Channel succession and sediment transport in the Black Vermillion River

Abstract

Tuttle Creek Reservoir in northeaster Kansas has lost approximately 47% of its capacity due to excessive sediment being supplied from upstream since completion in 1961. The Black Vermillion River (BVR) is a tributary of the Big Blue River which flows into Tuttle Creek Reservoir. This research utilizes three unique data sets within three separate but related studies to better understand sediment being produced by bank erosion along the BVR. The first study sought to understand how the channel evolution of the upper BVR could be influencing streambank erosion and sediment production. Seven study reaches in the upper BVR watershed, each with two cross sections, that were first monitored between 2007 and 2010 were resurveyed in 2021. The cross-sectional data suggest that channel incision is waning or has ceased, and that the channel has entered a phase of rapid widening with minor aggradation, as 12 of 14 cross section experienced widening over the 14-year study period. Analysis of sequential LiDAR data from 2010 (12) and 2018 of the entire upper BVR riverbank indicated that the widening observed at the cross sections was indicative of most of the riverbank in the study area, with 76 of 78 1-km river segments experiencing net erosion (widening) in the Irish Creek, Main Stem, and Weyer Creek subwatersheds. In the North Fork subwatershed, the same pattern of net erosion was observed upstream of the Weyer Creek confluence; however, downstream all ten 1-km river segments experienced net deposition, which was supported by cross sectional data in the same area. The second study expanded the use of the 2010, 2012, and 2018 lidar data to include the BVR main channel in the lower BVR watershed, which identified an area of severe erosion in the lower BVR watershed with area of heavy deposition downstream near the confluence of the BVR and the Big Blue River. The second study utilized the lidar data to evaluate the effect of the presence, amount, and density of woody vegetation on erosion and deposition rates. Banks with no, little, and ample woody vegetation were isolated. Rates of erosion were calculated for each category with banks void of woody vegetation experiencing between 57% and 288% more erosion when compared to banks with at least some woody vegetation. In contrast, no correlation was found between LiDAR-derived vegetation density metrics and erosion in any of the study reaches. No correlation was found between the LiDAR metrics and deposition except in the North Fork-Lower Main Stem channel in Marshall County, where a modest correlation (R² 0.35) was found. The third study evaluated the use of Structure-from-Motion (SfM) Photogrammetry to monitor erosion “hotspots” over a 1.5-year period at both a seasonal and annual temporal scale. The data suggest SfM is better suited for the monitoring of erosion than deposition, as vegetation establishment where deposition occurs interferes with the SfM image collection. Three study banks experienced large amounts of erosion up to 1.34m³ per square meter of bank face, while other banks became vegetated and experienced little change or minor net deposition. In addition, the average erosion per meter of the bank face of the three study banks in the upper watershed was greater than that of the six in the lower watershed. The three data sets and methodologies provided information that could be employed in other areas of the Tuttle Creek Reservoir watershed and guide BVR restoration efforts. Traditional cross-sectional and other data provided insight into the evolutional phase of the BVR. Lidar derived DEMs and subsequent DoDs provide estimates of erosion and deposition along the entire channel of the BVR and three of its tributaries, and may provide more accurate locations erosional “hotspots” than the methods previously used by the KWO. Erosion hotspots identified with repeat lidar surveys could then be monitored using more exact surveying methods, such as traditional surveying and SfM. The process of creating DoDs along riverbanks could easily be applied to all the Tuttle Creek watershed within Marshall and Nemaha Counties. It is also believed that SfM could be an effective tool in monitoring most of the erosion hotspots within the BVR, as most of them are well suited for the use of SfM.