oemObservationErrorCalc

Workspace.oemObservationErrorCalc(self, oem: OptimalEstimationData = self.oem) → None

Calculates the covariance matrix describing the error due to uncertainties in the observation system.

The uncertainties of the observation system are described by oem.measurement_vec_error_covmat, which must be set by the user to include the relevant contributions from the measurement and the forward model.

Prerequisite for the calculation of measurement_vec_error_covmat_observation_system is a successful oemCalc() computation where also the gain matrix has been computed.

The result is \(\mathbf{S}_{\rm obs}=\mathbf{G}\mathbf{S}_\epsilon\mathbf{G}^{\top}\). This is the observation contribution, not the full posterior covariance; see Gain, averaging kernel, and retrieval uncertainty for the assumptions and decomposition.

Author: Simon Pfreundschuh

Parameters:

oem (~pyarts3.arts.OptimalEstimationData, optional) – Numerical problem and results for oemCalc() and oemCalcReduced(). Defaults to self.oem. [INOUT]