全球暖化下陸地水含量的季節變化與陸氣交互作用

Abstract

陸地水含量是全球水循環中的基本要素，其變化在地球氣候系統中扮演重要的角色。自2002年起，重力反演與氣候實驗衛星(GRACE)提供了一個嶄新的方法以估計水含量的變化。藉由測量地球重力的差異，GRACE衛星資料可以在全球尺度下估計該區域每個月的水含量變化。 本研究使用第五期耦合模式相互比對計畫（CMIP5）之模擬結果，估計土壤水含量隨著全球暖化的改變。結果顯示，氣候增暖將增加北半球中高緯度土壤水含量之季節變化量。冬季的降雨增多與降雪減少，提高土壤水含量在冬季的補注；春季之融雪減少則減少土壤水含量在春夏季的補注，導致乾季越乾與濕季越濕。同時，暖化導致的土壤冰融化也會影響滲透率與增加土壤水的可變動量，提高土壤水之季節變化量增加的可能性。本研究顯示土壤水之季節變化量增加與在暖化的水資源分佈不均，對於氣候變遷下的水資源管理有相當重要的意義。 此外，本研究以澳洲為例，比較GRACE衛星觀測資料與陸地水文模式的陸地水含量差異，並提出造成差異之可能機制。進一步分析陸氣交互作用之強度後，我們可以得到陸氣交互作用在時空分布的變化。在洪災事件中，陸氣耦合強度可能先上升後下降，即蒸發量隨土壤溼度增加而增加，但在洪災後期蒸發量接近潛在蒸發量，此時蒸發量與土壤濕度則無顯著關聯。本研究顯現不同的土壤溼度可能對地表能量與水氣循環造成非線性的影響。此發現將有助於對於氣候模式的模擬以及短期氣候預報。

Terrestrial water storage (TWS) is a fundamental signal in the land hydrological cycle, and its changes play a crucial role in the earth’s climate system. Since 2002, the Gravity Recovery and Climate Experiment (GRACE) has offered a new method to estimate the variability of TWS by measuring gravity changes, in which GRACE can provide for the first time the TWS globally at monthly time scales. In this study, we used simulations from the Coupled Model Intercomparison Project Phase 5 (CMIP5) archives to investigate changes in the annual range of soil water storage under global warming. Results show that future warming could lead to significant declines in snowfall, and a corresponding lack of snowmelt water recharge to the soil, which makes soil water less available during spring and summer. Conversely, more precipitation as rainfall results in higher recharge to soil water during its accumulating season. Thus, the wettest month of soil water gets wetter, and the driest month gets drier, resulting in an increase of the annual range and suggesting that stronger heterogeneity in global water distribution (changing extremes) could occur under global warming. This has implications for water management and water security under a changing climate. In addition, we compared the GRACE data with the results of TWS simulated by several land surface models over Australia, where lots of dry and endorheic basins exist with low frequencies of water mass changes. We examined several factors and mechanisms that cause the bias between models and observations. Since land surface models provide the boundary conditions for the land-atmosphere interaction in the global climate models, the mechanisms whereby water transport influences terrestrial water storage might impact the climate. Furthermore, the highly spatial and temporal variability in water storage over Australia plays an essential role in affecting the variability of land-atmosphere coupling strength. In this study, we applied an index to diagnose the impacts of variabilities in water storage on the coupling strength. Results show that the sensitivity index first increases but then decreases during the flooding in semi-arid regions, which is the temporal transition between the soil moisture-limited regime and the energy-limited regime. The high sensitivity index indicates that the evaporation follows well with soil moisture variations, while the low sensitivity index reveals weaker land-atmosphere interactions. Therefore, the results have crucial implications for land-atmosphere interactions and climate predictions.