Urban Patterns and Flood Damage in Texas Coastal Watersheds

Abstract

This study develops a framework to conceptualize and measure multiple urban patterns and examines their relationship with flood damage in Texas coastal watersheds. Development and flood damage impacts are analyzed over a ten year period in 916 watersheds that overlap Texas’ 41 coastal watershed counties using the USGS National Hydrography Dataset. A cross sectional time series regression model is used to determine how changes in these patterns influence the amount of flood damage that occurs in the study area.

Results from the study provide clarity on how different dimensions of urbanization are related to flood damage. Using six landscape metric measurements for three different levels of urban land cover and two measures of residential property location in relation to the rest of the watershed, regression analyses conclude that most urban pattern metrics are significant in influencing the degree of flood damage at a watershed scale. Specifically, increases in percentage of impervious surface increases flood damage, as do most other metrics as they pertain to expansiveness of impervious surface across the landscape. Two metrics (Mean Shape and Average Distance of Residential Property to Water) did not behave as hypothesized; it is believed that mean patch shape was incorrectly hypothesized, and the metric representing average distance to water was measured inappropriately.

The results of the models and the significance and direction of the independent and control variables all provide evidence of the need to take urban form and environmental factors into consideration and an ecosystem-based approach should be taken when engaging in policy and planning activities to reduce residential property damage from flood events.