Author

Tracy Whelen

Date of Award

2020

Document Type

Open Access Thesis

Department

Geography

First Advisor

Susan L. Cutter

Abstract

Risk assessments enable fire departments to be better prepared for future incidents and to engage in more effective prevention activities. A combination of physical, demographic, and behavioral risk factors combined form a community’s level of risk. This research shows how spatial and nonspatial statistical methods can be used within a GIS framework to create such a risk assessment, with the Columbia-Richland Fire Department in Richland County, SC being used as a case study. Hot spot analysis and thematic mapping of incident rates were used to assess the first research question – what is the spatial variability of structure fires, carbon monoxide incidents, and emergency medical calls? Correlation analysis, principal component analysis (PCA), and factor analysis were applied to a few dozen social and physical risk factors at the block group level to assess the second research question - how are the risk factors correlated with each other, and how are these risk factors varied across the county? The results of all types of methods were compared against each other to assess how risk factors correlated with incident types. These methods were able to map hot and cold spots of incidents, identify the most relevant risk factors, and show which risk factors were most prevalent in hot spot areas. The primary hot spot for EMS and fire incidents was found in northern Columbia, with a secondary hot spot located in far Lower Richland. PCA identified nine primary factors, the top three of which were related to systematic hard times, older homeowners, and rural location. Factor analysis was able to cluster block groups into fourteen groupings of similar risk traits. There were very clear differences in incident rates between the fourteen groupings, although hot spots contained block groups from multiple groupings. Given the snapshot in time nature of risk assessments, this research builds a baseline for future risk assessments, both in terms of methods and results.

Rights

© 2020, Tracy Whelen

Included in

Geography Commons

Share

COinS