Northern Prairie Wildlife Research Center
Bear Lake Wetlands Center, 22752 County Road 7, Hutchinson, MN 55350
During 1990-91, the owner of a horse breeding facility filled a wetland with sawdust and horse manure. About 0.4 ha of the basin had been filled with >1.2 m of waste. The St. Paul District of the Corps of Engineers determined that the fill violated the provisions of the Clean Water Act. The Corps required removal of the fill and restoration of the wetland.
The rolling character of the terrain and the depth of the fill made it difficult to tell the location of the original wetland edge. To achieve restoration, the pre-existing boundaries of the wetland needed to be defined. Two alternative techniques were considered. One technique involved the use of time-consuming and expensive soil boring equipment. The second technique, and the one selected, involved the use of Agricultural Stabilization and Conservation Service 35mm color aerial slides that are taken annually in mid-summer. Slides from 1979-91 were examined and two slides were selected. One slide from 1989 represented the approximate maximum extent of the wetland prior to filling. A second slide taken in 1991 showed the area after filling. The two slides were scanned into WetMAPP, a GIS software package, using a still image video camera. The photo-image was used as the basis from which the edge of the preexisting wetland was traced and a base map created.
A reference grid system was then established on the ground. The grid points were marked with barn lime to make them visible from the air. A low altitude aerial slide was taken on 23 October 1992, which showed the altered condition of the wetland and the reference grid system. This slide was also scanned into WetMAPP. The distance between the original edge of wetland and the grid points was calculated using the "Distance" function of the software. These calculated distances were used in the field, along with the grid system, to reestablish the location of the obscured wetland edge.
We measured the calculated distance from each grid point and set a lath to locate the estimated wetland edge. To verify the edge of the wetland, we examined the soil for hydric characteristics using an Oakfield hand soil probe. Generally, the location of the calculated edge was within 3 m of the point where the soils had hydric indicators. Based on nearly 20 years of personal experience with wetland delineation and classification, the results were well within the variability expected when trying to establish a precise wetland boundary in an infinitely variable natural ecosystem.
The site was visited in early December, 1992 during the restoration process. The staked wetland edge generally matched the natural edge revealed during the removal of the fill.
Our experience with this process showed that high altitude 35mm slides used in combination with a video image capture system and GIS software was a cost effective tool in the restoration of a filled wetland.