Any modeling is based on a decent amount of assumptions of behavior, from climate to health to economic models, so definitely use caution in any of those. It’s just another preliminary study that will evolve over time as it gets reviewed. The problem is it’s just such a sensitive topic right now that it’s almost impossible for anyone to read it in a truly objectively, IMO.
At least it’s weather relevant factoring in temperature and precip to estimate indoor vs outdoor traffic .
Ygcst=β +β Sturgis +β Temp +β Precip + β 'Xst+α +τt +εgct