# Endangered and Threatened Wildlife and Plants; 12-Month Finding on a Petition To List Humboldt Marten as an Endangered or Threatened Species

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3A2015-07766

## Record

- **Collection:** Federal Register
- **Document type:** Proposed Rule
- **Published:** April 7, 2015
- **Citation:** 80 FR 18742

## Text

DEPARTMENT OF THE INTERIOR
Fish and Wildlife Service
50 CFR Part 17
[Docket No. FWS-R8-ES-2011-0105; 4500030113]
Endangered and Threatened Wildlife and Plants; 12-Month Finding on a Petition To List Humboldt Marten as an Endangered or Threatened Species

AGENCY:

Fish and Wildlife Service, Interior.

ACTION:

Notice of 12-month petition finding.

SUMMARY:

We, the U.S. Fish and Wildlife Service (Service), announce a 12-month finding on a petition to list the previously classified subspecies Humboldt marten (
Martes americana humboldtensis
), or the (now-recognized) subspecies of Humboldt marten (
Martes caurina humboldtensis
), or the Humboldt marten distinct population segment (DPS) of the Pacific marten (
M. caurina
) as an endangered or threatened species under the Endangered Species Act of 1973, as amended (Act). The petition and this finding also address populations of marten from coastal Oregon, which recent genetic analyses indicate are likely to be the same entity as the current classification of Humboldt marten. We recognize a coastal DPS of the Pacific marten (which includes coastal Oregon populations of marten and the current classification of Humboldt marten) and find that this DPS is not warranted for listing at this time. However, we ask the public to submit to us any new information that becomes available concerning the stressors that may be impacting the coastal DPS of Pacific marten or its habitat at any time.

DATES:

The finding announced in this document was made on April 7, 2015.

ADDRESSES:

This finding is available on the Internet at
http://www.regulations.gov
at Docket Number FWS-R8-ES-2011-0105. Supporting documentation we used in preparing this finding is available for public inspection, by appointment, during normal business hours at the U.S. Fish and Wildlife Service, Arcata Fish and Wildlife Office, 1655 Heindon Road, Arcata, CA 95521. Please submit any new information, materials, comments, or questions concerning this finding to the above street address.

FOR FURTHER INFORMATION CONTACT:

Bruce Bingham, Field Supervisor, U.S. Fish and Wildlife Service, Arcata Fish and Wildlife Office (see
ADDRESSES
); by telephone at 707-822-7201; or by facsimile at 707-822-8411. If you use a telecommunications device for the deaf (TDD), please call the Federal Information Relay Service (FIRS) at 800-877-8339.

SUPPLEMENTARY INFORMATION:

Acronyms and Abbreviations Used in This Document

We use many acronyms and abbreviations throughout this 12-month finding. To assist the reader, we provide a list of these here for easy reference:

Act = Endangered Species Act of 1973, as amended (16 U.S.C. 1531
et seq.
)

AR = Anticoagulant Rodenticides

BLM = Bureau of Land Management

CBD = Center for Biological Diversity

CDFG = California Department of Fish and Game (see below)

CDFW = California Department of Fish and Wildlife (formerly CDFG)

CDPR = California Department of Parks and Recreation

CESA = California Endangered Species Act

CEQA = California Environmental Quality Act

CFR = Code of Federal Regulations

DPS = Distinct Population Segment

EPIC = Environmental Protection Information Center

Forest Service = U.S. Forest Service

FR = Federal Register

GIS = Geographic Information System

HCP = Habitat Conservation Plan

HMCG = Humboldt Marten Conservation Group

IPCC = Intergovernmental Panel on Climate Change

IUCN = International Union for Conservation of Nature

LANDFIRE = Landscape Fire and Resource Management Planning Tools Project

LRMP = Land and Resource Management Plan

MDL = Multi-District Litigation

MOU = Memorandum of Understanding

MTBS = Monitoring Trends in Burn Severity

NMFS = National Marine Fisheries Service

NWFP = Northwest Forest Plan

OAR = Oregon Administrative Rules

ODF = Oregon Department of Forestry

RMP = Resource Management Plan

Service = U.S. Fish and Wildlife Service

SPR = Significant Portion of [a Species'] Range

USDA = U.S. Department of Agriculture

Background

Section 4(b)(3)(B) of the Act (16 U.S.C. 1531
et seq.
) requires that, for any petition to revise the Federal Lists of Endangered and Threatened Wildlife and Plants that contains substantial scientific or commercial information suggesting that listing a species may be warranted, we make a finding within 12 months of the date of receipt of the petition. In this finding, we will determine that the petitioned action is: (1) Not warranted, (2) warranted, or (3) warranted, but the immediate proposal of a regulation implementing the petitioned action is precluded by other pending proposals to determine whether species are endangered or threatened, and expeditious progress is being made to add or remove qualified species from the Federal Lists of Endangered and Threatened Wildlife and Plants (“warranted but precluded”). Section 4(b)(3)(C) of the Act requires that we treat a petition for which the requested action is found to be warranted but precluded as though resubmitted on the date of such finding, that is, requiring a subsequent finding to be made within 12 months. We must publish these 12-month findings in the
Federal Register
.

Previous Federal Actions

On September 28, 2010, we received a petition dated September 28, 2010, from the Center for Biological Diversity (CBD) and the Environmental Protection Information Center (EPIC), requesting that we consider for listing the (then-classified) subspecies Humboldt marten (
Martes americana humboldtensis
), or the (now-recognized) subspecies Humboldt marten (
M. caurina humboldtensis
), or the Humboldt marten DPS of the Pacific marten (
M. caurina
). The petitioners further stipulated that, based on recent genetic analyses indicating that populations of marten from coastal Oregon (considered members of
M. a. caurina
) are more closely related to
M. a. humboldtensis
than to
M. a. caurina
in the Cascades of Oregon (citing Dawson 2008, Slauson
et al.
2009a), the range of the subspecies or DPS of the Humboldt marten should be expanded to include coastal Oregon populations of martens. In a letter to the petitioners dated October 22, 2010, we responded that we reviewed the information presented in the petition and determined that issuing an emergency regulation temporarily listing the species under section 4(b)(7) of the Act was not warranted.

On January 12, 2012, we published in the
Federal Register
a 90-day finding (77 FR 1900) that the petition presented substantial information indicating that listing may be warranted and that initiated a status review. For purposes of the 90-day finding, the common name Humboldt marten referred to the then-classified American marten (
M. americana
) populations in coastal northern California and coastal Oregon.

On June 23, 2014, we published a scoping notice in the
Federal Register
(79 FR 35509) that summarized the uncertainty regarding the taxonomic classification of the subspecies (based on current genetics information) and indicated our intent to conduct an evaluation (for the 12-month finding) of

a potential DPS of martens in coastal northern California and coastal Oregon relative to the full species classification level.

According to section 3(16) of the Act, we may consider for listing any of three categories of vertebrate animals: A species, subspecies, or DPS (see the Service's 1996 DPS Policy at 61 FR 4722). We refer to each of these categories as a potential “listable entity.” We evaluated three possible listable entities for this 12-month finding based upon the best available published and unpublished information for martens in coastal northern California and coastal Oregon (for further details, please see the
Current Taxonomic Description
and
Listable Entity Evaluation and Distinct Population Segment Analysis
sections, below):

• Subspecies Humboldt marten (
Martes americana humboldtensis
): This entity was considered not reasonable for evaluation because its species-level name is no longer considered valid. Specifically, Dawson and Cook (2012, entire) split the then-classified American marten (
M. americana
) to recognize the Pacific marten (
M. caurina
) for all martens occurring west of the Rocky Mountain crest.

• Subspecies Humboldt marten (
Martes caurina humboldtensis
): This entity was considered not reasonable for evaluation because its description is (currently) specifically linked with the extant population that resides in coastal northern California and does not include the coastal Oregon populations, which the best available genetics data indicate are likely the same entity.

• DPS of the Pacific marten (
Martes caurina
): We considered it reasonable that a DPS of the Pacific marten constitute the listable entity for our status review based on our evaluations of the best scientific and commercial data currently available (including unpublished genetics information), and our consideration of the Service's February 7, 1996, Policy Regarding the Recognition of Distinct Vertebrate Population Segments Under the Endangered Species Act (DPS Policy; 61 FR 4722). As such, we considered in the scoping notice (79 FR 35509; June 23, 2014) that the DPS include the currently recognized
M. caurina humboldtensis
(
i.e.,
Humboldt marten) and the coastal populations of
M. caurina caurina
in Oregon (
i.e.,
Oregon Coast Range group). We solicited information regarding our consideration of the coastal northern California and coastal Oregon populations of Pacific marten as a single listable entity. See
Listable Entity Evaluation and Distinct Population Segment Analysis,
below, for additional discussion related to our decision that a coastal DPS of the Pacific marten (hereafter referred to as “coastal marten”) constitutes the listable entity for this status review.

This notice constitutes the 12-month finding on the September 28, 2010, petition to list the (then-classified) subspecies Humboldt marten (
Martes americana humboldtensis
), or the (now-recognized) subspecies Humboldt marten (
M. caurina humboldtensis
), or the Humboldt marten DPS of the Pacific marten (
M. caurina
) as an endangered or threatened species.

This finding is based upon the Species Report titled “Coastal Oregon and Northern Coastal California populations of the Pacific marten (
Martes caurina
)” (Service, 2015) (Species Report), a scientific analysis of available information prepared by a team of Service biologists from the Service's Arcata Fish and Wildlife Office, Oregon Fish and Wildlife Office, Pacific Southwest Regional Office, Pacific Regional Office, and National Headquarters Office. The purpose of the Species Report is to provide the best available scientific and commercial information about the species so that we can evaluate whether or not the species warrants protection under the Act. In it, we compiled the best scientific and commercial data available concerning the status of the coastal Oregon and northern coastal California populations of Pacific marten, including past, present, and future threats to these populations. As such, the Species Report, including the appendix, provides the scientific basis that informs our regulatory decision in this document, which involves the further application of standards within the Act and its regulations and policies. The Species Report can be found on the Internet at
http://www.regulations.gov
, Docket No. FWS-R8-ES-2011-0105.

Current Taxonomic Description

The American marten (
Martes americana
) was originally described as a single species by Turton (1806, entire), based on specimens from eastern North America. In 1890, Merriam (1890, entire) considered a new species,
Mustela
[=
Martes
]
caurina,
to be those martens found west of the Rocky Mountains. In 1926, the Humboldt [Pine] marten (
M. c. humboldtensis
) was described as a subspecies of
Martes caurina
(Grinnell and Dixon 1926, entire); historically, this subspecies was distributed throughout the coastal, fog-influenced coniferous forests of northern California from northwestern Sonoma County north to the Oregon border (Grinnell and Dixon 1926, entire). In 1953, Wright (1953, entire) described one species, the American marten (
M. americana
), which included as subspecies both the Humboldt [Pine] marten subspecies (
M. a. humboldtensis
), and the former western marten species (
M. caurina
), classified as
M. a. caurina.

As noted above, at the time of our 90-day finding (77 FR 1900; January 12, 2012), the Humboldt marten was classified as
Martes americana humboldtensis.
Subsequently, Dawson and Cook (2012, entire) split the American marten, recognizing the Pacific marten (
M. caurina
) for all martens occurring west of the Rocky Mountain crest, based on genetic and morphological differences. Currently, the classification of the Humboldt marten in coastal northern California is
M. c. humboldtensis,
and the marten populations occurring in adjacent coastal Oregon are
M. c. caurina.
In addition, as currently recognized, populations of martens in the Oregon Cascades northward through the State of Washington and into British Columbia, Canada, are also
M. c. caurina.

Ongoing genetic research indicates uncertainty in the currently accepted Pacific marten subspecies delineations in California and Oregon. Specifically, the best available data indicate that the
Martes caurina humboldtensis
population in coastal northern California (Humboldt, Siskiyou, and Del Norte Counties) and the two known
M. c. caurina
populations in coastal Oregon (Curry, Coos, coastal portion of Douglas, coastal portion of Lane, Lincoln, and Tillamook Counties) are likely a single evolutionary unit (clade) (Slauson
et al.
2009a, p. 1,340; Schwartz and Slauson 2015, pers. comm.) (as noted in the scoping notice that published in the
Federal Register
on June 23, 2014 (79 FR 35509), and was made available for review at
http://www.regulations.gov
, Docket No. FWS-R8-ES-2014-0023). Although questions regarding the taxonomy of marten subspecies in northern California and Oregon are not new (
i.e.,
both the petition we received (CBD and EPIC 2010) and our 90-day finding (January 12, 2012; 77 FR 1900) identified ongoing genetic research and taxonomic uncertainty), the best available information indicate that the original designation of two separate marten subspecies occurring in coastal northern California and coastal Oregon is likely invalid (Schwartz and Slauson 2015, pers. comm.).

Listable Entity Evaluation and Distinct Population Segment Analysis

Based on the September 28, 2010, petition, and information received both prior and subsequent to our June 23, 2014, scoping notice regarding the listable entity, we considered whether the potential coastal DPS of Pacific marten meets the definition of a DPS as described in the Service's DPS Policy (61 FR 4722; February 7, 1996).

Section 3(16) of the Act defines the term “species” to include “. . . any subspecies of fish or wildlife or plants, and any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature.” We have always understood the phrase “interbreeds when mature” to mean that a DPS must consist of members of the same species or subspecies in the wild that would be biologically capable of interbreeding if given the opportunity, but all members need not actually interbreed with each other. A DPS is a subset of a species or subspecies, and cannot consist of members of a different species or subspecies. The “biological species concept” defines species according to a group of organisms, their actual or potential ability to interbreed, and their relative reproductive isolation from other organisms. This concept is a widely accepted approach to defining species. The Act's use of the phrase “interbreeds when mature” reflects this understanding. Use of this phrase with respect to a DPS is simply intended to mean that a DPS must be comprised of members of the same species or subspecies. As long as this requirement is met, a DPS may include multiple populations of vertebrate organisms even if they may not actually interbreed with each other. For example, a DPS may consist of multiple populations of a fish species separated into different drainages. While these populations may not actually interbreed with each other, their members are biologically capable of interbreeding.

The National Marine Fisheries Service (NMFS) and the Service published a joint Policy Regarding the Recognition of Distinct Vertebrate Population Segments Under the Endangered Species Act (DPS Policy on February 7, 1996 (61 FR 4722). According to the DPS Policy, two elements must be satisfied in order for a population segment to qualify as a possible DPS: discreteness and significance. If the population segment qualifies as a DPS, the conservation status of that DPS is then evaluated to determine whether it is endangered or threatened.

A population segment of a vertebrate species may be considered discrete 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; or (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 is found to be discrete, then it is evaluated for significance under the DPS Policy on the basis of its importance to the taxon to which it belongs. This consideration may include, but is not limited to, the following: (1) Persistence of the discrete population segment in an ecological setting unusual or unique to the taxon; (2) evidence that loss of the discrete population segment would result in a significant gap in the range of a taxon; (3) evidence that the population represents the only surviving natural occurrence of a taxon that may be more abundant elsewhere as an introduced population outside of its historical range; or (4) evidence that the population differs markedly from other populations of the species in its genetic characteristics.

If a population segment is both discrete and significant (
i.e.,
it qualifies as a potential DPS), its evaluation for endangered or threatened status is based on the Act's definitions of those terms and a review of the factors listed in section 4(a) of the Act. According to our DPS Policy, it may be appropriate to assign different listing classifications to different DPSs of the same vertebrate taxon.

We were petitioned to list collectively two groups of the Pacific marten (two populations in Oregon and one in California) that are currently recognized as belonging to two separate subspecies (as described above). To ensure that we evaluated the most accurate listable entity based on the best scientific and commercial data currently available (including unpublished genetics information), we published a scoping notice in the
Federal Register
on June 23, 2014 (79 FR 35509), notifying the public that we considered it reasonable that a coastal DPS of the Pacific marten constitute the listable entity for our status review.

We received eight comment letters from six entities in response to our June 23, 2014, scoping notice. Four entities agreed with our proposed DPS, one was silent, and one disagreed with our evaluation of a coastal DPS of the Pacific marten as the listable entity; two entities commented twice reiterating their same positions. The commenter who disagreed with the proposed coastal DPS of the Pacific marten as the listable entity believed more information, including genetics, would be required and that the entity we proposed would not be a valid DPS according to Service criteria. Following publication of the scoping notice in the
Federal Register,
we received more genetics information (Schwartz and Slauson 2015, pers. comm.) that supports our consideration of a coastal DPS of the Pacific marten.

After taking into consideration the comments received and conducting further evaluation of the best available scientific and commercial information (including additional genetics information), we confirm here that this DPS is a listable entity, including the currently recognized
Martes caurina humboldtensis
(
i.e.,
Humboldt marten) and the coastal populations of
M. caurina caurina
in Oregon (
i.e.,
Oregon Coast Range group). This entity is reasonable given:

(1) The best available data (
e.g.,
new genetics information, similar habitat usage) suggest that the coastal northern California marten population and the coastal Oregon marten populations represent a single evolutionary entity as opposed to two separate entities (Schwartz
et al.,
In prep.). In particular, Schwartz
et al.
(In prep.) has provided substantive information (with both mitochondrial and nuclear DNA evaluations) that the marten populations occurring in coastal northern California and coastal Oregon are unique and more closely related to each other than to other groups/populations of Pacific martens, to the extent that they are diagnosably distinct from all other Pacific martens.

(2) Existing genetics information (Slauson
et al.
2009a, entire) suggests that subspecies-level taxonomy of
M. c. humboldtensis, M. c. caurina,
and possibly other subspecies of the Pacific marten as currently classified may be inaccurate.

(3) The DPS Policy (February 7, 1996; 61 FR 4722) states that the population segment under consideration must be evaluated for discreteness and significance in relation to the remainder of the taxon to which it belongs. Ordinarily, in the present case we would evaluate the marten populations relative to the subspecies to which they belong, but the populations in question currently represent two separate subspecies and there is uncertainty as to the legitimacy of those subspecies classifications, rendering such an evaluation invalid.

(4) Uncertainty in the subspecies-level taxonomy of Pacific marten logically necessitates that we elevate our evaluation of the DPS relative to the Pacific marten at the full species level. In other words, we apply the criteria for evaluating a coastal DPS of the Pacific marten relative to the full species Pacific marten (
Martes caurina
) as a whole.

(5) The DPS Policy (February 7, 1996; 61 FR 4722) states that “In all cases, the organisms in a population are members of a single species or lesser taxon.” Therefore, given (1) through (4) above, an evaluation at the species level is appropriate. Consequently, for purposes of this Finding, below we evaluate the Pacific marten populations that occur in coastal Oregon and coastal northern California under our DPS Policy.

For this 12-month finding and DPS analysis of the Pacific marten populations that occur in coastal Oregon and coastal northern California, we reviewed and evaluated all available published and unpublished information, including numerous publications, reports, and other data submitted by the public. Marten distribution in coastal northern California and coastal Oregon is discussed in detail in the “Species Distribution” section of the Species Report titled “Coastal Oregon and Northern Coastal California populations of the Pacific marten (
Martes caurina
)” (Service 2015, pp. 28-32), which is available on the Internet at
http://www.regulations.gov
, Docket No. FWS-R8-ES-2011-0105.

Discreteness

Under the DPS Policy, a population segment of a vertebrate taxon may be considered discrete 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. As the marten populations in question here do not transcend an international boundary, this criterion does not apply.

As described below, the Pacific marten populations that occur in coastal Oregon and coastal northern California are markedly separated from other Pacific marten populations by geographical isolation (
i.e.,
separated by areas of unsuitable habitat), and marked genetic differences between those coastal populations (coastal Oregon and coastal northern California) and other populations of Pacific marten are evidence of this long-standing separation. The extant population in coastal northern California is separated from the Sierra marten subspecies (
Martes caurina sierrae
) by unsuitable habitat to the east in the Klamath River canyon. The coastal central Oregon extant population is separated from Pacific marten populations to the east (in the Oregon Cascade Mountains) primarily by unsuitable habitat within the Willamette Valley. Although some suitable habitat occurs between the coastal southern Oregon extant population area and the southern Cascades population of Pacific martens to the east, the distance to large blocks of suitable habitat in the southern Cascade Mountains far exceeds the mean maximum dispersal distance for martens (see discussion below). Additionally, martens that occur in coastal Oregon and coastal northern California occur in areas without significant, persistent snowpack (Slauson 2003, p. 66; Slauson
et al.,
In prep.). Mountain ranges to the east that have both unsuitable marten habitat and are covered by significant, persistent snowpack stand between the coastal Oregon and coastal northern California populations of Pacific martens and other Pacific marten populations (
e.g.,
separation of Humboldt and Sierra Nevada populations), thereby effectively isolating the coastal marten populations from other Pacific martens. East-west movements that would potentially connect Pacific marten populations in coastal Oregon and coastal northern California with inland Pacific marten populations are likely rare because:

(1) Most juvenile marten dispersal distances (that are published in literature) in both logged and unlogged forests range from less than or equal to 5 km (3.1 mi) (Broquet
et al.
2006, p. 1,694) to approximately 15 km (9.3 mi) (Phillips 1994, pp. 93-94; Pauli
et al.
2012, p. 393). The distance between the coastal Oregon and coastal northern California populations of Pacific martens and other Pacific marten populations to the east exceeds the likely maximum dispersal distance.

(2) Pacific martens within the three extant populations in coastal Oregon and coastal northern California likely only need to disperse short distances to establish a home range because there are typically sufficient amounts of unoccupied suitable habitat available within their natal area.

(3) Large patches of unsuitable habitat on the eastern edge of the historical range in this region would likely deter juvenile martens from moving east. As described below in the section
Summary of Species Information,
the coastal Oregon and coastal northern California populations of Pacific martens require a dense shrub understory comprised of shade-tolerant shrub species within the conifer-dominated overstory that they occupy (Zielinski
et al.
2001, p. 485; Slauson
et al.
2007, p. 464), and in coastal Oregon and coastal northern California, this dense shrub layer generally does not occur outside of the coastal fog-influenced areas. Thus, martens in coastal northern California and coastal Oregon are functionally isolated from other marten populations by their dependence on the dense shrub layer found in the coastal coniferous forests of this region.

The coastal Oregon and coastal northern California populations of Pacific martens are also markedly separated from other populations of the Pacific marten as evidenced by quantitative measures of genetic discontinuity. The Humboldt marten was historically distributed throughout the coastal coniferous forests of northern California from northwestern Sonoma County northward to the Oregon border (Grinnell
et al.
1937, pp. 207-210). Recent phylogenetic analyses using mitochondrial DNA (mtDNA) support the distinctiveness of the Humboldt marten subspecies, based on the presence of distinct haplotypes shared by historical museum specimens and martens currently occupying portions of the historical range in northern coastal California (Slauson
et al.
2009a, entire). Marten populations in coastal Oregon, which were historically described as
M. c. caurina,
also share these haplotypes, leading Slauson
et al.
(2009a, pp. 1338-1339) to suggest that martens in the Coast Range of Oregon may also be
M. c. humboldtensis.
Furthermore, preliminary results of a subspecific genetic evaluation of the Pacific marten by Schwartz
et al.
(In prep.)--using nuclear DNA (nDNA) and samples from substantially more martens than used by Slauson
et al.
(2009a)--demonstrate that the coastal Oregon and coastal northern California populations of Pacific martens are clearly distinguishable from other populations of Pacific marten on the basis of their genetic characteristics. Schwartz
et al.
(In prep.) indicate that coastal Oregon and northern coastal California marten populations represent a single evolutionary clade, calling into

question the separation of the original subspecies range boundaries (
i.e., M. c. humboldtensis
in northern coastal California and
M. c. caurina
in coastal Oregon) at the California-Oregon border. Although some low degree of introgression indicates occasional past movement of individuals between coastal and inland populations, the evidence suggests this was an infrequent occurrence (Schwartz
et al.,
In prep.); thus, the coastal Oregon and coastal northern California populations of Pacific martens are effectively genetically discrete from other populations of Pacific marten.

In summary, the best available information indicates that Pacific marten populations in coastal Oregon and coastal northern California are geographically isolated and genetically discrete from all other populations of the Pacific marten. Therefore, the marked separation condition for discreteness under our DPS Policy is met.

Significance

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) while encouraging the conservation of genetic diversity. In making this determination, we consider available scientific evidence of the DPS's importance to the taxon to which it belongs.

Because 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 (significance) to the taxon to which it belongs. This consideration of the population segment's significance may include, but is not limited to, the following:

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

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

(3) Evidence that the discrete population segment 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 discrete population segment 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 conditions. Other classes of information that might bear on the biological and ecological importance of a discrete population segment may also be used as appropriate, to provide evidence for significance, as described in the DPS Policy (61 FR 4722; February 7, 1996). At least two of the significance criteria are met for the marten populations in coastal Oregon and coastal northern California. First, we find that populations of Pacific martens in coastal Oregon and coastal northern California differ markedly from other populations of the Pacific marten species in their genetic characteristics. As described above under “Discreteness,” the coastal Oregon and coastal northern California populations of Pacific martens are genetically distinct from all other populations of Pacific martens (Schwartz
et al.,
In prep.). As a result, loss of the marten populations from coastal Oregon and coastal northern California would result in a reduction in Pacific marten genetic diversity. Second, we find that the loss of martens from coastal Oregon and coastal northern California would result in a significant gap in the range for the Pacific marten. The coastal populations of martens in California and Oregon represent the only coastal populations of Pacific martens in these States and inhabit a habitat association unique from other non-coastal marten populations—that is, areas consisting of occasional, non-persistent snowpack (below 914 meters (m) (3,000 feet (ft)) with a mesic, shade-tolerant shrub layer (understory) within coastal coniferous forest habitat (see “Life History” section of the Species Report). The requirement of this dense (greater than 70 percent cover), shrubby understory is particularly unusual for martens, and is a unique habitat association not described elsewhere in the distribution of either Pacific martens or American martens in North America (Slauson
et al.,
In prep.(a)). The coastal Oregon and coastal northern California populations of Pacific martens are also the only martens known to utilize coastal serpentine habitat and dune forest habitat distributed on coastal terraces. These genetic differences and the evidence that a significant gap in the range of the taxon would result from the loss of the discrete population segment both individually satisfy the significance criterion of the DPS Policy. Therefore, under the Service's DPS Policy, we find that the populations of Pacific martens in coastal Oregon and coastal northern California are significant to the taxon to which they belong.

Conclusion of DPS Analysis Regarding Pacific Martens in Coastal Oregon and Coastal Northern California

As stated above under
Current Taxonomic Description,
the best available scientific and commercial information suggests that the coastal Oregon populations of Pacific marten (
Martes caurina caurina
) are likely the same entity as the currently classified Humboldt marten (
M. c. humboldtensis
). We find that the coastal Oregon and coastal northern California populations of Pacific martens collectively constitute a valid DPS under the Service's DPS Policy because this population segment is both discrete and significant to the taxon to which it belongs. We therefore consider the coastal Oregon and coastal northern California populations of Pacific martens collectively as the “coastal DPS of the Pacific marten,” which constitutes the listable entity for this status review. Throughout this document when we use the term “coastal marten,” we are using this term as shorthand for the coastal DPS of the Pacific marten.

Summary of Species Information

A thorough review of the taxonomy, life history, biophysical environment, habitat use, distributions, and population abundance/trends of the coastal DPS of Pacific marten is presented in the Species Report (Service 2015, pp. 1-40) available on the Internet at
http://www.regulations.gov
, Docket No. FWS-R8-ES-2011-0105). A summary of this information is presented below. We used data specific to coastal marten populations when they were available; when such information was lacking, we relied on information regarding North American martens in general (American or Pacific martens), and have made these distinctions in the text that follows.

Life History

Two species of marten, divided into 14 total subspecies, inhabit North America. Collectively, North American martens are characterized by the long and narrow body type typical of the mustelid family (Mustelidae;
e.g.,
weasels, minks, otters and fishers), overall brown pelage (fur) with distinctive coloration on the throat and upper chest that varies from orange to yellow to cream, large and distinctly

triangular ears, and a bushy tail that is proportionally equivalent to about 75 percent of the body length (Clark
et al.
1987, p. 2; Powell
et al.
2003, p. 636).

Marten activity patterns coincide with their prey species availability. Specifically, martens are active year-round and seasonally adjust their activity patterns to synchronize with those of their key prey species (Zielinski
et al.
1983, pp. 387-388). Overall, the diet of North American marten species is dominated by mammals, but birds, insects, and fruits are seasonally important (Martin 1994, pp. 298-301). Diet analysis for the coastal marten is currently limited to scats collected from the coastal northern California population during summer and fall, and includes mammals, berries, birds, and reptiles (Slauson and Zielinski, In prep.). Sciurid (members of the squirrel family) and cricetid rodents (
i.e.,
New World rats and mice) dominate the coastal marten's diet, with the most frequent prey species being chipmunks (
Tamias
spp.) and red-backed voles (
Myodes californicus
), and, to a lesser extent, Douglas squirrels (
Tamiasciurus douglasii
) and flying squirrels (
Glaucomys sabrinus
) (Slauson and Zielinski, In prep.).

Information on coastal marten reproduction and survivorship is lacking; therefore our analysis is based on knowledge of North American martens in general, which are polygamous mammals. Female martens mate no sooner than 15 months of age and first litters are produced no sooner than 24 months of age (Strickland
et al.
1982, p. 601). Mating occurs from late June to early August (Markley and Bassett 1942, pp. 606-607), and females give birth in March and April (Strickland
et al.
1982, p. 602). Female martens are capable of producing from one to five kits per litter, but the modal average is two to three (Strickland and Douglas 1987, p. 602; Mead 1994, p. 410). Information is not available on the average number of young raised to weaning, the average number of young recruited into the population per female, or the effects of annual variation in environmental conditions and prey populations on kit survival. Regarding longevity, captive Pacific martens are known to reach 15 years of age (Clark
et al.
1987, p. 3); however, data from American marten individuals in the wild in the Algonquin Region of Ontario, Canada, indicate that 10 percent (of 2,076 females trapped) were more than 5 years old (Strickland and Douglas 1987, p. 535). Finally, age structure of coastal martens has not been studied, although the best available information from an untrapped population of Pacific martens in the Sierra Nevada mountains indicates relatively consistent proportions of yearling and adult age classes (Slauson
et al.,
In prep.(a)).

Juvenile dispersal of the American marten is generally thought to occur as early as August, although fall, winter, and spring (the year after birth) dispersal periods have been reported (Clark and Campbell 1976, p. 294; Slough 1989, p. 993). Juvenile dispersal in coastal northern California and Sierra Nevada martens has been observed to occur as early as August and continues at least until the following summer season (Slauson and Zielinski 2014, unpubl. data). Information is not available regarding the timing of juvenile dispersal for coastal martens in Oregon. Pauli
et al.
(2012, p. 393) found that Pacific and American martens exhibit similar dispersal distances, averaging 15.5 km (9 mi). Most studies find that the majority of juvenile martens disperse relatively short distances to establish home ranges, ranging from less than or equal to 5 km (3.1 mi) (Broquet
et al.
2006, p. 1,694) to approximately 15 km (9.3 mi) (Phillips 1994, pp. 9394; Pauli
et al.
2012, p. 393). However, Broquet
et al.
(2006, p. 1695) also describe juvenile martens as capable of covering long distances during dispersal, up to 82 km (50 mi) in their study. Other researchers have reported instances of dispersal movements by martens ranging from 40 to 80 km (25 to 50 mi) (Thompson and Colgan 1987, pp. 831-832; Fecske and Jenks 2002, p. 310), up to 149 km (93 mi) or even 160 km (100 mi) in distance (Slough 1989, p. 993; Kyle and Strobeck 2003, p. 61). Based on minimal genetic structuring of marten populations in a heavily harvested forest landscape, Kyle and Strobeck (2003, pp. 60-61) suggested that habitat fragmentation may not necessarily impede marten movement to the degree formerly understood. However, Kyle and Strobeck (2003, p. 65) also caution that smaller scale disturbances may still act as partial barriers to marten gene flow. Johnson (2008, pp. 33-36) found that juvenile martens traveled slower, shorter distances, and suffered twice the mortality risk in logged versus unlogged landscapes. Therefore, the best available information suggest that landscape condition (
e.g.,
the spatial distribution of unlogged and logged stands) has important effects on dispersal dynamics, affecting both the distance dispersers can travel and the success rate they have in establishing home ranges and surviving to adulthood.

Intraguild predation and interspecific competition occurs naturally within the range of the coastal DPS of Pacific marten. Intraguild predation refers to killing and eating of potential competitors that utilize the same prey resources. Interspecific competition is a form of competition in which individuals of a different species compete for the same resource in an ecosystem (as opposed to intraspecific competition that involves organisms of the same species). Martens are susceptible to predation by larger mammalian and avian predators, typically habitat-generalist species, including coyote (
Canis latrans
), red fox (
Vulpes vulpes
), bobcat (
Felis rufus
), fishers (
Pekania pennanti
), and great horned owl (
Bubo virginianus
) (Thompson 1994, p. 276; Lindstrom
et al.
1995, entire; Bull and Heater 2001, p. 4; McCann
et al.
2010, p. 11). Marten predators may vary depending on the quality of the habitat. For example, American marten populations in highly altered forest landscapes show higher rates of predation by habitat generalist carnivores (and lower annual survival rates) than those in less-altered forest landscapes (Thompson 1994, p. 278)). Because marten populations are strongly influenced by adult and juvenile survivorship (Buskirk
et al.
2012, p. 89), predation of martens can have a meaningful effect on abundance and population growth rates. Additional discussion on predation as a stressor on the coastal marten is provided below in Summary of Information Pertaining to the Five Factors.

Habitat Description

The preferred habitat type for the coastal DPS of Pacific marten occurs in some of the most productive forests in the world. In unmanaged, late-seral stages, these forests are typically composed of long-lived, large trees, with multi-layered canopy structure, substantial large woody debris (standing and downed), and abundant ferns, herbs, and shrubs on the forest floor (Sawyer
et al.
2000, entire; Chappell
et al.
2001, entire; Sawyer 2007, entire; DellaSala
et al.
2011, entire). The forests are largely coniferous and typically dominated by coast Douglas-fir (
Pseudotsuga menziesii menziesii
), western hemlock (
Tsuga heterophylla
), and Sitka spruce (
Picea sitchensis
) in Oregon, and redwood (
Sequoia sempervirens
) and coast Douglas-fir in California (Ricketts
et al.
1999, entire; Sawyer 2007, entire). Higher elevation areas also include sub-dominant conifers such as western red cedar (
Thuja plicata
), Port Orford-cedar (
Chamaecyparis lawsoniana
), grand fir (
Abies grandis
), sugar pine (
Pinus lambertiana
), and white fir (

Abies

concolor

) (Chappell
et al.
2001, entire; Sawyer 2007, entire). Hardwood-dominated stands are uncommon, although hardwood species such as tanoak (
Notholithocarpus densiflorus
), golden chinquapin (
Chrysolepis chrysophylla
), and Pacific madrone (
Arbutus menziesii
) are common canopy subdominants. Red alder (
Alnus rubra
) can occur as an early successional overstory dominant in the uplands in some near-coast locations or post-logging sites. Riparian forests are dominated by broadleaf species such as red alder, black cottonwood (
Populus trichocarpa
), bigleaf maple (
Acer macrophyllum
), and mesic shrub species such as vine maple (
A. circinatum
).

A dense understory of shrubs and herbaceous plants are a key habitat requirement for the coastal marten (see “Habitat Use” section of the Species Report (Service 2015, pp. 18-27)). Species presence and dominance is shaped largely by the combination of soil nutrients and moisture, with herbaceous species such as sword fern (
Polystichum munitum
) dominating on nitrogen rich or very moist sites, and evergreen shrubs such as Pacific rhododendron (
Rhododendron macrophyllum
) and salal or wintergreen (
Gaultheria
sp.) dominating on nutrient poor or drier sites (Chappell and Kagan 2001, entire). Other dominant or co-dominant understory shrub species include evergreen huckleberry (
Vaccinium ovatum
), salmonberry (
Rubus spectabilis
), red huckleberry (
Vaccinium parvifolium
), and in serpentine habitats (see description below) dwarf tanbark (
Notholithocarpus densiflorus
var.
echinoides
) and huckleberry oak (
Quercus vaccinifolia
) (Jimerson
et al.
1996, pp. A13-A15; Sawyer
et al.
2000, entire; Chappell
et al.
2001, entire). Many of the dominant shrub species are adapted to fire by having lignotubers, which are basal swellings at the interface between the roots and shoots usually just below the soil surface, allowing these species to quickly sprout after fire kills the shoots and thus maintain site dominance (Agee 1993, p. 133).

Two additional, rare forest habitats are of particular relevance to coastal martens: Coastal serpentine and coastal dune forest. Forests in serpentine habitats are typically open and rocky with stunted trees that contrast sharply with the dense, rapidly-growing stands on more productive, non-serpentine soils that surround these sites (Jimerson
et al.
1995, pp. A8-A31). Martens are not known to occupy these more open, drier, interior areas. However, on the extreme coastal edge of the serpentine habitats that occur in coastal northern California and coastal Oregon, increased moisture and summer fog supports dense, spatially-extensive shrub layers; coastal martens have been found in this wetter variant of coastal serpentine habitat in both Oregon and California. The serpentine communities used by coastal martens are composed of a variety of coniferous trees, such as Douglas-fir, sugar pine, lodgepole pine (
Pinus contorta
), western white pine (
P. monticola
), Jeffrey pine (
P. jeffreyi
), knobcone pine (
P. attenuatta
), and Port Orford-cedar, and are dominated by mast-producing shrubs such as dwarf tanbark, huckleberry oak, and red huckleberry (Jimerson
et al.
1995, p. C1; Slauson 2003, pp. 5, 9, 13). The coastal dune forest communities where coastal martens have been found are predominantly in coastal Oregon and are typically dominated by shore pine (
P. contorta contorta
), the coastal form of lodgepole pine, and in some areas co-dominated by Sitka spruce occurring in stabilized dunes on marine terraces. Although martens have been found in these less-common habitat types, it is important to note that the more extensive dominant forest types (
i.e.,
coastal coniferous forests) support the majority of the historical marten distribution in coastal Oregon and coastal northern California.

Coastal martens select habitat at four primary spatial scales: Micro-scale (resting and denning structures), stand-scale, home range, and landscape-scale (facilitating movement, occupancy, and population dynamics).

(1) Micro-scale—Rest structures are used daily by martens between foraging bouts to provide thermoregulatory benefits and protection from predators (Taylor and Buskirk 1994, pp. 253-255). Reuse rates for individual rest structures are low and selection for structure type changes seasonally to meet thermoregulatory needs (
e.g.,
Spencer 1987), such that multiple resting structures meeting seasonal requirements are required across the home range. Large-diameter live trees, snags, and logs provide the main types of resting structures for martens (Spencer
et al.
1983, pp. 1182-1185; Schumacher 1999, pp. 26-58; Slauson and Zielinski 2009, pp. 41-42). Denning structures used by female martens to give birth to kits are called natal dens, and the subsequent locations where they move their kits are referred to as maternal dens. Ruggiero
et al.
(1998, pp. 665-669) found that both the characteristics of the den structures and the characteristics of the stands in which they were found influenced den-site selection. This is likely due to the importance of high-quality foraging habitat in close proximity to den sites, allowing females to simultaneously maximize the energy they gain from foraging during lactation and minimize the time spent away from kits, especially when they are dependent on their mothers for thermoregulation. The most common den structures used by Pacific and American martens are large-diameter, live and dead trees with cavities (Thompson
et al.
2012, p. 223).

(2) Stand-scale—Martens select forest stands that provide habitat structure supporting one or more life history needs that include foraging, resting, or denning. Coastal martens in California most strongly selected stands of old-growth, conifer-dominated forests with dense shrub layers (Slauson
et al.
2007, pp. 464-465). Other than the late-mature developmental stage, which was used in proportion to its availability, stands in earlier developmental stages were selected against (Slauson
et al.
2007, pp. 462-464). These old-growth and late-mature stands most often were dominated by Douglas-fir overstory, but also had mature hardwood understories composed of either tanoak or golden chinquapin. Shrub layers were dense (greater than 70 percent cover), spatially extensive, and dominated by evergreen huckleberry, salal, and rhododendron (Slauson
et al.
2007, p. 465). The majority of detections of martens in coastal southern Oregon share these same stand characteristics (Zielinski
et al.
2001, p. 485).

(3) Home Range—Pacific and American martens exhibit strong habitat selection at the home range scale, suggesting that this scale of selection most directly influences an individual's fitness (Thompson
et al.
2012, p. 210). Martens establish home ranges to encompass their year-round resource needs and, during the breeding season, gain access to members of the opposite sex. Marten home ranges are often positioned to maximize high-quality habitat (typically greater than 70 percent high-quality, late-successional forest (reviewed in Thompson
et al.
2012, p. 218)) and to minimize low-quality habitat (
e.g.,
recent clear cuts, partial harvest) (Phillips 1994, pp. 59-60). Females, due to their solitary role raising young, have unique needs that require access to suitable den sites located near reliable and nearby prey resources to support the energetic demands of lactation and providing food for kits. In coastal northern California, Slauson and Zielinski (2014, unpubl. data) found 97 percent (38 of 39) of the female within-home-range resting and active locations occurred in the core old-growth and late-mature

riparian habitat patches. In comparison, 77 percent (30 of 39) of the male within-home-range resting and active locations occurred in the core old-growth and late-mature riparian habitat patches (Slauson and Zielinski 2014, unpubl. data). Also of note is that there is an inverse relationship between the amount of high-quality habitat and marten home range size (
i.e.,
as the amount of high-quality habitat decreases, home range size increases) (Thompson 1994, p. 276; Potvin and Breton 1997, p. 462; Fuller and Harrison 2005, pp. 715-719).

(4) Landscape-scale—The pattern and composition of habitat at this scale affects: (a) The ability of martens to successfully disperse and find suitable home ranges; (b) survival and species occurrence over time and space; and (c) ultimately, population size and persistence. Successful dispersal requires the existence of functional habitat connectivity between patches of habitat suitable for reproduction to maintain or expand population size and distribution. Also, during dispersal, martens use a search strategy that is not random or linear, suggesting they are responding to habitat cues and that landscape pattern likely influences movement trajectories (Johnson 2008, pp. 27-29, 36-39). Compared to other species closely associated with late-successional forest, American and Pacific marten populations, including the coastal marten, are sensitive to the loss or fragmentation of high-quality habitat at the landscape scale. For example, martens exhibit a progression of responses to timber harvest as the proportion of habitat affected by intensive logging activities increases. Such activities include, but are not limited to, clear cutting (see review in Thompson
et al.
2012), partial harvest (Potvin
et al.
2000, pp. 851-854; Fuller and Harrison 2005, pp. 715-716; Godbout and Ouellet 2008, pp. 336-338), and shelterwood cutting (Ellis 1998, p. 41-49). As a result, the combination of habitat loss and fragmentation of remnant suitable habitat effectively lowers the density of martens by reducing the number of home ranges that can be supported (Thompson 1994, p. 276).

Historical and Current Distribution of Coastal Martens and Suitable Habitat

At the time of European settlement, the coastal marten occurred in all coastal Oregon counties and the coastal northern counties of California within late-successional coniferous forests. The majority of historical (pre-1980) verifiable marten detections (
i.e.,
occurrence records supported by direct physical evidence such as tracks, photographs, and carcasses) were within the fog-influenced coastal coniferous forest as opposed to interior forests (Grinnell and Dixon 1926, p. 413). Specifically, Slauson and Zielinski (2007, p. 241) reported 83 percent of the coastal northern California marten historical records occurring less than 25 km (15 mi) from the coast and no records occurring greater than 35 km (22 mi) from the coast, while our analysis (see Service 2015, pp. 6, 31) revealed greater than 90 percent of the coastal Oregon marten historical records occurring closer to the coast than to the interior portions of the coastal marten's range. Historical abundance of coastal martens is unknown. However, as is typical of mammalian carnivores, coastal martens likely never occurred in high densities.

Unregulated fur trapping occurred throughout the coastal marten's historical range, and by the late 1920s, few marten were captured where they were once considered relatively abundant (Zielinski and Golightly 1996, entire). A marked decline in the number of coastal marten harvested in coastal northern California led to the closure of marten trapping in northwestern California in 1946. In Oregon, marten fur trapping remains legal Statewide. Historical fur trapping is thought to have resulted in a significant contraction of coastal marten distribution and the extirpation of coastal marten from large portions of its historical range. Although we can make conclusions about the general historical distribution of coastal martens, information on historical population size is not available, thus precluding an accurate assessment of the impact of unregulated trapping on coastal marten population abundance.

Due to the lack of surveys for coastal martens, little information is available regarding their current distribution; this is particularly true for coastal Oregon. We do know, however, that there are at least three extant populations of coastal martens, one in coastal northern California, one in coastal southern Oregon, and one in coastal central Oregon, as described in detail below, and we have information regarding the extent of suitable habitat that is currently available to coastal martens throughout their range. It is therefore possible that coastal martens may occur in any of these areas of suitable habitat that have not been surveyed, or have been surveyed only sporadically. Here we briefly describe the areas of suitable habitat available to coastal martens.

Slauson
et al.
(In prep.(b)) developed a landscape habitat suitability model that we used to assess how much suitable habitat is currently available to coastal martens. The model was developed by identifying the combination of environmental, topographic, disturbance history, and vegetation variables that best described the distribution of marten detection/non-detection survey data. Specifics regarding model development and variables can be found in the “Current Landscape Habitat Suitability” section of the Species Report (Service 2015, pp. 26-27). The model categorizes the landscape into low, medium, and high suitability classes representing the relative probability of marten occupancy of habitat at the landscape scale.

Model results indicate that approximately 41 percent of the coastal marten's historical range contain suitable habitat (described as low, medium, and high suitability habitat) for coastal martens (see “Current Landscape Habitat Suitability” section of the Species Report). The model identified approximately 59 percent of the remaining lands within the historical range of the coastal marten to be unsuitable, which includes (but is not limited to) forested habitat that is not utilized by martens (
e.g.,
heavily managed timber lands), urban and suburban developments, and agricultural lands. However, it is important to note that, for the purposes of this analysis, we considered “low suitability habitat” as defined in this model to be “unsuitable” when examining the current and long-term stressors to the coastal marten and its habitat into the future. In other words, in evaluating stressors to the coastal marten and its habitat, we considered only areas that provide moderate- to high-suitability habitat as identified by the model. We came to this conclusion based on feedback from the species experts (Slauson
et al.,
In prep.(a)) who indicate that these “low suitability habitat” areas currently have a low probability of coastal marten occurrence. Including these areas as suitable habitat for the purposes of this analysis would bias the amount of actual suitable habitat present both currently and in the future.

Much of the coastal marten's historical habitat has been lost. Extensive logging of old-growth redwood habitat in coastal northern California began in the late 1800s, and coincided with unregulated fur trapping. Late-successional coniferous forests in coastal Oregon were also extensively harvested in the early 1900s. Currently, less than 5 percent of the redwood forests existing at the time of European settlement remain within the

historical range of the coastal marten in coastal northern California (Save the Redwoods League 2015, no page number). Based on the best available information, much of the coastal coniferous forest habitat in both States, especially within a few miles of the coast, appears to be currently owned (in general) by either private industrial timber companies or smaller land owners, and managed for timber production.

Within the coastal marten's historical range, the majority of remaining late-successional coniferous forests suitable for the coastal marten is within national forests, and national and State parks. Where martens are known to occur, relatively high amounts of moderate- to high-suitability habitat are still found, and much of this habitat occurs in areas that are managed for the maintenance or enhancement of late-successional forest conditions that are beneficial to coastal martens. For example, approximately 71, 79, and 90 percent of the total available suitable habitat on Federal lands in the coastal central Oregon, coastal southern Oregon, and coastal northern California population areas, respectively, occur within the Northwest Forest Plan (NWFP) Federal reserve lands, which are designed to retain and accelerate the development of late seral characteristics. Currently, the largest contiguous blocks of suitable coastal marten habitat occur within the Six Rivers National Forest in the extreme northern portion of the historical range in California, and in the adjacent Siskiyou portion of the Rogue River-Siskiyou National Forest in the southern portion of the historical range in Oregon. Large blocks of suitable habitat also occur in coastal central Oregon on the Siuslaw National Forest. Little suitable habitat is currently found in the southern half of the historical range in California. In the coastal northern portion of the historical range in Oregon, suitable habitat is limited to a narrow band along the coast. Finally, in the area between the Siskiyou and Siuslaw National Forests in the historical range in Oregon, there is some limited amount of suitable habitat on BLM ownership. Habitat conditions specific to each of the known extant population areas of coastal martens are discussed below.

Distribution and Abundance of Current Known Extant Populations

There are three known extant populations of coastal martens in coastal central Oregon, coastal southern Oregon, and coastal northern California, according to the best available scientific and commercial data (Figure 1; see section 8.1.2 (Delineation of Extant Population Areas) of the Species Report (Service 2015, p. 32)). These populations have been described as disjunct (
e.g.,
Slauson and Zielinski 2009, pp. 35-36). Survey effort has been limited in some portions of the coastal marten's range, however. Therefore, it is unknown whether additional coastal martens may be found in areas that have not yet been surveyed. In addition, a few coastal marten verifiable detections occur outside these three population areas, but these martens are currently not considered part of any known viable population (Slauson
et al.,
In prep.(a)). Surveys for martens have occurred in much of the California portion of the historical range and suitable interior habitat in southwestern Oregon, although minimal survey effort has occurred in coastal central Oregon and no surveys have occurred in coastal northern Oregon (see Figure 8.2 in the Species Report).

BILLING CODE 4310-55-P

EP07AP15.011

BILLING CODE 4310-55-C
Coastal Central Oregon Extant Population Area

This 4,150-km
2
(1,602-mi
2
) population area includes all coastal-draining watersheds from the Umpqua River north to the Yaquina River in Lincoln, Benton, western Lane, western Douglas, and northwestern Coos Counties. Lands within this extant population area are owned/managed by Siuslaw National Forest (41 percent), private landowners (40 percent), Bureau of Land Management (BLM; 10 percent), and Oregon Department of Forestry (ODF) and Oregon State Parks (9 percent). A total of approximately 2,348

km
2
(907 square miles (mi
2
); 56 percent) of the extant population area contains moderate- and high-suitability habitat (Service 2015, p. 33) for coastal martens. Of the currently available moderate- and high-suitability habitat, 23 percent is in private ownership and 71 percent is in Federal ownership, and 71 percent of the Federal lands are in Reserves, which are managed for late-seral characteristics (Service 2015, p. 76). The best available information suggests that most of the private forest land is owned by private, industrial timber companies (Lettman 2011, p. 33).

This population area comprises approximately 20 percent coastal marten habitat of high suitability, 36 percent of moderate suitability, 22 percent of low suitability (which has low probability of coastal marten occurrence currently and into the future), and 21 percent unsuitable (Slauson
et al.,
In prep.(b)). In total, suitable marten habitat composes 78 percent of the population area. However, we note that the model (which used data from northwest California and southwest Oregon) generated suitable habitat values for this population area that did not include coastal dune habitat, which is considered suitable for coastal martens based on visual observations and the presence of several verifiable marten detections (Slauson
et al.,
In prep.(a)). Thus the amount of potentially suitable habitat for coastal martens identified by the habitat model is an underestimate for this population area.

Population abundance information is not available for the coastal central Oregon population of coastal martens. Although only a single station had been surveyed in this population area since the late 1980s, presence/absence surveys began in this area in the summer of 2014. One marten was detected in 2014 (Slauson
et al.
2014, unpubl. data), and six more were detected in January and February 2015; as of the time of this publication, surveys in this area are ongoing (Moriarty 2015, pers. comm.). The area surveyed represents only about 4 percent of the currently delineated coastal central Oregon population area described herein, and 2014 was the first year of survey effort in this area. Based on the results to date and the availability of suitable habitat in this area, it is likely that more martens will be detected in this area as surveys continue.

Abundance or trend information is not available for any populations of coastal martens in Oregon. Although researchers note that martens in this area have likely declined relative to their historical condition, they cite to insufficient historical or contemporary data to allow evaluation of the status of martens in the coastal mountain ranges of central and northern Oregon (Zielinski
et al.
2001, p. 486). There are no data available for estimating current population abundance or trend for the coastal central Oregon population, and although survey efforts recently began in this area, data from these surveys will only be informative in terms of establishing presence or absence of coastal martens. Zielinski
et al.
(2001, pp. 486-487) could only suggest that marten numbers may be relatively low on the northern Oregon coast, based on the absence of reported road kills along coastal Highway 101 in this area, in contrast to several road-killed martens reported from the same highway in central Oregon. In sum, although coastal martens have likely declined relative to their historical abundance due to the past effects of overtrapping and timber harvest (Zielinski
et al.
2001, p. 487), there are no empirical data on which to base an estimate of either current population abundance or trend of martens on the central Oregon coast.

Coastal Southern Oregon Extant Population Area

This 4,696-km
2
(1,813-mi
2
) population area includes Chetco River, Pistol River, south Fork Rough and Ready Creek, and the North Fork Smith River watersheds in Curry, western Josephine, and southern Coos Counties. Lands within this population area are owned/managed by Rogue River-Siskiyou National Forest (78 percent), private landowners (13 percent), BLM (8 percent), and ODF (less than 1 percent). A total of approximately 3,641 km
2
(1,406 mi
2
; 78 percent) of the extant population area contains moderate- and high-suitability habitat (Service 2015, p. 35). As stated above for the coastal central Oregon population area, present moderate- and high-suitability habitat on private lands is expected to be harvested or not likely to retain late-seral characteristics into the future. Of the currently available moderate- and high-suitability habitat in the coastal southern Oregon population area, 10 percent is private ownership and 90 percent is Federal ownership, and 79 percent of the federally managed lands are Federal Reserves, which are managed for late-seral characteristics (Service 2015, p. 76). The best available information suggests that most of the private forest land is owned by private, industrial timber companies (Lettman
et al.
2011, p. 33).

This population area comprises approximately 52 percent coastal marten habitat of high suitability, 26 percent of moderate suitability, 17 percent of low suitability, and 5 percent unsuitable (Slauson
et al.,
In prep.(b)). In total, suitable marten habitat composes 95 percent of the population area.

Similar to the situation for the coastal central Oregon population, described above, population abundance information is not available for the coastal southern Oregon population of coastal martens. Although extensive grid-based surveys (which are used to estimate marten abundance or presence/absence) have not been conducted for this population, grid-based surveys began in this area in the summer of 2014. No coastal martens were detected in 2014 (Slauson
et al.
2015, unpubl. data), but surveys just beginning at the time of this publication have yielded a single marten detection (Moriarty 2015, pers. comm.). The area surveyed represents only a small portion of the currently delineated coastal southern Oregon population area described herein, and 2014 represented the first year of survey effort in this area. At this time, similar to the coastal central Oregon population area, there are no empirical data on which to base an estimate of either current population abundance or trend of martens on the southern Oregon coast.

Coastal Northern California Extant Population Area

This 812-km
2
(313-mi
2
) population area includes the south Fork of the Smith River, Blue Creek, Bluff Creek, Camp Creek, Cappell Creek, Pecwan Creek, Slate Creek, and Rock Creek (Siskiyou County, north of Orleans, California) watersheds in Del Norte, northern Humboldt, and western Siskiyou Counties. Lands within this population area are owned/managed by the U.S. Forest Service (Forest Service) (Klamath National Forest and Six Rivers National Forest; 65 percent); the Yurok Tribe of the Yurok Reservation, California (Yurok Tribe; 23 percent); private landowners, primarily Green Diamond Resource Company (11 percent); and Redwood National and State Parks (1 percent). A total of approximately 656 km
2
(253 mi
2
; 81 percent) of the extant population area contains moderate- and high-suitability habitat (Service 2015, p. 75). Currently present moderate- and high-suitability habitat on private lands is expected to be harvested or not likely to retain late-seral characteristics into the future. Of the currently available moderate- and high-suitability habitat in the coastal northern California population area, 11 percent is private ownership and 77

percent is Federal ownership, and 90 percent of the federally managed lands are Federal Reserves, which are managed for late-seral characteristics (Service 2015, p. 75). The best available information suggests that most of the private land is owned by private, industrial timber companies (Service 2014, unpubl. data).

This population area comprises approximately 67 percent coastal marten habitat of high suitability, 14 percent of moderate suitability, 7 percent of low suitability, and 12 percent unsuitable (Slauson
et al.,
In prep.(b)). In total, suitable marten habitat composes 88 percent of the population area.

As reported in 1996 by Zielinski and Golightly (1996, entire), this coastal northern California population has apparently recovered from numbers that were once so low (in the 50 years prior to 1995) that it was considered to be extremely rare or extinct. Martens in coastal northern California were first surveyed to estimate abundance in 2000-2001, and again in 2008 (Slauson
et al.
2009b, p.11) and 2012 (Slauson
et al.
2014, unpubl. data). A total of 31.5 martens (95 percent confidence interval = 24-40) were estimated for 2000-2001, and 20.2 martens (95 percent confidence interval = 11-30) were estimated for 2008, which represents a 42 percent decline in occupancy between those two time periods (Slauson
et al.
2009b, pp. 10, 11). In 2012, all locations sampled in 2008 were resampled (Slauson
et al.,
In prep.(a)). Preliminary occupancy estimates for the 2012 sampling were similar to results from 2008 (Slauson
et al.,
In prep.(a)), suggesting no further changes in marten population abundance in northern coastal California between 2008 and 2012. Slauson
et al.
(2009b, p. 13) advised that these population estimates should be considered minimum estimates because the sampling area did not fully cover all potentially occupied habitats; therefore, they suggested more realistic population estimates should be doubled (
i.e.,
60 coastal martens in 2000-2001, and 40 in 2008). Based on these samples, Slauson
et al.
(2009b, p. 13) concluded that as of 2008, it was likely that the entire coastal northern California population of martens contained fewer than 100 individuals. As noted above, subsequent survey efforts in 2012 indicated no further changes in estimated population size since that time; therefore, the best available data (preliminary estimates from surveys in 2012) suggest that the current population estimate for the coastal northern California population is similar to the estimate for 2008 (
i.e.,
fewer than 100 individuals).

Summary of Information Pertaining to the Five Factors

Section 4 of the Act (16 U.S.C. 1533) and implementing regulations (50 CFR 424) set forth procedures for adding species to, removing species from, or reclassifying species on the Federal Lists of Endangered and Threatened Wildlife and Plants. Under section 4(a)(1) of the Act, a species may be determined to be an endangered or threatened species based on any of the following 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.

In making this finding, information pertaining to the coastal DPS of the Pacific marten in relation to the five factors provided in section 4(a)(1) of the Act is discussed below. In considering what factors might constitute threats to a species, we must look beyond the mere exposure of the species to a particular factor to evaluate whether the species may respond to that factor in a way that causes actual impacts to the species. If there is exposure to a factor but no response, or only a positive response, that factor is not a threat. If there is exposure and the species responds negatively, the factor may be a threat and we then attempt to determine if that factor rises to the level of a threat, meaning that it may drive or contribute to the risk of extinction of the species such that the species warrants listing as an endangered or threatened species as those terms are defined in the Act. However, the identification of factors that could impact a species negatively is not sufficient to compel a finding that the species warrants listing. The information must include evidence sufficient to suggest that these factors are operative threats that act on the species to the point that the species meets the definition of an endangered or threatened species under the Act.

Potential stressors that may impact coastal martens in coastal Oregon and coastal northern California include actions that may affect marten individuals or populations (
i.e.,
trapping (for fur and research purposes), predation, disease, collision with vehicles, and exposure to toxicants) and actions that may lead to the loss, degradation, or fragmentation of suitable marten habitat (
i.e.,
wildfire, climate change, vegetation management, and development). To provide a temporal component to our evaluation of potential stressors (
i.e.,
impacts into the future), we first determined whether we had data available that would allow us to reasonably predict the likely future impact of each specific stressor over time. Where such data were available, we made predictions of future conditions over a period of time specific to that stressor (
i.e.,
wildfire, climate change, as described below). If we did not have such stressor-specific data available, we used IUCN's standard 3-generation timeframe to assess risk (International Union for Conservation of Nature (IUCN) 2014, pp. 14-21). Using a calculated marten generation time of 5 years (see the Species Report for more information on calculating marten generation time), this translated to a timeframe of 15 years, which we used in analyzing the foreseeable future for the majority of the stressors discussed below. This time period allows for analysis of multiple generations of coastal martens over a reasonable time period, as opposed to examining further into the future where assumptions or extensive uncertainty would not allow meaningful projections of potential future impacts.

To assess the stressor of wildfire, we used a longer future period consisting of 30 years based on more extensive data available regarding wildfires from the past approximate 30 years. This information was used to predict the future equivalent level of expected fire frequency, size, and severity. Using a longer foreseeable future timeframe for wildfire better incorporates the range of fire-related activity that may occur within the coastal Oregon and coastal northern California population areas. To assess the stressor of climate change, we used a longer foreseeable future period of 40-50 years, which coincides with the model projection timeframes available for climate change (
e.g.,
changes in temperature and precipitation) in coastal Oregon and coastal northern California. Climate projections beyond this approximate time period diverge with increasing uncertainty (see,
e.g.,
Lenihan
et al.
2008, pp. 16-17), including uncertainties in the magnitude and timing, as well as regional details, of predicted climate change, especially at smaller scales (IPCC 2015, no page number), which is why we cannot reliably project future climate change effects beyond this timeframe.

A thorough review of each of the potential stressors is presented in the Species Report (Service 2015, pp. 41-78), which is available on the Internet

at
http://www.regulations.gov
, Docket No. FWS-R8-ES-2011-0105. A summary of this information is presented below.

Each potential stressor was evaluated to determine the likely impact to coastal martens or their habitat.

• A low-level impact indicates: (1) Individual martens in one or more populations may be impacted, but not at the population level; or (2) minimal loss, degradation, or fragmentation of suitable habitat.

• A medium-level impact indicates: (1) Individual martens in one or more populations are being impacted, likely resulting in a population-level impact; or (2) moderate loss, degradation, or fragmentation of suitable habitat.

• A high-level impact indicates: (1) Individual martens in one or more populations are being impacted, likely resulting in a significant population-level impact; or (2) significant loss, degradation, or fragmentation of suitable habitat.

Factor A—The Present or Threatened Destruction, Modification, or Curtailment of the Species' Habitat or Range

Wildfire

Wildfire can impact individual coastal martens directly through mortality (Factor E); however, fires generally kill or injure a relatively small proportion of animal populations, particularly if they are mobile (Lyon
et al.
2000, pp. 17-20), and the best available data do not indicate that wildfire is causing loss of individual martens. If direct mortality of individual martens occurs, we expect the impact to be discountable because martens are capable of rapid evacuation from an approaching fire, and adequate suitable habitat likely exists within their extant population areas to establish a new home range (provided the majority of the suitable habitat within the extant population area is not subjected to an overly large, high-severity wildfire).

Wildfire is a major disturbance force of habitat within the range of the coastal marten in all but the wettest coastal forests and thus has been analyzed in terms of its effect on coastal marten habitat. Wildfire can affect the composition and structural characteristics of the forest communities at multiple spatial and temporal scales. Fire severity is often expressed in categories of high, medium, or low severity, as well as mixed severity. High-severity fire, also called stand-replacing fire, kills all or nearly all vegetation within a stand and may extend across a landscape (Jain
et al.
2012, p. 47). Medium-severity fire refers to fire that is intermediate in its effects between high-severity and low-severity fire; for example, a fire may kill scattered clumps of overstory trees within a stand. Low-severity fire burns at ground-level and does not kill most overstory trees, although it may consume understory vegetation and downed woody debris (Jain
et al.
2012, p. 47). Finally, mixed-severity fire includes patches of low-severity fire and patches of high-severity fire (Jain
et al.
2012, p. 47).

Regional moisture gradients result in wildfires occurring more frequently with increasing distance from the coast and farther south in the coastal marten's range. The effect of fire on coastal marten habitat varies from high-severity fires that consume much or all of the structural features (
e.g.,
large trees, snags, logs) that are important elements of suitable coastal marten habitat, requiring centuries to regrow, to low-severity fires that burn only the dense, shade-tolerant shrub layer preferred by the coastal marten (Slauson
et al.
2009b, p. 11). The shrub layer likely takes 1 to 2 decades to regrow to suitable size and density, depending on its fire resistance and adaptive response to disturbances (Slauson 2014, pers. comm.). However, some low-severity fires may burn ground cover without burning the dense, shade-tolerant shrub layer preferred by the coastal marten. Wildfires within the range of the coastal marten often burn at mixed severities (Landscape Fire and Resource Management Planning Tools Project (LANDFIRE) 2008a; LANDFIRE 2008b; LANDFIRE undated(a)), with some areas within the fire perimeter burning at a high severity, resulting in stand replacement, and other portions burning at low severity, resulting in the loss of only ground vegetation. Fire effects are complex; therefore, potential impacts of future wildfires on coastal marten suitable habitat are difficult to predict.

Historical fire records indicate that, compared to the coastal central Oregon population area, the coastal northern California and coastal southern Oregon population areas (including adjacent or intervening areas) have experienced larger and more severe wildfires (Monitoring Trends in Burn Severity (MTBS; 2013, entire), both also experiencing many small (less than 0.4 hectares (ha) (1 acre (ac)) fires. The potential for severe, stand-replacing wildfire has increased in some areas where fire suppression and regeneration timber harvest (
i.e.,
the intent to develop a new stand/forest) have played a role in raising fuel load to levels that place late-successional forest at increased risk (Forest Service and BLM 1994b, pp. 3, 4-49). Although fire suppression is known to contribute to the severity of wildfire in some areas, within at least parts of coastal northern California and coastal southern Oregon, fire suppression has had little effect on altering the structure and composition of the dominant forest types and has not caused an increase in high-severity fire compared to the historical patterns (Odion
et al.
2004, pp. 933-935; Miller
et al.
2012, p. 200). In other words, the period of fire suppression may not be long enough to manifest such effects in coastal forest types where the return intervals for high-severity, stand-replacing fires are on the order of centuries (
e.g.,
Veirs 1982, pp. 132-133; Oneal
et al.
2006, pp. 82-87).

The best available historical fire information and the more xeric nature (
i.e.,
environment containing little moisture) of the interior within the Klamath Ecoregion indicate that future loss, degradation, or fragmentation of moderate- and high-suitability coastal marten habitat from wildfires will likely result in a greater impact in the coastal southern Oregon and coastal northern California populations as compared to the coastal central Oregon population. However, the more coastal climate where most martens occur may have an ameliorating effect (
e.g.,
increased humidity, reduced temperatures) on fires, reducing the size of fires in the coastal area compared to those more characteristic of the rest of the Klamath Ecoregion. Historical data between 1984 and 2012 indicate that wildfires burned approximately 17 percent and 42 percent of the combined moderate- and high-suitability coastal marten habitat within the coastal northern California and coastal southern Oregon population areas, respectively, with a few large fires responsible for the majority of burned suitable habitat (MTBS 2013, entire). We note that these wildfires burned at varying levels of severity; in other words, although some suitable habitat was lost as a result of the wildfires, varying levels of suitable habitat remain throughout the population areas, with moderate- and high-suitability habitat remaining within the wildfire perimeters after the fires were extinguished (Service 2014, unpubl. Geographic Information System (GIS) analysis).

It is possible that fire frequency, size, and severity may increase in the future within coastal Oregon (both central and southern) and coastal northern California, based on projected increases in temperature and decreased precipitation (see “Climate Change,” below), with potentially greater

increases within coastal southern Oregon and coastal northern California based on the history of wildfire within these portions of the coastal marten's range. In contrast, little moderate- and high-suitability coastal marten habitat has burned (historically, between 1984 and 2012) within and adjacent to the coastal central Oregon population area (MTBS 2013, entire). Large, stand-replacing fires occur infrequently (at intervals greater than 200 to 250 years) within coastal central Oregon (Impara 1997, p. 92; Long
et al.
1998, p. 786; Long and Whitlock 2002, p. 223l; LANDFIRE 2008a). In general, most fires that have recently occurred within the range of coastal marten have burned at mixed severity (
e.g.,
LANDFIRE 2008a; LANDFIRE 2008b; LANDFIRE undated(a)), resulting in some areas burning at a lower intensity with loss of only ground or shrub understory vegetation, and retaining of a portion of the moderate- and high-quality habitat within the fire perimeters.

In our initial development of the Species Report, we identified an overall low-level impact across the northern portion of the coastal marten's range, and a medium-level impact across the southern portion of the coastal marten's range (see section 9.2.3.1 in the Species Report). These overall impact levels were based on the probability of occurrence of a wildfire over a 15-year time period. When considering historical fire data over a 30-year time period to predict the future equivalent level of expected fire frequency, size, and severity (see Appendix A in the Species Report), the overall level of impact (
i.e.,
probability of occurrence of a wildfire) is potentially the same. However, this impact level estimate does not take into account the historical fire data (
e.g.,
LANDFIRE 2008a; LANDFIRE 2008b; LANDFIRE undated(a)) that show most wildfires burned at low severity and retained moderate- and high-quality habitat post-fire.

Based on the analysis contained within the Species Report and summarized above, we expect that within the range of the coastal marten, the incidence of wildfire in the future will be similar to that recorded for 1984 to 2012. We note, however, that high-severity fires have been infrequent in the past and are considered to remain infrequent, overall, into the future. Our expectation is that fire frequency, size, and severity in the future will be fairly similar (or slightly higher in some areas based on climate change predictions). Based on these 30 years (
i.e.,
1984-2012) of data, we can reasonably estimate these effects will continue with the same approximate level of impact into the next 30 years as has occurred over the previous 30 years (
i.e.,
mixed severity wildfires will likely occur although most will be low severity and retain some moderate- and high-quality habitat post-fire); thus, we predict that, overall, these impacts do not rise to the level of a threat. We base this conclusion on:

(1) The persistence of moderate- and high-quality habitat that has remained following recent large wildfires (
i.e.,
wildfires that have burned at mixed severities (LANDFIRE 2008a; LANDFIRE 2008b; LANDFIRE undated(a)), which have not resulted in extensive stand-replacement within the coastal marten's range.

(2) The overall continued presence of relatively moist habitat conditions for coastal marten habitat, primarily along the western coast, including overall cooler, moist summer conditions that moderate the dry conditions that promote fire ignition and spread.

(3) Information indicating that parts of coastal northern California and coastal southern Oregon have experienced fire suppression with little effect on altering the structure and composition of the dominant forest types, and no increase in high-severity fire compared to the historical patterns (Odion
et al.
2004, pp. 933-935; Miller
et al.
2012, p. 200).

Climate Change

“Climate” refers to the mean and variability of weather conditions over time, with 30 years being a typical period for such measurements, although shorter or longer periods also may be used (Intergovernmental Panel on Climate Change [IPCC] 2013, p. 1,450). The term “climate change” thus refers to a change in the mean or variability of one or more measures of climate (
e.g.,
temperature or precipitation) that persists for an extended period, typically decades or longer, whether the change is due to natural variability, human activity, or both (IPCC 2013, p. 1,450). A recent synthesis report of climate change and its effects is available from the IPCC (IPCC 2014, entire).

Changes in climate may have direct or indirect effects on species. These effects may be positive, neutral, or negative, and they may change over time, depending on the species and other relevant considerations, such as interactions of climate with other variables (
e.g.,
habitat fragmentation, fire frequency) (IPCC 2007, pp. 8-14, 18-19). Typically, expert judgment and appropriate analytical approaches are used to weigh relevant information, including uncertainty, in various aspects of climate change.

Global climate projections are informative, and in some cases, the only scientific information available. However, projected changes in climate and related impacts can vary substantially across and within different regions of the world (
e.g.,
IPCC 2007, pp. 8-12). Therefore, we use “downscaled” projections (see Glick
et al.
2011, pp. 58-61, for a discussion of downscaling) when they are available and have been developed through appropriate scientific procedures, because such projections provide higher resolution information that is more relevant to spatial scales used for analyses of a given taxon. For this analysis across the range of the coastal marten, downscaled projections are used in addition to some regional climate models that provide higher resolution projections using a modeling approach that differs from downscaling. The geographic region of the projections is the southern terminus of temperate rainforests of the North American continent, which encompasses the range of the coastal marten.

Climate throughout the range of the coastal marten is projected over the next approximately 40 to 50 years to become warmer, and in particular summers will be hotter and drier, with more frequent heat waves (Pierce
et al.
2013, p. 848; Cayan
et al.
2012, p. 10; Salathé
et al.
2010, p. 69; Tebaldi
et al.
2006, pp. 191-200; Hayhoe
et al.
2004, p. 12423). However, the northern portion of the coastal marten's range will likely experience winters that may become wetter, although warmer temperatures may result in an overall water deficit (Pierce
et al.
2013, p. 848; Cayan
et al.
2012, p. 10; Salathé
et al.
2010, p. 69; Tebaldi
et al.
2006, pp. 191-200; Hayhoe
et al.
2004, p. 12423). The coastal marten's currently suitable habitat may be affected by climate change to some extent. At this time, nearly all models for the coastal northern California and coastal southern Oregon population areas predict shifts in vegetation type over time from conifer forest to mixed-conifer hardwood forest, as well as shifts toward woodland and chaparral, with some shifts predicted to be observable by 2030, but most by the end of the century (roughly 2070 through 2099) (Whitlock
et al.
2003, p. 16; Rehfeldt
et al.
2006, p. 1143; Lenihan
et al.
2008, p. 20; Doppelt
et al.
2009, p. 7; Littell
et al.
2011, pp. 11-12; Shafer
et al.
2010, pp. 180-181; Littell
et al.
2013, pp. 113-115). The predicted extent and nature of these shifts and the potential rate of change vary greatly, depending on

potential emissions scenarios, assumptions (for example, in how various plant species are likely to respond to changes in temperature, precipitation, and carbon dioxide concentration), and variables incorporated into the models. Despite these differences, most models produce qualitatively similar forecasts of the impacts of potential future climates on ecosystem distribution, function, and disturbances (Shafer
et al.
2010, p. 179). Although climate models have become increasingly sophisticated, the simulated future response of ecosystems remains subject to great uncertainty due to a number of factors, especially over longer timeframes (see,
e.g.,
Lenihan
et al.
2008, pp. 16-17). In sum, although there is general agreement in the direction and nature of changes anticipated, models continue to have limitations which lead to uncertainties in the magnitude and timing, as well as regional details, of predicted climate change, especially at smaller scales (IPCC 2015, no page number) Thus, although we anticipate the coastal marten's currently suitable habitat may be affected by climate change to some extent, there is a high level of uncertainty regarding the nature of any such effects and the likelihood and timing of their occurrence.

In coastal central and northern Oregon, models also project shifts by the end of this century in vegetation type from maritime conifer forest toward mixed conifer-hardwood and deciduous forests, although models differ in the extent of this change (Whitlock
et al.
2003, p. 16; Rehfeldt
et al.
2006, p. 1143; Lenihan
et al.
2008, p. 20; Doppelt
et al.
2009, p. 7; Littell
et al.
2011, pp. 11-12; Shafer
et al.
2010, pp. 180-181; Littell
et al.
2013, pp. 113-115). These shifts in future vegetation type may lead to range shifts for the coastal marten, although information is not available to indicate how rapidly this may occur. It is important to note that studies of climate change present a range of effects including some that indicate conditions could remain suitable for coastal martens. For example, in areas with stable or increasing total precipitation, overall warmer temperatures are expected to result in a decreased snowpack ((Cayan
et al.
2012, pp. 20-21; Littell
et al.
2011, p. 60; Salathé
et al.
2010, pp. 66-68; Hayhoe
et al.
2004, p. 12423), which would result in increased availability of habitat for coastal martens at higher elevations, as well as increased availability of prey during the winter months (Service 2015, p. 7). Overall, it is not clear how finer-scale abiotic factors may shape local climates and influence local vegetation trends either to the benefit or detriment of coastal martens, nor is the timeframe clear over which these influences may be realized.

We note that redwood forest habitat within coastal national and State parks to the west of the coastal northern California population area may remain suitable for coastal martens even with projected changes in climate (based on a moderate emissions scenario within 50 years; DellaSala 2013, entire). However, to reach this coastal redwood habitat, martens would need to traverse many kilometers of unsuitable habitat (
i.e.,
industrial timberlands). Martens actively select against these areas that do not have protective overstory cover; however, limited movement across unsuitable habitat areas may occur. In contrast, coastal martens currently occurring within the drier, interior portions of the coastal southern Oregon population area could migrate into other suitable habitat to the west as climate change alters the more interior habitat; a natural, westward migration is possible due to a lack of significant physical barriers to east-west movements within that region.

Overall, studies of climate change present a range of effects on vegetation, including some that indicate conditions could remain suitable for coastal martens in portions of the coastal range; furthermore, the severity of potential impacts to coastal marten habitat will likely vary across the range, with effects to coastal martens potentially ranging from negative, neutral, or beneficial. Thus, the Species Report described an estimated range of low- to medium-impact for this stressor for coastal southern Oregon and coastal northern California (Service 205, pp. 67-72). Modeling projections are done at a large scale, and effects to species' habitat can be complex, unpredictable, and highly influenced by local-level biotic and abiotic factors. Although many climate models generally agree about the changes in temperature and precipitation, the consequent effects on vegetation are more uncertain, as is the rate at which any such changes might be realized. Therefore, it is not clear how or when changes in forest type and plant species composition will affect the distribution of coastal marten habitat. How any such changes may in turn affect coastal marten populations is even more uncertain. Thus, uncertainty exists when determining the level of impact climate change may have on coastal marten habitat. Consequently, at this time and based on the analysis contained within the Species Report and summarized above, we have determined that we do not have reliable information to indicate that climate change is a threat to coastal marten habitat now or in the future, although we will continue to seek additional information concerning how climate change may affect coastal marten habitat.

Vegetation Management

Vegetation management includes activities such as timber harvest, thinning, fuels reduction, and habitat restoration, which can result in the temporary or permanent loss, degradation, or fragmentation of suitable coastal marten habitat. Once lost, structural elements found in suitable coastal marten habitat that are required for denning and resting (such as large diameter live trees, snags, and logs) require more than a century to develop (Slauson and Zielinski 2009, p. 43). Slauson (2014, pers. comm.) anticipates that loss of the dense, shade-tolerant shrub layer required by the coastal marten would take 1 to 2 decades to regrow.

Historically, vegetation management activities (particularly large-scale harvest of late-successional coniferous forest habitat) reduced the amount and distribution of suitable coastal marten habitat. At the present time, although the reduction and fragmentation of some suitable coastal marten habitat is expected to continue, the majority of suitable habitat for coastal martens is currently secure and expected to increase in the future. Habitat loss and degradation is expected to be realized primarily on private lands, which constitute a relatively small proportion of the suitable habitat available to martens in the three extant population areas (23 percent in coastal central Oregon, 10 percent in coastal southern Oregon, and 11 percent in coastal northern California). In contrast, most suitable marten habitat is in Federal ownership (71 percent in the coastal central Oregon population area, 90 percent in the coastal southern Oregon population area, and 77 percent in the coastal northern California population area), and the majority of those lands are in reserve allocations under the NWFP, which are managed for the maintenance or development of late-successional forest characteristics (71 percent of Federal lands in reserves in coastal central Oregon, 79 percent of Federal lands in reserves in coastal southern Oregon, and 90 percent of Federal lands in reserves in coastal northern California). We therefore expect not only the maintenance but further recruitment of suitable coastal marten

habitat on Federal reserve lands over time.

Some vegetation management activities (such as thinning, fuels reduction projects, and habitat restoration) have the potential to improve habitat suitability for the coastal marten in the long term by minimizing loss of late-successional stands due to wildfires and accelerating the development of late-seral characteristics (Zielinski 2013, pp. 419-422). This has been suggested for a similar mustelid, the fisher, where such activities may be consistent with maintaining landscapes that support fishers in the long term and sometimes even the short term, providing treatments retain appropriate habitat structures, composition, and configuration (Spencer
et al.
2008, entire; Scheller
et al.
2011, entire; Thompson
et al.
2011, entire; Truex and Zielinski 2013, entire; Zielinski 2013, pp. 17-20). Thus, it is reasonable to assume that these types of projects could increase the long-term, overall amount, distribution, and patch size of suitable coastal marten habitat, although some short-term degradation, loss, or fragmentation of suitable coastal marten habitat may occur in the interim.

On lands managed for industrial timber harvest, the past and current practice of managing coastal coniferous forests on a short-rotation system (40-60 years) to maximize wood production has reduced the complexity of the shrub and herb layers, which are important components of suitable marten habitat. These management practices have also precluded development of late-successional forest characteristics that are important to the coastal marten (such as large diameter logs, snags, and trees). Short-rotation forestry is prevalent on private lands, whereas only a small fraction of forested Federal lands (
i.e.,
“matrix” lands as defined under the NWFP) may be used for timber harvest.

Due to current and expected future intensive timber-harvesting activities, we do not anticipate that private lands would support viable marten populations or maintain important habitat elements in the future. Instead, the coastal marten relies on (and our analysis considers) the maintenance of suitable coastal marten habitat on Federal and State lands as the key element to support the long-term viability of coastal marten populations. Of the coastal marten suitable habitat within the three extant population areas, from 71 to 90 percent is on Federal lands and in reserve status under the NWFP, much of which is managed specifically for the development of late-successional characteristics that will be beneficial for coastal martens. Specifically, and at present:

(1) In the coastal central Oregon extant population area, 79 percent of the habitat is considered suitable for coastal martens (56 percent moderate to high suitability). Approximately 71 percent of the moderate- to high-suitability habitat occurs within Federal ownership, and 71 percent of that is Federal Reserve land.

(2) In the coastal southern Oregon extant population area, 95 percent of the habitat is considered suitable for coastal martens (78 percent moderate to high suitability). Approximately 90 percent of the moderate- to high-suitability habitat is in Federal ownership, and 79 percent of that is Federal Reserve land.

(3) In the coastal northern California extant population area, 87 percent of the habitat is considered suitable habitat for coastal martens (81 percent moderate to high suitability). Approximately 77 percent of that is in Federal ownership, and 90 percent of that is Federal Reserve land.

A small proportion of the moderate- and high-suitability habitat occurs on Federal matrix lands (
i.e.,
lands as defined under the NWFP that are used for timber harvest). The rate of loss of late-successional and old-growth forest on Federal lands due to timber harvest has declined substantially since the implementation of the NWFP (Mouer
et al.
2011, entire). Although the NWFP does not recognize marten habitat as a forest class or condition, late-successional old growth forest likely includes a subset of coastal marten habitat (if the necessary dense shrub layer is present).

Based on the analysis contained within the Species Report and summarized above, including the proportion of moderate- and high-suitability coastal marten habitat available and the favorably managed forested lands (primarily Federal Reserves) within each extant population area, we consider ongoing vegetation management to have a low impact on the loss, degradation, or fragmentation of suitable coastal marten habitat across the range of the DPS both currently and into the future. We note that loss of suitable habitat (primarily low-quality suitable habitat) is expected to continue to occur into the future on private lands within all three population areas, potentially to a greater extent in the coastal central Oregon population area due to a larger percentage of privately-owned timber lands within that population area. For the entire range, we considered vegetation management as a low-level impact on moderate and high suitability marten habitat for Federal lands, which constitute a majority of the extant population areas, have longer harvest rotations, and retain more structural features on the subset of that area in matrix, or where habitat will be retained on lands in Federal Reserves. In addition, because of the extent of Federal reserve land allocations that are designed to maintain and develop late-successional conditions, an unquantifiable amount of suitable habitat for coastal martens is expected to develop in the future. Overall, potential impacts from vegetation management do not rise to the level of a threat given the extensive beneficial land management practices expected to continue into the future (15 years) on public lands.

Development

Some impacts to suitable habitat are expected to occur within the range of the coastal marten as a result of development activities such as road building, dam construction and creation of new reservoirs, conversion of forest habitat for agricultural use, development and expansion of recreational areas (
e.g.,
golf courses, campgrounds, and trails), urban expansion, and rural development. Should these types of disturbances occur, they would likely result in the further loss, degradation, or fragmentation of suitable habitat. However, if these activities occur into the future, only a small amount of habitat may be impacted rangewide based on our evaluation of the best available data at this time because most of the potential development is expected on private lands that afford the coastal marten little suitable habitat to begin with. In addition, many of the areas that provide suitable habitat for coastal martens are areas of challenging topography that are not conducive to intensive or large-scale development.

In Oregon, the greatest rates of change from resource land use to more developed use occurred prior to 1984, before implementation of county land-use plans and land-use planning laws (Oregon Administrative Rule 660-015-00) that limit the conversion of designated resource lands, including forest lands, to other uses (Lettman
et al.
2011, p. 16). These laws encourage intensified development in areas already urbanizing, while limiting development in more rural areas (Lettman
et al.
2009, p. 4; Lettman
et al.
2011, p. 9). Consequently, conversion of non-Federal forest land has been limited in Oregon, with 98 percent of all non-Federal forest, agricultural, and range

lands in the State in 1974 remaining in those uses in 2009 (Lettman
et al.
2011, p. 11). Virtually all land-use change during this time occurred on private land (Lettman
et al.
2011, p. 11). However, development of private land within 1.6 km (1 mi) of Federal forest land is increasing, which can affect management along the periphery of adjacent Federal lands, such as increasing the need for fuel treatments on public lands to protect structures on adjacent private lands (Lettman
et al.
2009, pp. 33-34; Azuma
et al.
2013, pp. 1-2). Development of Federal forest lands in California and Oregon, however, is expected to be limited given past history (
e.g.
Lettman
et al.
2011, p. 11 for Oregon) and the management mandates of the land management agencies.

Based on the analysis contained within the Species Report and summarized above, and similar to the vegetation management discussion above, we estimate that development has a low impact on the loss, degradation, or fragmentation of suitable coastal marten habitat across the range of the DPS both currently and into the future, and thus does not rise to the level of a threat. If development occurs, the frequency and amount of habitat impacted may be greater in the coastal central Oregon population area as opposed to the other two population areas due to a larger percentage of privately-owned timber lands within the coastal central Oregon population area. However, as exhibited over the past 30 years, any loss is expected to be small.

Factor B—Overutilization for Commercial, Recreational, Scientific, or Educational Purposes

Trapping

Trapping for Fur

Historical unregulated fur trapping (prior to the 1930s) of coastal martens is considered by researchers as the likely cause of the marked contraction in coastal marten distribution. Legal marten fur trapping in coastal northern California ended in 1946. However, fur trapping remains legal and has continued in Oregon, and the number of martens harvested in coastal Oregon counties has declined since the 1940s (Zielinski
et al.
2001, p. 482), although it is not known whether trapping effort remained unchanged over this time period. By the 1970s, martens were considered rare along the Oregon coast (Zielinski
et al.
2001, p. 483; Mace 1970, pp. 13-14; Maser
et al.
1981, pp. 293-294). A total of 36 martens were harvested within coastal Oregon counties between 1969 and 1995 (Verts and Carraway 1998, p. 409). This harvest level excludes Lane and Douglas Counties because a substantial area of these counties is outside the DPS and fur trapping is only reported at the county level. The most recent data indicate that three coastal martens were trapped within coastal Oregon during the 2013 fur trapping season (Oregon Department of Fish and Wildlife, unpublished data). Overall, based on these data, the number of martens trapped in coastal Oregon has averaged fewer than two animals a year in recent decades. The fur trapping effort for martens in Oregon is relatively minimal; the Oregon Department of Fish and Wildlife reports that few trappers, generally from 4 to 8, trap for marten anywhere in the State in any given year. Most recent harvests of martens are from the Cascades and Blue Mountain Ranges; harvest of martens in the Coast Range is extremely rare (Hiller 2011, p. 17). Any potential population impacts of removing individual coastal martens as a result of fur trapping are difficult to estimate due to a lack of population size estimates in both Oregon population areas. The best available data indicate, however, that relatively few martens are removed from coastal populations as a result of fur trapping in Oregon, and we have no evidence to suggest that these populations may be in decline as a consequence of fur trapping.

Based on the analysis contained within the Species Report and summarized above, we consider the legal fur trapping of coastal martens as having no overall impact to the population in coastal northern California, as there is no legal fur trapping for martens in that State. Fur trapping effort for martens in Oregon is relatively minimal, and most martens harvested are not trapped in the coast ranges. We estimate a low- to medium-level of impact to the two extant populations in coastal Oregon, reflecting the uncertainty regarding the size of those populations. We estimate that the impacts of fur trapping on coastal martens in Oregon will continue at a similar level, both currently and into the future, because the best available data do not suggest that either fur trapping effort or impacts are likely to change. Additionally, of note for California, we expect that nearly all coastal martens that are accidentally captured in box traps (body-gripping traps are illegal in California) set for other furbearer species, or that are live-trapped for research purposes, will be released unharmed. As a result of this best available information for Oregon and California, we have determined that fur trapping, overall, does not have a significant population-level impact across the DPS's range and does not rise to the level of a threat.

Trapping for Research Purposes

Based on the analysis contained within the Species Report, we consider the potential impacts of live-trapping and handling for research purposes on coastal marten populations as discountable. We came to this conclusion based on the limited distribution of marten research projects in the three extant population areas (currently only a single project in the western half of the coastal northern California population area where no martens were injured or killed during live-trapping), and based on the strict trapping and handling protocols that must be adhered to by coastal marten researchers to ensure the safety of study animals. Available information does not suggest that there would be any change to the level of anticipated impacts of live-trapping and handling for research purposes into the future, and, therefore, we find that the potential impacts to the coastal marten from trapping for research purposes do not rise to the level of a threat.

Factor C—Disease or Predation

Disease

Numerous pathogens (
e.g.,
canine distemper, canine parvovirus, toxoplasmosis) are known to cause severe disease in mustelids. Infected domestic dogs that are allowed to roam within an extant marten population area could expose martens to lethal pathogens. Fur trappers could capture an infected carnivore (
e.g.,
marten, fisher, gray fox, bobcat) and inadvertently spread the disease to martens through contaminated traps. Marten researchers could also transfer lethal pathogens within and between extant population areas if traps and track-plate boxes are not disinfected after exposure to any carnivore species, including coastal martens.

An outbreak of a lethal pathogen within any of the three extant coastal marten populations could occur. Several serious pathogens have been detected in the related fisher less than 9 km (5.6 mi) from the nearest verifiable marten detection within the coastal northern California population (Brown
et al.
2008, entire), suggesting that martens could be exposed by infected juvenile fishers that disperse from their natal area into the coastal marten population area. However, despite possible exposure to pathogens, no outbreaks of

diseases have been detected in coastal martens, and we have no evidence to suggest that disease is currently present in any of the coastal marten populations.

The best available data do not indicate that disease has impacted coastal martens at any point in time in the past or currently. The prevalence of past exposure to lethal pathogens within the coastal northern California population and the coastal Oregon populations has not been demonstrated through a serosurvey (
i.e.,
a screening test of the serum of a marten to determine susceptibility to a particular disease). Additionally, if the known extant populations are disjunct from one another, as suggested by Slauson and Zielinski (2009, pp. 35-36), this would be beneficial in terms of reducing the ease of transmission of disease between the populations, should an outbreak occur. Thus, at this time, the best available data do not indicate that a disease outbreak has had, or is likely to have, a significant population-level effect on coastal martens.

In sum, there are currently no indications of disease in coastal marten populations. If an outbreak of a serious disease should occur, it could have a significant impact on the affected population. However, based upon the best available scientific and commercial data as presented in the Species Report and summarized here, there is a low probability that a disease outbreak may occur. We anticipate that if there should be an outbreak, it will likely have a low effect on all three coastal marten populations combined, as the distance between them makes it unlikely that the effects of such an outbreak would spread. Thus, we have determined that disease has a low-level population impact across the coastal marten's range and, therefore, does not rise to the level of a threat currently or into the future.

Predation

Predation is a natural ongoing source of mortality for the coastal marten and would not be expected to negatively impact the viability of marten populations in coastal Oregon and coastal northern California unless annual predation rates, combined with all other mortality sources, exceed annual juvenile coastal marten recruitment rates (estimated at 50 percent for the coastal marten; Slauson
et al.,
In prep.(a)). At this time, the only documented coastal marten predators are bobcats (Slauson
et al.
2014, unpubl. data). However, additional predator species have been documented for other marten species and populations:

(1) Strickland
et al.
(1982, p. 607) summarized reports of American martens being preyed upon by coyotes, fishers, red foxes, cougars, golden and bald eagles (
Aquila chrysaetos, Haliaeetus leucocephalus
), and great horned owls (
Bubo virginianus
).

(2) Bull and Heater (2001a, p. 3) conducted a study in northeastern Oregon and documented 18 martens (
i.e., Martes caurina vulpina
) killed by predators: 44 percent by bobcats, 22 percent by raptors, 22 percent by other martens, and 11 percent by coyotes.

Historical coastal marten predation rates are unknown, although the historical assemblage of predator species was likely similar to the current assemblage. It is possible that human-caused changes in vegetation composition, vegetation distribution, and extensive road building over time have increased predator densities and distribution within the range of the coastal marten. These changes in vegetation and infrastructure provide more access and avenues in which predators can exploit their prey base, especially in forested areas that w

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A2015-07766. Public record. Not legal advice.