Assessment of lateral erosion in three agriculture- dominated Minnesota streams: measurement tools, and factors affecting erosion rates

Abstract

Statewide, 14 percent of Minnesota’s impaired waters are listed for excessive turbidity. In-channel and near-channel erosion are commonly considered major contributors to Midwestern turbidity problems. This research sought to determine the primary drivers of channel erosion in the Elm Creek, Buffalo River and Whitewater River watersheds, with a goal of informing guidance and policy on in-channel and near-channel erosion control practices. Of special interest was whether a woody- or herbaceous- dominated riparian corridor was more stable. First, the definitions, history, and some stream erosion variables are explained. The variables were limited to a brief review of the effects of soil, chemistry, vegetation, hydrology and stream size characteristics. The three study watersheds were introduced. Second, two GIS-based lateral erosion tools (DNR Static Lateral Migration Tool and BBE Dynamic Lateral Migration Tool) and a common field-based methods (BANCS) were compared. The dataset allowed comparison of results from three tools and multiple users on three streams. The DNR Static Lateral Migration Tool was applied to three streams by one user, and to the Whitewater River by a second. The BBE Dynamic Lateral Migration Tool was applied to the Buffalo River by a third user. The BANCS tool was applied to three streams by a group of users, and to the Whitewater River by another group. The reach breaks for the DNR Static Lateral Migration Tool were chosen to allow comparison of erosion rates to reach-specific variables. The reach breaks for the BBE Dynamic Lateral Migration Tool were at set distances. Generally, the erosion rates across all tools, user groups and streams were between 0 and 0.6 meters (0 and 2 feet) per year, though some results were higher. All GIS-based tools and users returned erosion rates near or under 1.2 meters (4 feet) per year, with maximum BANCS results near 1.8 meters (6 feet) per year or more. The erosion data allowed for customization of a stream bank erosion prediction graph for comparable Minnesota streams. Third, using GIS, the lateral erosion rates of nearly 240 reaches of the three streams were compared to other stream characteristics. These characteristics included vegetation type, eroded area, reach length, valley length, sinuosity, water surface slope, low bank slope, high bank slope, water surface elevation, low bank elevation, high bank elevation, low bank height, high bank height, bankfull width, radius of curvature, near bank stress, stream mile (size), curve count, curve length, wetland presence, geomorphology, soils, and erosion to bankfull ratio. Of the measurements available to a GIS-user, near bank stress, and stream size were most correlated to erosion rate in these systems. Finally, a few of the values associated with Minnesota’s water economy are linked to the costs of preventative policy, and reactive restorations. Due to the local need for an accurate picture of erosion drivers and erosion rates, and for an efficient restoration prioritization tool, the University of Minnesota partnered with the Minnesota Department of Agriculture. The work can inform policy and restoration efforts.