How does the old saying go? Look to the past… something something… the future? Anyway. Sometimes when looking for solutions to spatial issues you have to look into the historic conditions that made those issues possible. In a digital world, the past is not always accessible in your desired working form. That’s why digitizing exists.
In Arc Desktop, you can use the georeferencing toolbar to take any image and give it spatial relevance. When bringing in image data, ArcMap will alert you that the data had no spatial reference. This is when the georeferencing toolbar will activate. When overlaid with a raster or shapefile of the same spatial area, you can add control points between the two to make the non-spatially referenced image line up with the spatially referenced map. Through a gradual game of point-and-click, we can move the non-referenced historic map image into place.
During my Urban GIS class, I delved into the litter crisis ongoing in Philadelphia. Why has Philadelphia become so infested by plastic snack bags and street mattresses? I started sifting through the information I could find easily: Demographics, poverty, locations of public waste bins, etc. The distribution of high rates of litter throughout the city affected impoverished and minority groups the most. What I didn’t know was why. Once I introduced data on vacant an abandoned property into my analysis, I realized that the conditions which breed excessive litter may have something to do with the marginalization of specific neighborhoods and groups of people.
Redlining historically restricted or eliminated any investment opportunities inside neighborhoods that were deemed “undesirable”. I’ll give you one guess to figure out what that meant. I wanted to know if a history of redlining in Philadelphia influenced the way litter is distributed throughout the city today. Upon failing to find a nice ready to use shapefile, I surrendered myself to the (then) scary world of digitizing. Here is the 1937 Home Owner’s Loan Corporation Map, also known as a redlining map. Areas marked in green were the most desirable, followed by blue, yellow indicated to be cautious with investment, and red referred to completely undesirable areas.
Before the HOLC map was digitized, I could only do a visual inspection to see how the areas that I found to heavily littered compared with the areas of redlining. With digitization, I could do a real overlay analysis to see how these areas lined up. First I used georeferencing to align the HOLC map with a shapefile of Philadelphia. Once the HOLC map had its spatial reference, I created a new shapefile and used the editor toolbar to draw my own polygons around each of the designated desirability sections on the HOLC map (I saved myself some time by only drawing out the polygons that were inside Philadelphia county). In the attribute table I assigned a number for each of the desirability scores so I could distinguish each of the polygons and be able to use them against the demographic data I have acquired.
What I found was that as the litter score in an area increased, the level of desirability according to the HOLC map decreased. My conclusion from this overlay was that litter could be symptom of a greater issue within a city – that of lack of investment. This is a conclusion I could not have come to had I not been able to access and manipulate historic maps, like the one seen here.