Date of Graduation

12-2014

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Biology (PhD)

Degree Level

Graduate

Department

Biological Sciences

Advisor

Daniel Magoulick

Committee Member

Gary Huxel

Second Committee Member

Steven Stephenson

Third Committee Member

Jason Tullis

Keywords

Coleoptera, Eco-hydrology, Geoinformatics, Insect Trap, Spatial Scale

Abstract

An investigation of endangered American burying beetle (Nicrophorus americanus) ecology led to development of a Geodata Crawler with applications in eco-hydrology. Geodata Crawler includes a national GIS (geospatial information systems) database with layers that quantify climate, land cover, soils, human development, and other attributes of the biosphere. For user-locations in the continental United States, Geodata Crawler can rapidly tabulate site-specific statistics within automatically delineated sample areas: points, site radii, watersheds, and riparian zones, among others. Geodata Crawler supported a multi-scale analysis of N. americanus habitat at a military installation in western Arkansas to produce a Landsat-based monitoring tool. Royle's N-mixture model was used to simultaneously account for 1) the detection process associated with baited pitfall traps, and 2) the ecological processes driving spatial patterns of abundance. Detection rates of N. americanus averaged 20% and were optimized at about 29º C on nights with high humidity and slight wind. Effective sample radii assessed using marked beetles released at known locations were no more than 800 m, and detection rates dropped below 5% beyond 400 m. Nicrophorus americanus abundance was associated with native grasslands and open-canopy oak woodlands with rolling topography, sandy loam soils, and moderate disturbances from wildfire. Habitat measured within 800 m site radii produced best fitting models compared to 100 or 1600 m radii. A new above-ground bucket trap was evaluated in comparison to standard pitfall traps. Compared to standard pitfall traps, above-ground bucket traps were safer for beetles, more resistant to scavengers, and more time-efficient for workers to install. The first application of Geodata Crawler for aquatic ecology was an eco-hydrology project that identified seven natural flow regimes of the Ozark-Ouachita Interior Highlands based on daily hydrological data from 64 reference streams. Geodata Crawler quantified climate and catchment characteristics necessary to predict natural flow regimes of 24,557 un-gaged stream segments. This dissertation demonstrated the utility of Geodata Crawler for eco-hydrology and species distribution modeling. Development will continue to expand potential applications to include landscape genetics and climate change, and also to support web-based project submission, cluster computing, and FTP-based data retrieval.

Share

COinS