Abstract
This paper focuses on the development and application of a new One-Dimensional Variational (1DVAR) data assimilation algorithm for estimating the spatial and temporal variations of soil moisture and temperature profiles, by grid-based analysis using remote sensing and in situ observations. This algorithm employs a heuristic optimization approach, simulated annealing (SA), which is capable of minimizing the Variational cost function without using adjoint models. The present assimilation scheme assimilates passive microwave remote sensing observations of brightness temperature into the land surface scheme (LSS), Simple Biosphere Model2 (SiB2). The LSS is used as a model operator, and a Radiative Transfer Model (RTM) is used as an observational operator. The modeling system has been applied, and validated, using data from the GAME-Tibet (Global Energy and Water cycle Experiment (GEWEX) Asian Monsoon Experiment in Tibet) mesoscale field experiment. Compared to SiB2, our assimilation scheme solves the major initialization problem, and estimates the soil temperature and soil moisture at the surface layer and in the root zone with significant improvements.
