Urban Growth vs. Development Suitability Using Raster Overlay

Using raster overlay to analyze suitability is usually very useful in spatial-related decision making. Here I would like to show the process to assess the threat and opportunity with respect to future development in Pennsylvania by identifying important resources lands threatened by likely urban development.

First, we would like to know which areas or grid cells change from non-urban in 1992 to urban in 2001.

Next, we would like to see which counties have been most efficient with their urban land conversion with respect to population increase, as shown by Table 1. There are 9 counties which lost population between 1990 and 2000, which are highlighted by light pink-orange. The most efficient urban land conversion is found in Lackawanna, whose ratio of land conversion to population growth is largest. Northumberland ranks the second. These two counties’ land conversion efficiency far exceed others. The most inefficient land conversion is found in Indiana.

*The cell size is 499.99*499.99* square meter.

We define water, farm, pasture, and forest as "sensitive land". Table 2 shows the total number of sensitive land cells in each county and all sensitive land in Pennsylvania in 1992.

*The cell size is 499.99*499.99* square meter.

Then, we want to see which sensitive lands were urbanized in 2001 in each county, as shown by Table 3. Among all counties, Westmoreland and Allegheny have large areas of sensitive lands that were urbanized in 2001. There are 71.25 and 44.5 square kilometer sensitive areas that were transformed to urban land in Allegheny and Westmoreland respectively. In contrast, in Forest, Cameron, and Sullivan, less than 2.5 square kilometer sensitive areas urbanized.

*The cell size is 499.99*499.99* square meter.

Now, we need to figure out which areas are likely to be urbanized going forward and then which areas are worth preserving. First, we should find those areas with more likelihood to be urbanized. To do so, we could map three decision factors of urbanization potential: areas within 6km to existing urban development, slope less than 2% grade, and within 10km of 4-lane highways. Giving them each a weight, we could map the future urbanization index.

Also, we need to figure out which areas are more sensitive. There are three decision factors of sensitivity: in active farm and forest use, hillsides with slopes of 15% or more, underdeveloped sites within 1000 meters of rivers.

Giving them each a weight, we can map the final environmental sensitivity. The top 3rd highest values of environmental sensitivity are deemed as “environmentally sensitive“, with the value of 1. The rest is 0.

By combining environmental sensitivity and future urbanization potential, we could categorize all cells into four types: not sensitive and not developed, sensitive and developed, not sensitive and developed, as well as sensitive and not developed.

As shown on that two maps, there are few areas that are environmentally sensitive and might be developed in Philadelphia. Meanwhile, in west and north Philly, there are a lot of areas that are not environmental sensitive and might be developed. In my opinion, more attention should be paid to these areas, since they are appropriate for future urban development without threatening the environment. Also, there are large number of non-white, mid-low income and not well-educated people living in these areas. To direct future development to these areas might be helpful to improve their living conditions.

#ArcGIS