以臺灣北部地區之時雨量資料推估降雨沖蝕指數

近年來山坡地不當開發使得土壤沖蝕現象日益惡化，為有效永續利用水土資源，土壤沖蝕量評估為一重要課題。目前以通用土壤流失公式(USLE)最為常見，公式中的降雨沖蝕指數計算方式為降雨動能與最大30分鐘降雨強度之乘積，但在許多地區的歷史雨量記錄資料中，缺乏計算降雨沖蝕指數所需的短時間間距雨量資料，因此如何以較容易取得之時雨量資料進行估算歷年降雨沖蝕指數，為本研究主要探討的重點。本研究蒐集臺灣北部地區(臺北市、新北市、桃園市、新竹縣及苗栗縣)內2002~2016年共61個中央氣象局雨量站之10分鐘雨量資料，根據Wischmeier and Smith(1958)之定義篩選出單場有效降雨事件共29,659場，分別以10分鐘雨量資料及累計而成的時雨量資料估算降雨動能E10與E60及最大降雨強度I30max與I60max，再求出單場降雨沖蝕指數R30j與R60j及年降雨沖蝕指數R30y與R60y，最後進行迴歸分析並求得不同時間間距之轉換係數。分析結果顯示，臺灣北部地區之降雨動能、最大降雨強度、單場降雨沖蝕指數及年降雨沖蝕指數的轉換係數(αE、αI、αRj及αRy)分別為1.057、1.504、1.350、及1.497。此外，雖然轉換係數皆隨著高程增加而呈現下降之趨勢，但變動幅度皆不大，建議未來若能以相鄰地區之降雨特性較相近的雨量站進行分析，應有助於使轉換係數與高程之關係更加明顯。

關鍵字

通用土壤流失公式；降雨沖蝕指數；轉換係數

並列摘要

In recent years, due to the improper development of slopeland, the soil erosion is increasingly serious. In order to effectively use water and soil resources perpetually, assessment of soil erosion is an important issue. The Universal Soil Loss Equation (USLE) is presently one of the most widely models to evaluate soil erosion. The Rainfall erosivity index (R30) in the USLE is calculated from the total kinetic energy and the maximum 30-min rainfall intensity of a storm. R30 values are calculated from rainfall information obtained from continuous recording. However, detailed chart-recorded rainfall data are not readily available, whereas hourly rainfall is available in many places. This study sets up in a simple method for estimating the rainfall erosivity index (R30) by using the value of R60 calculated from the rainfall kinetic energy (E60) and maximum intensity (I60max) measured at multiple rainfall stations. The data set consists of 29,659 storm events monitored by 61 rainfall stations located in northern Taiwan. It is shown that the average conversion factors (αE, αI, αRj and αRy) of the rainfall kinetic energy (E), the rainfall intensity (I), the rainfall erosivity index (Rj) and annual rainfall erosivity index (Ry) (i.e., the ratios of the 10-min to 60-min interval values of the corresponding factors) are respectively 1.057, 1.504, 1.350 and 1.497. The variations of average conversion factors (αE, αI, αRj and αRy) have a slight decrease with elevation. This study suggests that if the selections of rainfall stations have similar rainfall characteristics and locate around adjacent areas, the variation trend between average conversion factors and elevation may be more noticeably.