Metrics to describe the dynamical evolution of atmospheric moisture: Intercomparison of model (NARR) and observations (ISCCP)

The manuscript presents results from an exploratory study of metrics to describe the dynamical evolution of atmospheric moisture fields for interpretive studies. Specifically, we conduct diagnostic data analysis of the life cycle (initiation, growth, and decay) of instabilities of index fields of atmospheric moisture that can be related to cloudiness and rainfall from the NCEP North American Regional Reanalysis (NARR) and International Satellite Cloud Climatology Product (ISCCP) data in the southeast United States in 2000-2001. The two principal metrics consist of spatial fields of finite size Lyapunov exponents (FSLE, a measure of the growth rate of instability) and localization length (the characteristic time scale of instability, or alternatively a proxy measure of the local memory of the process). The results reveal (1) consistency among the spatial fields of the seasonal rates of instability growth (FSLE); (2) differences in the dynamical range of perturbation growth in the ISCCP observations (broad) as compared to the NARR (narrow), especially in the smaller perturbation/fast dynamics regime, which suggests stronger coupling of the physical processes in the NARR; and (3) stronger interseasonal and intraseasonal variability in the manifestation of nonlinear dynamics (e.g., strong summer/fall contrasts) in the NARR as compared to the satellite data. Larger disparities (r² ∼ 02-0.3) occur over the ocean on the west (summer) and eastern (fall) margins of the Gulf of Mexico and over land in the Trinity river basin (Texas) and at the foothills of the Ouachita mountains where agriculture and large freshwater reservoirs (large free water surfaces) exist and where the 2000 drought attained exceptional intensity. Whereas the differences over the ocean may be attributed to the subgrid-scale representation of warm core rings and the loop current over the Gulf of Mexico, the differences are likely associated with the representation of land-atmosphere interactions in the model over land. Copyright 2008 by the American Geophysical Union.

Duke Scholars

Author Ana P. Barros Civil and Environmental Engineering