Emily McRae's GIS Blog

This is an example of a bivariate choropleth map. There are two variables being shown here using a color scheme that fades across two color spectrums. The variables are percent obesity and percent inactivity, by US County. To prepare data for mapping in a bivariate map, you must be able to show a relationship between the variables. To do this in ArcGIS Pro at this time, there is a bit of manual work to be done, so you can follow the steps below to prepare your data set. 1. Establish your class breaks for each of your two variables. Decide what type of breaks you will use (natural, quantile, equal interval, etc.). Establish class breaks through either manual calculations, or by setting your map display and recording the class breaks that are automatically set. 2. Add three columns in your data set; a column to flag which class your values for the first variable will be in, a column to flag which class your values for your second variable will be in, and a column that concatenates

The map symbol sizes represent the difference in jobs, and are categorized to represent the same number of jobs for both gains and losses. The orange symbol indicates that that state lost jobs and the blue indicates that it gained jobs. The symbols are adjusted with Flannery's correction, making the difference between symbol size most apparent. We can see that the largest change in employment were with jobs gained, and that this occurred largely in Texas, California, and New York. This symbology provides an easy to understand visual of the change in employment by number across US states. This is not normalized however, and represent numbers of jobs, not percent employed or unemployed.

My variables are child poverty as a percent of child population and premature death. Premature death is measured as a rate of potential years of life lost (YPLL) per 100,000 people. I am relating these variables as a means of generalizing social determinants of health under poverty, as it tends to be a commonality across persons who have experienced a number of adverse childhood experiences (ACEs). As we know, your health impacts your life span, and this is true for physical health and socio-emotional well-being. That is to say, your mothers ability to find childcare is as intrinsically linked to your likelihood of developing cancer later in life as smoking is. For my map, I am using a purple color scheme across all of my map elements. It is a dark, but relaxing color and is also associated with cancer, mental health, and suicide awareness. My maps are sharing a color ramp, and the color ramp is also reflected on my scatter plot. The legend for the two maps is joined into one si

Above is a map of population change in Colorado from 2010 to 2014 by counties. I used a 6 class color palette with natural breaks, with a break at 0. Above zero are 3 classes of positive values to represent degrees in population increase, shown in greens. Below zero are three classes to represent population decline, shown in reds. The color palette was generated in Color Brewer. The legend shows the breaks in more of a sentence structure, with class breaks described in the format '_% to _%'. This way, it is clear what the symbols represent, both in terms of the span of values per color and the fact that we are looking at percents and not numbers. There is a callout extending from the legend stating what the states population change was.

The Color Brew ramp roughly shares the same class intervals as the adjusted progression ramp, with higher steps between G and B values and lower steps between R values. The intervals between classes on the Color Brewer ramp are not equal intervals though, while the ramps I selected in ArcGIS were exact intervals. The values calculated above for the class intervals are averages. Additionally, the values in Color Brewer don’t uniformly increase as you move to a lighter shade. For example, the R values increase from 185 to 223, then decrease to 201, the rise again to 241. You can also see visually that the saturation appears to change throughout the palette in a nonuniform way, such as the second color appearing very bright and intense compared to the other colors in the palette.