For decades biologists have focused almost exclusively on the presence or absence of certain indicator plant species in predicting salt marsh harvest mouse (Reithrodontomys raviventris) occupancy in the San Francisco Estuary. At the 2022 Annual Meeting we presented a new approach to assessing habitat for its potential to support the salt marsh harvest mouse which focused on function over form. This method guides the user to evaluate the habitat at a local and regional scale, and specifically addresses features important to individual survival, likely the single most important factor in population level persistence of this species which occurs in naturally isolated patches. During the 2022 Rangewide Survey for the Salt Marsh Harvest Mouse biologists utilized this habitat assessment tool to evaluate habitat and predict species presence before and after live trapping surveys at 55 sites. During this presentation we will present the results of the assessments, an evaluation of how accurately the assessments guided correct predictions, and recommendations for its use in the future. We will also address which habitat features from the assessment were most correlated with salt marsh harvest mouse occupancy (e.g., the value of marsh size versus vegetative cover). As salt marsh harvest mouse presence or absence has very important implications for many decisions made in the Estuary, the ability to accurately predict occupancy without costly and intensive live trapping is extremely valuable. Regional biologists should work together to continuously refine and improve these methods.

In 2022 a collaborative group of wildlife managers, researchers, and consultants in the San Francisco Estuary completed the first ever range-wide survey for the endangered salt marsh harvest mouse (Reithrodontomys raviventris). Since the species is restricted to the marshes of the Estuary and exists in discrete populations contained within isolated marshes, it was possible to identify all potential populations and target them for surveying. A large amount of data is collected for each harvest mouse captured (~15-20 attributes), in addition to 4-6 photos per individual, making data collection, management, and storage a challenge. Post-survey manual data entry is challenging due to factors like difficult to read handwriting. Further, linking tabular data with photo data is challenging using traditional data storage programs like Microsoft Excel. During this survey efforts 3-6 crews worked simultaneously to survey sites throughout the Estuary. Preparing and utilizing a different Field Maps map each week, which contained all sites, allowed teams to track progress and work adaptively. Pre-prepared maps of trap grids made it easier to track which traps had been checked, in addition to providing a dedicated container for tabular and photo data. Utilizing modern mobile app based technologies can improve survey efforts and data quality, though wildlife biologists themselves may not have the time or technical expertise to set them up. Here we share lessons learned and methods for utilizing ArcGIS Field Maps to perform better and safer wildlife surveys.

Habitat fragmentation is a leading threat to wildlife globally. Effects of habitat fragmentation are often inferred retrospectively based on genetic connectivity estimated from a single modern sampling effort. Alternatively, genetic data obtained from museum specimens can facilitate direct comparisons of genetic connectivity pre- and post-habitat fragmentation. We examined the effects of habitat fragmentation on the federally endangered salt marsh harvest mouse endemic to the San Francisco Bay Area by comparing genetic data from 11 microsatellite markers in modern samples (2010-2022) and historical museum specimens (1938-1959). Genetic diversity (allelic richness, expected heterozygosity) was greater in the historical samples in the total population and within each of the six subpopulations we sampled. Genetic distance (FST) among subpopulations was greater in the modern sample than historical sample, consistent with reduced connectivity and/or genetic drift within populations. Generally, sites with the smallest remaining habitat patches exhibited the greatest degree of genetic diversity loss. Our results directly link losses in genetic diversity and connectivity with ~80 years of habitat loss and fragmentation. Our work highlights the threat of anthropogenic landscape change to wildlife populations.

The endangered riparian brush rabbit (Sylvilagus bachmani riparius; RBR) is a brush species found in a few areas of the San Joaquin Valley of California. The subject of large restoration projects, little is known of RBR habits or interactions. Camera traps can capture activity and behavior in their dense habitat, providing insight into their habitat usage and, potentially, their recovery. Originally a Master’s camera trap project with goals of studying behavioral interactions and usage of restored sites. Extreme flooding in 2017 stranded the study population with concentrated prey and predator populations. In response, the camera study shifted to capture this pulse event at San Joaquin River National Wildlife Refuge from February to August 2017. Camera traps captured RBR performing a variety of behaviors, including territorial and reproductive behaviors, despite the limiting circumstances. In contrast to the crowded conditions, most behaviors recorded were individual, suggesting a solitary lifestyle. Interspecific interactions were primarily with competitor species, mostly the desert cottontail (Sylvilagus audubonii). Further, camera traps captured RBR interacting with exotic plants and supplemental feed in addition to native plant species. Overall, this study highlighted the additional knowledge gained from behavior and camera studies.

In arid and semi-arid environments, burrowing mammals play a key role in increasing landscape heterogeneity through facilitative (positive) species interactions. The loss of burrowing mammal populations can consequently lead to negative effects cascading through the ecosystem; it is therefore critical to understand these facilitative interactions for conservation and management. For instance, wildlife translocations are a popular management tool that are often not successful. Accounting for facilitative interactions during translocation could improve success rates. To investigate the importance of burrow facilitation on San Joaquin antelope squirrel (Ammospermophilus nelsoni) translocation success, we designed a program for antelope squirrels using a natural experiment, with paired sites selected for the presence and absence of a burrowing facilitator, the giant kangaroo rat (Dipodomys ingens). We radio collared 97 antelope squirrels from core habitat within the Carrizo Plain. We then translocated 67 to uninhabited lands in their historical range, half to a site with giant kangaroo rats and half without. Over two sessions we observed differences in survival, dispersal and movement, and home range size across the three treatment groups. Our study highlights the importance of considering facilitative interactions when selecting release sites for translocation and planning restoration campaigns.

Stephens' kangaroo rat (Dipodomys stephensi; SKR)--recently downlisted from federally endangered to threatened--is protected across a network of reserve lands in Riverside and San Diego Counties, California. The diversity of ownerships and jurisdictions involved has made it difficult to coordinate management and monitoring efforts, or to track progress towards species recovery goals. An interagency team recently completed a rangewide SKR management and monitoring plan, which uses an innovative habitat model, created using variables derived from updateable satellite imagery, as a foundation. During summer-fall of 2022, cooperating agencies completed the first year of systematic population monitoring under the plan using standardized live-trapping grids randomly distributed on reserves across the range and stratified by habitat value--thus establishing the first rangewide baseline for tracking SKR status and trends. In addition to providing one-stop shopping for information about SKR biology, populations, threats, and management techniques, the plan will help diverse agencies be more efficient and effective in using their limited resources to recover the SKR.

Karyotypes differences between two subspecies of Pacific kangaroo rat Dipodomys agilis (2N=62 in the north and 2N=60 in the south), combined with minor allozyme and morphological differences, resulted in their re-classification into two the agile kangaroo rat (AKR, D. agilis) and the Dulzura kangaroo rat (DKR, D. simulans) in 1997. The distribution of AKR is from Kern and Santa Barbara counties down to upland areas the San Gabriel, San Bernardino, and Santa Monica mountains, while DKR occurs at lower elevations south from the foothills of those mountains to northern Baja California. It has been proposed that these two species are potentially sympatric in the foothill region of the San Gabriel mountains. We tested the validity of their species level classification, as determined by genomic divergence and the absence of introgression. To this end, we sampled AKR/DKR populations along a ~100 km transect crossing the San Gabriel mountains and performed whole genome sequencing/resequencing. Our results showed that, although AKR is found north and south of the mountains, there was no evidence of hybridization with DKR, supporting the specific status of the taxa. The divergence time between AKR and DKR was estimated as 0.5 to 1 Mya, and there is no proof of secondary contact since then. These populations encompassed a substantial elevational range, and we examined patterns of intraspecific genetic differences potentially linked to elevation.

Capture-recapture is a powerful, if expensive and difficult, approach to estimate abundance. Noninvasive genetic sampling is a potential alternative, but methods vary among taxa with little research on its use in rodents. We addressed a series of questions to develop a noninvasive genetic sampling approach for the endangered giant kangaroo rat (Dipodomys ingens). Over 2 years, we used 2 sampling grids, first live trapping and then collecting and genotyping fecal pellets. We found that fecal pellets were successfully genotyped up to 9 days (93%) after exposure, but that rate precipitously fell soon after. Although giant kangaroo rats are territorial, multiple individuals deposited fecal pellets at the same sampling locations; however, single pellets contained sufficient DNA to recover genotypes and to identify individuals, so contamination was not a problem. Per occasion capture probabilities were lower in noninvasive genetic sampling (p = 0.26, SE = 0.01) than from live trapping (p = 0.42, SE = 0.06). Population estimates were similar using noninvasive genetic sampling, although they were quite a bit higher (Ntrap = 36, Nfecal = 64) on one grid. Noninvasive genetic sampling should be tested in additional taxa and systems to provide more generalizable recommendations.