Published works

We present the evaluation of high-resolution, JASMIN soil moisture analysis developed for Australia. The prototype JASMIN system has been developed primarily for use in fire and land management. JASMIN produce hourly soil moisture estimates over four soil layers at 5 km horizontal resolution. We evaluate JASMIN against three ground-based networks in Australia. Among the results, the median Pearson’s correlation obtained for surface soil moisture across the observation networks for JASMIN is between 0.78 and 0.85. JASMIN generally has a better skill than the Keetch-Byram Drought Index and Soil Dryness Index models used operationally in Australia. We also apply and evaluate a few rescaling approaches to the JASMIN soil moisture to facilitate its use in the current operational fire danger rating system. Minimummaximum matching, mean-variance matching, and cumulative distribution function (CDF) matching are the rescaling approaches applied. Validation of the rescaled products is performed using ground-based observations and MODIS fire radiative power data. The rescaling readily enables fire agencies to utilize the JASMIN product in their existing fire prediction models. However, the potential of JASMIN is greatest in the National Fire Danger Rating System (NFDRS), currently being prototyped across Australia. Particularly, the ability of JASMIN to estimate soil moisture at several levels is expected to be advantageous in the NFDRS. For example, the Spinifex fuel model implemented in the current NFDRS prototype uses 0-10cm soil moisture as an input.  This soil moisture information is available natively in JASMIN.