pysteps.verification.ensscores.ensemble_spread

pysteps.verification.ensscores.ensemble_spread(X_f, metric, **kwargs)

Compute mean ensemble spread for a given skill metric.

Parameters:

X_f : array-like

Array of shape (l,m,n) containing the forecast fields of shape (m,n) from l ensemble members.

metric : str

The skill metric to be used, the list includes:

Returns:

out : float

The mean skill compted between all possible pairs of the ensemble members, which can be used as definition of mean ensemble spread (as in Zacharov and Rezcova 2009 with the FSS).

Other Parameters:  

thr : float

Intensity threshold for categorical scores.

scale : int

The spatial scale to verify in px. In practice it represents the size of the moving window that it is used to compute the fraction of pixels above the threshold for the FSS.

References

[ZR09]