Abstract Excessive sediment is an important form of surface water impairment throughout the continental United States. Numerous studies have investigated the role of upland soil erosion as a source of sediment and phosphorus, but contributions of streambank erosion are still poorly understood. Current methods such as delineation and automated channel planform morphometric models are either too time-intensive, or do not provide adequate spatial resolution to measure smaller rivers over large scales. To estimate sediment contributions from river migration on large scales, we have created the Aerial Imagery Migration Model (AIMM), a Python and ArcPy based automated channel migration model designed to estimate volumes of erosion and deposition related to channel migration. AIMM utilizes the Normalized Difference Water Index (NDWI) to derive binary representations of river channels from aerial photography. The location of the channel is then compared between two time periods to identify zones of erosion and deposition and the volume loss related to channel migration is then calculated using a LiDAR-derived DEM. When compared to three delineations and the RivMAP model in the South Fork Iowa River watershed, AIMM was found to have a 98% agreement with RivMAP, 79% agreement with delineations, and predicted net sediment flux that was within one standard deviation of the mean prediction from the delineation analysis. Where public imagery is available, AIMM can be widely applied to estimate volumes of sediment loss in a time and cost-efficient procedure. In particular, the use of AIMM within the project-planning phase of conservation efforts could help focus resources in areas where they can have the most impact.

中文翻译：

---

从高分辨率航空影像和地形分析自动测量侵蚀河岸体积

摘要 沉积物过多是整个美国大陆地表水损害的一种重要形式。许多研究调查了高地土壤侵蚀作为沉积物和磷源的作用，但对河岸侵蚀的贡献仍知之甚少。当前的方法，例如划定和自动渠道平面形态测量模型，要么过于耗时，要么无法提供足够的空间分辨率来测量大尺度的小河流。为了估计大规模河流迁移的沉积物贡献，我们创建了航空影像迁移模型 (AIMM)，这是一种基于 Python 和 ArcPy 的自动河道迁移模型，旨在估算与河道迁移相关的侵蚀和沉积量。AIMM 利用归一化差值水指数 (NDWI) 从航空摄影中导出河道的二进制表示。然后在两个时间段之间比较通道的位置以识别侵蚀和沉积区域，然后使用 LiDAR 衍生的 DEM 计算与通道迁移相关的体积损失。与 South Fork Iowa River 流域的三个划界和 RivMAP 模型相比，发现 AIMM 与 RivMAP 的一致性为 98%，与划界的一致性为 79%，并且预测的净沉积物通量在平均预测的一个标准偏差内从轮廓分析。在公共图像可用的情况下，AIMM 可广泛应用于以具有时间和成本效益的程序估算沉积物损失量。特别是，