*** Data comes from here: https://www.ncei.noaa.gov/access/search/data-search/coop-hourly-precipitation?startDate=1988-01-01T00:00:00&endDate=1989-01-01T23:59:59&bbox=45.006,-79.763,40.506,-71.870&pageNum=1&place=State%20or%20Province:17 Type state into location, pick a date range somewhere in the 1980's Look at stations listed and choose those starting pre-1950 and ending recently (e.g. late 2020 or any time in 2021) click download remove the small files (if a lot of years are missing) Move all downloaded files to a data_directory cd data_directory gzip * *** Or simply use the cached data with no arguments gunzip stations-hourly.pkl.gz mkdir obj mv stations-hourly.pkl obj python3 hourly-rainfall.py *** To run the code you need python3 and pip3 pip3 install matplotlib pip3 install numpy pip3 install scipy Then run: python3 hourly-rainfall.py [list of station CSV files individually gzipped in data_directory]