Add support for RADOLAN data

[amotl]

Dear @mmaelicke,

within jeremiahpslewis/dwdbulk#5, you asked for RADOLAN support. Most probably, reading the binary files would prove to be more efficient.

Integrating wradlib [2] or radproc [3] into dwdweather2 should be possible, but would it make any sense? I am happy to receive any thoughts about this.

With kind regards,
Andreas.

[1] https://opendata.dwd.de/climate_environment/CDC/grids_germany/hourly/radolan/recent/bin/
[2] https://wradlib.org/
[3] https://github.com/jkreklow/radproc

[mmaelicke]

Hey Andreas,

For me personally, I can say that I implemented a quick&dirty script grabbing the RADOLAN data needed for analysis, which is not super complicated. The reason I asked was that the environment I am working in is a remote cluster and I usually prefer to build upon other packages that are maintained, as usually, it is exactly these scripts that screw up the toolchains.

I did not open an issue here, because fitting RADOLAN into the caching logic of dwdweather2 would be quite a heavy lift, wouldn't it?
I was not aware of wradlib, seems to be quite helpful, thanks for pointing out.
A possible use-case, that I can see would be an interface to either of these two libraries, based on a list of stations (from the observations_germany branch). This might be a common use-case to first download meteorological data and then the RADOLAN covering the stations downloaded. Does that make sense?

Best,
Mirko

[amotl]

Dear Mirko,

For me personally, I can say that I implemented a quick&dirty script grabbing the RADOLAN data needed for analysis, which is not super complicated. The reason I asked was that the environment I am working in is a remote cluster and I usually prefer to build upon other packages that are maintained, as usually, it is exactly these scripts that screw up the toolchains.

I see. Been there, seen that ;]. Thanks for outlining your scenario.

I did not open an issue here, because fitting RADOLAN into the caching logic of dwdweather2 would be quite a heavy lift, wouldn't it?

Well, it would definitively be some work, right.

I was not aware of wradlib, seems to be quite helpful, thanks for pointing out.

You are welcome. I hope it will be useful to you.

A possible use-case, that I can see would be an interface to either of these two libraries, based on a list of stations (from the observations_germany branch). This might be a common use-case to first download meteorological data and then the RADOLAN covering the stations downloaded. Does that make sense?

Absolutely. Thanks for outlining this. Let's keep this issue as a reference.

With kind regards,
Andreas.

[amotl]

Dear @mmaelicke,

thanks again for your input on this and for joining the discussion at earthobservations/wetterdienst#71 already. Let's focus our efforts on Wetterdienst instead of spending useless cycles with this library, right? So, I am deliberately closing the issue here.

With kind regards,
Andreas.