Covid-19 Tracker

A set of comprehensive map visualizations of the COVID-19 outbreak from cumulative cases to cases per square mile. Visit the actual site at
Time Frame —
Spring 2020,
2 weeks
Skills —
Data Visualization
Team —
Jason Zhu
Justin Chen
In the Spring of 2020, many tools emerged to track new confirmed cases of the coronavirus from online digital maps to detailed dashboards. These tools help public health officials coordinate response efforts and allow citizens to better understand the reach of the virus so that they can prepare accordingly.

However, many of these tools lack the granular scale necessary to contextualize new confirmed cases of the virus’ across large swaths of area. In large states such as California, seeing cases on a state-wide scale may be valuable to a national health official, but is not as helpful for state officials, local officials, or regular citizens. In addition, cases often skew towards larger urban areas due to their high population density at the expense of rural and suburban areas.
Visualization from Johns Hopkins University is overly specific and makes the data hard to grasp
Visualization from HealthMap is too broad and makes the data indecipherable
Our aim was to create a set of visuals that better represents the outbreak at the appropriate scale for everyday citizens, as well as local health officials.
“The level of granularity does help us to see where there may be specific hotspots within a state,” says Robert Bednarcyk, an epidemiologist at Emory University (Popular Mechanics)
Using Mapbox GL JS and a public data set, we developed a map of the United States that displayed cumulative cases. We then moved to chart each coordinate to its relevant counties. We felt this was an appropriate amount of granularity — useful for all levels of government and for the everyday citizen. To not dismiss the importance of seeing state-by-state numbers, we also pulled U.S. state data to display on the left.

We next added properties to each county, noting the number of confirmed cases for each date. We used a publicly available GeoJSON boundary file of the U.S. counties and sourced case data from a UW GitHub repository, that outlines each case, date of confirmation and coordinates.
Per Square Mile
Visualizes cases per square mile per county
Per Capita Map
Visualizes cases per capita per 100,000 people