Sampling strategies and assimilation of ground temperature for the estimation of surface energy balance components

The performance of a land data assimilation system for surface ground temperature sensing is demonstrated for the U.S. Southern Great Plains 1997 Hydrologic Field Experiment. Adjoint state formulation is used in a variational scheme to minimize the error of surface ground temperature predictions subject to constraints imposed by the system model, It is shown that continuous sampling of observations result in accurate estimation of the components of the surface energy balance and an index of soil moisture. Experiments on the effects of sparse temporal sampling (near the mean of minimum and maximum in the diurnal cycle) on the estimation show that observations at the peak of the diurnal cycle is the most suitable for the land data assimilation system. It is suggested that surface ground temperature within a similar to3 h window centered on this time in the diurnal cycle contains information on the cumulative heating and available energy partitioning at the land surface.