Exploring coronovirus over time

The GeoHealth Lab @ CRSSA has been exploring municipal patterns of COVID-19 in New Jersey. With 565 municipalities, this is a fairly detailed look at the movement of the virus across the landscapes of the Garden State. It is fraught with challenges since not all municipalities are reporting these numbers every day, they may not be using consistent definitions, and the municipalities range wildly in area and population. While we are culling these data from a variety of sources - tweets by mayors, county web sites, various newspapers - the reporters at NJ.com have not only been publishing numbers, but they published a great explanation of why you should be wary of these data. In that spirit, these early data visualizations are offered as a rapid response. 

While you should be careful in reading these data, as longitudinal data they display some remarkable patterns. Eventually there will be more carefully checked data that is presented in carefully balanced representations. But as we are currently surrounded by this pandemic, we should see what we can in the moment.  As a simple example, we can start just comparing the reported rates (cases per 100,000 residents) for just a few cities:

Early reports asked why New Brunswick (yellow) wasn't experiencing growth as fast as a few other cities, but the graph shows New Brunswick (blue) catching up with Newark's rate in  recent days. There is also a spatial pattern implied, as the southern cities lagged - it seems that northern New Jersey started first but Trenton is starting to catch up. Will Atlantic City (light blue at the bottom) see a later jump in its rate? The answer to that will have policy implications for the statewide shutdown.

Having suggested that the pattern is spatial, what we need is a map that shows us the changes over time. Here is an animated GIF tracking the municipal data from March 31 to April 27. (If it isn't animated for you, you can download it or refresh this page.)

The first time through you might just notice that the map gets darker. But with a second run through the animation you might notice that it starts with higher rates in Bergen County and spreads West and South. As it does, you can also see that it moves like a wave, washing across both large and small towns.

To see if large cities were having similar experiences, we selected out the 20 largest municipalities in New Jersey by population. For each of these, we graphed their change in reported rates over time. With so many squiggles it is hard to pick out just one town, but the legend shows their rankings and. more importantly, the graph shows that most of them are experiencing similar trends (consistent growth), but at decidedly different rates. Again, don't get lost in details, since these data could still use some polish. But the big pattern can help identify where to watch in the coming days.

Keep an eye out for updates. There is plenty more in the pipeline.