Satellite-based remotely sensed data have the potential to provide hydrologically relevant information about spatially and temporally varying physical variables. A methodology for estimating such variables from multichannel remotely sensed data is presented; the approach is based on a modified counterpropagation neural network (MCPN) and is both effective and efficient at building complex nonlinear input-output function mappings from large amounts of data. An application to high-resolution estimation of the spatial and temporal variation of surface rainfall using geostationary satellite infrared and visible imagery is presented. Test results also indicate that spatially and temporally sparse ground-based observations can be assimilated via an adaptive implementation of the MCPN method, thereby allowing on-line improvement of the estimates.
ASJC Scopus subject areas
- Water Science and Technology