Integrated validation of assimilating satellite derived observations over France using a hydrological model

Abstract

This study assesses the impacts of assimilating surface soil moisture (SSM) and leaf area index (LAI) observations on river discharge using the SAFRAN-ISBA-MODCOU (SIM) hydrological model. The SIM hydrological model consists of three stages: (1) An atmospheric reanalysis (SAFRAN) over France, which forces (2) a land surface model (ISBA-A-gs), which then provides drainage and runoff inputs to (3) a hydrogeological model (MODCOU). The river discharge from MODCOU is validated using observed river discharge over France from over 500 gauges. The SAFRAN forcing underestimates direct short-wave and long-wave radiation by approximately 5% averaged over France. The ISBA-A-gs model also significantly underestimates the grassland LAI compared with satellite retrievals during winter dormancy. These differences result in an under-estimation (overestimation) of evapotranspiration (drainage and runoff). The excess water flowing into the rivers and aquifers contributes to an overestimation of the SIM discharge. We attempted to resolve these problems by performing the following experiments: (i) a correction of the minimum LAI model parameter for grasslands, (ii) a bias-correction of the model radiative forcing, (iii) the assimilation of LAI observations and (iv) the assimilation of SSM and LAI observations. The data assimilation for (iii) and (iv) was done with a simplified extended Kalman filter (SEKF), which uses finite differences in the observation operator Jacobians to relate the observations to the model variables. Experiments (i) and (ii) improved the average SIM Nash scores by about 12 % and 20 % respectively. Experiment (iii) reduced the LAI phase errors in ISBA-A-gs but only slightly improved the discharge Nash effciency of SIM (by just 2 %). In contrast, experiment (iv) resulted in spurious increases in drainage and runoff, which degraded the discharge Nash effciency by about 35%. The poor performance of the SEKF is an artifact of the observation operator Jacobians. These Jacobians are dampened when the soil is saturated and when the vegetation is dormant, which leads to positive biases in drainage/runoff and insuffcient corrections to the LAI minimum, respectively. This motivates the development of a DA method that can take into account model errors and atmospheric forcing errors. The results also highlight the important role that vegetation plays on the hydrological cycle. It is recommended that a spatially variable LAI minimum parameter be introduced into ISBA-A-gs based on the lowest LAI values derived from satellite observations.