氣候變遷之影響層面相當廣泛,因此為了瞭解氣候變遷對水資源之影響,本研究以牡丹水庫集水區作為研究案例,使用動力降尺度模式模擬之未來氣候資料,以水文模式模擬水庫集水區入流量,探討未來在氣候變遷情境下(A2、B2)之水資源衝擊評估。本研究首先蒐集恆春氣象站(1897-2007)之氣象資料、牡丹雨量站(1998年-2007年)之雨量資料與牡丹水庫實際觀測入流量(1998年-2007年),以Mann-Kendall法針對所使用之氣象資料進行初步趨勢分析,並選用HSPF(Hydrological Simulation Program Fortran)建立牡丹水庫集水區之水文模式,依據牡丹水庫入流量進行參數率定。未來氣候資料則使用動力降尺度模式RSM(Regional Spectral Model)模擬之,最後將資料輸入至率定完成之HSPF中以進行流量衝擊評估。經初步歷史資料檢定發現雨量與蒸發散量無明顯增加或減少,輻射量與露點溫度則有減少趨勢,溫度則有增加之趨勢,並針對雨量資料進行機率分佈檢定,結果顯示未來降雨型態是有改變的。未來氣候變遷條件下水庫入流量評估結果顯示:中期(2040年-2050年)之帄均年流量將減少18%(A2情境)、17%(B2情境),帄均豐水期之流量減少,帄均枯水期之流量增加;長期(2090年-2100年)之帄均年流量減少3%(A2情境)、7.9%(B2情境),帄均豐水期之流量減少,帄均枯水期之流量增加。
Influence of climate change is expansive in the world. In order to investigate the impacts of climate change on water resources, taking Muden reservoir for example, this study applies dynamic downscaling method to obtain the future meteorologic data, and then simulates the inflow of Muden reservoir by hydrologic model to further assess the impacts of water resources in the scenario of climate change (A2.B2). First, we collect meteorologic data from weather stations in Hengchuen (1897-2007) and Mudan (1998-2007). In turn, we use Mann-Kendall method to roughly assess the trend of climate, and thus choose HSPF (Hydrological Simulation Program Fortran) to set up hydrologic model whose parameters are adjusted in term of observational inflow. The future climate data is simulated by downscaling model, RSM (Regional Spectral Model), eventually taken into calibrated HSPF to assess the impact of flow. From the results of Mann-Kendall test, we can find that precipitation and evaporation are not apparent bias; radiation and dew point are negative bias; temperature is positive bias. In addition we take a test for rain, and find that from of rain will change in the future. Under the climate change in the future, the results of assessment of inflow of Muden reservoir indicate: 1.the annual mean inflow will decrease 18% (A2) and 17% (B2). 2.mean inflow of rainy days will decrease and dry days will increase in 2040-2050 year. 3.annual mean inflow will decrease 3% (A2) and 7.9% (B2). 4.mean inflow of rainy days will decrease and dry days will increase in 2090-2100 year.