Occupancy patterns of western yellow bats (Lasiurus xanthinus) in palm oases in the lower Colorado Desert.
Our study focused on identifying spatial patterns of western yellow bat occupancy and characteristics of palm oases used for their roosts to create an important baseline to assess how habitat changes may impact their patterns of occurrence within the northwestern Colorado Desert. Those changes may include increased human recreational activity in roost areas, climate change, fire, and invasive species. Native palm oases in this region are popular for picnicking, camping, and destinations for recreational hiking. This use has sometimes resulted in wildfires. Although palm trees aren't necessarily killed as a result of fires, the attached dead leaf fronds (skirts) are removed (Cornett, 2010). While the lack of a palm skirt (immediately after fires) would clearly eliminate roosting habitat for bats, no studies have determined whether the bats use the shorter palm skirts that may be the only structure for roosting available for years following fire until young, unburned trees mature. It also has not been determined how long a time after a fire is required for a palm oasis to provide adequate roosting habitat for western yellow bats. Human activity in palm oases is another factor that may impact bat use of a site. Studies have shown some bat species to be sensitive to human activity, such as the gray bat (Myotis grisescens), which will abandon its roost when disturbed (Gore et al., 2012). While most tree-roosting bats select roosting sites high off the ground where humans are not able to directly disturb them, indirect disturbances such as trails and campsites may still have an effect in areas with heavy human disturbances. Disturbances such as light and sound can incite arousals in hibernating bat populations (Thomas, 1995) but have unknown effects on nonhibernating populations such as western yellow bats.
Native palm oases are often invaded by nonnative invasive plant species, primarily tamarisk or salt cedar (Tamarix ramosissima) and fountain grass (Pennisetum setaceum), which were originally introduced for landscaping or wind breaks (Barrows, 1993). The impact of these invasive species in isolated oases with finite water inflow can be the reduction or elimination of surface water, drying of the soil surface and, in the case of salt cedar, exuding salt from their leaves (Barrows, 1993). While a negative association between Lasiurus xanthinus and Tamarix species in desert riparian habitat along the Lower Colorado River was identified by Vizcarra et al. (2012), the impacts of invasive plant species on western yellow bats that occur in isolated oases have not been previously studied.
Material and Methods--Our study area was the Colorado Desert subdivision of the Sonoran Desert in southern California. Of the approximately 168 Washingtonia filifera palm oases occurring in the wild (Cornett, 2010), 41 (24%) were included in our surveys. Additional palm oases within this region were not surveyed due to difficulties gaining permitted access or to steep, inaccessible topography. Western yellow bat surveys included palm oases within the Santa Rosa and San Jacinto Mountains National Monument, Joshua Tree National Park, the Coachella Valley Multiple Species Habitat Conservation Plan, Anza Borrego Desert State Park, and on Bureau of Land Management lands (Table 1). Oasis names were derived from nomenclature used on maps or, in instances in which a name was not evident, we named the palm oasis after the canyon where the palms were located or after a designated trail nearest to the palms. in some instances the palm oases were partitioned into multiple sites if they were distinctly separated from the other palm sites spatially, structurally, or both (i.e., burned vs. unburned, tamarisk vs. no tamarisk).
Our survey protocol required surveying each of the 41 palm oases at least once from May through October of 2012. If L. xanthinus occurrence and roosting could not be confirmed through acoustic and visual affirmation on the first round of surveys, then oases were resurveyed up to two more times (for a maximum of up to three surveys at each site). Positive identification through visual and acoustic identification of a single yellow bat was sufficient to determine occupancy. Our objective was to determine occupancy, not density or abundance; however, other than family units consisting of an adult female and her dependent young, lasiurines are usually solitary (Carter and Menzel, 2007). Surveys were limited to nights when rain and lightning were not in the forecast and when winds did not exceed 15 km/h. While some bats have been shown to be lunar-phobic (Mello et al., 2013), previous research has found no lunar phase effect on bat activity for insectivorous bats in western North America (Hayes, 1997; Rogers et al., 2006). Although our study was not designed to test moon-phase sensitivity in L. xanthinus, we did take moon phase into consideration by ensuring that not all surveys at any given site were conducted during the same moon phase.
Where possible, surveys were conducted at or near level to the crown of palms in order to get a better visual on the emergence of the bat from its roost by accessing hillsides near palms. Talking was minimized, as there were numerous instances where we were able to hear rustling noises from the skirt of the palm, which was a helpful indictor of the yellow bat getting ready to take flight from its day roost in the palm skirt. Occupancy by western yellow bats in a palm oasis was determined through a combination of acoustic monitoring and visual identification at emergence from its roost. Lasiurus xanthinus was only considered to be confirmed roosting in the oases if the bat was seen dropping out of a palm tree skirt with the aid of a spotlight and calls were recorded of the species on an Anabat SD2 bat detector (Titley Scientific USA, Columbia, Missouri).
We employed the Anabat SD2 to actively monitor bat occurrences, moving around the palm oases to maximize bat detections. Active monitoring was selected over passive monitoring, as movement patterns and behaviors of western yellow bats were used as a guide for the next survey, if needed. Other studies using active monitoring of hoary bats (Lasiurus cinereus) showed that they used the same foliage roosts over days and weeks in spite of the unlimited number of roosting sites available (Willis and Brigham, 2005). Additionally, the quality of a call being recorded was higher as a direct result of being able to maintain contact with individual bats by adjusting the orientation of the microphone in relation to the movement of the bat in flight (Brigham et al., 2004). The western yellow bat shares the characteristic higher frequency, steeper call signature shared among tree-roosting bats in the genus Lasiurus; however, the different species calls each occupy different frequency bands (O'Farrell et al., 2000). Call sequences typically begin in the 60 kHz range and end at 32 kHz with a steep sweep, although there can be some variation within call sequences depending on the echolocation activity of the bat (Adams, 2003).
Monitoring began at the twilight-dark interface and continued for 1 h after sunset to confirm day roost location. Only one Anabat bat detector was used per site. All surveys began at a stationary point, sometimes sitting on a rock outcrop or standing a few meters away from the oasis with at least two surveyors on site. Once surveys began we stayed stationary, moving our locations only if there was no western yellow bat activity after 30 min of surveying or towards the end of the survey well after yellow bats had emerged from their day roost. Field notes were recorded on a voice recorder. By surveying from an initial stationary site near the perimeter of the oasis, we were able to see the direction the yellow bat came from on the first round of surveys and adjust our position for the second or third round. in most instances, this worked well for pinpointing the area of the oasis where the yellow bat was roosting.
While watching calls being displayed on the PDA screen in real time, we also listened for movement in the palm skirt and then spotlighted bats for visual identification. There can be acoustic overlap with eastern red bat (Lasiurus borealis), and big brown bat (Eptesicus fuscus) (Western Bat Working Group, http://www.wbwg. org/speciesinfo/survey_matrix/recommended_survey_methods. pdf); however, the eastern United States range of L. borealis does not overlap with L. xanthinus. As a result of identifying western yellow bat real-time in the field, any overlap in acoustic identification was eliminated by observing the species in flight with its medium-sized body, bright-yellow coloration of fur, and its tendency to fly high off the ground compared to other bat species that may also occur in a palm oasis. The only bat species that somewhat resembled the yellow bat in physical appearance was the pallid bat, Antrozous pallidus. The two species were differentiated by their calls as well as by the paler fur of pallid bats. Additionally, pallid bats flew low to the ground and unusually close to us as we monitored the palms, something yellow bats never did.
Habitat variables associated with palm oases were measured prior to active monitoring of bats at each palm oasis. The specific independent environmental metrics measured are described in Table 2. We constructed logistic regressions to determine which combination of these variables best explained the bat's occupancy and roost patterns. A logistic regression was used as it allowed inclusion of both the categorical and continuous predictor variables we measured at each oasis. Statistical analyses were conducted using SAS (Statistical Analysis System 9.0, Copyright [c] 2008, SAS Institute Inc., Cary, NC). All possible combinations of measured variables were analyzed to construct the model that best explained the western yellow bat occurrences. The best-fit model was determined by the variable combination that yielded a statistically significant model (P < 0.05, determined using a [chi square] analysis) and which had the lowest Akaike Information Criterion (AIC) value; models with AAIC scores within <2 points of the model with the lowest AIC value have strong support for also being best-choice models (Burnham and Anderson, 2002). The two dependent variables were yellow bat occupancy (1 = detected, 0 = undetected) and confirmed roosting (1 = confirmed, unconfirmed = 0).
A complete list of independent continuous variables used to construct our logistic regression models included: 1) elevation; 2) number of palm trees; 3) palm density; 4) distance to other palm clusters-oases (i.e., isolation--separation of clusters was based on topography as well as distance); 5) distance to closest water source (potentially important for the bats to drink and as a correlate of insect activity); 6) distance to urban areas; and the number of palms with palm skirt ratings of 7) 0-49% and 8) 50100%. Categorical variables included: 1) human disturbance; 2) invasion of Tamarix species; 3) the occurrence of recent or current palm germination; 4) the occurrence of owls (potential predators) roosting in the palms; and 5) fire history (Table 2).
Results--Forty-one palm oases were surveyed for the occurrence of L. xanthinus for a total of 106 nights of active monitoring. Roosting western yellow bats were confirmed at 19 of the palm sites, representing nearly half of all the palm oases surveyed. Yellow bats were found using, though not always roosting at, 33 oases; no yellow bats were detected at eight palm oases (Table 1). The majority of the sites with no yellow bat detections occurred in the northwestern end of the Colorado Desert near the San Gorgonio Pass (Fig. 1). These oases sites also exhibited much-less activity by other bat species compared to the rest of the locations we surveyed. Willow Hole Palms had no bat activity on all three occasions in which surveys were conducted. Even though 22 sites were identified as having yellow bats present but "unconfirmed" for roosting, it did not mean that they were not roosting in those oases, only that after three separate visits no yellow bats were observed emerging from palm skirts.
Logistic regression models were constructed for sites with confirmed roosting and for the presence (but unconfirmed roosting) of western yellow bats. The four best models all include new palm growth, elevation, and palms with skirt length ratings of 0-49% as important variables (Table 3). The distribution of palm oases where the bats were confirmed to be roosting is shown in Figure 2. The best two models for determining characteristics of palm oases in which yellow bats were detected (but not confirmed roosting) were similar to models for confirmed roosting sites (elevation and palms with skirt length ratings of 0-49%); however, additional variables emerged as important for describing the bats' occurrence. Those additional variables included (the absence of) owls roosting in the palms, distance to the next palm oasis, and palm skirt ratings of 50-100%. While the logistic models identified which variables were important for discerning sites where the bats roosted or were present, it is important to identify characteristics of those variables the bats may be selecting. We identified those variables that were consistent components of the highest ranking models and then contrasted those values for those sites with confirmed yellow bat roosting with sites where the bats were never detected (Table 4). Yellow bats roosted at oases with significantly higher elevation and with more palms with partial to full skirts. Of the 19 palm oases in which confirmation of a yellow bat roost was recorded, one had a mean skirt rating of 0-24%, nine had a 25-49% rating, three had a skirt rating of 50-74%, and six had skirt rating of 75-100%. Additionally, for palm oases with confirmed roosts, 74% showed evidence of new palm growth in the area whereas all locations in which the species was absent lacked any evidence of new growth.
Discussion--We found western yellow bats widespread at palm oases within the Colorado Desert of southeastern California, detecting them at 80% of the oases we surveyed. Confirmed roosting was observed at 46% of the oases surveyed, although it is likely that with a more-intensive survey effort some additional roost sites would be confirmed. sites with confirmed yellow bat roosting were significantly higher in elevation, with more palm skirts of variable lengths, and had more young palm growth than did sites where the bats were undetected. Additionally, there appeared to be an avoidance of palm oases at the northwestern edge of the palms' distribution in Colorado Desert.
Our original concern regarding the negative impact of increased fires in palm oases on the occurrence and roosting of western yellow bats was supported indirectly. Of the five oases surveyed that had the most-severe fire damage (charred trunks, only short skirts), 80% lacked detections of yellow bat roosting. Although fire history did not emerge as an important habitat variable, the existence of palm skirts of variable lengths was important. Many of the palm oases included trees of mixed skirt lengths due to fire history, packrat (Neotoma species) activity, or winds blowing off fronds. A benefit of having a grove of palm trees with assorted skirt lengths may be that the microclimate within different skirt lengths could provide the yellow bat with various thermal choices and protection, allowing the bats opportunities for seasonal as well as daily roost switching without having to do much traveling. More than half of confirmed roosting sites were in the 0-49% skirt length rating. While fires may be advantageous to diversifying palm habitat within a given oasis over long periods of time, oases with more-recent fire histories lacked diverse skirt lengths and appeared to have a negative effect on the bats. Sites in which the majority of the palm oases were recently burned lacked yellow bat activity and had reduced detection of any bat species.
Human disturbance within the oases did not emerge as an important habitat variable. Only two of the oases we surveyed had regular human visitation at night (camping), and those sites also lacked detections of roosting western yellow bats. However, no patterns emerged with moderate and infrequent human use to warrant any conclusions based on just two sites with regular camping activity, and so we believe additional research is need to answer the question of direct human disturbance.
The occurrence of tamarisk or salt cedar did not emerge as an important habitat variable. This may indicate that tamarisk is unimportant as an either positive or negative influence on western yellow bats or that the tamarisk abundance at the sites we surveyed had not yet reached a threshold of having a negative impact. The occurrence of new, young palm growth was an important variable and was indicative of sites with high (or at the surface) water tables. These same conditions support tamarisk germination and expansion; two of the five oases with the densest tamarisk growth also had new young palm growth. However, as tamarisk stands mature at sites where water is a limiting factor, they can reduce or eliminate water at or near the surface and then reduce new palm growth (Barrows, 1993).
Bat species have varying flight characteristics that allow some species greater maneuverability than others; thus, the size of the watering hole may prove to be more important than just the mere presence of surface water. As bats need to drink on the wing, water sources with larger surface areas tend to facilitate access by more bat species (Sara et al., 2010). Future research will focus on water sources used by western yellow bats and may show that the species has specific needs for drinking that may make palm oases with larger watering holes more important in roost selection.
Climate change may represent an additional stress for palm oases and, therefore, for western yellow bats. A limiting resource for native palms is surface water or near-surface groundwater, and predicted, reduced precipitation in arid regions, will likely deplete already stressed groundwater aquifers (Ali et al., 2012). Predicted increased drought frequency and intensity in the region including the Colorado Desert (Gao et al., 2012) may result in reducing the vigor of palm oases that are already water stressed. We did not measure the impact of climate change directly; nevertheless, two of the habitat variables that emerged as important for determining sites where western yellow bats roosted (elevation and new palm growth) could be impacted by the predicted levels of warming and drying for this region (Seager et al., 2007; Gao et al., 2012). Elevation was a component of every highly significant model assembled; sites with confirmed yellow bat roosts ranged from 25-948 m but averaged more than 170 m higher than those where the bats were never detected. The two lowest elevation sites (25 and 89 m) had abundant open water nearby that may have facilitated occupancy of these warmer sites. Assuming the bats' selection of middle-elevation sites for roosting reflected a thermal preference for themselves or their insect prey (or both), a warming climate could shift their roost sites to higher elevations. In a separate study, McCain (2007) found highest bat species richness at mid-slope and inferred a relationship with temperature and water.
The highest-elevation site we surveyed was at Dos Palms in the Pinyon Crest community within the Santa Rosa and San Jacinto National Monument (1,097 m), which was a transition area between desert and pinon-juniper forest. Although western yellow bats were not confirmed roosting at this location, they were detected at the oasis on two of the three nights surveyed. Additionally, there were five other palm sites over 800 m in elevation and yellow bats were confirmed roosting in all of them. Surveys conducted by the San Diego Natural History Museum also confirmed this species in nearby Taquitz Valley (in the San Jacinto State Park) (Drew Stokes, San Diego Natural History Museum, http://www.sdnhm.org/science/birds-and-mammals/ projects/san-jacinto-resurvey/results-and-products/), which is over 2,400 m in elevation. This finding demonstrates that the yellow bat is not entirely a desert-dwelling species. Yellow bats appear to occur in a greater range of suitable habitats than was previously known, and that plasticity may enable them to move foraging and perhaps roosting sites upslope as the climate warms. While the bats appear to be able to move upslope to avoid hotter-drier conditions, increased drought could reduce groundwater levels (Gao et al., 2012), reducing the vigor of palm oases and potentially reducing roost habitat suitability, especially at lower elevations.
Many tree-roosting bat species tend to be solitary and so their population levels and dynamics are difficult to study. There were no quantitative comparisons of long-term population trends of these foliage-roosting bats (Carter et al., 2003). In lieu of demographic or population-level data, identifying patterns of occupancy provides the best indication of distributional and particular changes in site occurrence with respect to shifting environmental conditions. This study provides land managers and biologists with important baseline data from which to evaluate the impacts of potential habitat and landscape changes due to land use, invasive species, and climate change.
We are very thankful for the help of our field assistants who spent many nights studying bats during the hottest time of the year: A. Nuckels, C. Barrows, G. Raymond, R. Jarvinen, S. Jarvinen, B. Rooney, L. Rooney, S. Robles, C. Robles, G. Decker, J. Purcell, J. Danos, T. Martin, K. Doran, B. Pope, V. James, J. Grossman, R. Vincent, S. Greely, G. Short, D. Scriven, D. Scriven, J. Futterman, I. Hawkins, and A. Ortiz. We would also like to thank agencies and organizations for allowing us to conduct surveys on their land: Santa Rosa and San Jacinto Mountains National Monument, Bureau of Land Management, U.S. Forest Service, Friends of the Desert Mountains, Center for Natural Lands Management, Indio Hills State Park, Anza-Borrego Desert State Park, California Department of Fish and Wildlife, Joshua Tree National Park, and the University of California Riverside, Boyd Deep Canyon Research Station. Bat Conservation International provided a scholarship to attend their Acoustic Monitoring Workshop in 2011 and gain experience and knowledge from C. Corben in using the Anabat SD2 bat detector.
Adams, R. A. 2003. Bats of the Rocky Mountain West: natural history, ecology, and conservation. university Press of Colorado, Boulder, Colorado.
Ali, R., D. McFarlane, S. Varma, W. Dawes, I. Emelyanova, G. Hodgson, and S. Charles. 2012. Potential climate change impacts on groundwater resources of south-western Australia. Journal of Hydrology 475:456-472.
Barrows, C. W. 1993. A success story. Restoration and Management Notes 11:35-38.
Bowers, N., R. Bowers, and K. Kaufman. 2004. Kaufman focus guides: mammals of North America. Houghton Mifflin Company, New York.
Brigham, R. M., E. K. V. Kalko, G. Jones, S. Parsons, and H. J. G. A. Limpens. 2004. Bat echolocation research: tools, techniques and analysis. Bat Conservation International, Austin, Texas.
Burnham, K. P., and D. R. Anderson. 2002. Model selection and multi model inference: a practical information-theoretic approach, 2nd Edition. Springer, New York.
Carter, T. C., and J. M. Menzel. 2007. Behavior and day-roosting ecology of North American foliage-roosting bats. Pages 61-81 in Bats in forests: conservation and management (M. J. Lacki, J. P. Hayes, and A. Kurta, editors). The John Hopkins University Press, Baltimore, Maryland.
Carter, T. C., M. A. Menzel, and D. A. Saugey. 2003. Population trends of solitary foliage-roosting bats. Pages 41-47 in Monitoring trends in bat populations of the United States and territories: problems and prospects. (T. J. O'Shea and M. A. Bogan, editors). U.S. Geological Survey, Biological Resource Discipline, Information and Technology Report, USGS/BRD/ITR-2003-0003.
Cornett, J. W. 2010. Desert palm oasis. Nature Trails Press, Palm Springs, California.
Gao, Y., L. R. Leung, E. P. SalathE, Jr., F. Dominguez, B. Nijssen, and D. P. Lettenmaier. 2012. Moisture flux convergence in regional and global climate models: implications for droughts in the southwestern united States under climate change. Geophysical Research Letters 39:1-5.
Gore, J. A., L. Lazure, and M. E. Ludlow. 2012. Decline in the winter population of gray bats (Myotis grisescens) in Florida. Southeastern Naturalist 11:89-98.
Hayes, J. P. 1997. Temporal variation in activity of bats and the design of echolocation-monitoring studies. Journal of Mammalogy 78:514-524.
Higginbotham, J. L., L. K. Ammerman, and M. T. Dixon. 1999. First record of Lasiurus xanthinus (Chiroptera: Vespertilionidae) in Texas. Southwestern Naturalist 44:343-347.
Higginbotham, J. L., M. T. Dixon, and L. K. Ammerman. 2000. Yucca provides roost for Lasiurus xanthinus (Chiroptera: Vespertilionidae) in Texas. Southwestern Naturalist 45:338-340.
McCain, C. M. 2007. Could temperature and water availability drive elevational species richness patterns? A global case study for bats. Global Ecology and Biogeography 16:1-13.
Mello, M. A. R., E. K. V. Kalko, and W. R. Silva. 2013. Effects of moonlight on the capturability of frugivorous phyllostomid bats (Chiroptera: Phyllostomidae) at different time scales. Zoologia 30:397-402.
O'Farrell, M. J., C. Corben, and W. Gannon. 2000. Geographic variation in the echolocation of the hoary bat (Lasiurus cinereus). Acta Chiropterologica 2:185-196.
O'Farrell, M. J., J. A. Williams, and B. Lund. 2004. Western yellow bat (Lasiurus xanthinus) in southern Nevada. Southwestern Naturalist 49:514-518.
Rogers, D. S., M. C. Belk, M. W. Gonzalez, and B. L. Coleman. 2006. Patterns of habitat use by bats along a riparian corridor in northern Utah. Southwestern Naturalist 51:52-58.
Sara, L., S. L. Jackrel, and R. S. Matlack. 2010. Influence of surface area, water level and adjacent vegetation on bat use of artificial water sources. American Midland Naturalist 164:7479.
Seager, R., M. Ting, I. Held, Y. Kushnir, J. Lu, G. Vecchi, H. Huang, N. Harnik, A. Leetmaa, N. Lau, C. Li, J. Velez, and N. Naik. 2007. Model projections of an imminent transition to a more arid climate in southwestern North America. Science 316:1181-1184.
Thomas, D. W. 1995. Hibernating bats are sensitive to nontactile human disturbance. Journal of Mammalogy 76:940-946.
Vicarra, B., L. Piest, and V. Frary. 2012. Lower Colorado River Multi-Species Conservation Program: Monitoring of covered and evaluation bat species for the Lower Colorado River Multi-Species Conservation Program - 2010 Final Report. Arizona Game and Fish Department http://www.lcrmscp. gov/steer_committee/technical_reports_species.html
Willis, C. K. R., and R. M. Brigham. 2005. Physiological and ecological aspects of roost selection by reproductive female hoary bats (Lasiurus cinereus). Journal of Mammalogy 86:85-94.
Submitted 20 September 2013
Acceptance recommended by Associate Editor, Troy A. Ladine, 6 January 2014.
Danielle D. Ortiz and Cameron W. Barrows *
Green Mountain College, Poultney, VT 05764 (DDO)
University of California Riverside's Center for Conservation Biology, Palm Desert, CA 92211 (CWB)
Present address of DDO: U.S. Forest Service, Big Bear Ranger Station, Fawnskin, CA 92333
* Correspondent: email@example.com
Table 1.--List of oasis names and land ownerships for those oases surveyed in this study. Bat occurrences were categorized as not detected (ND), detected (D), roosting confirmed (R), detected and roosting (D-R). Palm oasis name Land manager-owner Western yellow bat occurrence Oswit Canyon Palms Agua Caliente Band ND of Cahuilla Indians Borrego Canyon Palms Anza-Borrego Desert D-R State Park Mountain Palm Anza-Borrego Desert D-R Springs Palms State Park Seventeen Palms Anza-Borrego Desert ND State Park Art Smith Trail Bureau of Land D Palms no. 1 Management Art Smith Trail Bureau of Land D Palms no. 2 Management Bogert Trail Palms Bureau of Land D Management Corn Springs Palms Bureau of Land D Management Cox Palms Bureau of Land D Management Dead Indian Canyon Bureau of Land D-R Palms Management Devil Canyon Palms Bureau of Land D-R Management Folgers Palms Bureau of Land D-R Management Green Hill Palms Bureau of Land D-R Management Hidden Palms Bureau of Land D-R Management Hunter Palms Bureau of Land D Management Indian Palms Bureau of Land ND Management Ranch House Palms Bureau of Land D Management San Andreas Fault Bureau of Land D Palms Management Sheep Hole Palms Bureau of Land ND Management Vargas Palms Bureau of Land ND Management Willis Palms Bureau of Land D Management Willow Hole Palms Bureau of Land ND Management Carrizo Canyon Palms California D-R Department of Fish and Wildlife Grapevine Canyon California D-R Palms Department of Fish and Wildlife Hidden Palms California D-R Ecological Reserve Department of Fish and Wildlife Magnesia Canyon California ND Palms no. 1 Department of Fish and Wildlife Magnesia Canyon California D-R Palms no. 2 Department of Fish and Wildlife Magnesia Canyon California ND Palms no. 3 Department of Fish and Wildlife McCallum Palms Center for Natural D-R Lands Management Thousand Palms Center for Natural D Lands Management Applegarth- Friends of the D Lakeshore Palms Desert Mountains Biskra Palms Indio Hills State D Park Hidden Palms Indio Hills State D-R Park Macomber Palms Indio Hills State D Park Pushwalla Palms Indio Hills State D-R Park Cottonwood Springs Joshua Tree National D-R Palms Park Forty-nine Palms Joshua Tree National D-R Park Lost Palms Joshua Tree National D-R Park Dos Palms United States Forest D Service Bear Creek Palms University of D-R California Riverside Deep Canyon Palms University of D-R California Riverside Table 2.--Habitat parameters used in constructing logistic regression models to explain the presence and absence of western yellow bats (Lasiurus xanthinus) in palm oases. Variable name Description Elevation Derived from Google Earth; measurements recorded in meters New palm growth Identification of new palm tree growth by which plants were 1 ft in height or less and no "woody" tissue evident; new growth was found in areas of the oasis where a spring, seepage, or body of water was present Next oasis distance Derived from Goggle Earth; linear distance to the nearest palm oases Water distance Derived from Google Earth: distance to open water for drinking-watering holes that were at least 1 x 1 ft, open and unobstructed, including swimming pools Urbanization Derived from Google Earth: distance to urbanization was measured in meters Total palms Total number of individual palms in an oasis were counted Palm skirts 0-49 Number of palm trees in the oasis with 0-49% of palm skirt and not dead; only palms over 10 ft tall were included Palm skirts 50-100 Number of palm trees in the oasis with 50-100% of palm skirt; only palms over 10 ft tall were included Owl roosts Prior to bat surveys, each oasis was searched for evidence of owls roosting in palm oasis (whitewash and or pellets), visual of owl, or both: yes (1), no (0) Tamarisk The presence of Tamarisk species was denoted as present (1) or not present (0) Density Derived from Google Earth: a polygon around each oasis obtained the area in hectares of the palm oasis; using the total number of palms we divided the number of palms by the area to obtain a density Human activity A rating of (0) = rarely visited, (1) = regular daytime but no nighttime human use, (2) = daytime and nighttime use (i.e., campground) Fire A rating of (0) = no evidence of a recent fire with skirts nearly reaching the ground, (1) = palms in which a fire occurred within [greater than or equal to] 0 yr and skirts not exceeding half of the height of the palm tree, (2) = fire occurred < 10 years; palm trunks were heavily charred and there was little to no palm skirt development Table 3.--Akaike's Information Criterion scores (AIC), difference values (AAIC), and the P-value (P) for confirmed day roosts of western yellow bats (Lasiurus xanthinus) in desert fan palm (Washingtonia filifera) oases in the California portion of the Colorado Desert. Model variables AIC [DELTA]AIC P New growth+ 41.290 -- 0.0031 elevation+next oasis+water+0-49 Elevation+0- 42.013 0.723 0.0045 49+water+new growth New 42.833 1.543 0.0086 growth+elevation+ next oasis+water+owl roost+0-49+50-100 New 43.253 1.963 0.0065 growth+urbanization+ elevation+0-49+next oasis+water New growth+ 44.915 3.625 0.0187 urbanization+ elevation+0-49+ 50-100+next oasis+water Elevation+next 45.100 3.81 0.0272 oasis+owl roost+0-49+50-100 New growth+ 46.915 5.625 0.0322 urbanization+ elevation+0-49+50- 100+next oasis+ water+total palms Elevation+next 47.054 5.764 0.0487 oasis+water+owl roost+0-49+50-100 Elevation+owl 49.906 8.616 0.0039 roost+50-100+new growth New growth+ 51.344 10.054 0.0042 elevation+next oasis+water Table 4.--Variables identified in the logistic regression analyses compared between confirmed roosting habitats vs. oases where western yellow bats (Lasiurus xanthinus) were absent. Variables denoted with an asterisk (*) were found to be statistically different (P < 0.05) using a t-test for samples with unequal variances. Confirmed yellow bat roosting Mean SE Elevation (m) * 439.4 72.5 Proximity to water (m) 1886.4 647.6 Distance to next oasis (m) * 2312.2 629.2 Palm trees over 10 ft, 0-49% * 65.6 22.2 Palm trees over 10 ft, 50-100% 160.6 67.7 Evidence of new palm growth 74% Yellow bat absence Mean SE Elevation (m) * 263.3 39.3 Proximity to water (m) 1273.4 450.8 Distance to next oasis (m) * 2494.9 1124.2 Palm trees over 10 ft, 0-49% * 14.2 8.7 Palm trees over 10 ft, 50-100% 24.5 8.1 Evidence of new palm growth 0%
|Printer friendly Cite/link Email Feedback|
|Author:||Ortiz, Danielle D.; Barrows, Cameron W.|
|Date:||Sep 1, 2014|
|Previous Article:||Mountain lion habitat selection in Arizona.|
|Next Article:||Differences in benthic macroinvertebrate assemblages associated with a bloom of Didymosphenia geminata in the Lower American River, California.|