One of the greatest challenges in landscape ecology has been determining the degree to which landscape level environmental characteristics effect the distribution of freshwater mussels. Freshwater mussels have long been regarded as valuable indicators of lotic system health because they are often the first organisms to exhibit a response to changes in their environment. Understanding distributional patterns of mussels is consequently a valuable conservation and management tool. Here, I evaluated the improvement of predictive modeling software by applying MAXENT to the distributions of two freshwater mussel species, the Texas Pigtoe (Fusconia askewi), a state threatened species, and the Rock Pocketbook (Arcidens confragosus), a species of concern. Existing species distribution models for these two mussel species were evaluated by the addition of sampling data from two unique watersheds, the Sulphur and Cypress River drainages. The Sulphur River was historically modified and has a biogeographically unique geology compared to most other rivers in East Texas. In addition to locating both species of interest, a range extension for the White Heelsplitter (Lasmigona complanata) was produced on the lower Sulphur River. The Cypress River drainage is a set of watersheds that includes the Big Cypress Creek, Little Cypress Creek, and Black Cypress Creek. Compared to other East Texas Rivers the Cypress drainage is a relatively meandering bottomland system that is healthy along much of its length and empties into the Little Cypress Bayou near Louisiana. Each species distribution was modeled with 11 GIS derived environmental layers. I evaluated each environmental layer by the components these layers were built from and drew conclusions about associations of Texas Pigtoe and Rock Pocketbook with the most suitable habitat types in each important environmental layer. Improvement may depend on a complete understanding of a species fundamental niche and AUC values may not be an appropriate measure of model improvement. Freshwater mussel species will also associate significantly with specific components of each most highly contributing environmental layer.
Date of publication
Heffentrager, Kirian Bregan, "Utilizing MAXENT to Improve and Explain a Species Distribution Model for Freshwater Mussel Species in East Texas" (2013). Biology Theses. Paper 10.