GIS4007 - Module 5 - Choropleth Mapping

For this module, we had the opportunity to combine choropleth and graduated symbol mapping techniques to explore two variables across Europe which were population density and wine consumption per capita. The goal was to show a pattern in wine consumption while accounting for population distribution using the spatial analysis tools in ArcGIS Pro.

The map was developed using ArcGIS Pro. The process began with acquiring the appropriate datasets from the designated drive and exporting them into my project. I explored the attribute tables for both population density and wine consumption fields, which allowed me to identify key outliers such as Monaco and Malta. Since these two had extremely high population densities and the potential to distort the choropleth classification, I utilized SQL queries for data exclusion to remove them from the display, ensuring a more balanced visual representation (which wouldn't be the only time I use data exclusion).

I tested multiple classification methods, such as Equal Interval, Natural Breaks, and Quantile, and eventually settled on Natural Breaks since it represented the data better than the other methods. Quantile ended up making a lot of countries seem way more densely populated than they actually are and I noticed that Equal Interval didn’t really reflect the natural grouping of the data the way Natural Breaks does. It was kind of flattening the data out, which could have been misleading given how spread out the population density values are

Design considerations were probably the most intricate part of this process. I changed the labeling of countries by editing the names in the attribute table into English and repositioning them to avoid conflicts with the symbols. In especially dense areas, I used an inset map to maintain visual clarity and accessibility. Although there were easier way to position the labels, I was having difficulties getting the callout for the labels to work for the country names, so I ultimately had to annotate the labeling and change each one my hand to not overlap with the symbol. I used SQL for the main map and inset map to select and deselect the symbols for wine consumption.

My final map overlays a choropleth base layer representing population density (in people per square kilometer) with graduated circles showing wine consumption per capita (in liters/person) for each European country in 2012. An equal area projection (Europe Albers Equal Area Conic) was used since choropleth maps rely on accurate representation of spatial distribution and density and using an equal area projection makes sure that the comparisons between countries are correct and not distorted by size caused by the projection.

To visualize population density, I used a sequential color scheme moving from light greenish-blue (low density) to deep blue (high density). This approach made most sense the unipolar nature of the data that we were given and made sure that denser regions were more visually more saturated, helping the map viewers interpret the pattern easier.

For wine consumption, I went for graduated symbols over proportional ones. This allowed me to group the wine data into clearly defined classes, making patterns easier to read. Each red circle’s size corresponds to a range of wine consumption values, with the largest circles indicating the highest per capita.

Countries like Italy, France, and the Vatican City have both high population densities and high wine consumption. In contrast, countries like Poland or Ukraine showed relatively low wine intake despite relatively moderate densities.

This lab was a challenging mix of technical and artistic skills. It helped me realize the importance of data normalization, the accessibility in visual design (e.g., considering color blindness with sequential color ramps), and ultimately the thoughtfulness it takes to choose classification choices.