Logic for Precipitation Calculation - Underlying Differences

[renep86]

Dear Team,

Highly appreciate this great project to give easy access to complex weather data sets!

I have question regarding the reported precipitation data on stations. Is there any specific logic being implemented?

Is there somewhere a flag in the data that shows the original source, e.g. ISD?

https://github.com/meteostat/jasper/blob/60057d802434cbee740687ab1c5cf264c3de130e/cron/import/noaa/hourly/global.py#L22

For instance, when I compare the reported values from single stations where rainfall is reported

With the direct readings where no precipitation is present:

Also GSOD

https://www.ogimet.com/cgi-bin/gsodres?lang=en&ind=129920-99999&ord=DIR&ano=2024&mes=02&day=19&ndays=31

vs.

https://meteostat.net/en/station/12992?t=2024-01-20/2024-02-19

It would be great to get some additional insights on the underlying data process.

Thanks!