Research Project: Development of a Monitoring Network, Engineering Tools, and Guidelines for the Design, Analysis, and Rehabilitation of Embankment Dams, Hydraulic Structures, and Channels

Location: Agroclimate and Hydraulics Research Unit

2024 Annual Report

Objectives
1. Develop new and/or improve cloud-based technologies and engineering tools for data acquisition for dam, hydraulic structure, and channel monitoring and reservoir management that allow data to be used in off-site decision support systems and integrated into watershed modeling tools. 2. Develop new and/or enhance design guidance, engineering tools, software, and best management practice standards to monitor and assess the performance of dams and hydraulic structures as erosion control measures. 3. Enhance dam and/or spillway erosion prediction models through real-time monitoring and/or physical modeling of embankment dam and/or spillway erosion processes and breach. 4. Engage Missouri River Basin stakeholders through our University of Missouri Research and Extension partners to characterize water resource managers’ and producers’ behavior, attitudes, and economic considerations with respect to irrigation water use, conservation, and flood mitigation; and to introduce them to analytical based decision aides for evaluating new technologies, best management practices, and cost-benefit assessment. 5. Develop holistic stochastic optimization models, risk assessment, and decision support tools to improve sustainable agriculture production water management practices, while enhancing long-term landscape health in temperate environments. These models will focus on water availability, water storage, and flood mitigation with dynamic economic assessments. This objective will be met through a collaborative effort between HERU and University of Missouri partners.

Approach
Global environmental change and human activities are threatening water and land resources and economic growth. Further, water resources infrastructure is experiencing structural deterioration due to anthropogenic changes, which subsequently affect the water cycle and sediment and pollutant delivery to downstream waterbodies. Increasing occurrences of extreme weather exacerbate vulnerability of water infrastructure and threaten public health and safety. These challenges are acknowledged by bipartisan declarations to modernize water resources management and water infrastructure for a sustainable economy, advancements in agriculture and conservation, protection of public health, and support of healthy ecosystems. USDA plays an integral role in water resources management and availability across America and is equipped to meet the challenges through scientific discovery and engineering know-how. This project will focus a holistic approach to 1) develop new and/or improved cloud-based technologies and engineering tools for data acquisition for water resources infrastructure monitoring and reservoir management, which will allow data to be used in off-site decision support systems and integrated into watershed modeling tools, 2) expand hydrologic and hydraulic prediction models through real-time monitoring and/or physical modeling of water resources infrastructure through the implementation of a dam monitoring and inspection network, 3) develop new and/or enhanced designs, engineering tools, models, and best management practice standards to monitor and assess the performance of water resources infrastructure, 4) engage stakeholders through extension and outreach activities to assess the economic benefits of new and/or rehabilitated water resources infrastructure and conservation practices, and 5) develop stochastic optimization models and risk assessments to improve sustainable agriculture production and water management practices, while enhancing long-term landscape health in temperate environments. Federal and state agencies, agricultural producers and farmers, tribal organizations, emergency and floodplain managers, lending institutions, insurance agents, policy makers, and the international scientific community will reap the benefits of these advancements.

Progress Report
For Objective 1, ARS scientists in Stillwater, Oklahoma, in collaboration with University of Missouri (MU) scientists in Columbia, Missouri, have continued research on monitoring dams with unmanned aerial systems (UAS) and other sensor networks. Data was collected from the USDA-ARS Field Station Dam in Woodward, Oklahoma. Data gathered using field instrumentation and UASs during the 2022-2023 growing session was processed. Two ARS support engineers completed training and became certified UAS pilots, and three students from the University of Missouri were trained in UAS mission planning, image acquisition, image processing, data analysis, and interpretation. In addition, ARS scientists in Stillwater, Oklahoma, in collaboration with the ARS Partnerships in Data Innovations (PDI) are jointly working with a private consultant to develop a UAS fleet management system. ARS scientists in Stillwater, Oklahoma, in collaboration with a Virginia Tech University scientists continued the development and testing of low-cost meteorological sensors. New collaborations were forged with ARS scientists in Lincoln, Nebraska, on this effort. Scientists will continue this effort including working with the ARS (PDI) on the Internet of Things (IoT) for streamlining data transmission. ARS scientists in Stillwater, Oklahoma, are collaborating with Oklahoma State University scientists in Stillwater, Oklahoma, on smart stormwater sensing and developing a framework for integrating diverse hydrologic variables into a machine learning rainfall-runoff model to evaluate the impact of storm events on reservoirs. ARS scientists in Stillwater, Oklahoma, are collaborating with the ARS PDI, ARS scientists in Lincoln, Nebraska, and Oklahoma State University scientists on building, maintaining, and securing all PDI cloud resources in the Azure Commercial Environment to ensure functionality of ArcGIS Enterprise software suite, including upgrades for ArcGIS Knowledge, Apps, and other tools. The PDI cloud environment has continued to expand its available tools including the creation of an early version of a UAS fleet management platform that will allow pilots to file their flight plans, manage their UAS fleet, and upload UAS imagery into the PDI cloud. It allows for images to be processed with ArcGIS and be presented back to the pilots. These pathways are important resources for UAV-based monitoring features planned within the dam analysis modernization of tools, applications, and guidance standards. ARS scientists in Stillwater, Oklahoma, in collaboration with the ARS PDI and other ARS scientists are working to integrate Environment Systems Research Institute’s (ESRI’s) Knowledge Graph database system with the National Agricultural Library Thesaurus (NALT) concept space. The intent of this project is to lay a foundation for data upload, analysis, and retrieval unlike any other within ARS. The Knowledge environment includes graphical data analysis tools that have industry-changing potential. NALT provides a framework for developing ontologies and metadata that maximize machine learning capabilities and data retrieval. Creating a more robust NALT resource streamlines the process by which researchers will be able to label their data with the appropriate URI. ARS scientists and ARS PDI has forged a collaboration with University of Texas-Arlington (UTA) computer scientists and USDA soil scientist to integrate the project’s data dictionary with NALT, while the UTA team has been building a Knowledge graph based on USDA soil data. This work has potential to provide end users improved accessibility to soil data, which has widespread implications for dam design and rehabilitation and conservation practices and standards. ARS scientists in Stillwater, Oklahoma, in collaboration with ARS PDI is building on the success of first generation PDI IoT Hub. Additional capabilities have been added to not only push data to an ArcGIS GeoEvent server but to the PDI PostgreSQL server and to cold storage for archiving. Currently, an email alert system is being implemented that will actively monitor connected devices and incoming data and will send out automated alerts if anomalous activity is detected. ARS Scientists in Stillwater, Oklahoma, in collaboration with ARS PDI and Oklahoma State University scientists trained machine learning models (e.g., random forest, support vector machine, and XGBoost) to predict water levels of Lake Carl Blackwell in Stillwater, Oklahoma. Three additional watersheds were added for evaluation of their flood stage prediction in streamflow, surface runoff, and lake levels. ARS scientists in Stillwater, Oklahoma in collaboration with ARS PDI, USDA-NRCS, the Bureau of Land Management, and University of Montana made improvements to the Rangeland Analysis Platform (RAP). The RAP technology primarily helps land managers assess vegetation trends on public lands and to assist private ranchers maximize their grass production. Improvements to RAP resulted in new vegetation models that allow land managers forecast location-specific wildfire risk across the Great Basin. This forecast tool has implications to dam safety due to the changes that occur in hydrology due to wildfires and regrowth of vegetation that is different than vegetation prior to wildfires. For Objective 2, ARS scientists in Stillwater, Oklahoma, have completed modifications of a stepped spillway testing facility by installing one cycle of a labyrinth weir. Testing has initiated, and it will continue in the coming year. Site visit to the Sugar Creek Watershed has been scheduled with the USDA-Natural Resources Conservation Service (NRCS) to evaluate the performance of riffle-pool rock chutes that were constructed approximately 20 years ago. ARS scientists in Stillwater, Oklahoma, in collaboration with Oklahoma State University scientists in Stillwater, Oklahoma, have completed several USDA-NRCS National Engineering Handbook (NEH) chapters. A revised draft of the NEH stepped spillway chapter has been submitted to USDA-NRCS. NEH chapters on plunge basins and two-way covered risers were submitted to USDA-NRCS, and the technical writing team has begun revisions to the NEH chapter on plunge basins based on USDA-NRCS technical review of the document. ARS scientists in Stillwater, Oklahoma, in collaboration with Oklahoma State University scientists in Stillwater, Oklahoma, are working on the development of web-based applications that will aid hydraulic engineers in the design of rock chutes, plunge basins, and stepped spillways. This work includes transforming in some cases Excel spreadsheet style computations into applications using HTML, C#, and Python codes. For Objective 3, ARS scientists in Stillwater, Oklahoma, have focused on advancing prediction of earthen dam overtopping and erosion. For dam overtopping, artificial intelligence algorithms were utilized to forecast reservoir levels from meteorological observation networks (e.g., from NOAA and MESONET) and low-cost water level sensors. Real-time, twenty-four-hour ahead forecasts were developed from lakes with forecast accuracy of +/- 3 inches. Results from this work will inform ongoing efforts to install low-cost water level sensors and provide real-time dam overtopping forecasts. For prediction of earthen dam erosion, sensors were developed to continuously monitor erosion in laboratory tests. Prior to this work laboratory erosion tests utilized less-frequent manual measurements. Results from this work has shown significant differences in erosion processes between dry and wet soils. Research is ongoing to develop erosion models for dry soils aimed at informing in-house software used to predict dam failure times. Progress was made by ARS scientists in Stillwater, Oklahoma, in collaboration with Oklahoma State University scientists in creating training transcripts from videos for Windows Dam Analysis Modules (WinDAM) training. For Objective 4, University of Missouri scientists in Columbia, Missouri, in collaboration with ARS scientists in Stillwater, Oklahoma, complied survey results to gauge the public perception on the impacts of floods among the lower Missouri River basin. In addition, these scientists compiled and incorporated details related to adoption and use of extreme weather adaptation and mitigation strategies among producers and consumers in Missouri. These datasets have the potential to develop decision support tools for complex water management and climate change adaptation. For Objective 5, University of Missouri scientists in Columbia, Missouri, in collaboration with ARS scientists in Stillwater, Oklahoma, completed the development of hydrologic and biogeochemical transport models for the State of Missouri and the Mississippi River Basin with seven regional catchments. This development includes incorporating the operational details of reservoirs and agricultural operations including row crops, which are significant in rainfall-runoff processes. In addition, a climate database was compiled with future precipitation and air temperature projections. This database was used to (i) develop multiple climate change indices that capture the changes in extreme events in the Missouri River Basin, and (ii) prepare weather input database for the hydrological models to evaluate changes in runoff patterns through 2050. Training was provided to postdoctoral researchers, and students in model development, testing and validation, water quality assessment, and policy aspects related to water and natural resources management. Visiting scholars were trained on analyzing spatial variability of runoff and their relationship to agrochemical transport in streams in Ohio and Arkansas River catchments.

Accomplishments

Review Publications
Wise, J.L., Al Dushaishi, M. 2024. Prediction of soil erosion using 3d point scans and acoustic emissions. Water. 16(7). Article 1009. https://doi.org/10.3390/w16071009.
Weaver, S.M., Guinan, P.E., Semenova, I.G., Aloysius, N., Lupo, A.R., Hunt, S. 2023. A case study of drought: The dry summer of 2022 in the Midwest USA. Atmosphere. 14(9). https://doi.org/10.3390/atmos14091448.
Wise, J.L., Nygaard, R. 2023. Comparison of numerical methods that predict wellbore cement sheath integrity. Open Journal of Engineering. 2. Article 021048. https://doi.org/10.1115/1.4063342.