Super-scaling of wind gusts with thermal downdrafts in cloud-resolving simulations

Severe convective wind gusts ( ≥ 25 m s − 1 ) can cause significant local damage, but their small-scale nature makes them difficult to predict. This study aims to understand how gusts are simulated and to investigate the factors that control their strength, using high-resolution numerical simulations of three severe gust events. Our main finding is a linear downdraft-gust relationship with slopes greater than unity, suggesting that dynamical influences on gusts (other than the downdraft) scale with downdraft strength, resulting in enhanced gusts. Consistent with stronger low-level downdrafts, we found that severe gusts were up to four times more likely to be simulated with horizontal grid spacings ( Δ x) of 200 m compared to Δ x = 1 km. We also identify a contrasting model resolution dependence of low- versus high-level drafts: deep drafts appear stronger in the coarse grid, which evidence suggests is due to inhibited shallow convection and the subsequent build-up of instability, a mechanism we refer to as the “shallow release hypothesis”. The factors influencing gust strength, including the downdrafts’ thermal-like characteristics, may be under-resolved at Δ x = 1 km, resulting in weaker simulated wind gusts. The challenge of simulating such small-scale, complex phenomena could have implications for climate change risk assessments. Future research should examine whether even cutting-edge kilometer-scale climate projections may fail to adequately capture future severe wind events, thereby underestimating their impacts. Severe convective wind gusts ( ≥ 25 m s − 1 ) can cause significant local damage, but their small-scale nature makes them difficult to predict. This study aims to understand how gusts are simulated and to investigate the factors that control their strength, using high-resolution numerical simulations of three severe gust events. Our main finding is a linear downdraft-gust relationship with slopes greater than unity, suggesting that dynamical influences on gusts (other than the downdraft) scale with downdraft strength, resulting in enhanced gusts. Consistent with stronger low-level downdrafts, we found that severe gusts were up to four times more likely to be simulated with horizontal grid spacings ( Δ x) of 200 m compared to Δ x = 1 km. We also identify a contrasting model resolution dependence of low- versus high-level drafts: deep drafts appear stronger in the coarse grid, which evidence suggests is due to inhibited shallow convection and the subsequent build-up of instability, a mechanism we refer to as the “shallow release hypothesis”. The factors influencing gust strength, including the downdrafts’ thermal-like characteristics, may be under-resolved at Δ x = 1 km, resulting in weaker simulated wind gusts. The challenge of simulating such small-scale, complex phenomena could have implications for climate change risk assessments. Future research should examine whether even cutting-edge kilometer-scale climate projections may fail to adequately capture future severe wind events, thereby underestimating their impacts.