Research Interests ACER Lab
My research focuses on how spatial and temporal patterns of human stressors and land cover affect biodiversity. Land cover loss and degradation are the major threats to biodiversity, affecting 95% of imperiled terrestrial vertebrate species (Wilcove et al. 1988). There are few tools and little information to adequately understand and stem biodiversity loss, especially under scenarios of a changing climate. I use a hierarchical approach to address this problem in my research. First, I combine field experiments with spatial modeling to relate biodiversity and species distributions to human stressor intensity and land cover composition and configuration. Second, I study mechanisms of population limitation. I use land cover quality gradients identified in Step 1 to study variation in demography, micro-scale habitat use, and source-sink population dynamics. I also investigate methods to improve survey designs for sampling and monitoring biodiversity, and analytical approaches to estimate biodiversity. Lastly, I strive to produce tools and information from my research to guide local and regional conservation planning.
Below, I detail my research interests and outline my future research goals. Overall, my research focuses on five major areas. For each area, I briefly list up-to-date accomplishments.
1) How do spatial and temporal patterns of human stressors vary at regional scales?
A major thrust of my research is to assess spatial variation in cumulative effects of human stressors on ecological processes. I first published a paper on this question while working as an ecologist for the USGS (Leu et al. 2008). Focusing on the western U.S., I demonstrated that human stressor intensity is highest in low elevation areas that are highly productive. Since arriving at the College of William and Mary, my students and I have continued this type of research in both terrestrial and aquatic systems. The research has been funded by grants from the National Park Service, Fish and Wildlife Service, U.S. Geological Survey, and Virginia Space Grant Consortium.
For terrestrial systems, students and I work on three topics. First, Sean Mc Fall, Marie Pits, Corbet Wicks, and I modeled vehicular traffic volume for the conterminous U.S. Most studies evaluate road effects in two dimensions (length and width). In fact, we should evaluate road effects in three dimensions, where traffic volume is the third dimension. Traffic volume varies spatially and therefore so does road effects on ecological process. Over the past four years students and I collected traffic volume data from 48 State Departments of Transportation. We used kriging methods to interpolate traffic volume for primary roads (i.e., interstate highways) and secondary roads (i.e., state and federal highways). We finished traffic volume analyses in April 2014 and are now developing spatial models to predict where traffic volume disrupts connectivity for vertebrate species and how it influences human-caused fires and spread of exotic plants and insects. Our goal is to have the models published by spring 2015. Second, Jessica Pouder, Ellen Martin, and I assessed land use extent and intensity for the Mojave Desert region at scales overlapping with reptile, bird, and mammal home range sizes. Our paper indicating how vertebrates may perceive land use as a potential threat is currently in review and includes an undergraduate and graduate student co-author. The research produced one Master's Thesis (spring 2014). Third, Molly Teague, Aidan DeSena, Lauren August, Ellen Martin, Gabriel Games, and I assessed the accuracy of publically available land use spatial data either through field data or data that we hand digitized from aerial imagery. We found that available spatial data sets grossly underestimate the distribution of anthropogenic features. Many conservation and management plans are therefore potentially flawed because they underestimate land use effects on ecological processes. My students and I are currently working on publishing this research, having completed the Methods and part of the Results section thus far.
For aquatic systems, student and I have worked on two research topics. First, Robert Isdell, Char'Res Finney, Valerie Tran, Kimberly Zamuda, and I modeled dynamics of shoreline structure growth to evaluate estuary loss for the Virginia's portion of the Chesapeake Bay. We used aerial imagery to identify factors influencing changes in shoreline structure. Our analysis indicates that shoreline-structure change increases with the proportion of low-intensity development land cover, shoreline length, and housing density change, and decreases with proportion of marsh. This research produced a chapter in a Master's Thesis (fall 2013) and a paper that is currently in review in Applied Geography. Second, Ellen Martin, Bruce Pfirman, Aidan DesSena, and I assess factors influencing vernal pool persistence in Virginia's coastal region. In collaboration with James Vonesh and Anne Wright (Virginia Commonwealth University) and citizen scientists, we re-sampled vernal pools in 2012 that were originally sampled by Charles Blem in the late 1980's. My lab focused on relating land cover change between 1990 and 2011 to vernal pool persistence. Preliminary analyses found that vernal pool persistence is not influenced by land cover change. We aim to have our paper submitted for review by summer 2015. This research produced two presentations at a scientific meeting to educate citizen scientists.
Future Research Goals: D. M. Johnson (PI, Virginia Commonwealth University), S. J. Agosta (Co-PI, Virginia Commonwealth University), R. Dyer (Co-PI, Virginia Commonwealth), A. Eckert (Co-PI, Virginia Commonwealth), K. L. Grayson (Co-PI, Virginia Commonwealth), J. R. Brinkerhoff (Co-PI, Univ. of Richmond), J. D. Holland (Co-PI, Purdue University), K. J. Haynes (Co-PI, Univ. of Virginia), P. C. Tobin (Co-PI, U.S. Dept. of Agriculture, Forest Service), D. Parry (Co-PI, SUNY College of Environmental Science and Forestry), B. Aukema (Co-PI, Univ. of Minnesota-Twin Cities), K. F. Raffa (Co-PI, Univ. Wisconsin-Madison), and I have recently submitted a NSF proposal titled "Multi-scale drivers of invasion: Connecting local ecological processes and the human landscape to geographic spread." My section proposes to model probability of gypsy moth (Lymantria dispar) invasion from forest composition, forest configuration, and human aided dispersal of gypsy moth eggs via vehicular dispersal and fire wood transport. If funded, the proposed research will implement the vehicular traffic volume spatial data set developed by my lab.
2) What are the responses of species to human stressor intensity and patterns, land cover composition and configuration, and climate variability?
This major question forms the backbone of my current research while building on my work at the USGS. We investigate species distributions of terrestrial and aquatic vertebrates and invasive plants across regional/continental scales. From 2011-2015, my research has been supported by a grant from the Department of Defense's Strategic Environmental Research and Development Program (SERDP; RC-2202). The grant represents collaboration between Erica Fleishman (University of California, Davis), Brett Dickinson (Conservation Science Partners and Northern Arizona State University), Barry Noon (Colorado State University), and David Dobkin (High Desert Ecological Research Institute). My contribution to the proposal was development of the sampling designs for frogs, birds, and butterflies in eastern deciduous forests and modeling of biodiversity in relation to habitat fragmentation. Since 2012, Angela Zappalla, Vitek Jirinec, Alan Harris, Joseph Thompson, Kelly Watson, Daniel Ramos, Jessica Pouder, Chris Tyson, and I have collected data on the distribution of frogs, birds, butterflies, and exotic plants during a field season that extends from mid February to mid August. Within this research objective there are five separate projects, all of which spatially assess the effects of human stressors on biodiversity.
2.1) What are the effects of spatial and temporal land use patterns and intensity, and land cover composition and configuration on species occurring in the Intermountain West?
Since arriving at The College of William and Mary I completed six projects addressing this question, three of which I initiated when affiliated with the USGS. First, I finished writing a paper on human footprint impacts on Greater Sage-grouse (Centrocercus urophasianus). We found a negative correlation between Greater Sage-grouse population viability and human footprint intensity (Leu and Hanser 2011a). Second, I co-edited a peer-reviewed book in which I was lead author for two chapters (Leu et al. 2011b and 2011c) and co-author (see CV for complete list of citations) for nine chapters. The book summarizes a collaborative effort with Cameron Aldridge (USGS Fort Collins and Colorado State University) and Steve Knick and Steven Hanser (USGS Boise). All chapters were based on data collected over two years before I arrived at the College of William and Mary. We modeled the effects of human stressors like oil and gas development, biotic factors (i.e., land cover configuration and composition), and abiotic factors (e.g., terrain, climate, soil depth), on reptile, mammal, bird, and exotic plant distributions in the Wyoming Basin. We found that species responded differentially to land use and at different scales when controlled for abiotic and biotic factors. Third, I completed a paper on breeding-wintering ground connectivity for declining shrub-steppe bird species (Knick et al. 2014). Steve Knick, Steve Hanser, Kurt Fesenmeyer (all USGS), John Rotenberry (University of California, Riverside), and I used stable isotope signatures derived from feathers collected from birds captured on the breeding and wintering grounds to determine that breeding populations mix on the wintering grounds. We found that wintering location did not explain heterogeneity in breeding population trends. My contributions to this research consisted of developing the experimental design, collecting all of the field data, analyzing part of the data, and writing sections of the paper. Fourth, I finished co-advising my first graduate student with James Belthoff (Boise State University). Micah Scholer completed a study (spring 2011) on the effects of land cover configuration and composition and topography on the distribution of two small owl species using occupancy modeling (Scholer et al. In Press). This research identified both forest type and topography as important factors explaining heterogeneity in owl occupancy. Owls responded to these factors at scales between 0.27 - 3 km. Fifth, I modeled current habitat delineation, quality, and connectivity to project future location, quality, and connectivity of habitat under different scenarios of climate-induced land-cover change for 50 bird species breeding in the Great Basin (Fleishman et al. In press) in collaboration with Erica Fleishman (University of California, Davis), James Thomson (Monash University), Liz Kalies (Northern Arizona University), Brett Dickson (Conservation Science Partners and Northern Arizona University), and David Dobkin (High Desert Ecological Research Institute). My contribution to this paper consisted of conducting spatial analyses and writing sections of the paper. Sixth, Jessica Pouder, Ellen Martin and I developed habitat quality models for desert tortoise (Gopherus agassizii) and its predators in the Southwestern U.S. This research started in summer 2013 and is currently undergoing external review as a government publication.
Future Research Goals: E. Fleishman (PI, Univ. of California Davis), J. K. Balch (Co-PI, Pennsylvania State University), B. A. Bradley (Co-PI, University of Massachusetts), J. Dunham(U.S. Geological Survey), N. Horning (Co-PI, American Museum of Natural History), R. Mac Nally (Co-PI, University of Canberra), P. A. H. Westley (Co-PI, University of Alaska), and I have recently submitted a Department of Defense SERDP proposal titled "Interactions among non-stationary climate, land cover, fire dynamics, and species-environment relations in sagebrush steppe ecosystems" that was selected for full proposal after initial review (please see CV for complete author list). We propose to model how sagebrush ecosystem dynamics (Artemisia spp.) are influenced by invasion of the exotic cheat grass (Bromus tectorum), fire, land use, and climate change. We will use models to forecast sagebrush ecosystem dynamics under various climate regimes. We also propose to assess the effects of sagebrush loss on Greater Sage-grouse and other sagebrush-obligate species. For this proposal, I wrote sections on modeling land use patterns and sampling and modeling bird responses to sagebrush ecosystem loss.
2.2) What are the effects of land use pattern and intensity on interactions between large mammals, parasites, birds, and invasive plants?
Matt Feresten, Joanna Weeks, Corbet Wicks, Chris Tyson, Kelly Watson, Daniel Ramos, Morgan Nicolli, Vitek Jirinec, Angela Zappalla, and I initiated this research during my first field season at the College of William and Mary (2010). We are investigating how forest fragmentation affects the interaction of white-tailed deer (Odocoileus virginianus) habitat use, ticks densities, bird species densities and assemblages, and invasive plant distributions. White-tailed deer thrive in eastern deciduous forest landscapes fragmented by urbanization. As a result, deer populations have exponentially increased over the past three decades (Kilgo et al. 2010). We established a network of 130 sampling locations randomly placed along forest fragmentation and urban gradients on public lands on the Virginia peninsula, between Richmond and Hampton. We have collected four years of data on avian abundance, avian community structure, and deer habitat use. We collected data on tick abundance in 2010 and 2013, exotic plant distribution in 2010, 2011 and 2014, and vegetation composition in 2011 and 2014. Preliminary analyses indicate that white-tailed deer browsing significantly decreases avian diversity in the understory and that higher deer habitat use results in higher tick abundance. In collaboration with Oliver Kerscher (College of William and Mary), our research suggests that tick densities and prevalence of tick-borne disease, evaluated with DNA markers, are highest where humans recreate. This research has produced one Master's Theses (spring 2013), two Honors Theses (spring 2013), five presentations (posters = 2 and talks = 3) at local and international scientific meetings, and one publication in progress on how deer browsing affects the avian community. It has fostered collaborations with Daniel Cristol, Oliver Kerscher, and John Swaddle at the College of William and Mary, and Holly Gaff at Old Dominion University.
2.3) How does variation in frog colonization and extinction rates, species richness, and pond occupancy relate to land use and land cover configuration and composition?
Daniel Ramos, Kimberly Zamuda, and I use a novel approach to evaluate factors influencing pond-breeding frog occupancy by relating them to three life-history stages: breeding season, migration season (i.e., round-trip movement between breeding and wintering areas), and dispersal (one-way movement). We also evaluate whether species respond to land cover and land use at different scales in each of three life-history stages. We established a network of 51 randomly selected ponds on the Virginia Peninsula along an urban-rural gradient. By the end summer 2014, we will have collected four years of occurrence data for frog assemblages breeding either during the early, mid, or late breeding seasons (i.e., three surveys for each of three breeding seasons). Our analyses indicate that species respond differentially and at different scales to land use, land cover configuration, depending on the life-history stage. We are currently working on an analysis based on multi-stage occupancy models. This will allow us to relate chorus intensity to land use and landscape composition and configuration. Ultimately, this research will be used to develop connectivity models for pond-breeding frog species on the Virginia Peninsula. Thus far this research generated one Master's Thesis (fall 2012), two student oral presentations at national scientific meetings, and an invited seminar at the Biology Department, Old Dominion University. We are currently refining the Master's Thesis into a publication.
2.4) What are the effects of human land and aquatic modifications on the connectivity of the aquatic-terrestrial ecotone?
Robert Isell and I study the effects of anthropogenic terrestrial and aquatic stressors on terrestrial-aquatic connectivity in collaboration with Randolph Chambers (College of William and Mary) and Donna Bilkovic (Virginia Institute of Marine Science). We chose an estuarine turtle, the diamondback terrapin (Malaclemys terrapin), as a model species. This species nests on land like all turtle species, but occupies estuaries throughout the non-breeding season. In 2012 and 2013, we conducted surveys at 165 sites in the Virginia portion of the Chesapeake Bay from late spring to mid-summer. We used occupancy modeling to evaluate associations between diamondback terrapin occurrence, land use, salt marsh, shoreline armoring, and crabbing intensity at scales overlapping with daily and annual terrapin movements. We are the first to identify thresholds of terrapin responses to habitat and human stressors and to quantify negative associations between the distribution of diamondback terrapin and alterations to terrestrial-aquatic connectivity including agriculture and shoreline armoring. The research has produced one Master's Thesis (fall 2013), two student oral presentations at local and national meetings, and one publication in review in Diversity and Distributions.
2.5) What are the effects of land use pattern and intensity on butterfly diversity?
Angela Zappalla and I investigate the degree to which land cover configuration and composition affects diversity of butterflies in collaboration with Erica Fleishman (University of California, Davis). We predict that moderate land use will increase butterfly diversity because moderate disturbance increases edge effects and therefore nectar availability. We established 68 study sites that differ in forest type, upland vs. riparian forest, and edge effects, on the edge vs. forest interior. We have collected butterfly occurrence data for the past three years. This research currently supports a Master's student.
Future Research Goals: I plan to continue the long-term study on bird, frog and butterfly diversity. The study sites are embedded in rapidly changing landscapes in which forests are converted to low-density housing areas and associated infrastructure. This will provide opportunities for my students and I to evaluate effects of forest fragmentation/loss on biodiversity using before-and-after experiments.
3) What are the mechanisms of species distribution?
Since my arrival at the College of William and Mary, my lab has amassed extensive data sets on the distribution of species and biodiversity (birds = 5 years, frogs = 4 years, and butterflies = 3 years). We are now well positioned to test hypotheses on how biodiversity is affected by land use patterns and intensity, and on the mechanisms of species distributions. I am particularly interested in studying sink-source population dynamics for birds and frogs.
I currently mentor three graduate students who address mechanistic questions regarding species distributions for their Master's Thesis research. First, Vitek Jirinec and I investigate habitat use of the Wood Thrush (Hylocichla mustelina) using locations collected from radio-collared males and females. This species has experienced long-term range-wide declines over the past few decades. A number of studies indicate that the Wood Thrush is an area-sensitive species; however, our five-year point count study provides little evidence for decreased probability of occurrence in smaller forest fragments. We find Wood Thrush to occur in forest patches embedded in a matrix of low-density housing development. Second, Angela Zappalla and I study how variation in eastern swallow tail (Papilio glaucus) occupancy is influenced by land use, land cover, and nectar availability. Third, Katie MacCorkmick and I investigate the interaction of dispersal, anthropogenic disturbance, and competition for suitable micro-habitat sites for the invasive sweet fennel (Foeniculum vulgare). These research projects are projected to be finalized as Master's Theses in spring of 2015.
4) How can methods of assessment of species richness and occupancy across space, time, taxanomic groups, and ecoregions be improved?
This research investigates the efficacy of sampling techniques and estimation of biodiversity when detection of species is imperfect. It is in collaboration with Erica Fleishman (University of California, Davis), Brett Dickinson (Conservation Science Partners and Northern Arizona State University), Barry Noon (Colorado State University), and David Dobkin (High Desert Ecological Research Institute). Our goal is to develop and validate practical, statistically rigorous, and transferable methods to estimate species richness and occupancy across space and time. We focus on three taxonomic groups: birds, butterflies (in both the eastern and western U.S.), and pond-breeding frogs (in the eastern U.S.). We are in the third and final year of occurrence data collection. Our research shows that the closure assumption of occupancy modeling is violated when traditional bird and butterfly methods are employed. We also found that traditional anuran sampling protocols with one survey per breeding season are insufficient to capture heterogeneity in vocalization activity due to variability in environmental factors. Our research will suggest alternate sampling approaches. This research has produced one Honors Thesis and two publications in progress, which will be submitted for review by the end of July 2015. We project to publish about 15 additional publications from this research.
5) Implementation of my research to inform conservation strategies.
Land managers are required to make long term plans for ecosystem management and conservation of endangered and threatened species, such as Natural Resources Conservation Assessments (NRCA). However, they often have little information to make these decisions across such broad areas. The applied component of my research addresses this critical gap. Since summer of 2012, I collaborated with colleagues in academia (University of California, Davis, University of Massachusetts, and University of Nevada, Reno) and nonprofit conservation organizations (Conservation Science Partners and Great Basin Bird Observatory) to develop a NRCA for the National Park network in the Mojave Desert. Our research focused on current and projected land use and habitat quality models for threatened species and their predators. The NRCA is currently in review and includes one undergraduate and graduate student co-author. It produced one Master's Thesis (spring 2014) and two student poster presentations at a local and an international meeting. In addition, I collaborated with Aaron Haines, Kerry Reese, Mike Scott, Leona Svancara, and Keri Vierling (University of Idaho) on three related projects. First, we developed a conservation assessment tool that incorporates spatial-temporal changes in human footprint extent and intensity (Haines et al. 2008). Second, we identified areas of special conservation status for birds in the state of Idaho using a weighted biodiversity hotspot approach (Haines et al. 2010), where "hotspots" indicate species sensitivity to various human threats (i.e., urban development, agriculture, fire suppression, grazing, roads, and logging). Lastly, we delineated hotspots for terrestrial and aquatic species of concern and spatially related them to the extent of human stressors and the distributions of species of concern. We identified areas of conservation concern at the county level so that the management of threatened and endangered species can be integrated with county comprehensive plans (Haines et al. 2012). On the above three papers, I heavily contributed to spatial analyses and wrote major parts of the papers.
Fleishman, E., J. R. Thomson, L. Kalies, B. G. Dickson, D. S. Dobkin, and M. Leu. In Press. Projecting current and future location of habitat for breeding birds in the Great Basin. Ecosphere.
Haines, A. M., M. Leu, L. K. Svancara, J. M. Scott, K. P. Reese. 2008. A theoretical approach to using human footprint models to measure landscape level conservation success. Conservation Letters 1: 165-172.
Haines, A. M., M. Leu, L. K. Svancara, G. Wilson, and J. M. Scott. 2010. Using a distribution and conservation status weighted hotspot approach to identify areas in need of conservation action to benefit Idaho bird species. Northwest Science 84:170-182.
Haines, A. M., M. Leu, L. K. Svancara, J. M. Scott, K. Vierling, S. Martinuzzi, T. J. Laninga. 2012. Incorporating wildlife conservation into county comprehensive plans. Northwest Science 86:53-70.
Kilgo, J. C., H. S. Ray, C. Ruth, and K. V. Miller. 2010. Can Coyotes Affect Deer Populations in Southeastern North America? Journal of Wildlife Management 74:929-933.
Knick, S. T., M. Leu, J. T. Rotenberry, S. E. Hanser and K. T. Fesenmyer. 2014. Diffuse migratory connectivity in two species of shrubland birds: evidence from stable isotopes. Oecologia 174: 595-608.
Leu, M., S. E. Hanser, and S. T. Knick. 2008. The human footprint in the west: a large-scale analysis of anthropogenic impacts. Ecological Applications 18:1119-1139.
Leu, M., and S. E. Hanser. 2011a. Influences of the human footprint on sagebrush landscape patterns: implications for sage-grouse conservation. Pp. 253-271 in S. T. Knick and J. W. Connelly (editors). Greater sage-grouse ecology and conservation of a landscape species and its habitat. Studies in Avian Biology (Vol 38). University of California Press, Berkeley, CA.
Leu, M., S.E. Hanser, C.L. Aldridge, B.S. Cade, and S.T. Knick. 2011b. A sampling and analytical approach to develop spatial distribution models for sagebrush-associated species. Pp. 88-111 in Hanser, S.E., M. Leu, S.T. Knick, and C.L. Aldridge (editors). Sagebrush ecosystem conservation and management: Ecoregional assessment tools and models for the Wyoming Basins. Allen Press, Lawrence, KS.
Leu, M., S.E. Hanser, C.L. Aldridge, S.E. Nielsen, and S.T. Knick. 2011c. Occurrence of large and medium-sized mammal: occurrence but not counts models predict pronghorn distribution. Pp. 315-336 in Hanser, S.E., M. Leu, S.T. Knick, and C.L. Aldridge (editors). Sagebrush ecosystem conservation and management: Ecoregional assessment tools and models for the Wyoming Basins. Allen Press, Lawrence, KS.
Scholer, M. N., M. Leu and J. R. Belthoff. In Press. Factors associated with flammulated owl and northern saw-whet owl occupancy in southern Idaho. Journal of Raptor Research.
Wilcove, D. S., D. Rothstein, D. Jason, A. Phillips, and E. Losos. 1998. Quantifying Threats to Imperiled Species in the United States. BioScience 48:607-615.