| Popis: |
To protect and restore the biological integrity of streams, we need to be able to both detect biological degradation and infer likely causes of impairment. Managers often use biological indices to measure biological condition and detect degradation. However, the ability to detect degradation can be limited by the performance of the indices we develop. Index performance varies widely, but the sources of this variation are often unclear. In addition, although bioassessments are useful tools for detecting biological degradation, they do not identify stressors associated with impairment. My thesis research had two general goals: 1) develop statistically and ecologically robust indices to measure biological condition in Nevada streams and 2) quantify relationships between land uses, stressors, and biological condition to infer likely causes of degradation. I developed two biological indices for Nevada streams, a multimetric index (MMI) and observed to expected (O/E) taxa ratios, and determined if index performance was related to site isolation and sample evenness. The Nevada O/E indices were relatively imprecise compared with those from other regions, which likely results from low assemblage predictability associated with spatial isolation of aquatic habitats in arid regions. In contrast, the Nevada MMI was more precise than most previously developed MMIs, likely the result of using models to reduce natural variation in index scores. Sample evenness was positively associated with both O/E and MMI scores. Adjustments of index scores for sample evenness increased index precision, but also altered relative differences in index values and therefore inferences of biological impairment at specific sites. I also quantified relationships between biological condition, instream stressors, and land uses and used a weight of evidence approach to infer likely causes of degradation. Land uses such as agriculture, urbanization, and mining were associated with the spatial distributions of instream stressors, and these stressors were associated with variation in biological condition. Total dissolved solids and metal contamination were the stressors most strongly associated with biological condition. By detecting biological degradation and identifying important stressors and their potential sources, the tools I developed should help managers target conservation and restoration efforts and improve their ability to protect freshwater resources. |
