TERRAIN ANALYSIS BASED ON LIDAR-DERIVED DIGITAL ELEVATION MODELS TO IDENTIFY CRITICAL SOURCE REGIONS OF AGRICULTURALLY-DERIVED SEDIMENTS AND POLLUTANTS

The availability of a sub-meter resolution digital elevation model (DEM) for southeastern Minnesota provides the opportunity to use remote sensing and GIS technologies to identify portions of the landscape that are vulnerable to sediment erosion and contaminated surface runoff. Using an appropriate resolution DEM a number of topographic and geomorphic attributes of the landscape can be determined in GIS, including: slope, aspect, flow direction, flow accumulation, plan curvature, and profile curvature. Based on these primary attributes, secondary attributes such as the Topographic Wetness Index (TWI) and Stream Power Index (SPI) can be derived. The TWI is useful in identifying the spatial extent and distribution of zones of saturation and runoff generation. The SPI predicts the erosive power available across the landscape and is useful in identifying potential erosion and source areas of sediment.

This research focused on four small, agriculture and forest-dominated watersheds in southeastern Minnesota. There were two broad objectives: First, we wanted to develop a methodology for performing and validating terrain analyses within the context of our local land uses, geology, climate, and karst hydrology. Specifically, we wanted to determine the range of thresholds at which secondary terrain attributes such as SPI and TWI become reliable predictors of erosion and overland flow in our agricultural and forested watersheds. Secondly, using the results of the first objective, we wanted to map areas of overland flow generation and susceptibility to erosion to aid in focusing and identifying the best areas for the implementation of agricultural best management practices (BMPs) to mitigate suspended sediment and water quality problems in streams in the four project watersheds.

Preliminarily, we have established a protocol for validating predictive thresholds for the SPI and TWI based on a combination of statistical analysis, remote sensing, and field-based verification. We have also developed guidelines for our local landscapes for the appropriate resolution DEM to use based on the spatial scale of the study area analyzed. Moving forward, we will be using the SPI and TWI maps that we’ve generated to guide decision making related to agricultural BMP implementation in the next phase of our project.