Geomorphic responses of Duluth-area streams to the June 2012 flood, Minnesota

In 2013, the U.S. Geological Survey, in cooperation with the Minnesota Pollution Control Agency, completed a geomorphic assessment of 51 Duluth-area stream sites in 20 basins to describe and document the stream geomorphic changes associated with the June 2012 flood. Heavy rainfall caused flood peaks with annual exceedance probabilities of less than 0.002 (flood recurrence interval of greater than 500 years) on large and small streams in and surrounding the Duluth area. A geomorphic segment-scale classification previously developed in 2003–4 by the U.S. Geological Survey for Duluth-area streams was used as a framework to characterize the observed flood-related responses along a longitudinal continuum from headwaters to rivermouths at Lake Superior related to drainage network position, slope, geologic setting, and valley type. Field assessments in 2013 followed and expanded on techniques used in 2003–4 at intensive and rapid sites. A third level of assessment was added in 2013 to increase the amount of quantitative data at a subset of 2003–4 rapid sites. Characteristics of channel morphology, channel bed substrate, exposed bars and soft sediment deposition, large wood, pools, and bank erosion were measured; and repeat photographs were taken. Additional measurements in 2013 included identification of Rosgen Level II stream types. The comparative analyses of field data collected in 2003–4 and again in 2013 indicated notable geomorphic changes, some of them expected and others not. As expected, in headwaters with gently sloping wetland segments, geomorphic changes were negligible (little measured or observed change). Downstream, middle main stems generally had bank and bluff erosion and bar formation as expected. Steep bedrock sites along middle and lower main stems had localized bank and bluff erosion in short sections with intermittent bedrock. Lower main stem and alluvial sites had bank erosion, widening, gravel bar deposition, and aggradation. Bar formation and accumulation of gravel was more widespread than expected among all main stems, especially for sites upstream and downstream from channel constrictions from road crossings, or even steep sites with localized, more gently sloping sections. Decreases in large wood and pools also were observed throughout the longitudinal continuum of main-stem sites, with immediate implications for fish and benthic invertebrate aquatic habitat. Whether or not the geomorphic conditions will return to their preflood condition depends on the location along the longitudinal continuum. The amount of large wood and pools may return after more moderate floods, whereas bars with coarse material may remain in place, locally altering flow direction and causing continued bank erosion. Results from this study can be used by local managers in postflood reconstruction efforts and provide baseline information for continued monitoring of geomorphic responses to the June 2012 flood.