LAND SNAIL ABUNDANCE AND DIVERSITY WITH ASSOCIATED ECOLOGICAL VARIABLES IN SIX SOUTHERN ILLINOIS COUNTIES

Author

Marla Lee Coppolino, Southern Illinois University CarbondaleFollow

Date of Award

1-1-2009

Degree Name

Master of Science

Department

Zoology

First Advisor

Anderson, Frank

Abstract

Various ecological parameters, including soil pH, calcium, and habitat complexity have been suggested in the literature as having influence on land snail abundance and diversity. I compared relationships between 15 ecological parameters and snail abundance and diversity. 5,393 snails of 72 species were collected from 60 sites in 2007. Habitat observations were recorded in the field and soil core samples taken at each site and later analyzed for soil moisture, organic matter, cation exchange capacity, pH, calcium, magnesium, zinc, potassium, phosphorus, sulfur, boron, manganese, iron and copper. Statistical analyses were run using these soil factors plus a habitat complexity index (from combined values assigned to different levels of vegetation, topography and exposed rock), for a total of 15 parameters. To reduce the number of parameters and arrive at a more biologically meaningful model, Bayesian Information Criteria analysis (BIC) was run for abundance and diversity. The resulting best-fitted BIC model for abundance contained 3 parameters (pH, S, and habitat complexity) (R2 = 0.47), all of which were positively associated with abundance in a multiple regression analysis. For diversity, the best-fitted BIC model also contained 3 parameters (Ca, Fe, and habitat complexity) (R2 = 0.54). Ca and habitat complexity showed a positive association in the multiple regression analysis, but Fe was inversely associated with diversity, suggesting Fe could potentially act as an ecological limiting factor to it. These results suggest that land snail abundance and diversity are best treated separately in analyses, as they are influenced by different variables, with the exception of habitat complexity, which is suggested to have a positive association with both abundance and diversity. The regression model equations have potential value in that they can be used to predict snail abundance and diversity in areas that have not been assessed.