河床質粒徑於河川縱、橫向之變化-以頭前溪與鳳山溪為例

Abstract

河床質顆粒分布狀況為河川水理分析及整治之重要影響因子之一，其中以礫石質河床材料分布狀況最為複雜，其分布狀況具有空間上之不均質性，故欲獲得合理分析與整治效果，應先了解河床質粒徑分佈於河川縱、橫向之變化。 本研究以傅志偉(2005)提出之河床質調查規劃建議，以鳳山溪為例進行調查工作，規劃39組網格法與16組體積法，主要研究河床質縱、橫及垂直向分布情況，並將結果與鍾政良(2006)於頭前溪之作之調查結果比對，並討論各試驗取樣數量之合理性。另外本研究亦加入拍照法進行調查工作，以影像處理方式將結果與網格法結果比較，討論拍照法之適用性。 研究結果顯示河床質粒徑變化於縱向以及橫向確實有一明顯趨勢，但須對相同沉積環境的相同位置加以比較才能使變化趨勢突顯，如沙洲前端或邊灘靠近河道處，並發現沙洲之顆粒分布狀況極為明顯，沙洲兩側靠河道處顆粒明顯較粗，故沙洲上取樣位置對結果有相當大之影響，故建議於橫向趨勢調查上應將沙洲與邊灘分開研究。體積法結果發現以同一斷面之體積法分層結果可了解此斷面之沉積行為與洪水事件之影響範圍。而拍照法初步結果建議可以4-12組照片進行調查，並可與網格法結果比對。

Information on bed-material particle size is needed for a variety of purposes, which include computations of flow hydraulics, bedload transport rates transport capacity and flow competence to analyze, and predicting stream behavior and stream processes. Due to abrasion, sorting, weathering, landform types, and geological conditions, the grain size distribution along a river is non-uniform. In order to determine the grain size distribution efficiently and reasonably, the variation of the distribution in the longitudinal and transverse direction of a river needs to be understood. The river classification system propose by Fu(2005) was adopted in this thesis. The Fengshan River was selected as an example. Based on the classification system,the river was divided into 3 substreams and 39 sets of grid sampling and 16 sets of volumetric sampling were conducted. The study aims to study the variations of gravel-bed river material in longitudinal,transverse and vertical directions. The results was also compared with the results of the Toucian River studied by Zhong(2006). The study also applied sampling method to investigate river bed material for comparing with grid sampling. Base on the sampling results, the grain size distribution of gravel-bed river material has shown obvious trend in longitudinal and transverse directions. The study also found that the grain size distribution at the riffle is different with at the bar for a river. To investigate the variation of grain size distribution in the tranverse direction, the data analysis for the bar should be separated with the bar. Based on volumetric-sampling results along a transverse direction, the past flood events and sedimentation process could be inferred. The results of photo-sampling method show that 4-12sets of photo sampling are enough for obtaining the grain size distribution at a sampling site and the trend of the cure of grain size distribution is the same asthat of grid-sampling method.