熱帶降水趨勢的時空變動特徵分析

An Analysis of the Spatiotemporal Features of Precipitation Trends Over Tropical Oceans

孫既仁(Ji-Jen Sun)

指導教授 ： 隋中興

國立臺灣大學/理學院/大氣科學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.02031

摘要

本研究使用1988到2009年間SSM/I日降水以及GPCP五日平均降水資料，分析熱帶海洋（30oS-30oN）降水長期的趨勢變化。兩組資料顯示的降水趨勢特性雖有些許差異，但其重要結論一致。在熱帶海洋範圍內將所有降水資料依強度分級，累積各級雨量、頻率以及強度的趨勢變化，顯示強降雨（SSM/I為70%以上，GPCP則為95%以上的降雨強度）的降雨頻率及降雨量皆在增強，而弱降雨的降雨頻率與降雨量皆在減少。降雨強度(降雨量除上降雨頻率)的變化除了最微弱的降水之外皆有增強的趨勢（~0.05%DY）。空間分布上，計算熱帶海洋各網格降水時間上趨勢變化的特性，顯示降雨量與降雨頻率趨勢的分布特性一致，其二者所決定的降雨強度在空間上則幾乎都是增強的。從邊界層水氣收支的觀點診斷降水，計算邊界層水氣輻合與洋面蒸發的趨勢，顯示降水趨勢在空間上的分布特性，與邊界層水氣輻合趨勢分布一致；蒸發趨勢在區域上的影響可能較微弱，但整個熱帶海面上的蒸發皆為增加，且南半球的增加量較北半球大。
上述分析的降水趨勢特性，包含了人為暖化的作用，以及低頻氣候震盪的影響。一些氣候觀測分析的結果顯示，近20餘年的海溫趨勢與低頻氣候震盪的訊號關係相當密切，因此合理的預期降雨量的趨勢會受到低頻氣候震盪的影響。本研究使用部分最小平方法迴歸分析(Partial Least Squared Regression, 簡稱PLSRA)，將海溫場中受氣候震盪影響的降水場成分從原始降水資料中扣除，求取較不受氣候變異影響的降水，進而分析其趨勢。分析結果顯示，原始降水資料中熱帶海面平均降水具有很大的年際變異與增強趨勢（0.5%DY），而扣除氣候震盪影響後的平均降水，則顯示較小的年際變異與降雨量減少的趨勢變化(-0.7%DY)。從邊界層平均水氣源趨勢的變化來看，總降水量的趨勢變化特性主要是受到熱帶洋面總蒸發量變化影響的結果，此結果與觀測分析及模式模擬的結果較一致。即近二三十年來信風與降雨增強反映年代際的震盪，而剩餘平均降水減弱的趨勢，尚待進一步分析，確定相關的氣候水循環過程的改變。

關鍵字

熱帶海洋降水趨勢 ； 分級降雨 ； 年際變異 ； 部分最小平分法迴歸分析 ； 剩餘降水

並列摘要

We analyze long-term trend changes of tropical rainfall based on satellite products by SSM/I daily data and GPCP pentad data from 1988 to 2009. The important features derived from the two data sets are consistent although some differences do exist between the two data sets. All precipitation data in the tropical oceans (30oS-30oN) is categorized in percentile bins by intensity to estimate the trends of precipitation amount, frequency and intensity in each bin. Both amount and frequency of heavy rainfall (top 30% precipitation in SSM/I, and top 5% in GPCP) show increasing trends, and the opposite trend changes are found in weak rainfall. Precipitation intensity, however, shows an overall increasing trend (0.05%DY) in all percentile bins except the weakest bin. We further examine trends in total precipitation amount and frequency at each grid points within tropical oceans. The trend patterns of precipitation amount and frequency show similar regional distribution. The corresponding trend pattern of rainfall Intensity shows nearly uniform increasing trends over all grids except few small regions. Considering trends of moisture sources for precipitation in the boundary layer (moisture convergence and surface evaporation), local precipitation trends are dominated by moisture convergence. Evaporation shows increasing trends over tropical oceans with stronger magnitudes over southern hemisphere than that over northern hemisphere.
The observed trend changes from the above analysis consist of responses to anthropogenic warming and natural climate oscillations like ENSO and decadal to multi-decadal oscillations. Existing observational analyses show that SST in the recent decades since 1980s contain dominant multidecadal variability on top of global warming trends. Consequently, the low-frequency climate oscillations are anticipated to also influence the trends in total precipitation trend in our analysis. In this study, a partial least squared regression analysis (PLSRA) is applied to SST and rainfall data to suppress the natural oscillations in precipitation data. This approach is shown to process precipitation data by PLSRA to make the estimated precipitation trends less susceptible to effects of natural climate variability. Our analysis shows that the raw precipitation has large interannual to decadal variability with an increasing trend (0.5%DY). The application of PLSRA reduces the variability and the trend of adjusted rainfall becomes negative (-0.73%DY). A similar analysis of moisture budget in the boundary layer shows that the trend of evaporation is most dominant. The increasing rainfall trend is consistent with some recent observational studies that show strengthening trade winds and global monsoons associated with multidecadal climate oscillations. The decreasing rainfall trend after the natural oscillations are suppressed from the rainfall data awaits further investigations.

並列關鍵字

tropical ocean precipitation trends ； binned rainfall ； inter-annual variability ； Partial least squared regression analysis

參考文獻

Allen, M. R., W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature , 419 (6903), 224–232.

_____ and B. J. Soden, 2008: Atmospheric warming and the amplification of precipitation extremes. Science, 321, 1481–4.

_____, _____, V. O. John, W. Ingram, and P. Good, 2010: Current changes in tropical precipitation. Environ. Res. Lett., 5, 025205.

Bosilovich, Michael G., S. D. Schubert, G. K. Walker, 2005: Global Changes of the Water Cycle Intensity. J. Climate, 18, 1591–1608.

Chou, C., M.-H. Lo, 2007: Asymmetric Responses of Tropical Precipitation during ENSO. J. Climate, 20, 3411–3433.