HR: 1340h
AN: A53B-0891    [Abstracts]
TI: Validation and Projection of North American Daily Screen-Level Temperature and Precipitation in the CCM
AU: * Douville, H
EM: herve.douville@meteo.fr
AF: METEO-FRANCE/CNRM, 42 avenue Coriolis, Toulouse, 31057 France
AU: Gershunov, A
EM: sasha@ucsd.edu
AF: Scripps Institution of Oceanography, La Jolla, San Diego, CA 92093-0224 United States
AB: While coupled ocean-atmosphere general circulation models are generally validated on monthly to seasonal timescales, they are now widely used to predict potential climatic changes in extreme weather events under projected changes in radiative forcing due to human activities. In the present study, the daily minimum and maximum screen-level temperature and the daily precipitation simulated by the CNRM climate model (CCM) are validated over North America against in situ observations interpolated onto the CCM horizontal grid. The comparison between the simulated and observed distributions is made for both winter and summer seasons. It includes basic statistics such as moments and frequencies, but also more sophisticated diagnostics such as composites of daily temperatures for wet versus dry days, warm spells in summer, cold spells in winter, as well as wet spells for both seasons. Generally speaking, the CCM shows more realistic statistics for the winter than the summer season over North America. The main deficiency is an overestimation of summer precipitation, which is mainly due to large positive biases in the frequency of rainy days and particularly of low and medium precipitation rates. This drizzle problem is less pronounced in summer, but contributes significantly to the temperature biases found in both winter and summer. The daily temperature composites for wet versus dry days show realistic patterns for winter minima, but not for summer maxima. Cold spells and wet spells are well simulated in winter, which gives some confidence in the projected changes in frequency, date of occurence, duration, and intensity of these extreme events in the CCM. Conversely, the relatively poor simulation of warm spells and wet spells in summer emphasizes the need to improve the dynamical and physical processes that control these events in the CCM before providing credible scenarios on summer extremes with this model.
DE: 3309 Climatology (1620)
DE: 1836 Hydrologic budget (1655)
DE: 1854 Precipitation (3354)
DE: 1630 Impact phenomena
DE: 0325 Evolution of the atmosphere
SC: Atmospheric Sciences [A]
MN: 2004 AGU Fall Meeting