Soil Moisture Monitoring in the Goulburn Catchment
Riki Davidson and Jeffrey Walker
Information on how soil moisture varies in space and time has been largely restricted to point-scale ground measurements. Due to economic and logistical constraints, only a limited area can be monitored in this manner. However, recent advances in soil moisture remote sensing have made possible the monitoring of spatial and temporal variation in near-surface soil moisture. This project is concerned with the regional estimation of soil moisture by assimilating satellite observations of near-surface soil moisture into simple rainfall-runoff models. Optimal estimation techniques based on Kalman filtering will be further developed as a means of estimating depth-integrated soil moisture profiles from near-surface measurements.
Ground based monitoring of near-surface and profile soil moisture forms a major component of this study. These data will be used for evaluation of satellite measurements from the Advanced Microwave Scanning Radiometer (AMSR) and for soil moisture estimation from data assimilation. The region chosen for this study is the Goulburn River basin in the west of the upper Hunter River valley of New South Wales, an area of nearly 7 000 km2. This region was selected for: i) its large area of predominantly low to moderate vegetation cover (AMSR footprints are 40km x 70km); ii) lack of maritime effects (reducing the mixed pixel effects from water bodies); and iii) its proximity to Newcastle. Within the study region, the locations of soil moisture monitoring sites were chosen on the basis of their spatial distribution across the study region, ease of vehicular access, and a preference for monitoring sites exhibiting average rather than extreme soil moisture behaviour. These monitoring sites have been concentrated in the Krui and Merriwa River sub-catchments because i) there is either existing streamflow measuring equipment in place at the catchment outlet, or it is feasible to install relatively simple instrumentation to measure streamflow within these sub-catchments and ii) these catchments have predominantly low vegetation amounts meaning that this is where AMSR data will be of most use. It is envisioned that soil moisture and streamflow data assimilation will be integrated at a later stage of the project. A major component of locating soil moisture monitoring sites has been identifying and obtaining the cooperation of landholders.
To facilitate community participation we are in the process of establishing a website that provides information about the project and the ground based soil moisture measurement sites. As data are collected from these sites, they will be placed on a password protected section of the website and made available to interested landholders in the region. Climate data will also be collected from several locations, providing the atmospheric forcing necessary to drive the land surface model. The resultant model estimates of soil moisture (with and without assimilation) can then be compared with the measured data.