For example, a 100 cubic foot cab 
should be ventilated by a system having the capacity to move 50 to 100 
cubic feet per minute. Cabs should also be sealed to obtain a positive 
pressure greater than 0.2 inches of water gage.
    MSHA DPM-Related Compliance Assistance. As noted earlier, MSHA has 
engaged in extensive DPM-related compliance assistance since the 
existing rule was issued in 2001, and these activities are continuing. 
Compliance assistance has included seminars at various locations 
throughout the country, hands-on sampling training workshops, the 
online Filter Selection Guide, a compliance guide, a ``single source'' 
internet Web site devoted to underground M/NM DPM issues, DPM baseline 
sampling at all mines affected by the rule, online listings of MSHA-
Approved diesel engines and DPF efficiencies, the Estimator, and on-
site compliance assistance visits at dozens of mines, among others.
    MSHA continues to consult with the M/NM Diesel Partnership (the 
Partnership). The Partnership is composed of NIOSH, industry trade 
associations, and organized labor. MSHA is not a member of the 
Partnership due to its ongoing DPM rulemaking activities. The primary 
purpose of the Partnership is to identify technically and economically 
feasible controls to curtail particulate matter emissions from existing 
and new diesel-powered vehicles in underground metal and nonmetal 
mines.
    MSHA's diesel testing laboratory located in Triadelphia, WV has 
been active in evaluating many DPM control technologies. An example is 
the investigation to characterize NO2 emissions from 
catalyzed DPFs. As a result of this work, MSHA provided information to 
the mining community on the effects of catalyzed DPF's on 
NO2 production. MSHA's laboratory determined under steady 
state engine operating conditions, that a heavily platinum-catalyzed 
DPF would increase the NO2 concentration measured in the raw 
exhaust after the exhaust gas passed through the DPF. The increase in 
NO2 was compared to the required gaseous ventilation rate 
for the test engine without the DPF installed. The laboratory data 
showed that the gaseous ventilation rate would increase with a highly 
platinum-catalyzed DPF installed. MSHA's laboratory also tested DPFs 
that were either specially catalyzed with platinum (lower wash-coat 
platinum content) or a base metal wash-coat (no platinum used). The 
results of the laboratory tests showed no increase in the gaseous 
ventilation quantity when compared to the quantity without the DPFs 
installed. MSHA provided the industry with a Program Information 
Bulletin (PIB) P02-04, ``Potential Health Hazard Caused By Platinum-
Based Catalyzed Diesel Particulate Matter Exhaust Filters,'' dated May 
31, 2002. This PIB is located on MSHA's web page at the following 
internet address: http://www.msha.gov/regs/complian/PIB/2002/pib02-04.htm.
 The PIB states that mine operators that choose to use catalyzed 

DPFs that have shown an increase in NO2 in the laboratory 
need to ensure that the machines installed with these filters have 
adequate ventilation, and recommends that personal monitoring for 
NO2 should be performed.
    MSHA also provides an updated list on the internet of DPFs that 
have been

[[Page 32933]]

evaluated by MSHA. The internet address is: http://www.msha.gov/01-995/Coal/DPM-FilterEfflist.pdf.
 This list is divided into three tables. 

Table I includes paper and synthetic filters, mainly intended to be 
disposable. These DPFs are only used when the exhaust gas temperature 
is maintained to below 302[deg]F, as is required in inby areas of gassy 
mines. This is normally accomplished by the use of an exhaust gas heat 
exchanger. Temperature sensors and backpressure sensors must be used 
with these filters to protect the DPF from exhaust gas temperatures 
that would exceed 302 [deg]F or backpressures that would exceed the 
engine manufactures allowable limit. Table II lists ceramic and high 
temperature disposable pleated element media DPFs that do not increase 
the concentration of NO2 in the exhaust. Table III lists the 
DPFs that are platinum-catalyzed and have been determined in the 
laboratory to increase NO2 concentrations above the test 
engine's gaseous ventilation rate.
    MSHA's laboratory has also conducted limited tests on several 
control technologies other than DPFs. Evaluations have been conducted 
on an Ecomax which consists of a series of magnets installed on the 
fuel system lines, Rentar, an in-line fuel catalyst installed in the 
machine's fuel line, and the Fuel Preporator, a system for removing 
collected air from the fuel system design for better fuel combustion. 
The test results of the laboratory evaluations were inconclusive in 
demonstrating significant reductions in whole diesel particulate, 
however the data did not show any adverse effects on the raw DPM 
exhaust emissions.
    NIOSH also analyzed the Rentar and Fuel Preporator for their EC 
reduction potential. NIOSH's results were consistent with MSHA's 
results, and showed no significant EC reductions and no adverse effects 
on the engine emissions.
    MSHA's laboratory evaluated the changes in engine exhaust emissions 
when operating at high altitudes (greater than 1000 feet in elevation). 
MSHA used two electronic fuel injected engines for the test, a Mercedes 
904 and a Deutz BF4M 1013FC. MSHA first conducted field tests at engine 
laboratories located at 4000 feet and 6700 feet. Next, MSHA brought the 
two test engines to its laboratory. Using an altitude simulator setup, 
MSHA verified the accuracy of the simulator and ran various tests to 
evaluate the effects of altitude on the gaseous emissions and DPM. This 
high altitude work led to the development of guidelines that MSHA is 
using for approving diesel engines under 30 CFR, part 7, subpart E for 
engine operation above 1000 feet.
    MSHA received comments suggesting that its compliance assistance 
visits at various mine sites support the position that the DPM rule, 
even at the 400TC [mu]g/m\3\ interim limit, is economically 
and technologically infeasible. MSHA did visit a number of mines that 
were not in compliance with the interim DPM limit to provide compliance 
assistance, but at each such mine, the operator was presented with 
recommendations for utilizing feasible engineering and work practice 
controls for attaining compliance. MSHA determined that these mines 
were out-of-compliance not because it was infeasible for them to attain 
compliance, but because the respective mine operators had not yet fully 
implemented all feasible controls that were available to them.
    MSHA's compliance assistance work at the Greens Creek mine included 
an evaluation of DPM reductions obtained using heavily platinum-
catalyzed ceramic DPFs that relied on passive regeneration. The 
machines were equipped with engines ranging from 300 to 475 horsepower. 
The results of this testing showed that personal DPM exposures for the 
subject equipment operators (loaders and haulage trucks) were reduced 
by 57% to 70% when the DPFs were installed. The use of the ceramic DPFs 
reduced the average engine emissions by 96%.
    The Greens Creek report also showed that high DPM reductions (>90%) 
occurred even when a ceramic filter was compromised by cracking around 
the edges. This cracking was determined to be caused by a manufacturing 
defect related to the ``canning'' process (securing the ceramic filter 
in a stainless steel ``can'' for installation on the subject diesel 
equipment). Through discussions with the manufacturer, Greens Creek 
resolved the problem, and DPFs delivered since then have performed 
satisfactorily without any cracking. In addition, the use of 
environmental cabs reduced the DPM concentrations (i.e., concentration 
inside the cab versus outside the cab) by 75% when DPFs were used and 
80% when DPFs were not in use.
    As expected, NO2 increases were observed during these 
tests because the mine operator was using heavily platinum-catalyzed 
DPFs. However, the increases were so small (about 1 ppm in the 
downstream air flow compared to the upstream air flow in the area where 
a loader and two or three trucks were operating) that it was unclear 
whether the cause was data variability, slight changes in ventilation 
rate, or the use of heavily platinum-catalyzed DPFs. Greens Creek 
stated in its comments to this rulemaking that a 1-2 ppm increase in 
NO2 is experienced when highly platinum-catalyzed DPFs are 
used, but that this increase has been manageable for the mine.
    MSHA agrees that a highly platinum-catalyzed filter may increase 
NO2 levels based on engine duty cycle and ventilation. 
NO2 is formed from NO in the engine's exhaust in the 
presence of the catalyst. This reaction occurs at exhaust gas 
temperatures of approximately 325[deg]C. This temperature is also the 
temperature at which the platinum catalyst will allow for passive 
regeneration. Manufacturers of platinum-catalyzed DPFs have normally 
wash-coated their filters with large amounts of platinum to make sure 
that the DPFs will regenerate. This large concentration of platinum, in 
combination with the relatively long retention time of the exhaust gas 
in the filter, results in the formation of NO2.
    Manufacturers have been evaluating wash-coat formulations 
containing less platinum loading to lower the NO2 effects. 
Catalytic converters are also wash-coated with platinum; however, the 
loading used on catalytic converters is lower than ceramic DPFs, and 
due to faster movement of the exhaust gas through the catalytic 
converter compared to the ceramic filter, NO2 increases are 
minimal. One manufacturer provides an exhaust gas recirculation system 
(EGR) that reduces both oxides of nitrogen (NOX) and DPM 
when used in combination with a DPF.
    Mine operators also have the option of using DPFs that are not 
heavily wash-coated with a platinum catalyst. One manufacturer offers a 
lightly platinum-catalyzed DPF that is used in conjunction with a 
platinum-cerium fuel-borne catalyst (Fuel additive). This system has a 
slightly higher passive regeneration temperature requirement than 
heavily platinum-catalyzed DPFs, but it produces no excess 
NO2. Other options which do not produce excess 
NO2 include base metal catalyzed passive regenerating DPFs, 
and various on-board and off-board active regenerating DPFs. As noted 
earlier, part of the DPF selection process involves an evaluation of 
potential NO2 problems along with related ventilation 
issues. Where NO2 exposures could be problematic, MSHA 
recommends that heavily platinum-catalyzed DPFs be avoided.
    Table VII-1 provides information in the ``Comments'' column on the 
effects of DPF catalysts on NO2 emissions. MSHA has tested 
in their laboratory the types of DPFs listed, and has posted on

[[Page 32934]]

its website a list of the DPFs that can cause NO2 increases 
from the engine and those catalytic formulations that do not 
significantly increase NO2.
    MSHA is currently not aware of problems with overexposure to 
NO2 at mines using platinum-catalyzed DPFs on a routine 
production basis, where the overexposures are uniquely related to the 
DPFs. One mine operator that had been experiencing frequent 
overexposures to NO2 noted that these overexposures ceased 
after a major ventilation upgrade, despite increased use of heavily 
platinum-catalyzed DPFs.
    PIB 02-04 alerted mine operators that the platinum-
catalyzed DPFs identified on MSHA's website could increase 
NO2. MSHA continues to advise mine operators to monitor for 
any increases in ambient NO2 concentrations with the 
addition of platinum-catalyzed DPFs to their inventory.
    When NIOSH's Phase II study tests 2 and 3 were terminated 
prematurely due to high NO2 levels, the overexposures were 
determined to be due mainly to insufficient ventilation. As discussed 
previously, the average increase in NO2 from the use of 
platinum-catalyzed DPFs in the test area was approximately 1 ppm, but 
brief 3-5 ppm spikes were also observed. As stated above, mine 
operators are advised to sample for NO2 when platinum wash-
coated DPFs are used to ensure miners are not overexposed. Mine 
operators who use platinum-catalyzed DPFs should maintain ventilation 
systems that are able to remove or dilute the NO2 to a non-
hazardous level, and they must be aware of localized areas where 
NO2 could build up more quickly and create a health hazard 
for exposed miners.
    As discussed in the Greens Creek report, the use of catalyzed DPFs 
at that mine did not produce substantial increases in NO2 
levels. MSHA is continuing to work with filter manufacturers to 
evaluate catalytic formulations on NO2 generation.
    Stillwater mine DPM compliance. In its comments addressing the 2003 
NPRM, Stillwater Mining Company (SMC) provided discussion and several 
tables detailing its estimated DPM-related compliance costs. In its 
April 2004 comments in response to the February 20, 2004 limited 
reopening of the public record on this rulemaking, SMC provided further 
discussion and another compliance cost summary table which grouped cost 
elements into major categories. These estimates totaled about $114 to 
$117 million over a 10 year period.
    Using the Stillwater compliance cost estimates and other 
information obtained by MSHA during visits to the Stillwater mine, MSHA 
analyzed and evaluated Stillwater's estimated costs and developed a 
compliance cost estimate for this mine based on an alternative DPM 
control strategy. This analysis and evaluation is discussed below, and 
a summary is provided in Table VII-3. MSHA conducted this analysis and 
evaluation to demonstrate both to Stillwater and to other mines having 
some of the same or similar equipment, mine layouts, and operating 
practices that their choice of control strategy can significantly 
impact overall compliance costs, and therefore, the feasibility of 
compliance.
    MSHA's estimated yearly compliance costs for this mine, which are 
based largely on the itemized cost estimates provided by Stillwater, 
are between $1.24 million and $2.09 million per year. The lower end of 
this range relates to estimated compliance costs not including a recent 
$9 million ventilation upgrade. As discussed below, although Stillwater 
included the cost of this upgrade in its estimated DPM compliance 
costs, MSHA believes this cost item should not be considered DPM-
related, or is only partially attributable to DPM compliance because 
the ventilation system at this mine required a major upgrade anyway, 
independent of DPM issues. MSHA's $2.09 million yearly compliance cost 
estimate includes the $9 million ventilation upgrade.
    Although Stillwater's DPM-related compliance costs will be 
significant, they are not substantially different from expectations 
based on MSHA's 2001 REA. In the REA for the 2001 final DPM rule, MSHA 
determined that annual compliance costs would be about $128,000 for an 
average underground M/NM mine. However, Stillwater's mining operations 
are not representative of an average mine. Its fleet of 350+ pieces of 
diesel equipment is many times larger than the average mine's. MSHA's 
estimated yearly DPM-related compliance costs for large precious metals 
mines included in the REA was $659,987, based on a fleet size of 133 
diesel vehicles. Stillwater's fleet is about 2.6 times larger than the 
133 vehicle basis for this estimate. Thus, yearly compliance costs of 
2.6 x $659,987, or $1.72 million for Stillwater would be consistent 
with the 2001 REA's compliance cost estimate for a precious metals 
mining operation of this size.
    If the cost of Stillwater's recent ventilation system upgrade is 
not included as a DPM compliance cost, which as noted below, is a 
reasonable determination based on long-standing ventilation system 
deficiencies at this mine, Stillwater's estimated yearly compliance 
cost would be $1.24 million. As noted in the preceding paragraph, by 
way of comparison, an estimated compliance cost of $1.72 million for a 
precious metals mine of this size would be consistent with the 2001 
REA. If, however, the entire ventilation system upgrade is considered 
DPM-related, MSHA's estimated yearly compliance cost of $2.09 million 
for Stillwater would be about 22% higher than expected, based on the 
2001 REA. If the entire ventilation system upgrade is considered DPM-
related, but the annual savings resulting from the associated reduction 
in ventilation fan power consumption is deducted from the annualized 
cost of the upgrade, MSHA's estimated yearly compliance cost of $1.57 
million for Stillwater would be about 9.5% less than expected, based on 
the 2001 REA.
    For MSHA's analysis and evaluation, Stillwater's DPM compliance 
costs were grouped into six major cost categories. The analysis and 
evaluation of these six major cost categories is discussed below:
    1. Ventilation. As noted above, a $9 million ventilation upgrade 
was recently completed at the Stillwater mine, and the cost of this 
upgrade was included by Stillwater in its DPM compliance cost estimate. 
However, MSHA believes this upgrade would have been necessary with or 
without a DPM rule due to ongoing air quality problems and plans for 
increased mine development. Thus, this expenditure should not be 
considered a DPM compliance cost, or at most, only partially a DPM 
compliance cost.
    Total ventilation at the mine prior to the upgrade was about 
627,000 cfm, corresponding to approximately 52 cfm/actual utilized 
horsepower. After the upgrade, total ventilation volume increased to 
840,000 cfm, which is about 69 cfm/actual utilized horsepower.
    Most of Stillwater's diesel equipment has MSHA nameplate 
ventilation rates between 50 and 70 cfm/horsepower. These laboratory 
derived values indicate the ventilation necessary to maintain 
compliance with MSHA exposure limits for CO, CO2, NO, and 
NO2. Taking into account such practical in-mine factors as 
varying equipment duty cycles, imperfect mixing, use of DOCs, etc., 
acceptable air quality can sometimes be attained at ventilation rates 
somewhat less than the nameplate values. However, other factors, 
including out-of-tune engines, marginal auxiliary ventilation system 
performance, on-shift

[[Page 32935]]

blasting, and heavy concentrations of diesel equipment in particular 
sections of a mine can result in chronic localized noncompliance with 
gaseous emission limits.
    For example, Stillwater has had a persistent problem with 
NO2 overexposures for many years, indicating inadequate 
ventilation. Per company policy, whenever an NO2 monitor 
(carried by equipment operators) exceeded 5 dpm at the operator's 
location, that operator was removed to the surface. The mine operator 
has frequently removed miners to the surface for this reason over 
recent years. Thus, the ventilation upgrade was overdue, even without 
consideration for DPM levels underground.
    Other considerations also factored into the decision to carry out 
the ventilation upgrade, including planned production tonnage 
increases, the need to utilize trucks to haul ore up grade from below 
the level of the shaft bottom, an excessive number of booster fans 
(sometimes competing with each other for limited air), and the desire 
to increase the number of ventilation intakes into the mine (resulting 
in more fresh air escape routes and lower intake air velocities to 
improve miner comfort and dust conditions). By any number of measures, 
mine development had overreached the old ventilation system. The 
ventilation upgrade accomplished all of the above objectives, and 
resulted in a reduction of total fan power consumption by 1,000 
horsepower.
    Even if this ventilation upgrade could be entirely attributed to 
DPM compliance, the cost must be annualized over the expected 20+ year 
life of the asset, so the yearly cost (using a 7% discount rate) would 
be about $850,000. This yearly cost is partially offset by savings in 
electricity costs resulting from the 1,000 horsepower reduction in fan 
power consumption, so the ventilation upgrade actually resulted in a 
net annual cost to Stillwater of only about $197,000 (1,000 hp x 24 
hours/day x 365 days/year x 0.745 kw-hr/hp-hr x 10[cent]/kw-hr = 
$652,620; $849,536 - $652,620 = $196,916).
    2. Diesel Engines and Engine Upgrades. Only a portion of the 
expense of new diesel engines and engine upgrades should be considered 
a DPM compliance cost. Diesel engines have a finite life and need to be 
renewed and replaced periodically. Some new engines and engine upgrades 
would have been necessary with or without a DPM rule. Also, new, low-
emission engines enable improved operating efficiencies due to lower 
fuel consumption and better maintenance diagnostics, resulting in 
significant operating cost savings that partially off-set purchase 
costs.
    Like the ventilation upgrade, however, even if the total cost of 
engines and engine upgrades was attributable to DPM compliance, these 
costs (estimated by Stillwater at $1.2 million) must be annualized over 
the expected 10 year life of an engine, resulting in a yearly cost of 
about $171,000 (using a 7% discount rate).
    3. Soot Traps, Filters, Passive DPFs. The mine currently has fewer 
than 30 passive regeneration DPF systems and only one passive/active 
regeneration DPF system (fuel burner) in use, and reports no 
operational problems at this time, except one filter destroyed by a 
failed turbo-charger.
    In its comments to the 2003 NPRM, Stillwater outlined a plan for 
utilizing a combination of passive and active DPFs to control DPM in 
its mine. Passive filters would be used where equipment duty cycles and 
corresponding exhaust temperatures suggested the application would be 
successful, and active filters would be utilized on the remaining 
equipment. Stillwater reports $160,000 in passive filter costs to date. 
Assuming a filter life of two years, this results in a yearly cost of 
about $88,500 (using a 7% discount rate).
    4. Engine Test Equipment. The engine test equipment has a 5-year 
life, resulting in an annualized cost of about $68,000 (using a 7% 
discount rate).
    5. Emissions expenditure. The basis for Stillwater's ``Emissions 
expenditure'' line item cost of $43,000/month is unclear. As noted 
above, the mine currently has fewer than 30 passive regeneration DPF 
systems and only one active regeneration DPF system in use, and reports 
no operational problems at this time, except one filter destroyed by a 
failed turbo-charger. Engine-related emissions expenses are addressed 
in the diesel engines, engine upgrades, and engine test equipment line 
items above. However, ``emissions expenditures'' of $516,000 per year 
($43,000 per month x 12 months) are included as submitted by Stillwater 
in MSHA's estimated compliance cost.
    6. Active Regeneration Systems. Based on Stillwater's existing 
knowledge base relating to equipment duty cycles and exhaust 
temperatures, their plan for controlling DPM emissions included passive 
filters for only a small percentage of the mines' fleet: the large 
loaders and ore haulage trucks. In contrast, about 200 vehicles were 
expected to require active regeneration DPF systems.
    For costing the active systems, Stillwater made the following 
assumptions:
    a. Regeneration of the DPFs would be accomplished on-board the 
vehicles. Vehicles equipped with DPFs would travel from their normal 
work areas (stopes, develop ends, haulageways, etc.) to specially 
excavated regeneration stations provided with the necessary means of 
connecting the filters to power and compressed air. Upon arrival at a 
regeneration station, the filters would be ``plugged in'' to electrical 
power and compressed air utilities to accomplish regeneration.
    b. In addition to including the costs of filters and associated 
regeneration equipment, Stillwater's active DPF cost estimates also 
included excavating the regeneration stations and installing the 
required electrical power and compressed air.
    c. To insure reasonable travel distances to regeneration stations 
as mine workings advance over time, Stillwater's cost estimate was 
developed in the context of a 10-yr mine plan that included the 
excavation of new regeneration stations periodically over the 10 years.
    Stillwater's total estimated costs for active filter systems, 
regeneration equipment, and regeneration stations was about $104.4 
million over the 10-yr period of the mine plan. Of this total, $100.8 
million (96.6%) was for excavation of the regeneration stations, and 
$3.6 million was for active filter systems and regeneration equipment.
    Neither the number of active systems required at Stillwater, nor 
the estimated total cost of implementing active filters as specified in 
Stillwater's comments is disputed by MSHA. However, MSHA does not 
believe the particular plan developed by Stillwater is the optimal 
means of utilizing active DPM filters at this mine. Various alternative 
approaches for utilizing active filters exist which would be far less 
costly.
    Since excavating regeneration stations accounted for over 96% of 
the total cost of implementing Stillwater's active filter plan, 
alternatives that do not include such excavation costs would have a 
significant cost advantage over Stillwater's plan. It is somewhat 
curious that Stillwater developed its active DPF plan on the basis of 
this particular on-board active regeneration system, despite the 
extraordinarily high cost of excavating the regeneration stations, and 
Stillwater's prior experience with premature failure of the on-board 
heating elements built into the filters.
    A lower cost alternative to Stillwater's approach utilizes an on-
board fuel burner system to regenerate filters. The ArvinMeritor 
[reg] system has been on trial at this mine since February 
2004 with excellent results. This system actively

[[Page 32936]]

regenerates the filter media during normal equipment operations, and 
does not require the host vehicle to travel to a regeneration station 
to regenerate its filter.
    Another less costly alternative would be to utilize off-board 
regeneration instead of on-board regeneration. In off-board 
regeneration, a dirty filter is removed and replaced with a clean 
filter at the beginning of each shift. During shift change, the dirty 
filters are then transported by the equipment operator or a designated 
filter attendant to a central regeneration station or stations.
    Such stations could be a fraction of the size of the regeneration 
stations envisioned in Stillwater's plan, because they would only need 
to accommodate the filters, not the host vehicles. Since the host 
vehicles would not need to travel to the regeneration stations, the 
travel distance from normal work areas to the regeneration stations 
would be less important, greatly lessening the need for frequent 
construction of new regeneration stations as the workings advance. It 
is very likely that such stations could be co-located in existing 
underground shops, unused muck bays, unused parking areas, or other 
similar areas.
    Off-board regeneration might not be practical on larger machines 
due to the size of the filters. For larger machines that are not 
suitable for passive regenerating filters, the fuel burner approach 
might be preferable. But many of the machines targeted for active 
filtration are quite small, having 40 to 80 horsepower engines. Active 
filters for these engines are correspondingly small, and could be 
easily and quickly removed and replaced using quick disconnect 
fittings.
    Another lower cost option would be to utilize disposable high-
temperature synthetic fabric filters, especially on smaller, light duty 
equipment such as pickups, boss buggies, and skid steers. Depending on 
equipment utilization, such filters might only need to be replaced once 
or twice per week.
    In Table VII-3, the line for active filters shows the 10-year cost 
of Stillwater's plan for utilizing active filters along with MSHA's 
estimate of the yearly cost of alternatives to Stillwater's plan. 
MSHA's cost estimate for this line item is based on Stillwater's 
estimated cost for active filter systems, minus the cost of excavating 
regenerations stations, or $3.6 million over 10 years. Annualizing 
these active filter costs over the two-year expected life of these 
filters using a discount rate of 7% results in a yearly cost of about 
$398,000.

                       Table VII-3.--Stillwater's and MSHA's DPM Compliance Cost Estimates
----------------------------------------------------------------------------------------------------------------
                                        Stillwater's cost
             Cost item                      estimate           MSHA cost estimate           MSHA comments
----------------------------------------------------------------------------------------------------------------
Mine Ventilation Upgrade...........  >$9 million...........  $0....................  This upgrade necessary with
                                                                                      or without DPM rule to
                                                                                      address ongoing air
                                                                                      quality problems and plans
                                                                                      for mine development.
                                                             $849,536/yr \1\.......  Even if upgrade necessary
                                                                                      for DPM compliance, this
                                                                                      capital cost annualized
                                                                                      over expected 20+ year
                                                                                      life of the asset.
                                                             $327,440/yr \1\.......  Annualized cost over
                                                                                      expected 20+ year life of
                                                                                      the asset minus annual
                                                                                      power cost savings.
Engine upgrades, other misc.         >$1.2 million.........  $170,853/yr \1\.......  Some engines/upgrades part
 expenses.                                                                            of normal turnover of
                                                                                      engines and not DPM
                                                                                      compliance cost. Cost of
                                                                                      engines/upgrades
                                                                                      annualized over 10 year
                                                                                      expected engine life.
Test Equipment.....................  >$280,000.............  $68,289/yr \1\........  Cost of test equipment
                                                                                      annualized over 5 year
                                                                                      expected equipment life.
Soot traps, filters, passive DPFs..  $160,000..............  $88,495/yr \1\........  Cost of DPFs annualized
                                                                                      over 2 year expected
                                                                                      filter life.
Emissions expenditure..............  $43,000/month.........  $516,000/yr \1\.......  Cost element is unclear
                                                                                      based on current filter
                                                                                      use.
Active DPF systems, regeneration     $104.4 million over 10  $398,226/yr \2\.......  Less costly approaches for
 equipment, and regeneration          years.                                          implementing active
 station excavation.                                                                  regeneration were
                                                                                      overlooked. Approaches
                                                                                      that do not require
                                                                                      excavation of regeneration
                                                                                      stations save $100.8
                                                                                      million over 10 years.
                                                                                      $3.6 million would still
                                                                                      be required for filters
                                                                                      and regeneration
                                                                                      equipment, however, this
                                                                                      expense would be incurred
                                                                                      over 10 years.
                                    -------------------------
      Grand Total..................  $104.4 million over 10  Annual cost of $1.24    Certain cost elements
                                      years for active        to $2.09 million.       should not be considered
                                      DPFs, plus $10-$13     $1.24 million if cost    DPM compliance costs.
                                      million for other       of ventilation          However, even including
                                      costs over 10 years.    upgrade is not          ALL listed costs for
                                      Total cost $114-$117    included;.              ventilation, passive and
                                      million over 10 years. $2.09 million if cost    active DPFs, engines/
                                                              of ventilation          engine upgrades, test
                                                              upgrade is included;.   equip, and emissions
                                                             $1.57 million if cost    expenditures, MSHA
                                                              of ventilation          estimates total yearly
                                                              upgrade is included     cost for DPM compliance
                                                              minus power cost        will not exceed $2.09
                                                              savings.                million. Excluding
                                                                                      ventilation, estimated
                                                                                      total yearly cost is $1.24
                                                                                      million. Including
                                                                                      ventilation but
                                                                                      considering power cost
                                                                                      savings, estimated total
                                                                                      yearly cost is $1.57
                                                                                      million. Estimated yearly
                                                                                      compliance cost of $1.72
                                                                                      million for a precious
                                                                                      metals mine of this size
                                                                                      would be consistent with
                                                                                      2001 REA.
----------------------------------------------------------------------------------------------------------------
Notes:
\1\ Cost estimate based on commenter's estimated cost, annualized over the expected life of the item using a 7%
  discount rate. The annualization factor for a capital expenditure is 9.4% for 20 years, 14.2% for 10 years,
  24.4% for 5 years, and 55.3% for 2 years.
\2\ Cost estimate based on commenter's estimated cost for active systems minus the cost of excavating
  regeneration stations, annualized over the expected life of the active systems.


[[Page 32937]]

    Kerford Limestone DPM compliance. Kerford Limestone reported the 
results of a consultant's study that indicated compliance with the DPM 
limit for that mine would cost $348,000 for engine improvements, $1.15 
million for ventilation upgrades, and $25,500 to $38,500 per year for 
DPFs. They reported investing $975,000 to date toward DPM compliance.
    Kerford's engine costs of $348,000, when annualized over 10 years 
at a discount rate of 7%, results in a yearly cost of about $49,500. 
The $1.15 million ventilation cost, when annualized at the same 
discount over the expected 20+ year life of this asset, results in a 
yearly cost of about $108,600. When these two yearly costs are added to 
the maximum estimated annual DPF cost of $38,500, the total yearly cost 
for Kerford is about $196,600.
    Without commenting specifically on the reasonableness of Kerford's 
itemized cost estimates or whether the overall DPM control strategy 
proposed by its consultant was optimized for this mine, MSHA notes that 
Kerford's self-reported total yearly compliance cost of about $196,000 
is not excessive for an underground stone mine in its size category. By 
way of comparison, a yearly compliance cost of over $300,000 for a 
stone mine of this size would be consistent with MSHA's REA for the 
existing 2001 final rule.
    MSHA's REA for the existing 2001 final rule estimated compliance 
costs for a medium sized (20 to 500 employees) stone mine to be 
$150,738. However, this estimate was based on a fleet size of 9.5 
pieces of production equipment for this industry sector and mine size 
category. Kerford operates 19 pieces of production equipment. Adjusting 
the REA estimate of $150,738 for the larger fleet size at Kerford 
results in an estimated yearly compliance cost of $301,476. Thus, 
Kerford's estimated $196,600 yearly compliance cost is only about 65% 
of the level that would be expected for an underground stone mine of 
this size, based on the 2001 REA. The cost is virtually unchanged in 
the REA supporting this final rule.
    It was suggested by a commenter that MSHA underestimated Kerford 
Limestone's compliance costs by over $1 million, and it was further 
suggested that this underestimate, if extrapolated to the entire 
underground stone mining industry, resulted in industry-wide compliance 
costs exceeding $100 million. However, Kerford Limestone's yearly 
compliance costs, using its own cost estimates, are substantially less 
than expected, based on the 2001 REA for a medium sized underground 
stone mine.
    Bio-Diesel tests at Carmeuse Black River and Maysville mines. 
Commenters stated that in-mine tests with bio-diesel fuel produced 
measurable reductions in ambient DPM concentrations, but did not bring 
the subject mine into compliance. These comments refer to MSHA's 
compliance assistance work at the Carmeuse Black River and Maysville 
stone mines in Kentucky. At both mines, the use of bio-diesel fuel 
produced reductions in DPM. The recycled vegetable oil (RVO) with a 50% 
blend of bio-diesel to standard diesel fuel showed a 69% reduction in 
DPM, based on TC, for the area samples at the Maysville mine. Personal 
samples collected at the Black River Mine showed a 44% reduction in DPM 
with RVO at a 35% blend of bio-diesel to standard diesel fuel. The 
Virgin Soy Oil (VSO) mixtures showed reductions, but they were not as 
effective as the RVO at similar blends.
    The Maysville mine was in compliance with the interim limit based 
on the baseline samples and the samples taken with bio-diesel. In 
contrast, the Black River Mine was not in compliance with the interim 
limit based on the samples taken, even with the reduction in DPM using 
bio-diesel. One main difference between the two mines was that the 
Maysville mine had significantly more ventilation than Black River. 
This result indicates that the Black River mine will have to implement 
additional DPM controls to come into compliance, such as ventilation 
upgrades, cleaner engines, or DPFs.
    These commenters did not dispute the DPM reductions obtained. 
However, they indicated the following: That Deutz Corporation's 
Technical Circular does not approve the use of bio-diesel blends above 
20%; that a 50% bio-diesel fuel presented insurmountable equipment 
problems; and that the cost of bio-diesel has increased significantly, 
adversely impacting the feasibility potential of the 20% mixture.
    MSHA reviewed Deutz's Technical Circular (0199-3005en), and 
discussed this issue with Deutz. The Technical Circular provides a 
general statement that bio-diesel fuel is approved for Deutz brand 
engines. The Technical Circular does not mention any limitation on the 
use of bio-diesel above a certain percentage blend. Deutz requires that 
all fuels used in their engines meet Deutsches Institute fur Normung 
e.V. (DIN) specifications (German National Standards). The Deutz 
Technical Circular provides the DIN specifications for bio-diesel fuel.
    Comments regarding equipment problems relate to reports of bio-
diesel fuel causing clogging of fuel filters, resulting in excessive 
equipment downtime. One commenter expressed concern that Tier 2 engines 
used fuel filtering systems that would not be compatible with bio-
diesel. MSHA understands that engine manufacturers are working with the 
filter manufacturers to provide the best filtration for all engines. 
MSHA is not aware of any unique changes for EPA Tier 2 engines as 
related to fuel filtering systems or for utilizing bio-diesel fuel. As 
the engine technology continues to improve, especially in the area of 
the fuel system components, better fuel filtration systems will be 
utilized by the engine manufacturers.
    There are frequent references in the technical literature to bio-
diesel fuels initially cleaning old sediments out of fuel lines, 
thereby causing fuel filters to clog. It follows that fuel filters 
should be changed more frequently when bio-diesel is first used in a 
fuel system. However, the commenter suggests an entirely different type 
of incompatibility that is not limited to the transition period when 
bio-diesel is first used. This may or may not be a unique situation 
that may take additional work to resolve. The mine may have to install 
an additional by-pass filtering system on the machine to allow the 
operator to switch to another set of fuel filters instead of shutting 
down production if a fuel filter clogs.
    MSHA is not aware of long term filter clogging with the use of bio-
diesel fuel. However, through the NIOSH List-Server, mine operators 
have the opportunity to share experiences like the filter clogging 
problem with the mining community, and possibly receive a solution. A 
mine operator may use the List-Server to ask others in the mining 
community if their problem has been observed in other situations. 
Interested parties can respond, thus sharing experiences and solutions 
in a timely manner. The List-Server was established by the diesel team 
at NIOSH, Pittsburgh in response to the expressed and obvious need for 
a means to disseminate and share information and experiences concerning 
the application of available technologies for the reduction of miner 
exposures to DPM and gaseous emissions in underground mines.
    Regarding the cost of bio-diesel, MSHA acknowledges that users pay 
a premium for bio-diesel over standard diesel fuel. The cost for bio-
diesel can vary based on such factors as market price swings in the 
cost of feed-stocks, state tax incentives, proximity to production 
facilities, etc., but normally, where bio-diesel is available, the

[[Page 32938]]

premium is about one cent per gallon per percent bio-diesel in the fuel 
blend. At higher percentage bio-diesel blends, this premium can result 
in significantly higher overall fuel costs for the end-user. Depending 
on mine-specific factors, however, use of bio-diesel may be a cost-
effective DPM control option, either used by itself or in conjunction 
with other controls. Since the rule is performance oriented, the mine 
operator is free to choose the means of compliance.
    Based on these results and other data, MSHA's believes that bio-
diesel is a feasible DPM control. In the case of the Black River mine, 
bio-diesel would have to be used in combination with other controls for 
the mine to achieve compliance, or the mine operator may choose to 
abandon bio-diesel altogether and rely entirely on other controls for 
attaining compliance. MSHA disagrees with the commenters' assertion 
that a 50% bio-diesel blend presents ``insurmountable equipment 
problems.'' Bio-diesel is recognized by the EPA as an alternative clean 
fuel, engine manufacturers do not recommend against its use, and 
clogging can be prevented by the use of by-pass filtering systems.
    Water Emulsion Fuel: As discussed under the MSHA compliance 
assistance activities, we conducted tests at four mines to evaluate 
water emulsion fuel. These tests included a test at a small clay mine 
that used older technology engines, two single level limestone mines 
that used clean burning engines, and one multilevel limestone mine that 
used clean burning engines. Summer (20% water) and winter (10% water) 
blends of fuel were tested at two mines. Only summer blends of fuel 
were tested at the other two mines. MSHA evaluated the reduction in 
total mine DPM emissions by taking measurements at the mine exhaust 
openings, with and without the water emulsion fuel in use, and 
comparing these to similarly made measurements when standard No. 2 
diesel fuel was used. Table VII-4 summarizes the reductions in 
emissions measured for the tests.
    For clean burning engines the reduction in DPM emissions (as EC) 
ranged from 63 to 81 percent. For older engines the reduction in DPM 
emissions (as EC) was approximately 49 percent. Personal exposures were 
also reduced, however, this reduction was more variable than the 
reduction in engine emissions. This variability was attributed to the 
use of cabs, location in the mine and the specific ventilation rates at 
the work area in the mine.

     Table VII-4.--Emission Reductions for Water Emulsion Fuel Tests
------------------------------------------------------------------------
                                                    Percent     Percent
                                                   reduction   reduction
                      Mine                           in EC       in EC
                                                    (winter     (summer
                                                    blend)      blend)
------------------------------------------------------------------------
Clay............................................  ..........          49
Limestone.......................................          77          81
Limestone.......................................          63          73
Multilevel Limestone............................  ..........          80
------------------------------------------------------------------------

    For each mine test, equipment operators reported a noticeable loss 
of horsepower. However, this horsepower loss, even in the multilevel 
limestone mine, did not adversely effect production. In fact, during 
several of the mine tests, production was significantly above normal. 
The water emulsion fuel was favorably received by the employees. 
Workers reported that visibility improved. The water emulsion fuel has 
the same per gallon cost as No. 2 diesel fuel. Several operators 
reported as much as a 20 percent increase in fuel usage to compensate 
for the power loss.
    During the water emulsion fuel tests, a potential operating problem 
was observed when the fuel was used in Deutz engines. Simply put, some 
engines would not run. The source of this problem was traced by the 
engine and fuel manufacturers to a high efficiency water separator in 
the engine fuel line. The engine and fuel manufacturers have indicated 
that the problem can be corrected by replacing the standard high 
efficiency water separator with a less efficient unit.
    We believe that the use of water emulsion fuels provides a 
significant reduction in diesel engine emissions over a broad range of 
applications. Currently the biggest impediment to the use of the 
emulsified fuel is distribution. The manufacturer is making efforts to 
make the fuel more widely available.
    MSHA has not tested the fuel at high altitude mines (above 5000 
feet). At these elevations there are potential problems due to 
additional horsepower loss, steep grades and low winter temperatures. 
MSHA is working with the fuel manufacturer and mining industry to 
evaluate these concerns.
    Combining DPM Controls Into An Overall Strategy. The DPM rule 
allows mine operators flexibility in choosing engineering and 
administrative controls that are appropriate for site-specific 
conditions and operating practices. During its compliance assistance 
visits, MSHA urged mine operators to combine various engineering and 
administrative controls, including work practices, into an integrated 
DPM control strategy for their mines. For example, in stone mines where 
haulage trucks transport broken stone out of the mine to a surface 
crusher, and where the truck drivers are protected by effective 
environmental cabs with filtered breathing air, MSHA recommends that 
the main ramp used by the haulage trucks to travel out of the mine be 
maintained as an exhaust air course. Typically, the combined horsepower 
of the production loader and haulage trucks at a stone mine exceeds the 
horsepower of all other equipment combined. When haulage trucks travel 
loaded upgrade out of the mine, they generate significant amounts of 
DPM. If the ramp used by these trucks is maintained as an intake air 
course, the fresh air supply for the entire mine can become 
contaminated. Maintaining this ramp as an exhaust air course and 
requiring the loaded trucks to haul up this ramp as an administrative 
control enables the mine operator to provide better ventilation air 
quality along the face line. Depending on mine layout and ventilation, 
it may be possible to maintain all ramps traveled by the haulage trucks 
as exhaust air courses. It is especially important, however, that the 
ramps used for upgrade loaded haulage be maintained as exhaust air 
courses. This combination of engineering (cabs and ventilation) and 
administrative controls (loaded trucks haul up the ramps used as 
exhaust air course) particularly benefits powder crew workers who are 
required to work most of their shift outside of a protective cab.
    Some commenters stated that the industry has exhausted the ``easy'' 
methods of DPM control, and reducing DPM to lower limits would be 
prohibitively expensive. MSHA is not entirely certain what is meant by 
``easy'' methods, but suspects the commenter was referring to DPM 
controls other than major ventilation upgrades (new main fans, new 
ventilation shafts, etc.) and DPFs, which are either more costly than 
other options, or are perceived as more costly. At some mines, ``easy'' 
could also mean ``familiar,'' indicating the methods and strategies 
with which these mine operators have had actual first-hand experience. 
Based on this meaning, easy upgrades appear to be: Ventilation fans 
(main or booster), airflow distribution systems, environmental cabs, 
modern engines and alternate fuels.
    By either definition, MSHA believes that only a small portion of 
the industry has exhausted these control methods. For example, based on 
compliance assistance mine visits, baseline sampling results, and other 
data, MSHA

[[Page 32939]]

has observed that many mines have not yet implemented relatively low 
cost ventilation upgrades, and that at most mines that have initiated 
such programs, not all necessary upgrades have been completed.
    Another example involves environmental cabs with filtered breathing 
air. As noted above, even though most major pieces of production 
equipment in stone mines are provided with cabs, the corresponding 
health benefits are seldom fully realized due to open or broken 
windows, company policies that permit equipment to be operated with its 
doors open, inoperative or poorly maintained AC systems and cab 
pressurizing fans, damaged door seal gaskets, etc.
    A final example relates to the failure to employ effective work 
practices such as utilizing return air courses as truck haulage roads 
when the truck drivers are protected by environmental cabs with 
filtered breathing air.
    MSHA determined that compliance costs were economically feasible 
for the M/NM mining industry. In the REA for the 2001 final DPM rule, 
MSHA determined that annual compliance costs would be about $128,000 
for an average underground M/NM mine. Some mines, in particular mine 
size and commodity groups, because of mining methods used, equipment 
deployments, etc., would be expected to incur higher than average 
compliance costs. For example, the REA estimated yearly compliance 
costs for large precious metals mines to be $660,000. Based on its 
compliance assistance mine visits, baseline sampling results, and other 
data, MSHA believes that most mines have expended far less than the 
expected $128,000 yearly for DPM compliance. Though expenditures will 
undoubtedly need to rise in the future as the familiar and less costly 
DPM control methods are exhausted, they are not expected to exceed 
levels previously determined by MSHA to be economically feasible.

C. Economic Feasibility

    MSHA has determined that a PEL of 308 micrograms per cubic meter of 
air (308EC [mu]g/m3) is economically feasible for 
the M/NM mining industry. Economic feasibility does not guarantee the 
continued viability of individual employers, but instead, considers the 
industry in its entirety. It would not be inconsistent with the Mine 
Act to have a company which turned a profit by lagging behind the rest 
of an industry in providing for the health and safety of its workers to 
consequently find itself financially unable to comply with a new 
standard; See United Steelworkers of America v. Marshall, 647 F.2d 
1189, 1265 (1980). Although it was not Congress' intent to protect 
workers by putting their employers out of business, the increase in 
production costs or the decrease in profits would not be sufficient to 
strike down a standard. See Industrial Union Dep't., 499 F.2d at 477. 
On the contrary, a standard would not be considered economically 
feasible if an entire industry's competitive structure were threatened. 
Id. at 478; See also, AISI-II, 939 F.2d 975, 980 (DC Cir. 1991); United 
Steelworkers, 647 F.2d at 1264-65; AISI-I, 577 F.2d 825, 835-36 (1978). 
This would be of particular concern in the case of foreign competition, 
if American companies were unable to compete with imports or substitute 
products. The cost to government and the public, adequacy of supply, 
questions of employment, and utilization of energy may all be 
considered when analyzing feasibility.
    MSHA has also determined that there will be a small cost savings in 
economic impact on the mining industry under this final rule, because 
the requirements for meeting the PEL are similar to those in the 
existing DPM enforcement policy for the 2001 DPM standard. 
Specifically, MSHA will continue to require mine operators to 
establish, use and maintain all feasible engineering and administrative 
control methods to reduce a miner's exposure to the PEL. The final rule 
affords mine operators the flexibility to choose either engineering or 
administrative controls, or a combination of controls to reduce a 
miner's exposure. In the event that controls do not reduce a miner's 
exposure to the PEL, are not feasible, or do not produce significant 
reductions in DPM exposures, the operator must use and maintain 
controls to reduce the miner's exposure to as low as feasible and 
supplement controls with respiratory protection. Mine operators must 
establish a respiratory protection program when controls are 
infeasible. If MSHA confirms that mine operators have met all of the 
abovementioned requirements for addressing a miner's overexposure, and 
the miner's exposure continues to exceed the PEL (not counting 
respirators), MSHA will not issue a citation for an overexposure. 
Instead, MSHA will continue to monitor the circumstances leading to the 
miner's overexposure, and as controls become feasible, MSHA will 
require the mine operator to install and maintain them to reduce the 
miner's exposure to the PEL.
    MSHA believes that it has established in this final rulemaking that 
the new interim PEL is comparable to the TC interim concentration 
limit. Therefore, in determining the economic feasibility of 
engineering and administrative controls that the M/NM underground 
industry will have to use under this final rule, MSHA evaluated the 
cost of controls that are used to comply with the existing DPM TC 
interim concentration limit to that of the newly promulgated EC interim 
PEL. These controls include DPFs, ventilation upgrades, oxidation 
catalytic converters, alternative fuels, fuel additives, enclosures 
such as cabs and booths, improved maintenance procedures, newer 
engines, various work practices and administrative controls. MSHA's 
evaluation includes costs of retrofitting existing diesel-powered 
equipment and regeneration of DPFs.
    On the basis of evidence in the rulemaking record, including MSHA's 
current enforcement experience, MSHA has determined that this final 
rule results in a cost savings of $3,634 per year, primarily due to 
MSHA's determination to delete the DPM control plan.
    In highly unusual circumstances where the use of further controls 
may not be economically viable, the standard provides for a hierarchy 
of control strategy that allows specifically for the cost impact to be 
considered on a case-by-case basis. MSHA's DPM enforcement policy, 
therefore, takes into account the financial hardship on an 
individualized basis which MSHA believes effectively accommodates mine 
operator's economic concerns, particularly those of small mine 
operators.
    Whether controls are feasible for individual mine operators is 
based in part upon legal guidance from decisions of the independent 
Federal Mine Safety and Health Review Commission (Commission) involving 
enforcement of MSHA's noise standards for M/NM mines, 30 CFR 56.5-50 
(revised and recodified at 30 CFR 62.130). According to the Commission, 
a control is feasible when it: (1) Reduces exposure; (2) is 
economically achievable; and (3) is technologically achievable. See 
Secretary of Labor v. A. H. Smith, 6 FMSHRC 199, 201-02 (1984); 
Secretary of Labor v. Callanan Industries, Inc., 5 FMSHRC 1900, 1907-09 
(1983).
    In determining the economic feasibility of an engineering control, 
the Commission has ruled that MSHA must assess whether the costs of the 
control are disproportionate to the ``expected benefits,'' and whether 
the costs are so great that it is irrational to require implementation 
of the control to achieve those results. The Commission has expressly 
stated that cost-benefit analysis is unnecessary to determine whether a 
control is required.

[[Page 32940]]

    Consistent with Commission case law, MSHA considers three factors 
in determining whether engineering controls are feasible at a 
particular mine: (1) The nature and extent of the overexposure; (2) the 
demonstrated effectiveness of available technology; and (3) whether the 
committed resources are wholly out of proportion to the expected 
results. A violation under the final standard will entail an agency 
determination that a miner was overexposed, that controls are feasible, 
and that the mine operator failed to install or maintain such controls. 
According to the Commission, an engineering control may be feasible 
even though it fails to reduce exposure to permissible levels contained 
in the standard, as long as there is a significant reduction in a 
miner's exposure. Todilto Exploration and Development Corporation v. 
Secretary of Labor, 5 FMSHRC 1894, 1897 (1983).
    MSHA will consistently utilize its longstanding enforcement 
procedures under its other exposure-based standards at M/NM mines. As a 
result, MSHA will consider the total cost of the control or combination 
of controls relative to the expected benefits from implementation of 
the control or combination of controls when determining whether the 
costs are wholly out of proportion to results. If controls are capable 
of achieving a 25% reduction, MSHA will evaluate the cost of controls 
and determine whether their costs would be a rational expenditure to 
achieve the expected results.
    MSHA emphasizes that the concept of ``a combination of controls'' 
is not new to the mining industry. It is MSHA's consistent practice not 
to cost controls individually, but rather, combine their expected 
results to determine if the 25% significant reduction criteria, as 
discussed earlier in this section, can be satisfied.
    MSHA heavily weighs the potential benefits to miners' health when 
considering economic feasibility and does not conclude economic 
infeasibility merely because controls are expensive. Mine operators 
have the responsibility for demonstrating to MSHA that technologically 
feasible controls are so costly as to result in a significant economic 
hardship.
    In situations where MSHA finds that the mine operator has not 
installed all feasible controls, MSHA will issue a citation and 
establish a reasonable abatement date. Based on a mine's technological 
or economic circumstances, the standard gives MSHA the flexibility to 
extend the period within which a violation must be corrected. If a 
particular mine operator is cited for violating the DPM PEL, but that 
operator believes that the standard is technologically or economically 
infeasible for that operation, the operator ultimately can challenge 
the citation in an enforcement proceeding before the independent 
Commission.
    MSHA found that most of the practical and effective DPM controls 
that are available, such as DPFs, enclosed cabs with filtered breathing 
air, alternative diesel fuels, and low-emission engines, will achieve 
at least a 25% reduction in DPM exposure. Though this final rule 
affords each mine operator the flexibility to select the DPM control or 
combination of controls that are appropriate to their site-specific 
conditions, MSHA believes that the most cost effective DPM controls are 
DPF systems. MSHA believes that there are a number of available DPFs 
that do not increase production of NO2.
    MSHA estimates that DPFs for the M/NM underground mining industry 
range in cost from $5,000 to $12,000 per filter. This range of cost is 
consistent with the reported DPF costs from the NIOSH Phase I Study. A 
typical example is a 15 x 15 Engelhard DPX 
platinum-catalyzed DPF used on 475 horsepower haulage trucks at a 
multilevel metal mine in Alaska that costs $8,700.
    The average life expectancy of a DPF is approximately 8,000 hours. 
Some commenters, however, have reported life expectancies of between 
2,000 and 4,000 hours, while some other commenters have reported life 
expectancies for longer than 8,000 hours. However, in most of these 
cases the shortened DPF life was due to a malfunction of another piece 
of equipment, installation problems or a manufacturer's defect, 
depending on the type of DPF selected by an operator. MSHA's 8,000 hour 
estimate is based on an operation and maintenance guide prepared by DCL 
Incorporated and two technical papers given at the Mining Diesel 
Emission Conference in Toronto, Canada, November 1999. (See MSHA's REA 
for 2001 final rule.) Support for this estimate is provided by NIOSH in 
its publication titled ``Review Technology Available to the Underground 
Mining Industry for Control of Diesel Emissions'' (George H. 
Schnakenberg, PhD, Information Circular 9462, 2002) which reports that 
average ceramic DPF service life at Agrium's Canadian potash mines is 5 
years. This publication also references reports of a few Engelhard DPFs 
that have been in service 10 years.
    MSHA believes that the requirements for engineering and 
administrative controls clearly meet the feasibility requirements of 
the Mine Act, its legislative history and related case law.
    The trends in DPM control technology development to date, 
especially DPFs, indicate that manufacturers are creating more 
innovative designs. MSHA believes that more cost effective control 
methods are on the horizon. This reasoning is supported by a recently 
published EPA final rule for the control of emissions from nonroad 
diesel engines. The ``Clean Air Nonroad Diesel--Final Rule'' (Control 
of Emissions of Air Pollution from Nonroad Diesel Engines and Fuel, 69 
FR 38958 (2004)) sets emission standards for airborne contaminants, 
including DPM, for all diesel engine horsepower ranges. For engines up 
to 750 horsepower, the requirements will be phased in from 2008 through 
2014. For engines above 750 horsepower, the final compliance date is 
extended to 2015. EPA's Clean Air Nonroad Diesel Rule is a 
comprehensive national program to reduce emissions from future non-road 
diesel engines used in industries such as construction, agriculture and 
mining. To meet these emission standards, engine manufacturers will 
produce new engines with advanced emission-control technologies similar 
to catalytic technologies used in passenger cars. Exhaust emissions 
from these engines will decrease by more than 90%. Because the 
emission-control devices can be damaged by sulfur, the EPA is also 
adopting a limit to decrease the allowable level of sulfur in nonroad 
diesel fuel by more than 99% from current levels (from approximately 
3,000 parts per million [ppm] now to 15 ppm in 2010). This will be 
consistent with the on-highway fuel sulfur requirements. New engine 
standards take effect, based on engine horsepower, starting in 2008. 
Both the EPA and the diesel engine manufacturers agree that clean 
engine technology alone cannot achieve EPA's newly mandated emission 
limits; manufacturers will also have to use advanced technology options 
such as DPFs.
    MSHA believes DPFs are currently commercially available for any 
engine, application, or duty cycle used in underground M/NM mining. 
These new EPA rules, however, will undoubtedly be technology forcing 
and result in an increase in the variety, features, and capabilities of 
DPFs from which mine operators may choose, as well as lower the cost of 
DPFs and promote other technological innovation in this field.
    In spite of these trends in new technology, MSHA recognizes that, 
in a few cases, individual mine operators, particularly small 
operators, may have economic difficulty in achieving full

[[Page 32941]]

compliance with the interim limit immediately because of a lack of 
financial resources to purchase and install engineering controls. 
MSHA's revised enforcement strategy is designed to accommodate this 
problem. Under this enforcement strategy, MSHA allows mine operators 
with feasibility issues the necessary time to reduce exposures to the 
interim PEL.
    MSHA also has demonstrated that the effective date for this final 
rule does not pose an economic burden for underground M/NM mine 
operators. As stated earlier, the EC surrogate standard is comparable 
to the existing TC surrogate standard which has been in effect since 
July 2002, and has been enforced by MSHA since July 20, 2003. 
Consequently, MSHA cannot justify affording mine operators additional 
time to comply with an exposure limit currently enforced. MSHA believes 
that the startup date is justified by the rulemaking record and the 
mining industry's present capability of complying with the existing 
interim limit.
    Moreover, MSHA has afforded the underground M/NM mining industry 
additional consideration in relieving the financial impact of this 
final rule by delaying the period of time that was allowed for 
compliance with the 2001 comparable TC concentration limit. In response 
to concerns raised by the mining industry and the terms of the DPM 
settlement agreement, MSHA allowed as much as 2\1/2\ years for a DPM 
compliance phase-in strategy.
    Specifically, on March 15, 2001, MSHA published a notice delaying 
the effective date of the final DPM rule of January 19, 2001, (66 FR 
5706) until May 21, 2001 (66 FR 15032). By notice of May 21, 2001, (66 
FR 27863), MSHA delayed the final rule another 45 days, until July 5, 
2001. Furthermore, by notice of July 5, 2001, (67 FR 9180), MSHA 
delayed Sec.  57.5066(b), Maintenance standards, relating to 
``tagging'' requirements. MSHA also clarified that the interim 
concentration limit at Sec.  57.5060(a) and its related provisions in 
the final rule would not apply until after July 19, 2002, pursuant to 
its original effective date. By notice of July 18, 2002, MSHA stayed 
the effectiveness of: Sec.  57.5060(d), permitting miners to work in 
areas where DPM exceeds the applicable concentration limit with advance 
approval from the Secretary; Sec.  57.5060(e), prohibiting the use of 
PPE to comply with the concentration limits; Sec.  57.5060(f), 
prohibiting the use of administrative controls to comply with the 
concentration limits; and, Sec.  57.5062, addressing the DPM control 
plan. These provisions were stayed pending completion of this final 
rule.
    Finally, in the DPM settlement agreement, MSHA agreed to enforce: 
Sec.  57.5060(a), addressing the interim concentration of 400 
micrograms of TC per cubic meter of air; Sec.  57.5061, addressing 
compliance determinations; Sec.  57.5070, addressing miner training; 
and Sec.  57.5071, addressing environmental monitoring. However, to 
further assist the mining industry in instituting engineering controls, 
MSHA gave the mining industry an additional year, from July 20, 2002, 
until July 20, 2003, to begin to develop a written strategy of how they 
intended to comply with the interim DPM concentration limit. Operators 
with DPM levels above the concentration limit were to begin to order 
and install controls to reduce miners' exposures by July 20, 2003. 
Concurrently, MSHA provided comprehensive compliance assistance to M/NM 
underground operators. MSHA retained the discretion to take appropriate 
enforcement actions against operators who refuse either to cooperate in 
good faith with MSHA's compliance assistance, or to take good-faith 
steps to develop and implement a written compliance strategy for their 
mines. Mine operators had the obligation to develop a strategy to 
control DPM emissions and order engineering controls. MSHA began 
enforcing the interim limit at M/NM underground mines on July 20, 2003, 
under the terms of the settlement agreement.
    MSHA received a number of comments in response to its proposed 
economic feasibility discussion. Several commenters wanted MSHA to 
define ``economic feasibility.'' They believe that controls should be 
considered economically feasible if implementation would not bankrupt 
the company or force the mine to close. They also believe that MSHA's 
2003 NPRM did not indicate how MSHA will enforce the new language and 
wanted access to records of feasibility determinations made by MSHA. 
MSHA has chosen not to define ``economic feasibility'' nor 
``technological feasibility'' since the Supreme Court has done so in 
the OSHA Cotton Dust decision. As stated earlier in this part, the 
Supreme Court defined ``feasibility'' as ``capable of being done'' 
(American Textile Manufacturers' Institute v. Donovan (OSHA Cotton 
Dust), 452 U.S. 490, 508-509 (1981)). This preamble also discusses how 
the independent Commission explains the Secretary's burden of proof in 
establishing technological and economic feasibility of controls.
    Commenters criticized the high costs of DPM controls associated 
with attempts to achieve a significant reduction. These commenters 
stated that mine ventilation systems cost more than $100 million and 
provide a benefit only of a 3% to 4% DPM reduction, whereas a less-than 
$100 million administrative control could achieve a 21% to 22% 
reduction.
    First, MSHA disputes the assertion that a ventilation system costs 
$100 million. MSHA assumes mines already have some form of ventilation, 
since ventilation is needed whether or not DPM is a consideration. The 
existing system may be minimal, and rely partly or largely on natural 
ventilation, but a basic ventilation network must be present per 
existing MSHA ventilation regulations (Sec.  57.8518 through Sec.  
57.8535) and air quality standards (Sec.  57.5001 through Sec.  
57.5039) to support normal mining operations. Thus, in the context of 
the final rule, the question is not whether a ventilation system needs 
to be provided for compliance, but rather, whether an upgrade to an 
existing ventilation system is needed. If so, mine operators must 
examine whether major additions (new shaft, new main fan, etc.) are 
required, versus relatively minor improvements such as booster fans, 
auxiliary ventilation system upgrades, or repair or extensions to 
existing ventilation control structures. Even in an extreme case where 
a new ventilation shaft and main fan installation could be justified 
solely on the basis of DPM compliance, such upgrades cost far less than 
$100 million. Costs in the range of $5 million to as much as $20 
million would be more accurate.
    MSHA also notes that the level of DPM reduction obtained through a 
ventilation upgrade is proportional to the ratio of new ventilation air 
flow to the existing ventilation air flow. If overall air flow is 
doubled, DPM levels would be roughly cut in half. Of course factors 
such as imperfect mixing and effective distribution of air flow 
underground would ultimately determine the actual DPM reduction 
achieved. Major ventilation upgrades costing $5 to $20 million would 
typically result in DPM reductions of at least 20% to 30% or more, 
which is far greater than the 3% to 4% reduction that commenters 
estimated for a ventilation upgrade costing $100 million.
    It is also significant to note that some DPM controls that may be 
easier fixes for controlling DPM exposures may actually be quite high 
in overall life-cycle costs compared to other approaches that mine 
operators perceive

[[Page 32942]]

to be higher cost options. For example, if the operator of a stone mine 
determined that compliance could be achieved by installing a 150 
horsepower fan costing $25,000, this control option might appear to be 
advantageous compared to installing DPFs with an expected filter life 
of two years on the mine's production loader and three haulage trucks 
at a cost of $60,000 (4 filters x $15,000 per filter = $60,000). 
However, if the total cost of the ventilation upgrade is considered, 
including power costs to operate the fan 12 hours per day 6 days per 
week, the annual cost for ventilation surpasses the cost for filters. 
The $60,000 cost for DPFs, annualized over the two-year filter life is 
$33,186 (using a 7% discount rate). The fan power cost alone would be 
over $40,000 annually at $0.10 per kilowatt-hour (150hp x 12 hours/day 
x 6 days/week x 52 weeks/year x 0.745 kw-hr/hp-hr x .10 $/kw-hr).
    One commenter suggested that MSHA's failure to specify major 
ventilation upgrades for any mine in its 31-Mine Study results in a 
serious underestimate of compliance costs for those mines and the 
industry as a whole. This commenter states that the trona mines have 
already attained compliance with the final limit because of their high 
ventilation air flow rates, and that similarly high flows will be 
required at many other mines to attain compliance.
    MSHA notes that the final rule is performance oriented, and allows 
mine operators great latitude to choose the DPM control or controls 
that are most efficient and cost effective for a given mine. The trona 
mines are required to ventilate at very high rates for reasons other 
than DPM compliance to address methane issues, for instance. For them, 
ventilation is the logical DPM control because the control is already 
in place. Other type mines have more and varied choices, and selecting 
the optimum DPM control strategy involves evaluation of a broad range 
of factors such as current DPM levels, equipment and engines used, 
equipment deployments, mine layout, existing ventilation system, 
availability of alternate diesel fuels, and many more.
    For reasons of financial self-interest, mine operators would be 
unwise to implement high cost controls that achieve very little DPM 
reduction, such as a $100 million ventilation system that reduces DPM 
levels by only 3% to 4%. Such a choice would preclude less costly and 
more effective options available, such as DPFs, low emission engines, 
alternative diesel fuels, and cabs with filtered breathing air.
    As stated earlier, the final rule incorporates economic feasibility 
in its hierarchy of controls enforcement scheme. MSHA, likewise, could 
not require a mine operator to implement a control or combination of 
controls where the costs are wholly out of proportion to the expected 
results. MSHA would judge a ventilation upgrade costing $100 million, 
or even $5 to $20 million that achieves a DPM reduction of 3% to 4% as 
infeasible because the cost is wholly out of proportion to the expected 
results, and it is likely a mine operator would consider it a poor DPM 
compliance strategy for the same reason. The commenter suggests a lower 
cost administrative control that achieves a 21% to 22% reduction would 
be a better choice. MSHA agrees, if this control in combination with 
other controls would result in at least a 25% reduction.
    As noted previously, with some DPFs, filter efficiency is as high 
as 99+% for EC. MSHA, however, believes that both economic and 
technological feasibility must be considered. Whereas filter efficiency 
is a major component of technological feasibility, MSHA must consider 
all aspects of feasibility including implementation issues and cost of 
compliance to the mining industry. As stated earlier in this preamble, 
MSHA believes that some mine operators would need more time to meet a 
lower DPM limit presently based on economic feasibility and 
implementation issues with DPFs.
    Establishing a lower interim limit in this final rule would present 
complications with respect to economic feasibility, particularly where 
ventilation upgrades would be needed to meet a lower limit. Moreover, 
MSHA envisions that mine operators would have to filter larger numbers 
of diesel-powered equipment in order to meet a lower limit. Such a 
requirement could impose higher costs for the mining industry before 
experience is gained at the current level and the mining industry is 
given adequate time to meet a lower standard.
    Some commenters objected to MSHA's assessment of the number of 
mining operations that will need costly ventilation upgrades. These 
operators believe that a large number of mines will have to make 
ventilation improvements, provide cab improvements, add other 
engineering controls, implement other administrative controls, replace 
engines, and utilize DPFs. In response, the DPM rulemaking record does 
not sustain this position. MSHA found in its baseline sampling that 
only 37% of the mining operations covered by this DPM rule had miners 
overexposed to DPM. Consequently, at 63% of the mines sampled, MSHA 
found no overexposures to DPM. MSHA conducted this sampling in the same 
manner as it does its enforcement of the 2001 interim limit DPM rule. 
MSHA collected roughly 1,194 samples at 183 mines. Additionally, MSHA 
responded to each mine operator's request for compliance assistance and 
technical support for resolving engineering control implementation 
issues. The results of MSHA's work are included in the rulemaking 
record. Overall, the mining industry has been successful in reducing 
average DPM levels as demonstrated in the comparison of baseline 
sampling and 31-Mine Study data shown in Chart V-5.
    Also, in the 31-Mine Study, MSHA established that most mining 
operations would not need major ventilation changes, but rather, could 
implement less costly ventilation upgrades and DPFs. In most instances, 
the ventilation upgrades require no more than adding booster fans or 
auxiliary ventilation, and repairs or extensions to ventilation control 
structures such as brattice lines or air walls.
    A commenter suggested that ventilation costs for complying with the 
DPM rule for the Kerford Limestone mine were projected to be $1.15 
million, plus $348,450 for engine replacements, plus an additional 
$25,500 to $38,500 for DPF maintenance. According to the commenter, 
this mine has invested $975,000 since October 2001, primarily for 
ventilation improvements including sinking a shaft, consultant costs, a 
new blasting truck, and a new engine for a bolter. The commenter points 
out that in the 31-Mine Study, MSHA projected that first-year 
compliance costs for this same mine would be only $77,600, and suggests 
the discrepancy is an example of MSHA's underestimate of DPM compliance 
costs.
    MSHA notes that 13 DPM samples were taken during the 31-Mine Study 
at the Kerford mine. Sample results ranged from 143TC [mu]g/
m\3\ to 490TC [mu]g/m\3\. Per the 31-Mine Study methodology, 
DPM controls were specified based on the highest sample result. 
However, since the highest sample result only exceeded the interim DPM 
limit by about 23% (490TC [mu]g/m\3\ versus the interim DPM 
limit of 400TC [mu]g/m\3\), the controls necessary to attain 
compliance at this mine were not very extensive. Indeed, MSHA's 
analysis indicated that controlling DPM emissions from the mine's three 
loaders (two loaders used in normal operations plus one spare) using 
active DPF systems with filter efficiencies of 80% would enable the

[[Page 32943]]

mine to attain compliance with the interim limit. MSHA estimated the 
first year cost of three filter systems for the subject loaders plus an 
oven for regenerating the filters (active off-board regeneration) to be 
$77,600.
    MSHA has not seen the consultant's report that indicates new 
engines, DPFs, and a major ventilation upgrade would be required for 
the Kerford mine to comply with the interim DPM limit. However, these 
recommendations appear excessive based on MSHA's analysis in the 31-
Mine Study and also on the fact that compliance for this mine requires 
only a relatively small reduction in DPM levels from 490TC 
[mu]g/m\3\ to 400TC [mu]g/m\3\.
    As noted in the 31-Mine Study final report, MSHA is not suggesting 
that its findings represent the optimum compliance strategy for this or 
any mine. Rather, MSHA maintained merely that the controls specified in 
the final report are feasible and would be expected to attain 
compliance. MSHA suspects that the combination of controls recommended 
by Kerford's consultant, though capable of attaining compliance, is not 
the optimum and most cost effective approach available.
    As discussed in the Technological Feasibility section of this 
preamble, MSHA also notes that the total yearly cost represented by the 
consultant's recommended engine, ventilation system, and DPF 
expenditures is roughly in line with MSHA's 2001 REA estimate for an 
average mine, even though Kerford Limestone is substantially larger 
than average. The engine costs of $348,000, when annualized over 10 
years at a discount rate of 7%, results in a yearly cost of $49,500. 
The $1.15 million ventilation cost, when annualized over the expected 
20+ year life of this asset, results in a yearly cost of $108,600. When 
these two yearly costs are added to the maximum estimated annual DPF 
cost of $38,500, the total yearly cost for Kerford is about $196,600. 
When compared to the MSHA REA's estimated compliance cost of over 
$300,000 for a stone mine of this size, Kerford's costs are 
significantly less.
    Some mines, in particular mine size and commodity groups, because 
of their mining methods used, equipment deployments, etc., would be 
expected to incur higher than average compliance costs of $128,000 per 
year. For example, the REA estimated yearly compliance costs for large 
precious metals mines to be $660,000. Based on its compliance 
assistance mine visits, baseline sampling results, and other data, MSHA 
believes that most mines have expended far less than the expected 
$128,000 yearly for DPM compliance. Though expenditures will 
undoubtedly need to rise in the future as the easy DPM control methods 
are exhausted, they are not expected to exceed levels previously 
determined by MSHA to be economically feasible.
    Another mine that disputed MSHA's estimated DPM compliance cost 
estimates is the Stillwater Mine. MSHA estimated in the 31-Mine Study 
that DPM filters would be required on all LHDs and haulage trucks at 
this mine in order to attain compliance with the interim limit. 
Accordingly, MSHA estimated Stillwater's first year costs to be 
$470,100 and annual costs to be $108,163 for three loaders and twelve 
trucks used in normal mining production operations plus three more 
spare loaders and four more spare trucks. In its comments on the 2003 
NPRM, Stillwater indicated that its total diesel equipment inventory 
consists of over 350 pieces of diesel equipment, including over 90 
loaders and 40 haulage trucks, plus miscellaneous production equipment 
and spares. MSHA has since acknowledged that it had an inaccurate 
inventory of diesel equipment for the Stillwater mine when the 31-Mine 
Study was conducted. On the basis of the newly obtained inventory data, 
MSHA raised its compliance cost estimate for this mine to $935,000 to 
cover DPFs for the total production fleet.
    In its comments on the 2003 NPRM, Stillwater submitted its own 
compliance cost estimates. This estimate included a $9 million 
ventilation upgrade, $160,000 for passive DPFs, $1.2 million for engine 
upgrades, $280,000 for engine test equipment, $43,000 per month in 
emissions expenditures, over $100 million over ten years for active 
DPFs, plus various miscellaneous costs. Combining these items resulted 
in an estimated annual compliance cost for Stillwater of $11 to $12 
million.
    Clearly, the most significant cost item listed by Stillwater is 
active DPF systems. However, almost 97% of Stillwater's estimated 
active DPF systems costs are for excavation of parking areas. 
Stillwater's active DPF system implementation plan specified on-board 
active filter regeneration, wherein a vehicle would travel to a 
regeneration station and its DPF would be connected to electrical power 
and compressed air for regeneration. To insure reasonable travel 
distances between normal working areas and regeneration stations, 
Stillwater's active filter cost estimate was developed in the context 
of a ten-year mine plan, wherein new regeneration stations would be 
excavated periodically with the advance of the mine workings.
    As discussed in detail in the Technological Feasibility section of 
this preamble, MSHA analyzed and evaluated the Stillwater compliance 
cost estimate, and determined that compliance could be attained at a 
much lower cost. Since the cost of excavating regeneration stations was 
such a significant component of Stillwater's overall cost estimate, 
MSHA focused on eliminating this cost element. As explained in the 
Technological Feasibility section, MSHA described three feasible 
alternative approaches for utilizing active filtration that do not 
require excavation of regeneration station parking areas. Although MSHA 
disputed several of the remaining cost items, MSHA nonetheless accepted 
these costs as submitted by Stillwater in developing an alternate 
compliance cost estimate for this mine. The inclusion of these disputed 
items accounts for MSHA's estimated compliance cost of $1.57 million 
for the Stillwater mine being somewhat higher than the revised 31-Mine 
Study cost estimate of $935,000.
    As noted in the Technological Feasibility section of this preamble, 
MSHA's estimate of $1.57 million in annual DPM compliance cost is 
significant. However, it is less than MSHA estimated in the REA for the 
2001 final DPM rule for a large precious metals mine. The REA estimated 
annual compliance costs of $660,000 based on a fleet size of 133 
vehicles. Adjustment for Stillwater's fleet size of 350+ vehicles 
results in an estimated compliance cost of $1.7 million.
    Several other commenters suggested that MSHA's compliance cost 
estimates, in general, were unrealistically low. However, without 
specific examples to evaluate and analyze, such comments are difficult 
to refute. MSHA has supported its cost estimating methodologies in 
general, and where specific examples have been provided by commenters, 
MSHA has fully supported its compliance cost estimates, such as the