利用自動化雨滴譜儀進行北台灣雨滴粒徑分布量測及降雨動能推估之研究

Abstract

台灣目前所使用之降雨動能推估公式並非利用台灣本土降雨資料所建立。降雨動能有其地域性；因此，各個地區皆應建立本土化之降雨動能公式。本研究主要目的為利用自動化雨滴譜儀(Disdrometer)進行北台灣雨滴粒徑分布的量測，並利用量測所得資料建立本土化之降雨動能推估公式。 本研究利用於台灣北部地區中央、霞雲、翡翠等三處自動化雨滴譜儀測站自2012年至2016年6月所量測之降雨量及雨滴粒徑分布資料應用於降雨動能推估之研究。本研究使用3種方法進行降雨動能之推估，第一種為傳統之方法，即單位體積動能(KEmm)法，其降雨動能與降雨強度關係式之相關係數(R2)約為0.6；第二種為較新穎的方法，即單位時間動能(KEtime)法，其降雨動能與降雨強度關係式之R2可達到0.97以上；第三種為利用特性因子如雨滴粒徑分布(Drop Size Distribution)、降雨型態及降雨強度等，將降雨事件樣本進行分組並建立單位體積動能(KEmm)推估公式。結果顯示利用DSD進行分組之結果甚佳，其分組後之KEmm對降雨強度I為一定值，而降雨型態及降雨強度的分組結果則不理想。 本研究使用台灣本土資料建立數種降雨動能推估公式，其中KEtime推估公式及依DSD將KEmm進行分組所建立之推估公式兩者有較佳之表現。因此，本研究所建立之降雨動能推估公式可以用作未來降雨動能計算及降雨沖蝕指數推估之參考。

The equation for evaluating rainfall kinetic energy currently used in Taiwan is not derived from the local rainfall data. Because rainfall kinetic energy is quite related to the region, it is necessary to develop an equation of rainfall kinetic energy for each area. The main purposes of this study are to measure the rain drop size distribution in northern Taiwan using disdrometer and to establish the equation for evaluating rainfall kinetic energy using the measured data. In this study, the data of rainfall and rain drop size distribution collected at three observation stations, namely, National Central University station, Xiayun station and Feitsui station, using disdrometer from 2012 to the Jun. of 2016 were used to evaluate the rainfall kinetic energy. To accomplish this, three methods for evaluating rainfall kinetic energy were applied. The first is the method of energy per unit volume (KEmm ), a conventional method, and its coefficient of determination( R2) between rainfall kinetic energy and rainfall intensity is approximately 0.6. The second is the method of energy per unit time (KEtime ), a relatively newly developed method, and the coefficient of determination( R2) between rainfall kinetic energy and rainfall intensity is greater than 0.97. The third is the method using attribute characteristics such as drop size distribution, precipitation types and the rainfall intensity to categorize the rainfall events and develop the KE-I relationship. While the rainfall is categorized with the drop size distribution (DSD), the KEmm remains constant over various rainfall intensities, and the results are very good. However, while the precipitation types and the rainfall intensity are used for categorizing, the results are not satisfactory. In this study, several methods are used to develop the equations for evaluating rainfall kinetic energy using the local rainfall data. Among these, the results of the KEtime method and the third method using the drop size distribution (DSD) to categorize the rainfall are much better. Accordingly, the findings in this study are expected to be useful for evaluating rainfall kinetic energy and rainfall erosion index in the future.