Project Title: Road Networks, Land Cover and Ecological Services: Thirty Years of Change in Colorado and Florida
Mendenhall Fellow: Alisa Coffin , (970) 226-9480, firstname.lastname@example.org
Duty Station: Fort Collins, CO
Start Date: February 2, 2009
Education: Ph.D., Geography, University of Florida, 2009
Research Advisor: Ray Watts, email@example.com
Project Description: Road networks are among the most ancient and pervasive features in the landscape—the physical evidence of human social interactions, expressing cultural and economic interests. Roads provide conveyance for people, ideas and resources to be introduced, extracted, transported and processed for exchange. The aim of this research is to evaluate the consequences of road network development on key ecosystem services by comparing and contrasting two study areas: the Colorado Front Range and the Santa Fe River watershed of north-central Florida. This research builds on previous work examining road network development and landscape dynamics in both Florida and Colorado (Coffin, 2009; U.S. Geological Survey, 2007a). The similarities of the roads datasets and the differences of the study areas will allow for unique inter-site comparisons that consider temporal as well as spatial dynamics. The goal is to determine whether indices of road network structure and function are valid indicators of landscape changes that directly influence ecosystem services.
The thematic and temporal similarities of the datasets (that is, road networks for several decades of the 20th century) will allow for the comparison of similarities of structure and function of the road network while contrasting the differences, which may be attributed to biophysical and socio-economic contexts. But, more importantly, setting indicators of avian diversity against measures of road network structure and change in the twin study areas will provide insights into how road networks may be influential agents of change in these regions and ecological domains. In this study the BBS will be used to relate changes to the road network with land cover and changes in avian diversity. Parameters for study may include rates of change within guilds of birds, or changes in the structure of multiple guilds (for example, multi-guild population trends reduced to low order such as increase/decrease).
Biological diversity is an important indicator of ecosystem services. For this study, changes in breeding bird diversity will be evaluated relative to changes in the road networks in the study areas. Models that test questions of biodiversity rely on the results of animal and plant population monitoring programs. Included among these are efforts that regularly survey animal populations such as the North American Breeding Bird Survey (BBS) (U.S. Geological Survey, 2007b). The BBS is a valuable long-term dataset that can be used to model trends in bird populations (Link and Sauer, 1998). The BBS is an avian count survey where the same sample units, which are predetermined routes along roads, are evaluated every year. Each BBS route is assumed to represent an undefined area in which it is embedded. Studies using the BBS have been able to show links between changes to landscape context and bird populations. Many studies have used either buffers around the route or circles centered on the route centroid for areal analysis of land cover and pattern (Pidgeon and others, 2007). This approach runs counter to connectivity notions, which would suggest that the area surrounding a BBS route should morph functionally, which is to say uniquely for each route and potentially each time period. Even if the area remains constant, much more could be done to understand differences and changes in functional connectivity.
<<-- Aerial photo of a dead-end road and agricultural field in the Santa Fe River watershed, Florida. An increase in the proportion of dead-end roads contributed to road network connectivity changes in the watershed from 1975 to 2005.
Inferential statistical models will be used that incorporate landscape road network structure as independent variables and ecological service indicators of as dependent variables. Modeling methods may include a variety of global statistical models including a multivariate logistic regression model, as well as models that account for local variability and spatial autocorrelation such as spatial autoregressive models and geographically weighted regression models. Decisions about precisely which models to use will be deferred until the distributional properties of the parameters are more clearly understood.
The objectives of this research will be:
- To characterize, compare and contrast changes to the structure and function of the road networks and the landscape circumscribed by the roads (e.g. land cover, connectivity, roadless volume, network structure).
- To characterize, compare and contrast changes in avian diversity.
- To model the relationship between indicators of road networks and avian diversity.
Coffin, A. 2009, Road network development and landscape dynamics in the Santa Fe River watershed, north-central Florida, 1975 to 2005: Gainesville, University of Florida, Ph.D. dissertation.
Link, W.A., and Sauer, J.R., 1998, Estimating population change from count data: application to the North American Breeding Bird Survey: Ecological Applications, v. 8, p. 258–268.
Pidgeon, A.M., Radeloff, V.C., Flather, C.H., Lepczyk, C.A., Clayton, M.K., Hawbaker, T J., and Hammer, R.B., 2007, Associations of forest bird species richness with housing and landscape patterns across the U.S.A.: Ecological Applications, v. 17, p. 1989–2010.
U.S. Geological Survey, 2007a, CUES Comprehensive Urban Ecosystem Studies: Colorado CUES overview [http://rockyweb.cr.usgs.gov/cues/COcuesHome.html]. Last accessed on 21 April 2009.
U.S. Geological Survey, 2007b, North American Breeding Bird Survey [http://www.pwrc.usgs.gov/bbs/]. Last accessed on 21 April 2009.
Previous Profile Project Profiles Next Profile
Direct inquiries to Rama K. Kotra at firstname.lastname@example.org
Page Contact Information: Mendenhall Postdoctoral Fellowship Program Web Team
Last modified: 16:08:26 Thu 13 Dec 2012