GIS I - Lab 5: Application of Skills

Figure 1: This map shows the best suitable hunting land, located on publicly managed DNR land,
in Sauk County, Wisconsin. The green polygons denote land that matches the specified criteria. 

Goal

Background: The goal of this assignment was to develop a personal spatial question and carry out the correct means to answer that question. The spatial question that this lab seeks to answer is: Where are the best places in Sauk County, Wisconsin to hunt on land open to the public?

Purpose: The purpose of this assignment was to demonstrate knowledge of the skills learned from this course and apply them through various methods.

Methods

Data Collection: To begin the process of determining the best possible public hunting land in Sauk County, data had to be brought in from for streams, bodies of water, Wisconsin Department of Natural Resources (DNR) managed lands, county borders, state border, roads, and cities. Through ArcMap, database connections were made to the Wisconsin DNR Database and the ESRI Database. Through the DNR database the following data was acquired: Streams, DNR Managed Lands, County Borders, State Border. Through the ESRI database the following data was acquired: Bodies of Water, Roads, and Urban Areas. This data was brought into a file geodatabase created for this assignment. Concerns with this data stem from having to use a variety of different feature classes to examine one overall feature, such as the classes for streams and waterbodies.

Data Preparation:
Sauk County: In order to start utilizing the data, Sauk County had to be located and made into its own feature class. This allowed the analyst to use Sauk County to clip the other features to get details within only the target area. To do this, an attribute query was performed to locate Sauk County. Once the county was located, the option to make it into a feature class was chosen. Then, Sauk was projected to the NAD 1983 Wisconsin CRS Sauk (meters) projection to make the feature class more conducive to viewing solely the county.

Roads: To begin using the Roads data it was clipped using the Sauk County polygon. This data was then projected to the NAD 1983 Wisconsin CRS Sauk (meters) projection. This data had many different classes within it, depending on the size of the road. To differentiate between the different classes of roads, they were each given a different symbol, as seen in Figure 1. This class difference led to the need to create buffers based on different criteria to create different sized buffers. [Note: This was not a skill learned in this class, however, I was able to teach myself how to do it and used it for this feature class and for the Streams feature class]. To do this, a new field labelled "Buffer_Field" was added to the attribute table in the form of a Text field. Different values were assigned to this field, for example "0.5 Kilometers". This was added to each different class in order to create different sized buffers based on the size of the road and the perceived distance that a hunter would like to be away from the road in order to reduce the chance of traffic interfering with their hunt. The roads were then buffered by Field, rather than buffered by Distance. A dissolve was also applied to get rid of overlapping lines created by buffering. The resulting feature class was then clipped using Sauk County to get rid of buffers falling outside of the target area. Labels were activated for this data and a mask was applied to the labels to make them more visible on the map.

Urban Areas: To begin using the Urban Areas data it was clipped using the Sauk County polygon. This data was then projected to the NAD 1983 Wisconsin CRS Sauk (meters) projection. A dissolve was applied to remove boundaries between cities like Lake Delton and Wisconsin Dells, and Sauk City and Prairie du Sac, who share borders. Then a buffer was applied around the city to account for anthropogenic influence. The data was clipped again using the Sauk County polygon to remove buffers that extended outside of the county. Labels were activated for this data and a mask was applied to the labels to make them more visible on the map.

Bodies of Water: To begin using the Waterbodies data it was clipped using the Sauk County polygon. This data was then projected to the NAD 1983 Wisconsin CRS Sauk (meters) projection. A buffer was applied to buffer the area around the various bodies of water. This buffer assumed that wildlife would travel a certain distance to drink from a lake, pond, or river. Once the buffer was created and a dissolve was applied, it was clipped with Sauk County again to remove buffers that extended outside of the county. The data was then compared to the buffered urban class and an erase was applied, then to the buffered roads class and an erase was applied. This left waterbodies that were located a certain distance away from anthropogenic influence.

Streams: To begin using the Streams data it was clipped using the Sauk County polygon. This data was then projected to the NAD 1983 Wisconsin CRS Sauk (meters) projection. The data was then compared to the Waterbodies feature class and an "Erase" was applied to the data in order to remove data that was overlayed with lakes or rivers, as they were not properly displayed with a line feature class. This data had many different classes within it, depending on the size of the stream. This led to the need to create buffers based on different criteria to create different sized buffers. To do this, a new field labelled "Buffer_Field" was added to the attribute table in the form of a Text field. Different values were assigned to this field, for example "0.4 Kilometers". This was added to each different class in order to create different sized buffers based on the size of the stream and the perceived distance that animals would travel in order to drink from it. Smaller streams received smaller buffers because they would not supply as much water to the surrounding wildlife, or potentially because they only carried water during precipitation events (intermittent streams). The streams were then buffered by Field, rather than buffered by Distance. Once the buffer was created and a dissolve was applied, it was clipped with Sauk County again to remove buffers that extended outside of the county. The data was then compared to the buffered urban class and an erase was applied, then to the buffered roads class and an erase was applied. This left streams that were located a certain distance away from anthropogenic influence. Finally, a union was applied between the Waterbodies feature class and the Streams feature class in order to derive a new feature class labelled "Sauk Water". This feature class contained all units of water in Sauk County.

DNR Managed Lands: To begin using the DNR Managed Lands data it was clipped using the Sauk County polygon. This data was then projected to the NAD 1983 Wisconsin CRS Sauk (meters) projection. An attribute query was conducted for lands that were fully owned by the Wisconsin DNR or lands that were acquired through easements that allowed for public hunting. The interpretation of this data was assisted by Ann Runyard, GIS Analyst for the Wisconsin DNR. The query left only land that was available for public use of hunting. The data was then compared to the buffered urban class and an erase was applied, then to the buffered roads class and an erase was applied.

Data Utilization:
After all the data was prepared and the necessary feature classes were discerned, the data was finally able to be narrowed down to show only suitable public hunting land that fell within the proximity of a waterbody. To do this, an intersect was applied to show only data that met both of the necessary criteria of falling within the Sauk Water buffer and DNR huntable land. This left the answer to the initial query of the most suitable places to hunt publicly in Sauk County.

Cartographic Preparation: A proper map was created with a map to show location of the county within Wisconsin. A data flow model was also created to show workflow for the project.

Figure 2: Data flow model for the assignment. This shows the work flow carried out in order to achieve the desired output data.

Discussion
While I felt that most of this data provided a solid view of the potential lands to hunt in Sauk County, I also feel that more data would have better suited this query. It should be noted that the main focus of this project had deer in mind when the analysis was conducted. I feel that data regarding deer density would have helped show the most suitable land with the most potential for an encounter. There was data regarding deer management zones, but as they are not bound by county boundaries the data regarding density would have been skewed. I also feel that I ran into some issues with the hydrology aspect of this project. There are many different feature classes that deal with streams, rivers, ponds, and lakes, but I could not find one that contained a majority of those, so I had to settle for two feature classes and perform an erase followed by a union. With the streams feature class, lines were shown where lakes were present between an input stream to the lake and a drainage stream. The line between the two is what was erased, leaving only the input stream until the boundary of the lake and the drainage stream from the boundary of the lake. If I had the ability to input real world data gathering measures into my data I would attempt to get an overall idea on how many deer are registered per deer registration station. This would give me a general idea about the number of deer in the area, instead of a large scale classification based on spatially large zones.

Results
This project showed my aptitude to solve a question using applied GIS. It was a comprehensive project showing all that I had learned and gathered from previous coursework. As this is a potential field that I would like to go in to, understanding how to apply various tools and what those tools do is a crucial component to furthering my career.

Sources
Ann Runyard - GIS Analyst, Wisconsin Department of Natural Resources

ESRI Geodatabase

Wisconsin DNR Geodatabase

GIS I - Lab 4: Vector Analysis with ArcGIS

The map to the left shows a comparison of suitable habitat for bear and the ideal suitable habitat for bear in Marquette County, Michigan. The ideal habitat falls within pre-existing Michigan Department of Natural Resources land management zones. A small locator map in above the map on the left indicates the location of the study area, within Marquette County, in the state of Michigan. The map on the right shows the land use for areas within the study area. 

Goal

Background: The goal of this assignment was to create a map based on suitable habitat for bear in Marquette County, Michigan.

Purpose: The purpose of the assignment was to gain a better understanding of how to utilize and analyze data in order to properly show how the data relates.

Methods

Step One: In order to begin gathering the necessary data for this lab, a file geodatabase was created for storing feature class data. A Microsoft Excel table, containing XY coordinates for bears being monitored through radio tracking, was converted to be used in ArcGIS. This was done by adding the coordinates as an "event theme" in ArcMap, selecting "File" -> "Add Data" -> "Add XY Data" inputting the correct data, and exporting the final product to the file geodatabase. These points were then able to be saved as a feature class that provided the XY coordinates of all bears at the time of data collection.

Step Two: The next task was to prepare the data for analysis by running various tools to narrow down the search area for ideal bear habitat. The XY coordinates were spatially joined with a feature class showing landcover, in order to create a new feature class showing what landcover types bears were most likely to be found in. The results were then summarized to get a better, quantitative understanding of where bears were found.

Information was provided that stated that most bears seemed to be located near streams in order to have easy access to food and water. The stream feature class was selected and the boundaries surrounding the various stream segments were dissolved in order to create a single stream line. A buffer was created to encompass an area of 500 meters around the stream. Out of the sixty-eight bear locations recorded, forty-nine of the recorded locations were within the buffer created by the stream.

Both the landcover-bear spatial join and the stream buffer were intersected, showing the bear locations and landcover types within the boundaries of the stream buffer.

Step Three: With the newly acquired stream buffer showing bear locations and landcover, the next task was to use the acquired parameters to assess the best suitable habitat for bear management zones. This was achieved by first performing an attribute query to select the three most prominent landcover types that were inhabited by bears, determined by referencing the summarized table from the prior step. The newly clipped feature class was then dissolved so there were now unnecessary boundaries within the feature class. This selection was made into a feature class and the data was clipped from the stream buffer, to show the top three suitable habitats within 500 meters of a stream, where bears were present.

Step Four: After acquiring the data for the best suitable bear habitat within 500 meters of a stream, the next task was to compare this to current Michigan Department of Natural Resources management land, to determine the easiest place to expand upon a bear management program. The suitable habitat-stream buffer was intersected with the Michigan DNR lands in order to show all DNR land that fell within suitable lands within 500 meters of a stream. The resulting feature class was dissolved so there were no unnecessary boundaries.

Step Five: The final task was buffer urban areas so that potential bear management areas would not be found too close to urbanized areas. An attribute query was performed to isolate areas noted as "Urban" or "Built Up Land" and this was made into a feature class. Following this, the feature class was dissolved and a 5 kilometer buffer was applied to the feature to exclude areas within 5 kilometers of any urban areas. The buffer was then compared to the suitable lands within DNR lands within 500 meters of a stream. An erase tool was used in order to erase any data that fell within the urban buffer. This excluded all unwanted information and all that was left were the ideal areas for bear habitats, as noted in the left map, above.

Step Six: A proper map was created and a data flow model was also created to show workflow for the project.

Results

This assignment provided an opportunity to apply GIS to real world situations and queries. This was a particularly interesting assignment, as biogeography is a field that I am very interested in. Querying suitable bear habitats is a very applicable issue to work with. Many of the methods learned from this lab will be used to discern suitable hunting areas in the final project for this course.

Sources:
Michigan Geographic Data Library