Endangered and Threatened Wildlife and Plants; Endangered Species Status for Southern Sierra Nevada Distinct Population Segment of Fisher

DEPARTMENT OF THE INTERIOR

Fish and Wildlife Service

50 CFR Part 17

[Docket No. FWS-R8-ES-2018-0105; FF09E21000 FXES11110900000 201]

RIN 1018-BD85

Endangered and Threatened Wildlife and Plants; Endangered Species Status for Southern Sierra Nevada Distinct Population Segment of Fisher

AGENCY:

Fish and Wildlife Service, Interior.

ACTION:

Final rule.

SUMMARY:

We, the U.S. Fish and Wildlife Service (Service), determine endangered species status under the Endangered Species Act (Act), as amended, for the Southern Sierra Nevada Distinct Population Segment (DPS) of fisher (

Pekania pennanti

). This DPS occurs in California. The effect of this regulation will be to add this DPS to the List of Endangered and Threatened Wildlife.

DATES:

This rule is effective June 15, 2020.

ADDRESSES:

This final rule is available on the internet at

http://www.regulations.gov

in Docket No. FWS-R8-ES-2018-0105 and at

https://www.fws.gov/Yreka

. Comments and materials we received, as well as supporting documentation we used in preparing this rule, are available for public inspection at

http://www.regulations.gov

. Comments, materials, and documentation that we considered in this rulemaking will be available by appointment, during normal business hours at: U.S. Fish and Wildlife Service, Yreka Fish and Wildlife Office, 1829 South Oregon Street, Yreka, CA 96097; telephone 530-842-5763.

FOR FURTHER INFORMATION CONTACT:

Jenny Ericson, Field Supervisor, Yreka Fish and Wildlife Office, telephone: 530-842-5763. Persons who use a telecommunications device for the deaf may call the Federal Relay Service at 1-800-877-8339.

SUPPLEMENTARY INFORMATION:

Executive Summary

Why we need to publish a rule.

Under the Act, if we determine that a species may be an endangered or threatened species throughout all or a significant portion of its range, we are required to promptly publish a proposal in the

Federal Register

and make a determination on our proposal within 1 year. To the maximum extent prudent and determinable, we must designate critical habitat for any species that we determine to be an endangered or threatened species under the Act. Listing a species as an endangered or threatened species and designation of critical habitat can only be completed by issuing a rule.

What this document does.

This rule will add the Southern Sierra Nevada DPS of fisher (

Pekania pennanti

) (SSN DPS) as an endangered species to the List of Endangered and Threatened Wildlife in title 50 of the Code of Federal Regulations at 50 CFR 17.11(h).

The basis for our action.

Under the Act, we may determine that a species is an endangered or threatened species based on any of five factors: (A) The present or threatened destruction, modification, or curtailment of its habitat or range; (B) Overutilization for commercial, recreational, scientific, or educational purposes; (C) Disease or predation; (D) The inadequacy of existing regulatory mechanisms; or (E) Other natural or manmade factors affecting its continued existence. We identified multiple threats under various factors that are acting on, and will continue to act on, the SSN DPS, the full list of which can be found in our final Species Report 2016 (Service 2016, entire).

Of particular significance regarding implications for the DPS's status were loss and fragmentation of habitat resulting from high-severity wildfire and wildfire suppression (

i.e.,

loss of snags and other large habitat structures on which the species relies), climate change, and tree mortality from drought, disease, and insect infestations. Also of significance were threats related to potential direct impacts to individual fishers (

e.g.,

increased mortality, decreased reproductive rates, increased stress/hormone levels, alterations in behavioral patterns), including wildfire, increased temperatures resulting from climate change, disease and predation, exposure to toxicants, collisions with vehicles, and potential effects associated with small population size. These factors are resulting in a cumulative effect to such a degree that the best available information indicates the Southern Sierra Nevada DPS of fisher meets the definition of an endangered species.

Peer review and public comment.

In accordance with our joint policy on peer review published in the

Federal Register

on July 1, 1994 (59 FR 34270), and our August 22, 2016, memorandum updating and clarifying the role of peer review of listing actions under the Act, we sought comments from independent specialists to ensure that our consideration of the status of the species is based on scientifically sound data, assumptions, and analyses. We invited these peer reviewers to comment on both the draft Species Report (Service 2014) as well as the 2014 Proposed Rule (79 FR 60419, October 7, 2014). We also considered all comments and information received during three public comment periods (and one extension) for the 2014 Proposed Rule (79 FR 60419, October 7, 2014; 79 FR 76950, December 23, 2014; 80 FR 19953, April 24, 2015; 84 FR 644, January 31, 2019) and two comment periods for the 2019 Revised Proposed Rule (84 FR 60278, November 7, 2019; 84 FR 69712, December 19, 2019). All comments received during the peer review process and the public comment periods have either been incorporated in the final Species Report (Service 2016, entire), in this rule, or addressed in the Summary of Comments and Recommendations section of the preamble.

Acronyms and Abbreviations Used

We use several acronyms and abbreviations throughout the preamble of this final rule. To assist the reader, we list them here:

BLM = Bureau of Land Management

CAL FIRE = California Department of Forestry and Fire Protection

CBI = California Biology Institute

CCAA = Candidate Conservation Agreements with Assurances

CDFW = California Department of Fish and Wildlife

CESA = California Endangered Species Act

CEQA = California Environmental Quality Act

CFGC = California Fish and Game Commission

C.I. = confidence interval

DOI = Department of the Interior

DPS = distinct population segment

EKSA = Eastern Klamath Study Area

EPA = Environmental Protection Agency

ESU = evolutionarily significant unit

FPR = forest practice rules

GDRC = Green Diamond Resource Company

GNN = gradient nearest neighbor

HCP = Habitat Conservation Plan

MAUCRSA = Medicinal and Adult-Use Cannabis Regulation and Safety Act

MOU = Memorandum of Understanding

NCSO = Northern California/Southern Oregon

NEPA = National Environmental Policy Act

NFMA = National Forest Management Act

NPS = National Park Service

NSN = Northern Sierra Nevada

NWFP = Northwest Forest Plan

ODF = Oregon Department of Forestry

OGSI = old growth structure index

ONP = Olympic National Park

PECE = Policy for the Evaluation of Conservation Efforts

RCP = representative concentration pathways

RMP = resource management plan

SHA = Safe Harbor Agreements

SNAMP = Sierra Nevada Adaptive Management Project

SOC = Southern Oregon Cascades

SPI = Sierra Pacific Industries

SSN = Southern Sierra Nevada

USFS = U.S. Forest Service

USDA = U.S. Department of Agriculture

Previous Federal Actions

We first found the West Coast DPS of fisher (previously delineated as a contiguous area encompassing parts of the three States of Washington, Oregon, and California) to be warranted for listing in 2004 and each subsequent year in the annual Candidate Notice of Review. On October 7, 2014, we proposed to list the West Coast DPS of fisher as a threatened species under the Endangered Species Act of 1973, as amended (Act; 16 U.S.C. 1531

et seq.

) (79 FR 60419; Docket No. FWS-R8-ES-2014-0041) (hereafter referred to as 2014 Proposed Rule). On April 18, 2016, we withdrew the proposed rule to list the West Coast DPS of fisher (81 FR 22710), concluding that the potential threats acting upon the DPS were not of sufficient imminence, intensity, or magnitude to indicate that they were singly or cumulatively resulting in significant impacts at either the population or rangewide scales such that the DPS met the definition of an endangered or threatened species.

On October 19, 2016, the Center for Biological Diversity, Environmental Protection Information Center, Klamath-Siskiyou Wildlands Center, and Sierra Forest Legacy filed a complaint for declaratory and injunctive relief, alleging that our determination on the West Coast DPS of fisher violated the Act. By Order Re: Summary Judgment issued on September 21, 2018, the District Court for the Northern District of California vacated the listing withdrawal and remanded the Service's final determination for reconsideration. The Court's amended order, dated November 20, 2018, directed the Service to prepare a new determination by September 21, 2019.

On January 31, 2019, we reopened the comment period on the October 7, 2014, proposed rule to list the West Coast DPS of fisher as a threatened species (84 FR 644).

On May 17, 2019, the District Court for the Northern District of California granted a request by the Service for a 35-day extension to comply with the November 20, 2018, order as a result of delays due to the Federal Government's lapse in appropriations that prohibited the Service from working on this determination. The Court's amended order directed the Service to submit for publication a final listing determination or notice of a revised proposed rule by October 26, 2019, and in the event of publishing a revised proposed rule, submit for publication a final listing determination by April 25, 2020.

On November 7, 2019, we published a revised proposed rule to list the West Coast DPS of fisher (84 FR 60278) (hereafter referred to as 2019 Revised Proposed Rule). In the 2019 Revised Proposed Rule, we evaluated new information available since 2014 and reconsidered the best available information already in our files (including all peer, partner, and public comments received during previous comment periods as well as the two recent comment periods on the 2019 Revised Proposed Rule). In the 2019 Revised Proposed Rule, we concluded that the West Coast DPS of fisher continued to meet the definition of a threatened species based on cumulative effects associated with multiple threats across the DPS's range.

Additional information on Federal actions concerning the West Coast DPS of fisher prior to October 7, 2014, is outlined in the 2014 Proposed Rule (October 7, 2014, 79 FR 60419).

Summary of Changes From the 2019 Revised Proposed Rule

Our 2019 Revised Proposed Rule discussed how potential changes from the proposed rule to the final rule regarding status would constitute a logical outgrowth, stating that, “Because we will consider all comments and information received during the comment period, our final determination may differ from the proposed rule. Based on the new information we receive (and any comments on that new information), we may conclude that the species is endangered instead of threatened, or we may conclude that the species does not warrant listing as either an endangered or a threatened species. Such final decisions would be a logical outgrowth of this proposal as long as we: (1) Base the decisions on the best scientific and commercial data available after considering all of the relevant factors; (2) do not rely on factors Congress has not intended us to consider; and (3) articulate a rational connection between the facts found and the conclusions made, including why we changed our conclusion (84 FR at 60278-79, November 7, 2019).” Although this discussion centered on a final decision regarding the status of the previously singular West Coast DPS, and the logical outgrowth leading to that decision from our Revised Proposed Rule, we have followed this approach in developing this final rule in its totality, to include our re-evaluation of the DPS and the resulting status determinations that followed from our revised DPS determinations.

In our 2019 Revised Proposed Rule we presented our delineation of the DPS for West Coast populations of fishers, which was revised from the 2014 Proposed Rule. This revised delineation identified the West Coast DPS as comprising the two extant historically native subpopulations, Northern California/Southern Oregon (NCSO) and Southern Sierra Nevada (SSN), as well as the Northern Sierra Nevada (NSN, also known as the Stirling subpopulation, as referenced in specific text regarding the Stirling Management Unit) and Southern Oregon Cascades (SOC) subpopulations that resulted from reintroductions within a portion of the historical range of the DPS. These four subpopulation groups occur geographically in essentially two groupings: NCSO (including NSN and SOC subpopulations) and the wholly separate SSN subpopulation.

In the 2014 Proposed Rule, we explained that the DPS we proposed to list included all the fisher subpopulations in the three western States (Washington, Oregon, California) known to be extant at that time. Thus, the DPS included the fisher subpopulations in NCSO (including SOC and NSN), SSN, and Olympic National Park (ONP) in Washington. Both the ONP and SOC subpopulations were established with fishers translocated from areas outside the three western States,

e.g.,

British Columbia, Alberta, and Minnesota; the NCSO and SSN subpopulations were existing subpopulations historically indigenous to this three-State area, and NSN was established with fishers translocated from the NCSO source subpopulation.

However, we also included a discussion of potential alternative DPS configurations in the 2014 Proposed Rule, and we requested public comment and peer review on the two alternative DPS configurations.

DPS Alternative 1 consisted of a single DPS encompassing the extant subpopulations with unique genetic characteristics in California and southern Oregon (

i.e.,

NCSO, NSN, and SSN). Alternative 1 focused on conservation of known fishers indigenous to this California and southern Oregon region, and it excluded all reintroduced subpopulations established with non-California/Oregon fishers (

i.e.,

SOC and ONP). In addition, Alternative 1 excluded areas to the north of NCSO where subpopulations of historically indigenous fishers were likely extirpated. It included both SSN

and NCSO (which includes NSN), which each have unique genetic characteristics; this inclusion would allow for management of both these native subpopulations as a single DPS. In addition, this would allow for recovery efforts throughout the historical range in California and southern Oregon.

DPS Alternative 2 consisted of two narrowly drawn DPSs around each of the extant subpopulations with unique genetic characteristics in California and southern Oregon (

i.e.,

NCSO with NSN, and SSN). This alternative also focused on conservation of known fishers indigenous to this California and southern Oregon region with unique genetic characteristics, and it excluded all reintroduced subpopulations (

i.e.,

SOC and ONP) established with non-California/Oregon fishers. This Alternative excluded the areas to the north of NCSO where fisher subpopulations were likely extirpated; it included both NCSO (which includes NSN) and SSN subpopulations, which each have unique genetic characteristics; and it allowed for management of the subpopulations as separate DPSs, recognizing the unique genetic characteristics within each. In addition, if the magnitude of threats was found to be different in the two DPSs, this would allow for different management for each DPS with regard to recovery.

We received multiple comments on our DPS approach and possible alternative DPS configurations in response to the 2014 Proposed Rule. These comments spanned a broad range of responses from support for the full three-State DPS to support for each of the possible Alternatives to support for other configurations. The basis for the commenters' positions was equally varied; these positions ranged from supporting differing genetics between subpopulations to supporting the need for different management considerations. After consideration of all of these comments, we moved forward with a modified Alternative 1 in the 2019 Revised Proposed Rule, with the exception that we included SOC in the DPS (as part of NCSO). In the 2019 Revised Proposed Rule, we did not specifically state that the DPS was based on focusing on conservation of the extant subpopulations with unique genetic characteristics, but we did explain that the DPS was centered on what we called the “historically native” subpopulations (

i.e.,

those subpopulations of known fishers indigenous to the California and southern Oregon region with unique genetic characteristics) and included SOC because of the recent interbreeding with indigenous NCSO fishers.

Our 2019 Revised Proposed Rule further sought comment regarding its revised DPS determination (84 FR at 60279, November 7, 2019). We received numerous comments regarding the revised DPS determination in response to the 2019 Revised Proposed Rule, both during the initial 30-day comment period and in the subsequent 15-day comment period. Similar to the comments received on the 2014 Proposed Rule, the comments received on the 2019 Revised Proposed Rule expressed support for a wide range of DPS approaches. Various commenters suggested reverting back to the three-State DPS (

i.e.,

include Washington State again), making all subpopulations (NCSO, SSN, NSN, and SOC) individual DPSs, having two separate DPSs as in Alternative 2, and not including SOC in any DPS configuration.

While the comments presented a broad range of positions regarding DPS approaches, there was also a relatively consistent theme regarding management considerations. Many comments pointed to a concept we presented in the 2014 Proposed Rule that outlined alternative DPSs based on recognizing the unique genetic characteristics within each subpopulation and allowing for separate management of these two population segments (NCSO [including NSN and SOC] and SSN).

In light of the numerous comments received during multiple comment periods over the last 5 years recommending we reexamine our DPS configuration, we have again reevaluated our DPS approach. We determined that the most appropriate path forward was to evaluate the two population segments ((1) NCSO [including NSN and SOC] and (2) SSN) as individual DPSs (similar to Alternative 2 in the 2014 Proposed Rule). For each population segment, if both the discreteness and significance criteria were met, we would then evaluate the status for that individual DPS. We determined our analysis would focus on the conservation of extant subpopulations historically indigenous to the California and southern Oregon region with unique genetic characteristics (as outlined in the 2014 Proposed Rule) while also allowing for separate management of the two DPSs if either or both were warranted for listing. The concept of the possible need for different management between the two DPSs was further strengthened, in part, by the recent limited introduction of non-California/Oregon fisher genes into the NCSO subpopulation via interbreeding between NCSO and SOC fishers. We have now determined that the singular West Coast DPS configuration should instead be two separate DPSs: The NCSO DPS and the SSN DPS.

BILLING CODE 4333-15-P

ER15MY20.000

BILLING CODE 4333-15-C

The above discussion presents a logical outgrowth from our 2019 Revised Proposed Rule regarding our DPS determination for the following reasons. First, our 2014 Proposed Rule (79 FR 60419, October 7, 2014) recognized that for fisher, the Service's DPS analysis had started with the petitioned DPS, which included portions of California, Oregon, and Washington, but also pointed out that the Service had identified smaller areas within the larger DPS boundary that would also potentially constitute a valid DPS, and that may warrant listing under the Act (79 FR at 60438). The 2014 Proposed Rule further announced the Service's evaluation of a number of alternative DPSs that may potentially also be valid DPSs (covering a smaller entity or entities) and that the Service was considering in particular the appropriateness of two of these alternatives and seeking public and peer

review input on potential DPS alternatives (79 FR at 60438). One of these alternatives was Alternative 2, which consisted of two narrowly drawn DPSs around the extant subpopulations with unique genetic characteristics in California and southern Oregon; Alternative 2 is similar to the two DPS approaches we use here. Therefore, the public has seen this approach presented before, was aware that we were considering it and thus could anticipate that adoption of this approach was possible, and had several opportunities to provide comments on the approach.

Second, we outlined the uncertainty associated with our DPS approach in the 2014 Proposed Rule and alerted the public to this uncertainty. Specifically, our 2014 Proposed Rule stated that we sought peer review and public comment on the uncertainties associated with the specific topics outlined in the Information Requested section and in the Other DPS Alternatives section. Specific information from the peer reviewers and the public on the proposed DPS and the two alternatives informed our final listing decision (70 FR at 60441).

Third, our 2014 Proposed Rule explained to the public that the DPS approach in our final rule may differ from the proposed rule as a result of public comment. We stated that we may determine that the proposed DPS as set forth is the most appropriate for fisher conservation. Alternatively, through peer review and public comment, we could determine that one of the alternative DPSs set forth would be most appropriate for the conservation of fisher, and, therefore, any final listing determination may differ from this proposal (79 FR at 60438). As outlined above, we have explained the basis for this changed DPS and have articulated a rational connection between the facts found and our conclusion by which we have determined to separate the singular West Coast DPS configuration into two separate DPSs.

The Secretary has discretion when determining DPSs based upon the Congressional guidance that the authority to list DPS's be used `. . . sparingly' while encouraging the conservation of genetic diversity and in consideration of available scientific evidence of the discrete population segment's importance to the taxon to which it belongs (61 FR 4722, 4725, February 7, 1996). Our DPS approach of evaluating the two fisher population segments ((1) NCSO [including NSN and SOC] and (2) SSN) as separate DPSs encourages the conservation of genetic diversity by focusing on conserving extant native subpopulations with unique genetic characteristics.

Once we determined that the singular West Coast DPS should instead be two separate DPSs, we began individually evaluating the status of the NCSO DPS and the SSN DPS. In the 2019 Revised Proposed Rule (84 FR 60278, November 7, 2019), we proposed to list the then-singular West Coast DPS as a threatened species under the Act, and we also proposed a concurrent rule under section 4(d) of the Act for that DPS. While the magnitude of the threats discussed below have not changed substantially from our consideration of them in the 2019 Revised Proposed Rule, what has changed in this analysis is the consideration of their distribution across the ranges of the two separate DPSs, as opposed to applying an analysis for a singular West Coast DPS, and then how the impact of those threats affects each separate DPS where they occur. This final determination represents a change to that 2019 Revised Proposed Rule. We now add the SSN DPS as an endangered species to the List of Endangered and Threatened Wildlife, and we present our finding that the NCSO DPS does not warrant listing under the Act. As detailed below in the General Threat Information section and the specific threats discussions for each DPS, these final determinations are based on the best scientific and commercial data available, including new information received in response to the 2019 Revised Proposed Rule. Further, we have clearly articulated the rationales for our conclusions.

Distinct Population Segment Analysis

Under section 3(16) of the Act, we may consider for listing any species, including subspecies, of fish, wildlife, or plants, or any DPS of vertebrate fish or wildlife that interbreeds when mature (16 U.S.C. 1532(16)). Such entities are considered eligible for listing under the Act (and, therefore, are referred to as listable entities), should we determine that they meet the definition of an endangered or threatened species.

Under the Service's DPS Policy (61 FR 4722, February 7, 1996), three elements are considered in the decision concerning the determination and classification of a possible DPS as threatened or endangered. These elements include:

(1) The discreteness of a population in relation to the remainder of the species to which it belongs;

(2) The significance of the population segment to the species to which it belongs; and

(3) The population segment's conservation status in relation to the Act's standards for listing, delisting, or reclassification (

i.e.,

is the population segment endangered or threatened).

A population segment of a vertebrate taxon may be considered discrete under the DPS policy if it satisfies either one of the following conditions:

(1) It is markedly separated from other populations of the same taxon as a consequence of physical, physiological, ecological, or behavioral factors. Quantitative measures of genetic or morphological discontinuity may provide evidence of this separation.

(2) It is delimited by international governmental boundaries within which differences in control of exploitation, management of habitat, conservation status, or regulatory mechanisms exist that are significant in light of section 4(a)(1)(D) of the Act.

If a population segment is considered discrete under one or more of the conditions described in the Service's DPS policy, its biological and ecological significance will be considered in light of Congressional guidance that the authority to list DPSs be used “sparingly” (see Senate Report 151, 96th Congress, 1st Session). In making this determination, we consider available scientific evidence of the DPS's importance to the taxon to which it belongs. Since precise circumstances are likely to vary considerably from case to case, the DPS policy does not describe all the classes of information that might be used in determining the biological and ecological importance of a discrete population. However, the DPS policy describes four possible classes of information that provide evidence of a population segment's biological and ecological importance to the taxon to which it belongs. As specified in the DPS policy, this consideration of the population segment's significance may include, but is not limited to, the following:

(1) Persistence of the DPS in an ecological setting unusual or unique to the taxon;

(2) Evidence that loss of the DPS would result in a significant gap in the range of a taxon;

(3) Evidence that the DPS represents the only surviving natural occurrence of a taxon that may be more abundant elsewhere as an introduced population outside its historical range; or

(4) Evidence that the DPS differs markedly from other populations of the species in its genetic characteristics.

To be considered significant, a population segment needs to satisfy only one of these criteria, or other classes of information that might bear on the biological and ecological importance of a discrete population

segment, as described in the DPS policy. Below, we summarize discreteness and significance for each of the DPSs.

Northern California/Southern Oregon DPS of Fisher (NCSO DPS)

Discreteness

The NCSO DPS is markedly separate from other North American fisher populations to the east by enormous distances, geographical barriers, unsuitable habitat, and urban development. Fishers in this DPS are separated from the Rocky Mountains and the rest of the fisher taxon in the central and eastern United States by natural physical barriers including the non-forested high desert areas of the Great Basin in Nevada and eastern Oregon. Other physical barriers that separate the NCSO DPS from Rocky Mountain and eastern United States fisher populations include large areas without forests, including urban and rural open-canopied areas, agricultural development, and other non-forested areas.

The NCSO DPS is also markedly separate from fisher populations to the north by approximately 560 miles (mi) (900 kilometers (km)) (to the current populations of fishers in Canada) and 270 mi (430 km) (to the reintroduced fisher populations in Washington). These distances are well beyond the various reported fisher dispersal distances (as described in more detail in Service 2016, pp. 13-14). An additional component contributing to marked separation between the NCSO DPS and fishers in Washington is the Columbia River and adjacent human developments (

e.g.,

roads and towns); these likely act as a physical impediment to crossing by fishers dispersing in either direction. While juvenile fishers dispersing from natal areas are capable of moving long distances and navigating various landscape features such as highways, rivers, and rural communities to establish their own home range (Service 2016, pp. 13-14), the magnitude of these impediments and the distance between the NCSO DPS and Washington State fishers would preclude this possibility. Therefore, it is extremely unlikely that any transient individuals from the NCSO DPS could disperse far enough to reach the Washington range of reintroduced fishers, and even if they attempted to do so, they would likely not be able to cross the Columbia River. Not only is the river especially wide and deep year-round, but in the Cascade Range, it is bordered on one side by an interstate highway, a two-lane State highway on the other side, as well as a railroad track on both sides. These impediments further restrict the ability of fishers to surpass this obstacle.

In addition, the NCSO DPS is also markedly separate from the SSN DPS to the southeast by approximately 130 mi (209 km) from the southern end of the NCSO DPS to the northern end of the SSN DPS. This distance, although less than that between the NCSO DPS and Washington fishers, is still several times beyond the known maximum dispersal distances for fishers (Zielinski et al. 2005, p. 1402). The intervening habitat between the NCSO DPS and SSN DPS is additionally characterized by habitat that is highly altered with reduced forest density and increased human development of the landscape further limiting potential fisher dispersal across this region (Zielinski et al. 2005, p. 1,403).

In summary, the NCSO DPS is geographically isolated from all other populations of the species. Therefore, the marked separation condition for discreteness is met by geographical barriers, urban development, unsuitable habitat, and distances that are beyond the known dispersal distance of fishers.

Significance

For the NCSO DPS, we found that a combination of several of the criteria listed above provide evidence of its biological and ecological importance to the taxon. First, we note that the NCSO DPS represents a large portion of the taxon's range along the Pacific coast, and its loss would leave a significant gap between the SSN DPS and all fisher populations to the north. While we recognize that the NCSO DPS is geographically separated from other fisher populations, and this separation likely precludes the NCSO DPS from ever acting as a connection for a contiguous range of fishers from the SSN DPS to Canada, we note that its loss would still result in an even greater break in the west coast range of fishers than what currently exists. Furthermore, the NCSO DPS supports thousands of individuals, while the SSN supports just a few hundred, and populations in Washington are still small. Therefore, a loss of the NCSO DPS would mean the majority of the fishers in the West Coast States would be lost.

Significance is also demonstrated by the NCSO DPS's marked difference from other populations of the species in their genetic characteristics. The NCSO DPS is primarily composed of fishers native to this region of the country and which are genetically distinct from fishers in the remainder of North America (for example, Canada, Rocky Mountains, and Great Lakes). In addition, fishers in the NCSO DPS are also genetically distinct from those found in the SSN DPS, as we describe in Service 2016 (pp. 134-135). We note the NCSO DPS does include the translocated SOC subpopulation, which was established with fishers not native to this region (

i.e.,

British Columbia and Minnesota) and which do not share all the same genetic characteristics of the native fishers. However, it is highly unlikely that the unique genetic characteristics that have evolved over time as native fishers in the NCSO DPS have adapted to the environmental conditions of this area will be lost as a result of this very limited introduction of genes from fishers not indigenous to this region. Although there is interbreeding between SOC and indigenous fishers, we base our conclusion on the fact that SOC fishers do not appear to have expanded their range far from their original reintroduction area since their translocation over 40 years ago (Barry 2018, p. 23). We therefore conclude that the loss of fishers in the NCSO DPS would result in a reduction of the species' overall genetic diversity.

In light of the above, we conclude that the NCSO DPS is significant to the fisher taxon.

Summary

Given that both the discreteness and the significance elements of the DPS policy are met for fisher in the Northern California/Southern Oregon portion of its range, we find that the NCSO DPS of fisher is a valid DPS. Therefore, the NCSO DPS of fisher is a species under the Act.

Southern Sierra Nevada DPS of Fisher (SSN DPS)

Discreteness

Similar to the NCSO DPS, the SSN DPS is markedly separate from other North American fisher populations to the east by enormous distances, geographical barriers, unsuitable habitat, and urban development. Fishers in this DPS are separated from the Rocky Mountains and the rest of the taxon in the central and eastern United States by natural physical barriers including the non-forested high desert areas of the Great Basin in Nevada and eastern Oregon. Other physical barriers that separate the SSN DPS from Rocky Mountain and eastern United States fisher populations include large areas of unsuitable habitat such as urban and rural open-canopied areas, agricultural development, and other non-forested areas.

As noted above, the SSN DPS is markedly separate from the NCSO DPS

by approximately 130 mi (209 km). The intervening habitat between the NCSO DPS and SSN DPS is highly altered with reduced forest density and increased human development of the landscape, further limiting potential fisher dispersal across this region (Zielinski et al. 2005, p. 1,403). In addition, the SSN DPS is also considerably farther away from the Washington State and Canada fisher populations than the NCSO DPS, clearly meeting the marked separation condition of discreteness.

In summary, the SSN DPS is geographically isolated from all other populations of the species. Therefore, the marked separation condition for discreteness is met by geographical barriers, urban development, unsuitable habitat, and distances that are beyond the known dispersal distance of fishers.

Significance

For the SSN DPS, we also found that a combination of the criteria listed above provides evidence of the biological and ecological importance to the fisher taxon. First, we note that the SSN DPS represents the southernmost periphery of the taxon's range. Loss of the SSN DPS would shift representation of the taxon at its southern boundary approximately 400 miles northward to the range of the NCSO DPS.

We also note that the SSN DPS differs markedly from other populations of the species in its genetic characteristics. The SSN DPS is wholly composed of fishers native to this region of the country, and these fishers are genetically distinct from fishers in the remainder of North America (for example, Canada, Rocky Mountains, and Great Lakes). In addition, fishers in the SSN DPS are also genetically distinct from those found in the NCSO DPS. There is high genetic divergence between the SSN DPS and NCSO DPS with the populations being separated for thousands of years (Tucker et al. 2014, p. 3). The SSN DPS has only a single mitochondrial DNA haplotype, which is genealogically unique from the rest of the fisher taxon, including the NCSO DPS (Knaus et al. 2011, pp. 7, 11; Tucker 2019, pers. comm.). In addition, the SSN DPS has a unique distribution of alleles in comparison to the NCSO DPS (Tucker et al. 2012, p. 6). We therefore conclude that the loss of fishers in the SSN DPS would result in a reduction of the species' overall genetic diversity.

In light of the above, we conclude that the SSN DPS is significant to the fisher taxon.

Summary

Given that both the discreteness and the significance elements of the DPS policy are met for fisher in the Southern Sierra Nevada portion of its range, we find that the SSN DPS of fisher is a valid DPS. Therefore, the SSN DPS of fisher is a species under the Act.

Background

General Species Information

Species Information and Distribution

The fisher is a medium-sized, light brown to dark blackish-brown mammal found only in North America, with the face, neck, and shoulders sometimes being slightly gray, and the chest and underside often having irregular white patches. The fisher is classified in the order Carnivora, family Mustelidae, which is a family that also includes weasels, mink, martens, and otters (Service 2016, p. 8). The occurrence of fishers at regional scales is consistently associated with low- to mid-elevation coniferous and mixed conifer and hardwood forests with characteristics of mid- and late-successional forests (

e.g.,

diverse successional stages, moderate to dense forest canopies, large-diameter trees, coarse downed wood, and singular features of large snags, tree cavities, or deformed trees). Throughout their range, fishers are obligate users of tree or snag cavities for denning, and they select denning and resting sites with a high proportion of characteristics associated with late-successional forests, such as snags, down wood, and vertical and horizontal diversity. These characteristics are maintained and recruited in the forest through ecological processes such as fire, insect-related tree mortality, disease, and decay (

e.g.,

Service 2016, pp. 64, 123-124).

Fishers on the west coast of the continent have historically occurred in British Columbia, Washington, Oregon, and California. Fishers indigenous to the west coast in the contiguous United States were historically well distributed in the habitats described above, from the State of Washington south through Oregon, and into northern California and the Sierra Nevada mountains. Subpopulations of these indigenous fishers still occur in northern California/southwestern Oregon and the Sierra Nevada; however, populations of indigenous fishers were extirpated from Washington (Lewis and Hayes 2004, p. 1) and northern Oregon (Aubry and Lewis 2003, pp. 81-82). Recent surveys in the northern Oregon Cascades yielded no fishers (Moriarty et al. 2016, entire), suggesting they remain absent in this area, whereas surveys in the southern Oregon Cascades suggest fishers in this locale may be shifting to the south (Barry 2018, pp. 22-23) compared to their distribution in the late 1990s (Service 2014 and 2016, entire, though see current condition section for NCSO). Fishers in the southern Oregon Cascades were translocated from British Columbia and Minnesota circa 1980. In addition, a translocation of fishers from northwestern California to the northern Sierra Nevada (

i.e.,

NSN) occurred in 2009.

Fishers now occurring and reproducing in Washington were established using fishers translocated from outside this three-State region. Fishers from British Columbia were reintroduced to the Olympic Peninsula from 2008 to 2010 (Happe et al. 2017, p. viii; Happe et al. 2020, p. 345) and to the Washington Cascade Range south of Mt. Rainier from 2015 to 2017 (Lewis et al. 2018, p. 5). Reproduction has been documented in both areas. Beginning in 2018, fishers from Alberta were released in the northern Washington Cascades in North Cascades National Park; all animal translocations are expected to be completed in 2020 (Hayes and Lewis 2006, p. 35; Lewis et al. 2019, pp. 19-20).

Fishers were once well distributed throughout their historical range in the habitats described above. In Oregon and California, outside of the existing NCSO DPS and SSN DPS (see Figure 1, above), fishers are considered likely extirpated, though occasional sightings, verifiable and unverifiable, are reported. Additionally, in California, recent survey efforts have not detected fishers south of the reintroduced NSN subpopulation or north of the SSN DPS.

Additional information on the species' biology and distribution is described in the final Species Report (Service 2016, pp. 9-12, 25-53).

General Threat Information

Section 4 of the Act (16 U.S.C. 1533) and its implementing regulations (50 CFR part 424) set forth the procedures for determining whether a species is an “endangered species” or a “threatened species.” The Act defines an endangered species as a species that is “in danger of extinction throughout all or a significant portion of its range,” and a threatened species as a species that is “likely to become an endangered species within the foreseeable future throughout all or a significant portion of its range.” The Act requires that we determine whether any species is an “endangered species” or a “threatened species” because of any of the following

factors: (A) The present or threatened destruction, modification, or curtailment of its habitat or range; (B) Overutilization for commercial, recreational, scientific, or educational purposes; (C) Disease or predation; (D) The inadequacy of existing regulatory mechanisms; or (E) Other natural or manmade factors affecting its continued existence. These factors represent broad categories of natural or human-caused actions or conditions that could have an effect on a species' continued existence. In evaluating these actions and conditions, we look for those that may have a negative effect on individuals of the species, as well as other actions or conditions that may ameliorate any negative effects or may have positive effects.

We use the term “threat” to refer in general to actions or conditions that are known to or are reasonably likely to negatively affect individuals of a species. The term “threat” includes actions or conditions that have a direct impact on individuals (direct impacts), as well as those that affect individuals through alteration of their habitat or required resources (stressors). The term “threat” may encompass—either together or separately—the source of the action or condition or the action or condition itself.

However, the mere identification of any threat(s) does not necessarily mean that the species meets the statutory definition of an “endangered species” or a “threatened species.” In determining whether a species meets either definition, we must evaluate all identified threats by considering the expected response by the species, and the effects of the threats—in light of those actions and conditions that will ameliorate the threats—on an individual, population, and species level. We evaluate each threat and its expected effects on the species, and then analyze the cumulative effect of all of the threats on the species as a whole. We also consider the cumulative effect of the threats in light of those actions and conditions that will have positive effects on the species—such as any existing regulatory mechanisms or conservation efforts. The Secretary determines whether the species meets the definition of an “endangered species” or a “threatened species” only after conducting this cumulative analysis and describing the expected effect on the species now and in the foreseeable future. In our determination, we correlate the threats acting on the species to the factors in section 4(a)(1) of the Act.

Potential threats currently acting upon both the NCSO DPS and SSN DPS, or likely to affect them in the future, are evaluated and addressed in the final Species Report (Service 2016, pp. 53-162). The term “foreseeable future” extends only so far into the future as the Service can reasonably determine that both the future threats and the species' response to those threats are likely (50 CFR 424.11(d)). For fisher, in determining the foreseeable future, the immediacy of each threat was assessed independently based upon the nature of the threat and time period that we can be reasonably certain the threat is acting on fisher populations or their habitat. In general, we considered that the trajectories of the threats acting on fisher subpopulations across the DPS's range could be reasonably anticipated over the next 35-40 years. The reader is directed to the Species Report (Service 2016, entire) for a more detailed discussion of the threats summarized in this document (

http://www.fws.gov/cno/fisher/

). However, please note that our most recent consideration of new data since 2016 (including comments and information received during the two comment periods associated with the 2019 Revised Proposed Rule) coupled with our reevaluation of the entirety of the best available scientific and commercial information is represented and summarized in the various analyses below.

Our analyses below represent an evaluation of the biological status of the two DPSs, based upon our assessment of the effects anticipated for the identified threats, consideration of the cumulative impact of all effects anticipated from the identified threats, and how that cumulative impact may affect each DPS's continued existence currently and in the future. We used the best available scientific and commercial data, and the expert opinions of the analysis team members. The threats identified as having the potential to act upon both DPSs include: habitat-based threats, including high-severity wildfire, wildfire suppression activities, and post-fire management actions; climate change; tree mortality from drought, disease, and insect infestation; vegetation management; and human development (Factor A). We also evaluated potential threats related to direct mortality of fishers including trapping and incidental capture (Factor B), research activities (Factor B), disease or predation (Factor C), collision with vehicles (Factor E), exposure to toxicants (Factor E), and potential effects associated with small population size (Factor E). Finally, we evaluated the inadequacy of existing regulatory mechanisms (Factor D).

As we conducted our threats analyses, we began under the premise that those with the greatest potential to become significant drivers of the future status of both DPSs were: Wildfire and wildfire suppression; tree mortality from drought, disease, and insect infestation; the potential for climate change to exacerbate wildfire and tree mortality; threats related to vegetation management; and exposure to toxicants. Upon determining that the previous singular West Coast DPS configuration should instead be two separate DPSs, we then also modified our premise regarding threats with the potential to become significant drivers of status, and added to the above list of threats: The potential for effects from small population size; disease or predation; and collision with vehicles. While our assessment of the status of each DPS was based on analysis of all identified threats acting upon them, including the cumulative effects of those threats, we are only presenting our detailed analyses on these specific, potentially significant threat drivers common to both DPSs for the purposes of this rulemaking. We refer the reader to the Species Report (Service 2016, entire) for full detailed analyses of all the other individual threats.

As these potentially significant threat drivers were relevant to both DPSs, much of the fundamental information pertaining to the threats was also applicable to both DPS analyses. Although the ultimate conclusion about the significance of each threat varied between the DPSs, below we present scientific information about these threats common to both DPSs, followed by DPS-specific evaluations.

Wildfire and Wildfire Suppression

Our evaluation includes both the effects of wildfire on fisher habitat as well as those activities associated with wildfire suppression that may result in changes to fisher habitat (for example, backburning, fuel breaks, and snag removal). Naturally occurring fire regimes vary widely within the range of both the NCSO DPS and SSN DPS (Service 2014, p. 58), and fisher habitat has been burned across a spectrum from low- to high-severity.

Mixed-severity wildfire includes patches of low-severity wildfire and patches of high-severity wildfire (Jain et al. 2012, p. 47). At the landscape scale, mixed-severity wildfire effects to fisher habitat may affect an area's ability to support fishers for only a short period of time due to the patchy nature of burned and unburned areas. Additionally, a beneficial aspect of mixed-severity wildfires (as opposed to

just high-severity wildfires) is that these wildfires may contribute to the regeneration of the hardwood component of mixed-conifer forest used by fisher (Cocking et al. 2012, 2014, entire). Further these types of fires can sustain patches of unburned refugia that are important for maintaining patches of higher canopy cover, acting as a source for future tree regeneration, and providing habitat for fisher (Blomdahl et al. 2019, p. 1,049). Mixed-severity wildfire may reduce some elements of fisher habitat temporarily, but also helps to contribute to the ecological processes necessary to create tree cavities and other decay and structural abnormalities essential for denning and resting fishers (Weir et al. 2012, pp. 237-238). Low-severity wildfire is unlikely to remove habitat, and post-wildfire areas that burned at low-severity are likely still used by fishers (Naney et al. 2012, p. 6; Truex and Zielinski 2013, p. 90).

The potential for large, high-severity wildfires to affect fisher habitat and fisher populations is concentrated in northern California-southwestern Oregon and the Sierra Nevada areas as compared to the remainder of the fisher's historical range in the West Coast States (Service 2014, pp. 62-63). In general, high-severity wildfire can alter fisher habitat by removing forest canopy, large trees, and structurally diverse understories, which can take from decades to a century or more to regrow (Service 2014, pp. 59-60), but it may also provide foraging opportunities for fishers since these post-fire areas are often abundant with small mammals that fishers eat (Hanson 2013, p. 27; Service 2016, p. 66). For example, there is evidence of fishers associated with high-severity burned areas, or a mix of moderate- and high-severity burns (Service 2016, p. 66), particularly if the area was structurally complex prior to the fire (Hanson 2013, p. 28). However, another study found fishers avoiding areas of high- and moderate-severity fire (Thompson et al. 2019a, p. 15), so there is likely a threshold in high-severity patch size that influences fisher use of these areas (also see individual DPS sections).

Within shrub, grassland, and forested lands across the western United States (including the Sierra Nevada, southern Cascades, and Coast ranges), the wildfire season length increased over each of the last four decades, from 65 days in the 1970s to 140 days in the 2000s (Westerling 2016, pp. 3, 8, 10). The lengthening of the wildfire season is largely due to declining mountain snowpack and earlier spring snowmelt, which contributes to a decrease in vegetation moisture; this scenario causes wildfires to be more frequent and larger with an overall increase in the total area burned (Westerling 2016, pp. 8-9). Throughout the western United States there has been an increase in the patch size and total area of fires in recent decades. The evidence for an increasing area of high-severity fire is mixed given that studies present different historical levels of high-severity fire (Mallek et al. 2013, pp. 11-17; Stephens et al. 2015, pp. 12-16; Hanson and Odion 2016, pp. 12-17; Odion et al. 2016, entire; see Spies et al. 2018, p. 140 for summary of recent literature), but the scientific consensus accepts that mixed conifer forests were characterized by areas burned at low-, moderate-, and high-severity, with higher proportions of low-severity than is currently observed (Safford and Stevens 2017, p. 50). Given projected changes in climate, forests are expected to become more vulnerable to wildfires over the coming century.

Recent publications on wildfire occurrence and severity within the NCSO DPS and SSN DPS continue to support our conclusions that fire is likely to have a negative impact on fisher populations but will depend on fire size, burn severity, and proximity to occupied habitat (79 FR 60419, at 60429, October 7, 2014). Recent information on fishers' behavioral and localized population response to wildfires is available and discussed below in the NCSO DPS and SSN DPS specific discussions.

Climate Change

Overall, fisher habitat is likely to be affected by changing climate conditions, but the severity will vary, potentially greatly, among different regions, with effects to fishers ranging from negative, neutral, or potentially beneficial. Climate throughout the West Coast States is projected to become warmer over the next century, and in particular, summers will be hotter and drier, with heat waves that are more frequent (Hayhoe et al. 2004, p. 12,423; Tebaldi et al. 2006, pp. 191-200; Mote and Salathé 2010, p. 41; Salathé et al. 2010, p. 69; Cayan et al. 2012, pp. 4, 10; Mote et al. 2013, p. 34; Pierce et al. 2013, pp. 844, 848; Ackerly et al. 2018, pp. 6-8; Bedsworth et al. 2018, pp. 23, 26, 30; Dettinger et al. 2018, p. 5; Grantham 2018, p. 6).

In Oregon, Dalton et al. (2017, pp. 4, 8) evaluated greenhouse gas emissions via global climate models with future emission pathways called “representative concentration pathways” (RCPs). They considered multiple greenhouse gas emission scenarios, including both RCP 4.5 and RCP 8.5. Their analysis indicates that extreme heat events are expected to increase in frequency, duration, and intensity by the 2050s due to warming temperatures (RCP 4.5 = mean annual temperature increase predicted on average 3.6 degrees Fahrenheit (°F) (2.0 degrees Celsius (°C)); RCP 8.5 = mean annual temperature increase predicted on average 5.0 °F (2.8 °C). Summers are expected to warm more than the annual average and will likely become drier. Annual precipitation is projected to increase slightly, although with a high degree of uncertainty. Extreme heat and precipitation events are expected to increase in frequency, duration, and intensity.

In California, information from Pierce et al. (2013) and Safford et al. (2012) used multiple general circulation models and downscaling with regional climate models to develop probabilistic projections of temperature and precipitation changes over California by the 2060s. Predictions indicate an annual mean temperature increase of 4.3 °F (2.4 °C) by 2060 (Pierce et al. 2013, p. 844). Similarly, and more recently, Bedsworth et al. (2018, entire) summarizes 44 technical peer-reviewed reports to provide a California-wide climate change assessment. Under two modeled scenarios, average temperatures are projected to increase by 2.5 to 2.7 °F (1.4 to 1.5 °C) in the early century (2006 to 2039) and 4.4 to 5.8 °F (2.4 to 3.2 °C) in the mid-century (2040 to 2069) (Bedsworth et al. 2018, p. 23). Precipitation models suggest that northern California may become wetter, while most southern parts of California will become drier (Bedsworth et al. 2018, p. 25). The authors caution that “due to large annual variation, changes in annual mean or long-term precipitation are not the best metrics to understand” the effects to changes in precipitation in California (Bedsworth et al. 2018, p. 25). Specifically, the models project less overall precipitation with more extreme daily precipitation, inter-annual precipitation will be more erratic, and the number of dry years will increase (Bedsworth et al. 2018, p. 25 citing others; Polade et al. 2017, p. 1).

Higher temperatures during spring and summer, coupled with early snow melt, will reduce the moisture of both live fuels and dead surface fuels by increasing evaporative demands during the dry season and lengthening the fire season (Keeley and Syphard 2016, pp. 2-3; Restaino and Safford 2018, p. 500). In addition, models project an increase in lightning frequency that may be associated with an increase in potential fire ignitions (Restaino and Safford 2018, p. 500).

Studies specific to predicting the effects of climate change on suitable fisher habitat have produced a wide range of results. Ecotype conversion from conifer forest to woodland, shrubland, or grassland will result in the loss of suitable fisher habitat. This type of shift is predicted, for example, in the southern Sierra Nevada (Gonzalez et al. 2010, Figure 3; Lawler et al. 2012, p. 388; Dettinger et al. 2018, pp. 31-34; Restaino and Safford 2018, p. 500). On the other hand, shifts from conifer forest to hardwood-dominated mixed forest in the southern Sierra Nevada or Klamath region could either increase or decrease the habitat available to fishers (Lawler et al. 2012, pp. 384-386; Loarie et al. 2008, p. 4 and Figure 4). Given the more significant contribution of hardwood trees to fisher habitat in the drier parts of both the NCSO DPS and SSN DPS, a shift to increasing hardwoods in more coastal or higher elevation forest types could improve habitat, but shifts to hardwood-dominated stands may also reduce protective cover from rain and snowfall (Suffice et al. 2019, pp. 10, 11, 13). Nevertheless, trees are long-lived and mature forests can persist under suboptimal conditions, and these factors can prevent better-suited vegetation from becoming established until disturbance removes the original forest (Sheehan et al. 2015, p. 27). Consequently, the increase in the hardwood component of fisher habitat in predominantly conifer areas may not occur until after fires have changed the composition of the existing stand to allow hardwood establishment. All of these circumstances add to the uncertainty associated with climate change and how it relates to fisher.

Other studies suggest that climate change will adversely impact forest habitat by intensifying large-scale, high-severity wildfire, drought, and tree mortality (Kadir et al. 2013, pp. 132, 137; Westerling 2016, pp. 1-2; Westerling 2018, pp. 21-23; Bedsworth et al. 2018, p. 64; Dettinger et al. 2018, pp. 28-29; Stephens et al. 2018a, p. 77; Stephens et al. 2018b, p. 162; Restaino and Safford 2018, pp. 493-505). A wide range of assumptions and caveats typically accompanies these types of predictions. For example, fire modeling shows a decline in future (approximately 100 years) fire intensities after the existing woody vegetation is burned (Restaino and Safford 2018, p. 499), but it is uncertain if the resulting vegetation and composition will be suitable for fisher.

Variables predicting fisher resting habitat as described by Zielinski and Gray 2018 (p. 903) include stand characteristics such as high canopy closure, large basal area of conifer and hardwood trees, and diameter and age of dominant conifers. To date, climate change has not significantly affected resting habitat for fishers, which, according to Zielinski and Gray (2018, pp. 899, 903), has remained stable over the past 20 years across the California-portion of the range, although habitat suitability tends to be lower on private lands than public lands. However, when considering resting habitat trends over these 20 years to determine potential future resting habitat conditions in light of climate change projections, data from the Sierra National Forest (within a portion of the SSN DPS) indicates the beginning of a negative trend in resting habitat suitability (Zielinski and Gray 2018, p. 903), whereas resting habitat examined within the NCSO DPS varied greatly (

i.e.,

suitable resting habitat decreased in the Shasta-Trinity National Forest, increased in the Six Rivers National Forest, and remained unchanged over time for both the Klamath and Mendocino National Forests).

In addition to the potential climate change effects to fisher habitat discussed above, some researchers have suggested climate change may cause direct effects to fishers, including increased mortality, decreased reproductive rates, alterations in behavioral patterns, and range shifts. Fishers may be especially sensitive, physiologically, to warming summer temperatures (Zielinski et al. 2004, p. 488; Slauson et al. 2009, p. 27; Facka 2013, pers. comm.; Powell 2013, pers. comm.). As a result, researchers (

e.g.,

Burns et al. 2003, Zielinski et al. 2004, Lawler et al. 2012, Olson et al. 2014) theorize that fishers likely will either alter their use of microhabitats or shift their range northward and upslope, in order to avoid the thermal stress associated with increased summer temperatures. Preliminary research on fisher occupancy and climate begins to support these theories. For example, during a drought in central and southern California from 2012 to 2015, fisher utilized higher elevation areas that were otherwise inaccessible due to snowpack during other years (Tucker 2019, pers. comm.). Although fisher occur across a wide range of precipitation levels and minimum temperatures, and appear able to utilize higher elevations in years with less snowpack, it is unknown how the interaction of vegetation, fire regimes, and competition with other species will influence future fisher occupancy patterns in a changing climate (Zielinski et al. 2017, pp. 542-543).

The best available information indicates there is a link between changing climate conditions and the resulting changes to overall habitat suitability and availability for fishers throughout their range. There is also a link between changing climate conditions and the potential to increase fisher stress levels when habitat changes occur. More specifically, these changes affect the amount and distribution of habitat necessary for female fishers to be able to have places to den and raise their young. We provide three examples below.

First, ongoing climate change in California is likely to result in significant or amplified wildfire activity, with the area burned and fire severity likely to increase (Hurteau et al. 2019, pp. 1, 3; Moritz et al. 2018, p. 36). This in turn can result in reduced denning habitat availability for fishers (

e.g.,

Sheehan et al. 2015, pp. 20-22; Dalton et al. 2017, p. 46).

Second, under modeled increases in drought conditions, tree mortality and large-scale high-severity wildfire are likely to increase in frequency, size, and severity, especially if fuel loads in forests are not decreased (Young et al. 2017, p. 78; Westerling and Bryant 2008, pp. S244-S248; Abatzoglou and Williams 2016, pp. 11,770, 11,773; Bedsworth et al. 2018, pp. 29-30; Larvie et al. 2019, p. 1; Westerling 2018, pp. 21-23). Some models suggest that fire severity may be independent from fire intensity; thus, a lower-intensity fire could kill more trees if they are also experiencing a severe drought (Restaino and Safford 2018, p. 500). Although we can expect that seasonal summer dryness may prolong future droughts, it is unknown whether droughts in the future will be worse than our worst droughts in the past (Keeley and Syphard 2016, p. 6; Bedsworth et al. 2018, pp. 26, 57). Regardless, it appears that climate change is intensifying the effects of drought, given that changing climate conditions are estimated to have contributed 5 to 18 percent to the severity of one of the worst recent droughts in 20th-century California history (Williams et al. 2015, p. 6,819; Keeley and Syphard 2016, p. 6). The combination of drought and wildfire can result in loss of adequate forest-canopy cover and individual trees that provide habitat suitable for denning female fishers (

e.g.,

CBI 2019a, p. 9).

Third, the observed increases in wildfire activity in Oregon and California are partially due to climate change; increasing wildfire activity is expected under future warming, which in turn can increase tree mortality from disease and insects like mountain pine

beetles (Dalton et al. 2017, p. 46; Bedsworth et al. 2018, p. 64). Widespread tree mortality (climate related or not) is likely to result in fishers experiencing reduced fitness (

e.g.,

a positive relationship between higher amounts of tree mortality and higher cortisol levels in fishers; Kordosky 2019, pp. 14, 36) and an overall reduction in forest-stand conditions suitable for denning (CBI 2019a, entire; Green et al. 2019a, pp. 3-4). Most forests will experience some form of climate stress by the late 21st century and higher temperatures will result in more droughts in California, revealing the interconnected nature of climate, wildfire, and tree mortality that collectively can shift forest composition and structure (Larvie et al. 2019, pp. 12-14; Restaino and Safford 2018, p. 502) and further challenge the ability of fishers to locate suitable habitat.

Tree Mortality From Drought, Disease, and Insect Infestation

In our 2019 Revised Proposed Rule, this section was titled “Forest Insects and Tree Diseases”; we have changed the title to more accurately describe the threat. Localized tree mortality from insect outbreaks and tree diseases are natural processes, and they provide structures used by fisher for rest and den sites as well as their prey. However, widespread insect and disease outbreaks can alter the overall distribution and abundance of fisher habitat. For example, severe drought events in California since 2010, combined with insect outbreaks and tree diseases, have led to more than 147 million dead trees in California (California Department of Forestry and Fire Protections (CAL FIRE) and USFS 2019, no page number). Although both the NCSO DPS and SSN DPS experienced tree mortality during the recent drought, the magnitude of this effect on the landscape differed tremendously between each DPS (CAL FIRE and USFS 2019, no page number). The highest levels of tree mortality occur in the southern Sierra Nevada due to increased susceptibility to forest insects and tree disease from the severe drought while most of the NCSO DPS experienced background levels (0-5 dead trees per acre) of tree mortality (CAL FIRE and USFS 2019, no page number; California Tree Mortality Task Force 2020, entire).

Vegetation Management

Vegetation management techniques of the past (primarily timber harvest) have been implicated as one of the two primary causes for fisher declines across the United States. Many fisher researchers have suggested that the magnitude and intensity of past timber harvest is one of the main reasons fishers have not recovered in the western United States as compared to the northeastern United States (Service 2014, pp. 54-56). At the time of the 2014 Proposed Rule, we stated that vegetation management techniques have, and can, substantially modify the overstory canopy, the numbers and distribution of structural elements available for use by fisher, and the ecological processes that create them. An increase in open areas, such as those resulting from timber harvest, may increase the risk of predation on fishers by bobcats and other predators that frequent these areas (see the Predation and Disease section below). Overall, fisher home ranges comprise mosaics of forest-stand types and seral (stand age) stages but often with a high proportion of mid- to late-seral forests (Raley et al. 2012, p. 231).

Fishers occupy managed landscapes and stands where timber harvest and other vegetation management activities occur; the degree to which fishers tend to be found in these areas often depends on a multitude of factors, including the scale, intensity, and rate of activities; the composition and configuration of suitable habitat; and the amount and type of retained legacy structures (Service 2016, pp. 59-60; Thompson and Clayton 2016, pp. 11-16, 22; Niblett et al. 2017, pp. 14-17; Marcot et al. 2018, p. 400; Powell et al. 2019, entire; Parsons 2018, pp. 31, 53-55, 63; Purcell et al. 2018, pp. 60-61, 69-70). Fishers tolerate some clearcuts in their home ranges, though the mean proportion tends to be below 25 percent of their home-range area (Powell et al. 2019, p. 23). Fishers are also observed denning in areas where as much as 25 percent of the area near the den sites is in openings (Niblett et al. 2017, p. 17). Some level of open areas or younger stands may provide suitable prey for fishers (Parsons 2018, pp. 26-29, 53-55). Yet even in these situations, fishers are associated with forests that contain structures associated with older forests, such as complex canopies, down wood, hardwoods, and trees with microsites conducive to denning, resting, or supporting prey (Niblett et al. 2017, pp. 16-17; Powell et al. 2019, pp. 19-23). Therefore, for vegetation management it is important to maintain decadent structures that serve as den and rest trees and that likely required much time and site-specific conditions to develop (Matthews et al. 2019, p. 1,313). Overall, it appears fishers can tolerate management activities that promote forest heterogeneity (variation) and that consider the natural range of variation in forest structure, distribution, and composition when identifying and protecting valuable habitat elements (Thompson et al. 2019b, pp. 13-14).

While historical loss of mature and older forests via timber harvest through much of the 1900s resulted in a substantial loss of fisher habitat in California and Oregon, harvest volume has sharply declined throughout this area since 1990, primarily on Federal lands, but also on non-Federal lands. Although timber harvest is still ongoing throughout the NCSO and SSN DPSs, habitat ingrowth (

i.e.,

forest stands becoming habitat as a result of forest succession) is also occurring, offsetting some of those losses. We address this for each of the DPSs below.

Exposure to Toxicants

Wildlife can encounter a wide range of chemicals in the environment. Fertilizers and pesticides (

e.g.,

herbicides, insecticides, and rodenticides) are among the most common chemicals wildlife are exposed to and impacted by, especially near urban and agricultural areas. Of these chemicals, the rodenticides are the longest lasting and therefore the easiest to test for, track, and understand impacts to species. Both the draft and final Species Reports detail the exposure of fishers to rodenticides in Oregon and California (Service 2014, pp. 149-166; Service 2016, pp. 141-159).

The rodenticides impacting fishers include first- and second-generation anticoagulant rodenticides and neurotoxicant rodenticides. First-generation anticoagulant rodenticides are in a bait form that rodents consume for several consecutive feedings (

i.e.,

sublethal doses) to deliver a lethal dose. Second-generation rodenticides are significantly more potent than first-generation rodenticides, and a lethal dose can be ingested in a single feeding. Additionally, second-generation rodenticides are more likely than first-generation rodenticides to poison predatory wildlife (

e.g.,

fishers) that eat live or dead poisoned prey because they are more persistent in the environment. Neurotoxicant rodenticides are delivered in either single or multiple doses and have highly variable potency (multiple hours or days). Both first- and second-generation anticoagulant rodenticides as well as neurotoxicant rodenticides are used to kill small mammals that are destroying crops. Rodenticides impair an animal's ability to produce several key blood-clotting factors (anticoagulant rodenticides) or affect brain and liver function

(neurotoxicant rodenticides). Anticoagulant rodenticide exposure causes bleeding from the nose and gums, extensive bruises, anemia, fatigue, difficulty breathing, and also damage to small blood vessels, resulting in spontaneous and widespread hemorrhaging.

A sublethal dose of a rodenticide can produce significant clotting abnormalities and hemorrhaging, leading to a range of symptoms, such as difficulty moving and a decreased ability to recover from physical injury. Ingestion of the neurotoxicant bromethalin, which has been detected in DPS fisher carcasses, has fast-acting and physical effects such as unsteadiness and weakness, and at higher dosage levels, seizures. Both anticoagulant and neurotoxicant rodenticides can change or impede normal fisher movement and foraging behaviors and therefore may increase the probability of mortality from other sources such as predation or vehicle collision. In addition, anticoagulants bioaccumulate and become increasingly prevalent in predators; as they continue to eat contaminated prey, they accumulate more and more anticoagulant (Lopez-Perea and Mateo 2018, p. 165). Contaminated rodents are found within and adjacent to treated areas weeks or months after bait application (Geduhn et al. 2014, pp. 8-9; Tosh et al. 2012, pp. 5-6; Sage et al. 2008, p. 215).

Rodenticide use in agricultural or urban areas is common and wildlife exposure rates can be high. For example, in California 70 percent of tested mammals were positive for at least one anticoagulant rodenticide (Hosea 2000, p. 238). And across the world, 58 percent of tested predators were positive for anti-coagulant rodenticides (Lopez-Perea and Mateo 2018, p. 172). Not surprisingly, mammals are most impacted by rodenticides, when compared to birds, reptiles, and insects; and generalist species that eat a variety of prey species are more likely to be contaminated relative to specialist species that feed on one or a few species (Lopez-Perea and Mateo 2018, pp. 163, 173).

Predators that are (a) nocturnal, (b) opportunistic in feeding habitats where rodents are an important part of their diet, and (c) nonmigratory and live close to or within landscapes that are heavily impacted by human activities are more likely to be exposed to rodenticides and have relatively high liver-residue concentrations of multiple rodenticide compounds (Hindmarch and Elliott 2018, p. 251). Because fishers are territorial, nonmigratory mammals, and females remain particularly tied to their territories (Arthur et al. 1993, p. 872), they are among the species that are more vulnerable to rodenticide exposure. Additionally, fisher diets consist primarily of small mammals (Golightly et al. 2006, entire), which are the target species for rodenticides (Gabriel et al. 2015, entire; Thompson et al. 2014, pp. 97-98). Top predators within the range of fishers, including northern spotted owls (

Strix occidentalis caurina

) and barred owls (

S. varia

), have also been exposed to rodenticides (Franklin et al. 2018, p. 1; Gabriel et al. 2018, p. 1).

Data available since completion of the final Species Report in 2016 continue to document exposure and mortalities to fishers from rodenticides in both the NCSO and SSN DPSs (Gabriel and Wengert 2019, unpublished data, entire; Powell et al. 2019, p. 16). Here we discuss data specific to both the NCSO and SSN DPS; more DPS-specific information is found in the NCSO DPS and SSN DPS discussions below. Fisher carcasses have been collected and tested for their cause of death and their exposure to rodenticides (Gabriel and Wengert 2019, unpublished data). Data for 97 fisher carcasses collected in California in the period 2007-2014 indicate 81 percent of fishers tested positive for one or more rodenticides, and 48 fishers collected from 2015-2018 indicate 83 percent tested positive (Gabriel and Wengert 2019, unpublished data). Using data from both the SSN and the NCSO DPS and comparing the periods 2007-2011 and 2012-2014, mortalities due to rodenticide toxicosis increased from 5.6 to 18.7 percent (Gabriel and Wengert 2019, unpublished data, p. 2). And, from 2015 to 2018, additional fisher mortalities due to both anticoagulant and neurotoxicant rodenticides have been documented, including the toxicosis of neonatal kits in the womb (Gabriel and Wengert 2019, unpublished data, p. 4). The probability of fisher mortality increases with the number of anticoagulant rodenticides a fisher has been exposed to, and most fishers are exposed to more than one (Gabriel et al. 2015, p. 15).

The primary source of rodenticide exposure to fishers is from illegal marijuana grow sites on public, private, and tribal lands in California and Oregon (Gabriel et al. 2015, pp. 14-15; Thompson et al. 2014, pp. 97-98). In the mid- to late 1970s, 90 percent of the marijuana consumed in the United States came from abroad (Brady 2013, pp. 70-71). Marijuana cultivation in California really began in 1974 or 1975, and by 1979, 35 percent of the marijuana consumed in California was from California (Brady 2013, pp. 70-71). By 2010, 79 percent of all the marijuana consumed in the United States came from California (Brady 2013, pp. 70-71).

Information on the amount and types of rodenticides have been collected at more than 300 illegal grow sites in California from 2012 through 2018 (Gabriel and Wengert 2019, unpublished data, pp. 5-7). Through this time period the use of second-generation rodenticides decreased. This is likely because of regulation changes in 2014 that placed additional restrictions on the use of second-generation rodenticides in California (California Department of Pesticide Regulation 2014). The change in policy has led to a more intensive use of first-generation anticoagulant rodenticide and the highest amount of neurotoxicant rodenticide use since 2012 (Gabriel and Wengert 2019, unpublished data, pp. 5-7).

In order to evaluate the risk to fishers from illegal grow sites and any differences between populations, we use a Maximum Entropy model to identify high and moderate likelihood of illegal grow sites being located within habitat selected by fisher in California and Oregon (Gabriel and Wengert 2019, unpublished data, pp. 7-10). This model indicates that 44 percent of the habitat modeled (combined NCSO and SSN DPSs) for fishers is within areas of high and moderate likelihood for illegal grow sites—see also the individual DPS sections below. However, the extent to which the use of toxicants occurs on marijuana grow sites on private land, as well as other agricultural, commercial, and public land sites within the range of the fisher (and habitats that fishers select for), is unknown.

Illegal grow sites are regularly discovered in California (617 from 2012 through 2018, and 2,039 from 2004 through 2018) (Gabriel and Wengert 2019, unpublished data, p. 7). Law-enforcement specialists estimate they locate and raid roughly 20 to 40 percent of sites each year and only about 10 percent of those are remediated (Thompson et al. 2017, p. 45). If these estimates are accurate, it is reasonable to conclude that thousands of illegal grow sites—known and unknown, and with an undetermined amount of toxicants present—remain scattered within both the NCSO DPS and SSN DPS (Gabriel et al. 2015, entire; Thompson et al. 2017, p. 45). Rodenticides persist in the landscape, with first-generation rodenticides having a half-life of up to 16 days and second-generation rodenticides having a half-life up to 307 days (Shore and Coeurdassier 2018, p. 146).

As discussed, both the draft and final Species Reports detail the exposure of fishers to rodenticides (Service 2014, pp. 149-166; Service 2016, pp. 141-159). Below we summarize new information:

(1)

Rodent diversity

—Illegal grow sites that were treated with rodenticides contained only mice, as compared to untreated sites where rodenticides were not used and where large-bodied rodents (

e.g.,

woodrats, squirrels, chipmunks) were found. The absence of larger rodents at treated sites suggests that larger-bodied rodents may be impacted by rodenticides more than smaller bodied rodents. These large-bodied rodents are the prey species fishers prefer (Gabriel et al. 2017, p. 10). Further, illegal grow sites may act as “sinks” for prey moving in from neighboring areas meaning less prey is available for fisher (Gabriel 2018, pers. comm.).

(2)

Law Enforcement Activities

—During the “Operation Forest Watch, Department of Justice” campaign in California between October 2017 and September 2018, more than 20,000 pounds of fertilizer, pesticides, and chemicals were removed from 160 illegal grow sites (Department of Justice (DOJ) 2018, p. 2). Of these, 89 percent were confirmed or strongly suspected to have carbofuran or methamidophos (

i.e.,

insecticides (non-rodenticides) that cause central nervous system dysfunction), up from the previous year's total of 75 percent (DOJ 2018, p. 2). Estimates vary of the number of illegal grow sites that necessitate reclamation of toxicants, but as of 2018, 766 known illegal grow sites are still in need of reclamation (DOJ 2018, p. 2).

(3)

Effect of legalization

—Since the 2014 Proposed Rule, recreational marijuana cultivation and use became legal in Oregon (2015) and California (2016). The data are mixed with respect to how legalization is affecting illegal grows sites on public lands. Some studies find that illegal grow sites on National Forests have decreased in States where marijuana was legalized (Klassen and Anthony 2019, p. 39; Prestemon et al. 2019, p. 1). Conversely, many law-enforcement officials have found no indication that illegal grow sites have decreased with cannabis legalization, and may in fact be increasing, in part due to legalization providing an effective means to launder illegal marijuana (Hughes 2017, entire; Bureau of Cannabis Control California 2018, pp. 28, 30; Sabet 2018, pp. 94-95; Fuller 2019, no page number; Klassen and Anthony 2019, p. 45). Data from fisher monitoring suggests that illegal grow sites are dropping in number but are getting larger (impacting more fisher home ranges) (Gabriel 2018, pers. comm.). And, law-enforcement actions have caused illegal grow sites to disperse further, which makes them more difficult to locate (Gabriel 2018, pers. comm.). Other uncertainties make it difficult to reach conclusions about trends in the abundance and frequency of illegal grow sites this soon after legalization, including legal marijuana market forces, the clandestine nature of the black market, Federal illegality and trends of legalization in other States, State taxation of marijuana, local employment and economic conditions, and regulatory and law enforcement responses (Hughes 2017, entire; Bureau of Cannabis Control California 2018, pp. 28, 30; Sabet 2018, pp. 94-95; Fuller 2019, no page number; Klassen and Anthony 2019, pp. 45-46; Prestemon et al. 2019, pp. 9-11).

Legalization has resulted in an increase in legal marijuana cultivation. At this time, we have limited data about the prevalence of rodenticide use on legal private grow sites and whether fishers are at risk from rodenticide use on private land. In urban-wildland interfaces, or where private lands abut public forestland or occur as inholdings, legal grow sites are more likely within fisher home ranges (

e.g.,

Franklin et al. 2018, p. 3).

(4)

Reclamation Efforts

—Existing law enforcement cannot keep up with illegal marijuana activities (Bureau of Cannabis Control California 2018, p. 30; Wendt 2019, pp. 4-6). In addition, support from States and local governments to Federal law enforcement on public lands (

e.g.,

U.S. Forest Service (USFS)) has dwindled as they redirect resources to regulate the legalized marijuana industry (Bureau of Cannabis Control California 2018, p. 30; Klassen and Anthony 2019, p. 45).

The California Comprehensive Medical Cannabis Regulation and Safety Act of 2016 specifies that, after control and regulation of the program, 20 percent of the marijuana tax fund (established by this Act) shall be given to California Department of Fish and Wildlife (CDFW) for (1) cleanup, remediation, and restoration of environmental damage in watersheds affected by marijuana cultivation (a portion of which may be distributed through grants); and (2) the stewardship and operation of State-owned wildlife habitat areas and State park units to prevent illegal cultivation, and use (Comprehensive Medical Cannabis Regulation and Safety Act 2016, pp. 43-44). This language is not included in the 2017 Medicinal and Adult-Use Cannabis Regulation and Safety Act (MAUCRSA) that updates the 2016 Act (MAUCRSA 2017, entire).

In 2017, CDFW used their Regulation and Forest Restoration funds for their newly formed Cannabis Restoration Grant Program (CDFW 2017a, p. 3). The program funded the restoration of watersheds impacted by marijuana cultivation, including removing trash and equipment, diversion removal, riparian enhancements, and streambank stabilization (CDFW 2017b, p. 1). Funds for projects in 2017 totaled $1,300,000 (CDFW 2017a, p. 1). Monies from this program went to fund four efforts for watersheds within the range of the NCSO DPS (CDFW 2017a, p. 2). The largest and widest-ranging of these efforts included the removal and remediation of rodenticides at illegal grow sites. Monies were not made available in 2018 or 2019, but it is our understanding there are plans to add monies to this grant program in the future.

The CROP Project (Cannabis Removal on Public Lands) is a citizen-based organization established in 2018 with the primary goals of: (1) Securing and increasing State and Federal resources for illegal-grow-site reclamation; (2) increasing U.S. Department of Agriculture (USDA) USFS law enforcement and overall presence on National Forests; and (3) implementing a Statewide public education campaign, focusing on the human health risks associated with ingesting unregulated marijuana (

www.cropproject.org

). Successful accomplishment of these goals could substantially improve the discovery and reclamation of illegal grow sites, but it is too early to determine the degree to which this program reduces the threat of toxicants to fishers.

Please also see Existing Regulatory Mechanisms in both the NCSO and the SSN DPS discussions below for more information on voluntary conservation efforts that address illegal grow sites.

At this time, our evaluation of the best available scientific and commercial information regarding toxicants and their effects on fishers leads us to conclude that individual fishers within both DPSs have died from toxicant exposure, fishers suffer a variety of sublethal effects from exposure to rodenticides, and the potential for illegal grow sites within fisher habitat is high. The exposure rate of more than 80 percent of fisher carcasses tested in California has not declined between 2007 and 2018 (Gabriel and Wengert 2019, unpublished data, pp. 3-4), while poisoning has increased since 2007 (Gabriel et al. 2015, p. 7). We do not know the exposure rate of live fishers to

toxicants since this information is difficult to gather and has not been collected. In addition, the minimum amount of anticoagulant and neurotoxicant rodenticides required for sublethal or lethal poisoning is unknown. Specific information on fishers and toxicants within the NCSO DPS and the SSN DPS is described in the DPS-specific sections below.

Potential for Effects Associated With Small Population Size

Small populations are vulnerable to a rapid decline in their numbers and localized extinction due to the following: (1) Loss of genetic variability (

e.g.,

inbreeding depression, loss of evolutionary flexibility), (2) fluctuations in demographic parameters (

e.g.,

birth and death rates, population growth rates, population density), and (3) environmental stochasticity or random fluctuations in the biological (

e.g.,

predation, competition, disease) and physical environment (

e.g.,

wildfire, drought events, flooding) (Primack 2014, pp. 252-268). We note that forest carnivore populations, including fisher, are often isolated and generally occur in low densities (Service 2016, p. 29). While we do not have data across the entire fisher range on the West Coast demonstrating that fishers are exhibiting specific effects associated with small population size, consideration of these three elements along with life-history traits can provide an extinction-vulnerability profile for both the NCSO DPS and SSN DPS. Fishers in Oregon and California are currently restricted to two historically extant indigenous populations (NCSO and SSN), one extant reintroduced subpopulation (NSN, established with fishers from NCSO), and one subpopulation established with fishers from outside this region (SOC). We recognize the two geographic areas of fisher, SSN and NCSO (the latter of which includes the SOC and NSN for this analysis), are geographically isolated from one another with no evidence of and very little opportunity for genetic interchange. Our evaluation of the best scientific and commercial information available indicates that the separation of the SSN and NCSO populations occurred a very long time ago, possibly on the order of more than a thousand years, pre-European settlement (Tucker et al. 2012, pp. 1, 7; Knaus et al. 2011, p. 11). Despite their isolation and the small size of the SSN DPS, the native NCSO DPS and SSN DPS have persisted over a long period of time.

At this point in time, fishers in both the NCSO DPS and SSN DPS are reduced from their original/historical range within the West Coast States. The best available information suggests these populations are expected to remain isolated from one another (as has been apparent since pre-European settlement). Estimates of fisher population growth rates for the NCSO DPS and the portion of the SSN DPS surveyed do not indicate any overall positive or negative trend (see Current Condition section for the NCSO DPS below), with the exception of the recently reintroduced subpopulation in the NSN, which has steadily grown since its translocation beginning in 2009. The vulnerabilities related to small population size for each DPS are further described below.

Disease and Predation

We evaluated information on disease and predation in our 2016 Species Report (Service 2016, pp. 128-132). In addition, we evaluated the following new information available regarding disease or predation since the time of our 2014 Proposed Rule (

e.g.,

Gabriel et al. 2015, pp. 5-8, 12-16; Sweitzer et al. 2016a, pp. 444-448; Integral Ecology Research Center 2017, p. 2; Barry 2018, pp. 39-40; Green et al. 2018a, p. 549; Purcell et al. 2018, pp. 39-40, 50-51, 53, 72; CDFW 2019, entire). Although we did not identify this threat in the 2019 Revised Proposed Rule as one that may have been a potentially significant driver of future status, we are considering this new information in this Final Rule in light of our DPS determination that has resulted in two separate DPSs; the magnitude and scale of the effect disease or predation may have on each DPS may differ as a result of the DPS-specific demographics and distribution. Predation and disease are the two greatest sources of mortality for fishers of identified mortality sources studied in California (Gabriel et al. 2015, p. 6; Sweitzer et al. 2016a, p. 447). Of 183 California fishers where the mortality source was identified, 67 percent died from predation and 13 percent from a combination of disease, injury, or starvation (Sweitzer et al. 2016a, p. 447). Gabriel et al. (2015, p. 7) was able to separate disease from other mortality sources and found that 15 percent of 136 necropsied fishers died of disease.

Several viral and bacterial diseases are known to affect mustelids, including fishers. Known diseases that have caused fisher mortality in the area of the NCSO and SSN DPSs include canine distemper virus,

Toxoplasma gondii

(a protozoal infection), and several bacterial infections (Gabriel et al. 2015, pp. 7-8; see Service 2016, pp. 128-130 for diseases summary). Disease only has a minor impact where it has been studied in the SSN DPS (Spencer et al. 2015, p. 66), and it comprises a substantially smaller portion of fisher mortalities compared to predation.

We do not know if current predation rates are similar to historical rates in the area of the NCSO DPS and SSN DPS. Comparing predation rates to populations outside of the West Coast is not informative because most of those populations are trapped, skewing the mortality source results (

e.g.,

Lofroth et al. 2010, p. 62, Table 6.3). Recent research in California suggests that landscape changes as a result of disturbances over the past century may have altered the carnivore community and affected predation rates on fishers by bobcats (Wengert 2013, pp. 59-66, 93, 97-100) where an increased proximity to open and brushy areas (vegetation selected for by bobcats) increases the risk of predation on fishers. Mountain lions and bobcats are major predators of fishers. Of 90 fishers that died from predation or were killed by other animals, 90 percent were killed by members of the cat family (Felidae) (Gabriel et al. 2015, p. 5). Sublethal effects of toxicants may also result in higher than normal mortality rates associated with disease and predation, but we do not know what portion of identified mortalities would not have occurred but for the presence of sublethal levels of toxicants in the individual (Gabriel et al. 2015, p. 16; Sweitzer et al. 2016a, p. 448).

Disease and predation are naturally occurring sources of mortality, although the associated mortality rates may be increased by human-caused factors such as vegetation management or toxicants (Gabriel et al. 2015, pp. 14, 16). Predation has been identified as the most important factor limiting fisher populations in California (Sweitzer et al. 2016a, p. 448). High levels of predation may explain why fisher populations have not expanded into unoccupied suitable habitat throughout much of the NCSO and SSN DPSs (Gabriel et al. 2015, p. 16). However, the reintroduced NSN subpopulation appears to be growing despite mortalities due to predation, indicating that other factors such as fisher dispersal distance through unsuitable habitat may also limit fisher expansion (Powell and Zielinski 1994, pp. 60-61; Aubry and Lewis 2003, p. 88) and that reintroductions can play an important role in recovery for the species (Green et al. 2020, p. 13).

Vehicle Collisions

Fisher collisions with vehicles have been documented at multiple locations

within the NCSO DPS and SSN DPS. We summarize this information in the final fisher Species Report (Service 2016, pp. 137-138). Although we did not analyze this threat in the 2019 Revised Proposed Rule, this information warrants consideration in this Final Rule, particularly because we expect this threat to act differently in each of the newly-identified NCSO DPS and SSN DPS based on population size and proximity to human development. In general, fisher collisions with vehicles documented in California are relatively rare, representing less than 2 percent of documented mortalities (Gabriel et al. 2015, p. 15). And, vehicle-related mortalities may be a more local concern associated with specific high-traffic areas (Gabriel et al. 2015, pp. 7 and 15, Table 2).

Existing Regulatory Mechanisms

Many Federal and State existing regulatory mechanisms provide a benefit to fishers and their habitat. For example, trapping restrictions have substantially reduced fisher mortality throughout the NCSO DPS and SSN DPS of fisher. In some places, forest-management practices are explicitly applied to benefit fishers or other species with many similar habitat requirements, such as the northern spotted owl. State and Federal regulatory mechanisms have abated the large-scale loss of fishers to trapping and minimized the loss of fisher habitat, especially on Federal land (Service 2014, pp. 117-141). Additionally, rodenticides are regulated under Federal and State laws. However, fishers are still exposed to rodenticides where they are used (see NCSO and SSN DPS specific sections on Exposure to Toxicants and Existing Regulatory Mechanisms).

Finally, voluntary conservation measures are in place that provide a benefit to fishers and their habitat. These measures include Habitat Conservation Plans (HCPs), Candidate Conservation Agreements with Assurances (CCAAs), Safe Harbor Agreements (SHAs), Memoranda of Understanding (MOUs), and other conservation strategies, as described for each DPS below (see NCSO and SSN DPS specific sections on Voluntary Conservation Measures below).

Final Listing Determination for NCSO DPS of Fisher

Current Condition

The NCSO DPS comprises a mix of ownerships, with similar amounts of private and Federal ownership (Table 1). The USFS is the predominant Federal land manager within the DPS.

Table 1—Land Ownership or Management for the Northern California/Southern Oregon Distinct Population Segment of Fisher

Agency

California (CA)

Acres

(ac)

Percent (%)

for CA

Oregon (OR)

ac

% for OR

NCSO total

ac

%

Bureau of Land Management

864,221

4.0

945,910

17.8

1,810,130

6.8

Forest Service

8,433,567

39.5

2,332,813

43.8

10,766,380

40.4

Bureau of Indian Affairs

211,998

1.0

72

0.0

212,070

0.8

National Park Service

353,235

1.7

186,934

3.5

540,170

2.0

State and Local

473,997

2.2

20,637

0.4

494,635

1.9

Private

10,951,353

51.3

1,824,961

34.3

12,776,315

47.9

Total Acres *

21,346,412

100.0

5,327,797

100.0

26,674,209

100.0

* Acres and % may not sum due to rounding and because some other owners with less land are not included.

Population condition and abundance information for the NCSO DPS is presented for three different geographic portions of this DPS. First, the SOC portion west and south of Crater Lake in the Southern Oregon Cascade Range is predominantly represented by reintroduced individuals from British Columbia and Minnesota. However, recent analyses have documented that at least some of these reintroduced SOC individuals and native NCSO individuals are overlapping in range, with confirmed interbreeding (Pilgrim and Schwartz 2016, entire; Pilgrim and Schwartz 2017, entire). Second, the NSN portion is represented by native, reintroduced fishers whose genetic stock is from fishers relocated from the Klamath-Siskiyou and Shasta-Trinity subregions (in the historically native NCSO DPS). These animals were relocated into the northern Sierra Nevada. This geographic portion of the NCSO DPS occurs on land known as the Sierra Pacific Industries (SPI) Stirling Management Unit in Butte, Plumas, and Tehama Counties, California (Powell et al. 2019, p. 2). Third, the remainder of the native fishers in the NCSO DPS occupy the Klamath-Siskiyou Mountains in southern Oregon and northern California, the California Coast Range Mountains, the Shasta-Trinity subregions in northern California, and the western portion of the southern Cascades in northern California.

Fishers in the SOC portion of the NCSO DPS stem from a translocation of 30 fishers from British Columbia and Minnesota to the southeastern Cascade Range and west of Crater Lake between 1977 and 1981, after an earlier reintroduction in 1961 failed (Aubry and Lewis 2003, p. 84; Lofroth et al. 2010, pp. 43-44). Based on survey and research efforts starting in 1995, genetic evidence shows these fishers continue to persist (Drew et al. 2003, p. 57; Aubry et al. 2004, pp. 211-215; Wisely et al. 2004, p. 646; Pilgrim and Schwartz 2014-2017, entire; Moriarity et al. 2017, entire; Barry 2018, pp. 6, 22-24; Moriarty et al. 2019, p. 23).

Prior to 2015, survey work in the Oregon Cascades north of the NCSO DPS was limited to opportunistic or small-scale efforts. Verifiable fisher detections did not exist, except for two single fishers: One just north of the SOC subpopulation in 2014 (Wolfer 2014, pers. comm.) and a single dispersing juvenile male detected in the same general area in the 1990s (Aubry and Raley 2006, p. 5); this finding suggests occasional individuals may disperse north through the central Oregon Cascades. Over the winter of 2015-2016, systematic camera surveys occurred in the northern Oregon Cascades (specifically, the southern portion of the Mt. Hood National Forest and northern portion of the Willamette National Forest). No fishers were detected (Moriarty et al. 2016, entire), suggesting fishers may not reach this far north in the Oregon Cascades. Additionally,

surveys over the past 3 years have not detected fishers north of the Rogue River in the central Oregon Cascades (Barry 2018, pp. 22-23) (see below).

Information is not available on population size for the SOC portion of the NCSO DPS. In the northern portion of the SOC area, fishers were detected in the northern and eastern portions of Crater Lake National Park between 2013 and 2015 (Mohren 2016, pers. comm.). Outside of the Park, large-scale systematic surveys were conducted in 2016 and 2017 north and west of Crater Lake National Park and south to the Klamath Falls Resource Area (south of the reintroduction area) of the Bureau of Land Management (BLM) Lakeview District (Barry 2018, entire). Few fishers were detected in an area west of Crater Lake National Park where fishers were captured and radio-collared in the early 1990s by Aubry and Raley (2002, entire). Within the Klamath Plateau (generally the Klamath Falls Resource Area described above, but including surrounding non-Federal lands), Moriarty et al. (2019, pp. 5, 21) identified 31 to 41 individuals from 2015 to 2018, concluding that fishers in the SOC area do not appear to be expanding from where they were initially reintroduced.

In comparing this range estimate with a coarse baseline range estimate provided by the Service, Barry (2018, pp. 22-24) determined that there was a 67 percent range reduction for the SOC subpopulation, concluding that SOC fishers “appear to have contracted, shifted south, or the previous population extent was incorrectly estimated” (Barry 2018, pp. 22-24). Given the lack of systematic range-wide fisher monitoring in Oregon, the author, however, urged caution when comparing his analysis with the baseline range estimate provided by the Service, and we agree. Our baseline range estimate used by Barry (2018, p. 31, Figure 3) was derived by encompassing verifiable fisher locations since 1993 in southwest Oregon. Our boundaries were based on modeled fisher habitat and readily identifiable features such the Rogue River. These range maps included scattered, disjunct detections with intervening areas of few to no fisher detections (

e.g.,

see Service 2016, p. 34, Figure 7); consequently, our range map likely encompassed areas with limited fisher occurrence. Hence, comparing our coarse range map with Barry's fisher distribution, which was quantitatively modeled from systematic detection surveys to delineate areas with a higher probability of fisher occurrence, should indeed be interpreted with caution. Our coarse range map certainly included areas with limited numbers or lack of fishers; consequently, a 67 percent range reduction using that map as a baseline comparison overestimates any change in fisher distribution in the SOC subpopulation to some extent. We do concur, however, that SOC fishers seem to have shifted their distribution, and acknowledge that their distribution may be contracting to some degree. Further, we acknowledge Barry's (2018, pp. 22-24) assertion that the SOC subpopulation has had ample time since their reintroduction to colonize beyond the reintroduction area and has failed to do so, suggesting that either our understanding of suitable habitat may be incorrect, there may be unknown barriers limiting their distribution, or other factors may limit this subpopulation.

Barry (2018, p. 23) also concluded that the SOC subpopulation appears small and relatively isolated given the number and spacing of detections. However, there is interbreeding with indigenous fishers near the Klamath Plateau area, suggesting fishers in the southern part of the SOC subpopulation are not isolated.

Fishers in the NSN portion of the NCSO DPS stem from a 2009 to 2011 translocation of 40 fishers (24 females, 16 males) from Humboldt, Siskiyou, and Trinity Counties, California, to the SPI Stirling Management Unit. Ongoing monitoring has confirmed that fishers born onsite have established home ranges and have successfully reproduced. Trapping efforts in the fall of 2017 as part of ongoing monitoring of the reintroduced subpopulation indicate a minimum of 61 fishers (38 females, 23 males), which is 21 more than were originally introduced (Powell et al. 2019, p. 2). Overall, 220 individual fishers were identified between 2009 and 2017 with a young age structure, suggesting healthy reproduction and recruitment (Powell et al. 2019, p. 2). Although the subpopulation appears to be stable or growing, statistical conclusions will be difficult to draw until year 10 in 2020 (Powell et al. 2019, p. 2). The authors also concluded that the subpopulation is unlikely to go extinct in the next 20 years, barring dramatic decreases in survival and reproduction caused by stochastic events. We also recently received a draft manuscript concluding that estimated recruitment and survival probability of fishers in the NSN subpopulation “had stabilized and were quite high, indicating that this new population of fishers may be self-sustaining” (Green et al. 2020, p. 11).

Older estimates for the NCSO DPS (minus SOC and NSN) using various methodologies range from a low of 258-2,850 individuals, based on genetic data (Tucker et al. 2012, pp. 7, 9-10), to a high of 4,018 individuals based on extrapolation of data from two small study areas within the NCSO DPS to the entire NCSO DPS (Self et al. 2008, pp. 3-5). In 2017, a new estimate was developed for the NCSO DPS that includes southern Oregon and coastal California but still excludes SOC and NSN (Furnas et al. 2017, pp. 2-3). This study used detection/non-detection survey data from across much of the NCSO DPS to calculate an average density of 6.6 fishers per 39 mi

2

(100 km

2

) across the area they defined for the NCSO DPS (Furnas et al. 2017, pp. 12-15). Using this estimate of fisher density, the NCSO DPS is estimated to be 3,196 individuals (2,507-4,184; 95 percent Confidence Interval (C.I.)) and fishers were detected at 41 percent of 321 paired camera stations (Furnas et al. 2017, pp. 10, 12). Density models indicate a core area of predicted high density (greater than 10 fishers per 39 mi

2

(100 km

2

) from between about 25 to 50 mi (40 to 80 km) inland from the coast in the California Coast Range and southern Klamath Mountains in California (Furnas et al. 2017, pp. 12-13). CDFW determined in their status assessment for fishers in California that the assessment done by Furnas, when applied to fishers in the California portion of NCSO, suggests that fishers are common and widespread (estimated to occur at 60 percent of sample units in California) (CDFW 2015, p. 55).

The indigenous population of fishers in Oregon was estimated to have a 26 percent range reduction compared to verifiable fisher records collected since 1993 (Barry 2018, p. 22). However, the author notes this comparison should be treated with caution, and we agree. This estimate is subject to the same limitations as described earlier in this section for the SOC fisher subpopulation. That is, the coarse range map the author used for a baseline comparison included areas with limited numbers or even lack of fishers, so a 26 percent range reduction overestimates any change in the indigenous fisher population in Oregon.

Trend information for fishers within the NCSO DPS is based on the following two long-term study areas. As indicated above, we now consider the NCSO DPS to include the areas previously represented as the SOC and NSN reintroduced fisher subpopulations.

The Hoopa study area is approximately 145 mi

2

(370 km

2

) on the Hoopa Valley Indian Reservation north

of California State Highway 299 and near State Route 96, which is largely surrounded by the Six Rivers National Forest and other private lands. The study area represents the more mesic portion (containing a moderate amount of moisture) of the NCSO DPS. Fisher studies have been ongoing since 1996. The population trend in the period 2005-2012 indicates declining populations with lambda of 0.992 (C.I. 0.883-1.100), with a higher lambda rate for females 1.038 (0.881-1.196) than males 0.912 (0.777-1.047) (Higley et al. 2014, p. 102, Higley 2015, pers. comm.). The authors concluded that “the population as a whole is essentially stable” (Higley et al. 2014, p. 31), but they raised concerns about declines in survival of males over the last 3 years of the study; they believed the decline was associated with toxicant poisoning associated with illegal marijuana growing and that males were at a higher risk because of their larger home ranges compared to females (Higley et al. 2014, pp. 32, 38).

The Eastern Klamath Study Area (EKSA) is approximately 200 mi

2

(510 km

2

) in size straddling the California/Oregon border. This study area represents the more xeric portion (containing little moisture; very dry) of the NCSO DPS. Monitoring has occurred since 2006 (Green et al. 2018b, entire). Fishers in this study area were a source for translocating fishers to the NSN reintroduction site elsewhere in the DPS. The removal of nine fishers over a 2-year period in 2009 and 2010 (equivalent to 20 percent of the population) did not affect fisher abundance or density in the study area (Green et al. 2017, p. 9).

After fires in this study area in 2014, the estimated number of fishers declined by 40 percent from the year before the fire (Green et al. 2019b, p. 8). Prior to the fire, this population varied in the annual number of fishers and lambda trends (increasing and decreasing) (Green et al. 2016, p. 15, Table 1) (Table 2), indicating “the population of fishers in the Klamath was relatively stable before the fires occurred and for the three years immediately following the removal of fishers for translocations” (Green et al. 2016, p. 8). Modeling results suggest the post-fire decline was because of the fire. Although the fire notably affected fishers in this population in the 2 years immediately following, the fate of the fishers affected by the fire is unknown; it is possible that some fishers may have emigrated out of the burned areas (Green et al. 2017, pp. 9-10) or may reoccupy areas that burned at lower severities in the future. Credible intervals (a statistical measure of uncertainty) surrounding abundance estimates of fishers both pre- and post-fire overlap; although the post-fire estimate is at the lower range of the pre-fire estimate, the fisher population estimate post-fire does not appear to be substantially different from the lowest estimates in the pre-fire years (Green et al. 2019b, p. 18; Matthews and Green 2020, pers. comm.). Hence, even with the immediate decline in the local fisher population after the fire, the latest population estimate still appears to be within the statistical range of variation of pre-fire estimates. Data since 2016 have not yet been analyzed to assess the EKSA population trend over the past few years.

In the absence of limiting factors, populations tend to steadily increase (lambda >1) until the population growth becomes restricted. Within the NCSO DPS, this situation has been occurring in the NSN reintroduced population as it expands to fill available habitat (Powell et al. 2019, pp. 2, 4). Healthy populations will then naturally fluctuate around their upper limit, or carrying capacity, increasing in some years and decreasing in other years (Figure 2). This trend is exhibited in the data from the EKSA, where annual estimates of abundance for fishers have varied, yielding increasing and decreasing growth rates from year to year prior to the 2014 fires (Table 2). This occurrence is consistent with normal variation for populations that are neither growing nor declining, but fluctuating near carrying capacity. For both the Hoopa and the EKSA studies, the authors' use of the term “stable” (Higley et al. 2014, p. 31; Green et al. 2016, p. 8) implies that the lambda rates are not swinging dramatically from year to year, but rather annual abundance estimates are fluctuating around a steady value consistent with normal population variation. There are still uncertainties regarding the post-fire declines from the EKSA study area (addressed below in Wildfire and Wildfire Suppression section) as well as the reduced male survival rates in the Hoopa study area. However, the best available data suggests that populations are exhibiting variability that may be consistent with populations at or near carrying capacity.

ER15MY20.001

Table 2—Derived Posterior Parameter Estimates of Annual Population Density, Abundance, and Population Growth of Fishers in the Klamath. Parameters Are Presented as Median [95% Credible Interval] (Green et al. 2016, p. 15)

[These estimates have since been reparameterized (Matthews and Green 2020, pers. comm.), indicating a population exhibiting typical fluctuations both increasing and decreasing around K for this time period]

Year

Density

(fishers/100 km

2

)

Abundance

Lambda

2006

6.64 [4.94, 8.35]

39 [29, 49]

2007

6.64 [4.94, 8.18]

39 [29, 48]

1 [0.71, 1.35]

2008

6.99 [5.62, 8.69]

41 [32, 50]

1.06 [0.78, 1.4]

2009

6.47 [5.11, 8.18]

38 [29, 47]

0.92 [0.67, 1.2]

2010

5.79 [4.43, 7.33]

34 [26, 43]

0.91 [0.64, 1.21]

2011

6.47 [5.11, 8.18]

38 [28, 46]

1.09 [0.78, 1.45]

2012

6.3 [4.94, 8.18]

37 [27, 46]

0.98 [0.72, 1.33]

2013

6.99 [5.62, 8.69]

41 [32, 50]

1.11 [0.81, 1.49]

Fishers in the NCSO DPS have rebounded substantially from their low in the late 1800s and early 1900s. Grinnell et al. (1937, p. 227) suggested no more than 300 fishers occurred in all of California. Fishers currently occupy much of their historical range in northwestern California, including the redwood region, which may be an expansion from their historical distribution (CDFW 2015, p. 23); fisher detections have increased in northern coastal California since the 1990s, though it is not known as to whether this increase is due to a range expansion, recolonization, increased survey effort, or whether fishers remained undetected in earlier surveys (CDFW 2015, p. 50). Recent monitoring information submitted during the public comment period on the 2019 Revised Proposed Rule indicates fishers continue to occur across much of northern coastal California; systematic camera surveys on private timber lands found fishers at 65 of 93 (70 percent) camera stations (Green Diamond Resource Company [GDRC] 2019, p. 8) during the 2018-2019 winter, suggesting fishers are well-distributed across the company's lands. In Oregon, fishers also appear to have expanded from low numbers in the 1940s, when fishers were considered extremely rare and perhaps close to extirpation (see Barry 2018, pp. 16-17 for summary), to being “relatively common” where the indigenous population is found (Barry 2018, p. 22). Fishers also appear to be widespread and common throughout much of the DPS (CDFW 2015, pp. 54-55).

The major habitat-based threats experienced by the NCSO DPS are loss of complex canopy forests and den/rest sites and fragmentation of habitat from high-severity wildfire, wildfire suppression activities (

e.g.,

backburning, fuel breaks, and snag removal), and vegetation management (

e.g.,

fuels reduction treatments, salvage, hazard tree removal). Major non-habitat related threats are exposure to toxicants and, in some areas, predation. In

addition to these threats acting on the NCSO DPS, several conservation efforts are also designed to benefit fishers. These efforts include those being implemented within the portion of the range covered by the Northwest Forest Plan (NWFP) including the conservation and retention of late seral habitats and a network of reserved land use allocations, which provide fisher habitat. We summarize conservation measures and regulation mechanisms that address some of these threats below in the Existing Regulatory Mechanisms section.

Threats

As described above in the General Threats Information section, we determined our foreseeable future timeframe for evaluating the status of the NCSO fisher based upon the period for which we can reasonably determine that both the future threats and the species' responses to those threats are likely. In general, we considered that the trajectories of the threats acting on fisher subpopulations across the DPS's range could be reliably predicted for 35-40 years into the future.

We estimated this timeframe as a result of our evaluation of an array of time periods used in modeling. For example, climate models for areas with fisher habitat, HCPs, and timber harvest models generally predict 50 to 100 years into the future, and forest planning documents often predict over shorter timeframes (10 to 20 years). We considered 40 years at the time of the 2014 Proposed Rule, and given the 5-year time period since, we are modifying the foreseeable future time period to a range of 35-40 years. This is a timeframe that we can reasonably determine that both the future threats and the species' responses to those threats are likely. This time period extends only so far as the predictions into the future are reliable, including a balance of the timeframes of various models with the types of threats anticipated during the 35- to 40-year time period.

Wildfire and Wildfire Suppression

Direct evidence of fisher population response to wildfire is limited. In a monitored fisher population in the Klamath-Siskiyou area, declines in the overall fisher population occurred after wildfires in the study area in 2014 (Green et al. 2019b, entire). This population of fishers has been monitored since 2006. As noted by Green et al. (2019b, p. 4): “Previous research indicates this population of fishers had been relatively stable up to 2013, despite approximately 20% of the population being translocated elsewhere between 2009 and 2011.” Fisher numbers in the study area declined 40 percent from 2013, the year prior to the fires. This decrease became apparent the first full year following the fires (2015) and persisted into the following year (Green et al. 2019b, p. 8, Figure 2). While the fate of the fishers affected by the fire is unknown, it is possible that some fishers may have emigrated out of the burned areas (Green et al. 2017, pp. 9-10) or may reoccupy areas that burned at lower severities in the future. The reduced population estimate appears to be within the statistical range of variation of pre-fire estimates, as evidenced by overlapping credible intervals. The post-fire population decline of 40 percent is based on a comparison with the population estimate from 2013, which was the highest measured population estimate compared to all previous years, with 39 animals estimated (Green et al. 2017, p. 19; 2019b, pp. 15-18). The post-fire population estimate was not evaluated in context with the overall pre-fire population trend and its overall variation; such a comparison would likely yield a less dramatic population change. In addition, monitoring data since 2016 is not yet fully evaluated. Both of these tasks are currently underway (Matthews and Green 2020, pers. comm.). Fisher densities declined across all wildfire severity types, but they declined the most in areas with more than a 50 percent loss of tree basal area, consistent with other studies (Green et al. 2019b, pp. 6, 9). The authors note that their data represent only the short-term effects of fires, and any negative effects may not persist. We do not know the fate of individual fishers that left the population after the fire and whether their fitness was ultimately compromised. But this analysis does suggest that high-severity fires can have immediate and substantial effects on local fisher numbers.

Within the Biscuit Fire area in southwest Oregon, which burned in 2002, surveys conducted in 2016 and 2017 did not detect fishers within the burn perimeter (Barry 2018, pp. 22-23), suggesting the fires have extirpated fishers from the burn area. However, detection records do not suggest fishers were ever abundant in the area prior to the fire (Service 2016, pp. 24, 33, 34, and 35, Figures 4, 6, 7, and 8). We do acknowledge, however, that a large part of this area, is within the Kalmiopsis Wilderness Area, where surveys were likely limited due to restricted access. Therefore, fisher occupancy in some areas of the Biscuit Fire remains unknown.

Given projected changes in climate, forests are expected to become more vulnerable to wildfires over the coming century. For example, the proportion of forests considered highly suitable for wildfire in the Klamath Mountains is projected to increase from 18 percent to 48-51 percent by the end of the century, with most of that increase projected to occur on Federal lands (Davis et al. 2017, p. 180). Fire return intervals in low- to mid-elevation forests in Northwest California and the Sierra Nevada Mountains have among the highest departure rates from historical fire return intervals in the State (Safford and Van de Water 2014, pp. iii, 17, 22, 36-37). And, fire return intervals in the Coast Range and Klamath Mountains in Oregon are expected to decrease by half, which would result in a near tripling of the annual area burned in this century compared to last (Sheehan et al. 2015, pp. 20-22; Dalton et al. 2017, p. 46). We note that the projected increases include fires of all severity types, so the potential wildfire areas do not translate directly to an amount of fisher habitat removed. In the case of low- and moderate-severity fires, these may actually create elements used by fishers.

An analysis of fire effects on fisher habitat was done centering on the Klamath Basin and encompassing the NCSO (CBI 2019b and 2019c, entire). The study looked at fisher habitat patches large enough to support five or more breeding female home ranges (CBI 2019b, p. 16) and labeled them as core habitat; the study also identified fisher linkage areas, which were areas on the landscape identified as least-cost pathways to connect the core habitats (CBI 2019b, pp. 3, 16). They found that 24 percent of modeled fisher core areas and 24 percent of modeled fisher linkage areas were considered at risk of at least temporary loss due to severe fires (CBI 2019c, pp. 22, 25). It is important to note that these percentages do not total to 48 percent of the fisher habitat in the study area; core areas are larger patches of fisher habitat, while linkage areas may or may not comprise suitable habitat, but instead represent “least cost” paths between core areas.

To update our 2014 analysis of wildfire effects within the NCSO DPS, we conducted an analysis similar to the one completed for the 2014 draft Species Report (Service 2014, pp. 62-64; Service 2019b, unpublished data). Using the fisher habitat map developed for the 2014 Proposed Rule (Service 2016, Appendix B) and USFS data for burn severity for 2008-2018 (USFS 2019), we estimated the effects of high-

severity wildfire to fisher habitat (high and intermediate categories) over the past 11 years. We assumed wildfires that burned at high severity (greater than 50 percent basal area loss) changed fisher habitat to a condition that would not be selected by fishers for denning and resting (although this result may not always be the case, as described above in the General Species Information section). Use of greater than 50 percent basal area loss is consistent with recent fire effects analyses on fishers based on the recent results as reported in Green et al. (2019b, p. 6). Overall, high- and intermediate-quality fisher habitat in the NCSO DPS decreased by 526,424 ac (213,036 ha) from 7,050,035 ac (2,853,047 ha) to 6,523,610 ac (2,640,011 ha), or approximately 7.5 percent was lost as a result of wildfires since 2008; this is an average loss of 6.8 percent per decade.

For comparison purposes, in our 2014 draft Species Report, we estimated 4 percent of fisher habitat would be lost over the next 40 years due to high-severity wildfire, or 1 percent per decade (Service 2014, p. 64). Our 2014 area of analysis for the NCSO subpopulation was based on 27 years of fire data from 1984 to 2011 and assessed approximately 24,080,693 ac (9,745,111 ha), compared to the 10,459,612 ac (4,232,855 ha) assessed in our recent analysis above. The results of our new analysis are based on fire data from the period 2008 to 2018, an 11-year period of the most recent fire activity, which suggests our earlier estimates of changes to fisher habitat from wildfire over the next 40 years may have been an underestimate. However, while this increase in area burned may be consistent with the projections for wildfire increases in the DPS, the magnitude of increase in burned fisher habitat (

i.e.,

from 1 percent per decade in our 2014 analysis to 6.8 percent in our 2019 analysis) may not be a true reflection of the rate of change between the two time periods because of the different temporal (28 years v. 11 years) and geographic (the area analyzed in 2014 was twice as large as the area assessed in 2019) scales used in the comparison. Nevertheless, we recognize the increase in fire activity within the NCSO.

The geography of the Klamath ecoregion, which makes up much of the NCSO where fishers occur, is steep and complex. The variation in elevation and aspect shapes vegetation composition and distribution. This environment influences fuels and ultimately fire behavior and location (Taylor and Skinner 1998, p. 297; Taylor and Skinner 2003, p. 714; Skinner et al. 2018, pp. 179-180). Consequently, fires tend to be more prevalent on drier sites, while less frequent on moister sites, which tend to be areas more consistent with fisher habitat. While these patterns may or may not continue with the effects of climate change, we can use management such as the recent fuels reduction MOUs (see Existing Regulatory Mechanisms below) to leverage existing topography and vegetation condition to better manage for wildfires.

We acknowledge that large-scale wildfires affect fisher habitat, particularly given the predicted increases in wildfire associated with climate change by the end of the century. We also acknowledge that fires, even large fires, are part of the natural fire regime within the NCSO DPS, and fishers have sustained themselves and coexisted with wildfire for centuries. Into the future, it will be important to have areas that can maintain reproducing fishers while severely burned areas can regenerate into fisher habitat again, whether that is foraging habitat within a decade or two, or denning and roosting habitat several decades beyond. Existing land allocations like late-successional reserves from the NWFP on Federal lands throughout much of the NCSO DPS, especially in the areas with the greatest fire severities, will be necessary to manage these areas to return to forest habitat with complex structure. This process will ensure suitable habitat lost to fires will be managed to develop the overstory and structural features conducive to fishers. In the interim, retaining important structural features in burned areas, per reserve land allocation standards and guidelines, will facilitate the use of these areas by prey and foraging fishers within a few decades following high-severity fires.

Although fire risk is expected to increase with climate change, it is not expected to be uniform across the DPS, as described above in this section. The sporadic and episodic nature of fires will help ameliorate some of the risk to fishers across the DPS as a whole. There are effects to local fisher populations immediately after a high-severity fire (

e.g.,

Green et al. 2019b, entire). But fishers are well distributed across the NCSO DPS, including coastal areas such as the redwood region that may be less prone to wildfire risk. This distribution provides redundancy to loss of fishers after a local fire event. Plus, fishers appear to use high severity burned areas, at least for dispersal and foraging (Service 2016, p. 66), suggesting that even severely burned areas can continue to provide some benefits to fishers within a decade or two after the fire. The redundancy exhibited by the NCSO DPS, with multiple subpopulations distributed across a substantial range of habitat (see Resiliency, Redundancy, and Representation section), will allow the NCSO DPS of fishers to absorb the impact of fires, demonstrating the DPS's ability to withstand catastrophic events.

Climate Change

The general climate change related effects discussed above (see General Threats Information) apply to the NCSO DPS, in addition to the following effects, which are more specific to the NCSO DPS. In particular, Siskiyou and Trinity Counties in interior northern California are projected to see the greatest temperature increases for the North Coast Region (Grantham 2018, p. 17). In the Klamath Mountains, models suggest precipitation is likely to fall increasingly as rain rather than snow, becoming mainly rain-dominated by mid-century (Dalton et al. 2017, p. 17). Significant or amplified wildfire activity, with increased area burned and severity can result in reduced denning habitat availability for fishers in the Coast Range and Klamath Mountains. These two areas are projected to experience wildfire return intervals decreased by half and thus result in a near tripling of the annual area burned in this century compared to last (Sheehan et al. 2015, pp. 20-22; Dalton et al. 2017, p. 46). Fire return intervals in low- to mid-elevation forests in Northwest California and the Sierra Nevada Mountains have among the highest departure rates from historical fire return intervals in the State (Safford and Van de Water 2014, pp. iii, 17, 22, 36-37).

Overall, the best available scientific and commercial information suggests that changing climate conditions (particularly warmer and drier conditions) are influencing other threats to fishers and their habitat within the NCSO DPS, in particular the potential for increased wildfire frequency and intensity. However, this is not to say that the DPS will experience widespread or a uniform distribution of climate-driven wildfire events. Even under conditions for a potential increase in wildfire frequency, wildfires will remain sporadic and episodic across the range of the DPS, further moderated by the slope and aspect of terrain throughout the range (

e.g.,

influencing susceptibility to wildfire, and creating a mosaic of fire severity). The DPS's wide variety of topography, vegetation, and climate conditions in its array of physiographic provinces (Service 2016, pp. 15-17, 28-29, 38-39) results in

unpredictable variability in how these provinces will respond to changing climate conditions. Please see additional discussion about potential impacts to fishers or their habitat associated with wildfire (Wildfire and Wildfire Suppression above).

Tree Mortality From Drought, Disease, and Insect Infestation

Specific to the NCSO DPS, sudden oak death (

Phytophthora ramorum

) has caused some tree mortality in southwestern Oregon and northwestern California, but it is not causing widespread losses of oaks (California Oak Mortality Task Force 2019, p. 1; Oregon Department of Forestry (ODF) 2016, pp. 1-2). This finding suggests widespread loss of oaks used by fisher or fisher prey is not occurring as a result of sudden oak death. Overall, warmer and drier climate conditions are projected for the NCSO DPS; however, the varied composition of the vegetation (

e.g.,

Lofroth et al. 2011, pp. 34-90) in the DPS suggests insect outbreaks and disease due to drought-related stress on trees are more likely to be localized should they occur; therefore, future widespread tree mortality impacts to fisher habitat are not anticipated in the NCSO DPS.

Vegetation Management

Although local analyses across the NCSO DPS have assessed fisher habitat at several scales (see Lofroth et al. 2011, pp. 34-90 for study summaries, and Raley et al. 2012, pp. 234-235 for list of additional studies), there is no analysis available that explicitly tracks changes in fisher habitat in recent decades across large portions of the DPS, and which includes fisher habitat ingrowth as well as habitat loss to specific disturbances. Therefore, we used other available information, as described below, to analyze the potential effects of this threat on fishers in the NCSO DPS. In addition to the draft Species Report (Service 2014, pp. 85-96), we used several different sources of information to depict forest vegetation changes caused by vegetation management activities and offset by ingrowth within the range of the NCSO DPS. With the exception of the non-Federal timber harvest database in California (CAL FIRE) 2013), all of these sources are either new or updated since 2014 (Davis et al. 2015, entire; USFS 2016, entire; Spencer et al. 2016, entire; Spencer et al. 2017, entire; gradient nearest neighbor (GNN) data/maps). With these available data, we did not need to rely on northern spotted owl habitat data as a surrogate for fisher habitat data in this evaluation. Our revised methodology is described in detail for the historical, three-State range of the DPS in the 2016 final Species Report (Service 2016, pp. 98-111); we summarize it below and describe how it applies to the NCSO DPS.

Within the portion of the NCSO DPS overlying the Northwest Forest Plan region (generally most of the NCSO DPS except for the northern Sierras), we used information from the draft late-successional and old-growth forest monitoring report (Davis et al. 2015, entire) to assess changes in structural habitat elements associated with fisher habitat (

i.e.,

large trees, down wood, snags) as a result of vegetation management. This information included use of the “old growth structure index” (OGSI), which is an index that consists of four structural elements associated with older forests: (1) The density of large live trees; (2) the density of large snags; (3) the amount of down wood cover; and (4) the tree size diversity of the stand. Over a 20-year period (1993-2012), Davis et al. (2015, pp. 5-6, 16-18) tracked changes in forests classed as OGSI-80, which represents forests that begin to show stand structures associated with older forests (

e.g.,

large live trees, snags, down wood, and diverse tree sizes). Though OGSI-80 forests are not a comprehensive representation of fisher habitat, the condition does track forests that contain structural elements consistently used by fishers in habitat studies across the DPS, even in areas with substantially open areas and managed young stands (Lofroth et al. 2010, pp. 81-121; Service 2016, pp. 15-21; Niblett et al. 2017, pp. 16-17; Powell et al. 2019, pp. 21-23; Matthews et al. 2019, pp. 1,309, 1,313; Moriarty et al. 2019, pp. 29-30, 46-49). We acknowledge there is some unknown level of overrepresentation of stands that may not be occupied by fishers and underrepresentation of stands that fishers may actually occupy (Service 2016, p. 102), and we do not suggest that OGSI-80 is a surrogate for fisher habitat proper. Hence, we do not consider it a model of fisher habitat.

However, OGSI-80 does cover a majority of the NCSO DPS and provides a way to assess regional-scale trends in forests that contain the structural elements consistently used by fishers (

e.g.,

large snags, down wood, and large live trees). This information was the only data set available that identified the number of acres lost to timber harvest or vegetation management (as well as disturbances from fire and insects) and the number recruited by forest ingrowth. This OGSI-80 data set allows us to track changes as a result of vegetation management and forest recruitment. In using the OGSI-80 data, we do not expect there to be substantial differences in relative trends for disturbances and ingrowth effects on OGSI-80 stands compared to trends in their effects on fisher habitat.

Details of our analysis of Davis et al. (2015, entire) are explained in the 2016 final Species Report (Service 2016, pp. 101-102). We have since modified that analysis to include only data for the areas (physiographic provinces) that cover the current range of fishers in the NCSO DPS. The California portion of the NCSO DPS covers all of the California physiographic provinces analyzed in Davis et al. (2015, pp. 10, 30-31). The Oregon portion of the NCSO DPS occurs mostly within the Oregon Klamath province, but overlaps somewhat into small portions of the western and eastern Cascades provinces (Davis et al. 2015, pp. 10, 30-31). We assessed the results of including and excluding the data from these two Cascades provinces. Because no substantial differences were revealed between the two data sets, we report here the results of including only the Oregon Klamath province data along with data for all of the California physiographic provinces that are covered by the NWFP.

Although loss of OGSI-80 forests due to timber harvest on non-Federal lands (11.1 percent since 1993) was substantially greater than on Federal lands (1.0 percent since 1993), in combining all ownerships, the percent loss due to timber harvest from 1993 to 2012 was low (5.0 percent). This translates to a 2.5 percent loss per decade. However, this may underestimate future harvest trends because timber harvest volume within the NWFP area on Federal lands has been on a general upward trend since 2000. During the first decade of NWFP implementation, Federal agencies offered, on average annually, 54 percent of the timber harvest sale goals (probable sale quantity or PSQ) identified in the Plan, whereas volume offered in 2012 was at about 80 percent of the PSQ identified in the NWFP, as agencies became more familiar with implementing the NWFP (BLM 2015, p. 340; Spies et al. 2018, pp. 8-9). In addition, BLM has recently revised their management plans in western Oregon and is no longer operating under the NWFP. Consequently, that agency is predicting an increase in timber volume above the NWFP sale quantity in the first decade of implementation (through circa 2025) (BLM 2015, pp. 350-352). Recent litigation may also increase timber harvest on BLM (see Existing Regulatory Mechanisms section). Hence,

overall harvest trends on Federal lands may be increasing and may be closer to or more than rates observed in the last decade of NWFP implementation (2003 to 2012).

The net loss of OGSI-80 conditions to timber harvest, however, is somewhat less because 2.5 percent per decade does not include ingrowth of OGSI-80 stands. Ingrowth represents those stands that did not meet the OGSI-80 structural thresholds at the beginning of the 20-year monitoring period but, through vegetation succession, reached those thresholds at the end of the monitoring period. Stands that grow into the OGSI-80 condition are assumed to offset the loss of other OGSI-80 to disturbance such as vegetation management. However, we acknowledge that OGSI-80 stands exist on a continuum, and OGSI-80 stands lost to timber harvest or some other disturbance are not necessarily equivalent in structural quality to stands that recently cross a threshold of being classified as OGSI-80. That is, the longer stands remain in the OGSI-80 classification, the more likely they are to contain more old-forest structural conditions that benefit fishers.

Ingrowth of OGSI-80 stands within the NWFP portion of the DPS occurred at a rate of 8 percent over the 20-year period, or 4 percent per decade (calculated from Davis et al. (2015, Tables 6 and 7, pp. 30-31)). This ingrowth more than offsets the OGSI-80 stands lost to vegetation management. However, there is still an overall net loss of OGSI-80 stands in the DPS because all disturbances (

i.e.,

wildfire and forest insects and pathogens) need to be considered. When all disturbances and ingrowth are factored in, there is a net loss of 1 percent per decade. However, vegetation management affects a small portion of those habitat components used by fisher within the NWFP area. Furthermore, ingrowth rates are expected to increase in the foreseeable future on Federal lands within the NWFP area because forests regenerating from the post-World War II harvest boom starting in the 1940s are beginning to meet the OGSI-80 threshold (Davis et al. 2015, p. 7).

We note that we incorporated the loss of OGSI-80 stands to wildfire into this analysis of vegetation management only to fully consider the degree to which ingrowth can offset loss of OGSI-80 stands to disturbance. We use a different metric to address the loss of fisher habitat to wildfire (see the Wildfire and Wildfire Suppression section). For the wildfire analysis, we were able to obtain data from past wildfires and overlay it on fisher habitat to better represent fisher habitat loss to high-severity wildfires as well as to incorporate the effects from more recent wildfires than those analyzed by Davis et al. (2015, p. 29).

Outside of the NWFP portion of the DPS (primarily Sierra Nevada region), while we could track vegetation changes over time, the available data did not indicate the amount or types of disturbances affecting the specific vegetation types; that is, we could determine net change in a particular vegetation type, but could not quantify the amount lost to a specific disturbance type, unlike in the NWFP area. Timber harvest records were available for the Sierra Nevada region, but idiosyncrasies in the FACTS (Forest Service Activity Tracking System) database (see Spencer et al. (2016, p. A-30)) and the fact that the available private lands database (CAL FIRE timber harvest plans) did not indicate types of treatment or what portion of the plans may have actually been implemented, led to concerns in translating acres of “treatment” as depicted in these databases into on-the-ground changes in forest vegetation types that could represent fisher habitat. Instead, we relied on net vegetation change data to display actual changes in forests that approximate conditions suitable for fisher habitat, although we realize that net changes include other disturbances and that vegetation management will be some unknown portion of that change.

For the Sierra Nevada Range (note that this includes the entire range, as we were not able to split out the SSN DPS from the NCSO DPS), we approximated fisher habitat change using a vegetation trend analysis to track changes in forests with large structural conditions thought to be associated with fisher habitat (see Service 2016, p. 106 for a description related to using GNN data). The vegetation category tracked in this analysis is not equivalent to the OGSI-80 forests used by Davis et al. (2015, entire). Instead, the available data limited us to using predefined structure conditions describing forests with larger trees (greater than 20 in (50 cm)). We realize this process may not include all vegetation types used by fishers. This analysis showed that net loss of forests with larger structural conditions in the Sierra Nevada Range was 6.2 percent across all ownerships over the past 20 years, which equates to a loss of 3.1 percent per decade. However, this amount is loss associated with all disturbance types, including wildfire, insects, and disease, that occurred from 1993 through 2012. Hence, vegetation management is some unknown subset of this loss.

Vegetation management is not affecting large areas of the NCSO DPS, though fragmentation could be restricting fisher movements in localized areas or increasing predation risk. For example, fishers continue to persist in actively managed landscapes (GDRC 2019, no page numbers), and fishers reintroduced into the Sierra Nevada portion of the NCSO DPS on SPI lands, which are managed for timber production, suggest that fisher populations can become established and persist in a landscape where substantial portions were historically and are currently managed for timber production (Powell et al. 2019, entire; Green et al. 2020, entire). Hence, we conclude that vegetation management is a low-level threat because of the small proportion of area harvested in the NCSO DPS and because of the widespread distribution of fishers and their occurrence in actively managed landscapes.

Exposure to Toxicants

As described above in the General Threat Information section, rodenticides analyzed as a threat to the NCSO DPS of fishers include first- and second-generation anticoagulant rodenticides and neurotoxicant rodenticides. Both the draft and final Species Reports detail the exposure of the NCSO DPS of fishers to rodenticides in northern California and southern Oregon (Service 2014, pp. 149-166; Service 2016, pp. 141-159). Data available since the completion of the final Species Report in 2016 continue to document exposure and mortalities to fishers from rodenticides in the NCSO DPS (Gabriel and Wengert 2019, unpublished data, entire). Data for 48 fisher carcasses collected in the range of the NCSO DPS in the period 2007-2018 indicate 36 fishers (75 percent) tested positive for one or more rodenticides (Gabriel and Wengert 2019, unpublished data), while 13.5 percent of fisher mortalities with a known cause in the NCSO DPS from 2007 through 2014 were attributable to rodenticides (7 of 52 mortalities) (Gabriel et al. 2015, p. 6). Using data from both the SSN and the NCSO DPSs, mortalities due to rodenticide toxicosis increased from 5.6 to 18.7 percent since the collection and testing of fisher mortalities using data comparing the periods 2007-2011 to 2012-2014 (Gabriel and Wengert 2019, unpublished data, p. 2). From 2015 to 2018, additional NCSO DPS fisher mortalities due to both anticoagulant and neurotoxicant rodenticides have been documented (Gabriel and Wengert 2019, unpublished data, p. 4). At the Hoopa study site, population monitoring found

“the population as a whole is essentially stable” (Higley et al. 2014, p. 31), but there are concerns about declines in survival of males over the last 3 years of the study. The authors speculate this decline in male survival is attributed to toxicant poisoning associated with illegal grow sites and that males were identified as being at a higher risk for poisoning because of their larger home ranges compared to females (Higley et al. 2014, pp. 32, 38).

To evaluate the risk to NCSO DPS fishers from illegal grow sites, we use a Maximum Entropy model to identify high and moderate likelihood of illegal grow sites being located within fisher habitat (Gabriel and Wengert 2019, unpublished data, pp. 7-10) in Oregon and California. This model indicates that 54 percent of habitat modeled for NCSO DPS fishers is within areas of high and moderate likelihood for marijuana cultivation.

The majority of our illegal grow site data comes from California, and data are limited for the amount of pesticides used in Oregon. The USFS documented 63 trespass grows between 2006 and 2016, with toxicants present at all these sites (Clayton 2019, pers. comm.). In a separate effort, only one illegal grow site in southern Oregon has been sampled using the same protocol as 300 illegal grow sites in California where the amount and type of rodenticide at a site is tracked. This southern Oregon location had 54 pounds (lb) (24.5 kilograms (kg)) of first-generation anticoagulant rodenticide and 8 lb (3.6 kg) of neurotoxicant rodenticide dispersed around the site (Gabriel and Wengert 2019, unpublished data, p. 7).

As of January 24, 2020, 2,138 legal marijuana cultivation permits were active in counties within the NCSO and SSN DPSs in California (California Department of Food and Agriculture 2020, entire), and 423 legal marijuana operations have been approved as of January 17, 2020, in Oregon counties occupied by fishers (Oregon Liquor Control Commission 2020, entire).

Toxicant use on the landscape, and especially anticoagulant rodenticides, is a problem for fisher. However, the NSN subpopulation has grown to the point of becoming self-sustaining (Green et al. 2020, p. 11; Powell et al. 2019, p. 4) even with 11 of 12 fishers testing positive for anticoagulant rodenticides (Powell et al. 2019, p. 17). This finding suggests that toxicants may not be having a limiting effect on growth in this subpopulation. And, at EKSA only small annual variations were seen in the lambda value (Table 2) from 2006 to 2013 (Green et al. 2016, p. 15). This period is at the same time as toxicant data were being collected (Gabriel et al. 2015, entire; Gabriel et al. 2017, entire; Gabriel and Wengert, unpublished data 2019, entire), and presumably there were illegal grow sites distributed throughout the landscape. Illegal marijuana cultivation has been occurring in California since the mid-1970s. To some degree, the fisher's widespread distribution and relative commonness in the NCSO DPS diffuses the potential for a significant percentage of the subpopulation to be exposed to these toxicants. The presence of illegal grow sites on the landscape since the mid-1970s suggests that the fisher has been living with this threat for some time.

We do not know what level of toxicant exposure is occurring in live fishers in the wild. The best available mortality data are limited (19 individuals in California (Gabriel and Wengert 2019, unpublished data, p. 5), and of the 2 fishers found in Oregon that were tested for rodenticide exposure, both tested positive (Clayton 2016, pers. comm.). We also do not know how the legalization of marijuana will change grow-site location and potentially affect exposure and mortality rates of fishers due to rodenticides.

We view toxicants as a potentially significant threat to fishers in the NCSO DPS because of the reported exposure rate of toxicants in the NCSO DPS, the reported mortalities of fishers from toxicants in the NCSO DPS, the variety of potential sublethal effects due to exposure to rodenticides (including potential reduced ability to capture prey and avoid predators), and the degree to which illegal cannabis cultivation overlaps with the range and habitat of fisher in the NCSO DPS. The exposure rate of 75 percent of fisher carcasses tested in the NCSO DPS has not declined between 2007 and 2018 (Gabriel and Wengert 2019, unpublished data, pp. 3-4), while toxicosis has increased since 2007 (Gabriel et al. 2015, p. 7). As noted above, we do not know the exposure rate of live fishers to toxicants because this data is difficult to collect. In addition, the minimum amount of anticoagulant and neurotoxicant rodenticides required for sublethal or lethal poisoning of fishers is currently unknown. In spite of the widespread nature of illegal grow sites and their known association with illegal rodenticide use, as well as the prevalence of toxicants occurring in tested fishers, the NCSO subpopulation may be demonstrating an ability to withstand this threat with regard to population growth (see discussions above in Current Condition section regarding observed population growth and fluctuation information in NSN and at the EKSA and Hoopa sites).

Illegally used toxicants like rodenticides remain a threat to fishers within the NCSO DPS now and in the foreseeable future. Where illegal marijuana grow sites occur on the landscape and overlap with fisher ranges, illegally used pesticides have a high potential to harm those exposed individual fishers. However, while the threat of people developing illegal grow sites is widespread, we also note that such sites are generally widely dispersed within remote landscapes across the DPS range (

i.e.,

illegal growers look to be as isolated and hidden as possible). This situation would suggest that potential for significant exposure to fishers is generally limited to where the grow sites are located. However, while there is no certain discernible trend regarding whether illegal grow sites may increase or decrease as a result of marijuana legalization, it will still likely take many years before the currently existing sites can be found and remediated.

Potential for Effects Associated With Small Population Size

The NCSO DPS, which encompasses both the SOC and NSN reintroduction sites, covers a relatively large geographic area of approximately 15,444 mi

2

(40,000 km

2

). Overall, the NCSO DPS has not expanded beyond our previous estimates; however, the SOC subpopulation may have contracted (Barry 2018, p. 22; Moriarty et al. 2019, p. 5) while the NSN subpopulation continues to grow (Powell et al. 2019, p. 2). Please see the Current Condition section above for detailed information on subpopulation size estimates.

Generally, the ability of a species (or DPS) to withstand a catastrophic event (

i.e.,

bounce back from an event that may result in the loss of a population or large proportion of individuals) is lower with relatively few populations or a very limited distribution across the landscape. Overall, the NCSO DPS has not appeared to grow or expand, despite the availability of suitable habitat. However, multiple, well-distributed subpopulations (

i.e.,

NCSO, NSN, and SOC) continue to exist across the DPS; this occurrence includes aggregates of individuals in geographic areas within NCSO (

i.e.,

EKSA fishers, fishers in and around Redwood National Park, Hoopa fishers, or fishers spread downslope of the Siskiyou Crest). At this time, the best available information for monitored fishers within the DPS (

e.g.,

Green 2017, Higley et al. 2014, Powell et al. 2014, entire; Sweitzer et al. 2015a, entire) does not indicate whether the NCSO DPS is

increasing, stable, or declining. Tucker et al. (2012, pp. 8, 11) found low genetic diversity within the NCSO population (and SSN population), but the NCSO population (and SSN population) had also exhibited low genetic diversity from samples collected between 1880 and 1920, suggesting that the currently low diversity occurred prior to when the historical samples were taken, and thus prior to European settlement. However, fishers have rebounded from substantial population reductions that resulted from historical trapping and habitat loss, and they are currently widespread and common across the DPS. Fishers are well distributed across the NCSO DPS, without barriers for genetic exchange between and among its subpopulations (

e.g.,

genetically homogeneous fishers occupy either side of the Klamath River adjacent to a two-lane, paved highway (Service 2016, p. 113). Genetic diversity decreases moving southward with the peripheral areas having the lowest genetic diversity (Wisely et al. 2004, entire). Low genetic diversity can result in inbreeding depression, and one way to assess the risk of inbreeding depression is to determine the effective population size. An effective population size is the number of individuals in an ideal population that would result in the same level of inbreeding or genetic drift as that of the population under study (Jamieson and Allendorf 2012, p. 578). It is usually substantially smaller than the actual number of individuals in the population, often 10 to 20 percent of the census (actual) population size (Frankham 1995, p. 100). An effective population size estimate of 128 individuals for northwestern California suggests inbreeding depression is not a problem (Tucker et al. 2012, pp. 7-8, 10) when compared to thresholds of 50 or 100 individuals from the established literature discussing effective population sizes (Jamieson and Allendorf 2012, entire; Frankham et al. 2014, entire).

As we have described herein and previously, the NCSO DPS is isolated from other fisher populations, and small relative to the taxon as a whole. As such, the risks of small population size effects and of extinction exist. However, the broad distribution of the DPS across its range, in combination with the DPS occurring in multiple subpopulations with no barriers to genetic exchange within and between those subpopulations, and the low likelihood of a catastrophic event at a scale that could hypothetically affect the entire DPS, indicates that the risks of small population size effects and of extinction are very low.

Disease and Predation

A general description of disease and predation on fishers is provided above (see General Species Information and Summary of Threats). Specific to the California portion of the NCSO DPS, of 42 fisher mortalities analyzed, 54 percent were a result of predation and 19 percent were caused by disease (Gabriel et al. 2015, p. 7, Table 2). It is not unexpected that predation is the greatest source of mortality given the suite of larger, generalist predators that occupy the NCSO DPS (

e.g.,

coyotes, bobcats, and mountain lions). As noted in the General Species Information and Summary of Threats section, we do not know whether observed predation rates are substantially different from historical rates, or whether they are comparable with other populations not subjected to trapping. We acknowledge that sublethal effects of toxicants as well as a possible increase in exposure to generalist predators as a result of habitat modification may result in higher predation rates than what historically occurred (Gabriel et al. 2015, p. 14). However, fishers continue to remain widely distributed across the DPS, there is recent evidence of population growth from the NSN subpopulation, and the EKSA exhibits seemingly normal variability in spite of these stressors.

Vehicle Collisions

Vehicle-related mortalities make up a small portion of overall fisher mortality across California (see General Species Information and Summary of Threats above) and particularly in the NCSO DPS (Service 2016, p. 138). Although major paved highways with high-speed traffic occur throughout the DPS, available records do not indicate localized areas of concentrated mortalities that may substantially decrease local fisher populations. Hence, we do not consider vehicle collisions to be a substantial threat to fishers in the NCSO DPS.

Existing Regulatory Mechanisms

Forest Service (USFS) and BLM

A number of Federal agency regulatory mechanisms pertain to management of fisher (and other species and habitat). Most Federal activities must comply with the Natio

This text is long and has been trimmed here. Open the source document for the complete record.