[Federal Register: November 30, 2004 (Volume 69, Number 229)]
[Rules and Regulations]
[Page 69747-69774]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr30no04-16]
[[Page 69747]]
-----------------------------------------------------------------------
Part IV
Department of Agriculture
-----------------------------------------------------------------------
Animal and Plant Health Inspection Service
-----------------------------------------------------------------------
7 CFR Part 319
Mexican Avocado Import Program; Final Rule
[[Page 69748]]
-----------------------------------------------------------------------
DEPARTMENT OF AGRICULTURE
Animal and Plant Health Inspection Service
7 CFR Part 319
[Docket No. 03-022-5]
RIN 0579-AB81
Mexican Avocado Import Program
AGENCY: Animal and Plant Health Inspection Service, USDA.
ACTION: Final rule.
-----------------------------------------------------------------------
SUMMARY: We are amending the regulations governing the importation of
fruits and vegetables to expand the number of States in which fresh
Hass avocado fruit grown in approved orchards in approved
municipalities in Michoacan, Mexico, may be distributed. We are also
allowing the distribution of the avocados during all months of the
year. For the first 2 years following the effective date of this rule,
those avocados may be distributed in all States except California,
Florida, and Hawaii; after 2 years, the avocados may be distributed in
all States. We are taking this action in response to a request from the
Government of Mexico and based on our finding that the phytosanitary
measures described in this final rule will reduce the risk of
introducing plant pests associated with Mexican Hass avocados into the
United States.
DATES: Effective Date: January 31, 2005.
FOR FURTHER INFORMATION CONTACT: Ms. Karen Bedigian, Import Specialist,
Phytosanitary Issues Management Team, PPQ, APHIS, 4700 River Road Unit
140, Riverdale, MD 20737-1236; (301) 734-6799.
SUPPLEMENTARY INFORMATION:
Background
The regulations in ``Subpart--Fruits and Vegetables'' (7 CFR 319.56
through 319.56-8) prohibit or restrict the importation of fruits and
vegetables into the United States from certain parts of the world to
prevent the introduction and dissemination of plant pests, including
fruit flies, that are new to or not widely distributed within the
United States.
The regulations in 7 CFR 319.56-2ff (referred to below as the
regulations) have provided for the importation of fresh Hass avocado
fruit grown in approved orchards in approved municipalities in
Michoacan, Mexico, into specified areas of the United States, subject
to certain conditions. Those conditions, which include pest surveys and
pest risk-reducing cultural practices, packinghouse procedures,
inspection and shipping procedures, and restrictions on the time of
year (October 15 through April 15) that shipments may enter the United
States, are designed to reduce the risk of pest introduction. Further,
the regulations have limited the distribution of the avocados to 31
northeastern and north central States (Colorado, Connecticut, Delaware,
Idaho, Illinois, Indiana, Iowa, Kansas, Kentucky, Maine, Maryland,
Massachusetts, Michigan, Minnesota, Missouri, Montana, Nebraska, New
Hampshire, New Jersey, New York, North Dakota, Ohio, Pennsylvania,
Rhode Island, South Dakota, Utah, Vermont, Virginia, West Virginia,
Wisconsin, and Wyoming) and the District of Columbia.
In November 2000, the Government of Mexico requested that the
Animal and Plant Health Inspection Service (APHIS) amend the
regulations to allow Hass avocados to be imported year round into all
50 States. We did not act on Mexico's request at the time because we
did not have documentation available to support Mexico's position that
such importations would not present a risk of introducing plant pests
into certain States.
As part of our evaluation of Mexico's request, we prepared a draft
pest risk assessment (PRA), titled ``Importation of `Hass' Avocado
Fruit (Persea americana) from Mexico'' (June 2003), to evaluate the
importation of fruit to the entire United States throughout the year.
The draft PRA contained two components: (1) A risk assessment component
that identifies quarantine pests that are likely to follow the Mexican
Hass avocado import pathway, and (2) a risk management component that
evaluates the ability of the selected phytosanitary measures to
mitigate the risk posed by those quarantine pests.
The first component revealed that the quarantine pests of concern
remained the same as those identified in previous risk assessments.
After eliminating non-quarantine and non-pathway pests from the list,
eight pests of quarantine significance that follow the pathway remain:
Three fruit flies (Ceratitis capitata, Anastrepha ludens, A. striata),
three seed weevils (Conotrachelus aguacatae, C. perseae, and Heilipus
lauri), one stem weevil (Copturus aguacatae), and one seed moth
(Stenoma catenifer).
The second component of the draft PRA evaluated the selected
phytosanitary measures to mitigate the risk posed by the eight
identified pests. This component concluded that imports of Mexican
avocados subject to those phytosanitary requirements will result in the
following:
Fewer than 387 infested avocados will enter the United
States each year, estimated with 95 percent confidence.
Fewer than 49 avocados infested with stem weevil, seed
weevils, and seed moth will enter avocado producing areas each year,
estimated with 95 percent confidence.
Fewer than 208 avocados infested with fruit flies will
enter fruit fly susceptible areas each year, estimated with 95 percent
confidence.
Fewer than 3 avocados infested with stem weevil, seed
weevils and seed moth will be discarded in avocado producing areas each
year, estimated with 95 percent confidence.
Fewer than 11 avocados infested with fruit flies will be
discarded in fruit fly susceptible areas each year, estimated with 95
percent confidence.
There is an overall low likelihood of pest introduction.
Based on the statistical models we have used to estimate
sampling efficacy, it is slightly more likely that zero infested
avocados will enter the United States than one infested avocado;
however, we cannot rule out the possibility that some may enter the
country.
Only those avocados discarded in susceptible areas pose a risk of
establishment of the pests in the United States. In the PRA, the risk
associated with the importation of commercial shipments of avocados is
compared to the risks associated with infested avocados smuggled into
the United States. During the 17-year period from 1985 to 2002, an
average of 30 avocados each year (specific variety or cultivar not
recorded) infested with pathway pests were intercepted in baggage and
cargo and denied entry into the United States. Studies of port
efficiency, when searching for prohibited materials, indicate that
inspectors detect approximately 10 to 20 percent of what actually
arrives. That suggests that the number of prohibited avocados (i.e.,
smuggled or inadvertently imported non-program avocados) entering the
United States would average 150 to 300 per year.
While we state above that fewer than 387 infested avocados will
enter the United States each year, estimated with 95 percent
confidence, this number is based on statistical models. An examination
of over 10 million program fruit has not revealed any pests in 6 years
of fruit cutting and inspection and, also based on statistical models,
we determined that it is slightly more likely that zero infested
avocados will enter the United States than one infested
[[Page 69749]]
avocado. Prohibited transport of avocados in baggage and cargo poses a
substantially greater risk of introducing the above pests into the
United States than commercial imports of Hass avocados from Mexico.
Additionally, the 6 years' worth of data from the avocado import
program gives us confidence that the systems approach currently in
place provides adequate safeguards against avocado pests. The systems
approach mitigations include annual pest field surveys; orchard
certification; and packinghouse, packaging, and shipping requirements.
The efficacy of the systems approach depends on multiple measures.
Those measures are backed up by an inspection system that, when a pest
is detected, shuts down the imports from an affected area, depending on
the pest, until corrective actions are taken. An examination of over 10
million fruit has not revealed any pests in 6 years of fruit cutting
and inspection.
On May 24, 2004, we published in the Federal Register (69 FR 29466-
29477, Docket No. 03-022-3) a proposal to expand the number of States
in which fresh Hass avocado fruit grown in approved orchards in
approved municipalities in Michoacan, Mexico, may be distributed. We
also proposed to allow the distribution of the avocados during all
months of the year and to make other changes in the regulations, such
as removing restrictions on the ports through which the avocados may
enter the United States and the corridor through which the avocados
must transit the United States. We proposed this action in response to
a request from the Government of Mexico and based on our finding that
the phytosanitary measures described in this final rule will reduce the
risk of introducing plant pests associated with Mexican Hass avocados
into the United States.
We solicited comments concerning our proposal for 60 days ending
July 23, 2004. We received 17,022 comments by that date (including
11,000 form letters, both for and against the proposed rule). They were
from producers, exporters, researchers, members of Congress, and
representatives of State and foreign governments. They are discussed
below by topic.
After the comment period for the proposed rule closed on July 23,
2004, we updated the risk assessment \1\ based on comments that we
received. The updated risk assessment incorporates suggested changes to
the May 2004 version of the risk assessment that accompanied the
proposed rule and reflects new information received in public comments.
These changes include the following:
---------------------------------------------------------------------------
\1\ The updated risk assessment may be viewed on the Internet at
http://www.aphis.usda.gov/ppq/avocados and may be obtained from the
person listed under FOR FURTHER INFORMATION CONTACT.
---------------------------------------------------------------------------
A 47-State scenario was added in which the risk is
calculated for all States excluding California, Florida, and Hawaii.
Uncertainty was added to the estimate for sensitivity of
inspection in the model. The estimate of 50 percent was replaced with a
uniform distribution from 17.9 percent to 83.5 percent.
The estimate for the number of avocados imported was
changed for consistency with the economic analysis prepared for the
proposed rule.
Statistics including mean, mode, and standard deviation
were reported for all model output distributions.
Our new conclusions, based on the recalculations discussed above in
the second bullet, are as follows:
In the 50-State scenario, the risk assessment model results
present, with 95 percent confidence, the following estimates:
Fewer than 442 infested avocados will enter the entire
United States each year;
Fewer than 54 avocados infested with stem weevil, seed
weevils, and seed moth will enter avocado producing areas each year;
Fewer than 238 avocados infested with fruit flies will
enter fruit fly susceptible areas each year;
Fewer than three avocados infested with stem weevil, seed
weevils, and seed moth will be discarded in avocado producing areas
each year;
Fewer than 12 avocados infested with fruit flies will be
discarded in fruit fly susceptible areas each year.
Under the 50-State scenario, there is an overall low likelihood of
pest introduction.
In the 47-State scenario (excluding California, Florida, and
Hawaii), the risk assessment model results present, with 95 percent
confidence, the following estimates:
Fewer than 393 infested avocados will enter the 47 States
each year;
Fewer than seven avocados infested with stem weevil, seed
weevils, and seed moth will enter avocado producing areas outside of
California, Florida, and Hawaii each year;
Fewer than 98 avocados infested with fruit flies will
enter fruit fly susceptible areas outside of California, Florida, and
Hawaii each year;
Fewer than one avocado infested with stem weevil, seed
weevils, and seed moth will be discarded in avocado producing areas
outside of California, Florida, and Hawaii each year;
Fewer than five avocados infested with fruit flies will be
discarded in fruit fly susceptible areas outside of California,
Florida, and Hawaii each year; and
Under the 47-State scenario, there is an overall low likelihood of
pest introduction.
Even if some infested avocados entered the country, the likelihood
of pest establishment and spread would require that: (1) The infested
avocados must be in close proximity to host material; (2) the pests
must find mates; (3) the pests must successfully avoid predation; (4)
the adult pests must find host material; and (5) the climatological and
microenvironmental conditions must be suitable. These factors
substantially reduce the likelihood of establishment. The degree of
pest risk reduction attributable to each of the factors has not been
quantified. People generally consume the fruit they purchase and
dispose of the waste material in a manner (such as in plastic bags that
are landfilled or incinerated) that precludes the release of pests into
the environment.
In the preceding bullet points, the reader may note that the
estimated numbers of potentially infested fruit are in some cases
different than the similar bullet points presented in the proposed
rule. These differences are attributable to adjustments made in the
updated risk assessment to the 95th percentile estimates for ``N''
(number of Hass avocados imported from Mexico per year) and ``P1''
(proportion of avocados infested). P1 was revised upwards because the
detection sensitivity range 17.9 to 83.5 was used. P1 is the same for
the 47 and 50 State scenarios. N was revised downward based on the
revised economic analysis.
Based on comments that we received on the proposed rule, and taking
into account the findings of the updated risk assessment, this final
rule includes several provisions that differ from the proposed rule.
Specifically:
We proposed to allow the avocados to be distributed in all
50 States, but solicited comments on the possibility of delaying the
distribution of the avocados in California, Florida, and Hawaii for 1
year. In this final rule, we have adopted a delay in the distribution
of the avocados in California, Florida, and Hawaii for a period of 2
years based on the comments that we received. After that 2-year period,
the avocados may be distributed in all 50 States. The effective dates
for importing fruit into all 50 States are built into the final rule,
which precludes the need for APHIS to initiate further rulemaking in
order to expand the area into which the fruit may be
[[Page 69750]]
imported. If it is determined that the requirements of the export
program are not being observed routinely or uniformly, APHIS will be
able to act quickly to suspend the effective dates or even the entire
program, if warranted. The export program provides for the detection of
infested fruit at any point in the pathway, with that detection leading
to the rejection of the shipment containing the infested fruit and the
removal of the grove or municipality that produced the fruit from the
export program until it is determined by APHIS and the Mexican NPPO
that the grove or municipality may be readmitted to the program. Thus,
the detection of infested fruit will not, by itself, result in the
suspension of all or part of the export program. To determine whether
the requirements of the export program are being observed routinely or
uniformly and to ensure that the distribution restrictions of this rule
are being observed, APHIS personnel will be involved in monitoring
activities in both the United States and Mexico.
To reflect our proposal to allow the avocados to be
distributed in all 50 States, we proposed to remove the requirement in
Sec. 319.56-2ff(c)(3)(vii) that all boxes or crates of avocados be
clearly marked with, among other things, the statement ``Not for
distribution in AL, AK, AZ, AR, CA, FL, GA, HI, LA, MS, NV, NM, NC, OK,
OR, SC, TN, TX, WA, Puerto Rico, and all other U.S. Territories.'' In
this final rule, we have retained that marking requirement, specific to
California, Florida, and Hawaii, for the term of the 2-year delay in
distribution to those States.
To reflect our proposal to allow the avocados to be
distributed in all 50 States, we proposed to remove the provisions in
Sec. 319.56-2ff(f), which limited the ports of entry through which the
avocados may be imported, and Sec. 319.56-2ff(g), which described the
areas of the United States that avocados moving by truck or rail car
may transit while en route to approved States. In this final rule, we
continue to prohibit the movement of the avocados into or through
California, Florida, and Hawaii for the term of the 2-year delay in
distribution to those States.
To reflect our proposal to allow the avocados to be
distributed in all 50 States, we proposed to remove the provisions in
Sec. 319.56-2ff(j) that required any boxes used to repackage the
avocados in the United States to bear the same information that is
required to be displayed on the original boxes in which the fruit was
packed in Mexico. In this final rule, we have retained those
repackaging requirements due to the 2-year delay in distribution to
California, Florida, and Hawaii.
We proposed to add a requirement for the avocados to be
packed in insect-proof cartons, loaded in insect-proof containers, or
covered with insect-proof mesh or plastic tarpaulin prior to leaving
the packinghouse. This proposed requirement was intended to replace the
requirement in Sec. 319.56-2ff (c)(3)(viii) that, prior to leaving the
packinghouse, the truck or container transporting the avocados must be
secured by Sanidad Vegetal with a seal that will be broken when the
truck or container is opened. In this final rule, we retain the
requirement for seals and will not require the insect-proofing measures
we had proposed.
Although our adoption of a 2-year delay in distribution to
California, Florida, and Hawaii has led us to retain, at least in part,
the box marking, port of entry, and repackaging provisions discussed
above that we had proposed to remove, we have decided to follow through
with the removal of another measure related to limited distribution,
i.e., compliance agreements. The compliance agreement provisions that
were located in Sec. 319.56-2ff(k) were intended to ensure that
distributors and handlers of the avocados were familiar with the
distribution restrictions and other requirements of the regulations.
Given that the distribution restrictions established in this final rule
cover only three States, and only for a limited time, we believe that
the time, costs, and logistical difficulties involved in initiating
compliance agreements with all distributers and handlers of imported
Mexican Hass avocados in 47 States would outweigh the benefits that may
be gained by retaining the compliance agreement requirement. The fruit
stickering, box marking, and repackaging requirements of the
regulations will serve to ensure that the avocado's origin can be
determined, and the latter two requirements will ensure that the
limited distribution statement is present on all boxes of fruit. During
the 2-year delay, we will focus our efforts on education and outreach
so that distributors and handlers will be made aware of product origin
indicators and penalties for violation of regulations. We will be able
to concentrate our enforcement efforts more readily since the avocados
will be prohibited in only three States during the 2-year delay, and
will continue to take action and seek penalties for violations of the
regulations under the Plant Protection Act.
Determination by the Secretary
Under section 412(a) of the Plant Protection Act, the Secretary of
Agriculture may prohibit or restrict the importation and entry of any
plant or plant product if the Secretary determines that the prohibition
or restriction is necessary to prevent the introduction into the United
States or the dissemination within the United States of a plant pest or
noxious weed.
The Secretary has determined that it is not necessary to prohibit
the importation of Hass avocados from Mexico subject to the
phytosanitary requirements described in this final rule in order to
prevent the introduction into the United States or the dissemination
within the United States of a plant pest or noxious weed. This
determination is based on the findings of the risk assessment referred
to earlier in this document, and the Secretary's judgment that the
application of the measures required under Sec. 319.56-2ff would
prevent the introduction or dissemination of plant pests into the
United States.
Based on the Secretary's determination, and in response to the
Mexican Government's request, we are amending the regulations to expand
the number of States (plus the District of Columbia) in which fresh
Hass avocado fruit grown in approved orchards in approved
municipalities in Michoacan, Mexico, may be distributed and to allow
the distribution of the Hass avocados during all months of the year.
Comments
Comment: The U.S. Department of Agriculture's (USDA's) Mexican
avocado import program has never before operated during warm summer
months when pest population levels in Mexico are at their peak and
susceptible crops are in full production in the United States. The
proposed rule would allow avocados from Mexico to be imported during
all months of the year, across all States, including those currently
excluded under the existing rule. By increasing the timeframe to all
months, the proposed rule includes the time period when pest insects
are most active and sexually mature, in highly favorable environments
for infestation on avocados and other host plants.
Response: The goal of the program continues to be the exclusion of
any quarantine pests that could become established in the United
States. Under the modified systems approach semiannual surveys, rather
than annual, will be conducted at the municipality and orchard level.
Municipalities must be free of Ceratitis capitata,
[[Page 69751]]
Conotrachelus aguacatae, C. perseae, Heilipus lauri, and Stenoma
catenifer before they can be certified to export avocados to the United
States. In addition, orchards must be certified free of Copturus
aguacatae. Trapping is conducted in orchards for Anastrepha spp. fruit
flies. Both the regulations and the workplan specify what mitigation
measures must be taken when a pathway pest is detected in a certified
orchard or municipality. The time periods selected for the surveys were
based on the biology of the pests. Additionally, fruit cutting will be
conducted in the orchard, packinghouses, and at the port of entry.
Since the expansion of Mexican avocado imports in 1997, none of these
pests have been intercepted during inspections of fruit at
packinghouses or upon inspection at the U.S. border ports. Further, the
limited distribution plan that we will implement would delay the
importation of Hass avocados from the Mexico State of Michoacan into
the commercial avocado producing States of California, Florida, and
Hawaii for 2 years from the effective date of this final rule. This
restriction will provide APHIS an opportunity to further substantiate
the effectiveness of the mitigation measures under the expanded
program.
Comment: USDA has ignored the fact that stem weevils continue to be
found in alarming numbers in the Mexican production areas and that
these insects can easily migrate from backyard orchards to commercial
groves.
Response: Registered orchards and all contiguous orchards and
property are surveyed for the stem weevil, Copturus aguacatae and must
be found free of the pest. If Copturus aguacatae is detected in an
orchard requesting certification, eradication must be completed prior
to the orchard receiving certification to export avocados to the United
States. If the stem weevil is found in an orchard or property
contiguous to a certified orchard, eradication measures will be ordered
for that orchard or property. Surveys for the stem weevil will commence
in the registered orchard contiguous to the area where the detection
was found on a weekly basis until eradication of the stem weevil in the
contiguous orchard has been completed. Since 1997, the stem weevil has
been detected in 7 orchards that applied for certification.
Certification was denied due to the presence of the stem weevil. Since
the expansion of Mexican avocado imports in 1997, no stem weevils have
been intercepted during inspections of fruit at packinghouses or upon
inspection at the U.S. border ports.
Comment: USDA has mistakenly decided that other thrips-related
insects pose no threat because they are not associated with avocado
fruit; this is demonstrably false and inconsistent with research done
by prominent entomologists. Every year a significant percentage of the
California avocado crop is either downgraded to Grade 2 fruit, for
which growers receive a lot less money, or is culled or thrown out due
to thrips damage. Thrips-damaged fruit is unattractive and cannot be
sold in grocery stores. Why would the USDA allow Mexican fruit known to
be infested with thrips to be imported into avocado producing areas?
Response: The risk analysis does not list any thrips as pests that
would follow the pathway. There are 16 thrips listed in the Appendix A
pest list. Of the 16 which occur in Mexico, 5 occur in the United
States. All 16 pests are associated with a plant part other than
avocado fruit, or in rotting fruit on the ground. For example, research
(e.g., Hoddle, 2002; Yee et al., 2003, cited in the risk assessment)
has demonstrated that Scirtothrips perseae, lays eggs in small,
immature fruits and tender leaves, and does not feed on or lay eggs in
mature fruit, and is, therefore, unlikely to be imported with the
fruit. APHIS considers Scirtothrips perseae as probably representative
of other pest thrips species. Mitigation of these pests in rotting
fruit is addressed in the workplan and regulation. Avocado fruit that
has fallen from the trees must be removed from the orchard within 7
days and may not be included in field boxes of fruit to be packed for
export. In addition, damaged fruit must be culled at the packinghouse.
Although Frankliniella bruneri is listed as having been intercepted in
avocados at the U.S. border, the interception was made in fruit found
in baggage, not in a commercial shipment imported under the program.
Since the expansion of Mexican avocado imports in 1997, no thrips have
been intercepted in program fruit during inspections at packinghouses
or upon inspection at the U.S. border ports.
Comment: The USDA import program has grown exponentially since
1997, from approximately 3,700 to over 53,000 certified acres,
stretching USDA resources and increasing the probability of human
error. Believing a screening system can be set up to catch all infested
fruit at the border is not realistic. Control at the source with
identification and traceability through the entire chain is required.
Response: APHIS' role under the operational workplan provided by
the regulations is to provide management and monitoring of the
activities specified in the workplan, e.g., trapping, surveying, and
packinghouse inspections. While APHIS personnel do not necessarily
conduct these activities themselves, they do monitor Mexican officials'
compliance with workplan specifications. The staffing level of APHIS
personnel is sufficient to ensure that APHIS meets its requirements
under the workplan and that other signatories are in compliance with
the regulations. The lack of pest detections in the orchard,
packinghouse, and border inspection since the program began in 1997 is
evidence that the regulations and workplan are being complied with.
In addition, as more orchards have applied for certification, it
does take longer for inspectors to perform the initial inspection
before the first shipping date of October 15. The inspectors have had
to start inspections earlier before that date each year to finish
inspecting all of the orchards. Nevertheless, all orchards must be
inspected using the same workplan criteria, as the records show. APHIS
keeps lists of all the orchards inspected by name.
Comment: USDA should test the proposed expansion regulations under
actual production and distribution conditions over a 2-year period at a
minimum before allowing shipments into California, Florida, and Hawaii.
Response: As noted previously, APHIS has decided, based on the
comments we received on the issue, to implement a limited distribution
plan that would delay the importation of Hass avocados from the Mexico
State of Michoacan into the commercial avocado producing States of
California, Florida, and Hawaii for 2 years from the effective date of
this final rule. This restriction will provide APHIS an opportunity to
further substantiate the effectiveness of the mitigation measures under
the expanded program.
As stated in the economic analysis, the volume of Mexican Hass
avocado exported to the United States is expected to substantially
increase. In addition, some of the commenters stated that they believed
the expansion of this program could not be effectively managed. The
data collected during the first 2 years of the expanded program will
provide confirmation of the effectiveness of the mitigation measures
and management of the Mexican Hass avocado export program. Some of
those data will cover production and distribution periods not
previously covered by the current regulation.
Comment: USDA has never allowed untreated fruit to be imported from
a
[[Page 69752]]
region where quarantine pests are present into a region of the United
States where the same crop is produced.
Response: We disagree with the commenter's characterization of the
proposed action and statement that USDA has never allowed untreated
fruit to be imported from a region where quarantine pests are present
into a region of the United States where the same crop is produced. In
the case of tomatoes from France, for example, we allow the entry of
tomatoes from France under certain conditions. The tomatoes must be
produced under a systems approach to mitigate the pest risk of
Ceratitis capitata, which is a quarantine pest for the United States.
The systems approach includes trapping for Mediterranean fruit fly, the
tomatoes must be greenhouse grown, and the tomatoes must be safeguarded
from harvest to arrival in the United States. In addition, a
phytosanitary certificate is required. While phytosanitary treatment is
not an option to mitigate the fruit fly risk, the mitigation measures
applied to this commodity are equivalent to a pesticide treatment.
Similarly, under the Mexican avocado import program, Hass avocados are
subject to a systems approach to mitigating pest risk that produces
results similar to those achieved through treatment measures.
Comment: California growers have been unable to gain access to the
avocado market in Mexico since 1998. Rather than assisting U.S. growers
in dismantling false trade barriers erected by Mexico, USDA has
diligently worked with Mexican officials to open up the U.S. avocado
market to Mexican avocados.
Response: APHIS officials are aware that U.S. avocado producers
would like further access to Mexican markets. Initially, APHIS
officials requested market access for U.S. avocados only to Mexicali
and Tijuana and successfully opened these markets. Thereafter, APHIS
requested market access to the five northern Mexican States along the
border with the United States and more recently informed Mexico that
our avocado producers are interested in access to all of Mexico. In
response to APHIS' request, Mexican plant health officials began a pest
risk assessment for the importation of U.S. avocados to additional
Mexican markets. This process is similar to the pest risk assessment
process APHIS conducted prior to publishing the proposed rule to allow
Mexican avocados further access to our markets. USDA has continually
pursued the issue of expanded market access for U.S. avocados with
Mexico, however the next stage of the process cannot take place until
Mexico completes its risk assessment.
Comment: Mexican avocado growers have the ability to use pesticides
on their crops that American growers are prohibited from using.
American growers are at a disadvantage if they were ever faced with a
pest infestation, as USDA does not have a method of dealing with an
infestation that has occurred other than imposing an economically
devastating quarantine. Additionally, there are concerns about human
health issues that might arise as a result of pesticide residues.
Response: Considering the conclusions of the risk assessment and
given the fact that there have been no interceptions of pests in
commercial shipments of Hass avocados from Mexico, we do not believe
that a pest infestation will occur under the expanded program. USDA and
its State counterparts, however, have a variety of options for dealing
with pest outbreaks that may occur in this country. The response in any
particular instance depends largely on the specific circumstances of
the outbreak.
In addition, while the United States does not have direct control
over pesticides that are used on food commodities such as avocados in
other countries, there are regulations in the United States concerning
the importation of food to ensure that commodities do not enter the
United States containing illegal pesticide residues. Because DDT is a
pesticide that is banned in the United States, even if it were used on
food commodities in foreign countries, the current regulations
concerning the importation of food into the United States prevent the
entry of products treated with it.
Through section 408 of the Federal Food, Drug, and Cosmetic Act,
the Environmental Protection Agency (EPA) has the authority to
establish, change, or cancel tolerances for food commodities. These
tolerances are the maximum levels of pesticide residues that have been
determined, through comprehensive safety evaluations, to be safe for
human consumption. Tolerances apply to both food commodities that are
grown in the United States and food commodities that are grown and
imported into the United States from foreign countries. While EPA has
no authority in a foreign country, the tolerance levels are enforced
once the commodity enters the United States. Chemicals such as DDT that
are banned in the United States do not have tolerances on food
commodities.
Federal government food inspectors are responsible for monitoring
food commodities that enter the United States to confirm that tolerance
levels are not exceeded and that residues of pesticide chemicals that
are banned in the United States are not present on the commodities.
Tolerance levels for all chemicals that are acceptable for use on
avocados may be found in EPA's regulations in 40 CFR 180.101
through180.2020. Tolerance information can also be obtained at http://www.epa.gov/pesticides/food/viewtols.htm
.
Comment: If USDA expands the Mexican Hass avocado import program as
proposed, it should ensure that compensation is available for U.S.
growers in avocado producing areas should a pest infestation occur.
Response: The Plant Protection Act provides that the Secretary may
pay compensation to any person for economic losses incurred by the
person as a result of action taken by the Secretary under the
extraordinary emergency authority provided in section 415 of the Act (7
U.S.C. 7715). The determination of an extraordinary emergency would
depend on the circumstances of an infestation on a case-by-case basis
and APHIS cannot regulate on this issue at this time since infestation
has not occurred. Any decision as to the need to declare an
extraordinary emergency and, if declared, to pay compensation, rests
with the Secretary.
Comment: USDA should set up an insurance or indemnification program
to compensate domestic avocado growers for any damage incurred as a
result of any pest infestation that may occur as a result of the
proposed expansion.
Response: APHIS does not have the authority to establish such a
program under the Plant Protection Act.
Comment: Growers in Mexico should have to pay for the quarantine
insurance for the avocado growers in the United States. While it could
be argued that all of the previous infestations were the result of
illegal importation, it is inevitable that legal importation will
eventually create a domestic quarantine. Requiring the Mexican
producers to pay for the quarantine insurance would level the economic
costs. Paying for insurance would also encourage Mexico to rid other
areas of applicable pests.
Response: APHIS has no authority to require foreign producers to
pay quarantine insurance for domestic producers.
Comment: Mexican avocados should not be imported into the United
States because of the prevalence of stem weevils, seed weevils, fruit
flies, thrips, and persea mites in Mexico.
[[Page 69753]]
Response: APHIS has assessed the risk and determined that there is
low likelihood of introducing quarantine pests such as stem weevils,
seed weevils, seed moth, and fruit flies. Nine mites are identified in
the pest list in Appendix A of the risk assessment. Of the nine mites,
eight are also present in the United States so those species are not
considered to be quarantine pests. Only one mite, Brevipalpus
australis, occurs in Mexico but not in the United States. All the
identified mites are associated with a plant part other than the
avocado fruit, or in rotting fruit on the ground and, therefore, are
not likely to follow the pathway. Since the expansion of Mexican
avocado imports in 1997, mites have not been intercepted during
inspections at packinghouses or upon inspection at the U.S. border
ports.
The risk analysis does not list any thrips as pests that would
follow the pathway. There are 16 thrips listed in the Appendix A pest
list. Of the 16 that occur in Mexico, 5 also occur in the United
States. All 16 pests are associated with a plant part other than
avocado fruit, or in rotting fruit on the ground. For example, research
(e.g., Hoddle, 2002; Yee et al., 2003, cited in the risk assessment)
has demonstrated that Scirtothrips perseae lays eggs in small, immature
fruits and tender leaves, and does not feed on or lay eggs in mature
fruit, and is, therefore, unlikely to be imported with the fruit. APHIS
considers Scirtothrips perseae as probably representative of other pest
thrips species. Mitigation of these pests in rotting fruit is addressed
in the workplan and the regulations. Avocado fruit that has fallen from
the trees must be removed from the orchard within 7 days and may not be
included in field boxes of fruit to be packed for export. In addition,
damaged fruit must be culled at the packinghouse. Although
Frankliniella bruneri is listed as having been intercepted in avocados
at the U.S. border, the interception was made in fruit found in
baggage, not in a commercial shipment imported under the program. Since
the expansion of Mexican avocado imports in 1997, no thrips have been
intercepted in program fruit during inspections at packinghouses or
upon inspection at the U.S. border ports.
Comment: I am opposed to the expansion of the Mexican avocado
program into avocado-producing States due to a pest infestation that
resulted in a quarantine in San Diego County and other counties in
California in 2002 and 2003 (the Valley Center infestation). The
quarantine caused financial losses for which we were not compensated.
Response: The Valley Center infestation in California stemmed from
unknown origins and not from commercially imported Hass avocados from
Mexico which were prohibited from entering California at the time.
Comment: The avocado expansion will jeopardize not only domestic
avocado production, but all U.S. agricultural products susceptible to
pests found in Mexico.
Response: With the exception of fruit flies, the pathway pests
identified in the risk assessment are avocado specific, thus we expect
that the commenter is referring to fruit flies in speaking of ``pests
found in Mexico'' that would affect other U.S. agricultural products.
The risk assessment identified three fruit flies capable of following
the pathway. Ceratitis capitata can infest avocado (Liquido et
al.,1998) and is a quarantine pathway pest. The species is under
official control in Mexico and is found only on the Mexico-Guatemala
border (APHIS, 1999). Municipalities participating in the program must
be certified free from Ceratitis capitata.
Hass avocados are considered poor hosts for the other two fruit
flies, Anastrepha ludens and Anastrepha striata, thus those pests are
unlikely to follow the pathway. Anastrepha ludens, Anastrepha striata,
and Ceratitis capitata have not been intercepted in any of the more
than 10 million avocados cut during the course of the program.
The risk assessment further identifies the conditions that would be
necessary for fruit flies to be introduced in the United States. Even
if an infested avocado were to arrive at a region with host material,
several additional conditions are required for pest establishment: (1)
The pest must survive in the avocado during transportation and storage;
(2) the infested avocado must be discarded in close proximity to host
material; (3) the pest must find a mate; (4) the pest must successfully
avoid predation and other threats; (5) the adult pest must find
appropriate host material; (6) suitable climatological and
microenvironmental conditions must exist; and (7) they must escape
detection and subsequent eradication measures.
Comment: Mexican agricultural field workers are not reliable enough
to consistently follow procedures necessary for pest-free exports from
Mexico. In contrast, American avocado growers go through an abundance
of measures to monitor for disease and to contain outbreaks from
spreading to our neighbor growers.
Response: The commenter provided no information to support his
contention regarding the reliability of Mexican field workers. Avocados
are a commercially important crop in Mexico, and Mexico's continued
ability to export avocados to markets such as the United States and
Japan is dependent on its ability to meet the phytosanitary standards
of those importing nations. It has been our experience that avocado
producers in Mexico are fully capable of meeting the requirement of the
regulations and anticipate that they will continue to do so.
Comment: Eighty-five different Thysanoptera species of thrips have
been found in Mexican avocados and 24 different mite pest species,
pertaining to eight distinct families, have been found in Mexican
avocado plantings. More research should be done into these pests and
the damage they could incur should they reach an American avocado farm.
Response: In the latest risk assessment update, the list of
quarantine pest thrips has been updated. Mites and thrips are not
likely to be in the commercial import pathway because they are not
generally associated with mature fruit or remain on mature, harvested
fruit. None have been intercepted by APHIS with program Hass avocados
from Mexico. The risk assessment does not list any thrips as pests that
would follow the pathway. There are 16 thrips listed in the risk
assessment's Appendix A pest list as occurring in Mexico; of those, 5
also occur in the United States. All 16 pests are associated with a
plant part other than avocado fruit, or in rotting fruit on the ground.
For example, research (e.g., Hoddle, 2002; Yee et al., 2003, cited in
the risk assessment) has demonstrated that Scirtothrips perseae lays
eggs in small, immature fruits and tender leaves, and does not feed on
or lay eggs in mature fruit, and is, therefore, unlikely to be imported
with the fruit. APHIS considers Scirtothrips perseae as probably
representative of other pest thrips species. Mitigation of these pests
in rotting fruit is addressed in the workplan and the regulations.
Avocado fruit that has fallen from the trees must be removed from the
orchard within 7 days and may not be included in field boxes of fruit
to be packed for export. In addition, damaged fruit must be culled at
the packinghouse. Although Frankliniella bruneri is listed as having
been intercepted in avocados at the U.S. border, the interception was
made in fruit found in baggage, not a commercial shipment imported
under the program. Since the expansion of Mexican avocado imports in
1997, no thrips have been intercepted during inspections at
packinghouses or upon inspection at the U.S. border ports.
[[Page 69754]]
Nine mites are identified in the pest list in Appendix A of the
risk assessment. Of the nine mites, eight are also present in the
United States. Only one mite, Brevipalpus australis, occurs only in
Mexico. All the identified mites are associated with a plant part other
than avocado fruit, or in rotting fruit on the ground and, therefore,
not likely to follow the pathway. Since the expansion of Mexican
avocado imports in 1997, mites have not been intercepted in program
fruit during inspections at packinghouses or upon inspection at the
U.S. border ports.
Comment: The study that served as underlying research for the APHIS
risk assessment did not test Hass susceptibility to Mexican fruit fly
at all maturity levels during all-year weather conditions such as
temperature and humidity. Mexican fruit fly does reproduce in Hass
avocado, certainly so in harvested fruit. Fruit still on the tree but
ready to drop is a very probable host. A study should be performed by
APHIS experts, or by Agricultural Research Service (ARS) experts on
subtropical fruit pests, and Hass avocado host susceptibility should be
studied at all potential stages of the Hass avocado during its
marketing season, i.e., from just barely mature to very mature on-tree
Hass fruit, as well as fresh, naturally dropped from tree Hass fruit,
since the fruit could be harvested just before they drop.
Response: The host studies conducted by Aluja et al. for Anastrepha
species and Hass avocados in Mexico \2\ did test Hass susceptibility to
Mexican fruit fly at all maturity levels during all-year weather
conditions such as the temperature and humidity that occurs during the
summer months of June, August, and September. The study considered
fruits of a range of sizes that were commercially mature, and mature
fruit attached to the tree as well as off the tree. The study concluded
that this fruit was not considered to be a host for Mexican fruit fly
under any of these conditions.
---------------------------------------------------------------------------
\2\ Aluja, M., Diaz-Fleisher, F., and J. Arredondo. 2004. Non-
host status of Persea americana ``Hass'' to Anastrepha ludens,
Anastrepha obliqua, Anastrepha serpentina, and Anastrepha striata
(Diptera: Tephritidae) in Mexico. Journal of Economic Entomology,
volume 97, issue 2, April 2004.
---------------------------------------------------------------------------
The Aluja et al. study was subjected to rigorous peer review prior
to its publication and was likewise reviewed by USDA fruit fly experts
in ARS and APHIS. The input from ARS follows the tradition and
guidelines of peer review. The ARS experts offered their own
interpretation of the scope and applicability of the findings. This
information suggested that we should update our risk assessment, and we
considered the ARS input in preparing our updates and changing our
classification of the host status of Hass avocados. We fully intend to
continue to seek and incorporate ARS expertise and guidance in our risk
analysis products.
Comment: Paragraph (e)(1) of proposed Sec. 319.56-2ff provides
that if Heilipus lauri, Conotrachelus aguacatae, C. perseae, or Stenoma
catenifer are detected during the semiannual pest surveys, orchard
surveys, packinghouse inspections, or other monitoring or inspection
activity in the municipality, the municipality where the pest is found
will be suspended until APHIS and the Mexican NPPO agree that the
eradication measures taken have been effective and the risk of the pest
in the municipality has been eliminated. In order to harmonize
phytosanitary measures between Mexico and the United States, and given
that preclearance programs for exports from the United States to Mexico
do not in any case suspend the export program for a whole county when
there is a single detection of a quarantine pest, we request that the
suspension provided for in paragraph (e)(1) be applied to only the
grove involved, and not the entire municipality.
Response: Under the regulations, as well as in the proposed rule
and its final rule, area freedom for Heilipus lauri, Conotrachelus
aguacatae, C. perseae, and Stenoma catenifer is defined at the
municipality level. Mexico has requested that we adjust this to the
orchard level. Such an adjustment would require a change to the
regulations, and we believe that the public should have the opportunity
to comment on that change and its underlying basis. Therefore, APHIS
will take this suggestion under advisement and review whether a pest
risk analysis must be conducted to address the requested change, if the
change would provide an equivalent measure of phytosanitary security,
and whether or not to initiate the rulemaking that would be required to
make the requested change.
Comment: Different sensitivities in inspection have not been taken
into account in the risk assessment.
Response: The commenter is suggesting that the sensitivity of fruit
cutting may be different for the eight pathway pests. The estimate for
the sensitivity of fruit cutting used in the May 2004 risk assessment
is 50 percent (i.e., an infested fruit would be identified 50 percent
of the time). Our use of a point value (50 percent) in the quantitative
model did not include uncertainty about the estimate. APHIS had used
the average sensitivity of starfruit and grapefruit ([35 percent + 80
percent] / 2 = 57.5 percent) and rounded down to 50 percent. For
simplicity we used a point value, confident that this number is a
reasonable minimum and that the actual value is probably higher.
For our updated (August 2004) risk analysis, data were reanalyzed
to determine the effect of variation in the sensitivity parameter on
the model output. We replaced the 50 percent point estimate with a
uniform distribution from 17.9 to 83.5 percent. When we used the entire
range given in Gould (1995, table 3, as cited in the risk assessment)
of 17.9 percent to 83.5 percent, there was very little change in the
results. Gould (1995) reported that the sensitivity of detection by
experienced inspectors of six types of fruit (not including avocado)
infested with third instar Caribbean fruit fly (Anastrepha suspensa)
larvae ranged from 17.9 percent for green guavas to 83.5 percent for
carambolas. In order to account for uncertainty, a uniform distribution
was used in the analysis presented in Appendix D of the updated risk
assessment.
The sensitivity of detection could vary somewhat among pathway
pests. All of them can damage the fruit pulp when present in the fruit;
however, the stem weevil (Copturus aguacatae) produces tunnels that are
usually restricted to a small portion of the fruit close to the
peduncle. Stem weevil larvae rarely migrate into the fruit, but when
they do, they are usually localized to the area of the fruit near the
peduncle (APHIS, 1997; Gudino Juarez and Garcia Guzman, 1990, cited in
the risk assessment). Inspectors are specifically instructed and
trained to examine the peduncle end of the fruit for stem weevil larvae
(APHIS, 1997, cited in the risk assessment). Because of this training
and because the location of stem weevil larvae is highly predictable
and usually quite localized, APHIS has determined that the sensitivity
of detection for stem weevils and other internal avocado pests could
reasonably be considered to be close to the 50 percent point in the
17.9 to 83.5 percent range.
Comment: In the proposed rule, USDA points to the fact that no
pests of concern have been found in commercial shipments of Mexican
avocados since the program began. This fact may be true, but the
Department's reliance on it is misplaced; the dynamic nature of the
program ensures that the systems approach will not operate in the same
[[Page 69755]]
fashion from one year to the next, and possibility of human error
increases as the program grows in size. The importance of analyzing
human reliability factors in the estimation of risk is undisputed.
Nonetheless, USDA's consideration of human error in the operation of
its systems approach for Mexican avocados has, to date, been
inconsequential. The omission of the characterization of failure modes
and human reliability in the Department's risk assessment is a fatal
flaw that U.S. avocado growers cannot accept.
Response: While there will be additional acres under certification
in Michoacan as well as additional avocado imports, the systems
approach can be adapted to deal with these increasing requirements.
Additional staff or additional layers of mitigations may be added to
deal with issues that arise. APHIS meets yearly to negotiate a work
plan with Mexican plant health authorities and address issues that
arose in the previous year's operation. We are confident that adequate
resources will be available to ensure that the systems approach will
continue to be effective.
While past experience is not a perfect guide to future performance,
there is no reason to believe that we will not be able to rely on the
effectiveness of the systems approach under conditions that may exist
in the future. Additionally, there is also no reason to believe that
the systems approach will remain the same while demand for program
resources increases. The systems approach is a dynamic process that is,
and will continue to be, modified to address changes and future needs.
Comment: Given the very high long-term costs to California avocado
producers (and the State of California) of a pest introduction from
Mexican imports, why are all of the pest risk assessments on page 4 of
the analysis presented at a 95 rather than a 99 percent level of
confidence? I am sure that producers want a high level of confidence
that risks are very low or nonexistent.
Response: The risk assessment's Appendix D was modified to include
graphic representations of all percentile results for all of the model
outputs. Both the 95th percentile and the 99th percentile results are
included in the table of results in the body of the document and in
Appendix D. The 95th percentile results are as relevant as the 99th
percentile.
Comment: I have a problem reconciling the first and last pest risk
conclusions on page 4 of the proposed rule. The statement that ``fewer
than 387 infested avocados will enter the United States each year,
estimated with 95 confidence'' must be based on a different
distribution than the statement that ``it is slightly more likely that
zero infested avocados will enter the United States than one infested
avocado,'' or the distribution has to have a large standard error on
the high side.
Response: The number of infested avocados entering the United
States is not a point value but is represented by a probability
distribution. A probability distribution presents the range of values a
parameter can assume (x-axis), plotted against the relative likelihood
of assuming those values (y-axis). The probability distribution for the
number of infested avocados entering the United States is presented in
Appendix D, page 104. The figure indicates that the value with the
highest relative likelihood (the most likely value) is zero, and the
mean of the distribution is 122.6.
A cumulative probability distribution presents the range of values
a parameter can assume (x-axis), plotted against the likelihood of
assuming those values or less (y-axis). The cumulative probability
distribution for the number of infested avocados entering the United
States is presented in Appendix D, page 104. The 95th percentile value
for the number of infested avocados entering the United States is 387.
The 95th percentile of a parameter is the value in the data set for
which 95 percent of the values are below it and 5 percent are above.
The distribution for the number of infested avocados entering the
United States is skewed to the left, and has a tail to the right.
The most likely value or mode is the value that occurs most often
in a set of values. In a histogram and a result distribution, it is the
center value in the class or bar with the highest probability. In this
case, the most likely value is zero.
Comment: The statement in the APHIS risk assessment comparing the
probability of entry of zero versus one infested avocado is not at all
useful. A more relevant comparison of probabilities is between zero and
one or more infested avocados entering the United States. In addition,
a description of the statistical distributions (i.e., mean and
variance) that these statements are based on would help the reader to
better understand the nature of the risks.
Response: Appendix D has been modified to include mean and standard
deviations for all model output results. The mode of the distribution
is zero; therefore, zero infested avocados entering the United States
is more probable than one.
Comment: The consideration on behalf of the USDA to import foreign
fruit motivates foreign growers to purposely smuggle and introduce
insects into U.S. growing areas so as to overcome the opposition to
import, such as was suggested in the Valley Center Mexican fruit fly
infestation which cost growers millions.
Response: While the origin of the Valley Center outbreak has not
been determined, we have no information to suggest it was the result of
an intentional introduction of pests as the commenter contends.
Comment: The public must be informed about where the transportation
of avocados is prohibited and where it is not. The public must also be
informed that they are prohibited from personally transporting avocados
into the United States, even if commercial (inspected) avocados are
permitted. Public education is even more critical within the avocado
growing and producing States of Florida, California, and Hawaii. How
does the USDA propose to educate the public about this proposal? If
there is a trial period, how does the USDA plan to inform the public in
Florida, California, and Hawaii that commercially produced Mexican
avocados are still prohibited into those States? The public will not
see a difference between these two scenarios and it is not discussed
within the proposal. I fully expect to see serious increases in
inadvertent movement of fruit from Mexico.
Response: The general public will be notified of the change in the
Mexican avocado program and its specific restrictions through this
rulemaking process and through Agency outreach and the media by way of
press releases, fact sheets, publications, and other materials that
help explain APHIS programs. The Agency's outreach efforts are
coordinated with those of the States in order to extend their coverage.
Federal inspection officers at ports of entry will continue to inspect
members of the public returning to the United States and will seize any
agricultural items, including avocados, that are prohibited from
entering the country.
Comment: In light of the ARS conclusion that Hass avocados are a
very poor host for Anastrepha fruit flies, it would seem logical for
APHIS to at least lower some of the very costly elements of the systems
approach in Michoacan that are targeted at Anastrepha spp. fruit flies.
Yet, instead of decreasing the requirements aimed at fruit flies in
Michoacan, APHIS has left the requirements for fruit fly trapping
completely intact. This means that when Anastrepha spp. flies are
found, a list of unnecessary regulatory actions must take place,
including the needless application of pesticides.
[[Page 69756]]
Considering the consensus that the Hass avocado is a very poor
pathway for Anastrepha spp. fruit flies, it would seem logical to
replace the current restrictions requiring the need for a full fruit
fly trapping program with a fruit fly monitoring program. Additionally,
in light of the proposal to remove the specific details of the seed and
stem pest surveys from the regulation and insert them into the
phytosanitary work plan, it would also make sense to remove the
specific language referring to this fruit fly monitoring activity from
the regulation and, for the sake of consistency, insert it into the
phytosanitary work plan as well. This way, as the discussion on the
host status of the Hass avocado continues to evolve, there will be no
need to go through the rulemaking process to make adjustments to this
section of the regulation in the future.
Response: Removing the details for fruit fly trapping was not
considered at the time the proposed rule was published and, therefore,
we will not remove those details in the final rule. We will, however,
consider this issue for future rulemaking.
Comment: USDA should ensure that the surveys and detection
trappings in Mexico occur during all 12 months of the year to ensure
that monitoring for all potential pests is sufficient for all the
listed pests and occurs during all the potential detection periods
(instead of the proposed semiannual surveys).
Response: The semiannual municipality and orchard surveys are
required for initial certification and to maintain certification later
on. There is year-round trapping for fruit flies, which is performed in
support of a separate APHIS program, and packinghouse and border
inspections will occur on a continual basis. Other pests of concern are
surveyed at specific times of the year based on the biology of those
species. The regulations provide for the suspension of an orchard or
municipality from the program at any time as a result of the detection
of specified pests during the semiannual pest surveys, orchard surveys,
packinghouse inspections, or other monitoring or inspection activity.
Comment: We disagree with the APHIS proposal to replace the
requirement to seal each consignment moving from the packing shed to
the border with a new requirement for the avocados to be packed in
insect-proof boxes, loaded in insect-proof containers, or covered with
insect-proof mesh or plastic tarpaulin prior to leaving the packing
shed so that in the very infrequent occurrence of a shipment being
stopped for inspection by Mexican authorities, fruit flies or
hitchhiking pests will not enter. It is not logical to add those
requirements for the following reasons:
Refrigerated containers do not attract fruit flies or
other subtropical pests. In fact the opposite is true. Refrigerated
containers present a very inhospitable environment for tropical and
subtropical arthropods. Additionally, such conveyances do not provide
an adequate environment for insect activity such as oviposition.
If the Hass avocado is not a host to the fruit flies that
occur only in rather small numbers in Michoacan, then fruit fly-
proofing the shipment is completely unnecessary.
Actual fruit fly host materials such as mangos, citrus,
tomatoes, and peppers, which are routinely shipped from Mexico to the
United States, are not subject to such a requirement, nor is it
necessary.
Fruit fly/insect proof requirements on commodities such as
tomatoes from Israel or citrus from Spain are in place because the
fruit is often subject to long periods of exposure to the environment
while awaiting shipment to the United States at seaports or airports.
One hundred percent of the Hass avocado shipments destined to the
United States from Mexico are safeguarded in insect-proof warehouses
prior to being sealed in insect-proof, refrigerated trailers.
Based upon this reasoning, we believe that the current sealing
requirement is adequate and should remain in place.
Response: After careful consideration of the comments, APHIS has
decided to retain the provisions regarding sealing of containers. In
the proposed rule, we stated that our reason for changing from sealing
of containers to pest proof boxes, for safeguarding purposes, was
because some containers had been arriving at the port of entry with
broken seals. Seals could and were being broken by Mexican authorities,
to inspect containers for contraband. When the containers were
inspected by Mexican authorities, we were concerned that the shipment
could be exposed to possible infestation.
Upon further investigation, APHIS found that Mexico has effectively
addressed the issue of shipments arriving with broken seals. If a seal
is broken by a Mexican official, that official is to provide a specific
document stating that he/she has broken the seal. If the documentation
is not provided, U.S. inspectors use various methods to determine if
the shipment had been tampered with.
Data collected at the border reflects that Mexico has taken steps
to adequately address the broken seal issue. The number of shipments
that arrived at the ports of entry with broken safeguarding seals
decreased considerably, from 690 shipments in the 2002/2003 season to
231 in the 2003/2004 season. Of the 231 shipments arriving with a
broken seal during that last season, more than 86 percent had
documentation from the Mexican official who broke the seal. In
addition, the data show that none of the shipments arriving with broken
seals were compromised or infested with pests. As the commenter noted,
refrigerated containers present a very inhospitable environment for
tropical and subtropical arthropods and such conveyances do not provide
an adequate environment for insect activity such as oviposition. We
agree with the commenter as our data supports this statement.
Based on the above, we will continue to require sealing of shipping
containers at the packinghouse to safeguard each consignment as it
transits Mexico to the United States. This will not affect the results
or conclusions of the risk assessment. As long as sufficient measures
remain in place to safeguard the avocados during transit to the United
States, the conclusion of the risk assessment that likelihood of
introduction of quarantine pests is low will remain the same.
Therefore, this final rule does not require the avocados to be packed
in insect-proof cartons, loaded in insect-proof containers, or covered
with insect-proof mesh or plastic tarpaulin prior to leaving the
packinghouse.
Comment: Five years ago, the USDA representatives presented a plan,
called the ``systems approach to pest risk mitigation.'' The plan was
evaluated by two models, using the Monte Carlo modeling engine. The
first model, that of no import restrictions, was compared to the second
model, that of the systems approach. Had the evaluation established a
model for the current environment at the time, that of a complete ban
on Mexican avocado imports, the systems approach could never have
generated acceptable numbers in the modeling engine. Model 1, no
restrictions, indicates a likelihood of infestation by the seed weevil
in an average of 95 years, seed moth in 355 years, and the fruit fly in
72 years. Model 2, the systems approach, indicates a likelihood of
infestation by the seed weevil or fruit fly in 10,000 years, and by the
stem weevil in 11,000 years. In reality, a complete ban on Mexican
avocado imports into California in the current environment has led to
two Mexican fruit fly infestations in the last 5 years.
[[Page 69757]]
Infestations by persea mite, thrips, and other pests have also occurred
within the last 12 years.
Response: Contrary to the commenter's conclusion that prior
infestations in California were the result of APHIS policy, there is no
evidence linking any infestations with commercial Hass avocados from
Mexico imported under the program regardless of a model used to predict
risk. The current APHIS risk assessment based its predictive model on
detection samples for the quarantine pests of concern. The samples were
program fruits cut during orchard certification surveys, packinghouse
inspections, and at the border. This produced a sample of over 10
million fruit taken over 6 years of the import program. The results of
the model are presented as expected numbers of infested avocados
entering the United States annually.
Comment: The fruit fly study does not address susceptibility of
late season avocado to infestation. From my own grove operation, I have
noted the following late season fruit quality characteristics which
could influence fruit susceptibility to insect infestation:
Fruit oil content is higher than early season fruit;
Seed tap root pushes through bottom of avocado giving easy
access to fruit interior;
Fruits start ripening on the tree; and
Handling time window shortens; mature fruit ripen quicker.
The fruit fly is not the only insect pest of concern. How does late
season avocado fruit impact the occurrence of stem and seed weevils?
What other late season Mexican insect pests must the industry be
concerned with? USDA has failed to account for the possibility of the
harvest of mature, ripe avocados that could harbor fruit fly eggs and
larvae.
Response: The Aluja et al. fruit fly study included avocados tested
on the tree at maturity levels from low to high percentage dry matter,
indicating early and late season fruit. ARS reviewed the study and
concluded that commercial Hass avocados are a very poor host for
Mexican fruit fly and did not single out any maturity stage on the tree
as particularly vulnerable. APHIS has concluded, based partly on the
ARS findings, that there is a low likelihood of Anastrepha species of
fruit flies being in program fruit. APHIS recognizes that other
internal quarantine pests analyzed in the risk assessment may be
present in mature fruit, but that systems approach measures maintain
the low likelihood of their introduction in program fruit, which has
been validated by the fruit sampling that has been conducted over 6
years of the program.
Stem weevils are found in all varieties of avocados and can be
especially abundant in trees not managed under the program. Stem
weevils can be detected both by visual examination of cut fruit and by
the highly visible exudates the larvae leave on tree branches. Orchards
will be surveyed semiannually for stem weevil, and if weevils or weevil
signs are found, certification is denied or suspended. Additionally, if
stem weevil larvae are found in fruit cut at the packinghouse or at the
border, the regulations require the removal of the originating orchard
from the program immediately and avocado exports from that orchard will
be suspended until APHIS and the Mexican NPPO agree that the pest
eradication measures taken have been effective and that the pest risk
within that orchard has been eliminated.
Additionally, and as previously stated, USDA considers mature,
commercial Hass avocados to be a very poor pathway for thrips. Several
research studies (e.g., Hoddle, 2002; Yee et al., 2003, cited in the
PRA) have demonstrated that Scirtothrips perseae, the avocado thrips
(which APHIS considers as probably representative of other pest thrips
species) lays eggs in small, immature fruits and tender leaves, and
does not feed on or lay eggs in mature fruit, and is, therefore,
unlikely to be imported with the fruit. This is supported by the fact
that there have been no thrips interceptions by APHIS on commercial
Hass avocados from Mexico since the program began.
Comment: In order to provide time to reconcile critical issues on
safe agricultural import practices and create parity in U.S./Mexican
trade policy, there should be no expansion of Mexican avocado imports
beyond the 31 currently approved States for a period of 7 years. During
that time, U.S. avocado producers should have unrestricted access to
designated Mexican markets with allowances for comparable levels of
export (tonnage). Also during the 7-year period, the USDA should:
Conduct a comprehensive research program on U.S. avocado
farms to document existing exotic pest and disease problems;
Monitor U.S. avocado farms to measure the increase or
decrease to U.S. avocado production and costs from exotic pests; and
In conjunction with the California and Florida avocado
commissions, verify Mexico's compliance with and support of the U.S.
avocado export program.
After this 7-year period, USDA may consider easing restrictions
subject to the following conditions:
U.S. avocado farms experience no significant additional
impacts due to exotic pests or expanded quarantines;
U.S. avocado exports to Mexico have reached a comparable
equilibrium measured in tonnage and price with Mexican avocado exports
to the United States; and
No new research or data demonstrate greater future risk
from exotic pests or diseases from expanding Mexican avocado imports
into additional U.S. States.
Assuming these conditions have been met, USDA may propose to allow
further Mexican exports to U.S. States other than California and
Florida and the States which directly border California.
Response: APHIS has no authority under the Plant Protection Act to
prohibit or restrict the entry of an article merely to create parity in
trade between the United States and another nation. Further, as a
signatory to the International Plant Protection Convention, the United
States has agreed not to prescribe or adopt phytosanitary measures
concerning the importation of plants, plant products, and other
regulated articles unless such measures are made necessary by
phytosanitary considerations and are technically justified. Based on
the conclusions of the APHIS risk assessment, we do not believe that
there is a technical justification for the 7-year delay or other
measures suggested by the commenter. We have, however, implemented a 2-
year delay for imports into and distribution within California,
Florida, and Hawaii in response to other comments we received on the
proposed rule. This restriction will provide APHIS an opportunity to
further substantiate the effectiveness of the mitigation measures under
the expanded program.
Comment: USDA's proposed rule on the Mexican Hass avocado import
program includes several proposed changes to the protocol under which
the program operates. For example, the Department has proposed
conducting semiannual, rather than annual, pest surveys at the
municipality and orchard levels. We support the idea of semiannual
surveys during the wet and dry seasons. It is appropriate, too, to
leave the details of how and when surveys will be conducted to the
annual work plan developed by Mexico's national plant protection
organization and APHIS. In our view, it is imperative, however, that
areas with wild or backyard avocado trees continue to be included in
pest surveys conducted at the municipality level. These areas represent
the greatest
[[Page 69758]]
potential source of infestation or reinfestation of export orchards.
Response: The proposed rule did not include a proposal to eliminate
the surveying of areas with wild or backyard avocado trees during
municipality surveying. The proposal was to eliminate specific language
on the surveys from the regulations because this information would be
included in the workplan. Areas with wild or backyard avocado trees
will continue to be surveyed.
Comment: The regulations in Sec. 319.56-2ff (c)(2)(iii) state that
``avocado fruit that has fallen from the trees must be removed every 7
days and may not be included in field boxes of fruit to be packed for
export.'' I request and strongly recommend this permissive ``may'' be
strengthened to a mandatory ``must'' or ``shall.''
Response: The phrase ``may not be included'' does not contain a
``permissive'' element as the commenter suggests. As written, the text
of Sec. 319.56-2ff (c)(2)(iii) serves as clear prohibition on the
inclusion of fallen fruit in field boxes of fruit to be packed for
export.
Comment: Serious program infirmities must be addressed before
expansion can occur: There are no standardized procedures, training, or
oversight for fruit cutting; fruit cutting techniques are ineffective
at detecting the eggs, first instar, and second instar larvae of fruit
flies or the stem weevil, rendering USDA's risk probabilities
unreliable; improper pest survey timing has underrepresented pest
population levels; fruit fly trapping methodology and servicing are
flawed; fruit fly response and treatment procedures are inadequate. The
inspection process is not sufficient. USDA inspectors may simply drop
an avocado into a slicer and check for a mature worm rather than using
a loupe (a portable microscope lense). The larvae for almost every pest
are not visible to the naked eye. Additionally, the Department of
Homeland Security (DHS) is now doing port inspections that focus more
on drugs, guns, etc., than plant pests.
Response: Regarding training and oversight for fruit cutting,
inspectors are trained to detect pathway pests based on the biology of
the pest and what signs or symptoms of infestation to look for. They
have hand lenses that they may use, if they need them, to complete an
inspection. Pest damage, rot, and most stages of each of the internal
pests are visible to the trained inspector. APHIS is aware that there
is a possibility that a pest may escape detection and has accounted for
this uncertainty in the current risk assessment. Fruit cutting is only
one of the multiple measures of the systems approach that mitigates
pest risk.
Survey timing: Under the modified systems approach, semiannual
surveys will be conducted at the municipality and orchard level.
Municipalities must be free of Ceratitis capitata, Conotrachelus
aguacatae, C. perseae, Heilipus lauri, and Stenoma catenifer before
they can be certified to export avocados to the United States. In
addition, orchards must be certified free of Copturus aguacatae.
Trapping is conducted in orchards for Anastrepha spp. fruit flies. Both
the regulations and the workplan specify what mitigation measures must
be taken when a pathway pest is detected in a certified orchard. The
time periods selected for the surveys were based on the biology of the
pests. Additionally, the fruit cutting will be conducted in the
orchard, packinghouses, and at the port of entry. Since the expansion
of Mexican avocado imports in 1997, none of these pests have been
intercepted during inspections of fruit at packinghouses or upon
inspection at the U.S. border ports.
Fruit fly trapping is modeled after other APHIS programs for
consistency, and the actions are based on the biology of the pests and
fruit fly detections.
Comment: Because of insufficient analysis, USDA should subject its
risk assessment to rigorous, external peer review, to incorporate the
best science available and to establish a more credible research base
for its decision to allow imports to particular States. Serious program
deficiencies must be addressed before Mexico is allowed to expand
exports to additional States. For example, there are no standardized
procedures, training, or oversight for fruit cutting during pest
inspections. Fruit cutting techniques are ineffective at detecting the
eggs and larvae of fruit flies or the stem weevil, thus rendering
USDA's risk probabilities unreliable. Additionally, improper pest
survey timing has underrepresented pest population levels; fruit fly
trapping methodology and servicing are flawed; and fruit fly response
and treatment procedures are inadequate. I urgently request that this
program be suspended for further study by independent experts in the
field and in consultation with the industry because the scientific
basis for allowing Mexican fruit into the United States was based on a
joint USDA-Mexico study for one growing season in Mexico. This study is
a very small basis upon which to overthrow 80 years of exclusion and
contains much that is controversial and open to question.
Response: The APHIS risk assessment has been made available for
public review twice. First, we made the draft risk assessment available
for public comment for a total of 90 days through a notice of
availability published in the Federal Register on June 16, 2003 (68 FR
35619, Docket No. 03-022-1), and a subsequent extension of the comment
period (68 FR 48595-48596, Docket No. 03-022-2, published August 14,
2003). An updated version of the risk assessment was also made
available for public comment for an additional 60 days as part of our
May 2004 proposed rule. We received numerous comments regarding the
risk assessment in both instances, including comments from professional
risk analysts, private risk consultants, and university and government
scientists, and updates have been made to the risk assessment to
address those comments. Further, the fruit fly study (Aluja et al.
2004) noted by the commenter that is cited in the risk assessment was
subjected to rigorous peer review prior to its publication in the
Journal of Economic Entomology and was likewise reviewed by USDA fruit
fly experts in ARS and APHIS. The input that APHIS received from ARS
follows the tradition and guidelines of peer review. The ARS experts
offered their own interpretation of the scope and applicability of the
findings. This information suggested that we should update our risk
assessment, and we considered the ARS input in preparing our updates
and changing our classification of the host status of Hass avocados. We
believe that these documents do, in fact, provide a credible research
base for our decisionmaking with regard to the expansion of the Mexican
avocado export program to additional States and the Secretary's
determination is based on the findings of the risk assessment and her
judgment that the application of the measures required under Sec.
319.56-2ff would prevent the introduction or dissemination of plant
pests into the United States, thus we do not believe that the program
suspension recommended by the commenter is appropriate.
The specific issues raised by the commenter regarding fruit
cutting, pest surveys, inspection, and fruit fly trapping are addressed
in the response to the previous comment.
Comment: In the proposed rule, APHIS states that even if an
infested avocado were to arrive in an area of the United States where
host material was present, several additional conditions are required
for pest establishment (i.e., the pests survive during transportation
and storage; the infested avocados must
[[Page 69759]]
be discarded in close proximity to host material; the pests must find
mates; the pests must successfully avoid predation; the adult pests
must find host material; the climatological and microenvironmental
conditions must be suitable; and they must escape detection and
subsequent eradication measures). APHIS admits that information that
would allow quantifying these conditions is not currently available.
Without that information, how can the Secretary conclude that it is not
necessary to restrict Mexican avocados for phytosanitary reasons?
Response: As stated in the proposed rule and in this document, the
Secretary's determination is based on the findings of the risk
assessment and her judgment that the application of the measures
required under Sec. 319.56-2ff would prevent the introduction or
dissemination of plant pests into the United States. The risk
assessment contains both quantitative and qualitative elements, and our
final expression of a likelihood of introduction is a descriptive
statement. The results of the quantitative analysis do not equate to
likelihood of establishment. Rather, they express the likelihood of an
infested avocado being discarded in a suitable location; establishment
and spread would require the additional steps noted by the commenter,
which substantially reduce the likelihood of establishment. People
generally consume the fruit they purchase and dispose of the waste
material in a manner (such as in plastic bags that are landfilled or
incinerated \3\) that precludes the release of pests into the
environment. For these reasons, our final expression of a likelihood of
introduction is a descriptive statement. These factors, in combination
with the results of the quantitative analyses, led APHIS to conclude
that the likelihood of establishment of infested avocados through the
commercial pathway of Hass avocados imported from the State of
Michoacan and produced using the systems approach is low.
---------------------------------------------------------------------------
\3\ Citations pertaining to disposal patterns can be found on
pages 9 and 36 of ``Importation of Avocado Fruit (Persea americana
Mill. var. ``Hass'') from Mexico: A Risk Assessment,'' September 17,
2004.
---------------------------------------------------------------------------
Comment: Much is made in the risk assessment of the maximum
likelihood estimate of the number of infested avocados imported in any
year being zero. However, a more interesting statistic, from the point
of view of the agricultural industry, is the probability of no infested
avocados being imported in any year. From this one can calculate the
probability of importation of infested fruit over a period more
meaningful to agriculture. For tree crop agriculture, 20 years or more
is a meaningful planning horizon, and the formula p20 = 1-(pi )20 where
p20 = the probability that 1 or more infested fruit will be introduced
during a 20 year period and pi = the probability of no infested
avocados being imported (assuming pi for i = 1 to 20 are independent)
in any year. This p20 can be adjusted for the likelihood that any
imported infested fruit will result in establishment of the pest in a
producing area, as presented in the text. From this estimate of the
probability of establishment of the pest, multiplied by the cost of
eradication plus losses to growers, one can calculate the expected cost
of allowing the importation of Mexican avocados. This calculation has
much more meaning to the industry than the mere statement that the
probability of infestation from imported avocados (in any one year) is
``low.''
Response: The risk assessment was prepared to assist APHIS in
evaluating Mexico's request to expand the scope of the existing Hass
avocado import program. As such, its purpose was to analyze the risks
of expanding the Mexican Hass avocado import program to authorize
imports throughout the United States year-round. Although the method
presented by the commenter for estimating the likelihood of no infested
avocados being imported in any year could produce useful information
from the point of view of the agricultural industry, it does not appear
that the endpoint sought by the commenter--i.e., the ``expected cost of
allowing the importation of Mexican avocados''--could be achieved in a
meaningful way. Two of the factors that would be considered in the
calculation proposed by the commenter--the cost of eradication plus
losses to growers--could vary enormously, depending on the nature and
scope of the pest outbreak to be eradicated and the effects the
particular pest might have on production, so the final estimates would
necessarily be very broad in their range.
Comment: With respect to the risk assessment, USDA has ignored the
directive of the Congressional Appropriations Committee, which stated
``The Committee directs APHIS to include independent, third party
scientists in the development of any PRA for Mexican avocados prior to
the publication of any PRA in the Federal Register.'' Further, USDA
ignored fundamental disagreements between its own scientists regarding
the conclusions drawn by the underlying research data, and has not
released that research data, thus not allowing time for independent
review before the risk assessment was issued. Finally, the USDA has
assigned only a minimum crew of 11 men to survey and monitor an
increase of Mexican avocado acreage from 3,700 acres in 1997 to over
53,000 certified acres.
Response: On June 16, 2003, APHIS published a notice in the Federal
Register to inform the public of the availability of a new draft PRA
that was prepared in response to the Mexican Government's request that
the avocado import program be expanded to include all 50 States for the
entire year. In accordance with the Committees' direction to include
independent, third party scientists in the development of PRAs for the
avocado program, APHIS scientists consulted with independent subject
matter experts from a variety of accredited academic institutions
during the development of the draft PRA. These institutions included,
among others, Florida A&M University, the University of Florida, and
the Institute of Ecology in Veracruz, Mexico. APHIS scientists also
consulted with ARS researchers from various locations, including Hawaii
and Texas. APHIS extended the original comment period on the PRA for an
additional 30 days and accepted public comments on the assessment until
September 15, 2003. The public comment period served as an additional
opportunity for all members of the public, including independent
researchers and members of academia, to evaluate the draft PRA. After
reviewing all the comments, we determined that it was appropriate to
move ahead with a proposed rule. We reported our action to the
Committee prior to publishing the proposed rule.
As noted above, APHIS solicited the opinion of ARS scientists
regarding the fruit fly research presented in Aluja et al. (2004) and
worked with those scientists to understand the similarities and
differences between our Agencies' interpretations of the conclusions
drawn in the study. The APHIS position was to initially agree with Dr.
Aluja's findings that commercially packed Hass avocados are not a host
of Mexican fruit flies. ARS took a slightly more conservative position
that those avocados are a very poor host of the Mexican fruit fly. The
difference in the categorization of the Hass avocado's host status did
not effect the level of risk in the APHIS risk assessment. APHIS did,
in the final analysis, change its categorization of the Hass avocado
host status based on ARS's conclusions. Changing our conclusions on the
host status made it then necessary to calculate the pest risk for fruit
fly in our risk assessment. After performing these calculations, we
found the likelihood of
[[Page 69760]]
fruit fly infestation through commercial shipments of Hass avocado from
Mexico to be very low. The pest risk did not change as a result of
labeling the fruit fly as a poor host rather than a non-host.
The Aluja et al. research used in the APHIS risk assessment was
published in the Journal of Economic Entomology in April 2004. Because
the proposed rule was published on July 23, 2004, and was open for
public comment for 60 days, we believe that the research was available
to the general public with ample time for examination. Because the
Aluja et al. study had already been accepted for publication in 2003,
it was unavailable for distribution during the June-September 2003
comment period for the draft PRA.
Finally, as previously stated, APHIS' role under the operational
workplan provided by the regulations is to provide management and
monitoring of the activities specified in the workplan, e.g., trapping,
surveying, and packinghouse inspections. While APHIS personnel do not
necessarily conduct these activities themselves, they do monitor
Mexican officials' compliance with workplan specifications. The
staffing level of APHIS personnel is sufficient to ensure that APHIS
meets its requirements under the workplan and that other signatories
are in compliance with the regulations. The lack of pest detections in
the orchard, packinghouse, and border inspection since the program
began in 1997 is evidence that the regulations and workplan are being
complied with.
In addition, as more orchards have applied for certification, it
does take longer for inspectors to perform the initial inspection
before the first shipping date of October 15. The inspectors have had
to start inspections earlier before that date each year to finish
inspecting all of the orchards. Additional inspectors would be hired to
inspect the increased acreage within the required time-frame. If the
inspection does not occur within that timeframe, the orchards would not
be certified. All orchards must be inspected using the same work plan
criteria, as the records show. APHIS keeps lists of all the orchards
inspected by name.
Comment: Limiting Mexican fruit to colder climates makes sense.
Maybe Mexican fruit could be allowed only if it were packaged and
processed in a frozen type of product.
Response: Some types of processed avocado products are allowed into
the United States and can go to all States. Whether or not the
processed product is allowed in depends on whether the processing
mitigates any pest risk. Frozen avocados are allowed entry if they meet
these requirements:
An import permit is required,
The seeds must be removed; and
The avocados must be at or below 20 [deg]F at the time of
arrival at the port of entry. If the temperature of the avocados is
higher than specified, the avocados will be required to meet the import
requirements of fresh avocados.
As we discussed in the proposed rule and in this final rule, we do
not believe it is necessary to limit the importation of Mexican
avocados to cooler climates because of the mitigations in place and the
findings in our risk assessment.
Comment: Cultural practices used by Mexican avocado growers,
including unsanitary Mexican growing conditions in which human waste is
used, could cause illness in U.S. consumers.
Response: APHIS has no information to suggest that human waste is
used in avocado production in Mexico. Even if it were used in some
cases, the fact that avocados are a tree fruit make it unlikely that
avocados on the tree would be contaminated as a result.
Comment: Regarding this proposal to establish limitations on the
entry of Hass avocados into States with commercial production of
avocados in the United States, we believe that the scientific support
contained in the Aluja et al. study, which documents the scientific
evidence showing that the Hass avocado is not a host for Anastrepha
spp. fruit flies, together with there having been no detections of
quarantine pests throughout 7 years of exporting, ensures an adequate
level of plant health protection for the United States, including the
avocado-producing areas.
The above is also supported by the risk assessment prepared by USDA
in which the probabilistic analysis, based on the information from cut
fruit exported to the United States, establishes that the annual number
of fruits infested by quarantine pests imported into the United States
is zero.
Response: To provide APHIS an opportunity to further substantiate
the effectiveness of the mitigation measures under the expanded program
as discussed in our risk assessment on pages 4, 10, and 11, APHIS has
decided to delay for 2 years the implementation of the importation of
Hass avocados from Mexico into all 50 States. Rather, APHIS will allow
for the immediate importation of those avocados into all States except
for California, Florida, and Hawaii, which are avocado producing
States, to monitor the program and gather efficacy data under
production conditions during all months of the year. While APHIS has
concluded in the risk assessment that there is low likelihood of
introduction of fruit flies based on the findings of the Aluja et al.
study, as well as the conclusion of ARS that commercial program Hass
avocados are a ``very poor host'' of Mexican fruit fly and our own
analysis of detections based on over 10 million fruit sampled over 6
years of the import program, we believe the delay will offer the
opportunity to further substantiate these findings. In the risk
assessment, APHIS was asked to determine the likelihood of introduction
of quarantine pests of concern in program Hass avocados from Mexico.
Based on the above evidence presented in the risk assessment, a finding
of ``non-host'' for Anastrepha spp. fruit flies was not necessary for
APHIS to draw the conclusion of low likelihood of introduction. During
the near future, however, APHIS plans to work with Mexico, ARS, and
independent researchers in the scientific community to form a consensus
on the host status issue of the Hass avocado and Anastrepha spp. fruit
flies.
Comment: To protect our avocado-growing States from inadvertent
transmission of infested fruit, buffer zones of additional States is a
must. There is ample clientele for Mexico to sell, promote, and educate
outside our vulnerable areas. Mexican producers' insistence to sell
fruit in California and other growing areas shows proof of their intent
to destroy their competitors by causing pest infestations in an area
that would not have the tools to deal with the infestation because of
government regulations.
Response: APHIS has determined that the likelihood of introduction
of quarantine pests of concern would not be significantly reduced by
buffer States any more than just prohibiting movement into or through
California, Florida, and Hawaii for 2 years for the following reasons:
1. The likely buffer States, which would be Alabama, Arizona,
Georgia, Nevada, and Oregon, do not produce avocados or have special
quarantine regulations against avocados moving through their States or
moving into the prohibited States, but California and Florida do have
adequate quarantine regulations against certain agricultural products
moving within them. Since Hawaii is an island, it would not need
``buffer States.''
2. The avocado-growing area of Florida is confined to the southern
half of the peninsula, therefore the State's northern counties serve as
buffers to the producing counties. The avocado-growing areas of
California are more extensive, but they are either bordered
[[Page 69761]]
by the Pacific Ocean on the west, large expanses of mainly desert
counties of California or desert areas of Nevada and Arizona to the
east, a wide expanse of non-avocado-growing counties to the north, and
Mexico to the south.
Comment: I strongly oppose any provisions that would allow Mexican
Hass avocados to enter Florida until at least 1 year's worth of
monitoring data, both within Mexico and shipments to other States, has
been collected to demonstrate that the shipments are free of plant
pests of quarantine significance. Scientific data should be provided
that proves that the Mexican Hass variety of avocados is, under all
environmental conditions, resistant to all fruit flies known to be
established in Mexico. There is concern that environmental conditions
may cause the avocado to ripen prematurely while still on the tree and
therefore making it susceptible to fruit flies. I would also like to
receive information regarding the Mediterranean fruit fly trapping
program that is being utilized in Hass avocado production areas in
Mexico.
Response: As previously stated, APHIS is implementing a 2-year
delay on the importation of Hass avocados from Mexico into California,
Florida, and Hawaii. Additionally, in their review of the Aluja et al.
fruit fly research, ARS noted that there were some larvae that
developed in a few infested fruit, that the conditions enabling the
larvae to develop were not known, and that viable adults were not
produced. Even allowing for the few larvae, ARS experts still concluded
that the fruit were a ``very poor host'' of Mexican fruit fly. The
Aluja et al. research included fruit at all levels of maturity
including fruit off of the tree that had been allowed to ripen for
three hours. The conclusion was that fruit still on the tree was not a
host for the fruit fly.
The Mexican Mediterranean fruit fly trapping areas include all of
the avocado-growing areas of Michoacan. APHIS monitors all aspects of
the pest detection protocol in all avocado producing municipalities
that export to United States, including trapping for Medfly. Under our
regulations, Medfly is trapped at a density of 1 trap per 1 to 4 square
miles. Any findings of Medfly must be reported to APHIS.
Comment: There are a number of issues of concern not addressed in
the underlying research and the APHIS risk assessment used to justify
the expansion of the Mexican Hass avocado import program. These issues
are:
The APHIS risk assessment does not provide scientific data
covering phenology of fruit or the changes in soluble solids throughout
the year as it relates to pest infestation;
Traceback methods have been ineffective in the past;
The movement of fruit from an area of low pest prevalence
may not be accurate. The number of infested fruit could be much higher
than predicted;
There is no information about temperature ranges during
exposure period and effect of temperature changes on quarantine pests;
USDA has never clarified how the Mexican fruit fly
infestation entered Valley Center, CA, in 2002.
Response: The Aluja et al. study did include the summer months
June, August, and September as well as other months, fruits of a range
of sizes that were commercially mature, and mature fruit attached to
the tree as well as off the tree. This covered a range of fruit sizes
and soluble solid ranges. APHIS and ARS both concluded, based on the
study, that the fruit were a very poor host for Mexican fruit fly.
Fruit of a range of sizes and solid content have been present on the
trees during the 6-month shipping season, and only mature fruit are
exported, which may represent a range of soluble solid contents.
Sampling is done throughout the shipping period as well as in the
orchards before the season, so a range of soluble solid contents that
may occur in mature fruit would be in the sampled fruit. The shipping
season, which has occurred during a 6-month period with wide
temperature fluctuations, and the inspections conducted during that
period were considered in the risk assessment. The risk assessment
describes fruit sampling by other researchers that included most months
of the year when pests would be likely to be found.
Regarding tracebacks, because no infested exported fruit have been
detected in 6 years of sampling, no tracebacks have been necessary in
the program. However, because of required labeling on the boxes, the
necessary information is available to trace fruit back to packinghouses
and orchards if necessary.
Additionally, APHIS has monitored infestation through inspection of
program exported fruit. Predictions of infestation are based on the
inspections. The ``area of low prevalence'' concept is not an element
of the systems approaches that is relied upon under the importation
program. In relation to fruit flies, orchard trapping and subsequent
eradication if there is a detection are required under the program.
Finally, as stated previously, the Valley Center infestation
stemmed from unknown origins and not from a legally imported commercial
Hass avocado shipment from Mexico, which were prohibited from entering
California. The Mexican fruit fly could have been introduced into
California through a number of pathways, including the smuggling of
many different kinds of fruit.
Comment: Allowing Mexican avocados into California would be a
signal to the public that it is permissible to bring avocados across
the border from sources that have not been inspected. Therefore, USDA
should formulate a rule that includes a permanent provision to not
allow Mexican Hass avocados to be imported into California or any other
avocado-producing State.
Response: It is stated in the regulations that commercial shipments
of Hass avocados from Mexico cannot be imported or distributed into
California, Florida, and Hawaii for the first 2 years of the expanded
importation program. Inspectors will continue to check returning
travelers for unapproved agricultural commodities, including avocados.
Our regulations are enforceable under the provisions of the Plant
Protection Act.
Comment: Based on the results reported in the Aluja et al. study,
which established that the avocado should be considered a non-host for
Anastrepha spp. fruit flies, we request that A. ludens and A. striata
be removed from the list of pests of Hass avocados grown in Mexico that
follow the pathway.
Response: APHIS has concluded in the risk assessment that there is
low likelihood of introduction of fruit flies based on the finding of
ARS that commercial program Hass avocados are a ``very poor host'' of
Mexican fruit fly, and on analysis of detections based on over 10
million fruit sampled over 6 years of the import program. During the
near future, however, APHIS plans to work with Mexico, ARS, and
independent researchers in the scientific community to form a consensus
on the host status issue of the Hass avocado and Anastrepha spp. fruit
flies. At that time we will evaluate all Anastrepha spp. fruit flies
and determine which species should be removed from the pest list.
Comment: For reasons not known, it appears that APHIS has accepted
the unsupported and seemingly arbitrary opinion of ARS over the
conclusions of a team of scientific experts, headed by one of the
foremost fruit fly researchers in the world, after 2 years of exacting
research on the precise issue, and subsequently peer reviewed and
published in the Journal of Economic Entomology. In short, a three-page
memo based upon the opinion of two
[[Page 69762]]
ARS researchers with limited history in this area, containing one
reference to a study that was done on a different avocado, in a
different environment, with different insects has overruled the years
of painstaking research and the peer reviewed study published in the
Nation's leading relevant scientific journal.
Even the California Avocado Commission (CAC) supports this concept.
On December 20, 2002, in a letter to Dr. Richard Dunkle, Deputy
Administrator, USDA-APHIS-PPQ, an official of the CAC requested an
``alternative protocol'' that would allow Hass avocado growers in the
core area of the Valley Center Mexican fruit fly outbreak in California
the opportunity to harvest and distribute Hass avocados under a system
that mirrors the Mexican Hass avocado export program. Surely, the CAC
would not make this request if they thought that the Hass avocado would
pose the threat of moving Anastrepha spp. fruit flies out of the
quarantined area into other areas of California and the country.
Response: The claim that non-experts from ARS provided input is
incorrect. Whereas those particular ARS experts may not have published
extensively on fruit flies, a biological scientist is perfectly able to
review documents for scientific validity. The input from ARS follows
the tradition and guidelines of peer review, and the ARS experts did
not say that the article from Aluja et al. was invalid; rather the ARS
experts offered their own interpretation of the scope and applicability
of the findings. This information suggested that we should update our
risk assessment, and we considered the ARS input in preparing our
updates. We fully intend to continue to seek and incorporate ARS
expertise and guidance in our risk analysis products.
Comment: The Mediterranean fruit fly (Ceratitis capitata) is not
present in Mexico. Only two outbreaks have occurred along the border
between Chiapas and Guatemala, which were treated through the National
C. capitata Campaign, in which the Governments of the United States,
Mexico, and Guatemala participate, and those outbreaks were controlled.
For this reason, there is no justification for establishing measures to
prevent the introduction of this pest by means of the entry of Mexican
Hass avocados into the United States, since the avocado exporting zone
is located far away from the site of the outbreaks in Chiapas.
Response: While the Mediterranean fruit fly is currently considered
eradicated in Michoacan, there have been occasional limited outbreaks
beyond those cited by the commenter. Therefore, APHIS will continue
monitoring for Medfly. APHIS continues to consider Mediterranean fruit
fly monitoring and control as important elements of the Hass avocado
program systems approach, as avocados are considered to be a good host
of Mediterranean fruit fly. The pest is regarded in the risk assessment
as a quarantine pest that could be in the pathway if it is detected in
the avocado-producing area.
Comment: An expansion of the Mexican Hass avocado import program
will lead to increased air pollution and unsafe Mexican truck traffic.
Response: USDA has no authority over emissions or safety standards
for Mexican trucks.
Comments on the Economic Analysis
A number of commenters raised issues regarding the economic
analysis that accompanied the proposed rule. These issues are grouped
into three sections: The model and analysis, effects for California
avocado producers, and other comments.
The Model and Analysis
Comment: The base period for the analysis is October 15, 2000,
through October 15, 2002, with base figures being an average of these 2
years. A possible problem with the use of these 2 years is that the
rules for Mexican avocado imports changed, effective November 2001, and
only 1 of the 2 years included the expanded number of States and time
period that are currently effective. This choice of base period tends
to understate likely Mexican imports.
Response: We agree with the commenter and have changed the baseline
used in the analysis for this final rule to the 2-year period October
15, 2001, to October 15, 2003.
Comment: There is evidence that the producer level price elasticity
of demand may be even less than the -0.57 [used in the proposed rule].
If a more inelastic coefficient were used (-0.50 or lower), the
estimated price impacts of Mexican imports on California producers
would be greater, especially in the short run.
Response: USDA agrees with the commenter that price elasticity of
demand seems to have generally decreased over time. However, year-to-
year fluctuations occur, due to changes in real price levels. In the
analysis for the proposed rule, the elasticity of demand was adjusted
downward from an earlier study (``An Economic Evaluation of California
Avocado Industry Marketing Programs 1961-1995,'' by Hoy F. Carman and
R. Kim Craft, Giannini Foundation Research Report Number 345,
University of California, July 1998), from -0.75 to -0.57. In the
analysis for this final rule, the price elasticity of demand used for
California avocados is -0.63, based on the parameters estimated in
Carman and Kraft and the observed level of per-capita consumption and
the real producer price of California avocados from our baseline data.
This elasticity is somewhat higher than that used in the analysis for
the proposed rule due to a higher real producer price in the new
baseline.
The commenter provides an equation by which he has estimated a
price elasticity of -0.53 at average prices and quantities, and an
average of -0.44 for the period 1996/1997 through 2000/2001. (We
presume that the description of Qt as consumption of
avocados from all sources is a notational error, since the equation is
supposed to represent the demand function for only California
avocados.)
The small changes suggested by the commenter would affect the
results of the analysis insignificantly. As the commenter himself
states, ``Estimated coefficients from a recent demand function indicate
that the f.o.b. [free-on-board] level price elasticity of demand for
California avocados may be slightly more inelastic than -0.57, but this
should have only a small effect on the final estimates.'' The overall
conclusions of the study in terms of net social benefits of the rule
would still hold.
Comment: Mexico's market share in currently approved States during
Period 1 (October 15 through April 15) is understated in the analysis
for the proposed rule because the baseline is not current. A more
current baseline would show Mexico's larger market share, thereby
affecting the shift parameters and resulting in larger Mexican imports.
Response: We agree with the commenter. Based on the updated
baseline (October 15, 2001, to October 15, 2003) used in the analysis
for this final rule, we find nearly an 11 percent increase in Mexico's
market share in the currently approved region and time period, from
57.5 percent to 68.3 percent.
Comment: The retail food sector has significant market power. At
the very least, the analysis should point out that retailers
(middlemen) will use their market power to capture a portion of the
welfare gains.
Response: If food retailers do possess some degree of market power
in pricing avocados, we agree that retailers will
[[Page 69763]]
use their market power to capture some of the welfare benefits.
However, this will largely represent a change in the distribution of
the welfare gains from the proposed rule, with some of the benefits
being transferred from consumers to retailers. There will be some
increase in the deadweight loss due to an increase in the retail
margin, but the magnitude of deadweight loss is typically a very small
portion of the overall welfare change.
As the commenter suggests, we acknowledge in the analysis for this
final rule that if retailers do possess market power in the pricing of
avocados, a portion of the welfare gains to consumers will be
transferred to retailers, with a resulting small deadweight loss. In
this case, the overall welfare gain will be slightly overstated.
Comment: The price elasticity of supply used in the analysis, 0.35,
is overly elastic. Perennial crop acreage adjustments are lagged and
occur over many years. California production will change very little in
the first year or two after a rule change, with the result that prices
may be lower than projected for several years. The decrease in supply
will occur over time as some producers go out of business and others
remove trees in response to low prices and returns.
Response: The elasticity of supply used in the analysis for the
proposed rule was based on elasticities used in previous avocado
studies. We agree with the commenter that there will be limited
opportunity for producer adjustments during the first year following
publication of the final rule. In the analysis for the final rule, the
time assumed for moving to the new equilibrium is 2 years. Due to the
uncertainty of the magnitudes of the supply and demand elasticities in
the model, we conduct a sensitivity analysis that considers alternative
elasticity values. For the supply elasticity, we consider a range of
0.05 to 0.65. As is reported in the analysis accompanying this final
rule, the estimated standard deviations of the estimated changes in the
producer price of California avocados are relatively small. Thus, the
model results are not very sensitive to the choice of the value of the
supply elasticity within this range.
Comment: The initial values for the shift parameters for Region A
during Period 1 should be applied to Regions B and C during Period 1.
Adjustment of the shift parameters for Period 2 (April 15 to October
15) in all three regions will be a judgment call. I suggest that the
parameter for Mexico must be increased significantly (at least to
midway between Chile's parameter for Period 2 and Mexico's parameter
for Region A, Period 1), with proportional reductions in the parameters
for the United States and Chile.
Response: For time period 1, we disagree that the initial values of
the shift parameters for Region A should be applied to Regions B and C.
First, consumers in Region A have been able to purchase fresh Hass
avocados from Mexico for an extended period of time; since 1997 for
some consumers. Because individual preferences are usually thought to
evolve slowly over time, applying the shift parameters for Region A to
the other regions during Period 1 would likely overstate the increase
in demand for Mexican avocados. We believe that it is more likely that
consumers in Regions B and C will maintain a slight preference for
California avocados, at least in the short run. This belief is based on
the observation that the quantity market shares for avocados from
California and Chile for Regions B and C in the baseline data for the
final rule are almost equal. The wholesale price premium for California
avocados in both regions implies a preference towards California
avocados. (The initial values of the shift parameters are approximately
0.6 for California avocados and 0.4 for Chilean avocados.) This
preference may be a result of marketing activities by the Californian
Avocado Commission or consumer perceptions that fruit from California
is fresher than fruit from Chile. In the analysis of the final rule,
the shift parameters for Regions B and C during Period 1 are adjusted
to 0.4 for California avocados and 0.3 each for Chilean and Mexican
avocados.
For period 2, we disagree with the commenter that the shift
parameters for Mexican avocados should be increased significantly. Due
to seasonality in production, we believe that the preference parameter
for Californian avocados should be higher in Period 2 than in Period 1.
More fresh avocados are available from California than from Chile and
Mexico during the summer months and therefore the shift parameter for
California avocados should be larger for this time period. In the
analysis accompanying this final rule, the shift parameters for
California avocados in Regions A and B are approximately equal to 0.65,
and the preference parameters for Chilean and Mexican avocados are each
approximately equal to 0.175. Using this pattern of shift parameters,
the empirical model estimates that approximately 60 percent of total
Mexican and Chilean exports will occur during Period 1. This closely
matches historical seasonal export shares for both Mexico and Chile.
Comment: Mexican producers with avocados certified for export to
the United States have a choice to ship to the United States or to a
domestic Mexican market, with the choice of shipping destination based
on where the avocados will return the highest net price to the
producer. Given average wholesale prices in Mexico, I would expect
Mexican producers to continue to ship avocados to the United States
until U.S. prices decreased to that available for domestic shipments. I
do not have the ``break-even'' producer price for Mexican shipments to
the domestic or U.S. export markets, but it could be estimated. I would
expect this price to be significantly less than $0.63 per pound (the
producer price used in the analysis for the proposed rule); perhaps
less than $0.50 per pound. A realistic Mexican farm price for analysis
of the proposed rule change, one that accounts for domestic marketing
opportunities in Mexico, should be estimated and entered into the
model.
Response: Mexico exports only about one-tenth of its production (in
2002, about 94,243 metric tons out of 897,231 metric tons), so we would
generally not expect export prices to have a large effect on Mexico's
domestic prices. Moreover, the export and domestic markets are
different in their production requirements. For exports to the United
States (the destination of half of Mexico's exports in 2003), the
required risk mitigation measures mean higher costs of production--
costs readily borne because of the much higher net returns compared to
domestic sales. We expect that most of Mexico's avocado producers have
limited access to export opportunities because they cannot satisfy the
risk mitigation requirements, and perhaps because of commercial and
infrastructural limitations as well. Mexican growers, however, are
currently exporting to the United States only a fraction of the
avocados they could export from already approved orchards and
municipalities in the State of Michoac[aacute]n, thus the rule does
assume a substantial increase in imported Mexican avocados.
We agree with the commenter that producers with certified fields
will prefer to export to the United States as long as there is an
export premium to be gained. Because the average U.S. wholesale price
of Mexican avocados is substantially less than the wholesale prices of
California and Chilean avocados, we would not expect significant
decreases in the U.S. wholesale and farm prices of Mexican avocados.
Thus, we believe that any
[[Page 69764]]
price premium enjoyed by Mexican producers from exporting to the United
States will be maintained. Because exports are a small share of total
avocado production in Mexico, even a large increase in exports will not
likely substantially affect the Mexican domestic price.
Comment: The analysis assumes that recent price differentials
between California, Chilean, and Mexican Hass avocados will continue. I
believe that a portion of the differential will disappear over time as
Mexican shippers improve their quality of pack and as they establish
relationships with large U.S. buyers.
Response: The model does indicate a reduction in price
differentials with the increase in imports from Mexico, as shown in the
following table of the results of the analysis for the proposed rule.
------------------------------------------------------------------------
Wholesale
prices in Wholesale
the initial prices with
the rule
equilibrium
------------------------------------------------------------------------
Supply region:
California.................................. $1.63 $1.43
Chile....................................... 1.29 1.20
Mexico...................................... 1.14 1.14
Price differences:
California-Chile............................ 0.34 0.23
Chile-Mexico................................ 0.15 0.06
Mexico-California........................... 0.49 0.29
------------------------------------------------------------------------
Comment: We generally concur with the evaluation and offer the
following perspectives for consideration in reviewing the model. We
believe that the increase in consumption of 10.4 percent utilized in
the model may be overly conservative. The introduction of Mexican grown
avocados over a period of 7 years has resulted in an increase of over
400 percent in the consumption of avocados in the approved States. We
believe that the quality of Mexican avocados, coupled with targeted
promotional activities, may likely yield a higher growth in consumption
than is assumed within the model. We believe that the impact of future
imports from Chile will have a greater effect than is being projected
in the model. Chilean avocado growers have invested significantly in
new avocado groves that will increase the volume of fruit exported to
the United States in the near future. Approximately 95 percent of
Chilean avocado exports are destined for the U.S. marketplace, and it
is unlikely that these exports will be reduced regardless of the
opening of the U.S. marketplace to Mexican Hass avocados. We believe
that greater consideration should be given to the impact that the
proposed rule will have on the domestic Mexican avocado market. We
believe that historical domestic consumption rates coupled with the
reaction of prices in the domestic Mexican markets as a result of
decreases in the domestic supply of avocados will have a significant
impact. If Mexican domestic prices increase substantially, it is likely
that the Mexican avocado producers will choose to supply quantities
that are less than those contemplated in the model. A significant
adjustment in volumes from those assumed in the model may have a
considerable impact on the results of the analysis.
Response: We acknowledge the significant growth in consumption of
avocados from Mexico in the approved States, but also note that this is
the region and time period of weakest demand for California avocados.
We do not expect the same market dominance by Mexican imports in the
currently non-approved States, especially during the summer months.
The model shows continued strong participation in the U.S. avocado
market by Chile. Notwithstanding the expected decline in prices earned
and quantities supplied by that country, the model shows that with the
rule, prices and quantities of Chilean avocados will remain higher than
those of Mexican avocados.
The difference between Mexican domestic and export wholesale
avocado prices is significant. Compared to an average wholesale price
during the baseline period in the United