| description abstract | To study the effects of spatial variations of soil moisture and coverage coverage on the evolution of a prestorm environment, the Goddard mesoscale model (GMASS) was modified to incorporate a simple evapo-transpiration model that requires them two parameters. Soil moisture was estimated from an antecedent precipitation index. Relative fractional vegetation coverage was estimated from a normalized difference vegetation index (NDVI). The case study, 3?4 June 1980, is of particular interest because of the development of a tornado producing convective complex near Grand Island, Nebraska during a period of relatively weak synoptic-scale forcing. Three model simulators are compared. The first had no spatial variations in either sail moisture or vegetation; the second had soil moisture variability but no vegetation; and in the third, the observed variabilities of both soil moisture and vegetation are simulated. The modeled of effects spatial variations of vegetation and soil moisture include the enhancement of a stationary front oriented northwest-southeast through Grand Island. Prior to sunset, the unstable boundary layer collapses over a zone of cool surface temperature aligned with the observed front and coincident with an observed dry/moist soil boundary. Following the boundary layer collapse, the evolution of the ageostrophic flow exhibits a horizontally differential acceleration that amplifies the isolated upward motion over the frontal boundary. It is shown that the observed stationary front was strongly enhanced by differential heating caused by observed gradients of soil moisture, as acted upon by the vegetation cover. Thus, the run with realistic vegetation and soil moisture produces the best forecast of storm precursor conditions. | |
