Advances in Earth Science ›› 2017, Vol. 32 ›› Issue (4): 396-408. doi: 10.11867/j. issn. 1001-8166.2017.04.0396

• Orginal Article • Previous Articles     Next Articles

Decadal Prediction Skill of the Global Sea Surface Temperature in the BCC_CSM1.1 Climate Model

Zhenyu Han 1( ), Bo Wu 2, Xiaoge Xin 1   

  1. 1. National Climate Center, China Meteorological Administration, Beijing 100081, China
    2. State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China
  • Received:2016-10-17 Revised:2016-12-29 Online:2017-04-20 Published:2017-04-20
  • About author:

    First author:Han Zhenyu(1987-),male,Yuanping City,Shanxi Province,Associate professor. Research areas include climate modelling, East Asian climate change and climate variability.E-mail:hanzy@cma.gov.cn

  • Supported by:
    Foundation item:Project supported by the R&D Special Fund for Public Welfare Industry (Meteorology) “Development and research of ensemble decadal climate prediction system based on global climate models FGOALS-s, CAMS and CESM”(No.GYHY201506012);The LCS Open Funds for Young Scholars (2016) “Study on near-term climate prediction in China by BCC_CSM1.1 model”

Zhenyu Han, Bo Wu, Xiaoge Xin. Decadal Prediction Skill of the Global Sea Surface Temperature in the BCC_CSM1.1 Climate Model[J]. Advances in Earth Science, 2017, 32(4): 396-408.

This study assesses retrospective decadal prediction skill of Sea Surface Temperature (SST) variability in initialized climate prediction experiments (INT) with the Beijing Climate Center Climate System Model (BCC_CSM1.1). Ensemble forecasts were evaluated using observations, and compared to an ensemble of uninitialized simulations (NoINT). The results show as follows: ①The warming trend of global mean SST simulated by the INT runs is closer to the observation than that in the NoINT runs.②The INT runs show high SST prediction skills over broad regions of tropical Atlantic, western tropical Pacific and tropical Indian Oceans. ③ In the North Pacific and the east-central tropical Pacific Ocean, the prediction skills are very weak, and there are few improvements coming from the initialization in the INT runs. ④ In the southern Indian Ocean, the prediction skills of the INT runs are significantly larger than that of the NoINT runs, with the maximum skill at the 3~6 and 4~7 years lead time. The above-mentioned conclusions are similar to the results of other climate models. However, the prediction skill in the North Atlantic Ocean is much lower than that of other models, especially in the subpolar region. The low skills in the Atlantic Ocean may be attributed to the misrepresentation of the lead-lag relationship between the Atlantic meridional heat transport and the SST in the BCC_CSM1.1.

No related articles found!
Viewed
Full text


Abstract