Project:

Produce a series of thematic maps for the U of MN Equine Genetics Lab of the Veterinary Population Medicine Department. These maps will help illustrate a theory that laminitis, a disease that affects the feet of horses, is linked to the presence of certain dioxins (environmental pollutants) in soil where the animals live. The goal of the lab is to secure grant funding to study this theory. These maps will help them make their case by showing the survey results are unbiased and illustrating that a majority of suspect animals do live in close proximity to superfund sites.

The Data:

  • Human population density per county
  • Horse population density per county
  • Environmental superfund sites known to contain relevant soil dioxins
  • Suspect horse data collected through surveys of horse owners

The superfund data was the most difficult to find and synthesize. I ultimately found what we needed from the US Environmental Protection Agency and was able to pair it down to what was relevant–superfund sites known to contain the dioxins of interest. The survey data was messy but with some effort I was able to clean it up and geocode it. The US Department of Agriculture and US Census Bureau supplied the population data for horses and humans that was also needed.

Format: The map deliverables are categorized in the following formats:

MS PowerPoint presentation slides: A series of maps for Nichol Schultz’s PhD thesis defense whereby layers are added as she is presenting and introducing the concepts. In addition to layering, the data is also separated out by class and presented one class at a time to illustrate that the survey results are unbiased–in other words they do not all tend to come from areas highly populated by humans and/or horses. Nichol may or may not use the legend and title information when presenting thus I have included versions of the maps with, without, and containing minimal map elements as part of this submission that she can customize.

Tabloid Prints (11” x 17” documents): Four maps will be printed and included as part of a written grant proposal. These maps may be viewed by potential funders without interpretation and thus include common map elements. They are included in this submission as PDFs.

3. Google Earth: A KML version of all the data for exploration and possible use in the presentation as a Google Earth tour.

Methods: The million dollar question was how to represent all this data on one map?

Turns out the answer was you don’t. After much experimentation with proportional symbols on top of choropleth and multivariate choropleth we realized that the most effective way to illustrate the point was to show no more than 3 of the 4 datasets at a time. For the two population datasets, I chose to use choropleth grayscale symbology in an effort to provide detail without detracting from the point features of utmost importance. This decision, combined with the notion they wanted to show proximity, not actual distance, to superfund sites, lead me to utilize an equal-­area projection. The point features were challenging to get just right. There was significant overlap that I dealt with using transparency and point distribution. To highlight proximity I created a 30 mile buffer feature around the superfund sites and intersected it with the survey results. I used a thin circle outline to symbolize the points which fell within 30 miles of a superfund site. The net effect was almost like someone took a highlighter and circled all the survey result points that lie in the buffer zone.

Data used for these maps is confidential and thus the maps themselves cannot be published online here. Please contact me if you would like to see them in person. Below is an unlabeled Google Earth Screenshot to give you an idea.

Googe Earth Screenshot

<< Back