Effects of radiative cooling on the tropical convective response to sea surface temperature: 2-D large domain cumulus ensemble simulations

C. H. Sui, X. Li, C. H. Ho, K. M. Lau

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

The effects of radiative cooling on the tropical convective response to sea surface temperature (SST) are investigated using a series of two-dimensional large domain cloud-resolving simulations. The experiments are designed with an imposed warm-pool SST at 2.5°C, 3.5°C, and 4.5°C higher than the cold-pool SST, which is specified at 26°C. The area ratio of the model cold pool to the warm pool is set to 2.2. It is observed that the warm-pool convection intensifies when the SST difference between the warm and cold pools increases from 2.5°C to 3.5°C, whereas the warm-pool convection strength does not significantly change when the SST difference increases from 3.5°C to 4.5°C, which is qualitatively consistent with the observations. The analysis of the atmospheric heat budgets shows that the cap on the development of convection over the warm pool is due to a radiative cooling over the cold pool that has a small variability. The cloud-resolving model simulations indicate that the area of the radiatively driven subsidence expands to produce an enhanced mass exchange between the warm and cold pools in response to the enhanced SST difference between them. This acts against the tendency for increased clouds by the enhanced upward mass flux in response to the warmer SST over the warm pool; significantly stronger SST differences between the two pools (e.g., ≥3.5°C in the cloud-resolving model) do not proportionally produce further intensive convection activities over the warm pool.

Original languageEnglish
Article numberD08116
JournalJournal of Geophysical Research C: Oceans
Volume113
Issue number8
DOIs
StatePublished - 27 Apr 2008

Fingerprint

Dive into the research topics of 'Effects of radiative cooling on the tropical convective response to sea surface temperature: 2-D large domain cumulus ensemble simulations'. Together they form a unique fingerprint.

Cite this