Thesis Director: Barnali Dixon, Ph.D. College of Arts and Sciences
University of South Florida St. Petersburg
Habitat suitability modeling can reveal connections between seagrass and environmental variables namely water quality. This can help predict seagrass distribution affected by changing water quality variables. Seagrass, and its density distribution in Tampa Bay, Florida, is an important environmental and economic resource providing a habitat for fish and invertebrates, a food source for larger vertebrates, and a significant source of primary productivity. The overall goal of this study was to loosely couple GIS and logistic regression methods to analyze relationships between seagrass distribution and water quality variables. Specific objectives were i) to determine key water quality variables that influenced seagrass occurrence and ii) to analyze prediction error including difference between continuous and patchy seagrass beds. Water quality variables, such as light attenuation, salinity, and temporal variability of salinity (TVS), were used with GIS and the logistic regression model in order to predict their relationship with seagrass occurrence. Preliminary results showed that light attenuation was a significant predictor of distribution, with salinity and TVS to a lesser extent. The predictive model was validated using known seagrass polygon data. Results indicated that the model predicted continuous coverage more accurately (55%) than patchy coverage (33%). The difference of nearly 20% could be attributed to complex shape dynamics in patch delineation. Future error analysis should include incorporation of fractal dimensions of seagrass polygons for both continuous and patchy beds. Results of this study can give managers and planners information on the relationship between seagrass occurrence and the water quality variables investigated in this study.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Cope, Samantha, "Integration of GIS and logistic regression to develop a habitat suitability model for predicting seagrass distribution" (2015). USFSP Honors Program Theses (Undergraduate). 210.