Monitoring and quantifying suspended sediment concentration (SSC) along major fluvial systems such as the Missouri and Mississippi Rivers provide crucial information for biological processes, hydraulic infrastructure, and navigation. Traditional monitoring based on in situ measurements lack the spatial coverage necessary for detailed analysis. This study developed a method for quantifying SSC based on Landsat imagery and corresponding SSC data obtained from United States Geological Survey monitoring stations from 1982 to present. The presented methodology first uses feature fusion based on canonical correlation analysis to extract pertinent spectral information, and then trains a predictive reflectance-SSC model using a feed-forward neural network (FFNN), a cascade forward neural network (CFNN), and an extreme learning machine (ELM). The trained models are then used to predict SSC along the Missouri-Mississippi River system. Results demonstrated that the ELM-based technique generated R 2 > 0.9 for Landsat 4-5, Landsat 7, and Landsat 8 sensors and accurately predicted both relatively high and low SSC displaying little to no overfitting. The ELM model was then applied to Landsat images producing quantitative SSC maps. This study demonstrates the benefit of ELM over traditional modeling methods for the prediction of SSC based on satellite data and its potential to improve sediment transport and monitoring along large fluvial systems.

The Middle Mississippi River (MMR) and lower Missouri River (MOR) provide critical navigation waterways, ecological habitat, and flood conveyance. They are also directly linked to processes affecting geomorphic and ecological conditions in the lower MR and Delta. For this study, a method was developed to measure suspended‐sediment concentration (SSC) and turbidity along the MMR and the lower MOR using Landsat imagery. Data from nine United States Geological Survey water‐quality monitoring stations were used to create a model‐development dataset and a model‐validation dataset. Concurrent gaging data were identified for available Landsat images to generate the datasets. Surface‐reflectance filters were developed to eliminate images with cirrus cloud coverage or vessel traffic. Using the filtered model‐development dataset, unique reflectance‐SSC and reflectance‐turbidity models were developed for three Landsat sensors: Landsat 8 Operational Land Imager, Landsat 7 Enhanced Thematic Mapper Plus, and Landsat 4–5 Thematic Mapper. Coefficient of determination values for the models ranged from 0.72 to 0.88 for the model‐development dataset. The model‐validation dataset was used to evaluate the performance of the models and had coefficient of determination values ranging from 0.62 to 0.79.

A new water sample collection system was developed to improve representation of solids entrained in urban stormwater by integrating water-quality samples from the entire water column, rather than a single, fixed point. The depth-integrated sample arm (DISA) was better able to characterize suspended-sediment concentration and particle size distribution compared to fixed-point methods when tested in a controlled laboratory environment. Median suspended-sediment concentrations overestimated the actual concentration by 49 and 7% when sampling the water column at 3- and 4-points spaced vertically throughout the water column, respectively. Comparatively, sampling only at the bottom of the pipe, the fixed-point overestimated the actual concentration by 96%. The fixed-point sampler also showed a coarser particle size distribution compared to the DISA which was better able to reproduce the average distribution of particles in the water column over a range of hydraulic conditions. These results emphasize the need for a water sample collection system that integrates the entire water column, rather than a single, fixed point to properly characterize the concentration and distribution of particles entrained in stormwater pipe flow.

Channel bends are associated with secondary, helical currents, shifts of conveyance to the outer channel, and variable lateral sedimentation patterns. Outerbank erosion is of primary concern, resulting in stream-course migration that may place valuable infrastructure and land holdings in jeopardy. In such cases, in-stream river structures such as vane dikes are commonly implemented to restrict the channel course to desired boundaries. Vane dikes are normally installed in series, extending laterally from the outer channel bank into the stream, thereby reducing flow velocity at the outer bank and increasing flow velocity along the channel center and inner bank. Flows around vane dikes and similar transverse in-stream structures have been modeled both numerically and physically in the past, yet the effects on flow velocity within a channel bend due to vane-dike installations have not yet been fully realized. With a focus on the stabilization of two channel bends in the upper regions of the Rio Grande River, a scaled physical model was constructed for the evaluation of various vane-dike field configurations. Structure plan-form angle, spacing, and length were altered between configurations and comprehensive hydraulic data were collected at flow depths below, at, and above structure height. To address flow velocity effects from the structures, a dimensional analysis of influencing parameters was performed, and maximum conditions were used for regression analyses. A series of equations were generated which represent maximum changes in flow velocities at the outerbank, inner-bank, and centerline locations within a channel bend from the installation of vane-dike fields.