What Causes the Tropical Atlantic SST Bias in CCSM3?

点击翻译按钮获取中文摘要

Current atmosphere-ocean coupled climate models cannot reproduce the annual cycle of tropical Atlantic sea surface temperatures (SSTs) and thus do not have a skill to simulate or to forecast the tropical Atlantic climate variability. This project seeks to analyze the mixed layer heat budget of the National Center for Atmospheric Resesarch (NCAR) community climate system model version 3 (CCSM3) and version 3.5 to identify processes and/or parameterizations that contribute to this SST bias. The investigators will examine the possible role of local ocean dynamics in three key regions (southeastern Atlantic Ocean, eastern and central equatorial Atlantic Ocean, and tropical North Atlantic Ocean), and determine and quantify the external influences, with the goal of understanding if and how local atmosphere-ocean feedback processes amplify the remote signals.Specifically the investigators will perform some CCSM3(3.5) experiments by including (a) dynamic atmosphere with fixed SST run; (b) dynamic ocean with fixed surface flux run; (c) fully coupled atmosphere-ocean run; (d) coupled atmosphere-ocean runs with the regional flux-adjustment. Comparison of these model runs along with the observational data will provide some useful insights on the inherent biases in the ocean and atmospheric components of CCSM, and if and how such biases are transmitted to remote regions and/or locally amplified through the coupling processes. Through testing hypotheses with CCSM3(3.5), this study will increase and refine our understanding of the processes that affect the equatorial Atlantic SST and their broader involvement in tropical Atlantic climate variability. The outcome of the work will be merged into climate models, potentially improving understanding of tropical Atlantic variability and enhancing seasonal-to-decadal climate prediction skill, which will benefit society surrounding the tropical Atlantic.