IMIT   21220
INSTITUTO DE MODELADO E INNOVACION TECNOLOGICA
Unidad Ejecutora - UE
artículos
Título:
Estimating Model Parameters with Ensemble-Based Data Assimilation: Parameter Covariance Treatment
Autor/es:
RUIZ J.; PULIDO M.; MIYOSHI T.
Revista:
JOURNAL OF THE METEOROLOGICAL SOCIETY OF JAPAN
Editorial:
METEOROLOGICAL SOC JPN
Referencias:
Lugar: Tokio; Año: 2013 vol. 91 p. 453 - 469
ISSN:
0026-1165
Resumen:
In this work, various methods for the estimation of the parameter uncertainty and the covariance between theparameters and the state variables are investigated using the local ensemble transform Kalman filter (LETKF). Twomethods are compared for the estimation of the covariances between the state variables and the parameters: one using asingle ensemble for the simultaneous estimation of model state and parameters, and the other using two separateensembles; for the initial conditions and for the parameters. It is found that the method which uses two ensemblesproduces a more accurate representation of the covariances between observed variables and parameters, although thisdoes not produce an improvement of the parameter or state estimation. The experiments show that the former methodwith a single ensemble is more efficient and produces results as accurate as the ones obtained with the two separateensembles method. The impact of parameter ensemble spread upon the parameter estimation and its associatedanalysis is also investigated. A new approach to the optimization of the estimated parameter ensemble spread (EPES) isproposed in this work. This approach preserves the structure of the analysis error covariance matrix of the augmentedstate vector. Results indicate that the new approach determines the value of the parameter ensemble spread thatproduces the lowest errors in the analysis and in the estimated parameters. A simple low-resolution atmosphericgeneral circulation model known as SPEEDY is used for the evaluation of the different parameter estimationtechniques.