OOS 3-10 - Projecting future conditions of riverine ecosystems using a macroecological modeling approach

Monday, August 8, 2011: 4:40 PM
12A, Austin Convention Center
Kevin E. McCluney, Department of Biological Sciences, Bowling Green State University, Bowling Green, OH, N.L. Poff, Gdpe, Colorado State University, Fort Collins, CO, Geoffrey C. Poole, Department of Land Resources and Environmental Sciences, Montana State University, Bozeman, MT, James H. Thorp, Ecology and Evolutionary Biology, Kansas Biological Survey, University of Kansas, Lawrence, KS and Margaret Palmer, Chesapeake Biological Laboratory, University of Maryland, Cambridge, MD
Background/Question/Methods

There are many excellent models of hydrologic processes and water quality across a range of scales.  There are also many broad conceptual models of riverine ecosystems and much theoretical understanding of riverine ecology at small scales (1-100m).  However quantitative models of riverine ecology are typically limited to correlative relationships between habitat and species occurrence of one to a few species. Understanding the relative importance of ecological theory derived at small scales and landscape patterns in determining broad patterns of ecological response is critical in efforts to project future conditions and consequences of climate change, land-use change, and management scenarios.

Results/Conclusions

We posit a new approach to modeling riverine ecosystems, scaling local ecological theory to landscapes by abstracting the watershed, river, and floodplain into a network of heterogeneous patches connected by fluxes of materials.  This system could produce estimates of habitat characteristics and patterns useful in testing theories of ecological responses and developing new landscape-level theories.  For instance, we predict that: 1) increased patch complexity will result in greater diversity and greater persistence of species with immobile traits, 2) juxtaposition of two patches with dissimilar physical characteristics will result in greater diversity of both patches and 3) we should find a hump-shaped relationship between heterogeneity of patch types and diversity within any collection of contiguous patches.  Coupled with empirical data, such a system should allow identification of the most important predictors of broad patterns of ecological response, facilitating projection of future scenarios.  Additionally, by taking a traits-based approach, predictive power might be improved for multiple species and comparisons of risk could be made within and between large basins.  Our proposed approach can draw from a large body of existing work in riverine modeling and numerous data sources, connecting local and landscape predictors of ecological response across large basins.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.