大氣溫室氣體加倍對臺灣區域氣候和水資源衝擊評析

Abstract

本文藉由數値模擬探討溫室效應氣體倍增對臺灣區域氣候和水資源之影響。首先進行使用NCAR-CCM3和NCEP-RSM-96V(CCM3/RSM96Ⅵ模擬系統作不同模式解析度，目前二氧化碳濃度和二氧化碳濃度倍增情境下，臺灣區域氣候模擬之實驗。完成了三種不同解析度RSM的模擬。其為280km × 280km (RSMO和CCM3解析度相同)，50km X 50km(RSM1)和15km X 15km(RSM2)。作RSMO擬的目的是作為區域氣候模式的比較基準，以移除CCM3和RSMO之模式物理差異效應。對各別10個冬季和個夏季，目前二氧化碳濃度和二氧化碳倍增情境下RSM慔擬輸出統計來探討平均的衝擊。結果顯小CCM3和RSMO間仍有明顯差異，顯示不同模式物理的影響。臺灣區域降水季節和地理分佈的特徵可以RSM巢結於CCM3的二氧化碳等溫室效應氣體加倍造成全球尺度環流變遷的模擬中被突顯出來而RSM2的模擬最接近觀測。但是降水強度，特別是冬季可能被某些程度的高估。而模擬的降水頻率特性在夏季較為合理。RSMI和RSM2模擬顯示臺灣東北部在二氧化碳加倍後，冬季降水顯著增加，伴隨南部海岸背風面，地面溫度顯著升高，而全省冬夏雨日頻率都增加。其次以不同氣候模式模擬輸出輸入GWLF水文模式，模擬臺灣北中南東四個河川集水區的河川逕流量和地區地下水補注量，探討全球變遷對臺灣水資源供給面的影響。對氣候橂式模擬輸出的區域水文參變數，結合日氣象產出橂式產生日氣象參變數檔。其中包括降水量，平均氣溫，條件和無條件雨日機率等。並將其統計結果應用於水文模式預測河水流量和區域地下水補注量之變遷。結果顯示RSMO預測北中東三區河川流量皆將顯著增加。南部除春季外也有顯著增加，而東部將增加最多。高解析度模式RSM2預測各區氣候參數小而一致的增加，因而預測各區河川流量將少量增加。但RSMI預南部河川流量，在春季減少。中部河川流量全年減少。地下水補注方面除了RSMI預測中部地區全年和南部地區夏秋冬季減少之外，其餘地區的地下水補注量都呈現增加趨勢。本文進一步結合氣候模式、水文模式及灌溉用水評估模式評估農業用水探討全球變遷對臺灣地區水資源需求面之衝擊。農業用水量和氣溫關係密切，RSM模式預測作物生長期前移並且縮短。農業用水量除了中南部之外，其他地區則是減少。 It is to investigate the impacts of doubled greenhouse gases on the regional climate and water resource over Taiwan through numerical model simulation. At first, the impacts on the regional climate was evaluated through applying a nested NCAR Community Climate Model (CCM3) and NCEP Regional Spectral Model (RSM-96V) system to simulate the regional climates for the current and doubled CO scenarios (1 X CO and 2 X CO , respectively). Three different resolution RSM runs had been done. They were 280kIIl X 280km (RSMO, same resolution as CCM3), 50kIllX50kll 151m, x 151m (RSM2). RSMO was run as a comparison basis to remove climate model effects between CCM3 and RSM. The output of 10 winters and 10 summers of current CO scenario runs and doubled CO scenario runs respectively were examined. It was shown that noticeable differences between CCM3 and RSMO simulations appeared, indicating the mismatched of the model physics. Seasonally and geographically dependent precipitation patterns over Taiwan area were reasonably enhanced by RSM from the imposed CCM3 simulations, while the RSM2 reproduced the precipitation pattern most close to the observation. However, the magnitude, especially during winter time, seems to be somewhat overestimated. The precipitation frequency characteristics were reasonable only for summer time simulation. RSMI and RSM2 projected statistically significant increase of winter time precipitation over northeast Taiwan accompanied by an increase in surface temperature over the lee side of western coast plain and an increase of wet day probability over Taiwan in both seasons. In order to investigate the impacts of global change on the regional water resource over Taiwan from the supplying side, a GI, X/LF hydrological model was applied to simulate stream flows and ground water refill driven by the various climate model output. Four watersheds Over Taiwan were selected to represent northern, central, southern and eastern region of Taiwan. The outputs of climate model simulations as well as the observations were used in a daily weather generating model to generate area daily hydro-meteorological parameters. They included mean surface temperature, precipitation, and unconditional and conditional wet day probability. The statistics of these parameters were estimated and applied to the hydrological model. It was projected that annual stream flows of the four areas except the southern in spring will increase based on R SMO prediction, and those in the eastern watershed might have largest increase. The RSM2 tend to predict small and reasonable change of weather parameters, which further resulted in smaller and consistent increase of stream flows. However, RSMI predicted that the stream flow of the southern decreased in sprlng and the stream flow of the central decreased all year round. It was projected that the ground water refill over the four areas will be increased except those of R SMI over the central all year round and over the southern in Summer, fall and winter. Furthermore, to investigate the global change impacts on the regional water resource over Taiwan from the demanding side, a model system including climate model, hydrological model, and irrigation water demand assessment model was applied. Irrigation-water demand was quite closely related to the air temperature. The RSM projected that the irrigation demand will be decreased except in the central Taiwan due to the plant growth season become earlier and shorten.