Date of Award:

2016

Document Type:

Thesis

Degree Name:

Master of Science (MS)

Department:

Ecology

Advisor/Chair:

Charles P. Hawkins

Abstract

Two-thirds of United States stream length is in either fair or poor biological condition. However, we do not yet have reliable, quantitative tools to diagnose site-specific causes of biological impairment. One way to diagnose causes of impairment is to compare the environmental tolerances or preferences of the taxa expected at a site to those of the observed community. Stream ecologists have derived tolerance values (TVs) from field data for use in causal analysis, but inconsistencies across studies cast doubt on the accuracy of these TVs. Published TVs may not agree with one another for several reasons, including: differences in the methods used to estimate TVs, confounding among environmental variables, effects of environmental interactions on taxon abundance, and bias from incomplete sampling of a taxon’s niche space. My objectives were to determine if and how these four potential problems affect TVs and to determine if TVs can be used reliably in causal assessments. I collected macroinvertebrate, substrate, temperature, conductivity, and velocity data from 45 streams in the western U.S. Analyses of data from this field study suggest that incomplete sampling can profoundly influence TVs, whereas interactions may have little effect. Significant spatial correlations between environmental variables suggest that confounding may also affect the derivation of TVs, but the magnitude of this effect is unclear. Also, though each method used to estimate TVs has limitations, TVs derived from different methods appear to reveal the same environmental signal, and there may be little reason to prefer one method over another in a causal analysis. Over last few decades, researchers have begun to add biological traits, including estimates of preference and tolerance, to the suite of metrics used in bioassessments. As descriptors of important biological traits, TVs should help water quality managers reliably diagnose and combat the causes of biological stream impairment.

Checksum

d4e4f71ede50b27727323c1dbfff95e8

Share

COinS