HR: 1340h
AN: H33E-1423    [Abstracts]
TI: A nonhomogeneous stochastic weather typing approach for statistical downscaling of precipitation in Illinois
AU: * Vrac, M R
EM: vrac@uchicago.edu
AF: Center for Integrating Statistical and Environmental Science (CISES), The University of Chicago, 5734, S. Ellis Ave, Chicago, IL 60637 United States
AU: Hayhoe, K
H33E-1423 AF: Geosciences Department, Texas Tech University, 2500 Broadway, Lubbock, TX 79409 United States
AU: Stein, M
H33E-1423 AF: Department of Statistics, The University of Chicago, 5734 S. University Ave, Chicago, IL 60637 United States
AB: Downscaling methods try to derive local-scale values or characteristics from large-scale information such as AOGCM outputs. These methods can be useful to adress an issue of the climate change from a local point of view by understanding how this change will interact with existing local environmental features. Regional climate assessments require continuous time series for multiple scenarios and AOGCM drivers. This computational task is nowadays out of range of most of dynamical downscaling models. Here, advanced statistical clustering methods are applied to define original atmospheric patterns, that will be included as the bases of a nonhomogeneous stochastic weather typing approach. This method provides accurate and rapid simulations of local-scale precipitation features for 37 raingauges in Illinois at low computational cost. Two different kinds of atmospheric states are defined: "circulation" patterns - developed by a model based method applied to large scale NCEP reanalysis data - and "precipitation" patterns - obtained through a hierarchical ascending clustering method applied directly to the observed rainfall amounts on Illinois with an original metric. By modelling the transition probabilities from one pattern to another by a nonhomogeneous Markov model - i.e. influenced by some large scale atmospheric variables such as geopotential heights, humidity and dew point temperature depression - we see that the precipitation states allow us to model conditional distributions of precipitation given the current weather state - and then to simulate local precipitation intensities - more accurately than with the traditional approach based on upper-air circulation patterns alone.
DE: 1816 Estimation and forecasting
DE: 1839 Hydrologic scaling
DE: 1854 Precipitation (3354)
DE: 1869 Stochastic hydrology
SC: Hydrology [H]
MN: Fall Meeting 2005