Lab 4 Mini - Final Project

Introduction

For Lab 4 we were to come up with a simple question and answer it effectively using spatial tools.

My question was this, "Where can the Wisconsin DNR expand its forest fire management area near cities?"

This question entered my brain as I looked over data the DNR has available online. I noticed there was a fire amount of wildfires that occurred outside of the coverage area many of which were near cities. I decided to base my project around fire data in a 10 year period from 1998-2008.

Data Sources

To solve the question I posed I needed a variety of data including fire data, city locations, DNR protection area, county forest, national forest and WI state outline. This data could be found provided by the Wisconsin DNR and ESRI. The data can be found at this web address ftp://dnrftp01.wi.gov/geodata. Even though the data came from the DNR I had a few concerns mostly related to age of data. I was hoping for more up-to-date fire occurrence as the most recent data is from 2008 so I'm missing seven years worth of data. I also am concerned with the DNR protection area data as I could not find the age of the data, their protection area could be very different today if that data is 10 years old.

Methods

To solve my proposed question I used a variety of tools in ArcGIS including, unions, erase, buffer, intersect and clip. I first started by using the union tool on both county and national forest class to give me one Forest Land class. After, I used union again with Forest Land and Intensive and Extensive Protected Fire Area to give me my total protected area class (Figure 1). I made a feature class containing only fire occurrences from 1998-2008 (Figure 2) and used the erase tool on that class along with my protected area class to give me the locations of fire outside of the protection area. I buffered this class and found out which fires had occurred within 10 miles of a Wisconsin city. I intersected my cities with fires and was left with my areas in need of additional protection outside of the DNR's coverage area. For aesthetics I clipped this class so that the buffers stayed within the States outline. The entire process can be found in the model below (Figure 3).

Figure 1

Figure 2

Figure 3

Results

 

The results from the above steps can be seen below. The map (Figure 4) includes the proposed protection areas outside of the current DNR coverage in relation to the cities of Wisconsin. The proposed areas are within 10 miles of a Wisconsin city and were areas that had experienced issues with wildfires from 1998-2008. This map could be used if the DNR was looking to expand their coverage area and this would give them some justification to do so as there is huge amounts of property to protect near cities and many potential lives at stake.  

Figure 4

Evaluation

 

I'm happy with the result of this project. It allowed me to truly test what I've learned throughout the semester and accomplish something without really following step by step instructions. If I was to do the project over again I might factor some other variables in such as population of county in relation to the number of fires or something a long those lines. As far as challenges go I really only struggled at the end of the process figuring out how to only buffer near the cities which had wildfires in their proximity and not just every city in general but that was an easy problem to fix with a couple of tries.

 







Lab 3: Vector Analysis with ArcGIS

Goal:
The primary goal of this lab was to use a variety of geoprocessing tools we have learned within ArcGIS to find the best bear habitat for a study area found in Marquette county, Michigan.

Background:
Using recorded GPS data and tools in ArcGIS, the DNR of Michigan wanted to know what the best suitable habitat was for black bears following a couple of specific criteria.

Methods:
Included in the data was an excel file that had the bear locations. In order for it to be used I had to import it into ArcGIS. Once the data was imported and exported as a feature class I could add it to my geodatabase. Once adding the bear locations to my study area map I was able to see where the bears were located.

In order to find the top habitats based on the bear locations I had to join them with the land cover type class. After analyzing and summarizing the data I was able to find the top three land covers. They included mixed forest land, forested wetlands, and evergreen forest land. I created a separate feature class just including these three.

The next step was seeing how many bears were found near to streams. I put a 500 meter buffer around the streams in the study area. Along with adding the buffer it was also dissolved for clarity reasons. Once those two processes were finished the result was clipped with the bear locations. Over 70% of bears located were within 500 meters of a stream. This made sense as streams can be used as a major source of nourishment.

The next step was to find the best habitat within the 500m buffer. To do so I intersected the feature class that included the three land covers and the buffered stream. This resulted in a potential habitat feature class. I dissolved this class like the other for clarity and continuity purposes.

One of the last factors I had to account for was DNR owned land that fell into the potential bear habitat. I clipped the DNR feature class with the study area. This got rid of any impertinent data. Once again I dissolved the internal boundaries and intersected this feature with the potential habitat. This resulted in creating a feature class showcasing potential bear habitat on DNR land.

The last factor I had to consider was proximity to Urban or Built-Up lands. I ran the buffer tool on a feature class I created that just had that Major type of land cover. I then erased the result but left it on the map in a different color showing habitat that met all criteria except being outside of 5km on urban area.

The steps I used can be found in the model labeled Figure 1. Figure 2. consists of Python code that was used to make a stream buffer, intersect it with land cover and then erase the urban area near. Figure 3. shows the results of the lab.

Figure 1.

Figure 2.

Figure 3.

Results:
The map above highlights habitat areas for black bears in the study area. The DNR land is primarily in the middle of the study area. The suitable habitat near urban areas is found in the southern half specifically south west. There is a lot of suitable habitat found the northern half away especially in the northwest part away from urban areas so if the DNR was looking to expand it's coverage; there would be many suitable options.

Data Sources:

 

Land cover - USGS NLCD
http://www.mcgi.state.mi.us/mgdl/nlcd/metadata/nlcdshp.html

DNR management units http://www.dnr.state.mi.us/spatialdatalibrary/metadata/wildlife_mgmt_units.htm

Streams
http://www.mcgi.state.mi.us/mgdl/framework/metadata/Marquette.html