KNMI-data assigned to stress periods wrongly

[rubencalje]

Right now we make an error, when we add recharge from the KNMI-server to a model dataset. The recharge from the next stress period is assigned to the current stress period. See the code in read\knmi.py:

model_recharge = pd.Series(index=ds.time.data, dtype=float)
for j, ts in enumerate(model_recharge.index):
    if j < (len(model_recharge) - 1):
        model_recharge.loc[ts] = (
            timeseries.loc[ts : model_recharge.index[j + 1]].iloc[:-1].mean()
        )
    else:
        model_recharge.loc[ts] = timeseries.loc[ts:end].iloc[:-1].mean()

The index of the pandas Series timeseries is equal to the end of the period that the data describes. Therefore I will change this into the following code:

model_recharge = pd.Series(index=ds.time.data, dtype=float)
for j, ts in enumerate(model_recharge.index):
    if j == 0:
        start = ds.time.start
    else:
        start = model_recharge.index[j - 1]
    mask = (timeseries.index > start) & (timeseries.index <= ts)
    model_recharge.loc[ts] = timeseries[mask].mean()

Keep in mind this will change model output in the next version.

[rubencalje]

Fixed