An important component constraining atmospheric fields in Numerical Weather Prediction (NWP) models are the latent and sensible heat fluxes provided by Land Surface Schemes (LSSs). Coupled to a NWP model, LSSs generally ingest model rainfall along with a number of other near-surface fields in exchange for water and energy fluxes. Our attention is focused on the LSS developed by Viterbo and Beljaars (1995) (hereafter VB95). VB95 is used in operational NWP models such as the Australian Limited Area Prediction System (LAPS) or the European Centre for Medium-range Weather Forecasts (ECMWF) global model. Although in operational use in Australia, VB95 has never been verified against soil hydrology datasets from the Australian region which are believed to be markedly different from data sets of other regions.

Soil moisture and soil temperature data have been collected at various sites in southeast Australia and four sites on the North Island of New Zealand since 1995. To remove the NWP model error, we are testing VB95 in stand-alone mode, forced by observed rainfall and near-surface fields collected by Automated Weather Stations (AWS) located in close proximity to the soil hydrology sites. Comparisons between model-predicted and observed soil moisture and soil temperature fields and the sensitivity of VB95 to variations in soil and vegetation parameters will be presented.

This research contributes to the Murray Darling Basin Continental Scale Experiment (MDB CSE) within the Global Energy and Water Cycle Experiment (GEWEX). The MDB CSE aims to explore water and energy fluxes in the Murray Darling Basin in southeast Australia.