Modeling, Variability and Predictability of North American Hydrologic Extremes

Nevada System Of Higher Education, Desert Research Institute, Reno NV

Investigators

Abstract

This research is intended to produce tools to assess potential regional effects of global anthropogenic climate change on the structure of North American daily hydrology in different seasons. The PI will apply climate prediction methodology to forecast the probabilities of hydrological weather extremes. A new comprehensive statistical framework will be developed for the description of daily precipitation and streamflow data. Using this framework, the PI will examine variability and predictability of daily hydrological extremes. He will apply statistical methods designed for data with heavy-tailed distributions (distributions with relatively large numbers of extreme events that can't be accounted for by exponentially decaying tails of traditional Probability Density Functions (PDFs). He will then examine the influence of climate forcing in space and time on the shape of seasonal distributions of daily hydrology. The parameters describing the PDF's shape will become predictands from which statistical information, e.g., frequencies of extreme events, can be derived. The PI will work with scientists at UCSD, SIO and the University of Nevada, Reno. This research has broader impacts of research results applied to the important societal issue of extreme hydrologic events in the context of a changing climate.