Gerrymandering has been an established tradition in American politics and in other countries that similarly change their electoral district, as political parties try to change the boundaries of political districts in their electoral favor. The challenge has been to develop something potentially equitable that can be claimed to be independent of the political process. With spatial tools and technologies, there just might be a solution but wariness remains.
One approach could be to use K-means clustering that treats districts like finite resources where efficient boundaries are found relative to the population in a state. The algorithm optimizes efficient distancing from different cluster points. The most efficient solution is determined by the stability of the answer across a space where improved optimization or location of points, and thus the districts, cannot be made better in regards to the population’s location. Another approaches also include using Voronoi diagrams of equitable weighting and distribution, but this method is criticized for lacking consideration of population concentration. In other words, the K-means method better accounts for uneven distribution of population and resource spread across a region (or states in US districts).
One criticism of using GIS or spatial approaches for redistricting has been that GIS may not account for “communities” of interest and even a method that is abstract or neutral could simply divide communities because people’s understanding of community may not correspond easily to data used in a spatial approach. Nevertheless, despite the criticism, states have tried to use GIS and spatial techniques to redistrict more equitably. In a study from more than ten years ago that was conducted of states that did attempt to apply GIS methods, gerrymandering was still found to be common. Political parties, the Department of Justice, and state government all have applied GIS techniques. This has meant that all parties can now use data to manipulate outcomes to their favor. It was found that state parties now employee staff to help find data and structure data so that districts favor particular parties using GIS. In effect, gerrymandering has become easier due to GIS and spatial technologies, as before redistricting efforts often took very lengthy periods and many personnel to conduct the process. The complexity of the process may have helped limit it in places, while the fact that GIS makes redistricting easier makes it more likely parties could use it for their own benefit.
Gerrymandering does not only occur for political district boundaries but can occur in school districting. This was found in US public schools, using a GIS analysis of racial and ethnic segregation with attendance zones. In comparing the datasets, it was found that gerrymandering could be occurring in places where there is rapidly changing racial or ethnic change. The analysis was conducted by comparing Voronoi diagrams with the location of schools, where discrepancies are found. Simpson index is used to capture diversity of ethnic/racial backgrounds.
A recent spatial and simulation analysis of the combined political gerrymandering in Congress from all the states has suggested that neither Republicans nor Democrats have made notable advantages. The simulations show different scenarios desired by different parties would have likely results in similar national-level results since states could be balanced out by other states’ political party benefits. The simulations uses voting behavior and patterns across demographics from election outcomes and if populations’ votes counted in alternative districts than the district in which they found themselves in. (As a side note, for an in-depth look at voting behavior between Democrats and Republicans resulting in “unintentional gerrymandering” read Rigging Elections: A Spatial Statistics Analysis of Political and Unintentional Gerrymandering, by J. Petti. 2017)
Overall, while gerrymandering continues to be a political problem in states, particularly as GIS is now even used to aid in the gerrymandering process and evidence suggest that the use of GIS has even made it easier to gerrymander, other studies have suggested that the net results at a national level may indicate that gerrymandering has a negligible effect. Nevertheless, gerrymandering has important implications particularly for local services, including in education and access to good schools. New techniques may offer better ways to formulate districts, but criticism, as seen before, has been applied to GIS methods claiming to be neutral.
 For more on critique of using GIS or spatial methods for redistricting, see: Forest, B. (2004). Information sovereignty and GIS: the evolution of “communities of interest” in political redistricting. Political Geography, 23(4), 425–451.
 For more information on how GIS has been used to benefit parties in redistricting, see: Wachter, K. W. (2005). Spatial demography. Proceedings of the National Academy of Sciences, 102(43), 15299–15300. https://doi.org/10.1073/pnas.0508155102
 For more on gerrymandering for school districting, see: Richards, M. P. (2014). The Gerrymandering of School Attendance Zones and the Segregation of Public Schools: A Geospatial Analysis. American Educational Research Journal, 51(6), 1119–1157.
 For more on the net effect of Congressional gerrymandering in recent elections, see: Chen, J., & Cottrell, D. (2016). Evaluating partisan gains from Congressional gerrymandering: Using computer simulations to estimate the effect of gerrymandering in the U.S. House. Electoral Studies, 44, 329–340. https://doi.org/10.1016/j.electstud.2016.06.014
The first set of data that have returned have demonstrated GEDI is able to measure density of forests far better…
Recorded presentations from the 2019 QGIS International User Conference are now available online.
User can download this free handbook which provides insight into how different Synthetic Aperture Radar (SAR) methodologies can be used…
Derek Watkins has created a simple and easy to use interface for downloading 30-Meter SRTM data.