Compatibility Determination

Use: Mosquito Control, Monitoring, and Research

Refuge Name: Franz Lake National Wildlife Refuge, located in Skamania County, Washington, as part of the Ridgefield National Wildlife Refuge Complex.

Establishing and Acquisition Authority(ies):

Fish and Wildlife Act of 1956 (16 U.S.C. 742)

Refuge Purpose(s): "...for the development, advancement, management, conservation and protection of fish and wildlife resources..." 16 U.S.C. 742(a)(4).

"for the benefit of the United States Fish and Wildlife Service, in performing its activities and services. Such acceptance may be subject to the terms of any restrictive or affirmative covenant or condition of servitude..." 16 U.S.C. 742f(b)(1).

National Wildlife Refuge System Mission: The mission of the National Wildlife Refuge System is "to administer a national network of lands and waters for the conservation, management, and where appropriate, restoration of the fish, wildlife, and plant resources and their habitats within the United States for the benefit of present and future generations of Americans (National Wildlife Refuge System Administration Act of 1966, as amended [16 U.S.C. 668dd-668ee])."

Description of Use(s): The Skamania County (Washington) Mosquito Control Board has proposed a program to monitor, research, and control mosquito populations on the Franz Lake National Wildlife Refuge (Franz Lake NWR; refuge)(Map, Appendix 1). This work would be performed under a refuge-issued Special Use Permit (SUP) by Skamania County or by the Clark County (Washington) Mosquito Control Districts, under contract from Skamania County, for the 2002 mosquito season.

These activities are not priority public uses, as identified in the National Wildlife Refuge System Improvement Act of 1997. This legislation clearly states that no refuge use will be allowed unless it is first determined to be compatible with the mission of the National Wildlife Refuge System and the purposes of the refuge. This Act also mandated that all refuges would develop Comprehensive Conservation Plans (CCPs). As part of the development of the Comprehensive Conservation Plan for the three Columbia River Gorge refuges (Franz Lake, Pierce, and Steigerwald Lake NWRs), which we expect to be completed by spring 2003, the refuge is required to develop new Compatibility Determinations for mosquito monitoring, research, and control (for both non-emergency and emergency situations). For this reason, mosquito monitoring, and research, and control as proposed in this Compatibility Determination, are only proposed for one year, to be reassessed as part of the CCP process. This may result in a new Compatibility Determination for these uses at that time. In decision-making, each Refuge Manager is required to use sound professional judgement based on many factors, including experience, knowledge, the best science available, and adherence to the National Wildlife Refuge System Administration Act, as amended by the National Wildlife Refuge System Improvement Act of 1997, and other applicable laws, regulations, and policies.

During the past three years (1999-2001), the Clark County Mosquito Control District has been granted a Special Use Permit (SUP) to monitor Franz Lake NWR (see map, Appendix 1) for mosquito larvae, and apply the larvicide Bacillus thuringiensis israelensis (B.t.i.) when district personnel determined that larvae had reached threshold levels. The current threshold used by the local mosquito districts is 5 mosquito larvae per dip net sample. Threshold levels used by mosquito control districts on refuges around the country range from 1 larvae to 15 larvae per dip, with an average of five (Higgins 2002, pers. comm.). The Skamania County Mosquito Control Board has proposed that the refuge issue a SUP to the Clark County Mosquito Control District, working for Skamania County, to monitor the shores of Franz and Arthur Lakes for mosquito larvae and mosquito-borne diseases, and apply B.t.i. as needed.

B.t.i. is a somewhat selective microbial insecticide targeting mosquito larvae and some other non-target dipterans. B.t.i. produces protein endo-toxins that, when ingested by the susceptible insect, cause paralysis of cells in the gut, interfering with normal digestion and feeding (Siegel 1999). The proposed mosquito control activities would be the same as those conducted on the refuge between 1999-2001; B.t.i. would be used to control mosquitoes in wetlands and waters along Franz Lake NWR. As proposed, this larvicide may be applied as often as once per two to three weeks, depending on mosquito populations, using an application rate of 10 pounds per acre. The bacteria are grown on high-protein bases (fishmeal, soy flour), which are then formulated onto corncob pellets enabling it to be broadcast over the treatment area by a hand-held or all-terrain vehicle-mounted spreader. Mosquito control operations at Franz Lake NWR would be likely to coincide with the period of highest use of this area by subyearling fall chinook and other juvenile salmonids, March through August (Appendix 2).

The need for, and application of, larvicides would depend on reaching the dip threshold of

5 larvae per dip. Mosquito larvae numbers could be affected by a number of factors, such as weather conditions, precipitation, and time of year. Probably one of the most important factors contributing to the production of these floodwater mosquitos is the degree of water fluctuation in Franz and Arthur Lakes and on their shorelines, where the mosquitos lay their eggs and subsequently hatch. Although there is some fluctuation due to precipitation, tidal influences, and floodwaters, the primary source of fluctuating water levels in the Franz Lake basin are the result of changes in water releases from Bonneville Dam, which is located approximately 10 miles upstream. Due to the uncertainty of the mosquito hatches, the total amount of larvicide to be used, and the times, dates, and exact locations of application (magnitude and frequency) cannot be predicted.

Since 1998, Clark County Mosquito Control District has contracted with Multnomah County Vector and Nuisance Control (Multnomah County's Mosquito Control District) to conduct adult mosquito monitoring in Clark County, Washington, to include Ridgefield NWR. The monitoring consisted of setting out 5 standard encephalitis traps, each consisting of a CO2 (dry ice) mosquito attractant and small collection container. Mosquitos were trapped in adult traps in each of three locations on the refuge. Trapping provided information about local mosquito populations (species and relative numbers) and a source for samples to be sent to California for testing for Western and St. Louis encephalitis, and West Nile Virus. The Skamania County Mosquito Control Board has requested a Special Use Permit to allow the same monitoring on Franz Lake NWR, with samples sent to Olympia, Washington for testing.

Last year (2001), Clark County Mosquito Control District personnel, contracted by the Skamania County Mosquito Control Board to conduct research activities under their SUP, were scheduled to collect mosquito larvae from Franz Lake NWR. These larvae were to have been reared to adult stage at the Multnomah County Mosquito Control District laboratory, marked, then released back at Franz Lake NWR. This was to be the first phase of a mark/recapture study to determine if Franz Lake was the source of nuisance mosquitoes and/or mosquito-borne diseases in the City of Skamania area. The second phase would require an experienced entomologist be hired to capture adult mosquitoes using CO2 traps set around Skamania and surrounding areas. Adult mosquitoes would be brought back to the lab, identified as to species, and checked for previous marking. The number of marked individuals would be calculated as a proportion of the total number of mosquitoes captured. Last year the Columbia River experienced low water levels, with larvae numbers too low to be available for capture and rearing. Although the mosquito control districts do not currently have enough funds to conduct the second phase of the study, they have requested permission to conduct phase one activities. They hope to obtain funding in the future to complete this research, and obtain sufficient information to justify mosquito control on Franz Lake NWR. Both phases are jointly assessed for compatibility herein.

Herein, we will assess the potential impacts of the larvicide proposed for use, B.t.i., to control nuisance mosquitoes, as well as the impacts anticipated from the proposed monitoring and research efforts.

Background: On wetland habitats of the Franz Lake NWR, annual Columbia River high water flows flood much of the bottomlands surrounding Franz Lake and the adjacent Arthur Lake. This water permeates the ground cover of non-native reed canary grass, which provides large quantities of vegetative debris as egg-laying sites for two major species of floodwater mosquitoes, Aedes vexans and Aedes sticticus, when high-water levels recede. When water returns during subsequent flooding, the eggs hatch, larvae grow rapidly, and adults feed for two weeks or more until they die. The current water elevation fluctuations, caused by natural flood conditions and the operation of the Bonneville Dam by the U.S. Army Corps of Engineers upstream of the refuge, may provide optimum conditions for multiple hatches of mosquitoes in a single summer.

Residents living near the lake shore have indicated that, prior to the acquisition of the refuge, summertime mosquito populations were frequently high on the flats in the vicinity of Franz Lake. Historically, during years of mosquito outbreaks in Skamania, Washington, local residents thought that mosquitoes hatched in the area that is now the Franz Lake NWR would fly up the Indian Mary Creek drainage toward the town of Skamania and neighboring residences. The Skamania County Mosquito Control Board responded by fogging areas suspected of providing mosquitoes by vehicle, including what is now the Franz Lake NWR, using an unknown adulticide. When the refuge was acquired in 1990, the District had been inactive since 1979-1980 (Holm 2002, pers. comm.; Price 2002, pers. comm.).

During the early 1990s, grazing in the bottomlands surrounding Franz and Arthur Lakes was discontinued to eliminate damage to the wetlands and riparian habitat in the vicinity of the lakes. The Columbia River experienced extremely high flooding in 1996 and extended periods of flooding in 1997, which although not quite as high, were persistent until mid-July. Local residents believed that these factors, combined with extended rains during 1998 and 1999, caused increases in mosquito populations large enough that the mosquitos dispersed from the refuge up the Indian Mary Creek drainage into Skamania and the surrounding areas. This prompted numerous complaints of overabundant mosquitoes to refuge staff, news media, and local legislative staff. Residents expressed concerns about the nuisance caused by the mosquitoes, as well as the potential for the transmission of mosquito-born diseases, such as West Nile Virus, St. Louis encephalitis, and western equine encephalitis.

In 1999, the Skamania County Commissioner of District 1, learned that nearby Clark County, Washington, was using B.t.i. at the Ridgefield NWR for the control of mosquito larvae and requested that it also be used at Franz Lake NWR. Ridgefield NWR still allows treatment of mosquitos using B.t.i. One of the chief differences between the two sites is the prevalence of rearing salmonids. Franz Lake is a rearing site for endangered salmonids, whereas the mosquito control activities at the Ridgefield NWR occur on managed wetlands that are inaccessible to salmonids.

A Compatibility Determination for integrated pest management at Franz Lake NWR was issued in 1999, based on initial research into the effects of B.t.i. on non-target invertebrates. A literature search at that time revealed that its effects were relatively specific, affecting mosquitoes, craneflies, blackflies, and non-biting midges (chironomids). For this reason, B.t.i. was considered a significantly better choice for mosquito control than chemical adulticides, which may impact other invertebrates, fish, and wildlife. The justification for initiating pesticide applications at that time was that it did not appear to pose significant adverse effects to non-target species, and refuge staff sought to cooperate with nearby residents of Skamania County by reducing nuisance mosquitoes and the possibility of the transmission of mosquito-borne human diseases. Special Use Permits were issued for mosquito monitoring and control with B.t.i. in 1999, 2000, and 2001.

When mosquito monitoring and control were first conducted under authority of a Special Use Permit on Franz Lake NWR in 1999, Alpine Pest Management Specialists (subcontracted by Clark County Mosquito Control District for 1999 and 2000) conducted activities on approximately 60 acres in the easternmost section of the lake that contained reed canary grass and shallow water. In 2000, they extended their treatment area to the south edge of the shoreline, adding approximately 20 acres. Treatment in this area was outside the scope of their SUP due to the presence of a bald eagle nest and high salmonid concentrations. The areas previously authorized for monitoring and control were shallow wetland areas of Franz Lake, approximately 10 percent of the total 550 acres of the refuge. The other mosquito control districts (Multnomah County, Oregon; Columbia County, Oregon; and Clark County, Washington) do not conduct their control activities within at least 15 miles of Franz Lake, reducing the potential for any adverse cumulative impacts that may occur from temporarily disrupting the salmonid food source across a large area.

Since the issuance of the 1999 Compatibility Determination, refuge staff have expressed concerns about the lack of information documenting whether the mosquitoes using Franz Lake NWR are the mosquitoes found in the nearby community of Skamania, or if they are produced in residential backyards and nearby heavily vegetated lands. Therefore, since 2000, refuge staff have been working with personnel from three local mosquito control districts. Mosquito control personnel from Multnomah County, Oregon; Columbia County, Oregon; and Clark County, Washington have worked together for several years under a mutual agreement to conduct coordinated mosquito control activities, especially aerial spraying along their boundaries. Refuge personnel initially met with this group in January 2000 to share concerns about nuisance complaints and mosquito-borne diseases; U.S. Fish and Wildlife Service responsibilities for wildlife, fish, and public health; and to develop strategies and identify responsibilities for upcoming years. Annual meetings have continued, and one of the leading issues has been to investigate methods to learn the specifics of mosquito movement from Franz Lake NWR. A Mosquito Control Board was formed in Skamania County in 2001 and is currently part of this group.

Also, since the issuance of the 1999 Compatibility Determination, refuge staff have learned of additional information concerning the adverse effects of B.t.i. on several important aquatic invertebrates, and the listing and anticipated listing under the Endangered Species Act of several salmonid fish species present on Franz Lake NWR. The potential non-target impacts to aquatic invertebrate species from exposure to B.t.i. may reduce food resources for federally listed salmonids which use portions of Franz and Arthur Lakes as rearing areas. This information is summarized in the Wildlife and Wetland, and Fisheries Impacts from Mosquito Control sections below, with additional details about fisheries impacts described in Appendix 3 to this Compatibility Determination.

Our research on the potential for transmission of mosquito-borne diseases reveals that there have been 31 human cases state-wide of western equine and St. Louis encephalitis (with no deaths) between 1950 and 2000, and no reported cases of these diseases in Clark or Skamania Counties (Grendon 2000). Also, as of March 2002, the West Nile Virus (WNV) has been limited to the eastern half of the United States (U.S. Geological Survey - Center for Integration of Natural Disaster Information 2002), although some experts believe that it will be present in the west within two or three years. The State of Washington recently completed the first year of a West Nile Virus surveillance project, with 12 Washington counties participating. A mosquito species found in the eastern United States capable of carrying the West Nile Virus was found for the first time in the western United States near Seattle, Washington in January 2002. State health department officials were unable to determine if the species was new to the area or was identified as a result of new surveillance efforts. Although Clark County has had adult mosquito capture/testing conducted at Ridgefield NWR the past four years under a contract with Multnomah County, they are not planning to do so this year. Skamania County has not been involved in monitoring for mosquito-borne diseases to this date. This year, however, the Skamania County Mosquito Control Board has been provided with carbon dioxide traps and test tubes to capture adult mosquitos in locations of activity and send them to the Washington State Department of Health to test for mosquito-borne diseases. They would like to include Franz Lake NWR as one of their sampling sites.

In the event that the mosquitoes produced on Franz Lake NWR are determined to be causing a human-health emergency, the most appropriate control technique would be utilized and all Stipulations listed herein would be implemented.

Availability of Resources: While adequate funding and staff exist to manage the proposed mosquito monitoring and research activities at Franz Lake NWR, current and perceived future staffing levels are inadequate for the level of oversight that would be required for the proposed nuisance mosquito control program.

When staff from the Clark County Mosquito Control District visit Franz Lake NWR for monitoring/control activities, they usually spend a portion of the first day of the week taking samples, then they return and apply the treatment on another day. In the Gorge, there is also considerable travel time involved. The Control District schedule is fairly irregular and pinning their staff down to a schedule has been virtually impossible at Ridgefield NWR, and would be even more difficult to do at Franz Lake NWR. In 1999, mosquito activity was severe, but the Special Use Permit authorizing monitoring and control at Franz Lake NWR by Alpine Pest Management Specialists (subcontracted by Clark County Mosquito Control District for 1999 and 2000) was not in place until August. In spite of the late start, Alpine still made 8 monitoring/control trips. The year 2000 was a relatively low mosquito year because of few fluctuations in Columbia River water releases from the Bonneville Dam, but Alpine still made 7 visits to the refuge. The year 2001 was an extremely low-water year and resulted in two trips to Franz Lake by Clark County Mosquiot Control District staff, with no larvae dip thresholds reached due to the low-water levels.

Ideally, refuge staff would accompany mosquito control district personnel whenever they came onto the refuge to ensure their compliance with the stipulations presented herein and the terms and conditions of the SUP issued for the control treatment. That would amount to a full day or more of staff time every week or two when they schedule a visit, considering the disruptions in the regular work schedule, inability to schedule meetings, etc. The refuge complex does not have adequate staff available to monitor the mosquito control solely for nuisance purposes. That said, the refuge would ensure staff were made available to coordinate mosquito control by the districts in the event of a health-related emergency.

Anticipated Impacts of the Use(s):

Wildlife and Wetland Impacts Anticipated from Mosquito Control

The waters and wetlands of Franz Lake NWR provide habitat for a variety of wildlife that use mosquitoes (adults and larvae) and other invertebrates as food resources. Five species of swallows (tree, violet-green, barn, cliff, and rough-winged) and purple martins use the Franz Lake area, where they feed extensively on aerial insects (A. Poole and F. Gill, eds. 2002). These birds' primary food source is larger insects like dragonflies and damselflies, which feed on mosquitos (Higgins 2002, pers. comm.). These birds fly in groups of up to 25-50 over open water and wetland areas.

Breeding ducks and their young (mallards, cinnamon and blue-winged teals, gadwalls, northern shovelers, hooded mergansers, and wood ducks) and Canada geese are summer users of the lakes and wetlands, feeding in shallow waters along the shorelines. These species feed by filtering aquatic invertebrates (including mosquito larvae) from the water column, substrate, and vegetation.

It has been generally acknowledged that aquatic invertebrates comprise a significant portion of the diet of breeding waterfowl, but contributed little to the overall diet of non-breeding waterfowl. Recent studies, however, have shown that invertebrate foods have been found to be the dominant food of even non-breeding waterfowl when molting during the summer, fall, and winter (Heitmeyer 1988), egg production during the spring (Swanson et al. 1979), and duckling growth during the summer rearing period (Krapu and Swanson 1978).

Other wetland birds, such as great blue herons, great egrets, bitterns, sora, and Virginia rails are summer residents of the Franz Lake NWR, using vegetation in shallow waters along the shoreline for nesting and cover. The rails feed directly on selected invertebrates, while the larger wetland birds (i.e., great blue herons, great egrets, etc.) feed on small fish, reptiles, and amphibians, which themselves depend upon wetland invertebrates as their food supply. Midges of the family Chironomidae and others are important components of the diets of many of these wetland-using birds. These invertebrates, which are biologically, morphologically, and behaviorally similar to mosquitoes, are adversely impacted by most mosquito control treatments, including B.t.i.

It has been very difficult to develop any chemical pesticide which is entirely target specific, as most pesticides have some degree of impact on various non-target organisms. In recent years, the larvicide Bacillus thuringiensis israelensis (B.t.i.) has been viewed by some people as a "preferred" mosquito control material because of its status as a biological control.

Recent literature regarding the effects of B.t.i. on non-target organisms have highlighted and verified a number of concerns about mosquito control on wildlife refuges. The majority of initial studies have been accomplished in the lab or during one-season field studies. The first long-term research was a 3-year study (1998-1990) completed in Minnesota (Charbonneau et al. 1994). The first two years showed no reduction of the major aquatic invertebrates in the ponds, but the third year indicated significant adverse impacts to chironomids and several other taxa that are important dietary components of vertebrates such as waterfowl, fish, and amphibians. Further toxicity tests indicated that various environmental factors, such as temperature, larval instar development, water depth, water flow, tidal flow, substrate type, and the amount of organic material in the water column reduced the efficacy of the larvicide in the field.

A six-year study on 27 wetlands in Minnesota consisted of three years (1988-1990) of pre-treatment sampling of aquatic invertebrates and other parameters, followed by three years (1991-1993) of treatment with B.t.i. and subsequent sampling (Hershey et al. 1998). Insect densities and diversity were reduced by 57 to 83 percent in the second and third years of treatment. During this study, 179 genera of aquatic insects were examined, with chironomids (primarily midges) representing about half of the insect genera present at the beginning of the study. By the end of the study, however, only one to six genera dominated the treatment sites. Adverse impacts were primarily observed in the invertebrate tribes, primarily in the Chironomini and Tantarsini. These tribes are ubiquitous and are represented in almost every wetland with chironomids. There are over 300 species of midges within these two groups and not necessarily all of them would be adversely impacted by B.t.i,, but since it is likely that less than 10 species were examined, impacts of B.t.i. on individual species are unknown. This suggested the potential for indirect impacts on non-target species, creating a negative food web effect, and highlighted the need for further study, especially on a long-term basis. Without knowledge of what midge species are present in Franz and Arthur Lakes and what the impact to those species would be from B.t.i., the conservative approach based on the best available science would be to avoid application of B.t.i. (Higgins 2002, pers. comm.).

The potential negative impacts of B.t.i. treatments with respect to wildlife refuges center on direct impacts to chironomids and a general disruption of the aquatic food chain, both anticipated to be short-term, but depending on severity, may persist over the long-term. As stated earlier, chironomids are one of the more important food web components in wetlands. A single, highly-effective treatment (with accompanying side effects) can negate the important ecosystem values of individual wetlands and significantly reduce the value of these wetlands to wildlife species, which are the focus of refuge goals. Impacts of B.t.i. on non-target invertebrates may be highly variable and there are many unknowns.

Fisheries Impacts from Mosquito Control

The waters and wetlands of Franz Lake NWR provide habitat for several species of anadromous salmonids including chinook salmon, coho salmon, chum salmon, steelhead, and sea-run coastal cutthroat trout. Steelhead and fall chinook are listed under the Endangered Species Act as threatened, cutthroat trout are proposed for listing, and coho salmon are candidates for listing.

Coho salmon, steelhead, and cutthroat trout spawn in tributary streams, and chum salmon spawn in the outlet of Franz Lake. These species, as well as chinook salmon, rear in Franz and Arthur Lakes. Juvenile anadromous salmonids that originate from nearby tributaries, the mainstem Columbia River, and other streams throughout the Columbia River system also enter the Franz/Arthur Lake backwater to rear and to seek refuge from high flows in the Columbia River during the spring freshet. Shallow waters in these areas are especially important rearing areas for subyearling chinook salmon during their seaward migrations. Salmonids are moving through the Franz Lake NWR throughout the year. Other native and introduced fish species are also present throughout the year (see Appendix 3, Table 1, for list of species collected at Franz Lake NWR).

Bioassays have shown that B.t.i. is of low direct toxicity to various fish species. Therefore, salmonids and other fishes are not expected to suffer direct adverse effects from applications of B.t.i. for mosquito control. There is greater potential for indirect adverse effects to fish from disruption of the food web. B.t.i. is specific to mosquitoes, blackflies, and midges (chironomids). All three groups of insects are important food items for salmonids and other fishes, with midges (chironomids) having been documented as one of the most important food items for subyearling chinook salmon throughout the Columbia River and the Pacific Northwest. Subyearling chinook salmon inhabit the shallow backwater areas of Franz and Arthur Lakes from March through August (see Appendix 2, Graph 1), and during this time they feed extensively on chironomids. As noted earlier, studies have shown B.t.i. to have adverse effects on various species of chironomids. Lowered chironomid populations would reduce an important food source for rearing juvenile salmonids during the critical rearing period before seaward migration.

Mosquito control operations at Franz Lake NWR would be likely to coincide with the period of highest use of this area by subyearling fall chinook and other juvenile salmonids (March through August) (see Appendix 2, Graphs 2-4). Mosquito populations likely would be higher during years when Columbia River flows are high and water levels are fluctuating. These are also the years when more juvenile salmonids would be expected to enter backwaters and side channels to seek refuge from high-water velocities in the mainstem Columbia River. These fish would typically move along the shoreline of the Columbia River where velocities are slower, and enter into backwaters, where they would continue to move along the shorelines in search of food. This would lead them to the shallow waters vegetated with reed canary grass along the shorelines and the east end of Franz Lake. Mosquito control operations in these areas and during these years would likely adversely affect many juvenile salmonids.

Wildlife and Fisheries Impacts from Mosquito Monitoring and Research

The impacts anticipated from the proposed monitoring and research activities would be minimal. Foot traffic would be minimized; a single, all-terrain vehicle would be allowed for these activities and would be restricted to refuge service roads and a maximum speed limit of 5 miles per hour; and the number of mosquito larvae and adults anticipated to be taken would not be large enough to adversely impact any species or habitat. These activities would likely pose only minor, short-term disturbances. In the event of a health-related emergency, the data from these monitoring and research efforts would be invaluable to the refuge for determining an appropriate course of action.

Public Review and Comment: Public review of and comment on this Compatibility Determination will be conducted via a 30-day comment period, with notices placed at: the refuge headquarters; the Ridgefield NWR Complex headquarters; U.S. Fish and Wildlife Service, Lower Columbia River Fisheries Program Office, Washington State Office; National Marine Fisheries Service; Washington Department of Fish and Wildlife (Vancouver); Skamania County District 1 Commissioner; mosquito control districts (Skamania County, Clark County, Multnomah County, Columbia County); Stevenson, Camas, and Washougal post offices and libraries; the Skamania General Store; and in local newspapers (Vancouver Columbian, Skamania County Pioneer, Camas-Washougal Post Record, and Oregonian). Copies would also be sent to federal, state, and local legislative representatives. The proposed refuge uses evaluated herein will be re-evaluated during the CCP process and a new Compatibility Determination may be prepared at that time.

Determination for Mosquito Control (Non-emergency) (Check one below):

X Use is Not Compatible

Use is Compatible With Following Stipulations

Stipulations for Mosquito Control (Non-emergency):

Not applicable.

Justification for Mosquito Control (Non-emergency):

Based on the information available, the scale of impacts to chironomids and other non-target aquatic invertebrates from B.t.i. application depends on the species presence, abundance and distribution; substrate type; water depth during treatment periods; and the density of aquatic invertebrates at the treatment sites. Information about these parameters is largely unknown for the Franz Lake NWR. However, initial research in Minnesota indicates negative impacts on several non-target aquatic invertebrates, including chironomids.

Specifically, additional information is needed to properly assess the impacts of B.t.i. application on chironomids, wetland birds, and salmonids. These information needs include the habitat characteristics of areas to be treated, species of invertebrates (particularly chironomids) present, densities of invertebrate populations (particularly chironomid) in proposed treatment areas, and wildlife and fish use of these areas (see Appendix 3 for a discussion of research needs). Additionally, bioassays have not been conducted to examine the effect of B.t.i. on the chironomid species present. This should be done. The results would aid in predicting whether or not large-scale and repeated application of these insecticides would detrimentally impact the food web at Franz Lake NWR.

Subyearling chinook salmon are the species of salmon most likely to be using shallow water habitat in Franz Lake NWR and are also the species and life stage of salmonid that relies most heavily on chironomids. Mosquito treatment is likely to occur during the time when subyearling chinook salmon would be rearing in shallow waters of Franz Lake NWR, as these are also the times when mosquito populations in these areas are high. Subyearling chinook salmon use of Franz Lake NWR may also be highest in years of higher river flow when these fish seek refuge from strong river currents in backwaters.

Recent research into the effects of Bacillus thuringiensis israelensis (B.t.i.) on non-target invertebrate populations important to wildlife and fish species using the refuge, especially endangered/threatened species of salmon, have shown that treatment with B.t.i. may adversely affect listed fish species because of short-term food chain disruption, with unknown impacts in the long-term. This information, when combined with the absence of vital refuge-specific invertebrate information, has led to the conclusion that mosquito control using this product, which currently is considered to have the least negative consequences on invertebrate populations important to fish and wildlife, cannot be determined to be compatible at this time, as it is anticipated that this use would materially interfere with and detract from refuge purposes, Refuge System mission, and our legal mandate to ensure the biological integrity, diversity, and environmental health of the NWR System.

Determination for Mosquito Control (Emergency):

Use is Not Compatible

X Use is Compatible With Following Stipulations

Stipulations Necessary to Ensure Compatibility of Mosquito Control (Emergency):

1. Mosquito control measures will only be taken in the case of an emergency when there is real and imminent threat to the health of human, fish, or wildlife populations. For the purpose of treatment of refuge lands for disease-carrying mosquitoes, a public health emergency will be determined by the Regional Director of the U.S. Fish and Wildlife Service. The Regional Director will make this determination after consultation with the U.S. Centers for Disease Control and Prevention, the Washington State Department of Health, the Southwest Washington Department of Health (Clark County), and/or the Skamania County Department of Health. Once the Regional Director has determined the existence of a human health emergency, designated Service representatives will prescribe, following consultation with State and County health district personnel, and recognized biologists and entomologists, the type and duration of treatment for mosquitoes on refuge lands.

2. For the purpose of allowing the use of certain chemical pesticides or biological control pesticides to control mosquitoes, a mosquito-borne public health emergency is defined as:

Actual or threatened, imminent outbreak of western equine encephalitis, St. Louis encephalitis, West Nile encephalitis, malaria, or other mosquito vector-borne public health disease. The presence of western equine encephalitis, St. Louis encephalitis, West Nile encephalitis, or malaria viral titers or mosquito pool titers in the mosquito population, wild birds, or in sentinel chickens (in accordance with test protocols developed by the Washington State Department of Health and the U.S. Department of Health and Human Services, Centers for Disease Control and Prevention (CDC 2001)) will confirm that a public health emergency exists or is imminent. The West Nile Virus will also be monitored due to the discovery of its presence on the east coast in the vicinity of New York City and other locations in September 1999, and the January 2002 discovery in Seattle, Washington of a mosquito species found in the eastern United States capable of carrying the West Nile Virus.

3. Once a mosquito abatement district notifies the Refuge Manager of a laboratory test that is positive for any of the above viruses in mosquitoes documented to be on the refuge, the Refuge Manager will immediately inform the Regional Director of the U.S. Fish and Wildlife Service to make the ultimate determination of whether an emergency exists relative to the refuge.

4. The Service may allow mosquito control on System lands when all of the following exist:

A. There is documented incidence of mosquito-borne disease activity in those mosquito species that have been documented to breed on the refuge. The disease may be identified in birds, mosquitoes, humans, and/or other mammals.

B. Current monitoring data indicate that refuge-based mosquitoes pose an imminent threat to human health. The term "imminent threat" used here would mean high larval and/or adult populations of a mosquito species known to be present on the refuge.

C. The Regional Director, in consultation with appropriate health care professionals, determines there is a high probability that disease-carrying mosquitoes originating from the refuge may infect humans.

5. After receiving the determination of an emergency from the Regional Director, the Refuge Manager, in consultation with recognized experts (i.e., biologists/entomologists) and human health care authorities, will respond appropriately to a mosquito-borne disease threat when the above conditions have been met. The term "appropriately" means using the most effective agents practicable to address the specific human health risk, and would include larviciding or, in cases where infected mosquitoes have been found on System lands, adulticiding.

6. The use of pesticides on the refuge will only occur when:

A. there is evidence of an increasing risk for human infection;

B. this risk will be substantially lowered by the pesticide(s); and

C. the application of pesticides to the refuge is superior to other available approaches to manage disease risk.

7. Any emergency treatment lasting beyond 30 days will require a separate Compatibility Determination.

Justification for Mosquito Control (Emergency):

The National Wildlife Refuge System Improvement Act of 1997 authorizes the Secretary of the Interior to temporarily suspend, allow, or initiate any activity on a refuge necessary to protect the health and safety of the public or any fish and wildlife population on the National Wildlife Refuge System.

Determination for Mosquito Monitoring (Check one below):

Use is Not Compatible

X Use is Compatible With Following Stipulations

Stipulations Necessary to Ensure Compatibility of Mosquito Monitoring