台灣地區受到特殊的水文與地文條件影響,每當颱風或暴雨事件發生時,中下游平原地區容易造成淹水,其中又以都市區淹水造成的人民生命財產損失最為嚴重,然而都會地區建築物林立,不透水面積表層增加,且蓄水面積下降,且大量密集的建築物改變了原本土地的水流狀況,產生了束縮水流及增加額外形狀阻力的現象,造成了淹水現象更為嚴重。 本研究主要目的在透過大型水槽模型實驗,找出建蔽率與曼寧n值的相關性,並分為表面阻抗造成的曼寧n值變化與建物形狀阻抗造成的曼寧n值變化兩者來討論,因此則能在一般計算網格當中加入建蔽率α,反應出含建物淹水時的情形,之後進一步計算當淹水深度達一定高度時,水流由建物外部流入內部體積轉換的情形,以反應建物於淹水時的實際現象。 本文將模式應用於台中市的模擬區,進行實際案例模擬,以2008年卡玫基颱風雨量資料,分別比較於淹水模式中依建蔽率調整地表粗糙度前後淹水範圍及淹水深度的變化,結果顯示未調整地表粗糙度值前會嚴重忽略建物阻水效應,因此當我們以二維漫地流模式進行都會地區地表淹水模擬時,於淹水模式中加入建物因子並依照建物占網格大小即建蔽率α調整表粗糙度值,可使淹水情形較符合實際狀況。
Due to the special hydrographic and physiographic conditions in Taiwan, whenever a typhoon or rainstorm event comes, flooding is likely to occur in the middle and lower reaches of the plains. Noteworthily, the loss of lives and property caused by flooding will be the most considerable one in the metropolitan area. However, the modern tall building stands in great numbers in metropolitan areas, which leads to the increase of impervious area as well as the decline of water storage area. Furthermore, a large number of intensive building changes the original land flow conditions, resulting in a beam shrinking flow and the additional form drag phenomenon which makes the flooding phenomenon more serious. The main purpose of this research is to find out the correlation of building coverage and the Manning ’s n through flume model experiment. To probe into this issue, the Manning ’s n changes is further divided into those caused by surface impedance and those caused by the building impedance. Thus, the building coverage can be added to the general computing grid and reflects the flooding situation with building. Afterwards, a further calculation can be carried out to have in hand the volume conversion situations when flooding depth reaches a certain height and when the water flows into the building so as to reflect the actual situation when the building is flooded. The model will be applied to the simulation district of Taichung City for an actual case simulation. Using the rainfall data of Typhoon Kalmaegi in 2008, we calibrate the surface roughness of the model in accordance with building coverage from the comparision of the flooding scope and flooding depth. The results show that we can seriously neglect the water blocking effect of the building before the adjustment of surface roughness. Therefore, it might be more pertinent to take into consideration of the building factor in the flooded mode and adjust the surface roughness according to the ratio of the building coverage in the grid, when the two-dimensional model is used to undertake the flow simulation of the metropolitan areas. In this way, we can make the flooding situation more in line with the actual situation.