  SEQ CHAPTER \h \r 1 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

WASHINGTON D.C., 20460

MEMORANDUM	Date: May 9, 2006

SUBJECT:	Response to 60-Day Public Comments on the Draft Environmental
Fate and Ecological Risk Assessment Chapter in Support of the
Reregistration Eligibility Decision on Rotenone (PC Code 071003; DP
Barcodes D307382 and D307381 )

FROM:	R. David Jones, Ph.D., Senior Agronomist

		Thomas Steeger, Ph.D., Senior Biologist

		Environmental Risk Branch IV

Environmental Fate and Effects Division (7507C)

		Office of Pesticide Programs		

		

THROUGH:	Elizabeth Behl, Branch Chief

		Environmental Risk Branch IV

		Environmental Fate and Effects Division (7507C)

		Office of Pesticide Programs

		

TO:		Katherine Hall, Chemical Review Manager

		Tom Myers, Risk Manager

		Special Review and Reregistration Division (7508C)

Office of Pesticide Programs

	The Environmental Fate and Effects Division (EFED) has completed its
review of comments (Data Package (DP) Barcodes D307382 and D307381)
received during the 60-day public comment period on the draft
environmental fate and ecological risk assessment chapter for rotenone
(DP Barcode D307390).  A total of twelve comments were received during
the public comment phase (Phase 3) of the reregistration process on
rotenone.  All of the comments posted on the public docket
(EPA-HQ-2005-0494) are listed in Table 1.  Comments from Mr. Harold J.
Singer (EPA-HQ-0494-0026), Mr. Bob Broscheid (EPA-HQ-2005-0494-0027),
Mr. Douglas Stang (EPA-HQ-2005-0494-0028) and from Dr. Brian Finlayson
and Dr. Rosalie Schnick (EPA-HQ-2005-0494-0029) were posted to the
docket after the public comment period had closed on April 11, 2005;
however, responses to these comments are included in this memo.   

Table 1.  Summary of comments received on the draft environmental fate
and ecological risk assessment of rotenone during the 60-day public
comment phase (Phase 3) of reregistration (Rotenone Docket No.: 
EPA-HQ-OPP-2005-0494).

Docket Number  EPA-HQ-2005-0494	Submission

-0018	B. Schau

-0019	A. Campbell (duplicate of -0024)_

-0020	L. W. Ames (Board of California Watershed Network)

-0021 --  -0021.3	J.M. Horvath 

-0022	R. Huber (Wyoming Game and Fish)

-0023	N.A. Erman

-0024 -- -0024.4	A. Campbell

-0025	M. Smith (Missouri Department of Conservation)

-0026	H. Singer (California Regional Water Quality Control Board)

-0027	B. Broscheid (Arizona Game and Fish Department)

-0028 	D. Stang (New York Dept. of Environmental Conservation)

-0029	B. Finlayson and R. Schnick (American Fisheries Society)



	  SEQ CHAPTER \h \r 1 This review paraphrases issues and concerns in
order to efficiently address common concerns.  After comments are
summarized, EFED then provides a response to the comments.  The intent
of EFED’s review is to address issues regarding the underlying
science/data used to estimate potential risk to the environment from the
use of rotenone. Comments regarding human health and worker exposure
issues will be addressed separately by the Health Effects Division;
comments regarding potential mitigation measures will be addressed by
EPA regulatory staff in a separate review.  Many of the comments focused
on policy matters related to the use of piscicides to control fish
populations; however, the decision process used by resource managers to
renovate fish populations is beyond the scope of the environmental fate
and ecological risk assessment and will not be addressed by EFED..  The
ecological risk assessment briefly discusses alternative fish control
measures; mechanical measures to remove fish include primarily
electrofishing, netting and angling; however, mechanical means of
reducing fish numbers have varied widely in their effectiveness
depending on the size of the water body requiring management and the
management objective.  Integrating mechanical measures such as
electrofishing with rotenone treatments has in some cases increased the
overall effectiveness of rotenone by selectively eliminating fish in
isolated refuges where the rotenone may not be effective.

Comment: B. Sachau (EPA-HQ-2005-0494-0018) opposed the use or sale of
rotenone given the chemical’s “harmful effects” and the
uncertainty regarding potential effects on children.  In support of
these comments, information on rotenone from the Pesticide Action
Network (PAN) was included.

EFED Response:  The information taken from PAN on the environmental fate
and effects of rotenone is relatively consistent with that contained in
the environmental fate and ecological risk assessment.  As with most
pesticides, there is a potential for adverse effects to non-target
organisms and the ecological risk assessment attempts to identify the
risks associated with each of the supported uses.  PAN indicated that
the precautionary principle should be applied to all pesticides;
screening-level risk assessments are intended to be conservative
(precautionary) by making use of the most sensitive toxicity endpoints
and the highest estimated environmental concentrations to make point
estimates of potential risks.  

Agricultural uses of rotenone are no longer supported by the technical
registrants and thus, the only remaining supported use of rotenone is as
a piscicide.  The ecological risk assessment makes clear that based on
the highest treatment rate (250 µg/L) and using the most sensitive
toxicity endpoint, most aquatic animals in the targeted treatment area
will likely be killed. The extent to which rotenone may move outside of
the targeted treatment area is intended to be limited by rigorous
standard operating procedures used by resource managers to apply and
contain the chemical.  It should be noted however, that while the
ecological risk assessment made use of maximum treatment concentrations
to estimate risk quotients, applications rates in the field may be
considerably lower and the extent of mortality even within the targeted
treatment area may be significantly less than that indicated in the risk
assessment.

Concerns regarding the potential effects of rotenone on humans will be
addressed by the Health Effects Division. 

  

Comment:  Dr. Ann McCampbell (EPA-HQ-2005-0494-0019) wrote to express
concern that the use of rotenone as a piscicide should be cancelled
since it results in unacceptable water pollution and harm to other
species.  If rotenone is reregistered, Dr. McCampbell asserted that its
use should be limited to contained water bodies on private property and
that no formulations should contain piperonyl butoxide. In support of
her assertions, Dr. McCampbell provided three attachments which deal
primarily with another piscicide, antimycin A.  Dr. McCampbell comments
along with the attachments were also submitted to the docket under
EPA-HQ-0494-0024.1 through 0024.3.

EFED Response:Under the Federal Insecticide, Fungicide and Rodenticide
Act, the decision to (re)register a pesticide is based on whether a
compound causes an unreasonable risk to the environment and human
health.  As part of the decision process, the risks and benefits
associated with the use of a pesticide are considered. The intent of the
environmental fate and ecological risk assessment chapter is to identify
potential risks associated with the registered used of rotenone in order
to inform the decision process.Whether chemicals should be used to
manipulate the environment is a question that extends beyond the scope
of the ecological risk assessment. Additionally, human health-related
issues will be responded to separately by the Health Effects Division.
Since Dr. McCampbell’s comments and submissions are duplicated in the
Federal Docket, EFED provides a more thorough response below under
Docket No. EPA-HQ-0494-0024.

Comment:  Laurel W. Ames, an organizer and founding member of the Board
of the California Watershed Network (EPA-HQ-2005-0494-0020), expressed
concern regarding the unintended consequences of the use of rotenone in
its various formulations on non-target species.  Ms. Ames stated that
the risk assessment must disclose that each poisoning project will
adversely effect aquatic species and “must determine the amount of
extirpation that is caused by the use . . ..”  She asserted that the
assessment should address the problem of toxic residues that linger in
the water and sediments and noted that residues were detected two years
after treatment of Lake Davis; she suggested that applicators are not
following label instructions and that risk assessments should assume
“a high level of non-compliance”.  She also stated that a number of
recent studies have linked rotenone and its formulations to
Parkinson’s disease.  Additionally, Ms. Ames believed that the risk
assessment must compare the risks of using rotenone to well known
non-chemical alternative fish control measures such as electrofishing
and gill netting. Concern was expressed over the lack of a standard
operating procedure that would require pre-application surveys of
aquatic life so that rotenone’s acute and chronic effects on
non-target animals can be better documented.

EFED Response:  It is unclear from Ms. Ames’ comments whether she is
referring to the EFED environmental fate and ecological risk assessment
chapter or whether she is referring to environmental impact studies that
various states may perform to document pre- and post-treatment
populations.  The screening-level risk assessment makes clear that   SEQ
CHAPTER \h \r 1 rotenone is very highly toxic to aquatic animals and
that the direct application of rotenone to freshwater environments as a
piscicide is typically intended to kill all of the fish [and will likely
also kill aquatic invertebrates as well] in the target area.  Exposure
of aquatic organisms outside the treatment area is intended to be
limited through rigorous application standard operating procedures used
by trained fishery professionals. In situations where rotenone is likely
to move out of the intended treatment area, e.g., flowing water
environments, rotenone is typically deactivated with potassium
permanganate to prevent its movement out of the treatment areas. 
Additionally, rotenone is a restricted use pesticide and can only be
applied by certified pesticide applicators who have undergone training
in applying chemicals to aquatic environments.

Screening level risk assessments are intended to estimate risks based on
the labeled use of the chemical.  If would not be possible to evaluate
the effects of the misuse of any pesticide given the potential number of
ways that a chemical can be misused. The presumption underlying the
ecological risk assessment is that applications are being made legally,
according to the label. 

EPA is aware of the studies linking sub-chronic rotenone exposure to
Parkinson’s disease-like symptoms in laboratory rats. The relevancy of
these studies to potential exposure to wildlife from rotenone use in
fishery management is unclear.

The ecological risk assessment of rotenone is not intended to explore
alternative means of removing fish; however, the chapter does briefly
discuss mechanical alternatives to the use of chemicals.  Like most
sampling methods though, there are sampling biases that may limit the
effectiveness of mechanical fish removal methods.  Limitations may
include size selectivity or simple impracticability.

Comment:  In three documents, J. M. Horvath (EPA-HQ-2005-0493-0021.1 to
-0021.3) expressed concern regarding the link between rotenone and
Parkinson’s-like disease particularly related to exposures that may
occur through the agricultural uses of rotenone.  The commenter
recommended that neurotoxicity studies should be required and that these
studies should attempt to address potential age-dependent effects.  The
commenter also provided an annotated bibliography to support their
concerns. 

EFED Response:  These comments are primarily directed at the human
health risk assessment and will be addressed separately by the Health
Effects Division. However, it is important to note that the agricultural
uses of rotenone are no longer being supported by the technical
registrants and that the only remaining supported use of rotenone is
that as a piscicide. The ecological risk assessment discusses the
studies linking rotenone exposure to Parkinson’s disease-like symptoms
in mammals in the laboratory; however, it is uncertain how exposures
similar to those used to obtain these laboratory results would occur
based on the registered use of rotenone as a piscicide. 

Comment:  Mr. Rick Huber with Wyoming Game and Fish submitted comments
(EPA-HQ-2005-0493-0022) emphasizing the critical role that rotenone
plays in allowing Wyoming Game and Fish Department (WGFD) to meet their
species conservation goals and objectives.  Mr. Huber emphasized that
rotenone can only be applied by WGFD employees who hold Department of
Agriculture Commercial Pesticide Applicator  License  specifically
issued for fish control.  The comments included a brief description of
the methods and concentrations used by WGFD to treat streams, ponds and
reservoirs. 

EFED Response:  Comments on behalf of the WGFD are consistent with
EFED’s understanding of how rotenone is used for fishery management
purposes.

Comment:  Nancy A. and Don C. Erman (EPA-HQ-2005-0494-0023) with the
Department of Wildlife, Fish and Conservation Biology at the University
of California (Davis), wrote that technical grade rotenone is rarely
used as a piscicide and that EPA should “recognize and distinguish
among the many formulations” used and that some of the inerts
formulated with rotenone are toxic. Since rotenone is prone to kill a
range of non-target organisms [in addition to fish], the Ermans believed
that rotenone should not be referred to as a piscicide.  They asserted
that macroinvertebrates are sensitive to rotenone and that repeated
applications of the chemical can result in decreased species diversity
and thus potentially affect the entire aquatic food chain. According to
the Ermans, decreased species diversity and displacement of sensitive
species with more tolerant organisms may persist for a number of years
following treatment.

The Ermans also commented that rotenone and co-formulants have been
detected outside of targeted treatment areas and that efforts to
inactive rotenone with potassium permanganate have not been entirely
effective. They also believed that the risk assessment provided an
inadequate review of the toxicity of formulated end-products.

The Ermans commented that the risk assessment provided an insufficient
review of relevant literature linking Parkinson’s disease to rotenone
exposure.

EFED Response:  EFED concurs with the Erman’s comments that rotenone
will kill not only fish but aquatic invertebrates as well.  The use of
the term piscicide though is intended to reflect the labeled use of the
compound to kill fish.  Just as many insecticides may kill other
non-target organisms such as spiders and fish, the chemicals are
registered for the purpose of killing insects and thus are referred to
as insecticides.

The ecological risk assessment of rotenone states that   SEQ CHAPTER \h
\r 1 rotenone is very highly toxic to fish and invertebrates on an acute
exposure basis with median lethal concentration (LC50) values less than
10 µg/L and that   SEQ CHAPTER \h \r 1 the use of rotenone for fishery
management at maximum application rates would likely eliminate both
aquatic vertebrates and invertebrates in the treatment area.  The
chapter states that   SEQ CHAPTER \h \r 1 although the lowest toxicity
value for freshwater invertebrates (48-hr EC50=3.7 µg/L) was chosen for
risk assessment purposes, it is likely that more sensitive invertebrates
could be found in the wild. In this case, at maximum application rates,
acute mortality of aquatic invertebrates would be expected. Despite the
fact that invertebrates are less conspicuous members of the aquatic
community, they are a major component of aquatic ecosystems and food
webs. Any significant effects on invertebrates would most likely
influence other components of the ecosystem. Effects may not be limited
to merely a change in total biomass as a result of widespread mortality
but any changes associated with differential sensitivity could bring
about significant changes in the community structure, which could alter
system function.

 

  SEQ CHAPTER \h \r 1 The extent to which inerts, contaminants and other
active ingredients affect the toxicity of rotenone is not well known;
however, toxicity testing with formulated end-products suggests that in
general, co-formulants do not substantially affect the toxicity of
rotenone based on a species sensitivity distribution of fish acute 96-hr
LC50 values.  It is assumed that the distribution of fish species
sensitivities observed from laboratory tests represent the distribution
of sensitivities that are likely to be encountered in the environment.
This assumption may not be warranted though since laboratory test
species are not selected based on their sensitivity to chemicals but
rather are selected based on their ability to thrive under laboratory
conditions.

EFED concurs with the comment that not all of the literature linking
sub-chronic rotenone exposure to Parkinson’s disease-like symptoms in
laboratory rats was reviewed; however, the exposure methods (route and
duration) used to produce these results are not considered likely to
occur through the remaining registered use of rotenone as a piscicide.  
The Agency is not aware of any literature where environmentally relevant
concentrations of rotenone with similar exposure pathways that may occur
through the registered use of rotenone results in Parkinson’s
disease-like lesions.  If the Ermans are familiar with any such studies,
they are encouraged to submit these studies for review and possible
inclusion in the ecological risk assessment.  Additionally, given that
most aquatic animals in the target area are killed by rotenone, whether
the compound is capable of causing neurological effects may be academic.
 Potential worker exposure risks from rotenone are more thoroughly
discussed in the Health Effects Division’s chapter on human health
risks.

The environmental fate assessment of rotenone makes clear that the
chemical’s persistence depends on environmental conditions at the time
of application.  In warm alkaline waters, rotenone is likely to degrade
rapidly; however, in cold, neutral or acidic water, the chemical is more
likely to persist.  The risk assessment discusses the fact that residues
have been detected for some time following application and that the
chemical may move beyond the targeted treatment area. However, the
chapter also states that   SEQ CHAPTER \h \r 1 exposure of aquatic
organisms outside the intended treatment area is intended to be limited
through rigorous application standard operating procedures used by
trained fishery professionals.  Additionally, in lotic (flowing water)
environments, rotenone is typically deactivated with potassium
permanganate to prevent its movement out of the treatment areas.  The
efficacy of potassium permanganate in deactivating rotenone is likely
highly variable; the degradation kinetics of rotenone due to
permanganate treatment are poorly documented. Hhowever, it is also
likely that resource management agencies make every effort to confine
the compound to the targeted treatment area.

EFED concurs with the comment that the toxicity of all formulated
end-products was not evaluated in the chapter.  Only a limited amount of
toxicity data was available on formulated end-products.  However, the
data that were available suggest that technical grade rotenone is more
toxic than formulated end-products and that the chemicals co-formulated
with rotenone do not substantially affect the toxicity of rotenone. The
process used in the rotenone chapter for evaluating the toxicity of
formulated end-products is consistent with that described in the
document entitled “Overview of the Ecological Risk Assessment Process
in the Office of Pesticide Programs”.  With respect to inerts though,
the Agency is in the process of reviewing the potential ecological
effects of chemicals that are commonly co-formulated with pesticides.  

The ecological risk assessment of rotenone states that aquatic
macroinvertebrates exhibit roughly similar sensitivity to rotenone as do
fish, that it is likely that most if not all fish and macroinvertebrates
will be killed in the targeted treatment area, and that the entire
aquatic food chain can be affected. The expectation is that treated
streams/lakes will repopulate through immigration and/or restocking. 
Whether species density/richness is identical to pretreatment conditions
is uncertain; however, EFED concurs with the Ermans that it is possible
that more tolerant species can potentially displace those less tolerant
to rotenone if rotenone is repeatedly applied.  However, the logistics
of conducting a rotenone treatment are typically relatively complicated
and involve considerable resources. While a national Standard Operating
Procedure for rotenone use in fishery management has not been developed,
most resource managers are highly trained and would likely attempt to
limit the need for re-treatment and the extent to which rotenone would
extend beyond the desired treatment area as this could limit the extent
to which any aquatic population could recover. EFED has recommended that
a Standard Operating Procedure be developed by the registrants to
accompany the product.

Whether chemical means of manipulation should be used over other
mechanical control measures or to what extent other species should be
sacrificed to aid in the recovery of endangered species are important
questions which the Ermans raise; however, the answers involve policy
issues and are beyond the scope of screening-level risk assessment.

The risk assessment has been revised to include data recently submitted
by the California Department of Game and Fish.  While these data
indicate that rotenone was detected downstream of targeted treatment
areas in spite of efforts to detoxify the compound with potassium
permanganate, the concentrations were at or near the limit of detection
for rotenone.  Additionally, although the data are highly variable, they
indicate that at the treatment concentrations used, i.e., 2.5 – 25 µg
active ingredient/L, benthic macroinvertebrates were not eliminated
after several consecutive years of rotenone application.  Sampling
conducted one week after treatments shows that while macroinvertebrate
abundance and diversity were reduced [in some cases], the measurement
indices did not appear to have been significantly diminished based on
the limited statistics that were provided in the reports. Overall
abundance of benthic macroinvertebrates appeared to have recovered in
the treated stream reaches; however, there is uncertainty regarding the
extent to which species diversity may be affected.  Whether the
post-treatment benthic invertebrate sampling results are due to
immigration/relocation from upstream reaches or whether the
macroinvertebrates in the treatment area were less sensitive to the
rotenone concentrations used during the consecutive treatments is
uncertain. 

Comment:  Dr. Ann McCampbell (EPA-HQ-2005-0494-0024.1) submitted her
public testimony from a New Mexico Water Quality Control Commission
hearing on a petition to use piscicides in Rio Costilla drainage; Dr.
McCampbell objected to the use of chemicals to restore native fish
populations in New Mexico.

EFED Response:  None of the comments provided by Dr. McCampbell are
specific to the environmental fate and ecological risk assessment of
rotenone.  Although Dr. McCampbell states that EPA “will register a
product [pesticide] if it causes unreasonable harm if it chooses to
without telling anyone that’s what it did”, she does not provide any
information regarding either the environmental fate or toxicity of
rotenone to support her contention that rotenone use will result in
unreasonable harm.  The decision to (re)register a pesticide under the
Federal Insecticide, Fungicide and Rodenticide Act takes into account
risks and benefits; however, ecological risks is one factor considered
in that decision.  If Dr. McCampbell has data to support her statements,
she should submit the data to EPA for consideration.  Additionally, all
actions taken on the (re)registration of a pesticide are included in
public record and can be accessed under the Freedom of Information Act. 
The status of the reregistration of rotenone can be accessed on the
internet at    HYPERLINK
"http://www.epa.gov/oppsrrd1/reregistration/rotenone/" 
http://www.epa.gov/oppsrrd1/reregistration/rotenone/  .

 

Comment:  Dr. Ann McCampbell (EPA-HQ-2005-0494-0024.2) submitted a fact
sheet on antimycin A (PC Code 006314) in which she inserts comments
regarding how piscicides in general can have adverse effects on aquatic
communities.  In a brief section that is related to rotenone
specifically, Dr. McCampbell asserts that rotenone has a mechanism of
action similar to antimycin A and that rotenone exposure has been
related to Parkinson Disease-like symptoms in laboratory animals.  Dr.
McCampbell mentions that several organic solvents are used in formulated
end-products of rotenone and that these co-formulants are considered
toxic in their own right. She also raises concerns regarding the
potential effects of rotenone on plants.

EFED Response:  Although Dr. McCampbell raises several issues related to
whether a state elects to use piscicides to control fish populations
these comments are not related to the environmental fate and ecological
risk assessment of rotenone but are rather state policy issues.  In
sections of Dr. McCampbell’s comments that do address rotenone, Dr.
McCampbell is correct regarding the presence of organic solvents in
liquid formulations of rotenone.  The ecological risk assessment of
rotenone discusses this fact; however, for some constituents of the
liquid end-products, e.g. toluene, application rates are expected to
result in concentrations that are below established human health
criteria.  With respect to risk associated with formulated end-products,
the EFED chapter states that “  SEQ CHAPTER \h \r 1 [T]oxicity` data
are not available on all of the formulated products of rotenone for all
of the surrogate species typically evaluated.  However, based on
toxicity data collected on both technical grade rotenone (>95% active
ingredient) and formulated end-product, the technical grade active
ingredient is generally more toxic than formulated end-product
[corrected for active ingredient] by at least a factor of two.  These
data suggest that for the formulated products tested and the toxicity
endpoints measured, the inerts do not contribute substantially to the
toxicity of the active ingredient. These data also suggest that the
similarly structured rotenolones of plant resins (cube root resins)
contained to varying amounts in formulated end-products also do not
contribute substantially to the toxicity of rotenone.  The extent to
which the toxicity of untested formulations would be similar cannot be
determined from the currently available data.

To the extent that the surrogate species tested represent the range of
sensitivities of aquatic organisms, the ecological risk assessment
states that within the treatment area, most aquatic vertebrates
(including aquatic phase amphibians) and invertebrates are likely to be
killed at treatment concentrations used to restore native fish
populations. The chapter also goes on to say that  exposure of aquatic
organisms outside the intended treatment area is limited through
rigorous application standard operating procedures used by trained
fishery professionals and that in most lotic (flowing water)
environments, rotenone is typically deactivated with potassium
permanganate to prevent its movement out of the treatment areas.

With respect to the potential effects of rotenone on plants, the EFED
chapter states that   SEQ CHAPTER \h \r 1 no data are available to
evaluate the toxicity of rotenone to terrestrial plants. Although
rotenone is isolated from plants and has been routinely used as an
insecticide on plants, no information is available to evaluate potential
effects on plants.

  SEQ CHAPTER \h \r 1 Although several studies have linked sub-chronic
rotenone exposure to Parkinson’s disease-like symptoms in laboratory
rats, the exposure methods (route and duration) used to obtain these
results are not typically encountered through the current registered
uses of rotenone.  The relevancy of these studies to potential exposure
to rotenone from its registered use in fishery management is uncertain. 

 Comment:  Also included in the comments submitted by Dr. McCampbell is
an article (“Purity and the Rio Grande Cutthroat”) written by M. H.
Dutch Solomon (EPA-HQ-2005-0494-0024.3).  In this article written for
Country Sports, Mr. Solomon focuses on the use of antimycin to restore
native fish populations and the rationale to support this effort, i.e.,
restore genetic purity.

EFED Response.  None of the comments refer specifically to the
environmental fate and ecological risk assessment of rotenone.  The
extent to which chemical means are or should be used to restore native
fish populations is a policy issue and is not discussed in the
ecological risk assessment

Comment:  Michael S. Smith, Policy Coordinator for the Missouri
Department of Conservation (MDC; EPA-HQ-2005-0494-0025) wrote that the
MDA supports the continued availability of rotenone to control unwanted
fish in small impoundments and fish culture ponds.  Mr. Smith indicated
that all MDC personnel who apply rotenone are trained and licensed as
certified pesticide applicators by the Missouri Department of
Agriculture.

EFED Response:  Comments on behalf of the MDC are consistent with
EFED’s understanding of how rotenone is used for fishery management
purposes.

Comment:  Mr. Harold J. Singer (EPA-HQ-2005-0494-0026), Executive
Officer of the California Regional Water Quality Control Board Lahontan
Region, wrote to request that the EPA consider the water quality risks
associated with rotenone applications and requested that EPA impose
conditions/requirements on applicators to limit unintended impacts of
rotenone “formulation residues” on non-target organisms including
humans.  Mr. Singer goes on to list rotenone-specific and potassium
permanganate prohibitions and conditions for their use in the Lahontan
Basin and cites circumstances where past applications have
“violated” permit-prescribed water quality objectives.  Plans to use
rotenone have had to be abandoned because of a lack of certainty that
some potentially affected organisms may actually be relatively rare and
that rotenone residues may persist under certain environmental
conditions. Additionally, there is concern that rotenone may have
long-term effects on community structure, that rotenone formulations may
have differential toxicity, that both chemical and non-chemical
alternative exist, and that there are potential health risks to humans. 

EFED Response:  Mr. Singer raises valid concerns regarding the use of
rotenone where his experience has been that permit-prescribed conditions
have not been met. However, none of his concerns are contrary to what is
discussed in the ecological risk assessment of rotenone.  The chapter
indicates that rotenone can persist under certain conditions and that
the chemical can be equally toxic to both fish and aquatic invertebrates
particularly at maximum treatment rates. It is EFED’s
understanding(although a national Standard Operating Procedures have not
been developed) though that resource management agencies go to
considerable lengths to limit the extent of non-target mortality by
adhering to relatively rigorous application procedures. Based on Mr.
Singer’s comments, all of the violations that occurred though were
unintentional and were likely a result of the inability of anyone to
completely control the field environment. EFED encourages the
development of standard operating procedures that can effectively limit
non-target animal exposure

As stated previously, the ecological risk assessment has been revised to
include data recently submitted by the California Department of Game and
Fish (discussed in greater detail below).These data substantiate Mr.
Singer’s comments that rotenone has been detected downstream of
targeted treatment areas in spite of efforts to detoxify the compound
with potassium permanganate; however, concentrations were at or near the
limit of detection for rotenone. The water and sediment monitoring data
suggest though that potassium permanganate treatment was relatively
effective at limiting the movement of rotenone outside of the treatment
area.

Whether chemical or non-chemical means of control are used to meet
management objectives is a policy decision.  Also, the extent to which
pre- and post-monitoring of aquatic communities is required is not
within the scope of a national screening-level ecological risk
assessment. However, EFED has recommended the development of a
relatively comprehensive Standard Operating Procedure and that pre- and
post-treatment monitoring of non-target aquatic animals could be a
component of the procedure. 

Comment:  Mr. Bob Broscheid (EPA-HQ-2005-0494-0027), Habitat Branch
Chief with the State of Arizona Game and Fish Department (AGF), wrote to
emphasize the importance of rotenone as a management tool to implement
fish assemblage changes to all the state to meet management goals. 
Without rotenone as a management tool, he believed that recovery of
Federally-listed fish would not be possible in Arizona or nationally and
that effective management of recreational fisheries [in general] would
be jeopardized.   In his comments Mr. Brosheid provided examples of how
rotenone is used by the State of Arizona, and he described his
department’s efforts to engage stakeholders and limit human exposure
by rigorous application procedures.  The potential effects from nuisance
invasive species on aquatic communities is a concern for the State of
Arizona and rotenone offers a cost-effective means of control on large
bodies of water and streams with high flows.

EFED Response:  Comments on behalf of the AGF are consistent with
EFED’s understanding of how rotenone is used for fishery management
purposes.

Comment:  Mr. Douglas Stang, Chief of the New York State Department of
Environmental Conservation Bureau of Fisheries, wrote to emphasize New
York State’s reliance on rotenone to restore fish communities degraded
by introduced fish species and that the chemical has proven to be a
“safe and effective fish management tool ... essential to what has
become one of the premier fish restoration programs in the country.”

Mr. Stang expressed concern that the ecological risk assessment suggests
that all aquatic organisms will be eliminated during rotenone
treatments; however, in New York where treatments are typically at 1
μg/L, invertebrate and [aquatic phase] amphibian mortality it generally
low.  Mr. Stang believed that the risks to aquatic invertebrates are
overstated in the risk assessment and do not reflect the scientific
literature.

1 μg/L) considerably below the maximum label rate and has observed less
mortality of aquatic invertebrates and aquatic phase amphibians than may
occur at the maximum rate.  As evidenced by benthic macroinvertebrate
monitoring data collected by the California Department of Game and Fish,
lower treatment concentrations do not appear to substantially reduce
benthic macroinvertebrate abundance [one week after treatment] relative
to pretreatment conditions.

w as 0.1 μg/L have been reported used and that the reduced rates will
reduce the extent of mortality inflicted on the aquatic community.  The
effect of   reduced application rates on aquatic community structure is
uncertain though as the acute risk level of concern (risk quotient
≥0.5) would still be exceeded for both aquatic vertebrates and
invertebrates. 

and that the maximum application rate for use as a piscicide should be
200 μg/L, the solubility limit of rotenone in water at 20oC.

AFS asserted that concentrations and durations of rotenone exposure
during most stream treatments allows for the survival of
macroinvertebrates. AFS also stated that while the efficacy of the
rotenone treatment on Lake Davis, CA, [discussed in the ecological risk
assessment] will remain uncertain, they questioned the relevancy of
“EPA’s subjective assessment of this rotenone treatment” to the
reregistration of rotenone.

AFS had commented previously on behalf of the technical registrants
during the 30-day error correction phase (Phase I). The comments
essentially echoed those of the technical registrant Prentiss. 

EFED Response:  As discussed previously in this response document and in
the ecological risk assessment as well, EFED is uncertain regarding the
ecological relevancy of the studies linking rotenone exposure to
Parkinson disease-like symptoms in mammals. However, the
H慥瑬⁨晅敦瑣⁳楄楶楳湯眠汩⁬牰癯摩⁥⁡敳慰慲整
爠獥潰獮⁥湯琠敨爠汥癥湡祣漠⁦桴獥⁥瑳摵敩⁳潴栠
浵湡栠慥瑬⹨ഠ汁桴畯桧䄠卆栠獡爠灥慥整汤⁹獡敳瑲
摥琠慨⁴污⁬潲整潮敮愠灰楬慣楴湯⁳牡⁥潣摮捵整⁤
獵湩⁧⁡汣獯摥猠獹整⁭湡⁤桴瑡琠敨洠硡浩浵愠灰楬
慣楴湯爠瑡⁥㈨〵 μg/L) should be reduced to reflect the
solubility limit of rotenone (200 μg/L), it is OPP policy to evaluate
maximum label application rates in screening-level risk assessments.  If
the technical registrants are willing to reduce the maximum label rate
to 200 μg/L, then the ecological risk assessment will be revised to
reflect that change. Additionally, one technical registrant has
specifically requested to retain the 250 μg/L application since it is
useful in eradicating carp in lakes with high organic matter. However,
the ecological risk assessment has been revised to evaluate potential
risks at application rates of 250, 200 and 50 μg/L.

With respect to the discussion of the rotenone treatment of Lake Davis,
EFED is required to evaluate all of the available data on incidents
reported under FIFRA 6(a)2 adverse effect reporting requirements.  The
rotenone treatment of Lake Davis is contained in the Ecological Incident
Information System (EIIS) database as several thousand fish were
reported killed outside of the targeted treatment area.  Additionally,
both inerts (e.g., naphthalene and toluene) and a manufacturing
contaminant (i.e., trichloroethylene) were contentious issues that the
agency was notified about at the time of the incident and have continued
to raise public concerns relative to the use of rotenone. 

Comments received from AFS during Phase I of reregistration were not
responded to directly since Phase I is reserved exclusively for
technical registrant comments. However, since these comments were nearly
identical to those of Prentiss, EFED’s response would also be
identical.  The reader is therefore referred to EFED’s response to
Phase 1 comments (DP Barcode D307390).

In the following sections, EFED has reviewed the four studies included
with the AFS submission.  In general, the studies provide useful
information on water/sediment concentrations of rotenone and its major
degradate (rotenolone) following application of rotenone to alpine
streams. Additionally, there is useful information on benthic
macroinvertebrate monitoring in treated streams. The data indicate that
under the conditions studied, rotenone was relatively well confined to
the targeted treatment area through the use of potassium permanganate
deactivation and in circumstances where residues were detected outside
of the treatment area, rotenone concentrations were near the minimum
limit of detection.  At the treatment concentrations used, rotenone did
not kill all of the benthic macroinvertebrates; although, invertebrate
abundance and diversity were affected, indices appeared to recover over
time.  The ecological risk assessment of rotenone has been revised to
include some of these data qualitatively in the ecological risk
characterization discussion.

California Department of Game and Fish Water and Sediment Monitoring
Study:  Silver Creek.

Water and sediment monitoring data (Pesticide Laboratory Report No
P-1839; dated 11/12/95) were collected by the State of California
Department of Game and Fish from rotenone treatments of Silver Creek
(Mono County, CA) before, during and after application of NusynNoxfish®
(2.5% active ingredient [a.i.]) in 1996.  Silver Creek (water
temperature:  7 – 12oC; pH:  7.4 – 7.7; alkalinity:  10 – 25 mg/L
as CaCO3) and its minor tributaries were treated at a target
concentration of 1 mg formulated product/L (25 μg a.i./L) using a
combination of drip stations and hand spraying for 4 hours.  A potassium
permanganate (KMnO4) drip station was established downstream of the
treatment area and was run continuously for 28 hours.  Rotenone
concentrations 2-miles upstream of the detoxification station peaked at
roughly 50 μg/L  after 6 hours; rotenolone concentrations peaked at the
same time with concentrations of roughly 56  μg/L/.  In a 1-mile
transect of the treatment area during the last hour of treatment total
residues (rotenone plus rotenolone) concentrations ranged from roughly
20 μg/L to 150 μg/L; however, in general, half of the water samples
were between 60 and 80 μg/L.  Measurable residues (minimum detectable
level (LOD)≥2 μg/L) of rotenone were detected immediately upstream of
the detoxification station for 18 hours after application.  Measurable
residues (LOD≥2 μg/L) of rotenolone were detected for roughly 24
hours just upstream of the detoxification station.  Rotenone residues at
the limit of detection were reported in the water column a 15-minute
[walking] distance downstream of the detoxification site 2 hours
post-treatment; however, neither rotenone nor rotenolone were detected
in the water or sediment downstream of the detoxification station from 1
to 7 days post-treatment.  At no time were two of the inerts (methyl
naphthalene and xylene) or the manufacturing contaminant
trichloroethylene detected just above or downstream of the
detoxification station; however, naphthalene (LOD≥0.5 μg/L) was
detected (1.7 to 5.2 µg/L) immediately upstream of the detoxification
station roughly 5 hours post-treatment.  No naphthalene residues were
detected downstream of the detoxification station during the sampling
period.  It is not clear why the report is dated (11/12/95) prior to
when samples were received (9/96).

California Department of Game and Fish Water and Sediment Monitoring
Study: Silver King Creek

Additional monitoring data were provided (Pesticide Laboratory Report
No. P-1638; dated 2/1/94) from an earlier (1993) rotenone treatment of
Silver King Creek (water temperature:  1 – 10oC; pH:  7.19 – 7.78;
alkalinity:  16 – 26 mg/L as CaCO3) and its tributaries above
Llewellyn Falls (Alpine County, CA).  Rotenone (NusynNoxfish®; 2.5%
active ingredient) was applied at a target concentration of 0.5 mg/L
(12.5 μg a.i./L) using a combination of drip stations and hand
spraying.  A second application was conducted the following day using
drip stations alone on a downstream portion of the drainage and on a
tributary (Fourmile Canyon Creek) to Silver King Creek. A detoxification
station using KMnO4 was located just upstream of Llewellyn Falls and was
run for 10 hours during the first application with target treatment
concentrations ranging from 1 to 3 mg/L.During the second treatment,
KMnO4 concentrations were maintained at roughly 1 mg/L.  Rotenone
concentrations during the first treatment ranged from 3.1 to 20 µg/L
(mean=10.6 µg/L); during the second treatment, rotenone concentrations
averaged 11.8 µg/L.  Rotenone residues were detected a 30-minute travel
time downstream of the detoxification station on three sampling
occasions; however, residues (range:  2.2 – 4.0 µg/L) were close to
the detection limit of 2 µg/L.  None of the sediment samples collected
either 20-ft up-stream or 30-min travel time downstream of the
detoxification station had measurable residues of either rotenone or
rotenolone. No rotenone or rotenolone were detected in any of the
sediment samples collected above or below the detoxification station.
Water samples were analyzed for volatile and semi-volatile organic
compounds and revealed xylene (0.56 – 1.3 µg/L), ethyl benzene (0.9
µg/L), 1, 2, 4-trimethylbenzene (0.9 µg/L) and trichloroethylene (0.53
- 0.76 µg/L); these detections were limited in number and were close to
the limit of detection (0.5 µg/L for each of the compounds). 
Naphthalene and methyl naphthalene concentrations in the treatment area
ranged from 0.9 to 52µg/L and 30 to 50 µg/L, respectively; naphthalene
and methyl naphthalene concentrations as high as 36.2 and 40 µg/L,
respectively, were measured below the detoxification station during the
first treatment.  According to the report, rotenone concentrations along
with any of the measured constituents of the formulated product in the
treatment area were below detectable levels by 16-hours post-treatment. 


Although rotenone residues were detected downstream of the
detoxification station, the report concluded that “these detections
were near the analytical detection limit of 2.0 µg/L and were below
levels know to cause toxicity . . . and therefore do not represent
biologically significant concentrations” as evidenced by the lack of
[fish] mortality (wild or caged) below the treatment area during or
after application of rotenone.  Additionally, the detection of
naphthalene and methyl naphthalene was attributed by the report author
to the low water temperature (1 – 10oC) that was believed to have
reduced the volatility of these compounds.  The report indicated that
lower water temperatures reduced the efficacy of rotenone while
increasing its persistence in water in spite of efforts to decontaminate
the compound with potassium permanganate.

California Department of Game and Fish Benthic Macroinvertebrate
Monitoring Study:  Silver King Creek

Also included with the AFS submission is a report by the State of
California Department of Fish and Game entitled “Impacts of Rotenone
on Benthic Macroinvertebrate Populations in Silver King Creek, 1990
through 1996” discussing three consecutive years (1991 – 1993) of
rotenone applications to remove hybridized rainbow trout (Oncorhynchus
mykiss) and Lahonton cutthroat trout (O. clarkii henshawi) so that
native Paiute cutthroat trout (O. clarkia seleniris) could be
reintroduced.  Monitoring of macroinvertebrate populations was conducted
“the year before treatment (1990), before and after treatment during
each treatment year (1991 – 1993) and for three consecutive years
following the treatments (1994 – 1996).”  Macroinvertebrate biomass,
total number, total taxa, percent dominant taxa and total number of
stonefly (Plecoptera spp.) taxa were determined to quantify abundance
and diversity, overall stream health, and potential impacts to
Plecopterans that were considered a vulnerable indicator species for
gauging the environmental health of aquatic ecosystems.  Additionally,
the Biotic Condition Index (BCI), the Ephemeroptera, Plecoptera and
Trichoptera index (EPT) and the dominance and taxa diversity index (DAT)
were calculated.

Reference sites were located either on upstream portions of Silver King
Creek or on tributaries (Fourmile Canyon Creek) to the creek. 
Invertebrate sampling using a modified Surber (1 ft2) sampler was
conducted 1 to 4 weeks prior to treatment and approximately 1 week after
treatment; in non-treatment years, sampling was conducted once per year.
  

μg a.i./L.  Total biomass was identical between reference and treatment
sites during pretreatment sampling in 1990. During the first two
treatment years, total biomass declined in both the reference and
treatment sites; however, in 1993 (considered anomalous by the authors),
the reference site increased in biomass by 18% while the treatment site
declined by 15%.  By the third year of post-treatment sampling mean
biomass for reference and treated sites had increased to 2 g/m2 and 1.4
g/m2, respectively.  Based on Figure 2 in the report showing total
biomass though, it appears that across all sampling years other than
1995, the total biomass of invertebrates at reference sites either
equaled or exceeded the total biomass at rotenone-treated sites.  

Pretreatment mean total number of macroinvertebrates in both the
reference (9,971) and treated (9,572) sites declined over the next three
treatment years; post-treatment years exhibited marked fluctuations in
the mean total number of macroinvertebrates in both reference and
treatment sites.   However, based on the report graph of total number of
macroinvertebrates (Figure 3), except for sampling results in two year
(pre 1991 and 1995), the total number of macroinvertebrates at reference
sites exceeded that at rotenone-treated sites.  

BCI values for both reference and treated sites were classified as
“excellent” during the pretreatment year; although both sites
experienced declines in BCI over the treatment period (1991 – 1993),
declines in BCI values were consistently greater at treated sites.  Mean
BCI values at the reference sites eventually surpassed the pretreatment
level by the final year of sampling; however, treated sites had mean BCI
values 11.2% less than the pretreatment mean. 

Mean number of taxa at reference (32) and treated (37) both experienced
declines during the treatment years; however, according to the report,
the declines observed at treated sites were consistently greater (by a
factor of 2) than at reference sites in 1991 and 1992.  In 1993, both
the reference and treated sites experienced roughly similar declines. 
By the end of the study, the mean number of taxa for reference and
treated sites were 41.9% and 12.7%, respectively, higher than
pre-treatment values.

Mean percent dominant taxa values increased by 46% relative to
pretreatment values at treated sites and decreased by 13% at reference
sites in 1991, increased at both sites by relatively similar percentages
in 1992 and only the treated site increased again in 1993.  Although
pretreatment values for both reference and treated sites were around
20%, each had increased to between 65 and 72%, respectively, for the
reference and treated sites by the end of the study.

The mean DAT index values were considered “excellent” for reference
and treated sites during pretreatment sampling.  During the treatment
years, index values declined for both sites; however, during 1993,
rotenone-treated sites increased by 45% while the reference site
declined by 25%.  By the end of the study, the reference and treated
sites exceeded pretreatment DAT values by 52 and 13.6%, respectively.

The EPT index during the first year of treatment declined by 76% of the
pretreatment value while the reference site declined by 25%, while in
the second year, the pattern was reversed with treated sites showing a
29% decline and reference sites showing  a 58% decline.  In 1993, the
reference and treated sites had mean declines of 27% and 49%,
respectively.  According to the report, during the post-treatment
sampling period, mean reference site values were consistently lower than
treatment site values by 27.5 to 181%.  In comparison to pretreatment
(1990) values, the final (1996) EPT index mean had increased from 36 to
46 at reference sites and had increased from 52 to 59 at treated sites.

The total number of stonefly taxa at control (7) and treated (8) sites
were roughly similar in 1990; in 1991, the reference and treated sites
had declines of 33 and 44%, respectively.  By the second year of
treatment, the reference site was unchanged but the treated site had
decreased by 42% and by 1993, the reference and treated sites had
declines of 29 and 24%, respectively.  At the end of the sampling
period, reference and treated sites had 10 and 6 taxa, respectively,
representing roughly a 43% increase for the reference site and a 27%
decrease for treated sites.  Additionally, the total number of
stoneflies decreased following each treatment. In 1991, the number of
stoneflies at reference sites had increased by 83% while at treatment
sites, they had decreased by roughly 49%; in 1992, the total number
decreased by 16% at reference sites and by 52% at treated sites; in
1993, there was a 40% decrease in the number at reference sites while
there was a 58% decrease at treated sites.  By the end of the study, the
mean number of stoneflies at reference sites had increased by 215% of
the pretreatment value while the treated site mean had increased by 31%.
 

The report concluded that a comparison of measured concentrations of
rotenone in the water with toxicity values for aquatic invertebrates
would have “predicted little, if any, impact to aquatic invertebrates
in Silver King Creek and that the results of the study were consistent
with this prediction.  Further, the report concluded that while there
may be short-term impacts to the total number of taxa, the percent
dominant taxa, the EPT index and the total number of stonefly taxa,
impacts to abundance metrics (biomass and total number of invertebrates)
were not indicated. 

μg a.i./L)  to rotenone than rainbow trout (96-hr LC50=2.3 μg a.i./L),
the EFED ecological risk assessment presents data indicating that
freshwater invertebrates can exhibit roughly similar sensitivity to
rotenone as freshwater fish. It would be presumptuous to think that
laboratory test species are representative of the full range of
sensitivities that may exist in the wild and it is likely that more
sensitive species do exist.  The report does indicate that cladocerans,
copepods and other planktonic microcrustaceans can be sensitive to
rotenone; however, the authors suggest that these invertebrates inhabit
lakes. The marked declines in stonefly numbers at treated stream sites
suggest that they were also relatively sensitive to rotenone and that
while the number of stoneflies recovered to pretreatment levels, the
total number observed at rotenone treated sites was not as great as that
at reference sites. 

As with many field studies, the current study exhibits a considerable
amount of variability that makes it difficult to interpret what may be
treatment-related effects.  Whether changes observed in aquatic macro
and microinvertebrate populations within treated systems are due to the
direct effects of rotenone, a secondary effect due to reductions in the
number of predators (fish and zooplankton), or whether the effects are
do to some unrelated environmental factor is uncertain. However, in
spite of the variability, there appear to be some treatment related
effects on macroinvertebrate populations. Given the species abundance
and diversity upstream of the treated sites though, it is reasonable to
anticipate that the treated sites could recover to pretreatment
conditions through immigration/drift over time.  In spite of this
opportunity to immigrate/relocate into treated areas, the decreased
Biotic Condition Index appeared to be a treatment-related effect that
did not recover during the study period.  Potential impact trends were
also noted for the total number of taxa, the percent dominant taxa, the
EPT index and the total number of  stonefly taxa; however, the
significance of these effects could not be established in the current
study. 

The report discusses long-term effects on species diversity observed
from the rotenone-treatment of Strawberry Reservoir (Utah) and points
out the treatment concentrations used on Silver King Creek were lower
and that less of the water-shed was treated.  As stated previously, the
extent to which treated waters can be repopulated by immigration and/or
restocking will likely determine the extent to which sensitive species
will recover.

California Department of Game and Fish Benthic Macroinvertebrate
Monitoring Study:  Silver Creek

μg a.i./L).

According to the report, benthic macroinvertebrates were monitored
before and after treatment (1994 – 1996) and for two consecutive years
following treatments (1997 and 1998).  The same metrics as those
discussed above for Silver King Creek were measured on Silver Creek. 
Invertebrate sampling using a modified Surber (1 ft2) sampler was
conducted 1 to 4 weeks prior to treatment and approximately 1 week after
treatment; in non-treatment years, sampling was conducted once per year.
 The major difference between the two studies was that reference sites
were identified for the study on Silver King Creek; no reference sites
were identified for the Silver Creek study.

Over the three-year treatment period, 24-hr mean rotenone concentrations
were 3.5 (1994), 17.9 (1995) and 10.4 μg a.i./L (1996). Compared to
pretreatment values, total biomass increased by 6.2 and 20% following
the first two rotenone treatments; however, total biomass declined by
23.5% after the third rotenone treatment.  Only a single post-treatment
(1997) sample was collected and mean total biomass was 37.5% higher than
the pretreatment value.

The total number of macroinvertebrates increased by roughly 39 and 42%
following treatments in 1994 and 1995, respectively; however, in 1996
total biomass decreased by 54%.  During post-treatment sampling, mean
values were 11.5% and 32.3% lower than the pretreatment value.

Prior to treatment, the mean BCI (81.5) was categorized as “good” by
the authors.  Following each of the treatments, the BCI declined by
8.3%, 4.2% and 7.6% in 1994, 1995, and 1996, respectively.  During the
post-treatment sampling, the BCI in 1997 and 1998 exceeded the
pretreatment mean by 1.2% and 6.1%, respectively.

 

há

há

há

Ȁ਀ 

7

?

G

H

d

e

g

h

˜

³

¶

·

Â

Æ

Ç

Û

Ü

Ý

ã

ï

ú



&

=

N

S

Z

]

há

há

']

d

e

p

€

Ç

há

"g

@

@

há

@

@

@

@

@

 h&

 h&

 h&

h×.

h

h

h

 h

h

 h

h/

h

 h

h

h/

‚

ƒ

%by roughly 3% in 1997 over the pretreatment conditions; however, in
1998 the percent dominant taxa had declined by 54% of the pretreatment
value.

The DAT index was 15.7 prior to the use of rotenone and decreased by
7.9%, 16.1% and 1.1% in 1994, 1995, and 1996, respectively, following
application of rotenone.  During the post-treatment sampling, the mean
value in 1997 was 14.1; the metric was not determined for 1998.

The mean EPT index was 43.1 prior to use of rotenone and decreased by
51.4%, 45.1% and 32.3% of the pretreatment mean following application of
rotenone in 1994, 1995 and 1996, respectively.  During the
post-treatment sampling the mean values exceeded the pretreatment value
by 40% in 1997 and by 56% in 1998.   

The mean total number of stonefly taxa in pretreatment sampling was 6.5
and decreased by 23% in 1994,  14% in 1995 and 38% in 1996.   In 1997,
the mean number of stoneflies was 12% lower than the pretreatment value;
however, in 1998, the mean number was 3% higher than the pretreatment
value. 

 The mean total number of stoneflies in pretreatment samples was 543 and
decreased by roughly 75%, 47% and 77% in 1994, 1995 and 1996,
respectively, following treatments with rotenone.  In 1997, the mean
number increased to roughly 246% of the pretreatment level; however, in
1998, the metric decreased by 67% of the pretreatment value.

The report concluded that monitoring did not provide “any evidence
that rotenone use had affected macroinvertebrate abundance; however
there was an indication rotenone treatments had affected
macroinvertebrate diversity and the total number of stonefly taxa as
well as the total number of stoneflies.  Efforts to assess recovery were
confounded by a lack of reference sites and limited sample size;
however, according to the authors, “all of the metrics did return to
pre-treatment (1994) levels at least once before the end of the project
period in 1998.

Unlike the Silver King Creek report, figures in the Silver Creek report
provided vertical bars depicting the upper and lower 95% confidence
intervals around the mean.  These measures of dispersion illustrate the
broad range of variability associated with the measured indices. 
Without reference sites it is difficult to determine whether measured
changes can be attributed to rotenone treatments or “normal”
fluctuations in aquatic animal numbers due to environmental conditions.

Although treatment concentrations were selected to target brook trout
and have little to no impact on macroinvertebrates, the data suggest
that stoneflies are relatively sensitive to rotenone.  This is
consistent with what was observed on Silver King Creek. Additionally,
Ephemeroptera, Plecoptera and Trichoptera also appeared sensitive to
rotenone; however, based on at least one of the post-treatment sampling
years, representative indices appeared to recover to or exceed
pretreatment levels. As with the Silver King Creek data, while overall
abundance of macroinvertebrates did not appear to decline substantially
following rotenone treatment, the diversity of the macroinvertebrate
populations did appear to be affected.  The marked changes in several of
the indices in the final year (1998) of the study and the failure to
measure all of the indices during that year make it difficult to
determine the extent of macroinvertebrate recovery in the
rotenone-treated streams.

In general, the monitoring data on macroinvertebrate populations Silver
King Creek and Silver Creek provide useful information on the potential
effects of rotenone on non-target aquatic animals. This information will
be included in the revised risk assessment as an appendix and will be
discussed in the risk characterization.  

The AFS also included in their comments a reprint of an article entitled
“Rotenone Use in North America (1988 – 1997).  Information contained
in this article has already been included in the ecological risk
assessment chapter; therefore, this article will not be included in this
response to comments.

   HYPERLINK
"http://www.epa.gov/oppfead1/endanger/consultation/ecorisk-overview.pdf"
 http://www.epa.gov/oppfead1/endanger/consultation/ecorisk-overview.pdf 

 Email dated 05/02/06 from Robert Stewart on behalf of TIFA

 Trumbo, J.  1995. Pesticide Laboratory Report to the California
Regional Water Quality Control Board:  Lahontan Region.  State of
California Department of Fish and Game, 1701 Nimbus Road, Suite F,
Rancho Cordova, CA.  Lab No. P-1839. 

 Trumbo, J.  1994.  Pesticide Laboratory Report to the California
Regional Water Quality Control Board:  Lahontan Region.  State of
California Department of Fish and Game, 1701 Nimbus Road, Suite F,
Rancho Cordova, CA.  Lab No. P-1638.    

 Trumbo, J., S. Siepmann and B. Finlayson.  2000.  Impacts of Rotenone
on Benthic Macroinvertebrate Populations in Silver King Creek, 1990
through 1996.  State of California Department of Fish and Game,
Pesticide Investigation Unit, 1701 Nimbus Rd, Suite F, Rancho Cordova,
CA  95670.  Administrative Report 00-5.

Trumbo, J., S. Siepmann and B. Finlayson.  2000.  Impacts of Rotenone on
Benthic Macroinvertebrate Populations in Silver King Creek, 1994 through
1998.  State of California Department of Fish and Game, Pesticide
Investigation Unit, 1701 Nimbus Rd, Suite F, Rancho Cordova, CA  95670. 
Administrative Report 00-7. 

 McClay, W. 2000.  Rotenone Use in North America (1988 - 1997).  Pages
15-27 in Finlayson, B. J., R. A. Schnick, R. L. Cailteux, L. DeMong, W.
D. Horton, W. McClay, C. W. Thompson and G. J. Tichacek (authors), 
Rotenone Use in Fisheries Management: Administrative and Technical
Guidelines Manual.  American Fisheries Society, Bethesda, Maryland.

Page   PAGE  18  of   NUMPAGES  18 

Page   PAGE  1  of   NUMPAGES  18 

