# Endangered and Threatened Wildlife and Plants; Listing Lepidium papilliferum (Slickspot Peppergrass) as a Threatened Species Throughout Its Range

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3AE9-24039

## Record

- **Collection:** Federal Register
- **Document type:** Rule
- **Published:** October 8, 2009
- **Citation:** 74 FR 52014

## Text

DEPARTMENT OF THE INTERIOR
Fish and Wildlife Service
50 CFR Part 17
[RIN 1018-AW34]
[FWS-R1-ES-2008-0096]
[MO 922105-0008-B2]

Endangered and Threatened Wildlife and Plants; Listing
Lepidium papilliferum
(Slickspot Peppergrass) as a Threatened Species Throughout Its Range

AGENCY:

Fish and Wildlife Service, Interior.

ACTION:

Final rule.

SUMMARY:

We, the U.S. Fish and Wildlife Service (Service), determine that
Lepidium papilliferum
(slickspot peppergrass), a plant species from southwest Idaho, is a threatened species under the Endangered Species Act of 1973, as amended (Act). This final rule implements the Federal protections provided by the Act for this species. We have determined that critical habitat for
L. papilliferum
is prudent but not determinable at this time.

DATES:

This rule becomes effective December 7, 2009. The effective date has been extended to 60 days after publication in the
Federal Register
to allow the U.S. Bureau of Land Management (BLM) to finish conferring with the Service under section 7(a)(4) of the Act on the BLM's issuance of grazing permits within the range of
Lepidium papilliferum
.

ADDRESSES:

This final rule is available on the Internet at
http://www.regulations.gov
and also at
http://www.fws.gov/idaho
. Comments and materials received, as well as supporting documentation used in the preparation of this rule, will be available for public inspection, by appointment, during normal business hours at: U.S. Fish and Wildlife Service, Idaho Fish and Wildlife Office, 1387 S. Vinnell Way, Room 368, Boise, ID 83709; by telephone at 208-378-5243; by facsimile at 208-378-5262.

FOR FURTHER INFORMATION CONTACT:

Jeff Foss, Field Supervisor, at above address, telephone, and facsimile, or by electronic mail at:
fw1srbocomment@fws.gov
. Persons who use a telecommunications device for the deaf (TDD) may call the Federal Information Relay Service (FIRS) at 800-877-8339.

SUPPLEMENTARY INFORMATION:

Background

Lepidium papilliferum
is a small, flowering plant in the mustard family (Brassicaceae). The plant grows in unique microsite habitats known as slickspots, which are found within the semiarid sagebrush-steppe ecosystem of southwestern Idaho. The species is endemic to this region, known only from the Snake River Plain and its adjacent northern foothills (an area approximately 90 by 25 miles (mi) (145 by 40 kilometers (km)), or 2,250 square miles (mi
2
) (5,800 square kilometers (km
2
)), with a smaller disjunct population on the Owyhee Plateau (an area of approximately 11 by 12 mi (18 by 19 km), or 132 mi
2
(342 km
2
). The restricted distribution of
L. papilliferum
is likely due to its adaptation to the specific conditions within these slickspot habitats. The absence of all perennial plant species from these sites likewise demonstrates the specialization of
L. papilliferum
persisting in the unique conditions provided by slickspots (Fisher
et al.
1996, p. 16). The primary threat to
L. papilliferum
(as described under
The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range,
below) is the present or threatened destruction, modification, or curtailment of its habitat and range due to the increased frequency and extent of wildfires under a wildfire regime modified and exacerbated by the spread of invasive nonnative plants, particularly nonnative annual grasses such as
Bromus tectorum
(cheatgrass). In addition, even under conservative projections of the consequences of future climate change, the threats posed by wildfire and the invasion of
B. tectorum
are expected to further increase within the foreseeable future. Other threats to the species include competition and displacement by nonnative plant species, development, potential seed predation by harvester ants, and habitat fragmentation and isolation of small populations.

Previous Federal Actions

On July 15, 2002, we proposed to list
Lepidium papilliferum
as endangered (67 FR 46441). On January 12, 2007, we published a document in the
Federal Register
withdrawing that proposed rule (72 FR 1622). For a description of Federal actions concerning
L. papilliferum
prior to the 2007 withdrawal, please refer to that 2007 withdrawal document. The withdrawal of the proposal to list
L. papilliferum
was based on our conclusion that, while its sagebrush-steppe matrix habitat is becoming increasingly degraded, the best available data at the time provided no evidence indicating that this degradation was impacting
L. papilliferum
within its slickspot microsites. Furthermore, we concluded that, although we found that abundance on the Idaho Army National Guard's Orchard Training Area (OTA) had decreased in recent years, the observed rangewide fluctuations in population numbers appeared to be consistent with varying levels of spring rainfall, as expected. On April 6, 2007, Western Watersheds Project filed a lawsuit challenging our decision to withdraw the proposed rule to list
L. papilliferum.
On June 4, 2008, the U.S. District Court for the District of Idaho (Court) reversed the decision to withdraw the proposed rule, with directions that the case be remanded to the Service for further consideration consistent with the Court's opinion (
Western Watersheds Project
v.
Kempthorne,
Case No. CV 07-161-E-MHW (D. Idaho)).

After issuance of the Court's remand order, we published a public notification of the reinstatement of our July 15, 2002, proposed rule to list
Lepidium papilliferum
as endangered and announced the reopening of a public comment period on September 19, 2008 (73 FR 54345). The initial comment period closed on October 20, 2008. After the close of the comment period, new information became available that was relevant to our evaluation. Much of this information was contained in reports based on several independent analyses of the available information regarding
L. papilliferum
population trends on the OTA in southwest Idaho, the rangewide Habitat Integrity and Population (HIP) monitoring, and a recent analysis of
L. papilliferum
data collected on the Inside Desert (Owyhee Plateau) from 2000 to 2002. To ensure that our review of the species' status was complete, we announced another reopening of the comment period on March 17, 2009, for a period of 30 days (74 FR 11342). We posted several documents on
http://www.regulations.gov
for public review and comment, including the additional information and statistical analyses we received after the January 2007 withdrawal notice (72 FR 1622; January 12, 2007). A summary of the comments we received and our responses is provided in this document, following our finding.

Species Information

Description

Lepidium papilliferum
is an intricately branched, tap-rooted plant, averaging 2 to 8 inches (in) (5 to 20

centimeters (cm)) high, but occasionally reaching up to 16 in (40 cm) in height. Leaves and stems are covered with fine, soft hairs, and the leaves are divided into linear segments. Flowers are numerous, 0.1 in (3 to 4 millimeters (mm)) in diameter, white, and four petalled. Fruits (siliques) are 0.1 in (3 to 4 mm) across, round in outline, flattened, and two-seeded (Moseley 1994, pp. 3, 4; Holmgren
et al
. 2005, p. 260). The species is monocarpic (it flowers once and then dies) and displays two different life history strategies—an annual form and a biennial form. The annual form reproduces by flowering and setting seed in its first year, and dies within one growing season. The biennial life form initiates growth in the first year as a vegetative rosette, but does not flower and produce seed until the second growing season. Biennial rosettes must survive generally dry summer conditions, and consequently many of the biennial rosettes die before flowering and producing seed. The number of prior-year rosettes is positively correlated with the number of reproductive plants present the following year (ICDC 2008, p. 9; Unnasch 2008, p. 14; Sullivan and Nations 2009, p. 44). The proportion of annuals versus biennials in a population can vary greatly (Meyer
et al
. 2005, p. 15), but in general annuals appear to outnumber biennials (Moseley 1994, p. 12).

Seed Production

Depending on an individual plant's vigor, the effectiveness of its pollination, and whether it is functioning as an annual or a biennial, each
Lepidium papilliferum
plant produces varying numbers of seeds (Quinney 1998, pp. 15, 17). Biennial plants normally produce many more seeds than annual plants (Meyer
et al
. 2005, p. 15). Average seed output for annual plants at the OTA (an Idaho Army National Guard (IDARNG) training area on BLM land) was 125 seeds per plant in 1993 and 46 seeds per plant in 1994. In contrast, seed production of biennials at this site in 1993 and 1994 averaged 787 and 105 seeds per plant, respectively (Meyer
et al
. 2005, p. 16). Based on data collected from a 4-year demography study on the OTA, survivorship of the annual form of
L. papilliferum
was demonstrated to be higher than survivorship of biennials (Meyer
et al
. 2005, p. 16). For example, of the 4,065 plants counted in spring of 1993, a total of 2,503 survived to fruit as annuals, while only 85 survived to fruit as biennials in spring of 1994. Meyer
et al
. (2005, p. 21) hypothesize that the reproductive strategy of
L. papilliferum
is a plastic response, meaning that larger plants will flower and produce seed in their first season, whereas smaller plants that stand less chance of successfully setting seed in their first season will delay reproduction until the following year. The biennial life form is thus maintained, despite the higher risk of mortality.

Like many short-lived plants growing in arid environments, above-ground numbers of
Lepidium papilliferum
individuals can fluctuate widely from one year to the next, depending on seasonal precipitation patterns (Mancuso and Moseley 1998, p. 1; Meyer
et al
. 2005, pp. 4, 12, 15; Palazzo
et al
. 2005, p. 9; Menke and Kaye 2006a, p. 8; Menke and Kaye 2006b, pp. 10, 11; Sullivan and Nations 2009, p. 44). Mancuso and Moseley (1998, p. 1) note that sites with thousands of above-ground plants one year may have none the next, and vice versa. Above-ground plants represent only a portion of the population; the seed bank (a reserve of dormant seeds, generally found in the soil) contributes the other portion, and in many years constitutes the majority of the population (Mancuso and Moseley 1998, p. 1). Seed banks are adaptations for survival in a “risky environment,” because they buffer a species from stochastic (random) impacts, such as lack of soil moisture (Baskin and Baskin 2001, p. 160).

Seed Viability and Germination

The seeds of
Lepidium papilliferum
are found primarily within the slickspot microsites where the plants are found (Meyer and Allen 2005, pp. 5, 6). Slickspots, also known as mini-playas or natric (high sodium content) sites, are visually distinct openings in the sagebrush-steppe created by unusual soil conditions characterized by significantly greater sodium and clay content relative to the surrounding area (Moseley 1994, p. 7). The vast majority of
L. papilliferum
seeds in slickspots have been located near the soil surface, with lower numbers of seeds located in deeper soils (Meyer
et al
. 2005, p. 19; Palazzo
et al
. 2005, p. 3).
Lepidium papilliferum
seeds have been found in slickspots even if no above-ground plants are present (Meyer
et al
. 2005, p. 22; Palazzo
et al
. 2005, p. 10). When above-ground plants are present, flowering usually takes place in late April and May, fruit set occurs in June, and the seeds are released in late June or early July. Seeds produced in a given year are dormant for at least a year before any germination takes place. Following this year of dormancy, approximately 6 percent of the initially viable seeds produced in a given year germinate annually (Meyer
et al
. 2005, pp. 17, 18). When combined with an average annual 3 percent loss of seed viability, approximately 9 percent of the original seed cohort per year is lost after the first year. Thus, after 12 years, all seeds in a given cohort will likely have either died or germinated, resulting in a maximum estimated longevity of 12 years for seeds in the seed bank (Meyer
et al
. 2005, p. 18).

Billinge and Robertson (2008, pp. 1005-1006) report that both small and large
Lepidium papilliferum
populations share similar spatial structure, and that spatial structuring within its unique microsite slickspot habitats suggests that both pollen dispersal and seed dispersal are low for this species and occur over short distances (Robertson
et al
. 2006a, p. 3; Billinge and Robertson 2008, pp. 1005-1006). Modeling of dispersal and seed dormancy characteristics of desert annual plants predicts that plants with long-range dispersal will have few dormancy mechanisms and thus quick germination (Venable and Lawlor 1980, p. 272). Contrary to this prediction, however,
L. papilliferum
has delayed germination (Meyer
et al
. 2005, pp. 17-18), and, therefore, according to the model, may not disperse long distances. The primary seed dispersal mechanism for
L. papilliferum
is not known (Robertson and Ulappa 2004, p. 1708), although viable seeds have been found outside of slickspots, indicating that some seed dispersal is occurring beyond slickspot habitat (Palazzo
et al
. 2005, p. 10). Additionally, beginning in mid-July, entire dried-up biennial plants and some larger annual plants have been observed to break off at the base and are blown by the wind (Stillman, pers. obs., as reported in Robertson
et al
. 2006b, p. 44). This tumbleweed-like action may have historically resulted in occasional long-distance seed dispersal (Robertson
et al
. 2006b, p. 44). Ants are not considered to be a likely disperser despite harvesting an average of 32 percent of fruits across six sites (Robertson and White 2007, p. 11).

Lepidium papilliferum
seeds located near the soil surface show higher rates of germination and viability (Meyer and Allen 2005, pp. 6-8; Palazzo
et al
. 2005, p. 10) and the greatest seedling emergence success rate (Meyer and Allen 2005, pp. 6-8). Viable seeds were more abundant and had greater germination rates from the upper 2 in (5 cm) of soil (Palazzo
et al
. 2005, pp. 8, 10), while Meyer and Allen (2005, pp. 6-8) observed the upper 0.08 in (2 mm) optimal for germination. Deep burial of

L. papilliferum
seeds (average depths greater than 5.5 in (14 cm)) can entomb viable seeds and may preserve them beyond the 12-year period previously assumed as the maximum period of viability for
L. papilliferum
seeds (Meyer and Allen 2005, pp. 6, 9). However, seeds buried at such depth, even if they remain viable, are unlikely to regain the surface for successful germination. The effects of environmental factors such as wildfire on
L. papilliferum
seed dormancy and viability are currently unknown, although
L. papilliferum
abundance is reduced in burned areas (see discussion of Wildfire under
Summary of Factors Affecting the Species
).

Pollination

Lepidium papilliferum
is primarily an outcrossing species requiring pollen from separate plants for more successful fruit production and has a low seed set in the absence of insect pollinators (Robertson 2003a, p. 5; Robertson and Klemash 2003, p. 339; Robertson and Ulappa 2004, p. 1707; Billinge and Robertson 2008, pp. 1005-1006).
Lepidium papilliferum
is able to self-pollinate; however, with a selfing rate (rate of self-pollination) of 12 to 18 percent (Billinge 2006, p. 40; Robertson
et al
. 2006a, p. 40). In pollination experiments where researchers moved pollen from one plant to another, fruit production was observed to be higher with pollen from distant sources (4 to 12.4 mi (6.5 to 20 km) distance between patches of plants) compared to fruit production for plants pollinated with pollen from plants within the same patch (246 to 330 feet (ft) (75 to 100 meters (m)) distance within a plant patch) (Robertson and Ulappa 2004, p. 1705; Robertson
et al
. 2006a, p. 3).

Fruits produced from fertilized flowers reach full size approximately 2 weeks after pollination (Robertson and Ulappa 2004, p. 1706). Each fruit typically bears two seeds that drop to the ground when the fruit dehisces (splits open) in midsummer (Billinge and Robertson 2008, p. 1003).

Known
Lepidium papilliferum
insect pollinators include several families of bees (Hymenoptera), including Apidae, Halictidae, Sphecidae, and Vespidae; beetles (Coleoptera), including Dermestidae, Meloidae, and Melyridae; flies (Diptera), including Bombyliidae, Syrphidae, and Tachinidae; and others (Robertson and Klemash 2003, p. 336; Robertson
et al
. 2006b, p. 6). Seed set was not limited by the number of pollinators at any study site (Robertson
et al
. 2004, p. 14). Studies have shown a strong positive correlation between insect diversity and the number of
L. papilliferum
flowering at a site (Robertson and Hannon 2003, p. 8). Measurement of fruit set per visit revealed considerable variability in the effectiveness of pollination by different types of insects, ranging from 0 percent in dermestid beetles to 85 percent in honeybees (Robertson
et al
. 2006b, p. 15).

Genetics

The majority of species in the genus
Lepidium
have a base chromosome count of eight (Mummenhoff
et al
. 2001, p. 2051). Chromosome numbers for pollen mother cells in
L. papilliferum
ranged from 15 to 17 (n = 15.96 ± 0.16; Table 3; Figure 3), confirming that the plant is a tetraploid (has four sets of homologous chromosomes, as opposed to the more usual set of two) (Robertson
et al
. 2006b, p. 38).

The genetics of
Lepidium papilliferum
have been studied using samples collected from areas across the entire range of the species (Stillman
et al
. 2005, pp. 6, 8, 9; Larson
et al
. 2006, p. 14 and Fig. 4; Smith
et al
. in press, pp. 15-16). Genetic exchange can occur either through pollen or seed dispersal. Some researchers consider
L. papilliferum
to be closely related to
L. montanum
, and
L. papilliferum
was originally described as
L. montanum
var.
papilliferum
in 1900 by Louis Henderson. Results of genetic studies comparing
L. papilliferum
with
L. montanum
indicate that
L. papilliferum
forms a monophyletic group or subgroup that is genetically distinct from
L. montanum
(Larson
et al
. 2006, p. 13 and Figs. 4, 8; Smith 2006, pp. 5-7, Fig. 1). A more recent study examining the relationship between
L. montanum, L. papilliferum
, and
L fremontii
found that
L. papilliferum
is considered a sister taxa or closely related to
L. fremontii
, a native mustard of western North America (Smith
et al
. in press, pp. 15-16). Both
L. fremontii
and
L. papilliferum
are morphologically and ecologically distinct from
L. montanum,
and recent analyses reflect that both are monophyletic (organisms that share a common ancestor) with apparently little gene flow between them and
L. montanum
(Smith
et al
. in press, p. 18).

Some genetic differences have been observed between
Lepidium papilliferum
occurring on the Snake River Plain (now separated into the Boise Foothills and Snake River Plain regions) and the Owyhee Plateau. Plants in the Snake River Plain and the Owyhee Plateau populations are separated by a minimum of 44 mi (70 km), which is considered beyond the distance that insect pollinators can travel or that seed dispersal can occur. Sites in the Snake River Plain with fewer numbers of plants (16 to 746 flowering individuals) had less genetic diversity than sites with larger numbers of plants (more than 3,000 flowering individuals) (Robertson
et al
. 2006b, p. 42; Billinge and Robertson 2008, p. 1006), although this correlation between population size and genetic diversity was not evident in the Owyhee Plateau region (Stillman
et al
. 2005, p. 9; Robertson
et al
. 2006b, p. 41). The lowest values for average number of alleles per locus were detected in two of the smallest populations (Seaman's Gulch in the Boise Foothills region and Orchard in the Snake River Plain region); in contrast, the largest number of alleles per locus was detected in the second largest population (Kuna Butte SW in the Snake River Plain) (Robertson
et al
. 2006b, Table 4). Larson
et al
. (2006, p. 14 and Fig. 4) also found geographically well-defined populations of
L. papilliferum
between the Snake River Plain and Owyhee Plateau based on genetics. In contrast to the Stillman
et al
. (2005) study, Larson's findings indicate the possibility of depressed genetic diversity in
L. papilliferum
based on significantly greater average similarity coefficients within collection sites of
L. papilliferum
compared to those of
L. montanum
(Larson
et al
. 2006, p. 13).

In summary, recent genetic studies thus confirm that
Lepidium papilliferum
is a full species distinct from
L. montanum
. The currently accepted taxonomy recognizes
Lepidium papilliferum
(Henderson) A. Nels. and J.F. Macbr. as a full species (Taxonomic Serial No. 53383, Integrated Taxonomic Information System (ITIS), 2009). In addition, populations of
L. papilliferum
in the Owyhee Plateau demonstrate distinctive genetic differences from individuals in the Snake River Plain, likely a reflection of the isolation of these two populations due to limited seed dispersal and the limited range of pollinators, resulting in little current gene flow between them. Finally, there is some evidence that
L. papilliferum
has reduced genetic variability relative to other native species of
Lepidium
, such as
L. montanum
, and that smaller populations of
L. papilliferum
have less genetic diversity than larger populations.

Monitoring of
Lepidium papilliferum
Populations

There are several biological programs designed to monitor populations of
Lepidium papilliferum
over time, and, in some cases, its habitat as well. The primary monitoring programs are

described here to assist in understanding subsequent references to them in this document.

The Idaho Natural Heritage Program (INHP) uses element occurrences (EOs) to broadly describe the distribution of
Lepidium papilliferum
and assigns rankings to each EO based on measures of habitat quality and species abundance. EOs of
L. papilliferum
are defined by grouping occupied slickspots that occur within 1 km (0.6 mi) of each other; all occupied slickspots within a 1 km (0.6 mi) distance of another occupied slickspot are aggregated into a single EO. The definition of a single EO is based on the distance over which individuals of
L. papilliferum
are believed to be capable of genetic exchange through insect-mediated pollination (Colket and Robertson 2006). Due to the nature of their definition, individual EOs may differ greatly in size, based on whether there are many occupied slickspots distributed widely across the landscape relatively close to one another (which would comprise a single, large EO), or whether there are only a few (or even a single) slickspot(s) that occur close together but are relatively isolated from other occupied slickspots (which would comprise a single, small EO).

Each EO is assigned a qualitative rank defined by population size and habitat quality; EO ranks are periodically updated when new ranking information becomes available. Currently, no
Lepidium papilliferum
EOs are ranked A, which is defined as an EO with greater than 1,000 detectable above-ground plants occurring in the best habitat and landscape quality. The habitat quality rank diminishes from the highest of A to the lowest quality of D. An E ranking signifies that at least one plant was observed, but no abundance, habitat, or landscape data are available (Colket
et al
. 2006, p. 4). A rank of F indicates the most recent survey failed to find any
L. papilliferum
plants. A rank of H indicates
L. papilliferum
plants have not been documented at that location since 1970 based on old herbarium records with geographically vague location descriptions, such as a town name. A rank of X indicates
L. papilliferum
plants had been extirpated from that EO, based on agricultural conversion, commercial or residential development, or other documented habitat destruction where
L. papilliferum
plants had been previously recorded. An EO can also be ranked as X if it receives an F rank five times within a 12-year period (Colket
et al
. 2006, p. 4). The current rankings for
L. papilliferum
are reviewed below in the section
Element Occurrences Rangewide
.

The Habitat Integrity Index (HII) program conducted by the Idaho Conservation Data Center (ICDC, now the INHP) was the first rangewide effort aimed at monitoring
Lepidium papilliferum
and its habitat. The HII was initiated in 1998 and ran for 5 years through 2002 (Mancuso and Moseley 1998; Mancuso
et al
. 1998; Mancuso 2000, 2001, 2002, 2003). Although 52 transects were established over the years, a total of 17 transects were sampled during all years of HII monitoring (Mancuso 2003, p. 3); no rangewide monitoring of
L. papilliferum
was conducted in 2003. Monitoring was initially based on a system of transects of varying lengths across the range of
L. papilliferum
, each subjectively located to include 10 slickspots on sites known to contain
L. papilliferum
(summarized in Sullivan and Nations 2009, p. 33; see Mancuso
et al
. 1998 for details). The primary goal of the HII methodology was to assess the overall habitat condition, including attributes associated with the slickspots and the sagebrush-steppe habitat;
L. papilliferum
abundance was assessed categorically (assigned to a range of values) in this program.

In 2004, the HII was replaced by the Habitat Integrity and Population (HIP) monitoring protocol, also implemented by the ICDC. HIP monitoring has been conducted annually since its implementation, thus 5 years of HIP data are now available (through 2008) (ICDC 2008, p. 2; State of Idaho 2008). The HIP protocol was designed to provide data more replicable and specific to the monitoring required for the Candidate Conservation Agreement (CCA) developed by the State of Idaho, BLM, and others in 2003 (State of Idaho
et al
. 2003). HIP presents measures of habitat, disturbance, and plant community attributes at each transect as well as counts of
L. papilliferum
rosettes and reproductive plants observed (with the exception of 2004, which still utilized categorical assessments of plant abundance). Similar to the HII protocol, HIP is based on transects of varying lengths subjectively located to include 10 slickspots along their lengths (see Colket 2005 for details on the HIP methodology); however, the HIP protocol includes a significantly greater number of rangewide transects, having increased from the original 70 established in 2004 to 80 today (ICDC 2008, p. 3).

HIP monitoring has been annually conducted since 2004 and consists of the following procedures: (1) Establish and permanently mark HIP transects; (2) record location information; (3) take photographs; (4) measure population, habitat, and disturbance attributes at selected slickspots; (5) measure plant community attributes; and (6) analyze and describe the results (Colket 2008, p. 3).

The INHP's EO records and the HII-HIP monitoring programs cover the entire range of
Lepidium papilliferum
. In addition, monitoring that has occurred within a subset of the species' range, on the Idaho Army National Guard's Orchard Training Area (OTA), provides particularly important information on the status of
L. papilliferum
due to the long-term nature of the monitoring programs. The sagebrush-steppe on the OTA is considered to be some of the highest-quality habitat remaining within the range of
L. papilliferum
, and the OTA is home to one of the largest and most expansive EOs of the species (Sullivan and Nations 2009, p. 22). Two of the OTA programs have been monitoring the same locations annually (with a few exceptions) since the early 1990s, and hence provide up to 18 years of population data for
L. papilliferum
. These two monitoring programs are known as rough census areas and special-use plots; both are conducted by staff or contractors of the OTA.

The methods of the rough census monitoring areas are presented in Sullivan and Nations 2009 (pp. 28-29). Briefly, the program began in 1990 by monitoring 5 areas but expanded to the current total of 15 rough census areas by 1994; the combined extent of the rough census areas on the OTA is 866.1 ac (350.5 ha). Counts are conducted by technicians who walk across parallel transects 66 ft (20 m) apart and record the total number of
Lepidium papilliferum
individuals observed in any occupied slickspots that are found; reproductive status is not noted. The sizes of the 15 rough census areas differ, ranging from 4.1 ac (1.7 ha) to 138.3 ac (56.0 ha), and not all areas have been monitored in all years; thus, analyses of the data must be standardized by transforming the raw count data to plant density (number of plants per unit area) to account for these differences (Sullivan and Nations 2009, p. 36). Using density as the index of population abundance instead of total counts also allowed for the use of 18 years of rough census data, from 1990 through 2008 (there were no counts in 1999), although only a few of the rough census areas were monitored in the earlier years.

The special-use plots are also located on the OTA. Although called “plots,” these are actually a series of 16 belt transects, each containing a single

slickspot (see Sullivan and Nations 2009, pp. 29-33, for details). A stake is centered in the single slickspot, and each year the number of
Lepidium papilliferum
individuals with a 16.4-ft (5-m) radius of that stake (comprising a 32.8-ft (10-m) diameter circle) are counted (additional habitat information is collected from the remainder of the belt transect).
Lepidium papilliferum
abundance estimates for each of the 16 central circular plots has been collected annually each year from 1991 through 2008; thus, 18 years of special-use plot data are available. As all special-use plots were the same size and were surveyed in all years, estimates of abundance are based on reported total counts of individual plants (Sullivan and Nations 2009, p. 37). Beginning in 2000, the special-use plot data distinguished between blooming and nonblooming individuals.

All of these programs provide information regarding the status of
Lepidium papilliferum
and its habitat, and will be referenced throughout this rule. In addition, we reference
L. papilliferum
Management Areas, which are units containing multiple EOs in a particular geographic area with similar land management issues or administrative boundaries as defined in the 2003 CCA (State of Idaho, p. 9). At a larger scale is the
L. papilliferum
(or “LEPA”) Consideration Zone, an area also designated by the 2003 CCA and defined as all areas that may or do contain
L. papilliferum
(State of Idaho 2003, p. 21). The LEPA Consideration Zone includes the entire range of the species, including all Management Areas and all EOs.

Ecology and Habitat

The native, semiarid sagebrush-steppe habitat of southwestern Idaho where
Lepidium papilliferum
is found can be divided into two plant associations, each dominated by the shrub
Artemisia tridentata
ssp.
wyomingensis
(Wyoming big sagebrush):
A. tridentata
ssp.
wyomingensis-Achnatherum thurberianum
(formerly
Stipa thurberiana
) (Thurber's needlegrass) and
A. tridentata
ssp.
wyomingensis-Agropyron spicatum
(bluebunch wheatgrass) habitat types (Moseley 1994, p. 9). The perennial bunchgrasses
Poa secunda
(Sandberg's bluegrass) and
Sitanion hysrix
(bottlebrush squirreltail) are commonly found in the understory of these habitats, and the species
Artemisia tridentata
ssp.
tridentata
(basin big sagebrush),
Chrysothamnus nauseosus
(grey rabbitbrush),
Chrysothamnus viridiflorus
(green rabbitbrush),
Eriogonum strictum
(strict buckwheat),
Purshia tridentata
(bitterbrush), and
Tetradymium glabrata
(little-leafed horsebrush) form a lesser component of the shrub community (Moseley 1994, p. 9; Mancuso and Moseley 1998, p. 17). Under relatively undisturbed conditions, the understory is populated by a diversity of perennial bunchgrasses and forbs, including species such as
Achnatherum
(formerly
Oryzopsis) hymenoides
(Indian ricegrass),
Achillea millefolium
(common yarrow),
Phacelia heterophylla
(varileaf phacelia),
Astragalus purshii
(Pursh's milkvetch),
Phlox longifolia
(longleaf phlox), and
Aristida purpurea
var.
longiseta
(purple threeawn) (Moseley 1994, p. 9; Mancuso and Moseley 1998, p. 17; Colket 2005, pp. 2-3). Menke and Kaye (2006a, p. 1) describe high quality matrix habitat conditions for
L. papilliferum
as sagebrush-steppe habitat in late seral condition, and Fisher
et al
. (1996, p. 1) note that “habitat with vigorous
Lepidium
populations has not been recently burned, is not heavily grazed, has an understory of native bunchgrasses, and a well developed microbiotic soil crust.” Moseley (1994, p. 4) suggests that
L. papilliferum
serves as an indicator species for the health of the sagebrush-steppe ecosystem in the western Snake River Plain.

The biological soil crust, also known as a microbiotic crust or cryptogamic crust, is one component of quality habitat for
Lepidium papilliferum
. Such crusts are commonly found in semiarid and arid ecosystems, and are formed by living organisms, primarily bryophytes, lichens, algae, and cyanobacteria, that bind together surface soil particles (Moseley 1994, p. 9; Johnston 1997, p. 4). Microbiotic crusts play an important role in stabilizing the soil and preventing erosion, increasing the availability of nitrogen and other nutrients in the soil, and regulating water infiltration and evaporation levels (Johnston 1997, pp. 8-10). In addition, an intact crust appears to aid in preventing the establishment of invasive plants (Brooks and Pyke 2001, p. 4, and references therein; see also Serpe
et al
. 2006, pp. 174, 176). These crusts are sensitive to disturbances that disrupt crust integrity, such as compression due to livestock trampling or off-road-vehicle (ORV) use, and are also subject to damage by fire; recovery from disturbance is possible but occurs very slowly (Johnston 1997, pp. 10-11).

As described earlier,
Lepidium papilliferum
occurs in slickspot habitat microsites scattered within the greater semiarid sagebrush-steppe ecosystem of southwestern Idaho.
Lepidium papilliferum
has infrequently been documented outside of slickspots, on occasion being found on disturbed soils, such as along graded roadsides and badger mounds. These are rare observations and the vast majority of plants documented over the past 19 years of surveys and monitoring for the species are documented within slickspot microsite habitats (USFWS 2006, p. 20). For example, in 2002, a complete census of an 11,070-ac (4,480-ha) area recorded approximately 56,500 slickspots (U.S. Air Force, 2003, p. 15), of which approximately 2,450 (about 4 percent) were occupied by
L. papilliferum
plants (Bashore, pers. comm. 2003, p. 1). Of the approximately 11,300
L. papilliferum
plants documented during the survey effort, only 11 plants were documented outside of slickspots (U.S. Air Force 2002, in summary attachment of document).

Slickspots are visually distinct openings characterized by soils with high sodium content and distinct clay layers; they tend to be highly reflective and relatively light in color, which makes them easy to detect on the landscape (Fisher
et al
. 1996, p. 3). Slickspots are distinguished from the surrounding sagebrush matrix as having the following characteristics: microsites where water pools when rain falls (Fisher
et al
. 1996, pp. 2, 4), sparse native vegetation, distinct soil layers with a columnar or prismatic structure, higher alkalinity and clay content and natric properties (Fisher
et al
. 1996, pp. 15-16; Meyer and Allen 2005, pp. 3-5, 8; Palazzo
et al
. 2008, p. 378), and reduced levels of organic matter and nutrients due to lower biomass production (Meyer and Quinney 1993, pp. 3, 6; Fisher
et al
. 1996, p. 4). Fisher
et al
. (1996, p. 11) describe slickspots as having a “smooth, panlike surface” that is structureless and slowly permeable when wet, moderately hard and cracked when dry. Although the low permeability of slickspots appears to help hold moisture (Moseley 1994, p. 8), once the thin crust dries, out the survival of
L. papilliferum
seedlings depends on the ability to extend the taproot into the argillic horizon (soil layer with high clay content), to extract moisture from the deeper natric zone (Fisher
et al
. 1996, p. 13).

Slickspots have three primary layers: The surface silt layer, the restrictive layer, and an underlying moist clay layer. Although slickspots can appear homogeneous on the surface, the actual depth of the silt and restrictive layer can vary throughout the slickspot (Meyer and Allen 2005; Tables 9, 10, and 11). The top two layers (surface silt and restrictive) of slickspots are normally very thin; the surface silt layer varies in

thickness from 0.1 to 1.2 in (a few mm to 3 cm) in slickspots known to support
Lepidium papilliferum
, and the restrictive layer varies in thickness from 0.4 to 1.2 in (1 to 3 cm) (Meyer and Allen 2005, p. 3). The rangewide mean surface silt layer depth was 0.31 in (0.78 cm) based on a 2005 study of 769 slickspots of unknown occupancy sampled at 79 transects (Colket 2006, p. 38). Additionally, measurements of the depth of the clay layer next to
L. papilliferum
plants at the Juniper Butte Training Range were taken in 2007 and 2008 to assess if depth of the clay layer could be a significant factor for plant germination. The average depth of the clay layer next to plants measured in 2007 was 2.5 in (6.3 cm), with a range from 1.2 to 4.7 in (3.0 to 12.0 cm) (n=18), and in 2008 was 2.1 in (5.4 cm) with a range from 1.6 to 3.1 in (4.0 to 8.0 cm) (n=16) (CH2MHill 2008a, p. 13). It appears that depth to the clay layer is not as critical to germination at the Juniper Butte Training Range as other factors may be (such as depth to surface of the soil, the timing and amount of moisture, seed bank, and ability of the slickspot to capture and maintain adequate moisture).

It is not known how long slickspots take to form, but it is hypothesized to take several thousands of years (Nettleton and Peterson 1983, p. 193; Seronko 2006). Climate conditions that allowed for the formation of slickspots in southwestern Idaho are thought to have occurred during a wetter Pleistocene period. Holocene additions of wind-carried salts (often loess deposits) produced the natric soils (high in sodium) characteristic of slickspots (Nettleton and Peterson 1983, p. 191; Seronko 2006). It may take several hundred years to alter or lose slickspots through natural climate change or severe natural erosion (Seronko 2006). Some researchers hypothesize that, given current climatic conditions, new slickspots are no longer being created (Nettleton and Peterson 1983, pp. 166, 191, 206). As slickspots appear to have formed during the Pleistocene and new slickspots are not being formed, the loss of a slickspot is apparently a permanent loss.

Some slickspots subjected to light disturbance in the past may apparently be capable of re-forming (Seronko 2006). Disturbances that alter the physical properties of the soil layers, however, such as deep disturbance and the addition of organic matter, may lead to destruction and permanent loss of slickspots. For example, such techniques as deep soil tilling, the addition of organic matter, and addition of gypsum have been recommended for the elimination of slickspots from agricultural lands in Idaho (Peterson 1919, p. 11; Rasmussen
et al
. 1972, p. 142). Slickspot soils are especially susceptible to mechanical disturbances when wet (Rengasmy
et al
. 1984, p. 63; Seronko 2004). Such disturbances disrupt the soil layers important to
Lepidium papilliferum
seed germination and seedling growth, and alter hydrological function. Meyer and Allen (2005, p. 9) suggest that if sufficient time passes following the disturbance of slickspot soil layers, it is possible that the slickspot soil layers may regain their pre-disturbance configuration, yet not support the species. Thus, while the slickspot appears to have regained its former character, some essential component required to sustain the life history requirements of
L. papilliferum
has apparently been lost, or the active seed bank is no longer present.

Most slickspots are between 10 square feet (ft
2
) and 20 ft
2
(1 square meter (m
2
) and 2 m
2
) in size, although some are as large as 110 ft
2
(10 m
2
) (Mancuso
et al
. 1998, p. 1). Slickspots cover a relatively small cumulative area within the larger sagebrush-steppe matrix, and only a small percentage of slickspots are known to be occupied by
Lepidium papilliferum
. For example, a 2002 inventory of the 11,070 acre (ac) (4,480 hectare (ha)) Juniper Butte Range on the Owyhee Plateau found approximately 1 percent (109 ac (44 ha)) of the sagebrush-steppe area consisted of slickspot habitat, and of that slickspot habitat, only 4 percent (4 ac (1.6 ha)) was occupied by above-ground
L. papilliferum
plants (U.S. Air Force 2002, p. 9). It is not known why
L. papilliferum
is not found in a greater proportion of slickspot microsites (Fisher
et al
. 1996, p. 15).

The highest monthly temperatures within the range of
Lepidium papilliferum
normally occur in July (approximately in the low 90 degrees Fahrenheit (approximately 33 degrees Celsius)), and lowest monthly temperatures occur in January (approximately in the low 20 degrees Fahrenheit (minus 7 degrees Celsius)). Precipitation tends to fall as rain, primarily in winter and spring (November to May); the lowest rainfall occurs in July and August, with the months of June, September, and October receiving slightly more rainfall than July and August. Average annual precipitation patterns vary within the species' range, and are generally higher in the northern regions (e.g., 11.7 in (29.7 cm) near Boise, 7.4 in (18.8 cm) at the city of Bruneau, and 9.9 in (25.1 cm) at Mountain Home).

Several analyses have shown a positive association between above-ground abundance of
Lepidium papilliferum
and spring precipitation in the same year. Evaluating rangewide HII monitoring data collected over 4 years from 1998 to 2001, Palazzo
et al
. (2005, p. 9) found a positive relationship (p-value less than 0.01) between abundance of above-ground plants and February to June precipitation. Meyer
et al
. (2005, p. 15) found that an increase in February through May precipitation increased the number of
L. papilliferum
seedlings at the OTA based on
L. papilliferum
census and survival data collected from 1993 to 1995. CH2MHill (2007a, p. 14) analyzed data from 2005 to 2007 collected at the Juniper Butte Range in the Owyhee Plateau region and found a positive correlation between spring precipitation and plant numbers. Utilizing HII monitoring data collected from 1998 to 2002, as well as 2004 HIP monitoring data, Menke and Kay (2006a, b) found that March to May precipitation accounted for 99.4 percent of the variation in
L. papilliferum
abundance for the years 1998 to 2001 (2006a, p. 8), and 89 percent for the years 1998 to 2002, and 2004 (2006b, pp. 10-11). These results appear to have been strongly influenced by the data point for 1998, which was an unusually wet spring (Unnasch 2008, p. 16). Because the 1998 HII data represents an outlier with respect to both
L. papilliferum
abundance and precipitation, it largely determines the regression relationship by itself; thus, Menke and Kaye's 2006 conclusion that abundance increases with spring precipitation is not well supported (Sullivan and Nations 2009, p. 140). More recently, however, Sullivan and Nations (2009, pp. 30, 41) analyzed data collected at the OTA over a period of 18 years between 1990 and 2008, and found evidence that both plant density at the rough census areas and plant abundance at special-use plots were positively related to mean monthly precipitation in late winter and spring (January through May). Thus, analysis of this long-term dataset again points to a strong relationship between
L. papilliferum
abundance and spring precipitation. This correlation of abundance with spring rainfall is important, as it at least partially explains annual fluctuations in
L. papilliferum
population numbers.

In contrast, precipitation in the fall or early winter may have a negative effect on
Lepidium papilliferum
abundance the following spring (Meyer
et al
. 2005, p. 15; Sullivan and Nations 2009, p. 39). It has been suggested that this negative

relationship may be the result of prolonged flooding of the slickspot microsites, causing subsequent mortality of overwintering biennial rosettes (Meyer
et al
. 2005, pp. 15-16). This suggestion is supported by the analysis of 9 years of OTA data from the period 2000-2008 that shows a negative association between October to January precipitation and abundance of non-blooming
L. papilliferum
the following spring, although only the relationship with October to December precipitation is statistically significant (Sullivan and Nations 2009, p. 43). For blooming plants, the negative association between October to January precipitation and spring abundance was highly significant (Sullivan and Nations 2009, pp. 43-44).

However, Unnasch (2008, p. 2) found no relationship between precipitation and the abundance of
Lepidium papilliferum
in an analysis of HIP data collected over a 3-year period from 2005 to 2007. Unnasch hypothesized that
L. papilliferum
may manifest threshold effects in germination and that there is a pulse of germination following a requisite amount of rainfall that could lead to a major flush of
L. papilliferum
germination during very wet years. If total rainfall is below that threshold, annual germination is more random (Unnasch 2008, p. 16). Comparing his results to those of Menke and Kaye, Unnasch (2008, p. 15) suggests that the relationship with spring precipitation reported by Menke and Kaye was strongly affected by abundance data from the year 1998, although in turn the relatively short 3-year study period may have influenced Unnasch's study results. Sullivan and Nations (2009, pp. 140, 142) likewise suggested that the exceptionally high precipitation in 1998 likely influenced the results of Menke and Kaye's analysis. However, as described above, Sullivan and Nation's more robust analysis of 18 years of data from the OTA confirmed a positive correlation between spring precipitation and the abundance of
L. papilliferum
(Sullivan and Nations 2009, pp. 40-44). As both annual precipitation and plant abundance are highly variable, the numbers of years included in the data set for evaluation is of great importance in determining the degree of confidence in the outcome of any statistical analysis. For this reason, the Service believes the Sullivan and Nations (2009, pp. 40-44) evaluation of the 18-year dataset from the OTA is the best available data regarding the relationship between precipitation and abundance of
L. papilliferum
.

Recent analyses suggest that temperature also influences the annual abundance of
Lepidium papilliferum
. Although Menke and Kaye (2006b, p. 8) found that minimum and maximum temperatures were not statistically correlated with
L. papilliferum
abundance based on a limited number of years of data, Sullivan and Nations (2009, p. 46-57) used more precise temperature data in concert with the 18-year
L. papilliferum
abundance dataset from the OTA to evaluate the potential interaction between precipitation, temperature, and plant abundance. Their analysis of the data collected between 1990 and 2008 suggests a complex relationship between temperature and precipitation that influences the abundance of
L. papilliferum
on an annual basis. In short, they found that temperature and precipitation interact during the months of October through January such that the lowest density or abundance of
L. papilliferum
in the spring follows a fall or early winter when both precipitation and temperature are low, or both are high. Spring plant density or abundance is greatest following a fall or early winter when either precipitation is high and temperature is low, or precipitation is low and temperature is high (Sullivan and Nations 2009, p. 56). During late winter and spring, analysis of one OTA dataset (the “rough census” areas) suggested that temperature had a negative impact on
L. papilliferum
density, such that density is greater when precipitation is high but temperatures during March through May are lower (Sullivan and Nations 2009, p. 47), whereas the model of the OTA special-use plots suggests only a positive interaction of
L. papilliferum
abundance with precipitation during this time period, with no temperature effect (Sullivan and Nations 2009, p. 47). Sullivan and Nations caution that the limited geographic area within which the interactions of precipitation and temperature were studied limits the ability to extrapolate the observed relationship beyond the bounds of the OTA (Sullivan and Nations 2009, p. 57).

The sparse native vegetation naturally present at slickspots suggests that
Lepidium papilliferum
is more tolerant than surrounding vegetation at surviving in alkaline soils and spring inundation (e.g., Moseley 1994, p. 8, 14; Fisher
et al
. 1996, pp. 11, 16). Plant ecology literature suggests that plants tolerant of stress (e.g., plants that are capable of growing in harsh alkaline soils) are poor competitors (Grime 1977, p. 1185), making
L. papilliferum
a potentially poor competitor with other plants. In recent years, there are increasing observations of nonnative plants encroaching into slickspots, and consistent with theory, the evidence suggests that
L. papilliferum
is not able to successfully compete with these invasive exotics. Sullivan and Nations (2009, p. 111) report an “apparent mutual exclusivity” between nonnative plant species examined and
L. papilliferum
in slickspots. In other words, if plants such as
Bassia prostrata
(prostrate kochia or forage kochia, formerly
Kochia prostrata
) or
Bromus tectorum
are present in a slickspot,
L. papilliferum
is most often reduced in numbers or entirely absent.

Range and Distribution

The range of
Lepidium papilliferum
is restricted to the volcanic plains of southwest Idaho, occurring primarily in the Snake River Plain and its adjacent northern foothills, with a single disjunct a population on the Owyhee Plateau (Figure 1). The plant occurs at elevations ranging from approximately 2,200 ft (670 m) to 5,400 ft (1,645 m) in Ada, Canyon, Gem, Elmore, Payette, and Owyhee Counties (Moseley 1994, pp. 3-9). Based on differences in topography, soil, and relative abundance, we have further divided the extant
Lepidium papilliferum
populations into three physiographic regions: the Boise Foothills, the Snake River Plain, and the Owyhee Plateau. The nature and severity of factors affecting the species also vary between the three physiographic regions for the purposes of analysis. For example, urban and rural development, agriculture, and infrastructure development has been substantial in the sagebrush-steppe habitat of the Boise Foothills and the Snake River Plain regions, while very little of these types of development has occurred within the Owyhee Plateau region. Genetic analyses reveal some separation between the greater Snake River Plain and Owyhee Plateau populations of
L. papilliferum
(Larson
et al
. 2006, p. 14), as might be expected due to their relative isolation. We are not aware of any studies that may have examined the relative genetic differentiation, if any, of the Boise Foothills population from the remainder of the Snake River Plain.

Figure 1
. Range of
Lepidium papilliferum
in southwest Idaho, showing its distribution in the three physiographic provinces of the Snake River Plain, Boise Foothills, and Owyhee Plateau.

BILLING CODE 4310-55-S

ER08OC09.000

BILLING CODE 4310-55-C

As of February 2009, there were 80 extant EOs in the three physiographic regions that collectively comprise approximately 15,801 ac (6,394 ha) of total area that is broadly occupied by
Lepidium papilliferum
(Cole 2009b, Threats Table). The area actually occupied by
L. papilliferum
is a small fraction of the total acreage, since slickspots occupy only a small percentage of the landscape, and
L. papilliferum
then occupies only a fraction of those slickspots (see U.S. Air Force 2002, p. 9, for an example). Table 1 presents the distribution and landownership and management information for all
L. papilliferum
EOs, in total and by region.

Table 1. Distribution and Land Ownership of
Lepidium papilliferum
Element Occurrences by Physiographic Region (Cole 2009b, Threats Table; Sullivan and Nations 2009, p. 77).

All areas are estimates, and may not total exactly due to rounding.

Lepidium papilliferum
EOs

Number of EOs
[percent of total]

Federal ownership in acres
(hectares)
[percent of total]

State ownership
in acres
(hectares)
[percent of total]

Private ownership
in acres
(hectares)
[percent of total]

Total EO Area
(hectares)
[percent of total rangewide
EO area]

Snake River Plain

43
[54]

12,754 ac
(5,160 ha)
[98]

55 ac
(22 ha)
[0.5]

164 ac
(66 ha)
[1.5]

12,980 ac
(5,250 ha)
[82]

Boise Foothills

16
[20]

89 ac
(36 ha)
[48]

0 ac
(0 ha)
0

96 ac
(39 ha)
[52]

185 ac
(75 ha)
[1.2]

Owyhee Plateau

21
[26]

2,636 ac
(1,067 ha)
[99.7]

7 ac
(3 ha)
[0.3]

0 ac
(o ha)
[0]

2,643 ac
(1,070 ha)
[16. 8%]

All extant
EOs

80
[100]

15,479 ac
(6,264 ha)
[98.0]

62 ac
(25 ha)
[0.4]

260 ac
(105 ha)
[1.6]

15,801 ac
(6,394 ha)
[100]

The range of
Lepidium papilliferum
was first estimated in 1994 (Moseley 1994, p. 6). Expanded survey efforts in recent years have resulted in an increase in the amount of known occupied habitat, particularly on the Owyhee Plateau and in the Boise Foothill regions. Between 2003 and 2006, 16 new EOs were documented, all within 3 mi (4.8 km) of previously existing EOs: 2 on the Snake River Plain with a total area of 2.7 ac (1 ha), and 14 on the Owyhee Plateau with a total area of 46.6 ac (18 ha) (Colket
et al
. 2006, Tables and Appendix A). Since 2006, additional surveys of previously unsurveyed lands have resulted in the discovery of several new occupied sites. Because most of these newly discovered sites were within 1 km (0.6 mi) of a documented EO, they typically resulted in the expansion or merging of existing EOs rather than the creation of a new EO. For example, in 2007, 2,560 ac (1,036 ha) of BLM land on the Owyhee Plateau were inventoried for
L. papilliferum
just south of the U.S. Air Force's Juniper Butte Training Range. Of the 2,171 slickspots surveyed, 200 (9 percent) were occupied by
L. papilliferum
with a total of 1,059 flowering plants and 214 rosettes (ERO 2007, pp. 1, 7-8), resulting in the expansion of EO 16 (Cole 2009a, p. 38). Surveys conducted in 2008 in the vicinity of the Ada County landfill in the Boise Foothills region revealed nearly 5,000 plants in 75 slickspots (Cole 2008, p. 8), which expanded the size of existing EOs 38 and 65 (Cole 2009a, p. 39). Pre-development surveys conducted during 2007 by URS Corporation (URS) on BLM and private lands in the Boise Foothills region northwest of the City of Eagle detected 43 occupied slickspots out of 187 surveyed, with approximately 17,880
L. papilliferum
plants (URS 2008, p. 10). These observations expanded the total area of EO 76 (Cole 2009a, p. 39). Finally, additional survey efforts on previously surveyed areas at the OTA resulted in the documentation of 365 new occupied slickspots in 2005, resulting in further expansion of existing EO 27 (URS 2005, pp. 6-7).

Not all potential
Lepidium papilliferum
habitats in southwest Idaho have been surveyed, and it is possible that additional
L. papilliferum
sites may be found outside of areas that are currently known to be occupied. Recent modeling was completed to develop a high-quality, predictive-distribution model of
L. papilliferum
to identify potential habitat (Colket 2008, p. 1). Although surveys were conducted in 2008 in some areas identified as potential, previously unsurveyed habitat, these did not result in any new locations of the species (Colket 2008, pp. 4-6). There have also been searches for
L. papilliferum
in eastern Oregon, but the species has never been found there (Findley 2003, p. 1). We have no historical records indicating that
L. papilliferum
has ever been found anywhere outside of its present range in southwestern Idaho, as described in this rule.

Abundance and Population Trend

Forming a reliable estimate of any trend in the abundance of
Lepidium papilliferum
over time is complicated by multiple factors. For one, since individuals of the species may act as either an annual or a biennial, in any given year there will be varying numbers of plants acting as spring-flowering annuals versus overwintering rosettes. The relative proportions of these two life history forms can fluctuate annually depending on a variety of factors, including precipitation, temperature, and the abundance of rosettes produced the previous year (Unnasch 2008, pp. 14-15; Sullivan and Nations 2009, pp. 43-44, 134-135). Secondly,
L. papilliferum
has a long-lived seed bank, likely as an adaptation to unpredictable conditions, in which years of good rainfall favorable for germination and survival may be followed by periods of drought; a persistent seed bank provides a population buffer against years of poor reproductive potential in such a highly variable environment (Meyer
et al
. 2005, p. 21). Only a small percentage of
L. papilliferum
seeds germinate annually, resulting in an estimated maximum longevity of 12 years for seeds in the seed bank (Meyer
et al
2005, p. 18). The presence of this persistent seed bank confounds the ability to determine any trend in abundance over time, as the number of above-ground plants that can be counted in any one year represents only a subset of the latent population that is present in the seed bank. In effect, it takes at least 12 years to trace the fate of a single year's cohort of seeds, resulting in a significant lag effect in detecting any real underlying change in total population abundance over the long term.

An additional complicating factor in trying to detect any population trend for
Lepidium papilliferum
is the extreme

variability of annual abundance or density of the plant. As is common for desert annuals, the numbers of
L. papilliferum
can vary dramatically from year to year, depending on environmental conditions. As an example, the total number of plants on the 16 special-use plots at the OTA went from 624 individuals in 1997 to 3,330 plants in 1998, subsequently dropping back down again to 756 plants in 1999; total abundance over the years 1991 through 2008 ranged from a low of 249 plants to 15,236 individuals (Weaver 2008). Some of the great variation in yearly plant numbers is likely due to the relationship between
L. papilliferum
and precipitation, as described above. The annual abundance or density of
L. papilliferum
shows a significant positive association with levels of spring rainfall, roughly from March through May (Meyer
et al
. 2005, p. 15; Palazzo
et al
. 2005, p. 9; Sullivan and Nations 2009, pp. 39-41), and survival of potential biennials is associated with increased summer rainfall (Meyer
et al
. 2005, p. 15). There is also some suggestion that increased winter precipitation may show a negative association with plant abundance, although not all analyses are consistently significant on this point (Meyer
et al
. 2005, pp. 15-16; Sullivan and Nations 2009, pp. 39-41). Temperature also appears to play a role in annual abundance of
L. papilliferum
in concert with precipitation, although the exact nature of the relationship is complex and not well understood (Sullivan and Nations 2009, p. 57). Furthermore, the interaction between temperature, precipitation, and
L. papilliferum
abundance appears to vary regionally between the Boise Foothills, Owyhee Plateau, and Snake River Plain (Sullivan and Nations 2009, pp. 103-104).

Because the population dynamics of
Lepidium papilliferum
are complicated, surrogate methods of monitoring the status of the species, such as monitoring the status of the ecosystem upon which it depends, may be preferable to counts of individual plants. For example, due to the extreme annual fluctuations in annual plant abundance and the complicating nature of the long-lived seed bank for this species, Mancuso and Moseley (1998, p. 1) note that “estimating the number of above-ground plants is by itself not a reliable measure to evaluate population and species viability.” As an alternative or supplement to population monitoring, they suggest monitoring the ecological integrity of
L. papilliferum
habitat, essentially using measures of habitat quality and quantity as a surrogate for assessing the status or viability of
L papilliferum
. Habitat monitoring is a recommended method of monitoring annual plants with a long-lived seed bank, where in some years the majority of the plant population is expressed in the seed bank rather than as above-ground plants (Elzinga
et al
. 1998, p. 55). For these reasons, we consider that data regarding the trends in habitat quality and quantity for
L. papilliferum
provide us with information that is equally important, if not more so, than direct counts of individual plants in evaluating the overall status of the species. Trends in habitat quality are discussed in the
Habitat Quality
section of this document, as well as under
The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range
in the
Summary of Threats Affecting the Species
section, below.

From a statistical standpoint, the extreme variability in annual abundance or density estimates greatly reduces the ability to reliably detect a long-term trend in the population without many years of standardized data. The presence of the persistent seed bank adds further uncertainty to the determination of population trend, as 12 years may effectively be considered to represent a single generation of the plant. Relatively short-term analyses of abundance estimates for the purposes of estimating a population trend are thus of limited utility due to the high variance observed in the data (Sullivan and Nations 2009, p. 93). In our evaluation, we weighed the relative quality of the available datasets for discerning population trend in
Lepidium papilliferum
according to the degree of confidence we had in the results of any analyses, given the great degree of variability observed and the multiple factors potentially influencing annual counts of the plant.

Four data sets are available that provide some index or measure of
Lepidium papilliferum
abundance: Rangewide EO records, rangewide HII-HIP transects, rough census data collected on the OTA, and special-use plot data from the OTA. Each of these programs is described in the
Monitoring of

Lepidium papilliferum

Populations
section, above, and the degree to which we relied on the information provided by them is described below.

The INHP records of
Lepidium papilliferum
EOs provide only estimated ranges or categorical estimates of abundance, and are so variable in both size and space over time that we considered these records to be informative in terms of evaluating the current overall condition of the species, but we did not rely on EO records for temporal population trend estimates.

Five years of HII monitoring data (1998 to 2002) and 5 years of HIP monitoring data (2004 to 2008) are available on
Lepidium papilliferum
abundance and habitat condition rangewide. Although the HII-HIP program provides valuable information regarding the relationship between
L. papilliferum
abundance and measures of habitat quality or disturbance, the time series of this data set is considered too short to reliably detect any trend in rangewide population abundance, due to the extreme annual variability in the data (Sullivan and Nations 2009, p. 93).

We consider the best available data regarding
Lepidium papilliferum
abundance to be the long-term datasets from the OTA, including the rough census areas and special-use plots, which provide 18 years of population monitoring information. The relative value of the OTA dataset is supported by the analysis of Sullivan and Nations (2009), a report resulting from our contract with an independent consulting firm to evaluate the available population trend data for
L. papilliferum,
as well as to analyze any information available regarding potential relationships between the abundance of
L. papilliferum
and measures of habitat quality or disturbance. Considering the available data from the HII-HIP monitoring, and the rough census area and special-use plot monitoring from the OTA, Sullivan and Nations considered that the long-term nature of the datasets from the OTA make these data the best available data when attempting to model trends through time (Sullivan and Nations 2009, p. 56). Furthermore, they placed slightly greater confidence in the analyses based on the rough census areas as opposed to the special-use plots, since the special-use plots are in effect a subset of the rough census areas and are based on counts from only a single slickspot, and are therefore subject to greater variability in response to localized impacts (Sullivan and Nations 2009, pp. 55, 96). They also noted that the HII and HIP programs do not yet have sufficient data to determine population trends rangewide (Sullivan and Nations (2009, p. 93). However, they determined that all three programs—rangewide HIP, OTA rough census areas, and OTA special-use plots— track annual changes in
L. papilliferum
abundance similarly, and each can act as an index of abundance. Based on their analysis, they concluded that the trend observed on the OTA may be considered likely representative of

the trend across the entire range of the species (Sullivan and Nations 2009, p. 96).

Analysis of Population Trend

Sullivan and Nations analyzed the data on
Lepidium papilliferum
numbers (density or total abundance) from both the rough census areas and the special-use plots at the OTA, assuming a simple linear trend and using a repeated measures implementation of the general negative binomial regression model to account for the large variances in the data (a statistical technique for determining whether a statistically significant trend exists when using a data set with counts from the same areas every year and large changes in the values between years). The model was not intended to describe the complex pattern in the relative density or abundance of
L. papilliferum
over time, but only to determine whether there is evidence of any overall population trend (Sullivan and Nations 2009, p. 38).

Based on this model, of the two OTA datasets, Sullivan and Nations (2009, pp. 3, 55, 96) considered the rough census data to be slightly more reliable. Their analysis of this rough census data showed a negative trend in density with a slope of -0.086 over the years 1990 to 2008; this trend was statistically significant (p = 0.0087, two-sided p-value) (Sullivan and Nations 2009, pp. 38-39). Because plant density was unusually high on a single rough census area, the Study 4 Site, the data were reanalyzed, removing that site as a potentially highly influential data point. The result was a more shallow negative slope (-0.059), but the trend remained statistically significant (p = 0.0046) (Sullivan and Nations 2009, p. 39).

Rough census area densities were further regressed against 3-month running averages of precipitation.
Lepidium papilliferum
density was positively associated with mean monthly precipitation in each of the January to March, February to April, and March to May periods, and negatively associated with mean monthly precipitation for the periods October to December, November to January, and December to February; these relationships were all significant at p < 0.0001 (Sullivan and Nations 2009, pp. 39-40). These findings are consistent with those of Meyer
et al
. 2005 (pp. 15-16), which reported a positive association between
Lepidium
seedlings recruited and spring precipitation, and a likely negative association with winter precipitation, which is postulated to drown overwintering rosettes.

The analysis of abundance data from the special-use plots on the OTA reveals a similarly negative slope over the years 1991 through 2008, but the results were not statistically significant (p = 0.2857) (Sullivan and Nations 2009, p. 4). In other words, based on the count data from the special-use plots, there was not sufficient evidence to conclude that the slope of abundance over time was significantly different from zero. The relationship between abundance and spring precipitation on the special-use plots was similar to that observed on the rough census areas; mean monthly precipitation in January to March, February to April, and March to May were all positively associated with abundance and all were statistically significant (p < 0.0001). There was no significant relationship, however, between fall or winter precipitation and
Lepidium papilliferum
abundance on the special-use plots (Sullivan and Nations 2009, p. 41). Using a shorter time-series of data from 2000 to 2008, Sullivan and Nations (2009, pp. 43-44) found that the abundance of blooming plants was positively associated with both the current year's precipitation and the number of rosettes present in the previous year, and that the number of rosettes was negatively associated with precipitation in the prior October to December period.

The researchers concluded that there is “limited evidence for declining populations,” because trends on the OTA are negative but only statistically significant for the rough census areas (Sullivan and Nations 2009, pp. 2, 44). In earlier analyses of
Lepidium papilliferum
population HII-HIP data, Menke and Kaye had initially reported a negative rangewide population trend for the periods 1998 through 2002 (Menke and Kaye 2006a) and for 1998 through 2004 (Menke and Kaye 2006b). However, Sullivan and Nations (2009, p. 141) point out that the fact that the HII transects were first monitored during a higher-than-average abundance year in 1998 greatly influenced the interpretation of the short time-series dataset, and suggest that the negative trend in abundance is not supported when abundance in subsequent years is included. Additionally, as described above, the HII-HIP data collection has not yet occurred over a long enough period to allow for reliable trend analyses (Sullivan and Nations 2009, p. 93). In comparing the mean number of
L. papilliferum
per transect resulting from his own analyses of HIP data from 2005 through 2007 with the results reported by Menke and Kaye (2006b), Unnasch (2008, p. 14) suggests that, since 1999, there has been no consistent rangewide population trend for the species.

Although Sullivan and Nations did not attempt to discern a trend in population numbers based on the HIP data, they did compare mean total abundance of
Lepidium papilliferum
per transect between physiographic regions, based on the HIP data from 2004 through 2008. They found that relative abundance was significantly different between regions, being greatest in the Boise Foothills region and lowest on the Owyhee Plateau region; abundance on the Snake River Plain region was intermediate between the other two (Sullivan and Nations 2009, p. 103).

In summary, we have reviewed all of the best available scientific and commercial data available to us to determine whether we can discern a long-term trend in the abundance of
Lepidium papilliferum
. The extreme variability in annual counts of the species makes it difficult to discern a trend in numbers with statistical confidence. For this reason, we place greater confidence in the longest time series of monitoring data available to us, that from the OTA (up to 18 years of data for some rough census areas and all special-use plots). In addition, as described above, Sullivan and Nations suggest that the data from the rough census areas may be considered slightly more reliable than that from the special-use plots (Sullivan and Nations 2009, pp. 3, 55). The long-term data from the OTA, which we considered to be the best available data for attempting to model trends through time in agreement with Sullivan and Nations (2009, pp. 3, 56), suggest that population numbers may be trending downward on the OTA. Although numbers on both the rough census areas and the special-use plots showed a slightly negative slope over time, only the analysis of the rough census areas was statistically significant (Sullivan and Nations 2009, pp. 38-40). We considered this to be relatively limited evidence of a downward trend in the population, given the lack of consistently significant results between the two monitoring programs. Furthermore, the slope is not steep, annual variation in plant numbers continues to be extremely high, and the plant has demonstrated an ability to rebound from low numbers due to the persistent seed bank.

We do recognize, however, that the OTA provides some of the highest quality habitat remaining for
Lepidium papilliferum
. Therefore, we believe it is reasonable to infer that if the population is trending downward there, then conditions are likely worse in the

remainder of the plant's range where habitat conditions are more degraded. This conclusion is supported by the analysis of Sullivan and Nations (2009, p. 96), which suggests that the trends on the OTA, as a general index of abundance, might reasonably be considered representative of trends rangewide (Sullivan and Nations 2009, p. 96). Direct evidence in support of this argument, however, is lacking. In addition, since the abundance of
L. papilliferum
is associated with annual precipitation, we considered whether any trend in precipitation over the same time period for which the rough census areas and special-use plot data were collected might be correlated with the observed negative trend in plant numbers. Assuming a simple linear trend, analogous to the model used by Sullivan and Nations in their analysis of
L. papilliferum
density and abundance at the OTA over time, we found no significant trend in precipitation at the OTA over the years 1991 through 2007 (data were not available for 2008). Although we evaluated total annual precipitation, total and mean winter precipitation, total and mean spring precipitation, and 3-month moving averages across the year, least squares regression did not yield any slopes of precipitation over time that were statistically significant from zero (Zwartjes 2009, p. 1). Any observed negative trend in
L. papilliferum
density or abundance at the OTA thus appears to be independent of any trend in precipitation over the time period of interest.

In weighing all of this information, we conclude that the best available evidence suggests that
Lepidium papilliferum
numbers may be trending downward. The dataset from the rough census areas on the OTA shows a significant downward trend in density over the last 18 years. Furthermore, we believe it is reasonable to infer that this negative trend may be similar or possibly even greater rangewide in areas outside the high quality habitat of the OTA, and this trend appears to be independent of any trend in precipitation. The best available scientific and commercial data therefore suggest that over the past two decades,
L. papilliferum
has likely significantly declined in abundance.

In terms of projecting this trend into the future, however, there are many uncertainties associated with both the data and the model that preclude our ability to do so; these include, but are not limited to: Great annual variability in plant numbers, the confounding influence of the long-lived seed bank, the complications associated with annual variability in both precipitation and temperature, and the inconsistent results between the special-use plots and the rough census areas on the OTA. The evaluation of Sullivan and Nations was based on a simple model of
Lepidium papilliferum
abundance or density as a linear function of time, and intended only to discern whether there was any general trend in the population. The authors acknowledge that the dynamics are complicated, and note their model is not intended to describe (nor explain) the details of the temporal pattern of abundance or density of
L. papilliferum
(Sullivan and Nations 2009, p. 38). In addition, we do not have any models for
L. papilliferum
based on multivariate analyses, which would simultaneously take into account additional variables such as precipitation, to potentially allow for the prediction of abundance or density of
L. papilliferum
over time based on projected conditions. Although the currently available model is helpful in terms of interpreting the population information available to date and indicates that
L. papilliferum
has likely been trending downward, for all of the reasons outlined above, it would be inappropriate to rely on this model to predict any future population trajectory for
L. papilliferum
.

Habitat Quality

As described above under “
Ecology and Habitat
,” the natural sagebrush-steppe community that surrounds the slickspot microsites in which
Lepidium papilliferum
occurs is dominated by sagebrush (primarily
Artemisia tridentata
ssp.
wyomingensis
) with a diverse understory of native perennial bunchgrasses and forbs. Historically, fires were relatively infrequent in this ecosystem, likely occurring on the order of every 100 years (Whisenant 1990, p. 4). Data on the plant community and fire history pattern are some of the habitat quality attributes collected as part of
Lepidium papilliferum
HIP monitoring, which has been conducted rangewide since 2004. Results from the 2008 HIP monitoring conducted at 80 HIP transects indicated that over the past 5 years, 14 of the transects (18 percent) that were initially characterized by predominantly native vegetation have undergone overall declines in habitat quality, primarily due to increased nonnative species cover (Colket 2009, pp. 10). Furthermore, this increase in nonnatives was observed not only in the surrounding plant community, but in the slickspots occupied by
L. papilliferum
as well.
Bromus tectorum
was the most common nonnative species in slickspots, followed by
Agropyron cristatum
(crested wheatgrass),
Ceratocephala testiculata,
formerly
Ranunculus testiculatus
(bur buttercup), and
Lepidium perfoliatum
(clasping-leaf pepperweed) (ICDC 2008, p. 9). Noxious or aggressive nonnatives detected in HIP transect slickspots include
Linum perenne
(‘Appar' blue flax),
Centaurea cyanus
(garden cornflower),
Bassia prostrata
(prostrate kochia or forage kochia),
Chondrilla juncea
(rush skeletonweed), and
Cardaria draba
(whitetop) (Colket 2009, pp. 8-9).

A review of the rangewide HIP transect data for evidence of fire history reveals that 38 of 80 HIP transects (48 percent) currently show no effects from wildfire and 6 others (7.5 percent) were predominantly unburned. Five transects (6.25 percent) had partially burned (with approximately half of the area unburned), 13 (16.25 percent) were predominantly burned, and 18 (22.5 percent) have completely burned (Colket 2009, Table 5). HIP classifies areas as burned if they are devoid of shrub cover or have patchy shrub cover in areas that exhibit the site capacity to support a healthy sagebrush-steppe community; this may include areas that have recently or historically burned. Four HIP transects were burned in 2007 in the Murphy Complex Fire in the Owyhee Plateau geographic region (Colket 2009, p. 23). Sixty-six of the 80 HIP transects (83 percent) have nearby wildfire effects within 1,640 ft (500 m) (Colket 2009, p. 26). A recent geospatial data analysis evaluating the total
Lepidium papilliferum
EO area affected by wildfire from 1957 to 2007 found that the perimeter of 107 wildfires that had occurred encompassed approximately 11,442 ac (4,509 ha), or 73 percent of the total EO area rangewide (Stoner 2009, p. 48). However, caution should be used in interpreting this geospatial information, as this represents relatively coarse vegetation information that may not reflect that some EOs may be located within remnant unburned islands of sagebrush habitat within fire perimeters.

Several features of slickspots and their surrounding habitat were consistently more degraded in areas that had burned. Slickspots in burned areas had lower soil crust cover and greater exotic (nonnative) species cover, and the total native species cover and shrub cover were consistently lower in burned transects, while total exotic species cover,, including
Bromus tectorum
, was consistently higher in burned transects (Menke and Kaye 2006b, p. 19). Sullivan and Nations (2009, p. 3) found a significantly negative relationship

between the abundance or density of
Lepidium papilliferum
and both the presence of
B. tectorum
and past fire. The positive association between the abundance of
B. tectorum
and fire frequency is well established (Whisenant 1990, p. 6). The complex and positive feedback loop between the encroachment of invasive annual grasses such as
B. tectorum
, increased fire frequency, and decreased integrity of biological soil crusts contributes to the degradation of sagebrush-steppe habitat quality for
L. papilliferum
(for additional details, see the
Modified Wildfire Regime
and
Invasive Nonnative Plant Species
discussions under Factor A of
Summary of Factors Affecting the Species
).

Element Occurrences Rangewide

The EO ranking system utilized by the INHP is described above in the
Monitoring of

Lepidium papilliferum

Populations
section. In brief, occurrences of
Lepidium papilliferum
are ranked based on measures of habitat quality and species abundance. The first EO ranks for
L. papilliferum
were assigned in 1993 (Colket
et al
. 2006, Tables 1-13). In 2006,
L. papilliferum
EO specifications and ranking were updated and revised by the ICDC to apply more consistent EO specifications rangewide (Colket
et al
. 2006, pp. 15-44). Due to the change in methods in 2006, EO rankings assigned before 2006 are not comparable to those assigned after 2006. Currently, EO ranks are more consistently assigned, are useful as an assessment of estimated viability or probability of persistence, and help prioritize conservation planning or actions (NatureServe 2002).

As of February 2009, the INHP has ranked 80 extant EO records for
Lepidium papilliferum
based on habitat quality and abundance (Cole 2009b, Threats Table). In addition, nine EOs are ranked as extirpated or probably extirpated, and seven EOs are considered historical (information is too vague for relocation of the sites). All nine extirpations were formerly verified locations from old herbarium collections (the most recent from 1955) where the habitat is now completely developed or converted to agricultural lands (Colket
et al
. 2006, Table 13). The 80 extant (as of February 2009) EOs represent a reduction in the number of extant EOs (85) known in 2006. However, this reduction in the number of EOs is due to the merging of EOs associated with new locations of plants rather than from the loss of individual EOs. As of February 2009, there are no A-ranked EOs for
L. papilliferum
; the most common EO ranks for
L. papilliferum
rangewide are C and D (Table 2). EO ranks also vary by physiographic region. A little more than one-half of the extant EO area in the Boise Foothills region is ranked as C, which means there are 50 to 399 above-ground plants, low to moderate introduced nonnative plant species cover, and EOs are partially burned. Approximately three-quarters of the total EO area in the Snake River Plain is ranked B, meaning there are 400 to 999 above-ground plants, the native plant community is intact with low introduced nonnative plant species cover, and EOs are largely unburned. The majority of the B-ranked EO acreage rangewide occurs on the Idaho Army National Guard's Orchard Training Area (OTA). The majority of the total EO area in the Owyhee Plateau physiographic region is also ranked B.

EO size can also influence the ranking of an EO as a percentage of total rangewide EO area. For example, one EO (number 27) located on the OTA in the Snake River Plain region has a total area of 7,163 acres (2,899 ha) and accounts for roughly 59 percent of all the area within
Lepidium papilliferum
EOs assigned a B rank throughout the entire range of the species. There are less than 2.2 ac (1 ha) of B-ranked area in the Boise Foothills region, and nearly 2,540 B-ranked ac (1,028 ha) on the Owyhee Plateau. Therefore, according to the EO rankings, the majority of the highest quality remaining habitat for
L. papilliferum
occurs on the Snake River Plain (see Table 2), with most of that occurring within the OTA.

Table 2. Extant Element Occurrence (EO) Ranks across the entire range of
Lepidium papilliferum

(INHP data from February 2009).

Element Occurrence Rank
No. EO's
Hectares
Acres
Percent of Area

Boise Foothills

B
1
0.84
2.07
1.65

BC
1
1.79
4.41
3.53

C
5
28.34
70.03
56.05

D
6
15.37
37.99
30.40

F
3
4.23
10.46
8.37

TOTAL

16

50.57

124.96

100.00

Snake River Plain

B
5
3,875.14
9,575.47
73.77

BC
1
1.42
3.51
0.03

C
19
935.06
2,310.53
17.80

D
12
350.44
865.94
6.67

D?
1
0.78
1.93
0.01

F
4
89.82
221.94
1.71

NR
1
0.20
0.48
0.00

TOTAL

43

5,252.86

12,979.81

100.00

Owyhee Plateau
1

B
5
1,027.50
2,537.00
96.02

C
4
21.85
53.99
2.04

D
5
18.42
45.52
1.72

E
0
0.00
0.00
0.00

F
7
2.36
5.83
0.22

TOTAL

21

1070.13

2,644.35

100.00

1
Note that Sullivan and Nations (2009, pp. 79-81) differed in their overview of extant EOs in the Owyhee Plateau as they presented EO 16 as each of its 27 individual sub-EOs (sub-EOs 700-726). Table 2 combines all Owyhee Plateau sub-EOs into the single EO 16 and also incorporates changes as described in the February 2009 INHP
Lepidium papilliferum
data.

Summary of Factors Affecting the Species

Section 4 of the Act and its implementing regulations (50 CFR 424) set forth the procedures for adding species to the Federal Lists of Endangered and Threatened Wildlife and Plants. A species may be determined to be an endangered or threatened species due to one or more of the five factors described in section 4(a)(1) of the Act: (A) The present or threatened destruction, modification, or curtailment of its habitat or range; (B) overutilization for commercial, recreational, scientific, or educational purposes; (C) disease or predation; (D) the inadequacy of existing regulatory mechanisms; or (E) other natural or manmade factors affecting its continued existence. Listing actions may be warranted based on any of the above threat factors, singly or in combination. Each of these factors relevant to
Lepidium papilliferum
is discussed below.

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

Several threat factors are contributing to the destruction, modification, or curtailment of
Lepidium papilliferum
's habitat or range. The sagebrush-steppe habitat of the Great Basin where
L. papilliferum
occurs is becoming increasingly degraded due to the impacts of multiple threats, including the invasion of nonnative annual grasses, such as
Bromus tectorum
, and increased frequency of fire. As described below,
B. tectorum
can impact
L. papilliferum
directly through competition, but also indirectly by providing continuous fine fuels that contribute to the increased frequency and extent of wildfires. Frequent wildfires have numerous negative consequences in the sagebrush-steppe system, which is adapted to much longer fire-return intervals, ultimately resulting in the conversion of the sagebrush community to nonnative annual grasslands, with associated losses of native species diversity and natural ecological function. Because the modified wildfire regime and invasion of
B. tectorum
create a positive feedback loop, it is difficult to separate out the effects of each of these threat factors independently. We have attempted to do so here, but much of the discussion may overlap due to the inherent synergism between these two threat factors.

In addition to wildfire and nonnative plants, development poses a threat to
Lepidium papilliferum
, both directly through the destruction of populations and loss of slickspot microsites, and indirectly through habitat fragmentation and isolation (discussed separately under Factor E, below). The loss of slickspots is a permanent loss of habitat for
L. papilliferum
, since the species is specialized to occupy these unique microsite habitats that were formed in the Pleistocene, and once lost, slickspots cannot be recreated on the landscape.

Livestock pose a threat to
Lepidium papilliferum
, primarily through mechanical damage to individual plants and slickspot habitats. However, the current livestock management conditions and associated conservation measures address this potential threat such that it does not pose a significant risk to the viability of the species as a whole.

All of these threats have long been recognized as contributing to the ongoing degradation of the sagebrush-steppe ecosystem of southwestern Idaho. However, we have only recently received independent evaluations of the direct relationship between the more significant threats and indicators of population viability specifically for
Lepidium papilliferum
. New evidence suggests that there is a significant negative association between cover of nonnative plant species and wildfire and the abundance of
L. papilliferum
, such that the species appears to be in decline across its range, with adverse impacts continuing and likely increasing into the foreseeable future. Each of the threat factors contributing to the present or threatened destruction, modification, or curtailment of
L. papilliferum
's habitat or range is assessed in detail below.

Modified Wildfire Regime

Fire was historically infrequent in the desert shrublands of the Great Basin, as

the native plant communities of the native annuals and bunchgrasses did not provide sufficient fine fuels to carry large scale wildfires. The bare spaces between widely spaced shrubs and relatively low fuel loads in such ecosystems as the sagebrush-steppe generally prevented fires from spreading very far, and any fires that did burn were usually restricted to relatively small, isolated patches (Brookes and Pyke 2001, p. 5; Whisenant 1990, pp. 4, 6). Natural fire return intervals in sagebrush-steppe prior to the arrival of European settlers are estimated to have ranged from 60 to 110 years; the estimate for the more xeric
Artemisia tridentata
ssp.
wyomingensis
sagebrush community inhabited by
Lepidium papilliferum
is estimated to have been as long as 100 years (Wright and Bailey 1982, p. 158) and possibly up to 240 years (Baker 2006, p. 181). Beginning in the early 1900s, however, the widespread invasion of nonnative plant species, particularly annual grasses such as
Bromus tectorum
and
Taeniatherum caput-medusae
, has created a bed of continuous fine fuels across the southwest Idaho landscape. The continuous fine fuels provided by these nonnative annual grasses result in more frequent fires due to greater horizontal fuel continuity, increased fuel surface-to-volume ratio, and various properties that facilitate wildfire ignition, such as lower moisture content and thus increased flammability (Whisenant 1990, p. 6; Pellant 1996, p. 3 and references therein; Brooks
et al
. 2004a, p. 679). Nonnative annual grasses also provide for more continuous and uniform fires, burning across extensive areas of the landscape. Native bunchgrasses provide a patchy, discontinuous fuelbed such that fires are not easily carried and tend to burn only in small patches. The continuous fires carried by nonnative annual grasses such as
B. tectorum
, on the other hand, leave few or no patches of unburned vegetation, which can inhibit the post-fire recovery of native sagebrush-steppe vegetation by eliminating seed sources for regrowth of the native species (Whisenant 1990, p. 4; Pyke 2007).
Bromus tectorum
, in particular, apparently alters the soil environment such that it creates a positive feedback loop, enhancing the environment for its own growth and generating conditions conducive to further invasion (Pyke 2007). As
B. tectorum
has become more dominant in the sagebrush-steppe habitat of the Snake River Plain over the past several decades, wildfire frequency intervals have become shortened from the historical average of 60 to 110 years to the current frequency intervals of 5 years or less (Wright and Bailey 1982, p. 158; Billings 1990, pp. 307-308; Whisenant 1990, p. 4; USGS 1999; West and Young 2000, p. 262; Launchbaugh
et al
. 2008, p. 3; Zouhar
et al
. 2008, pp. 40-41).

The dramatic increase in the frequency of wildfires has a particularly negative effect on the native plant community in this region that has historically experienced fire relatively infrequently, and thus is dominated by plants that are not adapted to short fire-return intervals. Many of the native species of the sagebrush-steppe ecosystem are killed outright by wildfires and do not have adaptations such as underground rhizomes for post-fire vegetative regrowth, but must reproduce by seed. As a result, under a regime of increasingly frequent fire, perennial plants tend to be lost from the landscape (Whisenant 1990, p. 9). Sagebrush (
Artemisia
spp.), for example, are easily killed by fire (Baker 2006, p. 178 and references therein; Cooper
et al
. 2007, p. 8; USDA Forest Service Fire Effects Information System 2009). Because they are not adapted to frequent fires, sagebrush does not resprout after burning, as many fire tolerant species do (Young and Evans 1978, pp. 283, 287; Brooks and Pyke 2001, pp. 6-7; USDA Forest Service Fire Effects Information System 2009), but must rely upon seed sources for reestablishment. Natural revegetation requires a nearby remnant seed source, as from an unburned patch of sagebrush, which now rarely occurs because of the more continuous and extensive fires that occur if a
B. tectorum
understory is present (USDA Forest Service Fire Effects Information System 2009). In addition, when fires occur as frequently as every 3 to 5 years, even if seedlings should begin to grow there is not sufficient time for sagebrush to regenerate prior to the next fire cycle. Thus, sagebrush is eliminated from the plant community, which in turn allows for conversion to annual grassland (Whisenant 1990, p. 9; Pyke 2007; USDA Forest Service Fire Effects Information System 2009). The short fire-return intervals now experienced in this region prevent the sagebrush-steppe community from recovering and attaining late seral stage condition, thus eliminating high quality habitat for
L. papilliferum
.

The dramatic increase in frequency and extent of wildfires has contributed to the conversion of vast areas of sagebrush-steppe into invasive annual grasslands (USGS 1999). Since post-fire conditions are favorable for further invasion and establishment of nonnative annual grasses, invasive grasses soon dominate the community, leading to the establishment of an invasive grass-increased fire frequency cycle (Whisenant 1990, p. 4; Brooks and Pyke 2001, p. 5; D'Antonio and Vitousek 1992, pp. 73, 75; Brooks
et al
. 2004a, p. 678). Invasive grasses promote recurrent fires, which in turn convert high diversity native shrublands to low diversity alien grasslands; these grasslands then burn more frequently and expansively across the landscape, creating disturbance conditions that promote the further expansion of the invasive grasses, and so on. This invasive grass-fire cycle has been recognized in Great Basin shrub ecosystems since the 1930s (Brooks and Pyke 2001, p. 5, and references therein). As an example, at the Snake River Birds of Prey National Conservation Area in the Snake River Plain area of southern Idaho, nearly half of the native sagebrush-steppe habitat (a total of 494,211 ac (200,000 ha)) converted to nonnative annual grasslands in less than 10 years by a series of 200 fires (Smith and Collopy 1998, as cited in Brooks and Pyke 2001, p. 7).

The rate of conversion from sagebrush-steppe to annual grasslands continues to accelerate in the Snake River Plain of southwest Idaho (Whisenant 1990, p. 4). As the coverage of
Bromus tectorum
continues to increase in the region, it is reasonable to expect that the extent and frequency of wildfires will likewise continue to increase, given the demonstrated positive feedback cycle between these factors (Whisenant 1990, p. 4; Brooks and Pyke 2001, p. 5; D'Antonio and Vitousek 1992, pp. 73, 75; Brooks
et al
. 2004a, p. 678). Climate change models also project a likely increase in fire frequency within the semiarid Great Basin region inhabited by
Lepidium papilliferum
(see
Climate Change
under Factor E, below).

Wildfire therefore contributes to the continuing invasion and establishment of nonnative annual grasslands within the range of
Lepidium papilliferum
, which in turn further increases the likelihood of more frequent and intense wildfires across the range of the species (Brooks
et al
. 2004a, pp. 677-687). But wildfire's role in promoting the invasion of annual grasses goes beyond its circular positive impact on the fire cycle, as nonnative annual grasses and other nonnative plant species that are likely to invade following fire have numerous other negative effects on
L. papilliferum
, slickspots, and the surrounding sagebrush-steppe ecosystem as well, as described below under
Invasive Nonnative Plant Species.

Wildfire also damages biological soil crusts, which are important to the sagebrush-steppe ecosystem and slickspots where
Lepidium papilliferum
occur, because the soil crusts stabilize and protect soil surfaces from wind and water erosion, retain soil moisture, discourage annual weed growth, and fix atmospheric nitrogen (Eldridge and Greene 1994 as cited in Belnap
et al
. 2001, p. 4; Johnston 1997, pp. 8-10; Brooks and Pyke 2001, p. 4). Fires can cause severe damage to soil crusts, altering their ecological function and creating an opportunity for invasion by weedy annual plant species (Johnston 1997, p. 10; Brooks and Pyke 2001, p. 4, and references therein). In a statistical analysis of HII and HIP data between 1998 and 2004, burned areas had less soil crust cover and higher nonnative plant cover (Menke and Kaye 2006b, p. 3). In general,
L. papilliferum
abundance is greatest in areas that also have the greatest cover of soil crust (Boise Foothills and Snake River Plain), although the populations in the Owyhee Plateau contrasted in showing a slightly negative (but not statistically significant) relationship with soil crust cover (Sullivan and Nations 2009, p. 135). Fire in the presence of shrubs, particularly sagebrush, tends to be greater in intensity, which decreases the potential for soil crust recovery (Johnston 1997, p. 11); therefore, recovery of these crusts after a fire is less likely in the sagebrush-steppe habitat where
L. papilliferum
occurs. Given the generally positive association between soil crust cover and
L. papilliferum
, the compromised integrity of the microbiotic crust in response to fire likely has a negative impact on
L. papilliferum
as well.

More frequent wildfires also promote soil erosion and consequent sedimentation, as perennial grasses that normally limit erosion are eliminated in arid environments such as the sagebrush-steppe ecosystem (Bunting
et al
. 2003, p. 82). Increased sedimentation can result in a silt layer that is too thick for optimal
Lepidium papilliferum
germination (Meyer and Allen 2005, pp. 6-7). Wind erosion following wildfire can also remove the top silt layer of slickspots, exposing the clay vesicular layer below, as observed at HIP transect 721 following the 2007 Murphy Complex Fire (U.S. BLM 2007, p. 23). However, effects of the loss of the upper slickspot silt layer on
L. papilliferum
are not known.

The threats of wildfire and nonnative invasive species working in concert are considered the predominant factor affecting
Lepidium papilliferum
, particularly its habitat quality. In a statistical analysis of HII data over 5 years between 1998 and 2001, areas that had burned earlier in the study and were left with depleted shrub and soil crust did not recover (Menke and Kaye 2006a, p. iii). Burned areas had less native plant cover, greater nonnative plant cover, increased slickspot perimeter compromise (the slickspot boundaries lose definition), and increased organic debris accumulation (Menke and Kaye 2006a, p. iii). As mentioned above, analysis of additional HII and HIP data from 1998 through 2004 showed that burned areas had less soil crust cover and greater nonnative plant cover (Menke and Kaye 2006b, p. 3). Past wildfires thus appear to have had a lasting negative impact on the plant community surrounding slickspots, including increased nonnative species cover and decreased soil crust cover (Menke and Kaye 2006b, p. 19). Although we recognized wildfire as one of the primary threats affecting the matrix habitat of
L. papilliferum
in our 2007 finding, at that time we did not have any data that directly tied wildfire with a negative impact on the species itself, as would be demonstrated, for example, by a corresponding decline in
L. papilliferum
abundance (72 FR 1622, 1635; January 12, 2007).

As discussed above, several researchers have noted signs of increased habitat degradation for
Lepidium papilliferum
, most notably in terms of exotic species cover and wildfire frequency (e.g., Moseley 1994, p. 23; Menke and Kaye 2006b, p. 19; Colket 2008, pp. 33-34), but only recently have analyses demonstrated a statistically significant negative relationship between the degradation of habitat quality, both within slickspot microsites and in the surrounding sagebrush-steppe matrix, and the abundance of
L. papilliferum
. Sullivan and Nations (2009, pp. 114-118, 137) found a consistent, statistically significant negative correlation between wildfire and the abundance of
L. papilliferum
across its range. Their analysis of 5 years of HIP monitoring data indicated that
L. papilliferum
“abundance was lower within those slickspot (sic) that had previously burned” (Sullivan and Nations 2009, p. 137), and the relationship between
L. papilliferum
abundance and fire is reported as “relatively large and statistically significant,” regardless of the age of the fire or the number of past fires (Sullivan and Nations 2009, p. 118). The nature of this relationship was not affected by the number of fires that may have occurred in the past; whether only one fire had occurred or several, the association with decreased abundance of
L. papilliferum
was similar (Sullivan and Nations 2009, p. 118).

The evidence also points to an increase in the geographic extent of wildfire within the range of
Lepidium papilliferum
. Since the 1980s, 59 percent of the total
L. papilliferum
management area acreage rangewide has burned, more than double the acreage burned in the preceding three decades (from the 1950s through 1970s). Based on available information, approximately 11 percent of the total management area burned in the 1950s; 1 percent in the 1960s; 15 percent in the 1970s; 26 percent in the 1980s; 34 percent in the 1990s; and as of 2007, 11 percent in the 2000s (data based on GIS fire data provided by BLM Boise and Twin Falls District; I. Ross 2008, pers. comm. and A. Webb 2008, pers. comm., as cited in Colket 2008, p. 33). Based on the negative relationship observed between fire,
L. papilliferum
, and habitat quality as described above, we conclude that this increase in area burned translates into an increase in the number of
L. papilliferum
populations subjected to the negative impacts of wildfire.

An evaluation of
Lepidium papilliferum
EOs for which habitat information has been documented (79 of 80 EOs) demonstrates that most have experienced the effects of fire. Fifty-five of 79 EOs have been at least partially burned (14 of 16 EOs on the Boise Foothills, 30 of 42 EOs on the Snake River Plain and 11 of 21 EOs on the Owyhee Plateau), and 75 EOs have adjacent landscapes that have at least partially burned (16 of 16 EOs on the Boise Foothills, 39 of 42 EOs on the Snake River Plain, and 20 of 21 EOs on the Owyhee Plateau) (Cole 2009b, Threats Table).

In 2008, 38 of the 80 HIP transects were unburned, 6 were predominantly unburned, 5 approximately half burned and half unburned, 13 were predominantly burned, and 18 were completely burned. Sixty-six HIP transects had been at least partially burned to within 1,500 ft (500 m) (Colket 2009, p. 26). In 2007, the Inside Desert Fire on the Owyhee Plateau burned 2,695 ac (1,041 ha) within Management Area 11, and the Elk Mountain Fire burned 11,868 ac (4,083 ha) within Management Area 11; both fires were part of the 652,016 ac (263,862 ha) Murphy Complex Fire in the Owyhee Plateau region (Colket 2009, p. 65). In 2008, the first year of HIP monitoring following the fire was completed in the four transects (Transects 701, 711, 719, and 721) that burned in the Murphy Complex Fire

(Colket 2009, p. 24). All 10 slickspots at HIP transect 701 had been previously burned before being burned again in 2007. At HIP transect 711, only 1 slickspot had been previously burned, but 9 of its 10 slickspots were burned in the Murphy Complex Fire. HIP transects 719 and 721 were completely unburned high quality big sagebrush habitat before the Murphy Complex Fire burned all 10 slickspots at both HIP transects (Colket 2009, p. 24).

A 2009 geospatial data analysis evaluating the total
Lepidium papilliferum
EO area affected by wildfire from 1957 to 2007 found that 107 wildfires have occurred, the fire perimeters of which included approximately 11,442 ac (4,509 ha), or 73 percent of the total EO area (Stoner 2009, p. 48).

Table 3 shows the evidence of wildfires documented through HIP rangewide transect monitoring in 2008 and includes both recent and historical fires. Wildfire evidence can remain on the landscape for up to 20 years.

Table 3. Evidence of Wildfire Documented at HIP Transects in 2008 (Colket 2009, Table 5, pp. 50-62).

Physiogeographic Region
Number of HIP transects at least partially burned
Number of HIP transects not burned
Total HIP transects

Adjacent landscapes
within 0.31 miles (500
meters) of HIP
transects either burned or partially burned

Boise Foothills
7
3
10
10

Snake River Plain
21
26
47
38

Owyhee Plateau
14
9
23
19

TOTAL

42 (52.5 percent)

38 (47.5 percent)

80 (100 percent)

67 (84 percent)

The effects of fire disturbance and habitat degradation are evident in some of the earliest photographs of HII and HIP transects, which show habitats lacking shrubs and dominated by
Bromus tectorum
. However, photographs from the early 1990s of transects that had not burned prior to being established were comprised primarily of native
Artemisia tridentata
with a nonnative
B. tectorum
or
Ceratocephala testiculata
understory. As of 2008, 14 of 80 total HIP transects had changed from a higher to a lower habitat quality classification since 2004, or had been partially or completely burned (Colket 2009, pp. 8-9). The photographs demonstrate that many of the transects that burned are now devoid of
A. tridentata
and are instead dominated by
B. tectorum
(Colket 2009, pp. 63-64).

At present, ongoing control efforts may slow the incidence of wildfire in some areas, but are not capable of preventing wildfires across the range of
Lepidium papilliferum
. For example, four established HIP transects on the Owyhee Plateau burned in 2007 in the Inside Desert and Murphy Complex fires, even though wildfire control measures were in place and implemented (Colket 2009, p. 24). In the Snake River Plain region, portions of two EOs (EO 32, EO 26) were burned in 2006 by the Ten York Fire and Cold Fire respectively. No EOs or portions of known EOs are documented to have burned in the Snake River Plain and Boise Foothills regions in 2007 (U.S. BLM 2008a, p. 21). On the OTA, the IDARNG has demonstrated intensive management efforts implemented to suppress wildfire and using wildfire-rehabilitation activities with minimal ground disturbance have been effective in reducing the threat of wildfire and the rate of spread of nonnative invasive species (for additional information, see
Wildfire Management and Post-Wildfire Rehabilitation
section below). However, such intensive management is currently concentrated within
L. papilliferum
EOs and is possible only within a limited range of
L. papilliferum
. This may explain why the highest quality habitat remaining is on the OTA, where the greatest infrastructure is in place to manage and control wildfires.

Summary of Modified Wildfire Regime

The observed increases in frequency and geographic extent of wildfires, the negative consequences for
L. papilliferum
and its habitat associated with the invasion of nonnative grasses and wildfire, the strong positive feedback loop between wildfire and conversion of sagebrush-steppe to annual grasslands, and the lack of effective rangewide control mechanisms all contribute to the current modified wildfire regime being the greatest ongoing threat to
L. papilliferum
's existence. In addition, the best available data indicates that fire frequency is likely to increase in the foreseeable future due to increases in cover of
B. tectorum
and the projected effects of climate change (see
Invasive Nonnative Plant Species
, below, and also
Climate Change
under Factor E, below). Ongoing habitat loss and degradation is a result of the current wildfire regime, which is interrelated with several other negative factors, including: Increased nonnative species cover, especially annual grasses; increased sedimentation and organic debris accumulation in slickspots, which could alter slickspot function and hinder germination of
L. papilliferum
; the loss of native matrix vegetation, particularly shrubs; decreased native plant species diversity; decreased cover of microbiotic crusts; and habitat fragmentation due to isolation of habitat patches following fire.

Given the observed negative association between the abundance of
Lepidium papilliferum
and the increased frequency of fire, as well as the demonstrated negative impacts of frequent fire on the components that normally provide high quality habitat for
L. papilliferum
, such as late seral stage sagebrush and high microbiotic crust cover, we consider the current wildfire regime to pose a significant threat to
L. papilliferum.
Recurrent fire promotes the continued invasion of nonnative annual grasses and other invasive nonnative plants, along with all of their associated negative effects (see
Invasive Nonnative Plant Species
below). Based on the observed increases in the cover of
Bromus tectorum
throughout the range of the species, the lack of effective control mechanisms, and projections under most climate change models, we expect the degree of this threat will continue and likely increase within the foreseeable future. The significant threat posed by the current modified wildfire regime is pervasive throughout the range of the species.

Invasive Nonnative Plant Species

Invasive nonnative plants have become established in
Lepidium papilliferum
habitats by spreading through natural dispersal (unseeded) or have been intentionally planted as part of revegetation projects (seeded). Invasive nonnative plants can alter multiple attributes of ecosystems, including geomorphology, wildfire regime, hydrology, microclimate, nutrient cycling, and productivity (Dukes and Mooney 2003, pp. 1-35). They can also negatively affect native plants through competitive exclusion, niche displacement, hybridization, and competition for pollinators; examples are widespread among native taxa and ecosystems (D'Antonio and Vitousek 1992, pp. 63-87; Olson 1999, p. 5; Mooney and Cleland 2001, p. 1). Geospatial analyses indicate that approximately 20 percent of the total area of all
L. papilliferum
EOs rangewide is dominated by introduced invasive annual and perennial plant species (Stoner 2009, p. 81), and monitoring of HIP transects rangewide indicates that nonnative plant cover is continuing to increase at a relatively rapid pace (Colket 2008, pp. 1, 3). Although, historically, disturbance of native communities tended to pave the way for invasion by nonnative plants, today nonnative annual plants such as
Bromus tectorum
are so widespread that they have been documented spreading into areas not impacted by disturbance (Piemeisel 1951, p. 71; Tisdale
et al
. 1965, pp. 349-351; Stohlgren
et al
. 1999, p. 45). The known impacts of nonnative plants on
L. papilliferum
are discussed in this section.

One of the characteristics of slickspots is that they are largely devoid of native shrubs, grasses, and forbs, with the exception of
Lepidium papilliferum
; this is one of the features that make slickspots relatively easy to detect on the landscape (Moseley 1994, pp. 8, 14; Fisher
et al
. 1996, pp. 3-4, 11; Colket 2008, p. 1).
Lepidium papilliferum
has adapted to the unique edaphic and hydrological (soil and water) properties of the slickspot microsites that it inhabits, and has thus evolved with little competition from other native plants (Moseley 1994, p. 14). Weedy, nonnative plants have begun to invade these slickspots, however, including
Agropyron cristatum, Bromus tectorum
,
Lepidium perfoliatum, Ceratocephala testiculata
, and, in some areas,
Bassia prostrata
(Colket 2009, p. 3; Fisher
et al
. 1996, p. 4; Sullivan and Nations 2009, p. 99).

In our January 12, 2007, finding (72 FR 1622), we recognized invasive nonnative plants as one of the primary factors degrading the quality of
L. papilliferum
's habitat, but at the time we had no evidence demonstrating any negative association between the presence of nonnative plant species and either the abundance of
L. papilliferum
itself or the proportion of
L. papilliferum
in flower. For example, Menke and Kaye (2006b, p. 15) originally reported no correlation between the abundance of
L. papilliferum
and weedy species cover, either within slickspots or in the surrounding matrix vegetation. However, more recent analyses of the additional years of data now available have revealed a significant negative association between the presence of weedy species and the abundance or density of
L. papilliferum
, to the point that
L. papilliferum
may be excluded from slickspots (Sullivan and Nations 2009, pp. 109-112). Although the specific mechanisms are not well understood, some of these plants, such as
A. cristatum
and
B. tectorum
, are strong competitors in this arid environment for such limited resources as moisture, which tends to be concentrated in slickspots (Pyke and Archer 1991, p. 4; Moseley 1994, p. 8; Lesica and DeLuca 1998, p. 4), at least in the subsurface soils (Fisher
et al
. 1996, pp. 13-16). The available information, detailed below, indicates that nonnative plants in both slickspots and the surrounding matrix vegetation are negatively affecting
L. papilliferum.
Furthermore, we now have additional evidence that areas occupied by
L. papilliferum
formerly dominated by native vegetation are experiencing relatively rapid increases in cover of nonnative plant species; for example, Colket (2008, pp. 1, 3) reports that 22 of the 80 HIP transects (28 percent) have shown increases in nonnative plant species cover of 5 percent or more over the last 4 to 5 years. Here we discuss the effects of nonnative plant species on
L. papilliferum
and its habitat, detailing the evidence related to unseeded and seeded nonnative plants separately.

Unseeded Nonnative Invasive Plants

The most common unseeded nonnative annual grasses known to occur in
Lepidium papilliferum
's habitat include
Bromus tectorum
and
Taeniatherum caput-medusae
. Annual nonnative forbs now commonly associated with slickspots include
Lepidium perfoliatum, Salsola kali
(tumbleweed, also known as Russian thistle),
Sisymbrium altissimum
(tumble mustard, also known as tall tumble mustard), and
Ceratocephala testiculata
(Colket 2009, pp. 8-9).

As discussed under
Modified Wildfire Regime
above,
Bromus tectorum
in particular has become dominant in many sagebrush-steppe habitat areas during the last century due to livestock grazing, agriculture, and wildfire impacts (Pickford 1932, p. 165; Piemeisel 1951, p. 71; Peters and Bunting 1994, p. 34; Vail 1994, pp. 3-4; Brooks and Pyke 2001, pp. 4-6). Vast areas of sagebrush shrublands have been converted to
B. tectorum
in the past century (about 31,000 mi
2
(80,000 km
2
) in the Great Basin alone) (Menakis
et al
. 2003, p. 284). Low-elevation sites, which are relatively dry and experience wide variation in soil moisture, appear to be more vulnerable to
B. tectorum
invasion than higher elevation sites with more stable soil moisture.
Bromus tectorum
plants tend to be larger and more fecund in a post-wildfire environment than on unburned sites, potentially leading to subsequent increases in density on burned sites under favorable climatic conditions (Zouhar 2003a, as summarized in Zouhar
et al
. 2008, p. 154). The invasion of nonnative plant species, particularly annual grasses, has had a greater effect on the lower elevation sagebrush shrublands in the Snake River Plain of Idaho that historically experienced less frequent fire than higher elevation sites in the region; the higher elevation sites have higher precipitation and historically had more fine grasses and more frequent wildfires (Gruell 1985, pp. 103-104; Peters and Bunting 1994, p. 33). These lower elevation sagebrush shrublands include the range of
Lepidium papilliferum
. As detailed under
Modified Wildfire Regime
, above, the
B. tectorum-
fire cycle modifies and degrades the native sagebrush-steppe ecosystems on which
L. papilliferum
depends, and recurrent fire prevents the system from achieving the late seral stage condition that characterizes high-quality habitat for the species.

In addition to perpetuating the cycle of increased wildfire within the range of
Lepidium papilliferum
, nonnative plants such as
Bromus tectorum
and
Taeniatherum caput-medusae
can have additional negative impacts on
L. papilliferum
through competition, displacement, and altering the ecological function of slickspots. Invasive grasses can replace native plants such as
L. papilliferum
by outcompeting them for resources, such as soil nutrients or moisture (Brooks and Pyke 2001, p. 6, and references therein).
Bromus tectorum
in particular appears to displace native plants by prolific seed production, early germination, and superior competitive abilities for the

extraction of water and nutrients (Pellant 1996, pp. 3-4; Pyke 2007). In addition,
B. tectorum
is capable of modifying the ecosystems by altering the soil temperatures and soil water distribution (Pellant 1996, p. 4). Evidence that
B. tectorum
is likely displacing
L. papilliferum
is provided by Sullivan and Nations' (2009, p. 135) statistical analyses of
L. papilliferum
abundance and nonnative invasive plant species cover within slickspots. Working with 5 years of HIP data collected from 2004 through 2008, Sullivan and Nations found that the presence of other plants in slickspots, particularly invasive exotics such as
Bassia prostrata
and
Bromus tectorum
, was associated with the almost complete exclusion of
L. papilliferum
from those microsites (Sullivan and Nations 2009, pp. 111-112). Of all the factors considered in their analysis, only the amount of
B. tectorum
in the plant community around slickspots showed a consistent relationship with the abundance of
L.

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

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3AE9-24039. Public record. Not legal advice.