This is just an ongoing personal R project aimed to see if there is any discernible relation at the municipal level for climate conditions (temperature and humidity) and the spread of COVID-19
Today (ie. last update) = 2 May 2020
As of today, to run this you just need R and a couple basic libraries (Tidyverse, lubridate).
You'll see one script and one folder: data. It contains what it seems to contain.
The data folder is where action comes from and goes to. CO.csv and MX,csv contain individual-level datasets for COVID cases in Colombia and Mexico. municipiosmx.csv is just a list of official statistical codes for Mexico municipalities. All the data comes from official sources (INEGI for the codes, Secretaría de Salud for Mexico's case dataset, and Ministerio de Salud for Colombia's). Download corresponds to today (2 May 2020).
script.R simply where the action happens. It takes municipal data and merges it with real-time weather data. Where does it come from? I downloaded it from OpenWeatherMap project with my own API (you can obtain yours by signing up!). In case you don't want or cannot sign up, I also left in the folder some weather data with its corresponding date of download: see data folder, files with the name structure weather_______.csv where ________ corresponds to the download date.
If I find the time, I might include some way to estimate real-time reproductive numbers for each municipality. I might also add a dataset containing temperatures and humidity levels for several days.
Jorge Galindo - @JorgeGalindo
To download and deal with OWM data, I use the super-useful owmr package maintained by Stefan Kuethe. Stefan was kind enough to respond to an inquiry I made to him by providing a fundamental chunk of code to properly download OWM data. So we can consider him a necessary contributor to this little project.
Free to use however you see fit! But I wish you read some specialized literature before getting too excited about trends, etc. I'm no epidemiologist myself, that's why I know you'll need it in case you are neither. I hope you can provide something useful to the public debate. I'm sure you will :)