台灣土地利用型態對於局部環流與降雨模擬之影響

許志禎(Chih-Chen Hsu)

指導教授 ： 林沛練

國立中央大學/地球科學學院/大氣物理研究所/碩士(2008年)

摘要

地表過程不僅牽涉地表與大氣之間的交互作用，並且可以影響天氣與區域氣候的特性。在模式之中的地表過程要能發揮正確的作用，真實的土地利用型態分佈以及良好的參數化是決定性的因子。隨著台灣地區的經濟發展，MM5所使用美國地質調查局(USGS)所提供的舊版土地利用型態已經太過於老舊，而無法正確反映出台灣真實的土地利用情形。MM5更新到3.5版之後，也更新了新的土地利用型態資料，其中台灣全島的資料與舊版有相當的不同。此外，台灣中鼎公司(CTCI)利用內政部空照圖，將其數位化成台灣的高解析度土地利用資料。三種土地利用資料(USGS-OLD、USGS-NEW和CTCI)差異最大的地方為都市區的分布範圍以及濕作物、旱作地、山區混合林的分布範圍。而不同的土地利用型態皆對應不同的土壤參數，例如：土壤濕度、反照率、粗糙長度等。本論文延續曹(2007)的研究結果，針對冬季的個案利用MM5模式進行模擬比較，討論不同土地利用型態對局部環流以及氣象場模擬的影響。
在東北季風盛行的天氣型態之下，三種土地利用型態對臺灣北部與中部地區氣象場的模擬結果差異並不明顯。反之更新過後的USGS-NEW以及CTCI兩種土地利用型態在南部皆模擬出較高的溫度，USGS-NEW地表溫度最高，因此土壤中有較多水氣被蒸散到大氣之中，加上向岸風的向內傳送，所以在山區有較高的空氣塊相當位溫； CTCI的土地利用型態是以灌木與濕作物的型態分布較廣，土壤可用濕度低，地表溫度高因此有較多的能量以可感熱通量呈現，空氣塊加熱明顯而上升運動較為強烈。三種土地利用型態模擬結果與觀測資料進行比對後可以發現，冬季溫度的模擬結果都有低估的現象。不過更新過後的USGS-NEW和CTCI模擬結果因為模擬的溫度較高而較接近於觀測值。
土地利用型態對於降雨模擬影響的結果顯示，冬季個案因為降雨位置主要都位於海上，因此土地利用型態資料的更新與否，對降雨模擬並沒有明顯的影響。反之，夏季午後對流降雨個案模擬結果顯示，不同的土地利用型態資料對模擬的降雨位置與分布範圍造成影響。USGS-OLD和CTCI之土地利用型態以粗糙長度設定較短的農作地為主，模擬結果風速較強，因此其對流降雨位置偏北；而USGS-NEW都會區分布較多，粗糙長度設定較長，風速模擬較小，降雨位置偏南。而地表狀態偏乾且上升運動較為明顯的CTCI模擬降雨較早開始，但較早結束；USGS-NEW則因為地表溫度較高，土壤較多水氣蒸散到近地表之空氣中，因此降雨範圍較為廣泛；USGS-OLD則是因為加熱效應較不明顯，故降雨現象較晚發生，並且因為地表狀態較為潮濕，降雨時間持續較長。

關鍵字

土地利用 ； 局部環流 ； 降雨

並列摘要

The land surface processes not only involve the interaction among the earth''s surface and atmosphere but also influence the characteristics of weather and regional climate. Therefore, if the course of the land surface processes in the way can play a correct role, the true land use type distribution and the good parameter is decisive factors. With the economic development of Taiwan, the U. S. Geological Survey (USGS) land use data used nowadays in the Fifth-generation Mesoscale Model (MM5) is out of date and unable to reflect the real land use condition correctly. After the upgraded 3.5 version of MM5, the land use type data also had been renewed, there is quite difference land use distribution compared with the old edition on Taiwan. In addition, The China Technical Consultants Inc. (CTCI) land use data was digitized with the aerial photograph from Taiwan Ministry of the Interior (MOI). The apparent difference in these three kinds of land use data is the range and location of urban area, cropland and forest. And different land use type is corresponding to different soil parameters, for example: moisture, albedo, roughness length, etc. This research extends Cao (2007), to simulate and compare to the case of winter in MM5, and discuss the land use type impact on local circulation and meteorological field simulation.
Under the northeasterly wind, three land uses have not much difference between with each other in the result of simulation in north and middle part of Taiwan. However, the simulations with USGS-NEW and CTCI all imitate higher temperature in the south of Taiwan. There is stronger heating effect in USGS-NEW, so there are more water vapor in the soil can evaporate into the atmosphere. Because of the transfer by onshore wind, air has higher equivalent potential temperature in mountain area. In other way, due to relatively widely distribution with the type of the shrub and irrigated cropland in the land use type of CTCI, the soil has more energy is transported in sensible heat flux have low humidity. Therefore, air parcels heating and upward motions are much obviously. Compared the simulation result with observe data can find, there is clearly underestimate in the simulation of temperature in winter, but the USGS-NEW and CTCI simulation results are relatively near observed value.
In winter case, because of precipitation mainly fell area in the near coast, there is no obvious influence of land use change to rainfall simulation in winter time. On the contrary, the simulation result of the convective precipitation in summer case show that different land use type causes obviously influence on the distributed and intensity of the rainfall simulation. The primary land use type in USGS-OLD and CTCI are irrigated cropland and shrub, which the setting of roughness length is shorter then others. Consequently, the simulation in wind speed is relatively strong, and the convection rainfall system occurred in slight northern area. The primary land use type in USGS-New is urban, which setting in roughness length is longer, so the simulated wind speed is relatively small. Because of that, the rainfall area leaned towards the south. Because of the surface state drier and stronger upward motion obviously, CTCI simulation rainfall began earlier, and ended earlier also; USGS-NEW heat comparatively obvious, more water vapor can evaporate into the air of the surface layer, so the precipitation distribution area is extensively. It is not more obvious that the heating effect in USGS-OLD, so the rainfall phenomenon takes place later, but the soil available moisture of the surface is higher, so the precipitation time sustain longer.

並列關鍵字

land-use ； local circulation ； precipitation

參考文獻

曹嘉宏，2007：台灣土地利用型態對MM5模擬局部環流之影響。國立中央大學碩士論文，96頁。

戴俐卉，洪景山，莊秉潔，蔡徵霖，倪佩貞，2006：WRF模式中台灣地區土地利用型態織更新與個案研究。95年天氣分析與預報研討會論文彙編，中央氣象局，P.2-43。

Alapaty K, J.E. Pleim, S. Raman, D.S. Niyogi and D.W. Byun, 1997: Simulation of atmospheric boundary layer processes using local- and nonlocal-closure schemes. J. Appl. Meteor., 36 , 214–233.

Anthes, R. A., and T. T. Warner, 1978: Development of hydrodynamic modelssuitable for air pollution and other mesometeorological studies. Mon. Wea. Rev., 106, 1045-1078.

Betts, A. K., J. H. Ball, A. C. M. Beljaars, M.J. Miller and P. A. Vitebo, 1996: The land surface-atmosphere interaction :A review based on observational and global ,modeling perspective. J. Geophys. Res., 101,7209-7225.