Synoptic circulation and land surface influences on convection in the Midwest U.S. "Corn Belt" during the summers of 1999 and 2000. Part II: Role of vegetation boundaries

In Part I of this observational study inquiring into the relative influences of "top down" synoptic atmospheric conditions and "bottom up" land surface mesoscale conditions in deep convection for the humid lowlands of the Midwest U.S. Central Corn Belt (CCB), the composite atmospheric environments for afternoon and evening periods of convection (CV) versus no convection (NC) were determined for two recent summers (1999 and 2000) having contrasting precipitation patterns and amounts. A close spatial correspondence was noted between composite synoptic features representing baroclinity and upward vertical motion with the observed precipitation on CV days when the "background" (i.e., free atmosphere) wind speed exceeded approximately 10 m s^-' at 500 hPa (i.e., "stronger flow"). However, on CV days when wind speeds were <∼10 m s^-1 (i.e., "weaker flow"), areas of increased precipitation can be associated with synoptic composites that are not so different from those for corresponding NC days. From these observations, the presence of a land surface mesoscale influence on deep convection and precipitation is inferred that is better expressed on weaker flow days. Climatically, a likely candidate for enhancing low-level moisture convergence to promote deep convection are the quasi-permanent vegetation boundaries (QPVBs) between the two major land use and land cover (LULC) types of crop and forest that characterize much of the CCB. Accordingly, in this paper the role of these boundaries on summer precipitation variations for the CCB is extracted in two complementary ways: 1) for contrasting flow day types in the summers 1999 and 2000, by determining the spatially and temporally aggregated land surface influence on deep convection from composites of thermodynamic variables [e.g., surface lifted index (SLI), level of free convection (LFC), and lifted condensation level (LCL)] that are obtained from mapped data of the 6-h NCEP-NCAR reanalyses (NNR), and 0000 UTC rawinsonde ascents; and 2) for summer seasons 1995-2001, from the statistical associations of satellite-retrieved LULC boundary attributes (i.e., length and width) and precipitation at high spatial resolutions.