Improving the Representation of Organized Convection in Numerical Weather Prediction (NWP) Models

点击翻译按钮获取中文摘要

The goal of this research is to improve the convective and microphysical parameterizations that directly impact the motion of mesoscale convective systems (MCSs) and their associated impacts on larger-scale moisture transports and quantitative precipitation forecasts (QPF). Despite continuing advances in numerical weather prediction (NWP), limitations in the ability of numerical models to accurately represent convective systems are widely recognized. Current numerical models have difficulty representing processes such as interactions between convective parameterization (CP) schemes and model grid scale processes; convective-scale momentum, heat, and moisture fluxes; and the motion of MCSs. These problems, in turn, are linked to difficulties in the prediction of convection's impacts upon the larger-scale environment. For example, prior work demonstrated that the inability of operational NWP models to adequately predict the translation speed of MCSs adversely impacts large-scale QPF. The misrepresentation of one or more physical processes likely accounts for operational models' inability to properly forecast MCS movement, but as yet these misrepresented processes have not been elucidated. The PIs' working hypothesis is that two aspects of current CP configurations lead to poor MCS forecasts: (i) the neglect of momentum adjustment in most operational CP schemes, and (ii) the lack of representation of convective organization. These and other factors preclude the development of realistic cold pools in the post-MCS model atmosphere. In part because of these limitations, many NWP efforts are now utilizing explicit convection (EC) model configurations. However, many of the other physical parameterizations in NWP models, including PBL and microphysics schemes, were specifically developed for use in models with CP schemes. Therefore, omitting CP schemes in EC models may actually result in unforeseen consequences, yielding convective forecasts that are still not accurate. Intellectual Merit. The outcomes of the research will be 1) determination of precisely why model runs employing CP schemes often move organized convection too slowly; 2) improvements to CP schemes such that convective organization and translation speed are more realistically represented and better predicted; and, 3) development of a more complete understanding of physical process representation and improved configurations for operational models that do not employ CP schemes (explicit convection). Collectively, these outcomes will be significant because they will markedly improve the forecasting of QPF and the downstream impacts of organized convection. Broader Impacts. Both PIs are committed to an integrated approach that involves undergraduates and graduate students in the research process, includes new research results and tools in courses on forecasting and modeling, and facilitates interaction between students and operational collaborators. A critical aspect of the research is the participation of NOAA Researchers, providing a conduit for the transfer of research results directly into the realm of operational NWP. The following aspects of numerical forecasts are expected to improve: 1) improved QPF downstream of quickly propagating MCSs; 2) improved representation of convectively generated QPF associated with MCSs; and 3) improved near-surface temperature and wind forecasts in the vicinity of propagating convection. Model improvements of this type could provide benefits to daily forecasts.