We calculate it with respect to the previous sample like you say. But I think there are better ways to calculate precipitation rate. This calculation is just based on a correction calculation from Jason. We implemented it to match what Jason was doing. But if you want to compare across stations, then maybe we can find a better solution?
@ladsmund
I think it will make sense to define the rate wrt an explicit interval such as hourly or daily.
@PennyHow
We could definitely do that... so a resample everything to hourly or daily intervals before performing the correction and rate calculation
From @ladsmund:
Response from @PennyHow:
@ladsmund
@PennyHow