Impacts of Vertical Grid Resolution on the Representation of a Convective Environment

Abstract

In a pre-convective environment, modeled atmospheric conditions can be presented as discrete parameters, which serve as valuable resources both in forecasting potential severe weather events and also in assessing changes in convective potential over time. However, the value of these parameters may depend on the vertical resolution of the simulation that is used to do the calculation. Upper-air radiosonde sounding data from various research campaigns (e.g., PECAN, DC3, TRMM-LBA, etc.) in both midlatitude and tropical convective regimes are analyzed and later interpolated to the pressure levels in a range of atmospheric reanalysis and operational models. Indices such as CAPE, CIN, bulk shear, and strength of capping inversion are then calculated using both the raw sounding and the interpolated soundings. Interpolated soundings consistently underestimated CAPE, although models with a larger number of vertical levels underestimated CAPE less (e.g., ERA5). Kinematic parameters; however, were well correlated to the observed. The accurate representation of the capping inversion was found to be sensitive to both the number and distribution of levels in the first several kilometers. Increasing vertical resolution from previous model versions, may improve accuracy of parameters (e.g., ERA5, GFSv16). Based on vertical resolution alone, this study shows that decreased vertical resolution can impact the ability of convective parameters to represent a potentially convective environment.