|Stochastic Systems Group|
Virat Chatdarong - McLaughlin Research Group, MIT
Rainfall is a complex environmental variable that is difficult to describe either deterministically or statistically. It is controlled by turbulent and chaotic physical processes, and it varies over a wide range of spatial and temporal scales. In addition, it is intermittent, e.g. often zero, in both space and time. Rainfall can be measured with many types of instruments each with particular characteristic scale, coverage and accuracy. It would be desirable to combine all different types of measurement to make maximum use of all available information.
In our study, we propose an ensemble Kalman smoothing algorithm to merge and estimate rainfall from multiple measurement sources. A stochastic point-Poisson-process is used to simulate and propagate spatially distributed rainfall thru time. A preliminary study shows that the algorithm is able to merge multiple rainfall measurement provide physically reasonable estimates of rainfall.
Problems with this site should be emailed to email@example.com