Modeling the Effects of Low Impact Development Practices on Streams at the Watershed Scale

Abstract

Urban growth contributes to increasing storm water runoff which in turn causes an increase in the frequency and severity of flooding. Moreover, increased storm water runoff contributes to changing the character and volume of energy inputs to the stream. Traditionally, storm water management controls such as detention pond had been extensively studied and evaluated with respect to reducing and controlling peak flows. Nonpoint source pollutants due to urbanization and expanding of agricultural fields have become a big burden on municipalities and states.

Low Impact Development practices were developed to negate the negative impacts of urbanization on water resources by reducing the runoff volume and peak flows as well as improving outflow water quality. Though these practices have the capability of reducing runoff volumes and enhancing outflow water quality, they can be costly. Therefore, understanding the impact of installing LID practices on a watershed scale is becoming increasingly important.

In this study, field experiment and model study were applied to evaluate the effectiveness of LID practices on a watershed scale in the Blunn Creek Watershed located in Austin, Texas. The three LID practices which were evaluated in this study are permeable pavements, a bioretention area, and a detention pond. The main objective of this study was to investigate the influences of these practices at a watershed scale on: potential reduction on channel bank erosion, potential reduction on flood, and potential impact on aquatic life.

This study was one of very few studies that take place in the Blackland clay soil in Texas. A combination of different levels of LID practices such as permeable pavement and bioretention area resulted with achieving the main goal of this study of reducing stream bank erosion, bankfull exceedance, and maintaining acceptable flows for the integrity of aquatic life habitat. All LID practices have shown significant difference with respect to a control treatment at 95% confidence ratio. Performance of the modeled LID practices was validated by showing acceptable agreement in the percentage of reductions in total runoff between field experiments and model data.