Overestimated hydro inflow

[energyLS]

Checklist

• [ ] I am using the current main branch or the latest release. Please indicate. -> 58 commits behind
• [ ] I am running on an up-to-date pypsa-earth environment. Update via conda env update -f envs/environment.yaml. -> Approx 2 months old

Describe the Bug

In my model results, the hydro capacity factor is at 100%. This is probably due to extremely high inflow values, which are in the range of TWh within the first hour already:

This is tested for New Zealand. A possible source of error could be, that I have selected normalization:method: hydro_capacities in the config but didn't add the data for New Zealand in data/hydro_capacities. In this case, there should be a hint to do so in the config?

Currently running normalization:method:eia to see if it works.

[energyLS]

I have traced it down, the reason is that the installed_hydro flag is not set correctly if alternative_clustering: True is applied. This is the case because hydro_ppls.bus is chosen but hydro_ppls.region_id is correct for GADM, see here:

Happy to set up a PR which includes this fix, if there are any smarter ideas on how to avoid another case separation I am open for ideas. @davide-f I have mentioned this error a while ago, but now stumbled across it again. What is your take on it?