GEOS Skin SST Analysis: Results from Evaluations with a Hybrid 3dVar System

As part of the development of an Integrated Earth System Analysis (IESA), the 3dVar version of the NASA-GMAO atmospheric data assimilation system (ADAS) has been extended to model and analyze skin SST using an air-sea interface layer. This layer modifies the near-surface ocean temperature (or bulk SST) to include near-surface effects, such as diurnal warming due to solar insolation, cool-skin, and changes in surface stresses due to waves which were previously not felt by the lower atmosphere. Recent updates to ADAS have turned the GMAO 3dVar into a hybrid ensemble-variational system. In this upgrade a hybrid version of the Gridpoint Statistical Interpolation (GSI) analysis is supported by an ensemble Kalman filter (EnKF), responsible for analyzing an ensemble of model integrations. To properly combine the ensemble with the skin SST analysis the EnKF must be modified to update the skin SST of the background ensemble.

This presentation evaluates the impact of analyzing the skin SST of the members of the ensemble, and having the diurnal variations due to skin layer effects participate in the specification of the backgroud error covariance used in GSI.