Estimation of Large-Scale Vertical Motion over the Central United States for Summer

Abstract

We investigate the method-dependence of large-scale vertical motion (LSVM) estimates given by four variants of the kinematic approach (Endlich-Clark triangle; Chien-Smith pentagon; Objective analysis; Kung optimization) and the Limited-area Fine-Mesh II (LFM-II) model of the U.S. National Weather Service. The treatment spans 54 rawinsonde sounding times from three contrasting periods during the 1979 summer in the central United States that each included both widespread and abundant rainfall and intervening dry spells. Quantitative LSVM intercomparisons and evaluations for 500 and 700 mb are are with respect to cloud cover and rainfall data for 138 locations. The Kung (especially) and LFM-II estimates have the narrowest frequency distributions. A particularly broad distribution is evident for the Endlich-Clark method. The Kung and LFM-II approaches yield significantly more frequent estimates of upward LSVM for overcast and rainy conditions (64–74 percent, depending on level and weather category) than the other methods (57–66 percent). However, the LFM-II also gives significantly more frequent estimates of upward LSVM for the fair weather condition of “0–5/10 cloud cover” (44–47 percent, depending on level) than all of the other techniques (36–41 percent). The Kung method gives very low such frequencies (38–39 percent). The above fair weather result, together with more detailed frequency distribution information, suggest that the LFM-II may be biased toward giving upward LSVM at 500 and 700 mb. The foregoing findings are supported by additional analyses that intercompare the LSVM methods with respect to (i) their full frequency distributions when the more extreme cloud/rainfall conditions prevail and (ii) the frequency of occurrence of various cloud/rainfall categories when strong LSVM (defined separately for each method) is estimated. The results collectively suggest the Kung method to be superior to the other kinemtic approaches, in some cases substantially so, and also to the LFM-II. They offer guidance for the treatment of LSVM in meso- and synoptic-scale studies and climate dynamics.