Author(s): Lei Ye; Jian Wu; Chenchen Wu; Zhuohang Xin; Chi Zhang
Linked Author(s):
Keywords: No Keywords
Abstract: Channels are generally surrounded by well-defined banks that have a distinct signature in the contours. The contour curvature is an important topographic attribute usually used for channel head identification; however, the curvature at channel heads may vary considerably between and even within watersheds. Therefore, uncertainty exists in the extracted channel heads based on curvature due to the specified curvature threshold. Due to the complexity associated with channel heads location, the purpose of capturing the spatial variation and identifying a single probability distribution that could approximate the distribution of channel head curvature in this study is very practical. Based on a non parameter method, the channel heads are extracted from high-resolution digital elevation data at submeter obtained from LiDAR in fourteen mountainous watersheds in United States. Previous researches on channel head identification are mainly concentrated on spatially constant threshold, this study built off of earlier studies by analyzing the spatial variation of curvature within individual watersheds. For these study watersheds we selected, the channel head curvature has been shown to vary within and between watersheds. The paper, by the means of L-moment analysis, concluded that channel head curvature can be well represented by the Gamma distribution. Meanwhile, we tested the curvature extraction and distribution analysis on field mapped channel heads in Indian Creek and Mid Bailey Run, and the results is consistent with that in selected study sites. We also discussed the performance of curvature threshold within and between watersheds. The results indicate that: (1) the channel head curvature is sensitive to the local terrain and varies within and across watersheds, the channel head curvature has been shown to vary within and between watersheds; (2) Gamma distribution can effectively fit the spatial distribution of channel head curvature within watersheds in four kinds of two parameter distribution; and (3) a constant curvature threshold is not suitable for channel head identification ignoring the spatial variation of curvature between watersheds, however, the mode of curvature as constant threshold for channel head identification is acceptable in some watersheds.
Year: 2018