Previous PageTable Of ContentsNext Page

Spatial distribution of rainfall and storm movement (using remote sensing)

Elliott, J. Mr: ph. (03) 9669 4518; fax: (03) 9669 4725; J.Elliott@bom.gov.au

Seed, A. Dr: ph. (03) 9669 4591; fax: (03) 9669 4725; A.Seed@bom.gov.au

Research Organisation: Hydrology Unit , Bureau of Meteorology, GPO Box 1289K, Melbourne Vic. 3001

Collaborators: Bureau of Meteorology Research Centre, Bureau of Meteorology, GPO Box 1289K, Melbourne Vic. 3001

Sponsor: Cooperative Research Centre for Catchment Hydrology, Monash University, Clayton Vic. 3168

Objectives: To improve the determination of the quantity and spatial and temporal variation of rainfall.

Methodology:

New ways of determining the quantity and spatial and temporal distribution of rainfall through the use of remote sensing are being investigated. In a project being undertaken as part of the research program of the Cooperative Research Centre for Catchment Hydrology, the Bureau of Meteorology is investigating ways of combining observations of rainfall from ground-based gauges, with information remotely sensed by radar and satellite, to provide improved rainfall inputs for a range of operational hydrology applications. The particular focus is for real-time flood forecasting, but the work also has application to other water management problems where improved real-time estimation of spatial and temporal rainfall is important.

Progress:

Geostatistical techniques for combining radar and raingauge data have been investigated with emphasis given to cokriging and principal component analysis. Overall, cokriging was found to produce the best results, with an improvement in accuracy of around 20-40 per cent for daily point and average rainfall measurement. Principal component analysis is more computationally efficient and was shown to provide a consistent approximation to cokriging. (Sun et al., 1998). The benefit of this improvement in areal rainfall estimation to flood forecasting was also shown (Sun et al. 1998). The project has also developed a space-time model of Australian rainfall (Seed et al. 1998) to provide improved inputs for design hydrology applications such as flood estimation and applications involving the management of hydrologic risk.

Period: Starting date: 1997-09; completion date: 2000-03

Status: ongoing

Keywords: rainfall distribution, remote sensing, radar, hydrology

Publications:

Seed, A.W., Menabde, M. and Srikanthan, R. (1998). A space and time model for design storm generation. Sixth International Conference on Precipitation: Predictability of rainfall at all scales, 29 June - 1 July 1998, Hawaii, USA.

Sun, X., Keenan, T. and Mein, R. (1998). Principal component analysis for geostatistical rainfall estimations of using radar and raingauge data. Fourth International Symposium on Hydrological Applications of Weather Radar, 5-9 April 1998, San Diego, California. USA.

Sun, X., Mein, R., Keenan, T. and Elliott, J. (1998). Flood forecasting using cokriging and probability matching rainfall estimate inputs: A comparison. Fourth International Symposium on Hydrological Applications of Weather Radar, 5-9 April 1998, San Diego, California. USA.

Previous PageTop Of PageNext Page