Endangered and Threatened Wildlife and Plants; 12-Month Findings for Petitions to List the Greater Sage-Grouse (Centrocercus urophasianus) as Threatened or Endangered

DEPARTMENT OF THE INTERIOR

Fish and Wildlife Service

50 CFR Part 17

[FWS-R6-ES-2010-0018]

[MO 92210-0-0008-B2]

Endangered and Threatened Wildlife and Plants; 12-Month Findings for Petitions to List the Greater Sage-Grouse (

Centrocercus urophasianus

) as Threatened or Endangered

AGENCY:

Fish and Wildlife Service, Interior.

ACTION:

Notice of 12-month petition findings.

SUMMARY:

We, the U.S. Fish and Wildlife Service (Service), announce three 12-month findings on petitions to list three entities of the greater sage-grouse (

Centrocercus urophasianus

) as threatened or endangered under the Endangered Species Act of 1973, as amended (Act). We find that listing the greater sage-grouse (rangewide) is warranted, but precluded by higher priority listing actions. We will develop a proposed rule to list the greater sage-grouse as our priorities allow.

We find that listing the western subspecies of the greater sage-grouse is not warranted, based on determining that the western subspecies is not a valid taxon and thus is not a listable entity under the Act. We note, however, that greater sage-grouse in the area covered by the putative western subspecies (except those in the Bi-State area (Mono Basin), which are covered by a separate finding) are encompassed by our finding that listing the species is warranted but precluded rangewide.

We find that listing the Bi-State population (previously referred to as the Mono Basin area population), which meets our criteria as a distinct population segment (DPS) of the greater sage-grouse, is warranted but precluded by higher priority listing actions. We will develop a proposed rule to list the Bi-State DPS of the greater sage-grouse as our priorities allow, possibly in conjunction with a proposed rule to list the greater sage-grouse rangewide.

DATES:

The finding announced in the document was made on March 23, 2010.

ADDRESSES:

This finding is available on the Internet at

http://www.regulations.gov

and

www.fws.gov

. Supporting documentation we used to prepare this finding is available for public inspection, by appointment, during normal business hours at the U.S. Fish and Wildlife Service, 5353 Yellowstone Road, Suite 308A, Cheyenne, Wyoming 82009; telephone (307) 772-2374; facsimile (307) 772-2358. Please submit any new information, materials, comments, or questions concerning this species to the Service at the above address.

FOR FURTHER INFORMATION CONTACT:

Brian T. Kelly, Field Supervisor, U.S. Fish and Wildlife Service, Wyoming Ecological Services Office (see

ADDRESSES

). If you use a telecommunications device for the deaf (TDD), call the Federal Information Relay Service (FIRS) at (800) 877-8339.

SUPPLEMENTARY INFORMATION:

Background

Section 4(b)(3)(B) of the Act (16 U.S.C. 1531

et seq.

), requires that, for any petition containing substantial scientific or commercial information that the listing may be warranted, we make a finding within 12 months of the date of the receipt of the petition on whether the petitioned action is (a) not warranted, (b) warranted, or (c) warranted, but that immediate proposal of a regulation implementing the petitioned action is precluded by other pending proposals to determine whether species are threatened or endangered, and expeditious progress is being made to add or remove qualified species from the Lists of Endangered and Threatened Wildlife and Plants. 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 Action

Greater Sage-Grouse

On July 2, 2002, we received a petition from Craig C. Dremann requesting that we list the greater sage-grouse (

Centrocercus urophasianus

) as endangered across its entire range. We received a second petition from the Institute for Wildlife Protection on March 24, 2003, requesting that the greater sage-grouse be listed rangewide. On December 29, 2003, we received a third petition from the American Lands Alliance and 20 additional conservation organizations (American Lands Alliance

et al.

) to list the greater sage-grouse as threatened or endangered rangewide. On April 21, 2004, we announced our 90-day petition finding in the

Federal Register

(69 FR 21484) that these petitions taken collectively, as well as information in our files, presented substantial information indicating that the petitioned actions may be warranted. On July 9, 2004, we published a notice to reopen the period for submitting comments on our 90-day finding, until July 30, 2004 (69 FR 41445). In accordance with section 4(b)(3)(A) of the Act, we completed a status review of the best available scientific and commercial information on the species. On January 12, 2005, we announced our not-warranted 12-month finding in the

Federal Register

(70 FR 2243).

On July 14, 2006, Western Watersheds Project filed a complaint in Federal district court alleging that the Service's 2005 12-month finding was incorrect and arbitrary and requested the finding be remanded to the Service. On December 4, 2007, the U.S. District Court of Idaho ruled that our 2005 finding was arbitrary and capricious, and remanded it to the Service for further consideration. On January 30, 2008, the court approved a stipulated agreement between the Department of Justice and the plaintiffs to issue a new finding in May 2009, contingent on the availability of a new monograph of information on the sage-grouse and its habitat (Monograph). On February 26, 2008, we published a notice to initiate a status review for the greater sage-grouse (73 FR 10218), and on April 29, 2008, we published a notice extending the request for submitting information to June 27, 2008 (73 FR 23172). Publication of the Monograph was delayed due to circumstances outside the control of the Service. An amended joint stipulation, adopted by the court on June 15, 2009, required the Service to submit the 12-month finding to the

Federal Register

by February 26, 2010; this due date was subsequently extended to March 5, 2010.

Western Subspecies of the Greater Sage-Grouse

The western subspecies of the greater sage-grouse (

Centrocercus urophasianus phaios

) was identified by the Service as a category 2 candidate species on September 18, 1985 (50 FR 37958). At the time, we defined Category 2 species as those species for which we possessed information indicating that a proposal to list as endangered or threatened was possibly appropriate, but for which conclusive data on biological vulnerability and threats were not available to support a proposed rule. On February 28, 1996, we discontinued the designation of category 2 species as candidates for listing under the Act (61 FR 7596), and consequently the western subspecies was no longer considered to be a candidate for listing.

We received a petition, dated January 24, 2002, from the Institute for Wildlife

Protection requesting that the western subspecies occurring from northern California through Oregon and Washington, as well as any western sage-grouse still occurring in parts of Idaho, be listed under the Act. The petitioner excluded the Mono Basin area populations in California and northwest Nevada since they already had petitioned this population as a distinct population segment (DPS) for emergency listing (see discussion of Bi-State area (Mono Basin) population below). The petitioner also requested that the Service include the Columbia Basin DPS in this petition, even though we had already identified this DPS as a candidate for listing under the Act (66 FR 22984, May 7, 2001) (see discussion of Columbia Basin below).

We published a 90-day finding on February 7, 2003 (68 FR 6500), that the petition did not present substantial information indicating the petitioned action was warranted based on our determination that there was insufficient evidence to indicate that the petitioned western population of sage-grouse is a valid subspecies or DPS. The petitioner pursued legal action, first with a 60-day Notice of Intent to sue, followed by filing a complaint in Federal district court on June 6, 2003, challenging the merits of our 90-day finding. On August 10, 2004, the U.S. District Court for the Western District of Washington ruled in favor of the Service (Case No. C03-1251P). The petitioner appealed and on March 3, 2006, the U.S. Court of Appeals for the Ninth Circuit reversed in part the ruling of the District Court and remanded the matter for a new 90-day finding (

Institute for Wildlife Protection

v.

Norton

, 2006 U.S. App. LEXIS 5428 9th Cir., March 3, 2006). Specifically, the Court of Appeals rejected the Service's conclusion that the petition did not present substantial information indicating that western sage-grouse may be a valid subspecies, but upheld the Service's determination that the petition did not present substantial information indicating that the petitioned population may constitute a DPS. The Court's primary concern was that the Service did not provide a sufficient description of the principles we employed to determine the validity of the subspecies classification. On April 29, 2008, we published in the

Federal Register

(73 FR 23170) a 90-day finding that the petition presented substantial scientific or commercial information indicating that listing western sage-grouse may be warranted and initiated a status review for western sage-grouse.

In a related action, the Service also has made a finding on a petition to list the eastern subspecies of the greater sage-grouse (

Centrocercus urophasianus urophasianus

). On July 3, 2002, we received a petition from the Institute for Wildlife Protection to list the eastern subspecies, identified in the petition as including all sage-grouse east of Oregon, Washington, northern California, and a small portion of Idaho. The petitioners sued the Service in U.S. District Court on January 10, 2003, for failure to complete a 90-day finding. On October 3, 2003, the Court ordered the Service to complete a finding. The Service published its not-substantial 90-day finding in the

Federal Register

on January 7, 2004 (69 FR 933), based on our determination that the eastern sage-grouse was not a valid subspecies. The not-substantial finding was challenged, and on September 28, 2004, the U.S. District Court ruled in favor of the Service, dismissing the plaintiff's case.

Columbia Basin (Washington) Population of the Western Subspecies

On May 28, 1999, we received a petition dated May 14, 1999, from the Northwest Ecosystem Alliance and the Biodiversity Legal Foundation. The petitioners requested that the Washington population of western sage-grouse (

C. u. phaios

) be listed as threatened or endangered under the Act. The petitioners requested listing of the Washington population of western sage-grouse based upon threats to the population and its isolation from the remainder of the taxon. Accompanying the petition was information relating to the taxonomy, ecology, threats, and the past and present distribution of western sage-grouse.

In our documents we have used “Columbia Basin population” rather than “Washington population” because we believe it more appropriately describes the petitioned entity. We published a substantial 90-day finding on August 24, 2000 (65 FR 51578). On May 7, 2001, we published our 12-month finding (66 FR 22984), which included our determination that the Columbia Basin population of the western sage-grouse met the requirements of our policy on DPSs (61 FR 4722) and that listing the DPS was warranted but precluded by other higher priority listing actions. As required by section 4(b)(3)(C) of the Act, we have subsequently made resubmitted petition findings, announced in conjunction with our Candidate Notices of Review, in which we continued to find that listing the Columbia Basin DPS of the western subspecies was warranted but precluded by other higher priority listing actions (66 FR 54811, 67 FR 40663, 69 FR 24887, 70 FR 24893, 74 FR 57803). Subsequent to the March 2006 decision by the court on our 90-day finding on the petition to list the western subspecies of the greater sage-grouse (described above), our resubmitted petition findings stated we were not updating our analysis for the DPS, but would publish an updated finding regarding the petition to list the Columbia Basin population of the western subspecies following completion of the new rangewide status review for the greater sage-grouse.

Bi-State Area (Mono Basin) Population of Sage-grouse

On January 2, 2002, we received a petition from the Institute for Wildlife Protection requesting that the sage-grouse occurring in the Mono Basin area of Mono County, California, and Lyon County, Nevada, be emergency listed as an endangered distinct population segment (DPS) of

Centrocercus urophasianus phaios

, which the petitioners considered to be the western subspecies of the greater sage-grouse. This request was for portions of Alpine and Inyo Counties and most of Mono County in California and portions of Carson City, Douglas, Esmeralda, Lyon, and Mineral Counties in Nevada. On December 26, 2002, we published a 90-day finding that the petition did not present substantial scientific or commercial information indicating that the petitioned action may be warranted (67 FR 78811). Our 2002 finding was based on our determination that the petition did not present substantial information indicating that the population of greater sage-grouse in this area was a DPS under our DPS policy (61 FR 4722; February 7, 1996), and thus was not a listable entity (67 FR 78811; December 26, 2002). Our 2002 finding also included a determination that the petition did not present substantial information regarding threats to indicate that listing the petitioned population may be warranted (67 FR 78811).

On November 15, 2005, we received a petition submitted by the Stanford Law School Environmental Law Clinic on behalf of the Sagebrush Sea Campaign, Western Watersheds Project, Center for Biological Diversity, and Christians Caring for Creation to list the Mono Basin area population of greater sage-grouse as a threatened or endangered DPS of the greater sage-grouse (

C. urophasianus

) under the Act. On March 28, 2006, we responded that emergency listing was not warranted and, due to court orders and settlement agreements for other listing actions, we would not be able to address the petition at that time.

On November 18, 2005, the Institute for Wildlife Protection and Dr. Steven G. Herman sued the Service in U.S. District Court for the Western District of Washington (

Institute for Wildlife Protection et al.

v.

Norton et al.

, No. C05-1939 RSM), challenging the Service's 2002 finding that their petition did not present substantial information indicating that the petitioned action may be warranted. On April 11, 2006, we reached a stipulated settlement agreement with both plaintiffs under which we agreed to evaluate the November 2005 petition and concurrently reevaluate the December 2001 petition (received in January 2002). The settlement agreement required the Service to submit to the

Federal Register

a 90-day finding by December 8, 2006, and if substantial, to complete the 12-month finding by December 10, 2007. On December 19, 2006, we published a 90-day finding that these petitions did not present substantial scientific or commercial information indicating that the petitioned actions may be warranted (71 FR 76058).

On August 23, 2007, the November 2005 petitioners filed a complaint challenging the Service's 2006 finding. After review of the complaint, the Service determined that we would revisit our 2006 finding. The Service entered into a settlement agreement with the petitioners on February 25, 2008, in which the Service agreed to a voluntary remand of the 2006 petition finding, and to submit for publication in the

Federal Register

a new 90-day finding by April 25, 2008. The agreement further stipulated that if the new 90-day finding was positive, the Service would undertake a status review of the Mono Basin area population of the greater sage-grouse and submit for publication in the

Federal Register

a 12-month finding by April 24, 2009.

On April 29, 2008, we published in the

Federal Register

(73 FR 23173) a 90-day petition finding that the petitions presented substantial scientific or commercial information indicating that listing the Mono Basin area population may be warranted and initiated a status review. Based on a joint stipulation by the Service and the plaintiffs to extend the due date for the 12-month finding, on April 23, 2009, the U.S. District Court, Northern District of California, issued an order that if the parties did not agree to a later alternative date, the Service would submit a 12-month finding for the Mono Basin population of the greater sage-grouse to the

Federal Register

no later than May 26, 2009. On May 27, 2009, the U.S. District Court, Northern District of California, issued an order accepting a joint stipulation between the Department of Justice and the plaintiffs, which states that the parties agree that the Service may submit to the

Federal Register

a single document containing the 12-month findings for the Mono Basin area population and the greater sage-grouse no later than by February 26, 2010. Subsequently, the due date for submission of the document to the

Federal Register

was extended to March 5, 2010.

Both the November 2005 and the December 2001 petitions as well as our 2002 and 2006 findings use the term “Mono Basin area” to refer to greater sage-grouse that occur within the geographic area of eastern California and western Nevada that includes Mono Lake. For conservation planning purposes, this same geographic area is referred to as the Bi-State area by the States of California and Nevada (Greater Sage-grouse Conservation Plan for Nevada and Eastern California, 2004, pp. 4-5). For consistency with ongoing planning efforts, we will adopt the “Bi-State” nomenclature hereafter in this finding.

Biology and Ecology of Greater Sage-Grouse

Greater Sage-Grouse Description

The greater sage-grouse (

Centrocercus urophasianus

) is the largest North American grouse species. Adult male greater sage-grouse range in length from 66 to 76 centimeters (cm) (26 to 30 inches (in.)) and weigh between 2 and 3 kilograms (kg) (4 and 7 pounds (lb)). Adult females are smaller, ranging in length from 48 to 58 cm (19 to 23 in.) and weighing between 1 and 2 kg (2 and 4 lb). Males and females have dark grayish-brown body plumage with many small gray and white speckles, fleshy yellow combs over the eyes, long pointed tails, and dark green toes. Males also have blackish chin and throat feathers, conspicuous phylloplumes (specialized erectile feathers) at the back of the head and neck, and white feathers forming a ruff around the neck and upper belly. During breeding displays, males exhibit olive-green apteria (fleshy bare patches of skin) on their breasts (Schroeder

et al.

1999, p. 2).

Taxonomy

Greater sage-grouse are members of the Phasianidae family. They are one of two congeneric species; the other species in the genus is the Gunnison sage-grouse (

Centrocercus minimus

). In 1957, the American Ornithologists' Union (AOU) (AOU 1957, p 139) recognized two subspecies of the greater sage-grouse, the eastern (

Centrocercus urophasianus urophasianus)

and western (

C. u. phaios)

based on information from Aldrich (1946, p. 129). The original subspecies designation of the western sage-grouse was based solely on differences in coloration (specifically, reduced white markings and darker feathering on western birds) among 11 museum specimens collected from 8 locations in Washington, Oregon, and California. The last edition of the AOU Check-list of North American Birds to include subspecies was the 5

th

Edition, published in 1957. Subsequent editions of the Check-list have excluded treatment of subspecies. Richard Banks, who was the AOU Chair of the Committee on Classification and Nomenclature in 2000, indicated that, because the AOU has not published a revised edition at the subspecies level since 1957, the subspecies in that edition, including the western sage-grouse, are still recognized (Banks 2000, pers. comm.). However, in the latest edition of the Check-list (7

th

Ed., 1998, p. xii), the AOU explained that its decision to omit subspecies, “carries with it our realization that an uncertain number of currently recognized subspecies, especially those formally named early in this century, probably cannot be validated by rigorous modern techniques.”

Since the publication of the 1957 Check-list, the validity of the subspecies designations for greater sage-grouse has been questioned, and in some cases dismissed, by several credible taxonomic authorities (Johnsgard 1983, p. 109; Drut 1994, p. 2; Schroeder

et al.

1999, p. 3; International Union for Conservation of Nature (IUCN) 2000, p. 62; Banks 2000, 2002 pers. comm.; Johnsgard 2002, p. 108; Benedict

et al.

2003, p. 301). The Western Association of Fish and Wildlife Agencies (WAFWA), an organization of 23 State and provincial agencies charged with the protection and management of fish and wildlife resources in the western part of the United States and Canada, also questioned the validity of the western sage-grouse as a subspecies in its Conservation Assessment of Greater Sage-grouse and Sagebrush Habitats (Connelly

et al.

2004, pp. 8-4 to 8-5). Furthermore, in its State conservation assessment and strategy for greater sage-grouse, the Oregon Department of Fish and Wildlife (ODFW) stated that “recent genetic analysis (Benedict

et al.

2003) found little evidence to support this subspecies distinction, and this Plan refers to sage-grouse without reference to subspecies delineation in this document” (Hagen 2005, p. 5).

The Integrated Taxonomic Information System (ITIS), a database representing a partnership of U.S., Canadian, and Mexican agencies, other organizations, and taxonomic specialists designed to provide scientifically credible taxonomic information, lists the taxonomic status of western sage-grouse as “invalid - junior synonym” (ITIS 2010). In an evaluation of the historical classification of the western sage-grouse as a subspecies, Banks stated that it was “weakly characterized” but felt that it would be wise to continue to regard western sage-grouse as taxonomically valid “for management purposes” (Banks, pers. comm. 2000). This statement was made prior to the availability of behavioral and genetic information that has become available since 2000. In addition, Banks' opinion is qualified by the phrase “for Management purposes.” Management recommendations and other considerations must be clearly distinguished from scientific or commercial data that indicate whether an entity may be taxonomically valid for the purpose of listing under the Act.

Although the Service had referred to the western sage-grouse in past decisions (for example, in the 12-month finding for a petition to list the Columbia Basin population of western sage-grouse, 66 FR 22984; May 7, 2001), this taxonomic reference was ancillary to the decision at hand and was not the focal point of the listing action. In other words, when past listing actions were focused on some other entity, such as a potential distinct population segment in the State of Washington, we accepted the published taxonomy for western sage-grouse because that taxonomy itself was not the subject of the review and thus not subject to more rigorous evaluation at the time.

Taxonomy is a component of the biological sciences. Therefore, in our evaluation of the reliability of the information, we considered scientists with appropriate taxonomic credentials (which may include a combination of education, training, research, publications, classification and/or other experience relevant to taxonomy) as qualified to provide informed opinions regarding taxonomy, make taxonomic distinctions, and/or question taxonomic classification.

There is no universally accepted definition of what constitutes a subspecies, and the use of subspecies may vary between taxonomic groups (Haig

et al.

2006, pp. 1584-1594). The Service acknowledges the diverse opinions of the scientific community about species and subspecies concepts. However, to be operationally useful, subspecies must be discernible from one another (i.e., diagnosable); this element of “diagnosability,” or the ability to consistently distinguish between populations, is a common thread that runs through all subspecies concepts. The AOU Committee on Classification and Nomenclature offers the following definition of a subspecies: “Subspecies should represent geographically discrete breeding populations that are diagnosable from other populations on the basis of plumage and/or measurements, but are not yet reproductively isolated. Varying levels of diagnosability have been proposed for subspecies, typically ranging from at least 75% to 95% * * * subspecies that are phenotypically but not genetically distinct still warrant recognition if individuals can be assigned to a subspecies with a high degree of certainty” (AOU 2010). In addition, the latest AOU Check-list of North American Birds describes subspecies as: “geographic segments of species' populations that differ abruptly and discretely in morphology or coloration; these differences often correspond with difference in behavior and habitat” (AOU 1998, p. xii).

In general, higher levels of confidence in the classification of subspecies may be gained through the concurrence of multiple morphological, molecular, ecological, behavioral, and/or physiological characters (Haig

et al.

2006, p. 1591). The AOU definition of subspecies also incorporates this concept of looking for multiple lines of evidence, in referring to abrupt and discrete differences in morphology, coloration, and often corresponding differences in behavior or habitat as well (AOU 1998, p. xii). To assess subspecies diagnosability, we evaluated all the best scientific and commercial information available to determine whether the evidence points to a consistent separation of birds currently purported to be “western sage-grouse” from other populations of greater sage-grouse. This evaluation incorporated information that has become available since the AOU's last subspecies review in 1957, and included data on the geographic separation of the putative eastern and western subspecies, behavior, morphology, and genetics. If the assessment of these multiple characters provided a clear and consistent separation of the putative western subspecies from other populations of sage-grouse, such that any individual bird from the range of the western sage-grouse would likely be correctly assigned to that subspecies on the basis of the suite of characteristics analyzed, that would be considered indicative of a likely valid subspecies.

Geography

The delineation between eastern and western subspecies is vaguely defined and has changed over time from its original description (Aldrich 1946, p. 129; Aldrich and Duvall 1955 p. 12; AOU 1957, p. 139; Aldrich 1963, pp. 539-541). The boundary between the subspecies is generally described along a line starting on the Oregon-Nevada border south of Hart Mountain National Wildlife Refuge and ending near Nyssa, Oregon (Aldrich and Duvall 1955, p. 12; Aldrich 1963, pp. 539-541). Aldrich described the original eastern and western ranges in 1946 (Aldrich 1946, p. 129), while Aldrich and Duvall (1955, p. 12) and Aldrich (1963, pp. 539-541) described an intermediate form in northern California, presumably in a zone of intergradation between the subspecies. All of Aldrich's citations include a portion of Idaho within the western subspecies' range, but the 1957 AOU designation included Idaho as part of the eastern subspecies (AOU 1957, p. 139).

Our evaluation reveals that a boundary between potential western and eastern subspecies may be drawn multiple ways depending on whether one uses general description of historical placement, by considering topographic features, or in response to the differing patterns reported in studying sage-grouse genetics, morphology, or behavior. In their description of greater sage-grouse distribution, Schroeder

et al.

(2004, p. 369) noted the lack of evidence for differentiating between the purported subspecies, stating “We did not quantify the respective distributions of the eastern and western subspecies because of the lack of a clear dividing line (Aldrich and Duvall 1955) and the lack of genetic differentiation (Benedict

et al.

2003).” Based on this information, there does not appear to be any clear and consistent geographic separation between sage-grouse historically described as “eastern” and “western.”

Morphology

As noted above, the original description of the western subspecies of sage-grouse was based solely on differences in coloration (specifically, reduced white markings and darker feathering on western birds) among 11 museum specimens (10 whole birds, 1 head only) collected from 8 locations in Washington, Oregon, and California (Aldrich 1946, p. 129). By today's standards, this represents an extremely small sample size that would likely

yield little confidence in the ability to discriminate between populations on the basis of this character. Furthermore, the subspecies designation was based on this single characteristic; no other differences between the western and eastern subspecies of sage-grouse were noted in Aldrich's original description (Aldrich 1946, p. 129; USFWS 2010). Banks (1992) noted plumage color variation in the original specimens Aldrich (1946) used to make his subspecies designation, and agreed that the specimens from Washington, Oregon, and northern California did appear darker than the specimens collected in the eastern portion of the range. However, individual morphological variation in greater sage-grouse, such as plumage coloration, is extensive (Banks 1992). Further, given current taxonomic concepts, Banks (1992) doubted that most current taxonomists would identify a subspecies based on minor color variations from a limited number of specimens, as were available to Aldrich during the mid-1900s (Aldrich 1946, p. 129; Aldrich and Duvall 1955, p. 12; Aldrich 1963, pp. 539-541). Finally, the AOU Committee on Classification has stated that, because of discoloration resulting from age and poor specimen preparation, museum specimens “nearly always must be supplemented by new material for comprehensive systematic studies.” (AOU, Check-list of North American Birds, 7

th

ed., 1998, p. xv.)

Schroeder (2008, pp. 1-19) examined previously collected morphological data across the species' range from both published and unpublished sources. He found statistically significant differences between sexes, age groups, and populations in numerous characteristics including body mass, wing length, tail length, and primary feather length. Many of these differences were associated with sex and age, but body mass also varied by season. There also were substantial morphometric (size and shape) differences among populations. Notably, however, these population differences were not consistent with any of the described geographic delineations between eastern and western subspecies. For example, sage-grouse from Washington and from Northern Colorado up to Alberta appeared to be larger than those in Idaho, Nevada, Oregon, and California (Schroeder 2008, p. 9). This regional variation was not consistent with differences in previously established genetic characteristics (Oyler-McCance

et al.

2005, as cited in Schroeder 2008, p. 9). Thus our review revealed no clear basis for differentiating between the two described subspecies based on plumage or morphology.

Behavior

The only data available with respect to behavior are for strutting behavior on leks, a key component of mate selection. One recent study compared the male strut behavior between three sage-grouse populations that happen to include populations from both sides of the putative eastern-western line (Taylor and Young 2006, pp. 36-41). However, the classification of these populations changes depending on the description of western sage-grouse used. The Lyon/Mono population falls within the intermediate zone identified by Aldrich and Duvall (1955, p. 12) but would be classified as eastern under Aldrich (1963, p. 541). The Lassen population may be considered either western (Aldrich 1946, p. 129) or intermediate (Aldrich and Duvall 1955, p. 12; Aldrich 1963, p. 541). The Nye population falls within the range of the eastern sage-grouse (Aldrich and Duvall 1955, p. 12; Aldrich 1963, p. 541). The researchers found that male strut rates were not significantly different between populations, but that acoustic components of the display for the Lyon/Mono and Lassen populations (considered intermediate and/or western) were similar to each other, whereas the Nye population (eastern) was distinct. We consider these results inconclusive in distinguishing between eastern and western subspecies because of the inconsistent results and limited geographic scope of the study.

Schroeder (2008, p. 9) also examined previously collected data on strutting behavior on leks, including Taylor and Young (2006). He noted that, although there was regional variation in the strut rate of sage-grouse, it was not clear if this variation reflected population-level effects or some other unexplained variation. Based on the above limited information, we do not consider there to be any strong evidence of a clear separation of the western sage-grouse from other populations on the basis of behavioral differences.

Genetics

Genetic research can sometimes augment or refine taxonomic definitions that are based on morphology or behavior or both (discussed in Haig

et al.

2006, p. 1586; Oyler-McCance and Quinn in press, p. 19). Benedict

et al.

(2003, p. 309) found no genetic data supporting a subspecies designation. To investigate taxonomic questions and examine levels of gene flow and connectedness among populations, Oyler-McCance

et al.

(2005, p. 1294) conducted a comprehensive examination of the distribution of genetic variation across the entire range of greater sage-grouse, using both mitochondrial and nuclear deoxyribonucleic acid (DNA) sequence data. Oyler-McCance

et al.

(2005, p. 1306) found that the overall distribution of genetic variation showed a gradual shift across the range in both mitochondrial and nuclear DNA data sets. Their results demonstrate that greater sage-grouse populations follow an isolation-by-distance model of restricted gene flow (gene flow resulting from movement between neighboring populations rather than being the result of long distance movements of individuals) (Oyler-McCance

et al.

2005, p. 1293; Campton 2007, p. 4), and are not consistent with subspecies designations. Oyler-McCance and Quinn (in press, entire) reviewed available studies that used molecular genetic approaches, including Oyler-McCance

et al.

(2005). They examined the genetic data bearing on the delineation of the western and eastern subspecies of greater sage-grouse, and determined that the distinction is not supported by the genetic data (Oyler-McCance and Quinn in press, p. 4). The best available genetic information thus does not support the recognition of the western sage-grouse as a separate subspecies.

Summary: Taxonomic Evaluation of the Subspecies

The AOU has not revisited the question of whether the eastern and western subspecies are valid since their original classification in 1957. We have examined the best scientific information available regarding the putative subspecies of the greater sage-grouse and have considered multiple lines of evidence for the potential existence of western and eastern subspecies based on geographic, morphological, behavioral, and genetic data. In our evaluation, we looked for any consistent significant differences in these characters that might support recognition of the western or eastern sage-grouse as clear, discrete, and diagnosable populations, such that either might be considered a subspecies.

As described above, the boundaries distinguishing the two putative subspecies have shifted over time, and there does not appear to be any clear and consistent geographic separation between sage-grouse historically described as “eastern” and “western.” Banks (1992) and Schroeder (2008, p. 9) both found morphological variations between individuals and populations, but Banks stated that the differences would not be sufficient to recognize

subspecies by current taxonomic standards, and Schroeder noted that the differences were not consistent with any of the described geographic or genetic delineations between putative subspecies. Schroeder (2008 p. 9) also noted regional behavior differences in strut rate, but stated it was not clear if this variation reflected population-level effects. Finally, the best available genetic information indicates there is no distinction between the putative western and eastern subspecies (Benedict

et al.

2003, p. 309; Oyler-McCance and Quinn in press, p. 12).

Because the best scientific and commercial information do not support the taxonomic validity of the purported eastern or western subspecies, our analysis of the status of the greater sage-grouse (below) does not address considerations at the scale of subspecies. (See Findings section, below, for our finding on the petition to list the western subspecies of the greater sage-grouse.)

Life History Characteristics

Greater sage-grouse depend on a variety of shrub-steppe habitats throughout their life cycle, and are considered obligate users of several species of sagebrush (e.g.,

Artemisia tridentata

ssp.

wyomingensis

(Wyoming big sagebrush),

A. t.

ssp.

vaseyana

(mountain big sagebrush), and

A. t. tridentata

(basin big sagebrush)) (Patterson 1952, p. 48; Braun

et al.

1976, p. 168; Connelly

et al.

2000a, pp. 970-972; Connelly

et al.

2004, p. 4-1; Miller

et al.

in press, p. 1). Greater sage-grouse also use other sagebrush species such as

A. arbuscula

(low sagebrush),

A. nova

(black sagebrush),

A. frigida

(fringed sagebrush), and

A. cana

silver sagebrush (Schroeder

et al.

1999, pp. 4-5; Connelly

et al.

2004, p. 3-4). Thus, sage-grouse distribution is strongly correlated with the distribution of sagebrush habitats (Schroeder

et al.

2004, p. 364). Sage-grouse exhibit strong site fidelity (loyalty to a particular area even when the area is no longer of value) to seasonal habitats, which includes breeding, nesting, brood rearing, and wintering areas (Connelly

et al.

2004, p. 3-1). Adult sage-grouse rarely switch between these habitats once they have been selected, limiting their adaptability to changes.

During the spring breeding season, male sage-grouse gather together to perform courtship displays on areas called leks. Areas of bare soil, short-grass steppe, windswept ridges, exposed knolls, or other relatively open sites typically serve as leks (Patterson 1952, p. 83; Connelly

et al.

2004, p. 3-7 and references therein). Leks are often surrounded by denser shrub-steppe cover, which is used for escape, thermal, and feeding cover. The proximity, configuration, and abundance of nesting habitat are key factors influencing lek location (Connelly

et al

., 1981, and Connelly

et al

., 2000 b, cited in Connelly

et al

., in press a, p. 11). Leks can be formed opportunistically at any appropriate site within or adjacent to nesting habitat (Connelly

et al

. 2000a, p. 970), and, therefore, lek habitat availability is not considered to be a limiting factor for sage-grouse (Schroeder 1999, p. 4). Nest sites are selected independent of lek locations, but the reverse is not true (Bradbury

et al

. 1989, p. 22; Wakkinen

et al

. 1992, p. 382). Thus, leks are indicative of nesting habitat.

Leks range in size from less than 0.04 hectare (ha) (0.1 acre (ac)) to over 36 ha (90 ac) (Connelly

et al

. 2004, p. 4-3) and can host from several to hundreds of males (Johnsgard 2002, p. 112). Males defend individual territories within leks and perform elaborate displays with their specialized plumage and vocalizations to attract females for mating. Although males are capable of breeding the first spring after hatch, young males are rarely successful in breeding on leks due to the dominance of older males (Schroeder

et al

. 1999, p. 14). Numerous researchers have observed that a relatively small number of dominant males account for the majority of copulations on each lek (Schroeder

et al

. 1999, p. 8). However, Bush (2009, p. 106) found on average that 45.9 percent (range 14.3 to 54.5 percent) of genetically identified males in a population fathered offspring in a given year, which indicates that males and females likely engage in off-lek copulations. Males do not participate in incubation of eggs or rearing chicks.

Females have been documented to travel more than 20 km (12.5 mi) to their nest site after mating (Connelly

et al

. 2000a, p. 970), but distances between a nest site and the lek on which breeding occurred is variable (Connelly

et al

. 2004, pp. 4-5). Average distance between a female's nest and the lek on which she was first observed ranged from 3.4 km (2.1 mi) to 7.8 km (4.8 mi) in five studies examining 301 nest locations (Schroeder

et al

. 1999 p. 12).

Productive nesting areas are typically characterized by sagebrush with an understory of native grasses and forbs, with horizontal and vertical structural diversity that provides an insect prey base, herbaceous forage for pre-laying and nesting hens, and cover for the hen while she is incubating (Gregg 1991, p. 19; Schroeder

et al

. 1999, p. 4; Connelly

et al

. 2000a, p. 971; Connelly

et al

. 2004, pp. 4-17, 18; Connelly

et al

. in press b, p. 12). Sage-grouse also may use other shrub or bunchgrass species for nest sites (Klebenow 1969, p. 649; Connelly

et al

. 2000a, p. 970; Connelly

et al

. 2004, p. 4-4). Shrub canopy and grass cover provide concealment for sage-grouse nests and young, and are critical for reproductive success (Barnett and Crawford 1994, p. 116; Gregg

et al

. 1994, p. 164; DeLong

et al

.1995, p. 90; Connelly

et al

. 2004, p. 4-4). Published vegetation characteristics of successful nest sites included a sagebrush canopy cover of 15-25 percent, sagebrush heights of 30 to 80 cm (11.8 to 31.5 in.), and grass/forb cover of 18 cm (7.1 in.) (Connelly

et al

. 2000a, p. 977).

Sage-grouse clutch size ranges from 6 to 9 eggs with an average of 7 eggs (Connelly

et al

. in press a, pp. 14-15). The likelihood of a female nesting in a given year averages 82 percent in eastern areas of the range (Alberta, Montana, North Dakota, South Dakota, Colorado, Wyoming) and 78 percent in western areas of the range (California, Nevada, Idaho, Oregon, Washington, Utah ) (Connelly

et al

. in press a, p. 15). Adult females have higher nest initiation rates than yearling females (Connelly

et al

. in press a, p. 15). Nest success (one or more eggs hatching from a nest), as reported in the scientific literature, varies widely (15-86 percent Schroeder

et al

. 1999, p. 11). Overall, the average nest success for sage-grouse in habitats where sagebrush has not been disturbed is 51 percent and for sage-grouse in disturbed habitats is 37 percent (Connelly

et al

., in press a, p. 1). Re-nesting only occurs if the original nest is lost (Schroeder

et al

. 1999, p. 11). Sage-grouse re-nesting rates average 28.9 percent (based on 9 different studies) with a range from 5 to 41 percent (Connelly

et al

. 2004. p. 3-11). Other game bird species have much higher re-nesting rates, often exceeding 75 percent. The impact of re-nesting on annual productivity for most sage-grouse populations is unclear and thought to be limited (Crawford

et al

. 2004, p. 4). In north-central Washington State, re-nesting contributed to 38 percent of the annual productivity of that population (Schroeder 1997, p. 937). However, the author postulated that the re-nesting efforts in this population may be greater than anywhere else in the species' range because environmental conditions allow a longer period of time to successfully rear a clutch (Schroeder 1997, p. 939).

Little information is available on the level of productivity (number of chicks per hen that survive to fall) that is necessary to maintain a stable population (Connelly

et al

. 2000b, p.

970). However, Connelly

et al

. (2000b, p. 970, and references therein) suggest that 2.25 chicks per hen are necessary to maintain stable to increasing populations. Long-term productivity estimates of 1.40-2.96 chicks per hen across the species range have been reported (Connelly and Braun 1997, p. 20). Productivity declined slightly after 1985 to 1.21-2.19 chicks per hen (Connelly and Braun 1997, p. 20). Despite average clutch sizes of 7 eggs (Connelly

et al

. in press a, p. 15) due to low chick survival and limited renesting, there is little evidence that populations of sage-grouse produce large annual surpluses (Connelly

et al

. in press a, p. 24).

Hens rear their broods in the vicinity of the nest site for the first 2-3 weeks following hatching (within 0.2-5 km (0.1-3.1 mi)), based on two studies in Wyoming (Connelly

et al

. 2004, p. 4-8). Forbs and insects are essential nutritional components for chicks (Klebenow and Gray 1968, p. 81; Johnson and Boyce 1991, p. 90; Connelly

et al

. 2004, p. 4-9). Therefore, early brood-rearing habitat must provide adequate cover (sagebrush canopy cover of 10 to 25 percent; Connelly

et al

. 2000a, p. 977) adjacent to areas rich in forbs and insects to ensure chick survival during this period (Connelly

et al

. 2004, p. 4-9).

All sage-grouse gradually move from sagebrush uplands to more mesic areas (moist areas such as streambeds or wet meadows) during the late brood-rearing period (3 weeks post-hatch) in response to summer desiccation of herbaceous vegetation (Connelly

et al

. 2000a, p. 971). Summer use areas can include sagebrush habitats as well as riparian areas, wet meadows, and alfalfa fields (Schroeder

et al

. 1999, p. 4). These areas provide an abundance of forbs and insects for both hens and chicks (Schroeder

et al

. 1999, p. 4; Connelly

et al

. 2000a, p. 971). Sage-grouse will use free water although they do not require it since they obtain their water needs from the food they eat. However, natural water bodies and reservoirs can provide mesic areas for succulent forb and insect production, thereby attracting sage-grouse hens with broods (Connelly

et al

. 2004, p. 4-12). Broodless hens and cocks also will use more mesic areas in close proximity to sagebrush cover during the late summer, often arriving before hens with broods (Connelly

et al

. 2004, p. 4-10).

As vegetation continues to desiccate through the late summer and fall, sage-grouse shift their diet entirely to sagebrush (Schroeder

et al

. 1999, p. 5). Sage-grouse depend entirely on sagebrush throughout the winter for both food and cover. Sagebrush stand selection is influenced by snow depth (Patterson 1952, p. 184; Hupp and Braun 1989, p. 827), availability of sagebrush above the snow to provide cover (Connelly

et al

. 2004, pp. 4-13, and references therein) and, in some areas, topography (e.g., elevation, slope and aspect; Beck 1977, p. 22; Crawford

et al

. 2004, p. 5).

Many populations of sage-grouse migrate between seasonal ranges in response to habitat distribution (Connelly

et al

. 2004, p. 3-5). Migration can occur between winter and breeding and summer areas, between breeding, summer, and winter areas, or not at all. Migration distances of up to 161 km (100 mi) have been recorded (Patterson 1952, p.189); however, distances vary depending on the locations of seasonal habitats (Schroeder

et al

. 1999, p. 3). Migration distances for female sage-grouse generally are less than for males (Connelly

et al

. 2004, p. 3-4), but in one study in Colorado, females traveled farther than males (Beck 1977, p. 23). Almost no information is available regarding the distribution and characteristics of migration corridors for sage-grouse (Connelly

et al

. 2004, p. 4-19). Sage-grouse dispersal (permanent moves to other areas) is poorly understood (Connelly

et al

. 2004, p. 3-5) and appears to be sporadic (Dunn and Braun 1986, p. 89). Estimating an “average” home range for sage-grouse is difficult due to the large variation in sage-grouse movements both within and among populations. This variation is related to the spatial availability of habitats required for seasonal use, and annual recorded home ranges have varied from 4 to 615 square kilometers (km

2

) (1.5 to 237.5 square miles (mi

2

)) (Connelly

et al

., in press a, p. 10).

Sage-grouse typically live between 3 and 6 years, but individuals up to 9 years of age have been recorded in the wild (Connelly

et al

. 2004, p. 3-12). Hens typically survive longer due to a disproportionate impact of predation on leks to males (Schroeder

et al

. 1999, p. 14). Juvenile survival (from hatch to first breeding season) is affected by food availability, habitat quality, harvest, and weather. Based on a review of many field studies, juvenile survival rates range from 7 to 60 percent (Connelly

et al

. 2004, p. 3-12). The variation in juvenile mortality rates may be associated with gender, weather, harvest rates, age of brood female (broods with adult females have higher survival), and with habitat quality (rates increase in poor habitats) (Schroeder

et al

. 1999, p. 14; Connelly

et al

., in press a, p. 20). The average annual survival rate for male sage-grouse (all ages combined) documented in various studies ranged from 38 to 60 percent and 55 to 75 percent for females (Schroeder

et al

. 1999, p. 14). Higher female survival rates account for a female-biased sex ratio in adult birds (Schroeder 1999, p. 14; Johnsgard 2002, p. 621). The sex ratio of sage-grouse breeding populations varies widely with values between 1.2 and 3 females per male being reported (Connelly

et al

., in press a, p. 23). Although seasonal patterns of mortality have not been thoroughly examined, over-winter mortality appears to be low (Connelly

et al

. 2000b, p. 229; Connelly

et al

. 2004, p. 9-4). While both males and females are capable of breeding the first spring after hatch, young males are rarely successful due to the dominance of older males on the lek (Schroeder

et al

. 1999, p. 14). Nesting rates of yearling females are 25 percent less than adult females (Schroeder

et al

. 1999, p. 13).

Habitat Description and Characteristics

Sage-grouse are dependent on large areas of contiguous sagebrush (Patterson 1952, p. 48; Connelly

et al

. 2004, p. 4-1; Connelly

et al

. in press a, p. 10; Wisdom

et al

. in press, p. 4), and large-scale characteristics within surrounding landscapes influence sage-grouse habitat selection (Knick and Hanser in press, p. 26). Sagebrush is the most widespread vegetation in the intermountain lowlands in the western United States (West and Young 2000, p. 259) and is considered one of the most imperiled ecosystems in North America (Knick

et al

. 2003, p. 612; Miller

et al

. in press, p. 4, and references therein). Scientists recognize 14 species and 13 subspecies of sagebrush (Connelly

et al

. 2004, p. 5-2; Miller

et al

. in press, p. 8), each with unique habitat requirements and responses to perturbations (West and Young 2000, p. 259). Sagebrush species and subspecies occurrence in an area is dictated by local soil type, soil moisture, and climatic conditions (West 1983, p. 333; West and Young 2000, p. 260; Miller

et al

. in press, pp. 8-11). The degree of dominance by sagebrush varies with local site conditions and disturbance history. Plant associations, typically defined by perennial grasses, further define distinctive sagebrush communities (Miller and Eddleman 2000, pp. 10-14; Connelly

et al

. 2004, p. 5-3), and are influenced by topography, elevation, precipitation, and soil type. These ecological conditions influence the response and resiliency of sagebrush and their associated understories to natural and human-caused changes.

Sagebrush is typically divided into two groups, big sagebrush and low sagebrush, based on their affinities for

different soil types (West and Young 2000, p. 259). Big sagebrush species and subspecies, such as

A. tridentata

ssp.

wyomingensis

, are limited to coarse-textured and/or well-drained sediments. Low sagebrush, such as

A. nova,

typically occur where erosion has exposed clay or calcified soil horizons (West 1983, p. 334; West and Young 2000, p. 261). Reflecting these soil differences, big sagebrush will die if surfaces are saturated long enough to create anaerobic conditions for 2 to 3 days (West and Young 2000, p. 259). Some low sagebrush are more tolerant of occasionally supersaturated soils, and many low sage sites are partially flooded during spring snowmelt. None of the sagebrush taxa tolerate soils with high salinity (West 1983, p. 333; West and Young 2000, p. 257). Sagebrush that provide important annual and seasonal habitats for sage-grouse include three subspecies of big sagebrush (

A. t.

ssp.

wyomingensis, A. t

. ssp.

tridentata

and

A. t

. ssp.

vaseyana

), two low forms of sagebrush (

A. arbuscula

(little sagebrush) and

A. nova

), and

A. cana

ssp

. cana

(Miller

et al

. in press, p. 8).

All species of sagebrush produce large ephemeral leaves in the spring, which persist until reduced soil moisture occurs in the summer. Most species also produce smaller, over-wintering leaves in the late spring that last through summer and winter. Sagebrush have fibrous tap root systems, which allow the plants to draw surface soil moisture, and also to access water deep within the soil profile when surface water is limited (West and Young 2000, p. 259). Most sagebrush flower in the fall. However, during years of drought or other moisture stress, flowering may not occur. Although seed viability and germination are high, seed dispersal is limited. Sagebrush seeds, depending on the species, remain viable for 1 to 3 years. However, Wyoming big sagebrush seeds do not persist beyond the year of their production (West and Young 2000, p. 260).

Sagebrush is long-lived, with plants of some species surviving up to 150 years (West 1983, p. 340). They produce allelopathic chemicals that reduce seed germination, seedling growth, and root respiration of competing plant species and inhibit the activity of soil microbes and nitrogen fixation. Sagebrush has resistance to environmental extremes, with the exception of fire and occasionally defoliating insects (e.g., webworm (

Aroga

spp.); West 1983, p. 341). Most species of sagebrush are killed by fire (West 1983, p. 341; Miller and Eddleman 2000, p. 17; West and Young 2000, p. 259), and historic fire-return intervals were as long as 350 years, depending on sagebrush type and environmental conditions (Baker in press, p. 16). Natural sagebrush recolonization in burned areas depends on the presence of adjacent live plants for a seed source or on the seed bank, if present (Miller and Eddleman 2000, p. 17), and requires decades for full recovery.

Plants associated with the sagebrush understory vary, as does their productivity. Both plant composition and productivity are influenced by moisture availability, soil characteristics, climate, and topographic position (Miller

et al

., in press, pp. 8-14). Forb abundance can be highly variable from year to year and is largely affected by the amount and timing of precipitation.

Very little sagebrush within its extant range is undisturbed or unaltered from its condition prior to EuroAmerican settlement in the late 1800s (Knick

et al

. 2003, p. 612, and references therein). Due to the disruption of primary patterns, processes, and components of sagebrush ecosystems since EuroAmerican settlement (Knick

et al

. 2003, p. 612; Miller

et al

. in press, p. 4), the large range of abiotic variation, the minimal short-lived seed banks, and the long generation time of sagebrush, restoration of disturbed areas is very difficult. Not all areas previously dominated by sagebrush can be restored because alteration of vegetation, nutrient cycles, topsoil, and living (cryptobiotic) soil crusts has exceeded recovery thresholds (Knick

et al

. 2003, p. 620). Additionally, processes to restore sagebrush ecology are relatively unknown (Knick

et al

. 2003, p. 620). Active restoration activities are often limited by financial and logistic resources and lack of political motivation (Knick

et al

. 2003, p. 620; Miller

et al

. in press, p. 5) and may require decades or centuries (Knick

et al

. 2003, p. 620, and references therein). Meaningful restoration for greater sage-grouse requires landscape, watershed, or eco-regional scale context rather than individual, unconnected efforts (Knick

et al

. 2003, p. 623, and references therein; Wisdom

et al

. in press, p. 27). Landscape restoration efforts require a broad range of partnerships (private, State, and Federal) due to landownership patterns (Knick

et al

. 2003, p. 623; see discussion of landownership below). Except for areas where active restoration is attempted following disturbance (e.g., mining, wildfire), management efforts in sagebrush ecosystems are usually focused on maintaining the remaining sagebrush (Miller

et al

. in press, p. 5; Wisdom

et al

. in press, pp. 26, 30).

Greater sage-grouse require large, interconnected expanses of sagebrush with healthy, native understories (Patterson 1952, p. 9; Knick

et al

. 2003, p. 623; Connelly

et al

. 2004, pp. 4-15; Connelly

et al

. in press a, p. 10; Pyke in press, p. 7; Wisdom

et al

. in press, p. 4). There is little information available regarding minimum sagebrush patch sizes required to support populations of sage-grouse. This is due in part to the migratory nature of some but not all sage-grouse populations, the lack of juxtaposition of seasonal habitats, and differences in local, regional, and range-wide ecological conditions that influence the distribution of sagebrush and associated understories. Where home ranges have been reported (Connelly

et al

. in press a, p. 10 and references therein), they are extremely variable (4 to 615 km

2

range (1.5 to 237.5 mi

2

)). Occupancy of a home range also is based on multiple variables associated with both local vegetation characteristics and landscape characteristics (Knick

et al

. 2003, p. 621). Pyke (in press, p. 18) estimated that greater than 4,000 ha (9,884 ac) was necessary for population sustainability. However, he did not indicate whether this value was for migratory or nonmigratory populations, nor if this included juxtaposition of all seasonal habitats. Large seasonal and annual movements emphasize the landscape nature of the greater sage-grouse (Knick

et al

. 2003, p. 624; Connelly

et al

. in press a, p. 10).

Range and Distribution of Sage-Grouse and Sagebrush

Prior to settlement of western North America by European immigrants in the 19th century, greater sage-grouse occurred in 13 States and 3 Canadian provinces—Washington, Oregon, California, Nevada, Idaho, Montana, Wyoming, Colorado, Utah, South Dakota, North Dakota, Nebraska, Arizona, British Columbia, Alberta, and Saskatchewan (Schroeder

et al

. 1999, p. 2; Young

et al

. 2000, p. 445; Schroeder

et al

. 2004, p. 369). Sagebrush habitats that potentially supported sage-grouse occurred over approximately 1,200,483 km

2

(463,509 mi

2

) before 1800 (Schroeder

et al

. 2004, p. 366). Currently, greater sage-grouse occur in 11 States (Washington, Oregon, California, Nevada, Idaho, Montana, Wyoming, Colorado, Utah, South Dakota, and North Dakota), and 2 Canadian provinces (Alberta and Saskatchewan), occupying approximately 56 percent of their historical range (Schroeder

et al

. 2004, p. 369). Approximately 2 percent of the total range of the greater sage-grouse

occurs in Canada, with the remainder in the United States (Knick in press, p. 14).

Sage-grouse have been extirpated from Nebraska, British Columbia, and possibly Arizona (Schroeder

et al

. 1999, p. 2; Young

et al

. 2000 p. 445; Schroeder

et al

. 2004, p. 369). Current distribution of the greater sage-grouse is estimated at 668,412 km

2

(258,075 mi

2

; Connelly

et al

. 2004, p. 6-9; Schroeder

et al

. 2004, p. 369). Changes in distribution are the result of sagebrush alteration and degradation (Schroeder

et al

. 2004, p. 363).

Sage-grouse distribution is associated with sagebrush (Schroeder

et al

. 2004; p. 364), although sagebrush is more widely distributed. However, sagebrush does not always provide suitable habitat due to fragmentation and degradation (Schroeder

et al

. 2004, pp. 369, 372). Very little of the extant sagebrush is undisturbed, with up to 50 to 60 percent having altered understories or having been lost to direct conversion (Knick

et al

. 2003, p. 612 ). There also are challenges in mapping altered and depleted understories, particularly in semi-arid regions, so maps depicting only sagebrush as a dominant cover type are deceptive in their reflection of habitat quality and, therefore, use by sage-grouse (Knick

et al

. 2003, p. 616). As such, variations in the quality of sagebrush habitats (from either abiotic or anthropogenic events) are reflected by sage-grouse distribution and densities (Figure 1).

EP23MR10.000

Sagebrush occurs in two natural vegetation types that are delineated by temperature and patterns of precipitation (Miller

et al

. in press, p. 7). Sagebrush steppe ranges across the northern portion of sage-grouse range, from British Columbia and the Columbia Basin, through the northern Great Basin, Snake River Plain, and Montana, and into the Wyoming Basin and northern Colorado. Great Basin sagebrush occurs south of sagebrush steppe, and extends from the Colorado Plateau westward into Nevada, Utah, and California (Miller

et al

. in press, p. 7). Other sagebrush types within greater sage-grouse range include mixed-desert shrubland in the Bighorn Basin of Wyoming, and grasslands in eastern Montana and Wyoming that also support

A. cana

and

A. filifolia

(sand sagebrush) (Miller

et al

. in press, p. 7).

Due to differences in the ecology of sagebrush across the range of the greater sage-grouse, the Western Association of Fish and Wildlife Agencies (WAFWA) delineated seven Management Zones (MZs I-VII) based primarily on floristic provinces (Figure 2; Table 1; Stiver

et al

. 2006, p. 1-6). The boundaries of these MZs were delineated based on their ecological and biological attributes rather than on arbitrary political boundaries (Stiver

et al

. 2006, p. 1-6). Therefore, vegetation found within a MZ is similar and sage-grouse and their habitats within these areas are likely to respond similarly to environmental factors and management actions. The WAFWA conservation strategy includes the Gunnison sage-grouse, and the boundary for MZ VII includes its range (Stiver

et al

. 2006, pp. 1-1, 1-8), which does not overlap with the range of the greater sage-grouse.

Table 1—The Management Zones of the greater sage-grouse as defined by Stiver

et al.

(2006, pp. 1-7, 1-11).

MZ

STATES AND PROVINCES INCLUDED

FLORISTIC REGION

I

MT, WY, ND, SD, SK, AL

Great Plains

II

ID, WY, UT, CO

Wyoming Basin

III

UT, NV, CA

Southern Great Basin

IV

ID, UT, NV, OR

Snake River Plain

V

OR, CA, NV

Northern Great Basin

VI

WA

Columbia Basin

VII

CO, UT

Colorado Plateau

EP23MR10.001

As stated above, due to the variability in habitat conditions, sage-grouse are not evenly distributed across the range (Figure 1). The MZs I, II, IV, and V encompass the core populations of greater sage-grouse and have the highest reported densities (Table 2, Figures 1, 2; Stiver

et al

. 2006, p. 1-12). The MZ III is composed of lower density populations in the Great Basin, while fewer numbers of more dispersed birds occur in MZ VI (Stiver

et al

. 2006, p. 1-7).

Table 2—Relative abundance of greater sage-grouse leks, and numbers of males attending leks by Management Zone, based on the mean number of individual leks and mean maximum number of males attending leks by MZ during 2005-2007.

MZ

Relative Abundance of Leks

Relative Abundance of Males

Attending Leks

I

0.17

0.15

II

0.48

0.50

III

0.06

0.07

IV

0.19

0.18

V

0.09

0.10

VI

0.004

0.005

VII

0.003

0.003

Land Ownership of Habitats

Greater sage-grouse extant habitats have multiple surface ownerships, as reflected in Table 3. Most of the habitats occur on Federal surfaces, a reflection of land disposal practices during EuroAmerican settlement of the western United States (Knick in press, pp. 5-10). Lands dominated by sagebrush that were disposed to private ownership typically had deeper soils and greater available water capacity or access to water (valley bottoms), reflecting their capacity for agricultural development or increased grazing activities (Knick in press, p. 15). The lands remaining in Federal ownership were of poorer overall quality. The resulting low productivity on Federal surfaces affects their ability to recover from disturbance (Knick in press, p. 17).

Federal agencies manage almost two-thirds of the sagebrush habitats (Table 3). The Bureau of Land Management (BLM) manages just over half of sage-grouse habitats, while the U.S. Forest Service (USFS) is responsible for management of approximately 8 percent of sage-grouse habitat (Table 3). Other Federal agencies, including the Service, Bureau of Indian Affairs (BIA), Bureau of Reclamation (BOR), National Park Service (NPS), Department of Defense (DOD), and Department of Energy (DOE) also are responsible for sagebrush habitats, but at a much smaller scale (Table 3). State agencies manage approximately 5 percent of sage-grouse habitats.

Table 3—Percent surface ownership of total sagebrush area (km

2

(mi

2

)) within the sage-grouse management zones (from Knick in press, p. 39). Other Federal agencies include the Service, BOR, NPS, DOD, and DOE. MZ VII includes both Gunnison and greater sage-grouse.

Sage-grouse MZ

km

2

mi

2

Sagebrush Management and Ownership

BLM

Percent

Private

Percent

USFS

Percent

State

Percent

BIA

Percent

Other

Federal

Percent

I Great Plains

50,264

19,407

17

66

2

7

4

3

II Wyoming Basin

108,771

41,996

49

35

4

7

4

1

III Southern Great Basin

92,173

35,588

73

13

10

3

1

0

IV Snake River Plain

134,187

51,810

53

29

11

6

1

0

V Northern Great Basin

65,536

25,303

62

21

10

1

1

6

VI Columbia Basin

12,105

4,674

6

64

2

12

13

3

VII Colorado Plateau

17,534

6,770

42

36

6

6

9

1

TOTALS

480,570

185,549

52

31

8

5

3

1

Population Size

Estimates of greater sage-grouse abundance were mostly anecdotal prior to the implementation of systematic surveys in the 1950s (Braun 1998, p. 139). Early reports suggested the birds were abundant throughout their range, with estimates of historical populations ranging from 1,600,000 to 16,000,000 birds (65 FR 51580, August 24, 2000). However, concerns about extinction were raised in early literature due to market hunting and habitat alteration (Hornaday 1916, pp. 181-185). Following a review of published literature and anecdotal reports, Connelly

et al

. (2004, ES-1-3) concluded that the abundance of sage-grouse has declined from presettlement (defined as 1800) numbers. Most of the historical

population changes were the result of local extirpations, which has been inferred from a 44 percent reduction in sage-grouse distribution described by Schroeder

et al

. 2004 (Connelly

et al

. 2004, p. 6-9).

Population numbers are difficult to estimate due to the large range of the species, physical difficulty in accessing some areas of habitat, the cryptic coloration and behavior of hens (Garton

et al

. in press, p. 6), and survey protocols. Problems with inconsistent sampling protocols for lek surveys (e.g., number of times a lek is counted, number of leks surveyed in a year, observer bias, observer experience, time counted) were identified by Walsh

et al

. (2006, pp. 61-64) and Garton

et al

. (in press, p. 6), and many of those problems still persist (Stiver

et al

. 2006, p. 3-1). Additionally, estimating population sizes using lek data is difficult as the relationship of those data to actual population size (e.g., ratio of males to females, percent unseen birds) is usually unknown (WAFWA 2008, p. 3). However, the annual counting of males on leks remains the primary approach to monitor long-term trends of populations (WAFWA 2008, p. 3), and standardized techniques are beginning to be implemented throughout the species' range (Stiver

et al

. 2006, pp. 3-1 to 3-16). The use of harvest data for estimating population numbers also is of limited value since both harvest and the population size on which harvest is based are estimates. Given the limitations of these data, States usually rely on a combination of actual counts of birds on leks and harvest data to estimate population size. Estimates of populations by State, generated from a variety of data sources, are provided in Table 4.

Table 4—Sage-grouse population estimates based on data from State wildlife agencies.

Location

Data Year

Source

Estimated

Population

CA/NV

2004

California/Nevada Sage-grouse Conservation Team (2004, p. 26)

88,000

CO

2008

2007 CO Conservation plan, based on adjusted male lek counts (count + 1.6 multiplier, sex ratio females:males) (Colorado Greater Sage-grouse Steering Committee 2008, p. 56)

22,646

ID

2007

Calculated based on assumption of 5% of population is harvested

(Service, unpublished data)

98,700

MT

2007

Calculated based on assumption of 5% of population is harvested

(Service, unpublished data)

62,320

ND

2007

2008 lek counts adjusted (assumes 75% of males counted at lek, & sex ratio of 2:1) (A. Robinson, NDGFD, pers. comm., 2008)

308

OR

2003

2003 Oregon Conservation Plan Estimate (Hagen 2005, p. 27)

40,000

SD

2007

South Dakota Game and Fish web page (last updated in 2007)

1,500

UT

2002

Utah Division of Wildlife Resources (2002, p. 13)

12,999

WA

2003

Washington Division of Fish and Wildlife (Stinson

et al.

2004, p. 21)

1,059

WY

2007

Calculated based on assumption of 5% of population is harvested

(Service, unpublished data)

207,560

Canada

2006

Government of Canada 2010

450

Braun (1998, p. 141) estimated that the minimum 1998 rangewide spring population numbered about 157,000 sage-grouse, derived from numbers of males counted on leks. The same year, State wildlife agencies within the range of the species estimated the population was at least 515,000 based on lek counts and harvest data (Warren 2008, pers. comm.). In 2000, we estimated the rangewide abundance of sage-grouse was between a minimum of 100,000 (taken from Braun 1998, p. 141) up to 500,000 birds (based on harvest data from Idaho, Montana, Oregon, and Wyoming, with the assumption that 10 percent of the population is typically harvested) (65 FR 51578, August 24, 2000). In 2003, based on increased lek survey efforts, Connelly

et al

. (2004, p. 13-5) concluded that rangewide population numbers were likely much greater than the 157,000 estimated by Braun (1998, p. 141), but they were unable to generate a rangewide population estimate. Garton

et al

., (in press, p. 2) estimated a rangewide minimum of 88,816 males counted on leks in 2007, the last year data were formally collated and reported. Estimates of historical populations range from 1,600,000 to 16,000,000 birds (65 FR 51580).

Population Trends

Although population numbers are difficult to estimate, the long-term data collected from counting males on leks provides insight to population trends. Periods of historical decline in sage-grouse abundance occurred from the late 1800s to the early-1900s (Hornaday 1916, pp. 179-221; Crawford 1982, pp. 3-6; Drut 1994, pp. 2-5; WDFW 1995; Braun 1998, p. 140; Schroeder

et al

. 1999, p. 1). Other noticeable declines in sage-grouse populations occurred in the 1920s and 1930s, and then again in the 1960s and 1970s (Connelly and Braun 1997, pp. 3-4; Braun 1998, p. 141). Declines in the 1920s and 1930s were attributed to hunting, and declines in the 1960s and 1970s were primarily as a result of loss of habitat quality and quantity (Connelly and Braun 1997, p. 2). State wildlife agencies were sufficiently concerned with the decline in the 1920s and 1930s that many closed their hunting seasons and others significantly reduced bag limits and season lengths as a precautionary measure (Patterson 1952, pp. 30-33; Autenrieth 1981, p. 10).

Using lek counts as an index for abundance, Connelly

et al

. (2004, p. 6-71) reported rangewide declines from 1965 through 2003. Declines averaged 2 percent per year from 1965 to 2003. The decline was more dramatic from 1965 through 1985, with an average annual change of 3.5 percent. The rate of decline rangewide slowed to 0.37 percent annually during 1986 to 2003 and some populations increased (Connelly

et al

. 2004, p. 6-71). Based on these analyses, Connelly

et al

. 2004 (p. 6-71) estimated that sage-grouse population numbers in the late 1960s and early 1970s were likely two to three times greater than current numbers (Connelly

et al

. 2004, p. 6-71). Using a statistical population reconstruction approach, Garton

et al

. (in press, p. 67) also demonstrated a pattern of higher numbers of sage-grouse in the late 1960s and early 1970s, which was supported by data from several other sources (Garton

et al

. in press, p. 68).

In 2008, WAFWA conducted new population trend analyses that incorporated an additional 4 years of data beyond the Connelly

et al

. 2004 analysis (WAFWA 2008, entire). Although the WAFWA analyses used different statistical techniques, lek counts also were used. WAFWA results were similar to Connelly

et al

. (2004) in that a long-term population decline was detected during 1965 to 2007 (average 3.1 percent annually; WAFWA 2008, p. 12). WAFWA attributed the decline to the reduction in number of active leks (WAFWA 2008, p. 51). Similar to Connelly

et al

. (2004), the WAFWA analyses determined that the rate of decline lessened during 1985 to 2007 (average annual change of 1.4 percent annually) (WAFWA 2008, p. 58). Garton

et al

. (in press, pp. 68-69) also had similar results. While the average annual rate of decline has lessened since 1985 (3.1 to 1.4 percent), population declines continue and populations are now at much lower levels than in the early 1980's. Therefore, these continuing negative trends at such low relative numbers are concerning regarding long-term population persistence. Similarly, short-term increases or stable trends, while on the surface seem encouraging, do not indicate that populations are recovering but may instead be a function of losing leks and not increases in numbers (WAFWA 2008, p.51). Population stability may also be compromised if cycles in sage-grouse populations (Schroeder

et al

. 1999, p. 15; Connelly

et al

. 2004, p.6-71) are lost, which current analyses suggest, minimizing the opportunities for population recovery if habitat were available (Garton 2009, pers. comm.).

Although the MZs were not formally adopted by WAFWA until 2006, the population trend analyses conducted by Connelly

et al

. (2004) included trend analyses based on the same floristic provinces used to define the zones. While the average annual rate of change was not presented, the results of those analyses indicated long-term declines in greater sage-grouse for MZs I, II, III, IV and VI. Population trends in MZs V and VII were increasing, but the trends were not statistically significant (Connelly

et al

. 2004, p. 6-71; Stiver

et al

. 2006, p. 1-7). WAFWA (2008) and Garton

et al

. (in press) population trend analyses did consider MZs. The WAFWA (2008, pp. 13-27) and Garton

et al

. (in press, pp. 22-62) reported that MZs I through VI had negative population trends from 1965 to 2007. All population trend analyses had similar results, with the exception of MZ VII (Table 5). However, this MZ has one of the highest proportions of inactive leks (Garton

et al

. in press, p. 65), which may imply that male numbers on the remaining leks are increasing as birds relocate. The analysis of this MZ also suffered from small sample sizes and therefore large confidence intervals (Garton

et al

. in press, p. 217), so the trend may not actually reflect the population status.

Table 5—Long-term population trend estimates for greater sage-grouse Management Zones.

MZ

States and

Provinces

Included

Population Trend Estimates 1965-2003* (Connelly

et al

. 2004)

Population Trend Estimates Based on Annual Rates of Change (%) 1965-2007(WAFWA 2008)

Population Trend Estimates Based on Annual Rates of Change (%) 1965-2007 (Garton

et al

. in press)

I

MT, WY, ND, SD, SK, AL

Long-term decline

-2.9

-2.9

II

ID, WY, UT, CO

Long-term decline

-2.7

-3.5

III

UT, NV, CA

Long-term decline

-2.2

-10**

IV

ID, UT, NV, OR

Long-term decline

-3.8

-4**

V

OR, CA, NV

Change statistically undetectable

-3.3

-2**

VI

WA

Long-term decline

-5.1

-6.5

VII

CO, UT

Change statistically undetectable

No detectable trend

+34**

*Average annual rate of change was not reported.

**Due to sample inadequacies for the statistical analyses used, only data from 1995 to 2007 could be used.

Differences in the MZ trends observed between the three analyses are minimal, with the exception of MZs III, V, and VII. While the results of Connelly

et al

. (2004) and WAFWA (2008) were similar for MZ III, Garton

et al

. (in press) showed a larger rate of decline. This difference may be due to the shortened time period (12 versus 42 years) Garton

et al

. (in press) used for the analyses because some earlier data were not suitable for the statistical procedures used. This increased rate of decline was not observed for MZ IV where Garton

et al

.'s (in press) analyses also spanned only 12 years, suggesting that declines in MZ III may have recently accelerated. No explanation was offered by WAFWA (2008) about the difference between their analyses and Connelly

et al

. (2004) for MZ V. However, Garton

et al

. (in press) results are similar to WAFWA for the same area.

The difference in the annual rate of change between Connelly

et al

. (2004) and WAFWA (2008) as compared to Garton

et al

. (in press) for MZ VII is substantial (Table 5). Garton

et al

. (in press) did not offer an explanation of this difference, but Connelly

et al

.

(2004; as cited by (Stiver

et al

. 2006, p. 1-7)) indicated population trends were increasing in this MZ, although those increases were not statistically significant. However, Garton

et al

. (in press, pp. 62-63) reported that the number of leks in MZ VII declined by 39 percent during the same analysis period. The increase in annual rate of change may simply reflect increases on remaining leks as habitat became more limited.

In addition to calculating annual rates of change by MZ, Garton

et al

(in press) also reported the percent change in number of males per lek from 1965 to 2007, the percent change of active leks from 1965 to 2007, and minimum male population estimates in 2007 (Table 6). The percent change in number of males per lek and the percent change in active leks reflect population declines, and possibly habitat loss in all MZs.

Table 6—Minimum male greater sage-grouse population estimates in 2007, percent change in number of males per lek and percent change in number of active leks between 1965 and 2007 by Management Zone (from Garton

et al.

in press, pp. 22-64).

MZ

Min Population Est in 2007

(# of males)

Percent Change in

# of Males per Lek (1965-2007)

Percent Change of Active Leks

(1965-2007)

I

14,814

-17

-22

II

42,429

-30

-7

III

6,851

-24

-16 ***

IV

15,761

-54

-11***

V

6,925

-17**

-21**

VI

315

-76

-57

VII

241

-13

-39*

*1995 to 2007 — due to sample sizes, only data from this time period were used.

**1985 to 2007 — due to sample sizes, only data from this time period were used.

***1975 to 2007 — due to sample sizes, only data from this time period were used.

In summary, since neither presettlement nor current numbers of sage-grouse are accurately known, the actual rate and magnitude of decline since presettlement times is uncertain. However, three groups of researchers using different statistical methods (but the same lek count data) concluded that rangewide greater sage-grouse have experienced long-term population declines in the past 43 years, with that decline lessening in the past 22 years. Many of these declines are the result of loss of leks (WAFWA 2008, p. 51), indicating either a direct loss of habitat or habitat function (Connelly and Braun 1997, p. 2). A recent increase in the annual rate of change for MZ VII may simply be an anomaly of small population numbers, as other indicators suggest this area is suffering habitat losses. A delayed response of sage-grouse to changes in carrying capacity was identified by Garton

et al

. (in press, p.71).

Connectivity

Greater sage-grouse are a landscape-scale species, requiring large expanses of sagebrush to meet all seasonal habitat requirements. The loss of habitat from fragmentation and conversion decreases the connectivity between seasonal habitats potentially resulting in the loss of the population (Doherty

et al

. 2008, p. 194). Loss of connectivity also can increase population isolation (Knick and Hanser in press, p. 4, and references therein) and, therefore, the probability of loss of genetic diversity and extirpation from stochastic events.

Analyses of connectivity of greater sage-grouse across the sagebrush landscape were conducted by Knick and Hanser (in press, entire). Knick and Hanser (in press, p. 29) found that the average movement between population centers (leks) of sage-grouse rangewide was 16.6 km (10.3 mi), with a standard deviation of 7.3 km (4.5 mi). Leks within 18 km (11.2 mi) of each other had common features when compared to leks further than this distance (Knick and Hanser in press, p. 17). Therefore, they used a distance of 18 km (11.2 mi) between leks to assess connectivity (movement between populations), but cautioned that this distance may not accurately reflect genetic flow, or lack thereof, between populations (Knick and Hanser in press, p. 28). Genetic evidence suggests that exchange of individual birds has not been restricted, although there is a gradation of allelic frequencies across the species' range (Oyler-McCance and Quinn, in press, p. 14). This result suggests that widespread movements (e.g., across several States) are not occurring.

Population linkages primarily occurred within MZs, and connectivity between MZs was limited, with the exception of MZs I (Great Plains) and II (Wyoming Basin). Within MZs, the Wyoming Basin (MZ II) had the highest levels of connectivity, followed by MZ IV (Snake River Plain) and MZ I (Great Plains) (Knick and Hanser in press, p. 18). The MZ VI (Columbia Basin) and VII (Colorado Plateau) had the least internal connectivity, suggesting there was limited dispersal between leks and an existing relatively high degree of isolation (Knick and Hanser in press, p. 18). Areas along the edges of the sage-grouse range (e.g., Columbia Basin, Bi-State area) are currently isolated from other sage-grouse populations (Knick and Hanser in press, p. 28).

Connectivity between sage-grouse MZs and the populations within them declined across all three analysis periods examined: 1965-1974, 1980-1989, and 1998-2007. The decline in connectivity was due to the loss of leks and reduced population size (Knick and Hanser in press, p. 29). Historic leks with low connectivity also were lost (Knick and Hanser in press, p. 20), suggesting that current isolation of leks by distance (including habitat fragmentation) will likely result in their future loss (Knick and Hanser in press, p. 28). Small decreases in lek connectivity resulted in large increases in probability of lek abandonment (Knick and Hanser, in press, p. 29). Therefore, maintaining habitat connectivity and sage-grouse population numbers are essential for sage-grouse persistence.

Sagebrush distribution was the most important factor in maintaining connectivity (Knick and Hanser in press, p. 32). This result suggests that any activities that remove or fragment sagebrush habitats will contribute to loss of connectivity and population isolation. This conclusion is consistent with research from both Aldridge

et al

. (2008, p. 988) and Wisdom

et al

. (in press, p. 13), which independently identified the proximity of sagebrush patches and area in sagebrush cover as the best predictors for sage-grouse presence.

Summary of Information Pertaining to the Five Factors

Section 4 of the Act (16 U.S.C. 1533) and implementing regulations (50 CFR part 424) set forth procedures for adding species to the Federal Lists of Endangered and Threatened Wildlife and Plants. In making this finding, we summarize below information regarding the status and threats to the greater sage-grouse in relation to the five factors provided in section 4(a)(1) of the Act. Under section (4) of the Act, we may determine a species to be endangered or threatened on the basis of any of the following five factors: (A) Present or threatened destruction, modification, or curtailment of habitat or range; (B) overutilization for commercial, recreational, scientific, or educational purposes; (C) disease or predation; (D) inadequacy of existing regulatory mechanisms; or (E) other natural or manmade factors affecting its continued existence. Our evaluation of threats is based on information provided in the petition, available in our files, and other sources considered to be the best scientific and commercial information available, including published and unpublished studies and reports.

Differences in ecological conditions within each MZ affect the susceptibility of these areas to the various threats facing sagebrush ecosystems and its potential for restoration. For example,

Centaurea diffusa

(diffuse knapweed), an exotic annual weed, is most competitive within shrub-grassland communities where antelope bitterbrush is dominant (MZ VI), and

Bromus tectorum

(cheatgrass) is more dominant in areas with minimal summer precipitation (MZs III and V) (Miller

et al

., in press, pp. 20-21). Therefore, we stratify our analyses by these MZs because they represent zones within which ecological variation is less than what it would be across the range of the species. This approach allows us to better assess the impact and benefits of actions occurring across the species' range and in turn more accurately assess the status of the species.

Factor A. The Present or Threatened Destruction, Modification, or Curtailment of Habitat or Range

Several factors are contributing to the destruction, modification, or curtailment of the greater sage-grouse's habitat or range. Several recent studies have demonstrated that sagebrush area is one of the best landscape predictors of greater sage-grouse persistence (Aldridge

et al

. 2008, p. 987; Doherty

et al

. 2008, p. 191; Wisdom

et al

., in press, p. 17). Sagebrush habitats are becoming increasingly degraded and fragmented due to the impacts of multiple threats, including direct conversion, urbanization, infrastructure such as roads and powerlines built in support of several activities, wildfire and the change in wildfire frequency, incursion of invasive plants, grazing, and nonrenewable and renewable energy development. Many of these threat factors are exacerbated by the effects of climate change, which may influence long-term habitat trends.

Habitat Conversion for Agriculture

Sagebrush is estimated to have covered roughly 120 million ha (296 million ac; Schroeder

et al

. 2004, p. 365) in western North America, but large portions of that area have been cultivated for the production of agricultural crops (e.g., potatoes, wheat; Schroeder

et al

. 1999, p. 16; 2000, p. 11). Western rangelands were converted to agricultural lands on a large scale beginning with the series of Homestead Acts in the 1800s (Braun 1998, p. 142, Hays

et al

. 1998, p. 26; Knick in press, p. 4; Knick

et al

. in press, p. 11), especially where suitable deep soil terrain and water were available (Rogers 1964, p.13, Schroeder and Vander Haegen, 2009, in press, p. 3). Connelly

et al

. (2004, p. 5-55) estimated that 24.9 million ha (61.5 million ac) within the sage-grouse conservation area (SGCA) used for their assessment area (historic range of Gunnison and greater sage-grouse plus a 50-km (31-mi) buffer) for sage-grouse is now comprised of agricultural lands, although some areas within the species' range are not sagebrush habitat, and the SGCA is larger than the sage-grouse current distribution. An estimated 10 percent of sagebrush steppe that existed prior to EuroAmerican settlement has been converted to agriculture (Knick

et al

. in press, p. 13). The remaining 90 percent is largely unsuited for agriculture because irrigation is not considered to be feasible, topography and soils are limiting, or temperatures are too extreme for many crops (West 1996 cited in Knick

et al

. in press, p. 13).

Habitat conversion results in loss of habitat available for sage-grouse use. The actual effect of this loss depends on the amount of sagebrush lost, the type of seasonal habitat affected, and the arrangement of habitat lost (large blocks or small patches) (Knick

et al

. in press, p. 15). Direct impacts to sage-grouse depend on the timing of conversion (e.g., loss of nests, eggs). Indirect effects of agricultural activities adjoining sagebrush habitats include increased predation with a resulting reduced sage-grouse nest success (Connelly

et al

. 2004, p. 7-23), increased human presence, and habitat fragmentation.

To estimate the area possibly influenced by these indirect effects, Knick

et al

. (in press, p. 13) applied a “high effective buffer” out to 6.9 km (4.3 mi) from agricultural lands, based on foraging distances of synathropic (ecologically associated with humans) predators (e.g. red foxes (

Vulpes vulpes

) and ravens (

Corvus corax

)). Given the distribution of agricultural activities across the sagebrush range, nearly three quarters of all sagebrush within range of sage-grouse has been influenced by agricultural activities (falls within the high effective buffer) (Knick

et al

. in press, p. 13). This influence includes foraging distances for synathropic predators (Leu

et al

. 2008, p. 1120; Knick

et al

. in press, p. 13), and associated features such as irrigation ditches. Extensive conversion of sagebrush to agriculture within a landscape has decreased abundance of sage-grouse in many portions of their range (Knick and Hanser in press, p. 30, and references therein).

Soil associations have resulted in disproportionate levels of habitat conversion across different sagebrush communities. For example,

Artemisia tridentata

ssp.

vaseyana

is found at lower elevations, in soils that retain moisture 2 to 4 weeks longer than in well-drained, but dry and higher elevation soils typical of

A. t.

ssp.

wyomingensis

locations. Therefore, sagebrush communities dominated by basin big sagebrush (

A. t.

ssp

. tridentata)

have been converted to agriculture more extensively than have communities on poorer soil sites (Winward 2004, p. 29) (also see discussion below).

Large losses of sagebrush shrub-steppe habitats due to agricultural conversion have occurred in some areas within the range of the greater sage-grouse. This loss has been especially apparent in the Columbia Basin of the Northwest (MZ VI), the Snake River Plain of Idaho (MZ IV) (Schroeder

et al

. 2004, p. 370), and the Great Plains (MZ

I) (Knick

et al

. in press, p. 13). Hironaka

et al

. (1983, p. 27) estimated that 99 percent of basin big sagebrush habitat in the Snake River Plain has been converted to cropland. Between 1975 and 1992 alone, 29,762 ha (73,543 ac) of sagebrush habitat were converted to cropland on the Upper Snake River Plain, a 74-percent increase in cropland (Leonard

et al

. 2000, p. 268). The loss of this primarily winter sage-grouse habitat is significantly related to subsequent sage-grouse declines (Leonard

et al

. 2000, p. 268).

Prior to EuroAmerican settlement in the 19th century, Washington had an estimated 42 million ha (103.8 million ac) of shrub-steppe (Connelly

et al

. 2004, p. 7-22). Approximately 60 percent of the original shrub-steppe habitat in Washington has been converted to primarily agricultural uses (Dobler 1994, p. 2). Deep soils supporting shrub-steppe communities in Washington within sage-grouse range continue to be converted to agricultural uses (Vander Haegen

et al

. 2000, p. 1156), resulting in habitat loss. Agriculture is the dominant land cover within sagebrush areas of Washington (42 percent) and Idaho (19 percent) (Miller

et al

., in press, p. 18). In north-central Oregon (MZ V), approximately 2.6 million ha (6.4 million ac) of habitat were converted for agricultural purposes, essentially eliminating sage-grouse from this area (Willis

et al

. 1993, p. 35). More broadly, across the interior Columbia Basin of southern Idaho, northern Utah, northern Nevada, eastern Oregon (MZ IV), and Washington, approximately 6 million ha (14.8 million ac) of shrub-steppe habitat has been converted to agricultural crops (Altman and Holmes 2000, p. 10).

Braun concluded that development of irrigation projects to support agricultural production in areas where soils were sufficient to support agriculture, in some cases conjointly with hydroelectric dam construction, has resulted in additional sage-grouse habitat loss (Braun 1998, p. 142). The reservoirs formed by these projects impacted native shrub-steppe habitat adjacent to the rivers in addition to supporting the irrigation and direct conversion of shrub-steppe lands to agriculture. The projects precipitated conversion of large expanses of upland shrub-steppe habitat in the Columbia Basin for irrigated agriculture (65 FR 51578). The creation of these reservoirs also inundated hundreds of kilometers of riparian habitats used by sage-grouse broods (Braun 1998, p. 144). However, other small and isolated reclamation projects (4,000 to 8,000 ha (10,000 to 20,000 ac)) were responsible for three-fold localized increases in sage-grouse populations (Patterson 1952, pp. 266-274) by providing water in a semiarid environment, which provided additional insect and forb food resources (e.g., Eden Reclamation Project in Wyoming). Benefits of providing water through agricultural activities may now be negated due to the threat of West Nile virus (WNv) (Walker

et al

. 2004, p. 4).

Five percent of the areas occupied by Great Basin sagebrush have been converted to agriculture, urban or industrial areas (MZs III and IV) (Miller

et al

. in press, p. 18). Five percent has also been converted in the wheatgrass-needlegrass-shrubsteppe (MZ II, primarily in north-central Wyoming) (Miller

et al

., in press, p. 18). In sagebrush-steppe habitats, 14 percent of sagebrush habitats had been converted to agriculture, urban or industrial activities (MZs II, IV, V, and VI) (Miller

et al

., in press, pp. 17-18). Nineteen percent of the Great Plains area (MZ I) has been converted to agriculture (Knick

et al

. in press, p. 13). Conversions for sagebrush habitat types by State are detailed in Table 7.

Table 7—Current sagebrush-steppe habitat and agricultural lands within Great Basin sagebrush (as derived from LANDFIRE 2006 vegetation coverage) (from Miller

et al.

in press, pp. 17-18).

State

Percent Sagebrush

Percent

Agriculture

Washington

23.7

42.4

Montana

56.2*

7.5*

Wyoming

66.0*

3.4*

Idaho

55.0

18.6

Oregon

64.5

8.6

Nevada

58.7

1.3

Utah

37.6

9.7

California

49.8

8.0

Colorado

40.6*

11.8*

TOTAL

55.4

10.0

*Analyses did not include sagebrush lands in the eastern portions of Colorado, Montana, and Wyoming.

Aldridge

et al

. (2008, pp. 990-991) reported that sage-grouse extirpations were more likely to occur in areas where cultivated crops exceeded 25 percent. Their results supported the conclusions of others (e.g., Schroeder 1997, p. 934; Braun 1998, p. 142; Aldridge and Brigham 2003, p. 30) that extensive cultivation and fragmentation of native habitats have been associated with sage-grouse population declines. Wisdom

et al

. (in press, p. 4) identified environmental factors associated with the regional extirpation of sage-grouse. Areas still occupied by sage-grouse have three times less area in agriculture and a mean human density 26 times lower than extirpated areas (Wisdom

et al

., in press, p. 13). While sage-grouse may forage on agricultural crops (see discussion below), they avoid landscapes dominated by agriculture (Aldridge

et al

. 2008, p. 991). Conversions to croplands in southern Idaho have resulted in isolation of sagebrush-dominated landscapes into less productive regions north and south

of the Snake River Plain (Knick

et al

. 2003, p. 618). Therefore, formerly continuous populations in this area are now disconnected (Knick and Hanser in press, p. 52).

Sagebrush habitat continues to be converted for both dryland and irrigated crop production (Montana Farm Services Agency (FSA) in litt, 2009; Braun 1998, p. 142; 65 FR 51578, August 24, 2000). The increasing value of wheat and corn crops has driven new conversions in recent years. For example, the acres of sagebrush converted to tilled agriculture in Montana increased annually from 2005 to 2009, with approximately 10,259 ha (25,351 ac) converted, primarily in the eastern two-thirds of the State (MZ I) (Montana FSA in litt, 2009). In addition, in 2008, a single conversion in central Montana totaled between 3,345 and 10,000 ha (10,000 and 30,000 ac) (MZ I) (Hanebury 2008a, pers. comm.). Other large conversions occurred in the same part of Montana in 2008, although these were unquantified (Hanebury 2008b, pers. comm.). We were unable to gather any further information on crop conversions of sagebrush habitats as there are no systematic efforts to collect State or local data on conversion rates in the majority of the greater sage-grouse range (GAO 2007, p. 16).

In addition to crop conversion for traditional crops, recent interest in the development of crops for use as biofuels could potentially impact sage-grouse. For example, the 2008 Farm Bill authorized the Biomass Crop Assistance Program (BCAP), which provides financial incentives to agricultural producers that establish and produce eligible crops for conversion to bioenergy products (U.S. Department of Agriculture (USDA) 2009b, p. 1). Further loss of sagebrush habitats due to BCAP will negatively impact sage-grouse populations. However, currently we have no way of predicting the magnitude of BCAP impacts to sage-grouse (see discussion under Factor D, below).

Although conversion of shrub-steppe habitat to agricultural crops impacts sage-grouse through the loss of sagebrush on a broad scale, some studies report the use of agricultural crops (e.g., alfalfa) by sage-grouse. When alfalfa fields and other croplands are adjacent to extant sagebrush habitat, sage-grouse have been observed feeding in these fields, especially during brood-rearing (Patterson 1952, p. 203; Rogers 1964, p. 53; Wallestad 1971, p. 134; Connelly

et al

. 1988, p.120; Fischer

et al

. 1997, p. 89). Connelly

et al

. (1988, p. 120) reported seasonal movements of sage-grouse to agricultural crops as sagebrush habitats desiccated during the summer. However, use of irrigated crops may not be beneficial to greater sage-grouse if it increases exposure to pesticides (Knick

et al

. in press, p. 16) and WNv (Walker

et al

. 2004, p. 4).

Some conversion of cropland to sagebrush has occurred in former sage-grouse habitats through the USDA's voluntary Conservation Reserve Program (CRP) which pays landowners a rental fee to plant permanent vegetation on portions of their lands, taking them out of agricultural production. In Washington State (Columbia Basin, MZ VI), sage-grouse have declined precipitously in the Columbia Basin largely due to conversion of sagebrush habitats to cropland (Schroeder and Vander Haegen, in press, p. 4). Approximately 599,314 ha (1,480,937 ac) of converted farmland had been enrolled in the CRP, almost all of which was historically shrub-steppe (Schroeder and Vander Haegen in press, p. 5). Schroeder and Vander Haegen (in press, p. 20) found that CRP lands that have been out of production long enough to allow re-establishment of sagebrush and was juxtaposed to a relatively intact shrub-steppe landscape was most beneficial to sage-grouse. There appears to be some correlation with sage-grouse use of CRP and a slight increase in population size in north-central Washington (Schroeder and Vander Haegen in press, p. 21). Schroeder and Vander Haegen (in press, p. 21) concluded that the loss of CRP due to expiration of the program or incentives to produce biofuels would likely severely impact populations in the Columbia Basin.

Although estimates of the numbers of acres enrolled rangewide in CRP (and the number of acres soon to expire from CRP) are available, the extent of cropland conversion to habitats beneficial to sage-grouse (i.e., CRP lands planted with native grasses, forbs, and shrubs) is not known for any other area barring the Columbia Basin. Thus, outside this area, we cannot judge the overall impact of CRP land to sage-grouse persistence.

Direct habitat loss and conversion also occurs via numerous other landscape uses, including urbanization, livestock forage production, road building, and oil pads. These activities are described in greater detail below. Although we were unable to obtain an estimate of the total amount of sagebrush habitats that have been lost due to these activities, they have resulted in habitat fragmentation, as well as habitat loss.

Urbanization

Low densities of indigenous peoples have been present for more than 12,000 years in the historical range of sage-grouse. By 1900, less than 1 person per km

2

(1 person per 0.4 mi

2

) resided in 51 percent of the 325 counties within the SGCA, and densities greater than 10 persons per km

2

(10 persons per 0.4 mi

2

) occurred in 4 percent of the counties (Connelly

et al

. 2004, p. 7-24). By 2000, counties with less than 1 person per km

2

(1 person per 0.4 mi

2

) occurred in 31 percent of the 325 counties and densities greater than 10 persons per km

2

(10 persons per 0.4 mi

2

) occurred in 22 percent of the counties (Connelly

et al

. 2004, p. 7-25). Today, the Columbia Basin (MZ VI) has the highest density of humans while the Great Plains (MZ I) and Wyoming Basin (MZ II) have the lowest (Knick

et al

. in press, p. 19). Growth in the Great Plains (MZ I) continues to be slower than other areas. For example, population densities have increased since 1990 by 7 percent in the Great Plains (MZ I), by 19 percent in the Wyoming Basin (MZ II), and by 31 percent in the Colorado Plateau (MZ VII) (Knick

et al

. in press, p. 19).

The dominant urban areas in the sage-grouse range are located in the Bear River Valley of Utah, the portion of Bonneville Basin southeast of the Great Salt Lake, the Snake River Valley of southern Idaho, and the Columbia River Valley of Washington (Rand McNally Road Atlas 2003; Connelly

et al

. 2004, p. 7-25). Overall, approximately 1 percent of the amount of potential sagebrush (estimated historic range) is now covered by lands classified as urban (Miller

et al

., in press, p. 18).

Knick

et al

(in press, p. 107) examined the influence of urbanization on greater sage-grouse MZs by adding a 6.9-km (4.3-mi) buffer (an estimate of the foraging distances of mammalian and corvid predators of sage-grouse) to the total area of urban land use. Based the estimates using this approach, the Columbia Basin (MZ VI) was influenced the most by urbanization with 48.4 percent of the sagebrush area affected. The Northern Great Basin (MZ V) was influenced least with 12.5 percent affected. Wyoming Basin (MZ II), which has the majority of sage-grouse in the range, was at 18.4 percent affected.

Since 1950, the western U.S. population growth rate has exceeded the national average (Leu and Hanser in press, p. 4). This growth has led to increases in urban, suburban, and rural development. Rural development has increased especially rapidly in recent decades. For example, the amount of uninhabited area in the Great Basin

ecoregion has decreased from 90,000 km

2

(34,749 mi

2

) in 1990 to less than 12,000 km

2

(4,633 mi

2

) in 2004 (Knick

et al

. in press, p. 20). Urbanization has directly eliminated some sage-grouse habitat (Braun 1998, p. 145). Interrelated effects from urbanization include construction of associated infrastructure (e.g., roads, powerlines, and pipelines) and predation threats from the introduction of domestic pets and increases in predators subsidized by human activities. In particular, municipal solid waste landfills (landfills) and roads have been shown to contribute to increases in common raven (

Corvus corax

) populations (Knight

et al

. 1993 p. 470; Restani

et al

. 2001, p. 403; Webb

et al

. 2004, p. 523). Ravens are known to be an important predator on sage-grouse nests and have been considered a restraint on sage-grouse population growth in some locations (Batterson and Morse 1948, p. 14; Autenrieth 1981, p. 45; Coates 2007, p. 26). Landfills (and roads) are found in every State within the greater sage-grouse range and a number of these are located within or adjacent to sage-grouse habitat.

Recent changes in demographic and economic trends have resulted in greater than 60 percent of the Rocky Mountain West's counties experiencing rural sprawl where rural areas are outpacing urban areas in growth (Theobald 2003, p. 3). In some Colorado counties, up to 50 percent of sage-grouse habitat is under rural subdivision development, and an estimated 3 to 5 percent of all sage-grouse historical habitat in Colorado has already been converted into urban areas (Braun 1998, p. 145). We are unaware of similar estimates for other States within the range of the greater sage-grouse and, therefore, cannot determine the effects of this factor on a rangewide basis. Rural development has increasingly taken the form of low-density (approximately 6 to 25 homes per km

2

(6 to 25 homes per 0.4 mi

2

)) home development or exurban growth (Hansen

et al

. 2005, p. 1894). Between 1990 and 2000, 120,000 km

2

(46,332 mi

2

) of land were developed at exurban densities nationally (Theobald 2001, p. 553). However, this value includes development nationwide, and we are unable to report values specifically for sagebrush habitats. However, within the Great Basin (including California, Idaho, Nevada, and Utah), human populations have increased 69 percent and uninhabited areas declined by 86 percent between 1990 and 2004 (Leu and Hanser in press, p. 19). Similar to higher density urbanization, exurban development has the potential to negatively affect sage-grouse populations through fragmentation or other indirect habitat loss, increased infrastructure, and increased predation.

In modeling sage-grouse persistence, Aldridge

et al

. (2008, pp. 991-992) found that the density of humans in 1950 was the best predictor of sage-grouse extirpation among the human population metrics considered (including increasing human population growth). Sage-grouse extirpation was more likely in areas having a moderate human population density of at least 4 people per km

2

(4 people per 0.4 mi

2

). Increasing human populations were not a good predictor of sage-grouse persistence, most likely because much of the growth occurred in areas that are already no longer suitable for sage-grouse. Aldridge

et al

. (2008, p. 990) also reported that, based on their models, sage-grouse require a minimum of 25 percent sagebrush for persistence in an area. A high probability of persistence required 65 percent sagebrush or more. This result is similar to the results by Wisdom

et al

. (in press, p. 18) who reported that human density was 26 times greater in extirpated sage-grouse areas than in currently occupied range. Therefore, human population growth that results in exurban development in sagebrush habitats will reduce the likelihood of sage-grouse persistence in the area. Given the current demographic and economic trends in the Rocky Mountain West, we believe that rates of urbanization will continue increasing, resulting in further habitat fragmentation and degradation and decreasing the probability of long-term sage-grouse persistence.

Infrastructure in Sagebrush Habitats

Habitat fragmentation is the separation or splitting apart of previously contiguous, functional habitat components of a species. Fragmentation can result from direct habitat losses that leave the remaining habitat in noncontiguous patches, or from alteration of habitat areas that render the altered patches unusable to a species (i.e., functional habitat loss). Functional habitat losses include disturbances that change a habitat's successional state or remove one or more habitat functions; physical barriers that preclude use of otherwise suitable areas; and activities that prevent animals from using suitable habitat patches due to behavioral avoidance.

Sagebrush communities exhibit a high degree of variation in their resistance and resilience to change, beyond natural variation. Resistance (the ability to withstand disturbing forces without changing) and resilience (the ability to recover once altered) generally increase with increasing moisture and decreasing temperatures, and also can be linked to soil characteristics (Connelly

et al

. 2004, p. 13-6). However, most extant sagebrush habitat has been altered since European immigrant settlement of the West (Baker

et al

. 1976, p. 168; Braun 1998, p. 140; Knick

et al

. 2003, p. 612; Connelly

et al

. 2004, p. 13-6), and sagebrush habitat continues to be fragmented and lost (Knick

et al

. 2003, p. 614) through the factors described below. The cumulative effects of habitat fragmentation have not been quantified over the range of sagebrush and most fragmentation cannot be attributed to specific land uses (Knick

et al

. 2003, p. 616). However, in large-scale analysis of the collective effect of anthropogenic features (or the “human footprint”) in the western United States, Leu

et al

. (2008, p. 1130) found that 13 percent of the area was affected in some way by anthropogenic features (i.e., fragmentation). Areas with the lowest “human footprint” (i.e., no to slight development or use) experienced above-average human population growth between 1990 and 2000. There is significant evidence these areas will experience increasing habitat fragmentation in the future (Leu

et al

. 2008, p. 1133). Although the area covered by these estimates includes all western states, we believe the general points regarding effects of anthropogenic features apply to sage-grouse habitat.

Fragmentation of sagebrush habitats has been cited as a primary cause of the decline of sage-grouse populations because the species requires large expanses of contiguous sagebrush (Patterson 1952, pp. 192-193; Connelly and Braun 1997, p. 4; Braun 1998, p. 140; Johnson and Braun 1999, p. 78; Connelly

et al

. 2000a, p. 975; Miller and Eddleman 2000, p. 1; Schroeder and Baydack 2001, p. 29; Johnsgard 2002, p. 108; Aldridge and Brigham 2003, p. 25; Beck

et al

. 2003, p. 203; Pedersen

et al

. 2003, pp. 23-24; Connelly

et al

. 2004, p. 4-15; Schroeder

et al

. 2004, p. 368; Leu

et al

. in press, p. 19). The negative effects of habitat fragmentation have been well documented in numerous bird species, including some shrub-steppe obligates (Knick and Rotenberry 1995, pp. 1068-1069). However, prior to 2005, detailed data to assess how fragmentation influences specific greater sage-grouse life-history parameters such as productivity, density, and home range were not available. More recently, several studies have documented negative effects of fragmentation as a

result of oil and gas development and its associated infrastructure (see discussion of Energy Development below) on lek persistence, lek attendance, winter habitat use, recruitment, yearling annual survival rate, and female nest site choice (Holloran 2005, p. 49; Aldridge and Boyce 2007, pp. 517-523; Walker

et al

. 2007a, pp. 2651-2652; Doherty

et al

. 2008, p. 194). Wisdom

et al

. (in press, p. 18) reported that a variety of human developments, including roads, energy development, and other factors that contribute to habitat fragmentation have contributed to or been associated with sage-grouse extirpation. Estimating the impact of habitat fragmentation on sage-grouse is complicated by time lags in response to habitat changes (Garton

et al

., in press, p. 71), particularly since these long-lived birds will continue to return to altered breeding areas (leks, nesting areas, and early brood-rearing areas) due to strong site fidelity despite nesting or productivity failures (Wiens and Rotenberry 1985, p. 666).

Powerlines

Power grids were first constructed in the United States in the late 1800s. The public demand for electricity has grown as human population and industrial activities have expanded (Manville 2002, p. 5), resulting in more than 804,500 km (500,000 mi) of transmission lines (lines carrying greater than 115,000 volts (115 kilovolts (kV)) by 2002 within the United States (Manville 2002, p. 4). A similar estimate is not available for distribution lines (lines carrying less than 69,000volts (69kV)), and we are not aware of data for Canada. Within the SGCA, Knick

et al

. (

in press

, p. 21) showed that powerlines cover a minimum of 1,089km

2

(420.5 mi).

Due to the potential spread of invasive species and predators as a result of powerline construction the impact from the powerline is greater than the actual footprint. Knick

et al

. (in press, p. 111) estimated these impacts may influence up to 39 percent of all sagebrush in the SGCA. Powerlines can directly affect greater sage-grouse by posing a collision and electrocution hazard (Braun 1998, pp. 145-146; Connelly

et al

. 2000a, p. 974), and can have indirect effects by decreasing lek recruitment (Braun

et al

. 2002, p. 10), increasing predation (Connelly

et al

. 2004, p. 13-12), fragmenting habitat (Braun 1998, p. 146), and facilitating the invasion of exotic annual plants (Knick

et al

. 2003, p. 612; Connelly

et al

. 2004, p. 7-25). In 1939, three adult sage-grouse died as a result of colliding with a telegraph line in Utah (Borell 1939, p. 85). Both Braun (1998, p. 145) and Connelly

et al

. (2000a, p. 974) report that sage-grouse collisions with powerlines occur, although no specific instances were presented. There was also an unpublished observation reported by Aldridge and Brigham (2003, p. 31). In 2009, two sage-grouse died from electrocution after colliding with a powerline in the Mono Basin of California (Gardner 2009, pers. comm.). We were unable to find any other documentation of other collisions or electrocution of sage-grouse resulting from powerlines.

In areas where the vegetation is low and the terrain relatively flat, power poles provide an attractive hunting and roosting perch, as well as nesting stratum for many species of raptors and corvids (Steenhof

et al

. 1993, p. 27; Connelly

et al

. 2000a, p. 974; Manville 2002, p. 7; Vander Haegen

et al

. 2002, p. 503). Power poles increase a raptor's range of vision, allow for greater speed during attacks on prey, and serve as territorial markers (Steenhof

et al

. 1993, p. 275; Manville 2002, p. 7). Raptors may actively seek out power poles where natural perches are limited. For example, within 1 year of construction of a 596-km (372.5-mi) transmission line in southern Idaho and Oregon, raptors and common ravens began nesting on the supporting poles (Steenhof

et al

. 1993, p. 275). Within 10 years of construction, 133 pairs of raptors and ravens were nesting along this stretch (Steenhof

et al

. 1993, p. 275). Raven counts have increased by approximately 200 percent along the Falcon-Gondor transmission line corridor in Nevada within 5 years of construction (Atamian

et al

. 2007, p. 2). The increased abundance of raptors and corvids within occupied sage-grouse habitats can result in increased predation. Ellis (1985, p. 10) reported that golden eagle (

Aquila chryrsaetos

) predation on sage-grouse on leks increased from 26 to 73 percent of the total predation after completion of a transmission line within 200 meters (m) (220 yards (yd)) of an active sage-grouse lek in northeastern Utah. The lek was eventually abandoned, and Ellis (1985, p. 10) concluded that the presence of the powerline resulted in changes in sage-grouse dispersal patterns and caused fragmentation of the habitat.

Leks within 0.4 km (0.25 mi) of new powerlines constructed for coalbed methane development in the Powder River Basin of Wyoming had significantly lower growth rates, as measured by recruitment of new males onto the lek, compared to leks further from these lines, which were presumed to be the result of increased raptor predation (Braun

et al

. 2002, p. 10). Within the SGCA, Connelly

et al

. (2004, p. 7-26) estimated that the area potentially influenced by additional perches for corvids and raptors provided by powerlines, assuming a 5- to 6.9-km (3.1- to 4.3-mi) radius buffer around the perches based on the average foraging distance of these predators, was 672,644 to 837,390 km

2

(259,641 to 323,317 mi

2

), or 32 to 40 percent of the SGCA. The actual impact on the area would depend on corvid and raptor densities within the area, the amount of cover to reduce predation risk at sage-grouse nests, and other factors (see discussion in Factor C, below).

The presence of a powerline may fragment sage-grouse habitats even if raptors are not present. Braun (1998, p. 146) found that use of otherwise suitable habitat by sage-grouse near powerlines increased as distance from the powerline increased for up to 600 m (660 yd) and, based on that unpublished data, reported that the presence of powerlines may limit sage-grouse use within 1 km (0.6 mi) in otherwise suitable habitat. Similar results were recorded for other grouse species. Pruett

et al

. (2009, p. 6) found that lesser and greater prairie-chickens (

Tympanuchus pallidicinctus

and

T. cupido

, respectively) avoided otherwise suitable habitat near powerlines. Additionally, both species also crossed powerlines less often than nearby roads, which suggests that powerlines are a particularly strong barrier to movement (Pruett

et al

. 2009, p. 6).

Sage-grouse also may avoid powerlines as a result of the electromagnetic fields (Wisdom

et al

. in press, p. 19). Electromagnetic fields have been demonstrated to alter the behavior, physiology, endocrine systems, and immune function in birds, with negative consequences on reproduction and development (Fernie and Reynolds 2005, p. 135). Birds are diverse in their sensitivities to electromagnetic field exposures, with domestic chickens being very sensitive. Many raptor species are less affected (Fernie and Reynolds 2005, p. 135).

Linear corridors through sagebrush habitats can facilitate the spread of invasive species, such as

Bromus tectorum

(Gelbard and Belnap 2003, pp. 424-426; Knick

et al

. 2003, p. 620; Connelly

et al

. 2004, p. 1-2). However, we were unable to find any information regarding the amount of invasive species incursion as a result of powerline construction.

Powerlines are common to nearly every type of anthropogenic habitat use, except perhaps some forms of agricultural development (e.g., livestock grazing) and fire. Although we were

unable to find an estimate of all future proposed powerlines within currently occupied sage-grouse habitats, we anticipate that powerlines will continue to increase into the foreseeable future, particularly given the increasing development of energy resources and urban areas. For example, up to 8,579 km (5,311 mi) of new powerlines are predicted for the development of the Powder River Basin coal-bed methane field in northeastern Wyoming (BLM 2003) in addition to the approximately 9,656 km (6,000 mi) already constructed in that area. In November 2009, nine Federal agencies signed a Memorandum of Understanding to expedite the building of new transmission lines on Federal lands. If these lines cross sage-grouse habitats, sage-grouse will likely be negatively affected.

Communication Towers

Within sage-grouse habitats, 9,510 new communication towers have been constructed within recent years (Connelly

et al

. 2004, p. 13-7). While millions of birds are killed annually in the United States through collisions with communication towers and their associated structures (e.g., guy wires, lights) (Shire

et al

. 2000, p. 5; Manville 2002, p. 10), most documented mortalities are of migratory songbirds. We were unable to determine if any sage-grouse mortalities occur as a result of collision with communication towers or their supporting structures, as most towers are not monitored and those that are lie outside the range of the species (Kerlinger 2000, p. 2; Shire

et al

. 2000 p. 19). Cellular towers have the potential to cause sage-grouse mortality via collisions, to influence movements through avoidance of a tall structure (Wisdom

et al

. in press, p. 20), or to provide perches for corvids and raptors (Steenhof

et al

. 1993, p. 275; Connelly

et al

. 2004, p. 13-7).

In a comparison of sage-grouse locations in extirpated areas of their range (as determined by museum species and historical observations) and currently occupied habitats, the distance to cellular towers was nearly twice as far from grouse locations in currently occupied habitats than extirpated areas (Wisdom

et al

. in press, p. 13). The results may have been influenced by location as many cellular towers are close to intensive human development. However, such associations with other indicators of development and cellular towers were low (Wisdom

et al

. in press, p. 20). High levels of electromagnetic radiation within 500 m (547 yd) of all towers have been linked to decreased populations and reproductive performance of some bird and amphibian species (Wisdom

et al

. in press, p. 19, and references therein). We do not know if greater sage-grouse are negatively impacted by electromagnetic radiation, or if their avoidance of these structures is a response to increased predation risk.

Fences

Fences are used to delineate property boundaries and for livestock management (Braun 1998, p. 145; Connelly

et al

. 2000a, p. 974). The effects of fencing on sage-grouse include direct mortality through collisions, creation of predator (raptor) and corvid perch sites, the potential creation of predator corridors along fences (particularly if a road is maintained next to the fence), incursion of exotic species along the fencing corridor, and habitat fragmentation (Call and Maser 1985, p. 22; Braun 1998, p. 145; Connelly

et al

. 2000a, p. 974; Beck

et al

. 2003, p. 211; Knick

et al

. 2003, p. 612; Connelly

et al

. 2004, p. 1-2).

More than 1,000 km (625 mi) of fences were constructed annually in sagebrush habitats from 1996 through 2002, mostly in Montana, Nevada, Oregon, and Wyoming (Connelly

et al

. 2004, p. 7-34). Over 51,000 km (31,690 mi) of fences were constructed on BLM lands supporting sage-grouse populations between 1962 and 1997 (Connelly

et al

. 2000a, p. 974). Sage-grouse frequently fly low and fast across sagebrush flats, and fences can create a collision hazard (Call and Maser 1985, p. 22). Thirty-six carcasses of sage-grouse were found near Randolph, Utah, along a 3.2-km (2-mi) fence within 3 months of its construction (Call and Maser 1985, p. 22). Twenty-one incidents of mortality through fence collisions near Pinedale, Wyoming, were reported in 2003 to the BLM (Connelly

et al

. 2004, p. 13-12). A recent study in Wyoming confirmed 146 sage-grouse fence strike mortalities over a 31-month period along a 7.6-km (4.6-mi) stretch of 3-wire BLM range fence (Christiansen 2009).

Not all fences present the same mortality risk to sage-grouse. Mortality risk appears to be dependent on a combination of factors including design of fencing, landscape topography, and spatial relationship with seasonal habitats (Christiansen 2009, unpublished data). Although the effects of direct strike mortality on populations are not understood, fences are ubiquitous across the landscape. In many parts of the sage-grouse range (primarily Montana, Nevada, Oregon, Wyoming) fences exceed densities of more than 2 km/km

2

(1.2 mi/0.4 mi

2

; Knick

et al

. in press, p. 32). Fence collisions continue to be identified as a source of mortality for sage-grouse, and we expect this source of mortality to continue into the foreseeable future (Braun 1998, p. 145; Connelly

et al

. 2000a, p. 974; Oyler-McCance

et al

. 2001, p. 330; Connelly

et al

. 2004, p. 7-3).

Fence posts create perching places for raptors and corvids, which may increase their ability to prey on sage-grouse (Braun 1998, p. 145; Oyler-McCance

et al

. 2001, p. 330; Connelly

et al

. 2004, p. 13-12). We anticipate that the effect on sage-grouse populations through the creation of new raptor perches and predator corridors into sagebrush habitats is similar to that of powerlines discussed previously (Braun 1998, p. 145; Connelly

et al

. 2004, p. 7-3). Fences and their associated roads also facilitate the spread of invasive plant species that replace sagebrush plants upon which sage-grouse depend (Braun 1998, p. 145; Connelly

et al

. 2000a, p. 973; Gelbard and Belnap 2003, p. 421; Connelly

et al

. 2004, p. 7-3). Greater sage-grouse avoidance of habitat adjacent to fences, presumably to minimize the risk of predation, effectively results in habitat fragmentation even if the actual habitat is not removed (Braun 1998, p. 145).

Roads

Interstate highways and major paved roads cover approximately 2,500 km

2

(965 mi

2

) or 0.1 percent of the SGCA (Knick

et al

. in press, p. 21). Based on applying a 7-km (4.3-mi) buffer to estimate the potential impact of secondary effects from roads, interstates and highways are estimated to influence 851,044 km

2

(328,590 mi

2

) or 41 percent of the SGCA. Additionally, secondary paved roads are heavily distributed throughout most of the SGCA, existing at densities of up to greater than 5 km/km

2

(3.1 mi/mi

2

). Taken together, 95 percent of all sage-grouse habitats were within 2.5 km (1.5 mi) of a mapped road, and almost no area of sagebrush was greater the 6.9 km (4.3 mi) from a mapped road (Knick

et al

. in press, p. 21).

Impacts from roads may include direct habitat loss, direct mortality, barriers to migration corridors or seasonal habitats, facilitation of predators and spread of invasive vegetative species, and other indirect influences such as noise (Forman and Alexander 1998, pp. 207-231). Sage-grouse mortality resulting from collisions with vehicles does occur (Patterson 1952, p. 81), but mortalities are typically not monitored or recorded. Therefore, we are unable to determine the importance of this factor on sage-grouse populations. Data regarding how roads affect seasonal habitat availability

for individual sage-grouse populations by creating barriers and the ability of greater sage-grouse to reach these areas were not available. Road development within Gunnison sage-grouse (

C. minimus

) habitats impeded movement of local populations between the resultant patches, with grouse road avoidance presumably being a behavioral means to limit exposure to predation (Oyler-McCance

et al

. 2001, p. 330).

Roads can provide corridors for predators to move into previously unoccupied areas. For some mammalian species, dispersal along roads has greatly increased their distribution (Forman and Alexander 1998, p. 212; Forman 2000, p. 33). Corvids also use linear features such as primary and secondary roads as travel routes, expanding their movements into previously unused regions (Knight and Kawashima 1993, p. 268; Connelly

et al

. 2004, p. 12-3). In an analysis of anthropogenic impacts, at least 58 percent of the SGCA had a high or medium estimated presence of corvids (Connelly

et al

. 2004, p. 12-6). Corvids are important sage-grouse nest predators and in a study in Nevada were positively identified via video recorder as responsible for more than 50 percent of nest predations in the study area (Coates 2007, pp. 26-30). Bui (2009, p. 31) documented ravens following roads in oil and gas fields during foraging. Additionally, highway rest areas provide a source of food and perches for corvids and raptors, and facilitate their movements into surrounding areas (Connelly

et al

. 2004, p. 7-25).

The presence of roads increases human access and resulting disturbance effects in remote areas (Forman and Alexander 1998, p. 221; Forman 2000, p. 35; Connelly

et al

. 2004, pp. 7-6 to 7-25). Increases in legal and illegal hunting activities resulting from the use of roads built into sagebrush habitats have been documented (Hornaday 1916, p. 183; Patterson 1952, p. vi). However, the actual current effect of these increased activities on sage-grouse populations has not been determined. Roads also may facilitate access for rangeland habitat treatments, such as disking or mowing (Connelly

et al

. 2004, p. 7-25), resulting in subsequent direct habitat losses. New roads are being constructed to support development activities within the greater sage-grouse extant range. In the Powder River Basin of Wyoming, up to 28,572 km (17,754 mi) of roads to support coalbed methane development are proposed (BLM 2003).

The expansion of road networks contributes to exotic plant invasions via introduced road fill, vehicle transport, and road maintenance activities (Forman and Alexander 1998, p. 210; Forman 2000, p. 32; Gelbard and Belnap 2003, p. 426; Knick

et al

. 2003, p. 619; Connelly

et al

. 2004, p. 7-25). Invasive species are not limited to roadsides, but also encroach into surrounding habitats (Forman and Alexander 1998, p. 210; Forman 2000, p. 33; Gelbard and Belnap 2003, p. 427). In their study of roads on the Colorado Plateau of southern Utah, Gelbard and Belnap (2003, p. 426) found that improving unpaved four-wheel drive roads to paved roads resulted in increased cover of exotic plant species within the interior of adjacent plant communities. This effect was associated with road construction and maintenance activities and vehicle traffic, and not with differences in site characteristics. The incursion of exotic plants into native sagebrush systems can negatively affect greater sage-grouse through habitat losses and conversions (see further discussion in Invasive Plants, below).

Additional indirect effects of roads may result from birds' behavioral avoidance of road areas because of noise, visual disturbance, pollutants, and predators moving along a road. The absence of vegetation in arid and semiarid regions that may buffer these impacts further exacerbates the problem (Suter 1978, p. 6). Male sage-grouse lek attendance was shown to decline within 3 km (1.9 mi) of a methane well or haul road with traffic volume exceeding one vehicle per day (Holloran 2005, p. 40). Male sage-grouse depend on acoustical signals to attract females to leks (Gibson and Bradbury 1985, p. 82; Gratson 1993, p. 692). If noise interferes with mating displays, and thereby female attendance, younger males will not be drawn to the lek and eventually leks will become inactive (Amstrup and Phillips 1977, p. 26; Braun 1986, pp. 229-230).

Dust from roads and exposed roadsides can damage vegetation through interference with photosynthetic activities. The actual amount of potential damage depends on winds, wind direction, the type of surrounding vegetation and topography (Forman and Alexander 1998, p. 217). Chemicals used for road maintenance, particularly in areas with snowy or icy precipitation, can affect the composition of roadside vegetation (Forman and Alexander 1998, p. 219). We were unable to find any data relating these potential effects directly to impacts on sage-grouse population parameters.

In a study on the Pinedale Anticline in Wyoming, sage-grouse hens that bred on leks within 3 km (1.9 mi) of roads associated with oil and gas development traveled twice as far to nest as did hens bred on leks greater than 3 km (1.9 mi) from roads. Nest initiation rates for hens bred on leks close to roads also were lower (65 versus 89 percent) affecting population recruitment (33 versus 44 percent) (Lyon 2000, p. 33; Lyon and Anderson 2003, pp. 489-490). Lyon and Anderson (2003, p. 490) suggested that roads may be the primary impact of oil and gas development to sage-grouse, due to their persistence and continued use even after drilling and production have ceased. Braun

et al

. (2002, p. 5) suggested that daily vehicular traffic along road networks for oil wells can impact sage-grouse breeding activities based on lek abandonment patterns.

In a study of 804 leks within 100 km (62.5 mi) of Interstate 80 in southern Wyoming and northeastern Utah, Connelly

et al

. (2004, p. 13-12) found that there were no leks within 2 km (1.25 mi) of the interstate and only 9 leks were found between 2 and 4 km (1.25 and 2.5 mi) along this same highway. The number of active leks increased with increasing distance from the interstate. Lek persistence and activity relative to distance from the interstate also were measured. The distance of a lek from the interstate was a significant predictor of lek activity, with leks further from the interstate more likely to be active. An analysis of long-term changes in populations between 1970 and 2003 showed that leks closest (within 7.5 km (4.7 mi)) to the interstate declined at a greater rate than those further away (Connelly

et al

. 2004, p. 13-13). Extirpated sage-grouse range was 60 percent closer to highways (Wisdom

et al

. in press, p. 18). What is not clear from these studies is what specific factor relative to roads (e.g., noise, changes in vegetation, etc.) sage-grouse are responding to. Connelly

et al

. (2004, p. 13-13) caution that they have not included other potential sources of indirect disturbance (e.g., powerlines) in their analyses.

Aldridge

et al

. (2008, p. 992) did not find road density to be an important factor affecting sage-grouse persistence or rangewide patterns in sage-grouse extirpation. However, the authors did not consider the intensity of human use of roads in their modeling efforts. They also indicated that their analyses may have been influenced by inaccuracies in spatial road data sets, particularly for secondary roads (Aldridge

et al

. 2008, p. 992). However, Wisdom

et al

. (in press, p. 18) found that extirpated range has a 25 percent higher density of roads than occupied range. Wisdom

et al

.'s (in press) rangewide analysis supports the findings of numerous local studies

showing that roads can have both direct and indirect impacts on sage-grouse distribution and individual fitness (e.g., Lyon and Anderson 2003, Aldridge and Boyce 2007).

Railroads

Railroads presumably have the same potential impacts to sage-grouse as do roads because they create linear corridors within sagebrush habitats. Railways and the cattle they transport were primarily responsible for the initial spread of

Bromus tectorum

in the intermountain region (Connelly

et al

. 2004, p. 7-25).

B. tectorum

, an exotic species that is unsuitable as sage-grouse habitat, readily invaded the disturbed soils adjacent to railroads. Fires created by trains facilitated the spread of

B. tectorum

into adjacent areas. Knick

et al

. (in press, p. 109) found that railroads cover 487 km

2

(188 mi

2

) or less than 0.1 percent of the SGCA, but they estimated railroads could influence 10 percent of the SGCA based adding a 3-km (1.9-mi) buffer to estimate potential impacts from the exotic plants they can spread. Avian collisions with trains occur, although no estimates of mortality rates are documented in the literature (Erickson

et al

. 2001, p. 8).

Summary: Habitat Conversion for Agriculture; Urbanization; Infrastructure

Large losses of sagebrush shrub-steppe habitats due to agricultural conversion have occurred range wide, but have been especially significant in the Columbia Basin of Washington (MZ VI), the Snake River Plain of Idaho (MZ IV), and the Great Plains (MZ I). Conversion of sage brush habitats to cropland continues to occur, although quantitative data is available only for Montana. We do not know the current rate of conversion, but most areas suitable for agricultural production were converted many years ago. The current rate of conversion is likely to increase in the future if incentives for crop production for use as biofuels continue to be offered. Urban and exurban development also have direct and indirect negative effects on sage-grouse, including direct and indirect habitat losses, disturbance, and introduction of new predators and invasive plant species. Given current trends in the Rocky Mountain west, we expect urban and exurban development to continue. Infrastructure such as powerlines, roads, communication towers, and fences continue to fragment sage-grouse habitat. Past and current trends lead us to believe this source of fragmentation will increase into the future. Fragmentation of sagebrush habitats through a variety of mechanisms including those listed above has been cited as a primary cause of the decline of sage-grouse populations (Patterson 1952, pp. 192-193; Connelly and Braun 1997, p. 4; Braun 1998, p. 140; Johnson and Braun 1999, p. 78; Connelly

et al

. 2000a, p. 975; Miller and Eddleman 2000, p. 1; Schroeder and Baydack 2001, p. 29; Johnsgard 2002, p. 108; Aldridge and Brigham 2003, p. 25; Beck

et al

. 2003, p. 203; Pedersen

et al

. 2003, pp. 23-24; Connelly

et al

. 2004, p. 4-15; Schroeder

et al

. 2004, p. 368; Leu

et al

. in press, p. 19). The negative effects of habitat fragmentation on sage-grouse are diverse and include reduced lek persistence, lek attendance, winter habitat use, recruitment, yearling annual survival, and female nest site choice (Holloran 2005, p. 49; Aldridge and Boyce 2007, pp. 517-523; Walker

et al

. 2007a, pp. 2651-2652; Doherty

et al

. 2008, p. 194). Since fragmentation is associated with most anthropogenic activities, the effects are ubiquitous across the species range (Knick

et al

. in press, p. 24). We agree with the assessment that habitat fragmentation is a primary cause of sage-grouse decline and in some areas has already led to population extirpation. We also conclude that habitat fragmentation will continue into the foreseeable future and will continue to threaten the persistence of greater sage-grouse.

Fire

Many of the native vegetative species of the sagebrush-steppe ecosystem are killed by wildfires, and recovery requires many years. As a result of this loss of habitat, fire has been identified as a primary factor associated with greater sage-grouse population declines (Hulet 1983,

in

Connelly

et al

. 2000a, p. 973; Crowley and Connelly 1996,

in

Connelly

et al

. 2000c, p. 94; Connelly and Braun 1997, p. 232; Connelly

et al

. 2000a, p. 973; Connelly

et al

. 2000c, p. 93; Miller and Eddlemen 2000, p. 24; Johnson

et al

., in press, p. 12; Knick and Hanser, in press, pp. 29-30). In nesting and wintering sites, fire causes direct loss of habitat due to reduced cover and forage (Call and Maser 1985, p. 17). For example, prescribed fires in mountain big sagebrush at Hart Mountain National Antelope Refuge caused a short-term increase in certain forbs, but reduced sagebrush cover, making habitat less suitable for nesting (Rowland and Wisdom 2002, p. 28). Similarly, Nelle

et al

. (2000, p. 586) and Beck

et al

. (2009, p. 400) reported nesting habitat loss from fire, creating a long-term negative impact that will require 25 to 150 years of sagebrush regrowth before sufficient canopy cover becomes available for nesting birds.

In southeastern Idaho, sage-grouse populations were generally declining across the entire study area, but declines were more severe in post-fire years (Connelly

et al

. 2000c, p. 93). Further, Fischer

et al

. (1997, p. 89) concluded that habitat fragmentation caused by fire may influence distribution or migratory patterns in sage-grouse. Hulet (1983, in Connelly

et al

. 2000a, p. 973) documented the loss of leks from fire.

Fire within 54 km (33.6 mi) of a lek is one of two primary factors in predicting lek extirpation (Knick and Hanser in press, p. 26). Small increases in the amount of burned habitat surrounding a lek had a large influence on the probability of lek abandonment (Knick and Hanser, in press, pp. 29-30). Additionally, fire had a negative effect on lek trends in the Snake River Plain (MZ IV) and Southern Great Basin (MZ III) (Johnson

et al

. in press, p.12). Several recent studies have demonstrated that sagebrush area is one of the best landscape predictors of greater sage-grouse persistence (Aldridge

et al

. 2008, p. 987; Doherty

et al

. 2008, p. 191; Wisdom

et al

., in press, p. 17). While there may be limited instances where burned habitat is beneficial, these gains are lost if sagebrush habitat is not readily available (Woodward 2006, p. 65).

Herbaceous understory vegetation plays a critical role throughout the breeding season as a source of forage and cover for sage-grouse females and chicks. The response of herbaceous understory vegetation to fire varies with differences in species composition, pre-burn site condition, fire intensity, and pre- and post-fire patterns of precipitation. In general, when not considering the synergistic effects of invasive species, any short-term flush of understory grasses and forbs is lost after only a few years and little difference is apparent between burned and unburned sites (Cook

et al

. 1994, p. 298; Fischer

et al

. 1996, p. 196; Crawford 1999, p. 7; Wrobleski 1999, p. 31; Nelle

et al

. 2000, p. 588; Paysen

et al

. 2000, p. 154; Wambolt

et al

. 2001, p. 250). Independent of the response of perennial grasses and forbs to fire, the most important and widespread sagebrush species for greater sage-grouse (i.e., big sagebrush) are killed by fire and require decades to recover. Prior to recovery, these sites are of limited to no use to sage-grouse (Fischer

et al

. 1996, p. 196; Connelly

et al

. 2000c, p. 90; Nelle

et al

. 2000, p. 588; Beck

et al

. 2009, p. 400). Therefore, fire results in direct, long-term habitat loss.

In addition to altering plant community structure, fires can influence invertebrate food sources

(Schroeder

et al

. 1999, p. 5). Ants (Hymenoptera), grasshoppers (Orthoptera), and beetles (Coleoptera) are an essential component of juvenile greater sage-grouse diets, especially in the first 3 weeks of life (Johnson and Boyce 1991, p. 90). Crawford and Davis (2002, p. 56) reported that the abundance of arthropods did not decline following wildfire. Pyle (1992, p. 14) reported no apparent effect of prescribed burning to beetles. However, Fischer

et al

. (1996, p. 197) found that the abundance of insects was significantly lower 2-3 years post-burn. Additionally, grasshopper abundance declined 60 percent in burned plots versus unburned plots 1 year post-burn, but this difference disappeared the second year (Bock and Bock 1991, p. 165). Conversely, Nelle

et al

. (2000, p. 589) reported the abundance of beetles and ants was significantly greater in 1-year-old burns, but returned to pre-burn levels by years 3 to 5. The effect of fire on insect populations likely varies due to a host of environmental factors. Because few studies have been conducted and the results of those available vary, the specific magnitude and duration of the effects of fire on insect communities is still uncertain, as is the effect any changes may have on greater sage-grouse populations.

The few studies that have suggested fire may be beneficial for greater sage-grouse were primarily conducted in mesic areas used for brood-rearing (Klebenow 1970, p. 399; Pyle and Crawford 1996, p. 323; Gates 1983,

in

Connelly

et al

. 2000c, p. 90; Sime 1991,

in

Connelly

et al

. 2000a, p. 972). In this habitat, small fires may maintain a suitable habitat mosaic by reducing shrub encroachment and encouraging understory growth. However, without available nearby sagebrush cover, the utility of these sites is questionable. For example, Slater (2003, p. 63) reported that sage-grouse using burned areas were rarely found more than 60 m (200 ft) from the edge of the burn and may preferentially use the burned and unburned edge habitat. However, Byrne (2002, p. 27) reported avoidance of burned habitat by nesting, brood-rearing, and broodless females. Both Connelly

et al

. (2000c, p. 90) and Fischer

et al

. (1996, p. 196) found that prescribed burns did not improve brood-rearing habitat in Wyoming big sagebrush, as forbs did not increase and insect populations declined. Hence, fires in these locations may negatively affect brood-rearing habitat rather than improve it (Connelly and Braun 1997, p. 11).

The nature of historical fire patterns in sagebrush communities, particularly in

Artemisia tridentata

var.

wyomingensis

, is not well understood and a high degree of variability likely occurred (Miller and Eddleman 2000, p. 16; Zouhar

et al

. 2008, p. 154; Baker in press, p. 16). However, as inferred by several lines of reasoning, fire in sagebrush systems was historically infrequent (Baker in press, pp. 15-16). This conclusion is evidenced by the fact that most sagebrush species have not developed evolutionary adaptations such as re-sprouting and heat-stimulated seed germination found in other shrub-dominated systems, like chaparral, exposed to relatively frequent fire events. Baker (in press, p. 17) suggests natural fire regimes and landscapes were typically shaped by a few infrequent large fire events that occurred at intervals approaching the historical fire rotation (50 to 350 years - see discussion below). The researcher concludes that the historical sagebrush systems likely consisted of extensive sagebrush habitat dotted by small areas of grassland and that this condition was maintained by long interludes of numerous small fires, accounting for little burned area, punctuated by large fire events that consumed large expanses. In general, fire extensively reduces sagebrush within burned areas, and big sagebrush varieties, the most widespread species of sagebrush, can take up to 150 years to reestablish an area (Braun 1998, p. 147; Cooper

et al

. 2007, p. 13; Lesica

et al

. 2007, p. 264; Baker, in press, pp. 15-16).

Fire rotation, or the average amount of time it takes to burn once through a particular landscape, is difficult to quantify in large sagebrush expanses. Because sagebrush is killed by fire, it does not record evidence of prior burns (i.e., fire scars) as do forested systems. As a result, a clear picture of the complex spatial and temporal pattern of historical fire regimes in most sagebrush communities is not available. Widely variable estimates of historical fire rotation have been described in the literature. Depending on the species of sagebrush and other site-specific characteristics, fire return intervals from 10 to well over 300 years have been reported (McArthur 1994, p. 347; Peters and Bunting 1994, p. 33; Miller and Rose 1999, p. 556; Kilpatrick 2000, p. 1; Frost 1998,

in

Connelly

et al

. 2004, p. 7-4; Zouhar

et al

. 2008, p. 154; Baker in press, pp. 15-16). In general, mean fire return intervals in low-lying, xeric, big sagebrush communities range from over 100 to 350 years, and return intervals decrease from 50 to over 200 years in more mesic areas, at higher elevations, during wetter climatic periods, and in locations associated with grasslands (Baker 2006, p. 181; Mensing

et al

. 2006, p. 75; Baker, in press, pp. 15-16; Miller

et al

., in press, p. 35).

The invasion of exotic annual grasses, such as

Bromus tectorum

and

Taeniatherum asperum

(medusahead), has been shown to increase fire frequency within the sagebrush ecosystem (Zouhar

et al

. 2008, p. 41; Miller

et al

. in press, p. 39).

B. tectorum

readily invades sagebrush communities, especially disturbed sites, and changes historical fire patterns by providing an abundant and easily ignitable fuel source that facilitates fire spread. While sagebrush is killed by fire and is slow to reestablish,

B. tectorum

recovers within 1 to 2 years of a fire event (Young and Evans 1978, p. 285). This annual recovery leads to a readily burnable fuel source and ultimately a reoccurring fire cycle that prevents sagebrush reestablishment (Eiswerth

et al

. 2009, p. 1324). In the Snake River Plain (MZ IV), for example, Whisenant (1990, p. 4) suggests fire rotation due to

B. tectorum

establishment is now as low as 3-5 years. It is difficult and usually ineffective to restore an area to sagebrush after annual grasses become established (Paysen

et al

. 2000, p. 154; Connelly

et al

. 2004, pp. 7-44 to 7-50; Pyke, in press, p. 25). Habitat loss from fire and the subsequent invasion by nonnative annual grasses have negatively affected sage-grouse populations in some locations (Connelly

et al

. 2000c, p. 93).

Evidence exists of a significant relationship between an increase in fire occurrence caused by

Bromus tectorum

invasion in the Snake River Plain and Northern Great Basin since the 1960s (Miller

et al

., in press, p. 39) and in northern Nevada and eastern Oregon since 1980 (MZs IV and V). The extensive distribution and highly invasive nature of

B. tectorum

poses substantial increased risk of fire and permanent loss of sagebrush habitat, as areas disturbed by fire are highly susceptible to further invasion and ultimately habitat conversion to an altered community state. For example, Link

et al

. (2006, p. 116) show that risk of fire increases from approximately 46 to 100 percent when ground cover of

B. tectorum

increases from 12 to 45 percent or more. In the Great Basin Ecoregion (defined as east-central California, most of Nevada, and western Utah, MZs IV and V), approximately 58 percent of sagebrush habitats are at moderate to high risk of

B. tectorum

invasion during the next 30 years (Suring

et al

. 2005, p. 138). The BLM estimated that approximately 11.9 million ha (29 million ac) of public

lands in the western distribution of the greater sage-grouse (Washington, Oregon, Idaho, Nevada, Utah) were infested with weeds as of 2000 (BLM 2007a, p. 3-28). The most dominant invasive plants consist of grasses in the

Bromus

genus, which represent nearly 70 percent of the total infested area (BLM 2007a, p. 3-28).

Conifer woodlands have expanded into sagebrush ecosystems over the last century (Miller

et al

. in press, p. 34). Woodlands can encroach into sagebrush communities when the interval between fires becomes long enough for seedlings to establish and trees to mature and dominate a site (Miller

et al

. in press, p. 36). However, historical fire rotation appears to have been sufficiently long to allow woodland invasion, and yet extensive stands of mature sagebrush were evident during settlement times (Vale 1975, p. 33; Baker, in press, pp. 15-16). This suggests that causes other than active fire suppression must largely explain recent tree invasions into sagebrush habitats (Baker in press, p. 21, 24). Baker (in press, p. 24) and Miller

et al

. (in press, p. 37) offer a suite of causes, acting in concert with fire exclusion that may better explain the dramatic expansion of conifer woodlands over the last century. These causes include alterations due to domestic livestock grazing (such as reduced competition from native grasses and forbs and facilitation of tree regeneration by increased shrub cover and enhanced seed dispersal), climatic fluctuations favorable to tree regeneration, enhanced tree growth due to increased water use efficiency associated with carbon dioxide fertilization, and recovery from past disturbance (both natural and anthropogenic). Regardless of the cause of conifer woodland encroachment, the rate of expansion is increasing and is resulting in the loss and fragmentation of sagebrush habitats (see discussion in Pinyon-juniper section below).

Between 1980 and 2007, the number of fires and total area burned increased in all MZs across the greater sage-grouse's range except the Snake River Plain (MZ IV) (Miller

et al

., in press, p. 39). Additionally, average fire size increased in the Southern Great Basin (MZ III) during this same period. However, predicting the amount of habitat that will burn during an “average fire” year is difficult due to the highly variable nature of fire seasons. For example, the approximate area burned on or adjacent to BLM-managed lands varied from 140,000 ha (346,000 ac) in 1998 to a 6-fold increase in 1999 (814,200 ha; 2 million ac) returning back down to approximately the 1998 level in 2002 (157,700 ha; 384,743 ac) before rising again 10-fold in 2006 (1.4 million ha; 3.5 million ac) (Miller

et al

., in press, pp. 39-40).

From 1980 to 2007, wildfires have burned approximately 8.7 million ha (21.5 million ac) of sagebrush, or approximately 18 percent of the estimated 47.5 million ha (117.4 million ac) of sagebrush habitat occurring within the delineated MZs (Baker, in press, p. 43). Additionally, the trend in total acreage burned since 1980 has primarily increased (Miller

et al

., in press, p. 39). Although fire alters sagebrush habitats throughout the greater sage-grouse's range, fire disproportionately affects the Great Basin (Baker

et al

. in press, p. 20) (i.e., Utah, Nevada, Idaho, and eastern Oregon; MZ III, IV, and V) and will likely influence the persistence of greater sage-grouse populations in the area. In these three MZs combined, nearly 27 percent of sagebrush habitat has burned since 1980 (Baker, in press, p. 43). A primary reason for this disproportionate influence in this region is due to the presence of burned sites and their subsequent susceptibility to invasion by exotic annual grasses.

According to one review, range fires destroyed 30 to 40 percent of sage-grouse habitat in southern Idaho (MZ IV) in a 5-year period (1997-2001) (Signe Sather-Blair, BLM,

in

Healy 2001). This amount included about 202,000 ha (500,000 ac), which burned between 1999 and 2001, significantly altering the largest remaining contiguous patch of sagebrush in the State (Signe Sather-Blair, BLM,

in

Healy 2001). Between 2003 and 2007, Idaho lost an additional 267,000 ha (660,000 ac) of sage-grouse habitat, or approximately 7 percent of the total estimated remaining habitat in the State. Over nine fire seasons in Nevada (1999-2007), about 1 million ha (2.5 million ac) of sagebrush were burned, representing approximately 12 percent of the State's extant sagebrush habitat (Espinosa and Phenix 2008, p. 3). Most of these fires occurred in northeast Nevada (MZ IV) within quality habitat that has traditionally supported high densities of sage-grouse, which also is highly susceptible to

Bromus tectorum

invasion.

Baker (in press, p. 20) calculated recent fire rotation by MZ and compared these to estimates of historical fire rotations. Based on this analysis, the researcher suggests that increased fire rotations since 1980 are presumably outside the historic range of variability and far shorter in floristic regions where Wyoming big sagebrush is common (Baker in press, p. 20). This analysis included MZs III, IV, V, and VI, all of which have extensive

Bromus tectorum

invasions.

In addition to wildfire, land managers are using prescribed fire as well as mechanical and chemical treatments to obtain desired management objectives for a variety of wildlife species and domestic ungulates in sagebrush habitats throughout the range of the greater sage-grouse. While the efficacy of treatments in sagebrush habitats to enhance sage-grouse populations is questionable (Peterson 1970, p. 154; Swensen

et al

. 1987, p. 128; Connelly

et al

. 2000c, p. 94; Nelle

et al

. 2000, p. 590; WAFWA 2009, p. 12; Connelly

et al

. in press c, p. 8), as with wildland fire, an immediate and potentially long-term result is the loss of habitat (Beck

et al

. 2009, p. 400).

Knick

et al

. (in press, p. 33) report that more than 370,000 ha (914,000 ac) of public lands were treated with prescribed fire to address management

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