透過您的圖書館登入
IP:3.144.113.197
  • 學位論文

海氣耦合模式東亞季節預報評估

East Asian Monsoon Metrics for Model Evaluation

指導教授 : 許晃雄

摘要


到目前為止,海氣耦合模式已被廣泛的使用在季節預報上。相較於單純的大氣模式,同時考慮大氣和海洋的耦合模式,在短期的天氣、季節預報上較有可信度。而其中系集模式 (multi-model ensemble (MME)) 更能模擬出與實際情況相關係數較高的情形。但是,這些用於季節預報的海氣耦合模式,若用於更長時間氣候環流場、時間序列的年際變化甚至是主結構的模擬,仍有很大需要改進的空間。 在前人的研究中,已有許多對於海氣耦合模式的模擬能力進行探討,像是Wang et al. (2007) 利用Season-reliant Empirical Orthogonal Function (S-EOF) analysis method 來探討DEMETER和APCC/CliPAS中的MME在亞澳季風降雨年際變化上主要的結構和實際的差異。在本論文中,我們將利用DEMETER和NCEP/CFS中的海氣耦合模式驗證在東亞季風中的一些主要年際變化結構。 首先根據基本氣候環流場、太平洋和印度洋上兩米溫度、影響東亞季風重要大氣參數的年際變化結構、經驗正交函數分析的模擬,可發現模式們明顯低估東亞夏季季風平均強度與變異量,其中季風槽的模擬偏差最大。變異量的模擬以海上較佳,陸地上的年際變異遠低於觀測值。這可能反映出海氣耦合模式模擬海氣交互作用的能力較佳,但是對於氣陸交互作用則掌握能力相當不足。在印度洋和太平洋海溫的經驗正交函數的分析中,時間序列的趨勢大致相同,而觀測場的前兩個主要結構分別代表兩個不同的結構,但模式都在第一主結構的解釋變異度上都有高估的現象,顯示模式對於海溫影響大氣的掌握很好。 接著,由基本環流場可發現中高緯度和熱帶地區的表現有明顯差異,因此試圖進一步分析海氣耦合模式在不同區域的模擬情形,像是海洋大陸模擬的差異、高 (20°N-90°N)、低 (20°S-20°N)緯度變異度的模擬情形,並藉由多變數經驗正交函數 (Wang et al., 2008) 和奇異值分解法對各種不同的區域進行分析。利用多種統計方法規列出一系列量尺。 由以上分析得知:1. 海洋模擬的較陸塊上佳、2. 模式對於熱帶地區的掌握能力較中高緯度好、3. 模式對於年際變異有明顯低估的現象、4. 主結構的模擬仍有過於單一化且過於強調聖嬰現象造成的影響。因此,目前常用的季節預報海氣耦合模式在年際尺度上的模擬仍然有進步的空間。

關鍵字

量尺 模式評估

並列摘要


The coupled atmosphere-ocean seasonal prediction models have been extensively used nowadays. Compared with the stand-alone atmospheric models or ocean models, the coupled atmosphere and ocean models have higher skills in seasonal prediction, and especially their multi-model ensemble (MME), which more accurately make the 1-month lead seasonal forecast. On the other hand, the coupled atmosphere-ocean models still have deficiencies in reproducing realistic climatology. The year-to-year variation of the Asian-Australia monsoon exhibits enormous regional differences and depends strongly on the phase of the annual cycle. This poses a great challenge for the coupled atmosphere-ocean model in making realistic seasonal simulation. This problem was clearly demonstrated in Wang et al. (2007), which analyzed the seasonal hindcast by the DEMETER and APCC/CliPAS. This study is aimed to evaluate the capability of the DEMETER and NCEP/CFS in hindcasting the interannual variability of the East Asian monsoon for the 1980-2003 period. A preliminary analysis of the 2-meter temperature indicates that the hindcast skill of these coupled models is significantly higher in the ocean than over the continent. Especially, the simulated interannual variance in continent is significantly underestimated. Compared with the simulation skill between high latitude area (20N~90N) and Tropical ocean area (20S~20N), most of the models are able to simulate the variance in low latitude but high latitude. An EOF and SVD analysis of the East Asian monsoon indicates that while the models are able to simulate the leading circulation/precipitation coupled mode, they tend to put too much weight on the leading mode by exaggerating the fractional variance explained by the leading mode. It suggests that the models tend to produce lower variability than the observed and is likely to misrepresent the variability that is not associated with the leading mode. In our study, we found that the AOGCMs tend to show the feature of ENSO and lead by Tropical Ocean but ignore the variance in high latitude and land. This result reveals that the air-sea interaction is reasonably simulated, but the dynamical processes over land and the possible effect of air-land interaction are poorly handled in the models. Overall, the AOGCMs may overestimate the impact of Tropical Ocean, underestimate the influence of continent, and high latitude. Further analyses will focus on the simulation skill in high latitude. Besides, we will evaluate the performance of East Asian monsoon in IPCC models and define a set metrics for to evaluate the models’ capability in simulating the interannual variability of the East Asian monsoon. In addition to identify the deficiencies of the models for further improvement.

並列關鍵字

metrics model evaluation

參考文獻


Gleckler, P. J., K. E. Taylor, and C. Doutriaux, 2008: Performance metrics for climate models. J. Geophys. Res., 113, D06104, doi:10.1029/2007JD008972.
Kumar, K. K., M. Hoerling, and B. Rajagopalan, 2005: Advancing Indian monsoon rainfall predictions. Geophys. Res. Lett., 32, L08704, doi:10.1029/2004GL021979.
Liu, J., B. Wang, and J. Yang, 2008: Forced and internal modes of variability of the East Asian summer monsoon. Clim. Past, 4, 225-233.
Sperber, K. R., and T. N. Palmer, 1996: Interannual tropical, rainfall variability in general circulation model simulations associated with the Atmospheric Model Intercomparison Project. J. Climate, 9, 2727–2750.
Taylor, K. E., 2001: Summarizing multiple aspects of model performance in a single diagram, J. Geophys. Res., 106, 7183– 7192.

延伸閱讀