Water Resources Research Act Program

Details for Project ID 2017MO153B

Satellite-Imagery Based Method for Water-Quality Monitoring and Sediment Budgeting along the Middle-Mississippi River and Its Tributaries Phase 2

Institute: Missouri
Year Established: 2017 Start Date: 2017-03-01 End Date: 2018-02-28
Total Federal Funds: $22,000 Total Non-Federal Funds: $44,000

Principal Investigators: Amanda Cox, Abuduwasiti Wulamu

Abstract: Suspended sediment concentration (SSC) and suspended sediment load (SSL) along the Middle-Mississippi River (MMR) and its tributaries are of significant interest to researchers, engineers, scientists, and water resources managers as sediment erosion, deposition, and transport are fundamental to geomorphic and ecological conditions of the river. Sediment deficits can cause channel degradation that threatens adjacent infrastructure, reduces available terrestrial and aquatic habitat, and changes water quality to such a degree that species become threatened. Conversely, sediment surplus can lead to channel aggradation that is hazardous to navigation, reduces performance of water-intake structures, and increases the amount of nutrients and contaminants introduced to the system that contribute to basin-wide issues such as gulf hypoxia. Currently, in Missouri and Illinois 14 gauge stations on five rivers have monitored water-quality parameters. These gauges have been used to quantify a general trend of decreasing sediment supply from the Missouri River, which reinforces a growing need to understand trends in local sediment sources and sinks that contribute to the regional and basin-wide sediment budget. Spatial resolution of U.S. Geological Survey (USGS) monitoring stations and the fragmented nature of associated datasets make quantification of a local sediment budget difficult as there are a large number of unmonitored rivers and streams in the area. Recent advancements of remote sensing technologies have led to capacities to characterize surficial SSC in fluvial environments. The objective of the proposal is to develop a satellite-imagery based algorithm to monitor surficial SSC along the MMR between the Mississippi River-Missouri River and Mississippi River-Ohio River confluences. The efforts proposed in this work will generate a time history of surficial SSC and SSL to (1) supplement fragmented datasets at existing USGS gauge locations and (2) create new data for unmonitored tributaries along the MMR. Following development of the time histories of surficial SSC and SSL, these data will be used to estimate a local sediment budget of the study area. The resulting SSC data will be made available to the Missouri Water Resources Research Center for quality assurance and quality control with the intent that the proposed work may be integrated into existing USGS databases for public use. This proposal is for the second year of funding for an existing project. This proposal provides the results to date for the project and details the tasks to be completed during the second year. Over the past year, previous regression models were reviewed to provide the foundation for development of the regional model. Publically-available Landsat imagery with 30-meter spatial resolution was obtained from multiple Landsat sensors covering a 33-year period (1982 to present day). A database of available images was developed using pixels co-registered with the USGS gauge locations of interest. The images were filtered for cloud coverage, vessel traffic, and radiometrically and geometrically corrected. USGS gauge data of SSC and turbidity from the USGS Sediment Data Portal and USGS National Water Information System databases were retrieved and filtered for data anomalies. The USGS gauge stations and corresponding reflectance were divided into a development group and a test group. A best-fit linear regression model was created using the development group for both suspended sediment concentration and turbidity. The test group was used to validate the model through comparison of SSC values computed using the remote-sensing model and in-situ SSC measurements. The scope of work proposed for this coming year includes using the model to estimate SSC at approximately 17 tributaries along the Mississippi River (13 are completely unmonitored and 2 only monitor stage/discharge). A time series of surficial SSC will be developed for each tributary and SSL into the Mississippi River will be quantified for the 33-year period of record that Landsat images are available. Associated flow discharges for each of the tributaries will be estimated and used with the SSC to compute the SSL. Trends in the seasonal, annual, and decal sediment budget of the MMR will be evaluated using the computed SSL data.