EFFECTS OF FLOW REGIME ON FISHES AND FISHERIES: FROM LIFE HISTORIES TO ECOSYSTEM SERVICES

Author

Micah Bennett, Southern Illinois University CarbondaleFollow

Date of Award

12-1-2015

Degree Name

Doctor of Philosophy

Department

Zoology

First Advisor

Whiles, Matt

Second Advisor

Whitledge, Gregory

Abstract

Stream flow in freshwaters is considered a “master variable” influencing processes and traits from individual organisms to ecosystems. Due to this strong linkage, anthropogenic modification of flow regimes in freshwater ecosystems worldwide continues to have major impacts on populations, species, communities, and ecosystems and the many services they provide to humans. My dissertation investigated the impacts of flow regime and its variability on three levels of biological organization: populations, communities and ecosystems. The approach highlights links among evolutionary, community, and ecosystem ecology, while also testing basic models and demonstrating applied significance in freshwater conservation. At the population level, I evaluated the generality of the trilateral life history model (TLHM) for fishes - a trait-environment model well-studied at the assemblage level – finding that the TLHM adequately described major trade-offs in traits among populations in all species. Some TLHM flow-based predictions were confirmed, with periodic traits (high fecundity) favored at sites with greater flow seasonality and lower flow variability in two species, and equilibrium traits (large eggs) in more stable flow conditions in two species. However, relationships contradicting the TLHM were also found. In Chapter 3, I evaluated the effects of geographic location, scale, and sampling gear on agree with TLHM predictions using a fish community dataset from Louisiana. Generally, fewer than half of significant relationships supported TLHM predictions. These results suggest that, due to collinearity of hydrologic variables, effects of sampling gear, and scale of analysis, applying and operationalizing the predictions of the TLHM in terms of hydrology may not be straightforward. here is a continued need to match high-quality biological data with hydrologic data while also developing hydrologic modeling and datasets of correlated environmental variables at finer scales to match the grain of most biological sampling. Trait-environment models that are well-supported at multiple levels of biological organization could improve understanding of the impacts of environmental change on populations and communities and the valuable ecosystem services that they support. Chapters 4 and 5 focus on ecosystem services and how they are related to each other and influenced by flow regime in a large river-floodplain ecosystem – the Atchafalaya River in Louisiana. I first developed a model of denitrification in the Atchafalaya River. Denitrification rates ranged from 5,394 kg N y-1 (3.07 kg N km-2 y-1) in 1988 to 17,420 kg N y-1 (9.92 kg N km-2 y-1) in 1981, and rates were consistently higher in fall compared to spring. Total nitrate (NO3−) denitrified in the basin was negligible compared to total NO3− entering the GOM. If all N denitrified in the basin instead entered the Gulf, the hypoxic zone was predicted to increase only 5.07 km2 (0.06%). This negligible effect on N dynamics in the GOM agrees with other mass balance and isotopic studies in the region. Denitrification is only one of many ecosystem services provided by river-floodplain ecosystems. Because of the overriding influence of flow regime on river systems, an understanding of flow-ecology relationships in rivers is necessary to assess potential impacts of management decisions. However, translating complex flow-ecology relationships into stakeholder-relevant information remains a struggle. The concept of ecosystem services provides a bridge between flow-ecology relationships and stakeholder-relevant data. Flow-ecology relationships were used to explore complementary and trade-off relationships among 12 ecosystem services and related variables in the Atchafalaya River Basin, Louisiana. Results from Indicators of Hydrologic Alteration were reduced to four management-relevant hydrologic variables using principal components analysis. Multiple linear regression was used to determine flow-ecology relationships and Pearson correlation coefficients, along with regression results, were used to determine complementary and trade-off relationships among ecosystem services and related variables that were induced by flow. Seven ecosystem service variables had significant flow-ecology relationships for at least one hydrologic metric. There was a single complementary relationship mediated by flow and there were three such trade-off relationships; however, other trade-off and complementary relationships were not related to flow. These results give insight into potential conflicts among stakeholders, can reduce the dimensions of management decisions, and provide initial hypotheses for experimental flow modifications.