Goal
The goal of this lab was to use the geographic inquiry process, along with data and geoprocessing tools for vector analysis, in ArcGIS, to determine specific areas that would be suitable for bear habitat study areas in Marquette County, Michigan.
Background
In combination with data of the State, County, and areas of DNR management Lands, we were given the task of determining specific areas in Marquette County, Michigan, that would be suitable for bear habitat study areas. The potential study area had been predetermined, with data of bear locations inside of the study area marked by X,Y coordinates. Inside the study area we also had data on streams and land cover. The land cover feature class was divided into major and minor types. The major land type had data on the demographics of the area, more specifically what land was by type; forest, barren land, agricultural, or urban, while the minor land type data determined the specific use or vegetation types of these areas.
Methods
The first step in creating a map, with the specific areas of interest that could be used in the bear habitat study, was to use spatial vector tools to combine the bear location coordinates with the land cover feature class, in an attempt to determine the type of land cover that the bears preferred or were found most in.
Once this area was determined, information from Biologists indicated that the bears of interest might be found near streams. More spatial vector tools (buffer, dissolve and intersect) were used to find the percentage of bears that were within areas of 500 meter of streams when their location was collected. 49 of the 68 total bears that had their locations marked were found to be within a 500 meter distance of the streams that were confined in the study area.
The next task was to make a recommendation to the Michigan DNR for a bear management plan, based on suitable bear habitat located on DNR management lands. Analysis was preformed by overlaying the stream proximity buffer and the 3 most common land cover types which the bears were located in, but only within the study area in Marquette County. Once that area was determined, another intersect was preformed combining the most likely bear locations (near streams and by land cover) with areas of land that fall within the DNR management zones.
The final task was to take all of the information and data that we built up and exclude specific areas from our study. In particular the areas that the DNR did not want to be part of the bear habitat study areas, were areas that were near built up or urban areas. This task was done two ways, first using spatial vector tools, and secondly, using Python. With spatial vector tools, selecting land cover by built up or urban allowed us to create a new feature class based on those attributes. Creating a 3 kilometer buffer around built up or urban areas was the minimum distance which the DNR desired. Then simply erasing the potential bear habitat that fell within that buffer allowed us to exclude that area from our final analysis. In Python, code was used to accomplish the same task (see below).
Data Flow Model and Analysis:
*OB indicates what specific objectives were met at that stage of the model.
Python code:
Results:
Here is final map with analysis of the data, and a proposed bear habitat study area for the Michigan DNR. In the map on the left, we can see bear locations specifically near streams, and in most common land cover areas bears were found in, inside of the study area.
The map on the right is the final recommendation for the Michigan DNR, which includes most likely bear locations, in addition to being overlaid with specific criteria from the Michigan DNR, which included that the most likely locations for bear habitat study areas be on Michigan DNR land under their management, and being at least 3 kilometers away from urban areas.
Sources:
Michigan Center for Geographic Information
USGS NLCD
DNR Management Units
Stream Locations
