An Assessment of Representing Land-Ocean Heterogeneity via CAPE Relaxation Timescale in the Community Atmospheric Model 6 (CAM6)

Abstract

The time needed by deep convection to bring the atmosphere back to equilibrium is called convective adjustment timescale or simply adjustment timescale, typically denoted by $\tau$. In the Community Atmospheric Model|Community Atmosphere Model (CAM), $\tau$ is the convective available potential energy (CAPE) relaxation timescale and is 1 hr, worldwide. Observational evidence suggests that $\tau$ is generally longer than 1 hr. Further, continental and oceanic convection are different in terms of the vigor of updrafts and can have different longevities. So using $\tau =1$ hour worldwide in CAM has two potential caveats. A longer $\tau$ improves the simulation of the mean climate. However, it does not address the land-ocean heterogeneity of atmospheric deep convection. We investigate the prescription of two different CAPE relaxation timescales for land (${\tau }_L=1$ hr) and ocean (${\tau }_O=1$ to 4 hr). It is arguably an extremely crude parameterization of boundary layer control on atmospheric convection. We contrast a suite of 5-year-long simulations with two different $\tau$ for land and ocean to having one $\tau$ globally. The choice of longer $\tau$ over ocean is guided by previous studies and inspired by observational pieces of evidence. Nonetheless, to complement our variable ${\tau }_O$ experiments, we perform a simulation with ${\tau }_O=1$ hr and ${\tau }_L=4$ hrs. Most importantly, our key findings are immune to the exact values of prescribed ${\tau }_L$ and ${\tau }_O$. The CAM model, with two $\tau$ values $\left({\tau }_O>{\tau }_L\right)$, improves convective-stratiform rainfall partitioning and the Madden–Julian oscillation propagation characteristics.