Current Projects
Phyllostomid Bats in Costa Rica
Student: Kate Cleary
Our research investigates structural and functional connectivity for important frugivorous and nectivorous bat species (Phyllostomidae) in the San Juan-La Selva (SJLS) biological corridor in northern Costa Rica. Bats are an ideal taxon for addressing native species’ response to forest fragmentation and land use change in the matrix between habitat patches in fragmented landscapes. Of the 139 mammal species in the SJLS, 79 are bats. In addition, nectivorous and frugivorous bats are the primary seed dispersers and pollinators of hundreds of tree species, and thus play a key role in sustaining plant biodiversity. Fragmentation and land use change in the matrix, especially agricultural intensification, may present a significant challenge for bats to maintain reproductive connectivity within their own populations as well as for associated mutualistic tree species. Given their diversity, abundance, and key role as seed dispersers and pollinators, measuring how well bats are able to move through human-dominated landscapes for foraging and reproduction is an important component of estimating overall ecological resilience.
We use both established field methods and innovative genetic techniques to investigate structural (i.e., patch size and isolation) and functional (i.e., permeability of the agricultural matrix) connectivity for Phyllostomid bats at both the community and species level. At the community level, we aim to compare bat species richness, diversity, and relative abundance in continuous forest with these same metrics in patches of differing size and isolation and embedded in different land cover types. At the species level, we focus on two frugivorous species of differential mobility, Artibeus jamaicensis and Carollia castanea. For these species, we are using microsatellite markers to compare genetic diversity within and gene flow between forest patches of differing size and isolation and embedded in different land cover types. We are also using mitochondrial DNA markers to characterize historical genetic variation in these two species in the SJLS, which will improve our estimates of the proportion of genetic structure attributable to recent land use change.
This project will make an important contribution to wildlife conservation in human-dominated landscapes and develop methodologies applicable to any human-dominated landscape worldwide. This work is part of a larger interdisciplinary project focused on quantifying social and ecological resilience in the SJLS.
Detection of aquatic macroorganisms using environmental DNA
LEECG researchers: Caren Goldberg, Stephen Spear, Lisette Waits
UI collaborators: Katherine Strickler and Alexander Fremier
We have recently expanded the research capacity of the LEECG to include studies of environmental DNA (eDNA), where we use DNA found in wetlands, rivers, and streams to detect aquatic macroorganisms. We built on the established expertise of the LEECG in analyzing low-quality DNA samples to establish this program developing new methods and applications in eDNA research. Our pilot work in this field (Goldberg et al. 2011) demonstrated that it is possible to detect rare stream-breeding amphibians by identifying their eDNA in streamwater with high detection probabilities and we are continuing to work in this system testing different collection and analysis techniques. Our projects now include developing and demonstrating eDNA techniques for monitoring amphibian, aquatic reptile, mudsnail, and fish populations in Arizona, Florida, Georgia, Idaho, Montana, North Carolina, Tennessee, and Washington. Our collaborators on these projects include the Arizona Game and Fish Department, Columbia River Inter-Tribal Fish Commission, Department of Defense, Georgia Department of Natural Resources, Idaho Department of Fish and Game, Lee University, North Carolina Wildlife Resources Commission, North Carolina Zoo, The Orianne Society, Tennessee Wildlife Resources Agency, University of Georgia, Virginia Polytechnic Institute, Washington Department of Fish and Wildlife, Yakima Basin Fish and Wildlife Recovery Board, and USGS. A fact sheet we are developing with USGS on eDNA techniques is available here. More information about our work with the Department of Defense can be found here.
CSI Newfoundland: molecular determination of caribou calf predators
Student: Matt Mumma
The Newfoundland woodland caribou (Rangifer tarandus) population has decreased by greater than 66% since 1996. Habitat loss and overgrazing are likely implicated, but current recruitment is low due to high calf predation by black bears (Ursus americanus) and invasive coyotes (Canis latrans). Previously, kill site observations and necropsy results, when available, were used to assign the predator species to calf predation mortalities, but 26% of kills were unable to be assigned to a predator species. We used molecular techniques to identify the predator species at caribou calf kill sites in Newfoundland, Canada. In 2010, radio-telemetry collars were placed on 1 to 3 day-old caribou calves and calves were monitored from June through October. When a mortality signal was detected, the collar location was investigated to determine if predation had occurred. Calf mortality sites were searched for predator scat and hair. Calf carcasses were inspected for killing bite wounds as evidenced by hemorrhaging. An ethanol-soaked cotton swab was used to sample killing bite wounds for predator saliva cells. Other non-killing wounds were also sampled. In the absence of a carcass, bones, hide, and/or the collar were swabbed. Scat, hair, and swab samples were analyzed using several DNA species identification tests to distinguish among black bears, coyotes, lynx (Lynx canadensis) and red foxes (Vulpes vulpes). Molecular techniques identified a predator species at 92% of kill sites. None of these kill sites identified more than one predator species. 70% were attributed to coyotes and 30% were attributed to black bears. There was a 100% success rate in identifying the predator species from killing bite wound swabs. 75% were identified as coyotes and 25% were identified as black bears.
Evaluating noninvasive genetic monitoring for endangered pygmy rabbit recovery efforts
Student: Steph DeMay
Genetic monitoring of reintroduced pygmy rabbits (Brachylagus idahoensis) is crucial for assessing the success of recovery efforts for an endangered population in central Washington. Noninvasive genetic sampling of faecal pellets could be a valuable method. We evaluated the effect of sample age on DNA degradation in faecal pellets under summer field conditions. We placed 275 samples from known individuals in natural field conditions for 1 to 60 days before DNA analysis. We assessed DNA quality by amplifying a species-specific mitochondrial DNA (mtDNA) locus and 5 nuclear DNA (nDNA) microsatellite loci. Both mtDNA and nDNA exhibited high PCR success rates (94.%) in samples <1 day old. Success rates for microsatellite loci declined rapidly from 80.0% to 42.7% between days 5 and 7, likely due to increased environmental temperature. Success rates for mtDNA amplification were higher, with moderate success (66.7%) at 21 days. Logistic regression modelling indicated that DNA degradation was influenced by sample age, DNA type, locus length, and rabbit sex. Allelic dropout rates were relatively high (17.6% at < 1 day) and increased to 100% over time. False allele rates ranged from 0 to 30.0% and increased over time. We recommend collecting samples as fresh as possible for individual identification, ideally within 5 days. Our study suggests that this method can be useful for future monitoring efforts, including occupancy surveys, individual identification, population estimation, parentage analysis, and monitoring of demographic trends.
Student: Joel Tebbenkamp
The Bi-State population of greater sage-grouse (Centrocercus urophasianus) is a Distinct Population Segment and candidate for listing under ESA. The status of the Bi-State population is based largely on unique genetics and long-term geographic isolation from other greater sage-grouse populations. The Bi-State population includes multiple localized populations, with some that appear demographically isolated, exhibiting lower vital rates and population trends. Most notably, one of the populations exhibited low reproductive success due to nonviable eggs, which we hypothesized may be due to isolation and low genetic diversity. To help facilitate a more comprehensive understanding of the dynamics within the Bi-State population our objectives were to identify genetically distinct populations, assess genetic diversity, and identify first-generation dispersers using molecular techniques. Through capture and noninvasive genetic sample collection during 2007 - 2011 we sampled 331 sage-grouse throughout Mono County, California, characterizing each individual at 17 nuclear microsatellite loci. Pairwise FST between localized populations ranged from 0.065 - 0.246. Combining the results of the pairwise FST estimates and 2 Bayesian clustering analyses we identified 5 genetic populations. Genetic diversity indices (He = 0.58 - 0.60, AR = 2.86 - 3.19) were not significantly different between populations, suggesting low genetic diversity was not likely causing egg non-viability. We identified 9 first-generation dispersers, which combined with the relatively homogenous genetic diversity indices illustrates the population’s natural ability to avoid genetic isolation under the current landscape conditions.