HR: 08:30h
AN: NB21E-01    [Abstracts]
TI: Use Of Statistical Modeling To Assist Phosphorous Criteria For Midwestern Streams
AU: * Smith, T B
EM: tbsmith@inhs.uiuc.edu
AF: Illinois Natural History Survey, 607 E. Peabody, Champaign, IL 61820
AU: Hill, W
EM:
AF: Illinois Natural History Survey, 607 E. Peabody, Champaign, IL 61820
AB: Statistical models of complex ecosystems often struggle with limited generality and weak evidence for causality. Science designed to support nutrient criteria must deal with these problems as well. To help provide a scientific basis for nutrient criteria in Illinois streams we are creating regression models with the widest possible scope and are testing experimentally for evidence of causation. Our analyses employ long term data sets, contemporary observations, and experimental tests of phosphorous dose-responses. Response variables include algal biomass, fish and invertebrate diversity and dissolved oxygen. Analyses conducted on long term data from functionally independent watersheds (N=84) revealed non-linear relationships between both fish and mussel species richness, and P and NH₃-N. Nutrient-biodiversity relationships were not apparent in smaller streams and collinearity between P and NH₃-N limits inference of any causal relationships. A multivariate model developed from intensive sampling within a single stream demonstrated a clear relationship between phosphorous, canopy cover, current velocities and alga growth (R²=0.552; p<0.001). However, extensive empirical observations across stream ecosystems showed no relationship between phosphorous and algae growth. Our data do not yet support use of an extensive statistical model for phosphorous criteria across the state of Illinois.
DE: 9810 New fields (not classifiable under other headings)
SC: North American Benthological Society [NB]
MN: 2005 Joint Assembly