Abstract
The dependence of the RegCM3 (Regional Climate Model version 3) downscaling skill on initial conditions (ICs) and lateral boundary conditions (LBCs) are investigated for the 1998 summer flood along the Yangtze River Basin in China. The effect of IC uncertainties is depicted by 15 realizations starting on each consecutive day from April 1 to 15 while all ending on September 1, 1998 with identical driving LBCs, analyses are based on June, July and August simulations. The result reveals certain IC effect on precipitation for daily evolution but little for summer mean geographical distribution. In contrast, the effect of LBCs uncertainties as represented by four different reanalyses are notably larger in both daily evolution and summer mean distribution. The ensemble average among either 15 IC realizations or 4 LBC runs does not show important skill improvement over the individuals. None of the RegCM3 simulations (including the ensemble means) captured the observed main rain band along the Yangtze River Basin. This general failure suggests the need for further model physics improvement.
Highlights
Since Lorenz [1] first found that the atmosphere is a highly nonlinear system causing numerical solutions to be sensitive to small perturbations in initial conditions (ICs), and as computational capability is rapidly increasing, ensemble weather forecasts are becoming a common operational practice (Toth and Kalnay [2]; Mullen et al [3])
A comparison of the spatial frequency distribution for summer mean biases in precipitation and surface air temperature over all grids east of 100°E, where observations are abundant, shows no overall difference between IC_ENS and R-2 and all other IC runs. These results indicate that the ensemble mean has little improvement than individual realizations as far as the 1998 summer seasonal averages concerned
The RegCM3 downscaling skill dependence on initial conditions is examined by comparing summer simulations driven by R-2 starting from consecutive dates from April 1 – 15, 1998
Summary
Since Lorenz [1] first found that the atmosphere is a highly nonlinear system causing numerical solutions to be sensitive to small perturbations in initial conditions (ICs), and as computational capability is rapidly increasing, ensemble weather forecasts are becoming a common operational practice (Toth and Kalnay [2]; Mullen et al [3]). A Regional Climate Model (RCM) requires ICs at the initial step, and time-dependent lateral boundary conditions (LBCs) which provide large-scale atmospheric circulation through buffer zones that drive the development of mesoscale systems inside the RCM domain. Both ICs and LBCs can be provided by GCM simulations. The RCM sensitivities to these ICs and LBCs uncertainties are investigated for the summer of 1998 when severe flooding occurred along the Yangtze River Basin This case has been identified with active convection at regional to local scales under strong anomalous planetary forcings (Samel and Liang [12]) and is ideal for.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
