Quantifying the impact of the convective wind gust parameter on land model estimates of the surface energy balance at kilometer scales that resolve convective gusts

This study investigates the grey-zone problem of parameterizing convective wind gusts at kilometer grid scales that partially resolve secondary circulations; such as horizontal convective rolls (HCRs), boundary layer cells, and cold pools induced by thunderstorms. In this study Atmospheric Radiation Measurement (ARM) Doppler lidar and surface observations were combined with the DOE Energy Exascale Earth System (E3SM) land model (ELM) to study the impacts of simulated secondary circulations on the surface energy balance. Convection-permitting modeling (CPM) data from 7 case dates in the US Southern Great Plains with various convective boundary layers and convective storms are used as inputs to ELM. Three ELM experiments were conducted in which the meteorological input data was varied to explore the effect of secondary circulations: (1) using 3 km CPM hindcasts as meteorological input, (2) a control with secondary circulations removed by averaging 3 km CPM data to 30 km, and (3) a no gust parameter experiment as in experiment 1 but with the gust parameterization turned off. Observations support the connection between secondary circulations and surface wind speed found in ELM, with surface wind gusts lagging convective updrafts by a quarter cycle. At mesoscales, HCRs and cold pools enhanced ELM surface heat fluxes up to 50 W/m² and 150 W/m², respectively. Ground heat conduction and longwave fluxes decreased to accommodate the increased turbulent heat fluxes. Using the default subgrid wind gust parameterizations increased surface heat fluxes by 5 W/m² in HCRs, and up to 50 W/m² in cold pools. The increases persisted over time, leading to 50 W/m² (HCRs) and 120 W/m² (cold pools) cumulative hourly instantaneous flux increases from 18 to 24 UTC. Impacts of the gust parameter were found in all case dates and types studied, pointing to needed improvements in subgrid gust parameterizations. The results suggest that using unmodified wind gust parameters in kilometer-scale simulations could contribute to longer-term climate prediction biases.