ALBERT

Description: Comparing the Ensemble and Extended Kalman Filters for in situ soil moisture assimilation with contrasting soil conditions Hydrology and Earth System Sciences Discussions, 12, 7353-7403, 2015 Author(s): D. Fairbairn, A. L. Barbu, J.-F. Mahfouf, J.-C. Calvet, and E. Gelati Two data assimilation methods are compared for their ability to produce a deterministic soil moisture analysis on the Météo-France land surface model: (i) SEKF, a Simplified Extended Kalman Filter, which uses a climatological background-error covariance, (ii) EnSRF, the Ensemble Square Root Filter, which uses an ensemble background-error covariance and approximates random forcing errors stochastically. The accuracy of the deterministic analysis is measured on 12 sites with in situ observations and various soil textures in Southwest France (SMOSMANIA network). In the experiments with real observations, the two methods perform similarly and improve on the open loop. Both methods suffer from incorrect linear assumptions which are particularly degrading to the analysis during water-stressed conditions: the EnSRF by a dry bias and the SEKF by an over-sensitivity of the model Jacobian between the surface and the root zone layers. These problems are less severe for sandy soils than clay soils because sandy soils are less sensitive to perturbations in the initial conditions. A simple bias correction technique is tested on the EnSRF. Although this reduces the bias, it also suppresses the ensemble spread, which degrades the analysis performance. However, the EnSRF flow-dependent background-error covariance evidently captures seasonal variability in the soil moisture errors and should exploit planned improvements in the model physics. Synthetic experiments demonstrate that when there is only a random component in the precipitation forcing errors, the correct stochastic representation of these errors enables the EnSRF to perform better than the SEKF. But in the real experiments the same rainfall error specification does not improve the EnSRF analysis. It is likely that the actual rainfall errors are underestimated and that other sources of errors could limit the usefulness of this information. More comprehensive ways of representing the rainfall errors are suggested, which might improve the EnSRF performance.