Organization Title:United States Fish and Wildlife Service

A.Project Summary

Principle InvestigatorGregory R. Moyer (USFWS Regional Geneticist)

Co-PIs:Jason Duke (USFWS GIS Coordinator & IT Specialist)

Greg Rhinehart(USFWS, Cartographer)

PI Address:Conservation Genetics Lab; 5151 Spring Street, Warm Springs,

GA 31830. phone 706.655.3382 ext 1231; email

Project Title:Identifying priority areas for land protection in the south Atlantic: a landscape genetics pilot study

Focal Issue Addressed:Focal issue #3: Integrative projects that meet all aspects of the SALCC niche

Project Duration:Two years

Project Abstract:

This proposal seeks to develop a tool that strategically identifies priority areas for land protection. This is a pilot study to assess the extent of taxa that contain adequate genetic sampling within the south Atlantic ecoregion for characterization of intraspecific genetic variation. We seek to use genetic data from multiple taxa coupled with GIS data to provide a genetic landscape from which geographic patterns of intraspecific genetic diversity will be inferred. Joint analyses of the resulting genetic landscapes will be used to identify geographic areas where multiple species show atypical patterns of interpopulation divergence or intrapopulation diversity (i.e., a hotspot of high biological value). We will then assess the correlation (or lack thereof) between observed hotspots and current priority conservation areas.

B.Project Description:

Objectives

The goal of this project is to strategically identify priority areas for land protection or refuge design. Specific objectives of this pilot project include 1) compilation of previously published genetic datasets from species that contain adequate genetic sampling within the southeast Atlantic ecoregion; 2) characterization of geographic patterns of intraspecific genetic diversity and joint analyses of the resulting genetic landscapes for identification of geographic areas where multiple species show atypical patterns of interpopulation divergence or intrapopulation diversity (i.e., a hotspot) and 3) correlation of hotspots against current priority conservation areas.

As such, these data should provide a quantitative assessment of whether current priority conservation initiatives adequately protect areas of high biological value. If priority conservation areas are not correlated with areas of high biological value, then these data serve as a strategic approach to identify areas for land protection or refuge design.

Narrative

Background and Justification

The International Union for Conservation and Nature recognizes the need to conserve biodiversity at three levels: genetic, species and ecosystem diversity (McNeely et al. 1990). The importance of genetic diversity as a basis for future biological evolution and long term viability of populations, species, and ecosystems are well established (Frankel and Soule 1981; Frankham 2005; Laikre et al. 2010). Yet, while geographical patterns of species diversity are often considered in conservation assessments, the consideration of patterns of genetic diversity and recognition of conservation genetic concerns in practical management are largely lacking (Laikre 2010). This is unfortunate because an understanding of genetic diversity and how it is partitioned across the landscape can elucidate the evolutionary and contemporary processes that generate and maintain biodiversity, as well as, assist in monitoring of populations/species, defining conservation units, assessing population connectivity, determining critical habitat, and designating priority areas – many of which are important to the Service’s strategic habitat conservation initiative. Furthermore, these parameters can be modeled with existing climate change scenarios and urbanization projections to forecast future changes to critical habitat, population connectivity, and priority areas.

The spatial components of evolutionary and contemporary processes that generate and maintain biodiversity should be identified, mapped, and presented in an easily understood framework prior to their incorporation into conservation and management plans. Components of these processes (e.g., spatial connectivity) can be estimated by examining the spatial distribution of genetic variability within and among populations across multiple taxa. This type of an approach coupled with population genetic and GIS components (termed landscaped genetics) have the ability to reveal the geologic and ecological processes most important in shaping and maintaining contemporary genetic and species diversity patterns, presumably providing in sight for future management and conservation objectives. While our study focuses on the designation of priority areas, data gleaned from this study can be utilized in other aspects of conservation design and delivery (as mentioned abouve)– making landscape genetics an efficient and effective tool for conservation and management agencies.

Materials and Methods

Data and taxa. The primary purpose of this study is to assess the extent of taxa that contain adequate sampling within the south Atlantic ecoregion to characterize geographic patterns of intraspecific genetic diversity. If there is adequate coverage, implementation of our landscape genetics approach will not require any additional sampling or genetic data collection (thus a considerable cost savings to the SALCC). If there is not adequate coverage, our data will provide key information with regard to current knowledge gaps in taxon and genetic data collection for future landscape genetic proposals.

We will compile mitochondrial DNA and nuclear (microsatellite) datasets from as many aquatic and terrestrial species as possible (hopefully >20 species). Datasets for different taxa will be established from published and unpublished studies. In selecting studies, we will attempt to maximize taxonomic breadth while limiting inclusion of studies to those that sampled at least seven locations from throughout the south Atlantic ecoregion. We will also attempt to provide a relatively unbiased sample of genetic diversity within the ecoregion by sampling a breadth of taxa including invertebrates, amphibians, reptiles, birds, and mammals. For each species, collection location coordinates will be gathered from the primary studies or estimated using locality descriptions.

Justification of genetic data and priority of analyses. The utility of mtDNA data for landscape genetic inference is well-established (Avise 1992). Accordingly, it would be most informative to first focus on analyzing as many taxa as possible (our goal is > 20 taxa) using a single marker and rely on multiple taxa (rather than multiple genes) to cross-validate observed patterns. There are limitations to using any single gene to study evolutionary and demographic history (Ballard and Whitlock 2004); therefore, we will cross validate observed patterns by incorporating nuclear genetic variation where feasible (fewer microsatellite studies will be available when compared to mtDNA studies).

Estimating genetic divergence among populations. Each collection location (defined by each study) will be treated as a population for landscape level visualizations of genetic divergence. In most cases, large distances will probably separate these locations; however, multiple collection locations in close proximity may correspond to a single gene pool. In these cases, we will not combine any locations for landscape-level visualizations of genetic divergence, because (1) we are interested in mapping regions of high genetic similarity, as well as, high divergence (pooling samples would obscure these patterns), and (2) divergence patterns will be more accurately detected from multiple point locations scattered throughout a landscape than from fewer pooled samples each representing larger geographic areas.

We will calculate the mtDNA genetic divergence, DA, (Nei and Li 1979) between each pair of collection locations. Genetic distance estimates will be corrected for differing amounts of sequence evolution across data sets, using the 2-parameter model (Kimura 1980). Similarly, a chord distance (Nei 1987) will be used to estimate genetic divergence among populations for nuclear microsatellite data.

Visualization of genetic diversity across the landscape. Genetic distances will be visualized as genetic landscapes using ArcGIS 9.1. Euclidean distances will be calculated between all pairs of locations, and genetic distance will be regressed against (Euclidean) geographic distance using reduced major axis regression with the software IBDWS (Jensen et al. 2005).

The residuals from these regressions will be used to interpolate a genetic landscape for each species in ArcGIS9.1 using the following approach. First, a triangular irregular network will be constructed from the locations. The network connects all collection locations to their nearest neighbors with non-overlapping edges, forming irregularly distributed triangles. Second, residual DA values will be mapped to the geographic midpoints between locations along the edges of the network. By using residuals in these analyses we will be able to focus on regions of unusually high genetic divergence. Finally, a surface will be interpolated from the midpoint coverage using inverse distance weighted interpolation (power = 2, variable search radius with 12 points, grid cell size 1 km2). To avoid extrapolating beyond the original collection locations, individual species surfaces will be clipped to the extent of the original triangular irregular network or to the boundaries of the south Atlantic ecoregion in cases where this encompassed a smaller spatial extent.

All taxon specific genetic landscapes will be averaged to create a single multi-species genetic landscape to highlight areas of congruence. To assure that each species receives equal weight in the multi-species genetic landscape, we will compare a scaled vs. rank-scaled rescaling technique. The scaled divergence multi-species genetic landscape will be calculated by dividing each DAwithin a dataset by the maximum DA found in that data set. Thus, the maximum divergence for a species will be scaled to 1.0 before interpolating the surface. The resulting scaled multi-species genetic landscape will be an average of the individual species surfaces; therefore it could be subject to bias. To determine whether dissimilar skew and kurtosis across data sets is biasing results, we will also analyze a rank-scaled multi-species genetic landscape by rescaling the raw DA values

The spatial coverage of individual species’ genetic landscapes will not always be equal; therefore, the number of species represented in each 1 km2grid cell of the multi-species genetic landscape will also vary. To assess multi-species concordance, we will clip each multi-species genetic landscape extent to areas with coverage for three or more species. .

Estimating within-site genetic diversity. To estimate mtDNA genetic diversity within each collection location, we will calculate the average sequence divergence among individuals (Nei and Li 1979) under the Kimura (1980) 2-parameter model of sequence evolution. For nuclear microsatellite data, genetic diversity within collection locations will be calculated using the following indices: microsatellite allelic diversity, and FIS (FIS is the measure of departure from Hardy-Weinberg proportions within population).

Genetic diversity calculations are more dependent on sample size than the divergence calculations; therefore, we will pool collection locations less than five km apart in species datasets where individual sample sizes are low (1–3 individuals/location). Using inverse distance weighted interpolation as described above, we will interpolate the genetic diversity surface for each species and calculate the average diversity multi-species genetic landscape and variance surface for all datasets. In contrast to the genetic divergence multi-species genetic landscape, surface interpolation for the diversity multi-species genetic landscape will be conducted around the actual collection locations, rather than the midpoints between them.

Protection of hot spots. To determine whether hotspots are adequately protected under current land conservation efforts, the average genetic landscapes will be overlaid with a protected lands layer developed from a combination of sources. By examining the multi-species genetic landscapes in relation to these protected lands we will able to perform a gap analysis (Scott et al. 1993) to assess how well areas of atypical genetic divergence and diversity are currently protected.

Relationship to LCC Niche

Landscape-scale: How much of the LCC does the project cover? As envisioned the sampling strategy seeks to explore taxa that encompass the entirety of the SALCC. We will use taxa from aquatic, riparian, and terrestrial environments from throughout SALCC. It should be noted that the availability of genetic data is unknown which is why this is a pilot study. Once available knowledge of taxon specific data is gleaned from the literature, a sampling strategy can be designed that maximizes taxa from throughout the SALCC.

Cross-taxa: How well does the project support or integrate the needs of multiple natural and cultural resources? As stated previously, we seek to integrate multiple taxa from differing environments and can easily assess the correlation between presumed protected cultural resources and genetic hotspots.

Forward looking: Does the project predict future conditions or incorporate prediction of future conditions? Since this is a pilot study, this project does not seek to address future conditions; however, correlation of spatial genetic patterns with landscape or environmental variables can be applied to predict future changes in genetic structure and gene flow (connectivity or lack thereof) due to global and regional environmental changes (i.e., global warming, El Niño Southern Oscillation and hurricanes). These data can also assist in the determination key habitat requirements and migration corridors for analyzed species. Thus, these data allow for future analyses capable of predicting trends in habitat use and connectivity as well as loss of genetic diversity, reduction in census size, and population viability.

Decision focused: Does this project provide information vital to resource managers, policy makers, and conservation planners? The intent of this study is to provide resource managers, policy makers, and conservation planners a quantitative method to strategically assess current and future priority conservation initiatives. This study can also assist resource managers, policy makers, and conservation planners in monitoring of populations/species, defining conservation units, assessing population connectivity, and determining critical habitat, which are important parameters for strategic habitat conservation, protection of threatened and endangered species, and the Endangered Species Act.

Adaptive: Does the proposal provide information on how the product can be updated based on new data and information? Since this proposal, which is a pilot study, seeks to address data gaps and is also model –based, our proposal will be adaptive in the sense that as new data become available (estimates of genetic diversity, new GIS layers) these data will be incorporated in to all aspects of analyses to provide robust estimates of habitat connectivity, habitat use, and thus predictions of future habitat use and connectivity across the landscape. As new data become available, computer models can easily be rerun to assess model sensitivity and accuracy.

Making connections and filling gaps: How well does this project leverage and integrate existing efforts underway by various organizations? This study is unique in that it integrates genetic and GIS data to perform gap analysis. Thus the project leverages existing USFWS GIS support at no cost to the SALCC. Finally, once an initial pilot study is finished, we hope to fill in data gaps with assistance from the established USFWS and NPS Inventory and Monitoring programs.

Project Timeline

There are four major project tasks and include data compilation, data analysis, visualization of landscapes, and gap analysis (see Fig. 1). Data compilation and data analysis tasks will be completed in year-1 of the study with latter tasks being completed in year-2. A project report will be filed after year-1 with a synopsis of taxa and datasets available for landscape visualization. A final report summarizing our findings will be completed at the end of year-2. Depending on the number of taxa, taxon distribution, and number of genetic markers, the report will be submitted for peer-review publication; else, the report will outline necessary taxonomic gaps and sampling strategies necessary for accurate landscape visualization and gap analysis.

Figure 1. Estimated project time line.

C.Budget and Budget Justification

The PI requests funds totaling $71,750. All datasets will be collected from the primary literature; therefore, there will be no research associated costs; however, obtaining datasets from the primary literature will be time consuming. Funding is requested for a GS-9 full time employee for one year of initial data mining ($52,885 labor + $15,565 benefits). Addition travel expenses are also requested for discussion of project and collaboration between PIs ($3,000). In kind (matching) expenses totaling approximately $60,000 include two months of salary for G. Moyer for final report writing and approximately three months of salary for J. Duke and G. Rhinehart for GIS support and gap analysis.

D.Literature cited

Avise, J. C. 1992. Molecular population structure and the biogeographic history of a regional fauna: a case history with lessons for conservation biology. Oikos 63:62-76.

Ballard, J. W. O., and M. C. Whitlock. 2004. The incomplete natural history of mitochondria. Molecular Ecology 13:729-744.

Frankel, O. H., and M. E. Soule. 1981. Conservation and Evolution. Cambridge University Press, Cambridge, UK.

Frankham, R. 2005. Conservation biology: ecosystem recovery enhanced by genotypic diversity. Heredity 95:183.

Jensen, J. L., A. J. Bohonak, and S. T. Kelley. 2005. Isolation by Distance, Web Service. BMC Genetics 6:13.

Kimura, M. 1980. A simple method for estimating evolutionary rate of base substitution through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111-120.

Laikre, L. 2010. Genetic diversity is overlooked in international conservation policy implementation. Conservation Genetics 11:349-354.

Laikre, L., M. K. Schwartz, R. S. Waples, and N. Ryman. 2010. Compromising genetic diversity in the wild: unmonitored large-scale release of plants and animals. Trends in Ecology and Evolution 25:520-529.

McNeely, J., K. Miller, W. Reid, R. Mittermeier, and T. Werner. 1990. Conserving the World's Biological Diversity. IUCN, World Resources Institute, Conservation International, WWF-US and the World Bank: Washington, DC.

Nei, M. 1987. Molecular evolutionary genetics. Columbia University Press, New York.

Nei, M., and W.-H. Li. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences 76:5269-5273.