Journal of Advances in Modeling Earth Systems (Jun 2013)

Mechanisms of marine low cloud sensitivity to idealized climate perturbations: A single‐LES exploration extending the CGILS cases

  • Christopher S. Bretherton,
  • Peter N. Blossey,
  • Christopher R. Jones

DOI
https://doi.org/10.1002/jame.20019
Journal volume & issue
Vol. 5, no. 2
pp. 316 – 337

Abstract

Read online

Climate change sensitivities of subtropical cloud‐topped marine boundary layers are analyzed using large‐eddy simulation (LES) of three CGILS cases of well‐mixed stratocumulus, cumulus under stratocumulus, and shallow cumulus cloud regimes, respectively. For each case, a steadily forced control simulation on a small horizontally doubly periodic domain is run 10–20 days into quasi‐steady state. The LES is rerun to steady state with forcings perturbed by changes in temperature, free‐tropospheric relative humidity (RH), CO2 concentration, subsidence, inversion stability, and wind speed; cloud responses to combined forcings superpose approximately linearly. For all three cloud regimes and 2× CO2 forcing perturbations estimated from the CMIP3 multimodel mean, the LES predicts positive shortwave cloud feedback, like most CMIP3 global climate models. At both stratocumulus locations, the cloud remains overcast but thins in the warmer, moister, CO2‐enhanced climate, due to the combined effects of an increased lower‐tropospheric vertical humidity gradient and an enhanced free‐tropospheric greenhouse effect that reduces the radiative driving of turbulence. Reduced subsidence due to weakening of tropical overturning circulations partly counteracts these two factors by raising the inversion and allowing the cloud layer to deepen. These compensating mechanisms may explain the large scatter in low cloud feedbacks predicted by climate models. CMIP3‐predicted changes in wind speed, inversion stability, and free‐tropospheric RH have lesser impacts on the cloud thickness. In the shallow cumulus regime, precipitation regulates the simulated boundary‐layer depth and vertical structure. Cloud‐droplet (aerosol) concentration limits the boundary‐layer depth and affects the simulated cloud feedbacks.

Keywords