As an Environmental Planning Intern at the Delaware Valley Regional Planning Commission (DVRPC), I spent this past year working with staff in the Office of Environmental Planning to create a verified inventory of off-road walking, hiking, and biking trails in several New Jersey counties. This ongoing project has included several phases, including data collection and processing, feature editing, verification, and mapping/visualization, all of which I have had the privilege of contributing to. Though the inventory is still in the works for all four counties (including Burlington, Camden, Gloucester, and Mercer), much of the verification and mapping has been completed for Mercer County. The overall goal of the project is to create a single GIS-friendly file with selectable layers of verified and unverified trails, categorized by surface material and multi-use status, for each of the four counties.

After collecting data from planning staff in Mercer County, processing was completed using Excel and ArcGIS. Beyond basic processing, extensive feature editing was performed in order to ensure correct alignment and positioning of trail polylines. This was done using ArcMap’s editing tools with trail lines layered over high-resolution, leaf-off aerial imagery from Nearmap. Nearmap imagery was also used to verify trail surface material, in combination with Google Maps imagery and official published park and trail maps from organizations like the NJDEP and the New Jersey Trails Association. Figure 1 illustrates this process, visualizing trail surface types and verification status layered over aerial imagery.

Figure 1. Mercer County trail verification using Nearmap aerial imagery

In addition to verifying surface type and correcting geometry, I was tasked with classifying each trail segment according to whether it is multi-use or not. Where possible, I used ArcMap’s distance measuring capabilities to measure the width of trail segments to determine whether they conform to AASHTO standards for path width. In cases where this was not permitted by Nearmap or Google imagery, I referred to official park and/or trail brochures to determine the multi-use status of trails.

Finally, I used ArcMap to determine the percentage of each surface type, which enabled me to make the chart included in Figure 2, which is a county-wide visualization of verified and unverified trails of each surface material. As shown in the pie chart, I was able to successfully verify approximately 71% of the trails in Mercer County using the methods I’ve described so far. Future work may involve performing field checks to verify trails that are not visible in aerial imagery or included on park and trail maps. At this point in the project, next steps include verifying data for Camden and Gloucester Counties, merging the data into a single file for all four counties, and producing more maps and visualizations focused on multi-use status, verification status, surface type, spatial distribution, etc. Eventually, DVRPC hopes to combine this inventory of trails in its New Jersey counties with its existing Regional Trail network data to provide a comprehensive inventory of trails and connections throughout the Greater Philadelphia Region.

Figure 2. County-wide verification of trails

In May of 2019, I started an internship at the Economy League of Greater Philadelphia working as a Justice Policy Research Intern. During the first three months, I was working on a Criminal Justice Research project pertaining to the economic impact that a pardon could have on an individual with a criminal record, as well as their families and communities. My main tasks for this project were to conduct literature searches that would inform the basis of the project as well as design maps on socio-demographics, arrests rates, pardon applications filed and granted (2008-2018) throughout the state of Pennsylvania. Moreover, there was an emphasis placed on five key counties that had high arrest rates that included: Philadelphia, Bradford, Dauphin, Potter, and Allegheny County.

Following that, the data used in creating these maps came from the United States Census Bureau, Federal Bureau of Investigations, and the Pennsylvania Board of Pardons. I created approximately 60 maps for the project that my supervisor asked for but due to time restraints and the narrowness of the project, only two of the maps I created were included in the final product (http://economyleague.org/uploads/files/518454652334570386-impactofpardons-final.pdf). Those maps displayed rates for pardon applications filed and granted over top of county arrest rates for Pennsylvania. Pardon applications and grant rates were visualized using graduated symbols technique, while the arrest rates used a graduated color method.

Figure 1: Pardon application rates over top of arrest per capita in Pennsylvania.

Moreover, the pardon application data were allocated to the zip code level, and the arrest rates were designated at the county level. These two geographies were overlaid with one another to create the two maps. Figure 1 displays a map showing pardon application rates over county arrest rates. In looking at this map moderate and high arrest counties have lower rates of pardon applications filed, while low arrest counties have a higher rate of pardon applications filed. Additionally, low rates of pardon applications filed seem to be concentrated in the southeastern (Philadelphia County) and western (Allegheny County) portions of Pennsylvania. Next, figure 2 shows a map of granted pardon rates over county arrest rates. In this map grant rates for pardons appear to be lower across the board except for in counties such as Bradford, Venango, and

Figure 2: Granted pardon application rates over top of arrest per capita in Pennsylvania.

Washington County that have rates of 0.035. These two maps taken together show that the rate of pardon applications filed and pardons granted are higher in less crime-ridden communities, while higher crime-ridden communities have lower rates of pardon applications filed and granted. This spatial analysis revealed a disparity in how pardons are distributed to certain types of communities.

Overall, I found this project to be rewarding in that I got to utilize my knowledge of criminal justice that I obtained from my undergraduate education in conjunction with my coursework in the P.S.M program. Ultimately, this enabled me to craft nice map visualizations and contribute in a meaningful way to criminal justice policy, while gaining practical experience in the application of GIS for public policy.