In order to get a clearer picture of what the Shakopee Mdewakanton Sioux Community (SMSC) lands looked like prior to European settlement, I used GIS to map pre-settlement vegetation communities and other features like roads, trails, creeks, rivers, homesteads, and farm fields as recorded in the Public Land Survey (PLS) plat maps of 1854 and 1855. My primary data sources were the PLS plat images from the Land Management Information Center (LMIC), and bearing tree and PLS corners and vegetation type shapefiles from the Minnesota Department of Natural Resources.
I downloaded the PLS plats for Eagle Creek and Spring Lake Townships, in Scott County, Minnesota, where the SMSC owns land, and converted them from PDF to JPEG format. I then georeferenced the JPEGs to the Township-Range-Section shapefile for Scott County.
Next I exported the Eagle Creek and Spring Lake township boundaries to a separate shapefile which I could use as a base shapefile to map features on the plat images. I cut into the township polygons instead of digitizing individual polygons to represent the image features. This approach let me use the township polygons that already existed, and also avoided topological problems associated with creating new polygons by digitizing.
I used a scale of 1:10,000 to map the image features because, on the computer screen, it’s close to the size of a section; that meant that all features could be digitized consistently at the same scale. I digitized wetlands, prairies, forests, lakes, oak savanna, the Minnesota River, and farm fields as polygons, based on the plat images. I added homesteads as points, and roads, trails, and creeks as polylines.
The original surveyors who mapped this area clearly identified lakes, wetlands, prairies, and forests but I needed more information to determine the oak savanna vegetation community. I checked the PLS corners and vegetation type shapefile and found a cluster of points classified as "oak openings." Since the bearing tree data corroborated that all the tree species were oak (Quercus), I was fairly confident that this area was an oak savanna.
I finished the project by populating the attributes in the point, polyline, and polygon shapefiles to represent image features.
The project yielded good information. Calculating acreage for the mapped polygons showed roughly 8,000 acres of prairie on the northern end of Eagle Creek Township and about 30,000 acres of what was likely Big Woods forest for the southern half of Eagle Creek and all of Spring Lake townships. There were only 10 homesteads and about 350 acres of farm fields, so the townships were still relatively impact-free in 1854 and 1855.