| spelling |
ndltd-OhioLink-oai-etd.ohiolink.edu-osu13067824612021-08-03T06:02:58Z Ecological and management implications of multi-scale environmental influences on stream fish assemblages: evidence from Ohio and Idaho, USA Kautza, Adam Robert <p>Despite increased attention to spatial considerations in watersheds, the relative influences of multiscale abiotic factors on stream fish assemblages is not fully understood. To this end, I compared the relative influences and spatial relationships of abiotic environmental factors at landscape (i.e., watershed) and local (i.e., reach) scales on characteristics of stream fish assemblages between two geographic regions (Idaho and Ohio, USA) characterized by distinct environmental conditions and levels of anthropogenic impacts. I surveyed fish assemblages, collected habitat and geomorphic data, and used a GIS to quantify landscape-scale land use/land cover and drainage area at 32 stream reaches. Using a partial redundancy analysis, I found that environmental factors at the landscape-scale accounted for 36% of the variation in fish assemblages in Ohio, with land-use/land-cover factors driving the relationship. In contrast, a mixture of landscape- (17%) and local-scale (25%) factors explained variation in assemblage characteristics in Idaho. The spatial distribution of study reaches explained 19% of variation in fish assemblages in Idaho, but only 1% in Ohio. Likewise, spatially-structured environmental factors (e.g., the joint influence of environmental and spatial matrices) explained more variation in fish assemblages in Idaho (39%) than in Ohio (29%). </p> <p> Given the relative importance of broad-scale watershed processes in shaping stream ecosystems, I subsequently investigated the utility of a watershed-modeling approach for predicting fish-based indices of biotic integrity (IBI). I used the hydrologic model, Soil and Water Assessment Tool (SWAT), to model stream discharge, sediment flux and concentration, and yields of nitrogen and phosphorous from sixteen watersheds in Ohio. I used a suite of common field-based habitat and geomorphic assessment and measurements to characterize physical conditions at each study reach. Using principal component analysis (PCA) and multiple linear regression, I explored potential associations between environmental factors (SWAT-generated variables and field-collected data) and fish assemblages (IBI scores and fish assemblage descriptors, such as species richness, number of darter species, and percent intolerant species in an assemblage). Whereas field-collected data were somewhat important in explaining patterns in fish assemblages (12%, 14%, and 14% of variation explained for number of darter species, number of sensitive/intolerant species and species richness, respectively), my results suggested that SWAT-modeled, watershed-scale processes were, in most cases, more strongly associated with the same fish assemblage descriptors and also overall IBI scores. SWAT-generated streamflow variables explained > 60% of the variation in our significant environmental-fish assemblage models (e.g., species richness) and SWAT-generated sediment and nutrient variables were the most influential factors explaining variation in IBI. Overall, I found watershed-modeling techniques to have utility for evaluating assemblages at a coarse level and these techniques may prove to be an important complement to current reach-based field assessments. </p> <p>Results from my work suggest that (1) the relative importance of landscape- and local-level environmental factors, and the contribution of spatial structure, may be distinct in regions characterized by different levels of human impact, and (2) approaches that incorporate and evaluate watershed-scale processes may be particularly valuable not only for understanding complex environment-fish relationships, but also for monitoring and conservation applications related to stream fishes.</p> 2011-07-28 English text The Ohio State University / OhioLINK http://rave.ohiolink.edu/etdc/view?acc_num=osu1306782461 http://rave.ohiolink.edu/etdc/view?acc_num=osu1306782461 unrestricted This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.
|
