Introduction:
In this lab, I have been asked to pose a research question and carry out the steps necessary to answer it and present my findings.Research Question
Where are the best potential locations for a new park in the city of Eau Claire, WI based on community need and potential social impact?
Objectives
To help locate potential locations for a new park I established the following criteria:
- The new park site should not be located within .5 miles of a currently existing park
- The new park should be located in a census block group with population density > 2000 people/square mile.
- Block groups eligible for a new park should contain more than 100 children ages 5-9
- Median Household Income in eligible block groups should be < $40,000
These criterion aim to identify a population in need based on access. Furthermore they seek to maximize impact through population density and a relatively large population of young children compared to other block groups.
Intended Audience
City planners and community advocates working to improve the welfare of citizens in the city of Eau Claire will be able to utilize these findings to stimulate further discussion about future park locations. This is, by no means, exhaustive research into suitable park locations. It is meant as a starting point.
Data Sources:
In order to research the criteria I set for potential park locations, I employed data obtained on the UWEC Geography departmental server in their ESRI data folder as well as various geodatabases containing data specific to the city of Eau Claire. Shapefiles retrieved from this location were already equipped with U.S. Census Data used for later analysis. This data is marked with dates of 7/13/2009 and 11/16/2009 and was obtained from ESRI. The attribute tables revealed that 2000 and 2010 are included. Data retrieved includes the following:
- Eau Claire County boundary shapefile
- U.S. Block Groups shapefile
- Light Gray Canvas Base Map
- Eau Claire City_Limits_Area shapefile
- Water shapefile
- Eau Claire Zoning shapefile
Concerns
My initial concern had to do with locating Median Household Income data for block groups. Looking at the attribute tables for my block group shapefile revealed this information was not included. Going to U.S. Census American FactFinder revealed this information was only available at the block level from the year 2000. Due to the large gap with, not only my available data (2009,2010), but also with the current year (2014), I chose to forego this very useful (if it were current) piece of information.
Next, I realized that park data for the city of Eau Claire was not available in a clean form. It was necessary to draw out this information from zoning data. While I am confident I was able to locate all current relevant parks in the city of Eau Claire, I cannot claim this with 100% certainty.
City Parks do not exist in a separate file available to me. It will be left to my own interpretation to define which parks ought to be included for analysis. Others will certainly take issue with my interpretation. Choosing or excluding parks for any of a number of relevant reasons would most likely arrive at a different result than I will obtain.
Lastly, the city of Eau Claire has expanded in way where portions of the city are located in block groups that extend beyond the city boundaries. A decision was made to simply analyze the block groups as whole rather than trying to locate exactly which portions of the data could be attributed to sections of a block group actually in the Eau Claire city limits. These outlying areas are, most likely, irrelevant to this discussion due to their newly developing nature. Once again, though, there may be relevant areas on the outskirts of the city that were marginalized by the assessment methods employed.
By placing the city limits boundary over the U.S. block groups shapefiles, I extracted the necessary block groups and created a separate layer greatly reducing the data I would need to sift through. The resulting layer was queried twice to identify block groups that met two of my criteria (Population 2010/Square Mile-Density > 2000 and Population Age 5-9 > 100.
The former limit was chosen based on the average Density per square mile for the city of Eau Claire. Thus, block groups chosen would contain above average density. The latter limit was chosen based on looking at the various 5-9 yr. old populations for each block group. I wanted to eliminate a significant portion of block groups while still leaving several for further analyzation. I will admit the process of choosing eligible block groups based on 5-9 yr. old population was somewhat arbitrary, though still arrived at deductively.
Each of these queried selections were separated into their own layer and the Intersect tool formed a layer containing only the block groups that met both criteria
Public Parks
This portion proved to be a little more tricky than I had hoped. Because public parks were not available in their own layer, I had to query the zoning shapefile for all public land. This category did not possess any further sub-categories, so I had to digitize and select each park in order to turn them into a separate layer.
One benefit was achieved through this process; I chose to exclude parks such as Putnam Park which exists only as a walking and nature trail. Because I am putting focus on smaller children (ages 5-9), I felt parks such as these have far less relevance than other community parks. A .5 mile buffer was placed around existing parks and combined with my Density and 5-9 yr. old population layer through the Erase tool to exlude areas within eligible block groups that were too close to other parks.
City Limits
Eau Claire has many areas that have avoided annexation and an ever-expanding outer limit. As a result, the actual boundary of the city can be somewhat confusing. To alleviate this issue I utilized the Dissolve tool on the city boundary which presented a much more clear picture. This was combined with the resulting layer from my combination of Parks and Census Criteria data through the Clip tool.
This eliminated any area that was included in eligible block group and more than .5 miles away from existing parks but outside of the city boundaries. After implementing the Clip tool, I was left with the final result of my research.
Water
A shapefile containing the water for the Eau Claire county area was added to provide some color definition on the otherwise grey base-map. This was merely an aesthetic addition.
My initial concern had to do with locating Median Household Income data for block groups. Looking at the attribute tables for my block group shapefile revealed this information was not included. Going to U.S. Census American FactFinder revealed this information was only available at the block level from the year 2000. Due to the large gap with, not only my available data (2009,2010), but also with the current year (2014), I chose to forego this very useful (if it were current) piece of information.
Next, I realized that park data for the city of Eau Claire was not available in a clean form. It was necessary to draw out this information from zoning data. While I am confident I was able to locate all current relevant parks in the city of Eau Claire, I cannot claim this with 100% certainty.
City Parks do not exist in a separate file available to me. It will be left to my own interpretation to define which parks ought to be included for analysis. Others will certainly take issue with my interpretation. Choosing or excluding parks for any of a number of relevant reasons would most likely arrive at a different result than I will obtain.
Lastly, the city of Eau Claire has expanded in way where portions of the city are located in block groups that extend beyond the city boundaries. A decision was made to simply analyze the block groups as whole rather than trying to locate exactly which portions of the data could be attributed to sections of a block group actually in the Eau Claire city limits. These outlying areas are, most likely, irrelevant to this discussion due to their newly developing nature. Once again, though, there may be relevant areas on the outskirts of the city that were marginalized by the assessment methods employed.
Methods:
The data flow model below (as seen in Figure 1) was followed to obtain the visual results of this study.
Figure 1: Data flow model employed to identify ideal areas for a new park in Eau Claire, WI.
Density and 5-9 yr. PopulationBy placing the city limits boundary over the U.S. block groups shapefiles, I extracted the necessary block groups and created a separate layer greatly reducing the data I would need to sift through. The resulting layer was queried twice to identify block groups that met two of my criteria (Population 2010/Square Mile-Density > 2000 and Population Age 5-9 > 100.
The former limit was chosen based on the average Density per square mile for the city of Eau Claire. Thus, block groups chosen would contain above average density. The latter limit was chosen based on looking at the various 5-9 yr. old populations for each block group. I wanted to eliminate a significant portion of block groups while still leaving several for further analyzation. I will admit the process of choosing eligible block groups based on 5-9 yr. old population was somewhat arbitrary, though still arrived at deductively.
Each of these queried selections were separated into their own layer and the Intersect tool formed a layer containing only the block groups that met both criteria
Public Parks
This portion proved to be a little more tricky than I had hoped. Because public parks were not available in their own layer, I had to query the zoning shapefile for all public land. This category did not possess any further sub-categories, so I had to digitize and select each park in order to turn them into a separate layer.
One benefit was achieved through this process; I chose to exclude parks such as Putnam Park which exists only as a walking and nature trail. Because I am putting focus on smaller children (ages 5-9), I felt parks such as these have far less relevance than other community parks. A .5 mile buffer was placed around existing parks and combined with my Density and 5-9 yr. old population layer through the Erase tool to exlude areas within eligible block groups that were too close to other parks.
City Limits
Eau Claire has many areas that have avoided annexation and an ever-expanding outer limit. As a result, the actual boundary of the city can be somewhat confusing. To alleviate this issue I utilized the Dissolve tool on the city boundary which presented a much more clear picture. This was combined with the resulting layer from my combination of Parks and Census Criteria data through the Clip tool.
This eliminated any area that was included in eligible block group and more than .5 miles away from existing parks but outside of the city boundaries. After implementing the Clip tool, I was left with the final result of my research.
Water
A shapefile containing the water for the Eau Claire county area was added to provide some color definition on the otherwise grey base-map. This was merely an aesthetic addition.
Results:
The following map (as seen in Figure 2) identifies the ideal new park areas within the block groups meeting my criteria.
Figure 2: This map identifies the locations to be considered for a new park in Eau Claire, WI based on meeting the prescribed criteria.
Only two block groups toward the southeast of Eau Claire were able to meet the density and 5-9 yr. old population requirements employed. Two parks are located in one of them including the Meadowview School park as well as Fairfax Pool. Without the inclusion of Fairfax Pool, even more of these block groups would fall outside of a .5 mile buffer from surrounding parks. This further solidifies the need for an additional park within these two block groups.
There are three slivers toward the east of these block groups that are identified as eligible. While they may seem too small for a park, these areas indicate areas within the block group with the least amount of access and are not meant to exclude any sites within the block groups from consideration for a park.
That being said, the area to the north of these block groups would provide better access to the largest portion of these block groups with current low access. This is especially true considering that the park furthest north within these block groups is Fairfax Pool.
This study does not take into consideration the suitability of land for a potential park or its current usage. It merely identifies the areas where park accessibility is lowest and where the impact of a new park could be the greatest. Further study of these areas and their surroundings are necessary to identify a suitable park location.
Evaluation:
Overall Impression
I feel rather empowered upon the completion of this project. While my understand of GIS software is miniscule in comparison to its capabilities, I know that I can perform operations to achieve a desired result.
Personal Changes
Were I to engage in this project again, I would spend additional time considering what data would be most helpful to achieve my desired goal and what implementing that data would produce. While my goal was to identify ideal park locations, my project actually produced areas with relatively low access, high population, with a substantial population of young children that could benefit from new park. I did not really identify ideal park sites by any other measure.
I also should have spent more time understanding what data was accessible to me. I wasted a lot of time trying to find more up-to-date median household income information. Had I explored this sooner, I could have sought out other equivalents that might help produce the same kind of information.
Project Changes
I am having a hard time identifying changes I would make to this assignment. Having completed it in a way that meets my personal standards leaves me feeling each part of the exercise was essential to the learning experience.
Challenges
The challenges associated with this exercise were many. Envisioning a data flow model is very important but difficult to develop without more familiarity. Finding ways to get to data that is not in an ideal form, such as park locations, can be tricky and requires creativity.
Decisions about which elements to include and which to leave out are important and can really be the difference in a legible map and confusing map.
Managing a good geodatabase and keeping track of all of your data is a time-consuming, painstaking process. However, learning these habits is essential to producing responsible, accountable maps.
