|
| HYDROGRACE |
Soil moisture is a key component, meteorology and agricultural studies. However, the spatial and temporal variation in soil moisture is typically not well known or easy to measure. Ground-based soil moisture profile measurements may be made continuously at individual points but these are not always representative of the spatial distribution; microwave remote sensing may be used to give measurements of the soil moisture content in the top few centimetres for areas with low to moderate vegetation cover but do not provide any direct information on the total soil moisture content in the soil moisture profile or for areas with dense vegetation; and land surface models may be used to predict the spatial and temporal variation of soil moisture but these estimates suffer from inadequate model physics, parameter estimates and atmospheric forcing data. One current approach has been to reduce the uncertainty in the soil moisture predictions, by assimilating (the process of finding the model representation that is most consistent with the observations) the remotely sensed near-surface soil moisture measurements into the land surface model and relying on the point measurements for validation. However, there has been no way to close the water balance, at least not up till now.
The year 2002 saw the launch of NASA's Gravity Recovery And Climate Experiment (GRACE) satellites, which are mapping the Earth's gravity field at such a high level of precision that it will be possible to infer changes in terrestrial water storage (soil moisture, groundwater, snow, ice, lake, river and vegetation), thus allowing scientists to check the closure of global water balance predictions for the first time.
This project is concerned with making terrestrial observations of variations in the time dependant component of the Earth's gravity field so as to make effective use of the GRACE satellites for monitoring changes in soil moisture content. The timing of the GRACE mission also enables us to exploit a large amount of hydrological monitoring infrastructure installed for another purpose. The project presented here has three distinct yet inter-linked components that all leverage off the same ground-based monitoring and land surface modelling framework, resulting in a high level of synergy. These components are: (i) field validation of a relationship between soil moisture and changes in the Earth's gravity field, from ground- and satellite-based measurements of changes in gravity; (ii) development of a modelling framework for the assimilation of gravity data to constrain land surface model predictions of soil moisture content (such a framework enables the downscaling and disaggregation of space-borne gravity measurements by making use of remotely sensed information, such as the higher spatial (25 km) and temporal (daily) resolution remotely sensed near-surface soil moisture measurements from the Advanced Microwave Scanning Radiometer for the Earth observing system (AMSR-E).
This project will develop and test innovative techniques for the prediction of terrestrial water storage, with a particular emphasis on soil moisture content, from ground- and satellite-based monitoring of variations in the time-dependent component of the Earth's gravity field. This will be achieved by assimilating ground- and satellite-based gravity measurements at the point- and regional-scale respectively, and by the combined assimilation of gravity and near-surface soil moisture measurements into a land surface model. Extensive ground-based monitoring of soil moisture and groundwater changes will be undertaken for (i) correlation with the ground- and satellite-based gravity measurements and (ii) validation of the modelling and data assimilation developments. The important field work required by this project will be in the Murrumbidgee catchment, where an extensive soil moisture monitoring program by the University of Melbourne is already in place.