POPULATION STATUS OF THE RELIC LEOPARD FROG (RANA ONCA)

Bradford, D.F.¹, and R.D. Jennings <sup>2, 3

1</sup>U.S. Environmental Protection Agency
National Exposure Research Laboratory
Las Vegas, NV

²Marjorie Barrick Museum of Natural History
University of Nevada, Las Vegas
Las Vegas, NV

³Department of Natural Science
Western New Mexico University
Silver City, NM
(current address)

The relic leopard frog (Rana onca) of southern Nevada and southwestern Utah was thought to have become extinct in the 1950s. In 1991, however, leopard frogs in springs along the Overton Arm of Lake Mead, Nevada were determined to be R. onca based on morphological analyses. The objectives of this study were: (1) to determine the distribution of leopard frog populations in the lower Virgin River/Lake Mead area inclusive of the historic range of Rana onca, and (2) to evaluate the stability of populations of leopard frogs at the Overton Arm sites.

In 1991-1993 we surveyed 12 historic and 39 other localities for leopard frogs. Leopard frogs were found only at the three Overton Arm sites within an area of 8 km², and at a site near Littlefield, Arizona, encompassing <0.1 km². Between mid 1993 and 1996, we counted frogs approximately 6-7 times per year at two of the three Overton Arm sites. Numbers of frogs remained stable at Blue Point Spring (a large site), whereas the population went extinct in 1994 at Corral Spring (a small site). Extinction occurred concomitantly with encroachment of emergent vegetation into pools. Based on mark/recapture data for 96 adults in 1995-96 at Blue Point Spring, the estimated number of adults at a given time averaged 36 frogs (95% conf. limits, 27-45) over 555 m of stream habitat. Assuming a similar density elsewhere at the Overton Arm and Littlefield sites, we estimate the total number of Virgin River leopard frogs to be a few hundred adults.

ADDENDUM: We recently found leopard frogs in two springs approximately 4 km downstream from Hoover Dam. The taxonomic status of these frogs is under investigation.

FACTORS AFFECTING ALIEN ANNUAL PLANT ABUNDANCE AT A SITE IN THE WESTERN MOJAVE DESERT: EFFECTS OF HUMAN DISTURBANCE, MICROHABITAT, TOPOGRAPHY, AND RAINFALL.

Matthew L. Brooks

United States Department of the Interior
United States Geological Survey, Biological Resources Division
Western Ecological Research Center, Box Springs Field Station
41734 South Fork Dr. Three Rivers CA 93271
phone/fax: 559-561-6511
matt_brooks@usgs.gov

Four alien weeds (the grasses Bromus madritensis subsp. rubens, Bromus trinii, and Schismus spp, and the forb Erodium cicutarium) have invaded the Mojave Desert. To identify conditions that have promoted their spread, particularly those related to soil nutrients and human disturbances, I correlated temporal and spatial variation in their biomasses with environmental factors at the Desert Tortoise Research Natural Area (DTNA). Sampling was done during one low and one high rainfall year (49 and 209% of average) and stratified to sample all combinations of the following factors: (1) topographic position (upland-low nutrient, washlet-high nutrient); (2) microhabitat created by creosote bush (Larrea tridentata) (intershrub-low nutrient, Larrea-south canopy-medium nutrient, and Larrea-north canopy-high nutrient); and (3) human disturbance, in the combined form of sheep grazing and off-highway vehicle use (inside the DTNA-low disturbance, outside the DTNA-high disturbance).

Combined biomass of aliens was generally highest where and when soil nutrients were most abundant: in washlets, under creosote bushes, and during a year of high rainfall. Biomass of aliens was less affected by human disturbance, but was slightly higher in its presence outside the DTNA. Because this site was protected from disturbance for only 14 years prior to the beginning of this study, the contrast between disturbance levels was relatively small and larger disturbance gradients in the Mojave Desert may have greater affects on alien plant dominance.

Biomasses of individual species showed distinct patterns that often differed from each other. Biomasses of Bromus madritensis subsp. rubens and Bromus trinii were highest in the Larrea-north microhabitats during both years. In contrast, peak biomass of Schismus spp. and Erodium cicutarium differed between years, with greatest amounts in the Larrea-south microhabitat during the high-rainfall year and in intershrub spaces during the low-rainfall year. Apparently, Bromus spp is more limited by low soil nutrient levels than Schismus spp. and Erodium cicutarium are. These results highlight the importance of understanding the ecology of individual species in addition to patterns of overall alien dominance in managing alien plants.

Although absolute biomass of aliens was highest during the wet year, relative biomass of aliens was highest during the dry year. Annual plants that are alien to the Mojave Desert may have less stringent germination requirements than those that are native, which should cause the annual plant seedling cohort to be dominated by aliens during dry years. Germination and subsequent death of pre-reproductive alien plant seedling during dry years may explain why some alien annual plant populations decline following droughts in the Mojave Desert.

Two groups of alien species emerged in this study, one consisted of the very similar congeners Bromus madritensis subsp. rubens and Bromus trinii that were strongly associated with relatively high soil nutrient levels, and the other consisted of the very different monocot Schismus spp. and dicot Erodium cicutarium that were associated with a wider range of soil nutrients. In their regions of origin, Bromus madritensis subsp. rubens and Bromus trinii evolved in mesic ecosystems, whereas Schismus spp and Erodium cicutarium evolved in more arid ecosystems. All four species appear to have less stringent germination requirements than native species, but the apparent adaptations to arid conditions of Schismus spp. and Erodium cicutarium may allow them to survive drought periods better than the more mesic-adapted Bromus spp.

LANDSCAPE CHANGE OF THE LAS VEGAS VALLEY, 1972 TO 1998

Curtis M. Edmonds, Daniel T. Heggem, and Phillip Cuartas

U.S. Environmental Protection Agency
Office of Research and Development
Landscape Ecology Branch

Las Vegas has become one of the fastest growing metropolitan areas in the United States. The cities' population has doubled from 1980 to 1994 and in 1995 Las Vegas has surpassed the one million mark. The population of Las Vegas is currently growing at a rate of 7 percent annually. At this rate, the number of people will double again in ten years. Future growth may be limited by the availability of water due to it's location in the Mojave Desert. Water allocation and resource management will be a subject of great concern. Water usage and water quality can be linked to the landscape changes which occur in a region or watershed. It is possible to measure landscape change over a large area and determine the trends in ecological and hydrological condition. This method of analysis may provide an effective and economical way to evaluate watershed condition related to the disturbance caused by land use changes produced by human population growth. An example of this would be to analyze the explosive proliferation of development related vegetation in the Las Vegas Valley.

The watershed boundary for the Las Vegas Valley was defined. The pour point for the watershed was selected as the Las Vegas Wash where it enters Lake Mead. Precipitation which does not either percolate into the ground or evaporate into the air will eventually run off through this point into Lake Mead. The quality of the water entering Lake Mead is reflected in the type of land use found within this watershed boundary. This watershed boundary was created by using a 30 meter digital elevation model of the Las Vegas Valley and computer software designed to determine the flow, direction and accumulation of surface water.

The dramatic growth in the Las Vegas area can be seen in the multiple-date satellite imagery which uses Landsat Multi-Spectral Scanner imagery from the early 1970s, mid 1980s and early 1990s and Landsat Thematic Mapper imagery from 1998. The data from these images will be classified to determine the type, location and amount of land cover present in each year. The data can then be used to estimate the change which has occurred over 26 years. The data can also be used to forecast future development and how it may affect water availability and quality.

RIPARIAN RESTORATION IN THE MOJAVE DESERT: AN OVERVIEW OF SELECTED PUBLIC LAND SITES

Thomas B. Egan

U.S. Department of the Interior
Bureau of Land Management
2601 Barstow Road
Barstow CA 92311
tegan@ca.blm.gov

"Long enough in the desert a man like other animals can learn to smell water. Can learn, at least, the smell of things associated with water-the unique and heartening odor of the cottonwood tree....which in canyonlands is the tree of life. It signifies water, and not only water but shade, in a country where shelter from the sun is sometimes almost as precious as water."

Edward Abbey (1927-1989)

Riparian restoration is gaining momentum in the sands of the Mojave Desert, as the Bureau of Land Management continues its efforts at Afton Canyon, Amargosa Canyon, Harper Dry Lake and Salt Creek. These selected public land projects typify the techniques and issues common to restoration work on the three general types of Mojave riparian habitat, namely riverine, playa edge and desert spring habitats. Two projects representing the latter two riparian types, Harper Dry Lake and Salt Creek Projects, respectively, have been largely completed and focused on vehicle access management, perennial weed (saltcedar or Tamarix ramosissima) removal, minor native tree planting, wildlife "snag" installation, as well as pond construction and water source channeling at the playa edge site. Additional water procurement and well installation for the purposes of marsh hydration will be necessary at the playa edge site and low-level weed maintenance will be required at both sites. Work continues on the two large riverine-canyon sites, with the Afton Canyon Project roughly half completed and the Amargosa Canyon Project just beginning. The focus at both canyon sites is initial perennial weed (saltcedar) removal, low-level maintenance of previously treated weed areas and native tree planting. Techniques used in these efforts include an integrated weed management approach, utilizing manual cutting, controlled fire and herbicide application; a vehicle access management program entailing road closure, road re-routing and protective fencing; a livestock management program employing the use of cattle fence exclosures, water gaps and trespass cattle removal; a labor component emphasizing employment of inner city youth crews and Resource Conservation District personnel, who supervise prison inmate crews, as well as the use of interns and volunteers; and a revegetation component entailing native grass, shrub and tree plantings. A persistant funding grant submission program, specifically assigned project teams, as well as regular professional conference participation by team members, formed the foundation upon which these endeavors could be accomplished. Issues common to these restoration efforts are discussed and include upstream water regime alteration and excessive groundwater pumping; mixed land ownership patterns and upstream weed seed sources on private lands; securing year-long labor sources and simultaneous coordination of numerous work crews; as well as long-term project maintenance needs.

INTERAGENCY COOPERATION AND A LARGE-SCALE HABITAT MANIPULATION: THE EFFECTS OF ALIEN PLANTS AND GRANIVORES ON ECOLOGICAL RECOVERY FOLLOWING WILDFIRE

Todd Esque^{1, 3}, Sara Eckert¹, Louisa Evers², Timothy Duck², Brian Bracken², Tim Bartlett<sup>2

1</sup>United States Department of the Interior
United States Geological Survey, Biological Resources Division
Western Ecological Science Center, St. George Field Station
345 East Riverside Drive
St. George, Utah 84790
todd_esque@usgs.gov

²United States Department of the Interior
Bureau of Land Management -- Arizona
Arizona Strip District
345 East Riverside Drive
St. George, Utah 84790

³University of Nevada, Reno
Program in Ecology, Evolution, and Conservation Biology
Department of Biology / MS314
Reno, Nevada 89557

We describe the establishment of a large habitat manipulation designed to determine the mechanisms of habitat change due to fire, alien annual plants, and granivorous ants and rodents in the Mojave Desert. The Pakoon Basin of Mohave County in northern Arizona is an example of landscape level habitat change as a result of the alien annual grass/fire cycle. Wildfires and grass invasions are widespread, but localized, across the Mojave Desert. The Pakoon Basin has experienced large and frequent desert wildfires during the past 30 years. The research site occurs on an isolated, unburned mesa that is dominated by a Larrea tridentata -- Ambrosia dumosa vegetation association. The alien annual plants red brome (Bromus madritensis spp. rubens) and Mediterranean grass (Schismus spp.) occur in the area, but by no means dominate the landscape in this unburned habitat.

We established sixty 20 x 20-m treatment plots, half of which were burned in simulated desert fires. Treatments included burned vs. not burned; ants included vs. ants excluded, rodents included vs. rodents excluded, and interactions of these treatments. We also included procedural controls of unfenced plots. Response variables include soil seed bank diversity and density, annual plant production, perennial plant density and cover, and total soil inorganic nitrogen. Fire temperatures were measured with a datalogger attached to multiple thermocouples below and above ground (i.e. -5, -2, -1-cm, just below and above ground surface, +10, +25, and +50-cm) in the open and in Larrea and Ambrosia shrubs. In open sites, peak fire temperatures reached 350 °C. Peak temperature in Ambrosia was 768 °C, and Larrea 851 °C. These temperatures are within the known range for grass fires in the Great Plains and Sonoran Desert.

This research was initiated cooperatively between the US Geological Survey and Bureau of Land Management. We used grass hay as fuel to simulate a desert wildfire. We learned that all agencies with agricultural jurisdiction should be contacted prior to using agricultural hay. In addition, certified weed-free materials should be used in this type of activity. Fuels experts and fire fighters at the scene indicated that this fire was an accurate simulation of desert wildfire behavior. Damage to perennial plants was also similar to that observed on desert wildfires. We emphasize the benefits of such a large-scale ecological experiment by teaming together specialists in weed management, fire management and ecology.

ESTIMATION OF TAMARIX EVAPOTRANSPIRATION USING REMOTELY SENSED DATA

Lynn K. Fenstermaker¹, Dale A. Devitt², Stanley D. Smith³, and Ron Hammett<sup>3

1</sup>Desert Research Institute
Las Vegas, NV 89154
lynn@dri.edu

²University of Nevada, Reno
Reno, NV, 89557

³University of Nevada, Las Vegas
Las Vegas, NV 89154

A study was undertaken to evaluate the suitability of multispectral scanner data to estimate evapotranspiration (ET) from a Tamarix ramossisima stand located in the Virgin River floodplain approximately 2 kilometers north of the Overton arm of Lake Mead. A Bowen Ratio tower was established within this stand during the winter of 1994. 12-channel multispectral data (Daedalus 1260 and 1268) were acquired April 22 and July 28, 1994, May 26, 1995, and April 13 and August 29 1996. In addition sap flow, water potential, LI-COR 1600, tissue moisture and chlorophyll content were measured during the 1996 growing season. The data reported here are the result of a preliminary evaluation of the July 28, 1994 remotely sensed data. Significant relationships between canopy and air temperatures, NDVI and canopy temperature, and ET and canopy temperature were determined. An approach involving a NDVI coefficient and maximum latent heat flux was used to produce a map of approximate ET. It is apparent from this map that there is substantial variability within the fetch region of the Bowen Tower, which has an impact on the measurement of ET by the Bowen Ratio Energy Balance method. Further analyses are being performed with the 1996 data to produce a more accurate map of ET.

BEES, BEETLES, AND PLANTS: THE IMPORTANCE OF ECOLOGICAL INTERACTIONS TO CONSERVATION AND RESTORATION EFFORTS

John Hafernik and Leslie Saul-Gershenz

Department of Biology
San Francisco State University
1600 Holloway Avenue
San Francisco, CA 94132

There is growing awareness of the functional importance of insects in natural ecosystems and of the need to include them in conservation planning. Especially critical, for informed planning, is an understanding of the interactions and processes involving insects that affect the dynamics and composition of natural communities. Maintaining biodiversity requires preserving these important interactions. Our study addresses an important interaction, nest parasitism of the solitary bee Habropoda pallida (Anthophoridae) by larvae of the blister beetle Meloe franciscanus (Meloidae) and how this interaction may, in turn, affect two other critical interactions, pollination and herbivory. Habropoda pallida is an oligolectic bee primarily utilizing creosote bush Larrea tridentata, as a pollen source. Astragalus lentiginosus is a major nectar source for the bee in the sandy areas of the Mojave Desert. Individual Habropoda bees may visit up to 50,000 flowers in a lifetime and are thus probably important in influencing the floral diversity of sandy desert areas. Adult M. franciscanus are important winter and early spring herbivores of Astragalus lentiginosus and thus may also affect plant diversity, vegetation cover and nectar availability.

This study is being conducted at two sites within the East Mojave Preserve; the Desert Studies Center, near Baker California and the Kelso Dunes. We have sampled bees and beetles in these areas for the last 20 years, and have recently begun an intensive study. The results, so far, indicate that M. franciscanus is unusual among blister beetles in that its first instar larvae (triungulins) form large aggregations of up to 2000 larvae on bunch grasses and the dried remains of dicots in the dune areas of the Mojave Desert. These aggregations appear to be a novel adaptation for host-finding as they frequently attach, as a group, to males of H. pallida resulting in male bees carrying over 100 triungulins. We hypothesize that the triungulin larvae are dispersed to females by a type of "venereal transmission" as multiple males attempt to mate with a female. This mating behavior may also disperse triungulins among other males. Once on a female, a triungulin is carried to the bee's nest where it drops off and feeds on the bee's larvae and pollen and nectar provisions.

We believe this interaction is important for maintaining community structure in the Kelso Dunes and surrounding creosote bush flats. Loss of the bee could affect seed set and thus recruitment for several plant species. Loss of the beetle could affect winter growth rates of Astragalus lentiginosus and other species altering vegetation cover of it and other species. These events in turn could lead to a cascade of changes within the dune community. Managers wishing to conserve or restore dune habitat in the Mojave should include the following in their plans: 1) nesting sites for Habropoda pallida and other bees, 2) pollen and nectar sources for bees, 3) aggregation sites for triungulin larvae, and 4) foodplants for adult blister beetles.

HIBERNATION TEMPERATURES, TIMING, AND SITE CHARACTERISTICS FOR DESERT TORTOISES IN THE NORTHEAST MOJAVE DESERT

Dustin F. Haines¹, Todd C. Esque^1,2, and C. Richard Tracy<sup>2

1</sup> USGS-Biological Resources Division
St. George Field Station
45 East Riverside Drive
St. George, Utah 84790
dustin_haines@usgs.gov

² University of Nevada, Reno
Program in Ecology Evolution and Conservation Biology
Department of Biology / MS314
Reno, Nevada 89557

Desert tortoises (Gopherus agassizii) retreat to underground cover sites called hibernacula during winter months to escape cold temperatures in the northern part of their range. We studied tortoises while they were in hibernation using (a) radio telemetry and (b) miniature, temperature sensitive data loggers attached to tortoises at two sites I the northeastern Mojave Desert in 1995-1996 (N = 13), 1996-1997 (N = 16) and 1997-1998 (N = 20). The two sites differ in elevation, physiognomy and vegetation. The average dates that tortoises began hibernating combined for all the tortoises at both sites were: 4 November 1995 ( SD = 12.2 days), 24 October 1996 (SD = 8.0 days) and 28 Oct 97 (SD = 10.4 days). Onset and termination dates were the same between sites (P = 0.02, and P = 0.76, respectively). The average termination date of hibernation occurred on 24 March 1996 (SD = 14.59 days), 13 March 97 (SD = 4.37 days) and 27 March 98 (SD = 13.29 days). Onset and termination dates, however, were both significantly different between years (onset: P<0.02, termination: P<0.01). The duration of hibernation was not significantly different between sites (P = 0.67) or among years (P = 0.17). Thus , tortoises maintained the same amount of time in hibernation yet change the onset/termination dates.

Observed ambient temperatures inside hibernacula never dropped below 2 °C. However two tortoises emerged from their hibernacula to ambient air temperatures below 0 °C in April 1997. One experienced freezing temperatures for less than two hours, and the other between two and four hours (both tortoises survived). The hibernacula at one site were investigated with a fiberoptic light scope to determine depths to which tortoises retreated in hibernacula. We are using correlations among hibernacula temperatures and the depth of the tortoises as the basis of a predictive model of the efficacy of hibernacula. We anticipate that this model can be used by managers to establish reliable guidelines for anthropogenic, surface-disturbing activities. In particular, because understanding how hibernation conditions and site characteristics are important to how tortoises survive winter temperature extremes, we see the model as a way to minimize damage to tortoises in managed areas while remaining sensitive to the desire to conduct human activities in those areas.

CHARACTERISTICS OF THE MOJAVE AND GREAT BASIN DESERTS ON THE NEVADA TEST SITE

D. J. Hansen, W. K. Ostler, and D.B. Hall

Bechtel Nevada
P.O. Box 98521
Las Vegas, NV 89193-8521

The Nevada Test Site (NTS) located in south-central Nevada straddles the Mojave and Great Basin deserts. Transitional areas between the two deserts have been created by gradients in elevation, precipitation, temperature, and soils. In an effort to characterize and manage resources, over 1500 ecological landform units (ELU) were sampled at the NTS from 1996-1998. ELUs are small, homogeneous units of land having similar elevation, slope, soil, geological parent material, and vegetation. ELUs were delineated for the NTS using remote sensing data, topographic, soils, and geological maps. ELUs were sampled for biotic and abiotic factors, photographed, and a Global Positioning System coordinate taken from the midpoint along the 200-meter sampling transect. ELU boundaries were digitized into a Geographical Information System. Cluster analysis was used to classify ELUs into groups with similar relative abundance of shrub species. The vegetation classification system proposed by the Federal Geographic Data committee was used to establish the structure and nomenclature of vegetation types and a vegetation map was developed. This map and accompanying geographical information system database are an integral part in resource management at the NTS. Associations in the Great Basin Desert had the highest perennial species diversity measured by richness or evenness. Rodent sign and productivity of annual plants are least in the lower and higher elevations and most abundant in the mid-elevations. Microbiotic crusts are most abundant in fine-textured soils and of low abundance in soils with active erosional processes. Texture of surface soils differs little among all associations except for a few that are correlated with playas and steep mountain slopes. However, differences among associations are observed for deeper substrates comprised of limestone, basalt, and tuff parent materials.

CULTURAL RESOURCES OF THE GRANITE MOUNTAINS, CALIFORNIA

David Lee, Don Christensen, and Jerry Dickey

University of California
Sweeney Granite Mountains Desert Research Center
PO Box 101, Kelso CA 92351

The University of California's Sweeney Granite Mountains Desert Research Center is in the process of preparing a Cultural Resource Management Plan. To facilitate this process, a systematic and intuitive survey of archaeological resources was conducted from 1994 to the present. Surveys for rock art sites included the Granite Mountains, the adjacent Brown Buttes, and three areas of the southern Providence Mountains; vegetation was characterized at sites in the nearby Kelso Dunes; and lithic sources were documented in the Van Winkle Mountains just east of the Granite Mountains. These comprehensive surveys resulted in the recording of 93 new archaeological sites, and the re-recording, with more contemporary standards, 25 additional sites. Chronological controls were limited to surface finds of diagnostic projectile points and ceramics, but these findings provide evidence that pre-historic utilization of the Granite Mountains date back at least 4,000 years BP and continue to present the historic era. In total, more than 55 rock art sites and 2,200 elements have been documented in this study area. In particular, the Granite Mountains have the highest concentration of pictography currently recorded in the Mojave Desert, which underscores the ceremonial significance of the range. Because of the high biotic diversity and numerous water sources, the range offers the potential for "year-round" occupation by extended family-sized groups that were perhaps larger than those of the average seasonal subsistence patterns elsewhere in the Mojave Desert. Future research will attempt to more closely focus on this hypothesis. Preservation of cultural resources, particularly the highly visible and well-published rock art sites, presents some specific management challenges.

THE INFLUENCE OF EL NIÑO/SOUTHERN OSCILLATION CONDITIONS ON ACTIVITY AREA AND MOVEMENTS OF ADULT FEMALE DESERT TORTOISES

Kathie Meyer and Jeff Lovich

USGS/BRD Canyon
Crest Field Station
University of California, Riverside

We used radio-telemetry to monitor activity areas of adult female desert tortoises (Gopherus agassizii) at an industrial wind-energy park in the foothills of the San Bernardino Mountains near Palm Springs, California. We located tortoises weekly from late spring through the end of summer in 1997 and 1998 as part of a reproductive study. Precipitation totaled 96 mm and 231 mm in the 1996-1997 and 1997-1998 growing seasons (October through May), respectively. Rainfall was 2.4x greater during the 1997-1998 growing season due to an El Niño/Southern Oscillation (ENSO) event. Inordinate rainfall in the 1997-1998 growing season triggered the production of a substantial annual plant crop (mean dry weight 16.2 g/dm, SD = 15.6) the principal food and water source for desert tortoises. In comparison, biomass of annual plants in 1996 -1997 was significantly lower (mean dry weight 10.7 g/dm, SD = 7.2). We constructed minimum convex polygons for tortoise activity areas using all available observations. Mean activity areas in 1997 and in the ENSO year were 4.1 and 10.3 hectares, respectively. Tortoises increased the size of their activity areas during ENSO relative to1997 but there were significant individual differences as shown by a randomized block analysis of variance. Standardized mean minimum rates of movement for tortoises were 0.52 meters/hour in 1997 and 0.82 meters/hour during ENSO 1998. Activity areas and rates of travel were significantly greater in the ENSO year. Carapace length was positively correlated with activity area size during ENSO, and both years combined, but not in 1997 alone. Increased resource availability associated with abundant winter rainfall appears to have a strong influence on female desert tortoise activity and movements.

ANALYSIS AND ASSESSMENT OF MILITARY AND NONMILITARY IMPACTS ON BIODIVERSITY: A FRAMEWORK FOR ENVIRONMENTAL MANAGEMENT ON DOD WITHIN THE CALIFORNIA MOJAVE DESERT

David Mouat¹, Mary Cablk², Ross Kiester³, Jill Heaton⁴, Mark Meyers⁴, Richard Toth⁵, Rob Lilieholm⁵, and Robert Fisher<sup>6

1</sup>Desert Research Institute

²National Research Council

³U.S. Forest Service

⁴Oregon State University

⁵Utah State University

⁶San Diego State University

Our project is providing DoD with the techniques, training, and tools to carry out its military mission in the context of regional ecosystem management. Our goal is to enable the entire set of Mojave military installations to manage their resources unilaterally, as a single entity, within the context of the region, as opposed to the current system of independent resource managers operating only within the confines of their own jurisdictions. We are striving to design plausible future landscape for the Mojave Region based on trends, statewide population forecasts, existing plans, and other rational assumptions. This proactive approach is key to successful biodiversity management and continuation of effective military training and operations. We examine the impacts which are likely to result from these alternative future patterns of land use on the Desert Tortoise, focal vertebrate species, total terrestrial vertebrates, and landscape condition. We show the results of reptile transect data, the relationship between habitat condition and reptile demographics, changes in landscape condition from 1973 to the present, and initial designed futures.

We will show where people currently live in and adjacent to the Mojave and where they will be moving, and the implications of this population and attendant stressors on the Mojave environment. Finally, we discuss how these changes affect the military.

ANNUAL PLANT VEGETATION BOUNDARIES ON DISTURBANCE GRADIENTS IN CHIHUAHUAN DESERT RANGELANDS

Maliha Nash

U.S. Environmental Protection Agency
944 E. Harmon Ave.
Las Vegas, NV 89119

Piosphere gradients represent areas where the ecosystem is exposed to chronic disturbance of varying intensity while desertification transition gradients represent areas that changed in the past as a result of a combination of factors such as drought and grazing. We designed a study to test the hypothesis that the plant communities on gradients exposed to chronic disturbance would differ spatially and in composition from those on desertification transition gradients.

Spatial variation of annual plants in an area can be characterized by sampling along a suspected disturbance gradient at equi-spaced locations. The gradient can then be partitioned into homogenous zones by recognizing where the rate of change in composition and abundance of annuals is greatest with respect to distance. This allows the location of boundaries separating zones where disturbance has affected one or more of the variables that determine the germination and establishment success of species of annual plants.

DISTRIBUTION AND ABUNDANCE OF TWO KANGAROO RAT SPECIES ALONG AN ELEVATIONAL GRADIENT IN THE MOJAVE NATIONAL PRESERVE

Mary V. Price, Shauna A. McDonald

Department of Biology
University of California
Riverside, CA 92521
mary.price@ucr.edu

Since 1983 students at UC Riverside have censused small mammals along Kelbaker Road from Granite Pass to Kelso as part of field ecology class exercises. We have compiled 6 years of capture records for two kangaroo rat species; Dipodomys merriami and D. panamintinus, using captures per 100 trap-nights of effort as index of abundance. During the 1980s these two kangaroo rat species differed in their elevational distributions. D. panamintinus was most abundant at higher elevation and decreased at lower elevations. In contrast, D. merriami was most abundant at lower elevation and decreased with increasing elevation. Abundance of both species decreased dramatically following drought conditions in the early 1990s. During the low-density period the elevational pattern shifted to one in which D. merriami abundance either increased or did not change with increasing elevation, and D. panamintinus was less abundant than D. merriami at all elevations.

These results have several implications for wildlife management. First, abundance of desert rodents varies dramatically among years in response to yearly precipitation. Hence, management plans need to take temporal variability into account. Second, the shift in elevational distribution of D. merriami from moist to dry years suggests that low-elevation habitat may not be optimal for this species. Instead, D. merriami may be excluded from high elevation habitat by competition from D. panamintinus when the latter species is at high density; larger kangaroo rats are dominant over smaller ones. Alternatively, optimal conditions may vary with precipitation, occurring at lower elevations during moist years. In either case, patterns of distribution observed over a short time period may provide misleading information about habitat quality, and distributional data should not provide the sole basis for determining habitat requirements of sensitive wildlife species.

RECREATIONAL DISTURBANCE OF A DESERT STREAM FISH COMMUNITY

Mark Sappington

USGS - BRD Cooperative Studies Unit
Department of Biological Sciences
University of Nevada, Las Vegas
Las Vegas, NV
sapping@nevada.edu

As recreational use of public lands has grown over the past several decades, resource managers have increasingly faced the challenge of determining if recreational activities are impacting natural resources. Because natural variation in populations and communities may mask effects of human activities, an understanding of this natural variation is critical for establishing a baseline against which effects of human activities can be detected. This study examines the effects of recreational activities on native fish communities in the North Fork of the Virgin River within Zion National Park, Utah. During the summer of 1997 fish assemblages were sampled in 12 pools that were relatively free of recreational activities to determine natural spatial and temporal variation in native communities. Assemblages were concurrently sampled in 10 pools experiencing high levels of wading and float tubing to assess effects of recreational activities on fish populations and communities. Habitat features appeared to be an important determinant of native fish distribution among pools in low recreation areas. Although species abundance differed among these pools, communities were similar, and communities within individual pools were relatively persistent over the summer. Distributions of native fish appeared to be altered among pools in high recreation areas, suggesting that recreation may disrupt habitat associations. Communities in high recreation pools also appeared to be structured in an alternate state, with increased dominance by Virgin spinedace (Lepidomeda mollispinis mollispinis) and reduced abundance of Desert sucker (Catostomus clarki). Substrate disturbance by recreationists may account for decreased abundance of benthic feeding C. clarki in high recreation pools. Flash flood disturbances at the end of the summer appeared to "reset" the system, redistributing fish in the river and resulting in similar communities throughout the study area. Results suggest that recreational activities do disturb native fish communities; however, communities may return to a more natural state in the absence of recreational activities.

LITTER ACCUMULATION BENEATH MOJAVE DESERT SHRUBS EXPOSED TO PREDICTED 21ST CENTURY ATMOSPHERIC CO₂ LEVELS (FACE): COMPARISON OF EVERGREEN LARREA TRIDENTATA WITH VARIOUS DECIDUOUS SHRUBS

Zitzer Stephen, James Coleman, Dean Jordan, Robert Nowak, Jeff Seemann, and Stan Smith.

University of Nevada, Las Vegas
Las Vegas, NV 89154

University of Nevada Reno
Reno, NV, 89557

Desert Research Institute
Reno, NV 89517

The Mojave Desert is the most arid ecosystem of North America. Current climate change models predict the driest ecosystems will have the largest relative increases in primary productivity due to rising atmospheric CO₂ levels. We began continuous exposure of an undisturbed Mojave Desert plant community to 550 ppm atmospheric CO₂ using FACE (Free Air Carbon Dioxide Enrichment) technology on April 28, 1997. The perennial plant community in our study plots was dominated by Larrea tridentata an evergreen shrub, and deciduous shrubs Ambrosia dumosa, Lycium andersonii and Lycium pallidum which together comprised 64% of the mean perennial plant cover of 77.1 m² per plot. The experimental design consisted of 9 circular 25m diameter plots with three plots fumigated with 550 ppm CO₂, three control plots fumigated with ambient CO₂ and three control plots with no fumigation. Access to plants within the plots was via a specially designed apparatus which eliminated disturbance to the plants and soils. Paired litter trays, (0.0625 m²) were placed beneath shrub canopies with northeast and southwest aspects on 28 April 1998 and canopy depths above the trays were measured. Litter samples were collected at 4 to 8 week intervals, oven-dried to constant weight at 60^o C and weighed. On 1 February 1999 after 273 days, mean litter accumulation for all four species combined was 147.4, 119.0 and 110.3 g m^-3 of canopy volume for the 550 ppm CO₂ fumigation, non-fumigated and ambient fumigation treatments respectively. However, the 29% increase in mean litter accumulation beneath shrubs in the elevated CO₂ plots was not significantly greater than the controls plots because of the large variation between shrubs (coefficient of variation for the means ranged from 50-60%). Peak rates of litter accumulation for the deciduous shrubs occurred in July coincident with low volumetric soil moisture content. Peak rates for Larrea occurred in September, but temporal variation was less for evergreen Larrea than the deciduous shrubs. Species composition in the litter samples varied from 100 % of the overstory shrub species to a maximum of six species including annuals, grasses and other shrubs species. Litter quality also varied greatly relative to proportions of leaf, wood and seed in the samples. For an above average rainfall year, litter accumulation represented an annual turnover of about 10% of the aboveground biomass. Considering the low rates of decomposition in arid environments the increase in litter accumulation under Mojave Desert shrubs due to elevated CO₂ may represent a significant carbon sink under predicted levels of increasing atmospheric CO₂.

SOUTHWESTERN RIPARIAN SUSTAINABILITY ALONG A LOWER COLORADO RIVER IMPOUNDMENT

Nita Tallent-Halsell^{1, 2} and Lawrence R. Walker<sup>2

1</sup>U.S. Environmental Protection Agency
944 E. Harmon Ave.
Las Vegas, NV 89119

²University of Nevada, Las Vegas
Las Vegas, NV 89154

Riparian ecosystems are essential components of the southwestern landscape. They are extremely complex systems that are influenced by interactions among the hydrology, soils, geomorphology, and ecology of the entire watershed. The river hydrology, shoreline soils, geomorphology, and biology are often irreversibly changed with the construction of a dam and efforts to maintain and renew wildlife habitat are difficult. Riparian ecosystems on impounded Southwestern rivers are adversely affected by the dramatic water-level fluctuations and increased soil salinization. Revegetation efforts at Lake Mohave, a lower Colorado River impoundment bordering Nevada and Arizona, have not been successful (< 1% establishment). With the intent of describing the Lake Mohave riparian ecosystem, I examined the soils and vegetation composition at 34 2500-m² plot established along the approximately 100-km basin shoreline. In addition, greenhouse experiments were conducted to study the response of a native, Salix gooddingii (Gooding's willow) and an invader Tamarix ramosissima (saltcedar) to different water levels and soil types comparable to those influencing Lake Mohave riparian plant communities. Field observations indicate that Salix and Tamarix are the two dominant species. The flood-tolerant Salix is confined within the frequently inundated, drawdown zone as monospecific even-aged stands or mature individuals. The Tamarix, which cannot tolerate being inundated more than 70 days, is established in thickets immediately above the drawdown zone, suggesting that flooding suppresses Tamarix invasion into the flooded Salix-dominated area. No Salix seedlings or saplings were found within or above the drawdown zone suggesting that natural ecesis from seed source is being prohibited by elevated water levels during seed release. The average soil EC level was 4.76 ± 0.10 dS m^-1 which is toxic to Salix seedlings and cuttings. Greenhouse experiments revealed that survival was significantly higher for both species in saturated soils when compared to flooded and dry soils and the least growth and highest mortality resulted when plants were kept flooded, indicating that even though mature Salix is flood tolerant, cuttings are not and therefore should not be transplanted at the same lake elevational limit of mature Salix. Recommendations for introducing native vegetation include defining establishment goals and objectives, manipulating regulated water levels so they are aligned to native reproductive processes, remediating saline soils, removing Tamarix thickets, controlling predation, and monitoring the transplants into maturity.