Key points are not available for this paper at this time.
The main purpose of this paper is to demonstrate that it is possible to estimate soil moisture from the evolution of atmospheric parameters near the surface (temperature and relative humidity) if a realistic surface transfer model is available. Two methods to initialize soil moisture in meteorological models are then proposed: a variational method where the optimal soil moisture minimizes a penality function and a sequential method consisting of a set of predictions and static corrections of soil moisture. The algorithms are examined with a one-dimensional model including a detailed land-surface parameterization. A feasibility study is undertaken using the HAPEX-MOBILHY dataset where soil moisture has been measured together with atmospheric parameters. It is demonstrated that for three 48-h clear-sky periods the two methods are able to converge rapidly toward a realistic soil moisture content starting from arbitrary values.
Jean‐François Mahfouf (Fri,) studied this question.