The default behavior for adding and solving with noisemodels has changed from Pastas 1.5. Find more information here

pastas.stats.signatures.autocorr_time#

autocorr_time(series, cutoff=0.8, **kwargs)[source]#

Lag where the autocorrelation function exceeds a cut-off value.

Parameters
  • series (pandas.Series) – Pandas Series with DatetimeIndex and head values.

  • cutoff (float, optional) – Cut-off value for the autocorrelation function. Default is 0.7.

  • kwargs (dict, optional) – Additional keyword arguments are passed to the pastas acf method.

Returns

Lag in days where the autocorrelation function exceeds the cutoff value.

Return type

float

Notes

Lag in days where the autocorrelation function exceeds the cutoff value for the first time. The higher the lag, the more autocorrelated the time series is, and vice versa. In practical terms higher values mean that the groundwater system has a longer memory and the response to changes in the forcing are visible longer in the head time series.