 |
||||
|
||||
|
|
|
 |
||
|
||||
      |
Great Lakes Environmental Research
Laboratory
Distinguished Scientist Seminar Series
"Space-time rainfall organization and its role
in validating quantitative precipitation forecasts"
Dr. Efi Foufoula-Georgiou
Location: Great Lakes Environmental Research Laboratory 2205 Commonwealth Blvd. Ann Arbor, MI 48105-2945 Date: Friday, October 30th, 1998 Time: 10:00 am Room: 105 (Main Conference Room)
Dr. Efi Foufoula-Georgiou, of the Department of Civil Engineering, University of Minnesota, will be presenting a talk entitled "Space-time rainfall organization and its role in validating quantitative precipitation forecasts" on Friday, October 30th at 10:00 am in the main conference room of the NOAA Great Lakes Environmental Research Laboratory, Ann Arbor, Michigan.
Dr. Foufoula-Georgiou's major research interests are in the area of stochastic modeling of surface and subsurface hydrologic processes and systems. Specific areas of research include stochastic modeling of space-time rainfall using rain gauge, radar, and satellite data; estimation of exceedance probabilities of extreme design events; geomorphologic study of river networks and hydrologic response; optimal operation of water resources systems under uncertainty; and monitoring network design.
Her current research focuses on studying and understanding the statistical structure of hydrologic processes at a range of scales of interest to hydrology, and developing a theoretical framework for multi-scale analysis of stochastic processes. Of particular interest are remotely sensed rainfall fields and river network structures.
Several studies have presented evidence for the presence of scale-invariance in spatial, temporal or spatio-temporal rainfall patterns. Such an organization in observed patterns is hoped to be also statistically reproduced in patterns predicted by atmospheric numerical weather prediction models. However, this idea has not yet been explored for assessment of the predictive skill of atmospheric models and for validation of quantitative precipitation forecasts (QPFs). This talk reviews evidence of space-time organization in observed rainfall and advocates the need to develop new multiscale measures of forecast performance, i.e., measures that can depict how well the scale-to-scale variability (be that scale-invariant or not) of the forecasted fields matches that of the observed fields. We demonstrate that these measures are more informative and provide valuable feedback for atmospheric model improvements as compared to traditional measures of QPF verification.
Last updated: September 19, 2002 mbl