Contribution of lower and upper tropospheric forcing to the initiation and inhibition of convection over complex terrain
Final Report Abstract
Precipitating convection in a mountain region of moderate topography is investigated, with particular emphasis on its initiation in response to boundary-layer and mid- and upper-tropospheric forcing mechanisms. The data used in the study are from COPS (Convective and Orographically-induced Precipitation Study) that took place in southwestern Germany and eastern France in the summer of 2007. It is found that the initiation of precipitating convection can be roughly classified as being due to either surface heating and low-level flow convergence, surface heating and moisture supply overcoming convective inhibition during latent and/or potential instability, or mid-tropospheric dynamical processes due to mesoscale convergence lines and forced mean vertical motion. During Intensive Observation Period (IOP) 8b on 15 July 2007, deep convection developed east of the Black Forest crest although Convective Available Potential Energy (CAPE) was only moderate and Convective INhibition (CIN) was high. Data analysis revealed that convection was triggered by updrafts penetrating the capping inversion of the planetary boundary layer (PBL) as a result of low-level convergence. Although the numerical weather prediction (NWP) model COSMO with 2.8 km grid spacing simulated a convergence line and the evolution of a line of low clouds in good agreement with radar and satellite observations, no precipitating deep convection developed from this line of clouds. For an improved representation of orographic effects, simulations with a finer grid resolution of 1 km were performed. Despite almost optimal conditions, i. e. moderate amount of CAPE and almost vanishing CIN, the updrafts required to overcome CIN were not reached in both model configurations. Although both simulations did not initiate deep convection, the results suggest that in an air mass convection situation without mid-tropospheric forcing, the simulated location and timing of convergence lines with coexistent large values of CAPE and low values of CIN can be used as diagnostic parameters for deep convection nowcasting. For the same IOP, a model intercomparison effort of five different convection-resolving numerical models with eight different configurations was performed. It is found that besides an accurate specification of the thermo-dynamic and kinematic fields, low-level convergence lines and their ability to lift parcels up to the level of free convection need to be well represented in NWP models in order to account for their triggering effects of deep moist convection and to improve the overall forecast skill. A necessary but not sufficient requirement is a high horizontal grid spacing of the model to correctly resolve those features. The fact that a reasonably good representation of the convective storm was partly caused for the wrong reasons (e. g. a too moist PBL of Meso-NH) is of particular importance and highlights the benefits of having an ensemble of models for operational weather forecasting. Moreover, this study demonstrates how widely the model results can differ and that sensitivity tests for each model are needed for the decision to use a certain model for a certain application. Analyses of the process chain of soil moisture, surface fluxes, conditions of the convective boundary layer (CBL), and convection-related parameters revealed that knowledge of the boundary-layer status with an appropriate spatial resolution is needed, i. e. the resolving of all relevant scales is necessary to forecast the distribution of conditional instability and convective inhibition in the pre-convective environment over complex terrain. Furthermore, the CBL conditions are influenced to a great extent by advective processes in mountainous terrain, too. The convective indices in the whole COPS domain are found to depend on equivalent potential temperature in the CBL. CIN is positively correlated with the strength of the CBL-capping inversion and negatively with the CBL height: the higher the CBL, the lower the upper threshold of CIN. The frequency of low CIN is higher in the Black Forest mountains than in the Rhine valley, which facilitates convection initiation over the mountain sites. Based on a synopsis of processes, it was also shown that one process alone (synoptic forcing, soil-atmosphere interactions, PBL development, convergence features, density stratification) is not enough for identifying those regions where convection initiation is most likely. For this purpose, a superposition of appropriate synoptic-scale and boundary-layer forcing mechanisms with high temporal and spatial resolution is necessary.
