Correlations

Mean values of climate variables

some first-look insights

it’s good that the correlation between normal and warme strom consumption is very low. It is expected, and could be useful to use normal strom utilization as a proxy for other things (i.e. being at home basically)
also as expected, the correlation between normal strom utilization and temprature is low.
there is a nice and strong negative correlation between temperature and wärmestrom consumption. It is not strictly linear though. Rather polinomial (inverse) relatioship.
there seems to be some sort of weak positive correlation between wärmestrom and relative humidity. Not at all consistent though. It is positive with min and mean humidity, but negative or non-existent with max humidity. It may well be just an artifact or spurious correlation, explained by temperature, because there is indeed a correlation (not that strong, though) between temperature and humidity.
the number of observations are somewhat related to the temperature, which kinda makes sense, because the measurements are not random. I am much more likely to measure the thing when it’s cold!, and that seems to be reflected in the data.