py__init_obsvar_pdaf

py__init_obsvar_pdaf(step: int, dim_obs_p: int, obs_p: np.ndarray, meanvar: float) → float

Compute mean observation error variance.

This is used by ETKF-variants for adaptive forgetting factor (type_forget=1). This can be global mean, or sub-domain mean.

Parameters

step: int

Current time step

dim_obs_p: int

Dimension of process-local observation vector

obs_p: np.ndarray[np.float64, dim=1]

Process-local observation vector. shape: (dim_obs_p,)

meanvar: float

Mean observation error variance.

Returns

meanvar: float

Mean observation error variance.