Lab 5: Term Project

Introduction:

For my term project, I studied where the best place for a new outdoor music venue would be in the state of Wisconsin. The objectives of this project included coming up with this spatial question, finding relevant data, analyzing the relevant data using vector geoprocessing tools, creating a data flow model, and reporting the results. Geoprocessing and cartography was done in Esri ArcGIS.

Data Sources:

To answer my question I used several data layers, including US Cities, US Major Highways, Wisconsin State Boundary, and Wisconsin Outdoor Music Venues. Most of the data was retrieved from UW-Eau Claire's online data server. Most of the data from there was taken from the Esri Online database(US Cities, US Major Highways), while some was taken from the Wisconsin DNR database (Wisconsin State Boundary). The Wisconsin Outdoor Music Venues feature class I created myself- digitizing many of the outdoor concert venues in the state. I have a number of concerns about the data used in the project. Firstly, the data is not necessarily current. For example, I noted that the cities feature class's most current population field was from 2007, where the population of those cities may have changed greatly since then. Another data concern I had was the general error associated with digitizing my own feature class. I placed a point feature on the venue's land, when really the feature would be best reflected by a polygon. Also, I may have not included venues that I am not aware of.

Methods:

First, I added my feature classes to the map. This included US Cities, US Major Highways, WI State Boundary and Existing Music Venues. I then clipped those feature classes that contained data for the whole United States to the WI state boundary. I was left with a map of Wisconsin that had my data in it. I made sure to project my data frame to an appropriate projection- for this project I used NAD 1983 Wisconsin TM. Noting that all cities included in the clipped Cities feature class had populations of at least 10,000 I buffered it, leaving all cities with a boundary of 20 miles around it. Next, I used the select by attributes tool to select all cities with a population greater than 50,000, and created a buffer of 100 miles around it. Next, I selected all Interstate Highways and US Highways with the select by attribute tool, and buffered the results by 20 miles. I then buffered my Existing Venues by 50 miles. Next, I used the intersect tool, inputting the buffered feature classes of Cities >50,000, Cities > 10,000, and Interstate or US Highways. I was left with all potential locations of outdoor venues that were within 100 miles of cities with a population of at least 50,000, within 20 miles of cities with at least 10,000, and no more than 20 miles from Interstate or US highways. Next, I used the erase tool to remove the buffered Venues feature class from the results. This left me with the above criterion, this time excluding any location that was within 50 miles of existing outdoor concert venues. Finally, I used the intersect tool to only include results that were within WI state boundaries, as the buffers used previously had gone outside state boundaries. See below for data flow model and final map.

Results:

There are a number of proposed sites shown in the above map. These include a relatively large area right in the middle of the state, centered around Stevens Point, Plover and Wisconsin Rapids. There is another possible area near Marshfield, one near Lacrosse and another centered on River Falls. There is another possible area in NE Wisconsin, centered around Marinette. There are a couple small polygons near Waupun and Sheboygan as well. 

Evaluation:

This project was useful in fully understanding the geospatial process, as I was required to formulate my own spatial question, find my own data and process the data accordingly. If I were to repeat the project, I would first be sure that there was relevant data available, such that I wouldn't have to digitize my own data. This induced more error than I would have liked, and was time consuming. Also, conceptualizing my spatial question to the point that it wouldn't need to be revised would be beneficial.

Lab 4: Vector Analysis with ArcGIS

Introduction:

The goal of this lab was to determine suitable habitat for bears in the study area of Marquette County, Michigan using Esri ArcGIS vector analysis tools.

Methods:

First, I added the provided x, y coordinates of bears in Marquette county to a map in Arc GIS. Next, I used the Intersect tool to generate a feature class containing each bear's coordinates, and the land cover type in which it was found. I then summarized these results to find the top 3 land types that the bears were found in. Next, I used the Buffer tool to create a boundary around the streams in the area, and used the Intersect tool to create a feature class that showed me the bears that were found within 500 meters of a stream. Noting that 72% of bears were found within this distance, I decided to include this factor in my suitable habitat. I then used the Select by Attributes tool to create a feature class that only contained the top three land cover types mentioned earlier, and intersected this with the buffered streams feature class. I dissolved the results to remove unwanted internal boundaries. This left me with a suitable bear habitat based on land coverage type and proximity to a stream. Next, I restricted the results to DNR management areas by adding a feature class containing all of the DNR management lands in the county, and using the Clip tool to restrict this to my study area. Again, I used the Dissolve tool to remove the unwanted internal boundaries in the resulting feature class. I then intersected the DNR managed lands in my study area with the suitable habitat area, leaving me with all suitable bear habitat on DNR managed land. Next, I queried the land cover feature class to get all urban and built-up land. I gave the selection its own feature class and buffered it by 5 kilometers. I then used the Erase tool to remove any land within the 5 kilometers of any urban land, which left me with all DNR managed land with suitable bear habitat that is at least 5 kilometers away from urban land. See below for data flow model and map.

Results:

The above map shows the locations of bears in Marquette county, the total suitable bear habitat, and the suitable bear habitat that is in DNR managed land. The habitat model was based on having a coverage type of either mixed forest, forest wetland or evergreen forestland, being within 500 meters of a stream, and being at least 5 kilometers from urban land coverage.

Data flow model

Sources:

All data was downloaded from the Michigan Center for Geographic Information.

• Landcover is from 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 from 
• http://www.mcgi.state.mi.us/mgdl/framework/metadata/Marquette.html

Lab 2: US Census Bureau Information

Introduction:

The goal of this lab was to learn how to download and map data from the U.S. Census Bureau. In doing this, I was also introduced to a number of new data tools and file formats.

Methods:

I first navigated to the US Census Bureau website. Here I searched for some data, choosing a total population dataset by county. I downloaded this, extracted it, and inspected the Excel files. I then returned to the US Census Bureau website, and found an available shape file of Wisconsin's counties. I downloaded this and extracted it, and put it on a blank map. I then added the Excel file, and joined it to the counties shape file. This essentially added the content from the Excel file to the counties shape file, based on a common field (that being county ID). I then returned once more to the Census Bureau website and downloaded a sex by age data file, and added it to the map following the same steps as above. I created a new data frame and joined this data to another counties shape file. I placed these two maps side by side, and edited their symbology settings so that I had two maps- one showing population per county, and another showing male population under 5 by county.

Results:

The pattern for the county population map is logical, as where there are larger metropolitan areas, the population is higher. The darkest county is Milwaukee county, which houses the largest city in the state. On the percent of males under age five map, there is no obvious pattern. However, the counties with lower percentages tended to be towards the North of the state.

Sources:

http://factfinder2.census.gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t

Lab 3: GPS Data Collection

Introduction:

The goal of this lab was to become familiar with the Trimble Juno GPS unit and its Esri ArcPad software by using it to collect point, line, and polygon features on campus, then translating them to Esri ArcMap. The objectives of the lab were as follows:

1. Create a geodatabase.
2. Prepare the geodatabase for deployment to the Trimble Juno for field data collection using ArcPad Data Manager.
3. Load the Geodatabase onto the Trimble Juno.
4. Become familiar with the basics of the Trimble Juno GPS and ArcPad through an instructor led demo.
5. Collect point, line, and polygon features in the field using ArcPad on the Trimble Juno GPS
6. Check the collected data back into ArcGIS from the field.
 

Methods:

I began by creating a geodatabase in Esri ArcCatalog. I then created a number of feature classes for testing purposes, and a few more for collecting the data shown below. I imported a shapefile of the buildings on campus, and a raster image of the area of interest. Next, I used ArcPad Data Manager to check out these feature classes, and got them ready for use on the handheld GPS unit. Then, I connected the unit to the computer, and deployed my data to ArcPad. I then went outside, turned on the GPS unit, and began familiarizing myself with different data collection methods. I then used the point averaging method to delineate three grassy areas on campus. I also used the point streaming method for three more. I also collected six point features, three trees and three light posts. Finally, I used the point averaging method to digitize the footbridge on campus. After collecting my data, I re-connected to the computer and checked in my data. I organized the map to be more cartographically pleasing. This is shown below. (Note: The background image is outdated, so some of the digitized features may appear to be overlapping buildings etc.)

Results:

The different collection methods used yielded similar results, but there were a few small differences. In collecting polygon data, the point streaming method collected points automatically as I walked around the area I was digitizing. This made the polygon's corners duller, because of the way I walked around them. Also, this caused there to be more variablity around the edges, as the GPS unit has some degree of error. The point averaging method yielded a more precise deliniation, but it was more time consuming, as I had to add each individual vertex. It also worked well for deliniating the walking bridge.

Lab 1: Base Data

Goal and Background:

The confluence project is a collaboration of public and private parties, working together to redevelop an area of downtown Eau Claire, Wisconsin at the confluence of the Eau Claire and Chippewa rivers. A community arts center, public plaza, housing, commercial/retail space, and parking are all to be constructed on two parcels in this location.
The goal of my project was to prepare a report containing relevant mapping data. Below I have provided maps of civil divisions, census boundaries, PLSS divisions, parcel data, city land use, and the voting districts for the area of the confluence project.

Methods:

To create these maps, I used ESRI Arc GIS 10.2. I began by creating six data frames. In the first, I added a feature class containing Eau Claire County's border. Then, I added a civil divisions feature class. This is the Civil Divisions data frame. In the PLSS Features data frame, I added townships, sections, and quarter-quarter sections feature classes. I gave them unique colors so Eau Claire county was given a grid of PLSS features. In the Census Boundaries data frame, I added a block group feature class and a tracts feature class. I used the block groups data to display population density, with the tracts feature class over the top. In the zoning data frame I added a zoning feature class and grouped a number of zones together to simplify the map. I then provided a centerlines feature class to show roads, trails and the like. In each data frame, I zoomed in to a reasonable distance, and added the proposed site feature class into each data frame. It is denoted by two red polygons on each map.



Confluence Project. Click to expand.

 

Sources:

http://www.eauclairearts.com/confluence/
http://www.uwec.edu/News/more/confluenceprojectFAQs.htm
http://volumeone.org/news/1/posts/2012/05/15/3134_arts_center