==================
== Madhav Kumar ==
==================

Spartanburg Place:Non-Place Ratio

I recently read about Places and Non-Places and thought it would be fun to try it out on Downtown Spartanburg.



Results#

I calculated a Place:Non-Place ratio of 0.91:1 (47.55% place) for Downtown Spartanburg. This shows slightly more land is used for Non-Places than Places as represented by this map:

Caveats#

First, a few caveats. This is a very rough estimate. I created the map using a combination of hand-tracing and the publicly available data of the parcels in Spartanburg.1 As a result, not all of the Places/Non-Places line up perfectly.

Additionally, I was not super strict with the definition of of Place and Non-Place. For example, I designated the entirety of the County Courthouse as Place despite there being a fair amount of empty lawn space there. I also did not distinguish between open and closed businesses.

I could have easily spent ages to make this perfect, but then I might not have ever actually finished. In the spirit of incrementalism, I figure let’s start here and refine if needed.

Thoughts#

I am a little disappointed by this result. It shows that more than 50% of our downtown area’s land is occupied by Non-Places. The other side of this, however, is that it shows we have a lot of opportunities to improve! I also know the city is currently in the process of making changes that will improve this. For example, the upcoming changes to Daniel Morgan square will expand the Place area further.

Hopefully we can use this analysis to help identify morea areas where we can start to incrementally convert Non-Places to Places. More Places would also augment all the great work being done by PAL on the Dan. From a financial perspective, there’s also a huge potential gain in tax revenue from property and sales taxes that we’re missing out on by underusing our land!

This is also the densest area of our city which indicates the surrounding areas will likely have much worse Place:Non-Place ratios. I would love for our city to take steps to improve this. We could start by removing parking minimums entirely for example. If we want to even be more radical, we could institute parking maximums. I’d also love to see some creative adaptive reuse of some of the parking lots.

Methodology#

For the purposes of this project, I defined downtown as the area designated as the “Spartanburg Downtown Cultural District”. In future iterations it might be fun to see how the results change based on other definitions of downtown.2

I went through and traced out all of the parking lots and garages within that area in a Non-Place layer. Then, I imported the data for parcels in that area from the City of Spartanburg Zoning map into a Place layer.3 I moved any empty parcels over to the Non-Place layer and added unmarked Places like Daniel Morgan Square and Wall Street to the Place layer. To account for parking spaces on parcels, I subtracted the Non-Place polygons from the Place polygons to get the final Place area.

Subtracting the Place polygons from the base Cultural District polygon gives us all of the roads and areas like the greenspace near the railroads. Combining that with the existing Non-Place polygons gives us our final Non-Place area.

We can then calculate the areas4 of the Culutral District, Places, and Non-Places to get our ratios and percentages. The Python scipt is here.


  1. Also, shoutout to the City of Spartanburg for creating these interactive maps, making them easily accessible, and keeping them up to date! It makes my life of being a huge nerd so much easier. ↩︎

  2. Originally, I mapped out my own definition. I asked my partner where they would draw the boundaries of downtown and while we had some similarities there were enough differences for me to seek out a more formal definition. ↩︎

  3. Technical aside: It was quite an adventure to figure out how to convert the GeoJSON data I yoinked from Google into a geometry object format required to apply a spatial filter to the query. ↩︎

  4. One more technical note: When calculating the distribution of the Dan Trails, I used the ESPG:32718 reference system to get the length of the trails. In this case we’re calculating areas so we need to use an equal-area projection. I chose to use ESPG:9822. All cartographic projections distort some distances so different projections are required for different applications. I have a very rudimentary understanding of all of this but I think it’s so neat. ↩︎