Estimation of Soil Hydraulic Parameters via Surface Soil Moisture Observations with Reduce-Adjoint Variational Data Assimilation Scheme

Soil hydraulic parameters play a critical role as the input of numerus meteorological, hydrological and agricultural models and greatly affect flood forecasting, drought monitoring, and agricultural and irrigation management, all of which control the land and atmosphere system. Therefore, accurate estimation of these parameters is of critical importance for land surface and land-atmosphere interaction modeling. To estimate soil hydraulic parameters field measurements are required, which in large scale need extensive measurements. In this work, a synthetic study is conducted to explore the potential of using surface soil moisture observations for estimation of soil hydraulic parameters via reduced-adjoint variational data assimilation (RA-VDA) scheme and a 1-D soil moisture model (i.e. HYDRUS-1D). Proper orthogonal decomposition (POD) is a model reduction technique, which is used to approximate the gradient calculation in Variational Data Assimilation (VDA). RA-VDA method uses POD to approximate the adjoint model and calculate the gradient of the cost function. The accuracy of the proposed method is investigated through a series of synthetic experiments. To this end, synthetic dataset and observations are generated with the aid of forward model run (HYDRUS-1D). Next, using this virtual surface soil moisture observations, the assimilation algorithm is conducted to update soil hydraulic parameters by means of RA-VDA method. A set of feasibility tests will be conducted to test the utility of the approach to large scale using remotely sensed surface soil moisture observations.