Distributional Modeling Of Rare Fishes And Mussels: From Ecological Theory To Biodiversity Conservation

Author

Rob Hopkins, Southern Illinois University CarbondaleFollow

Date of Award

1-1-2009

Degree Name

Doctor of Philosophy

Department

Zoology

First Advisor

Burr,Brooks

Second Advisor

Oyana,Tonny

Abstract

Freshwater ecosystems are among the most threatened ecosystems on the planet. In recent years there has been increasing concerns over precipitous declines in population sizes and increasing rates of extinction of native freshwater fauna across North America. Nearly 50% of the mussel species and 25% of the fish species in North America are imperiled. Stream habitat degradation has been cited as the principal cause for declines, with anthropogenic land uses being the leading causes of stream degradation. Species distribution models (SDMs) have become an integral tool in ecological research and conservation planning. SDMs are reliant upon occurrence datasets for the taxa of concern and museum-based information has become a popular source for such data. In the first stage of my research, I developed a centralized database for Kentucky fishes based on museum-based information. Over the course of three years, the SIUC Ichthyology Lab built an occurrence dataset of Kentucky fishes consisting of more than 50,000 records dating back to the 1890s. Each record contains three pieces of information (1) species identification, (2) georeferenced locality, and (3) date of collection. In the second stage of my research, I investigated the use of multiscale landscape data in aquatic species distribution models using a case study of a freshwater mussel. The distribution of Rabbitsfoot (Quadrula cylindrica) in the upper Green River system (Ohio River drainage) was modeled with environmental variables from multiple spatial scales. Four types of landscape environment metrics were used, including: land use/land cover (LULC) pattern, LULC composition, soil composition, and geology composition. The study showed that LULC pattern metrics are very useful in modeling the distribution of Rabbitsfoot. Together with LULC compositional metrics, pattern metrics permitted a more detailed analysis of functional linkages between aquatic species distributions and landscape structure. Moreover, the inclusion of multiple spatial scales was necessary to accurately model the hierarchical processes in stream systems. Geomorphic features played an important role in regulating species distributions at intermediate and large scales while LULC variables appeared more influential at proximal scales. I then further tested the landscape-level approach to aquatic species distribution modeling using a case study of six narrow-range endemic fishes with contrasting biogeographies. Species biogeography did not appear to affect predictive performance and all models performed well statistically. Predictive maps showed accurate estimations of ranges for five of six species based on historical collections. The relative influence of each type of environmental feature and spatial scale varied markedly with between species. A hierarchical effect was detected for narrowly distributed species which were highly influenced by soil composition at larger spatial scales and land use/land cover (LULC) patterns at more proximal scales. Conversely, LULC pattern was the most influential feature for widely distributed at all spatial scales. Lastly, I developed a hierarchical approach to the selection and management of freshwater protected areas in the upper Green River system. By aligning the spatial scales and environmental variables analyzed at each stage in the conservation planning process, from species distribution modeling to reserve selection, I present a more robust methodology to conservation planning compared to traditional approaches. Using models of species richness fitted to landscape attributes, I also provided suggestions for landscape management strategies for each conservation unit. My research comprises the core conservation plan for the focal species in the upper Green River system.