Recipient Organization
CLEMSON UNIVERSITY
(N/A)
CLEMSON,SC 29634
Performing Department
South Carolina Water Resources Center
Non Technical Summary
With increasing flooding and drought events in the past fifteen years, the disparate water use patterns in the Southeast are becoming more apparent and forcing decision-makers to rethink water use/ allocation, and disaster response. At the same time, the Southeastern United States has experienced rapid growth and increased urbanization as individuals flee the climates of the Northeast and Midwest, and businesses seek out states that are relatively more labor and tax-friendly. Additionally, many of these regions have natural amenities, like beaches, lakes, and mountains, which are attractive to young professionals, retirees, and entrepreneurs. As such, many communities within the eight major river basins of South Carolina have experienced accelerated growth and development, placing current and future constraints on the region's existing resource base. With increased stressors, existing water use and allocation patterns are questioned as population and development increases the demand for water per capita both inside and outside source watersheds (Bidwell and Ryan, 2006; Renwick and Green, 2000; U.S. Environmental Protection Agency, 2002). Agricultural and natural resource management in South Carolina faces challenges from both human population growth (Milesi et al., 2003) and expected climate change (Mearns et al., 2003). These forces will likely increase pressures on our limited water resources, and it is critical that we develop new ways to monitor and understand water quantity and quality relationships across diverse landscapes. New water quality/quantity sensor technologies are rapidly emerging from the laboratory to the field and there is an urgent need to develop methodologies to evaluate and integrate these sensors which will provide a more accurate and spatially explicit understanding of our environment.One specific example in South Carolina relates to extreme growth limitations in the Lower Savannah River basin, ultimately attributable to water policy and management failure. Water quality and quantity are inextricably linked, and the Savannah River allocations of both reflect a bias for the state of Georgia. Georgia enjoys a water quality advantage in its appropriation of 97 percent of the existing waste load allocation under the EPA's 2006 Total Maximum Daily Load (TMDL) Lower Savannah standard for dissolved oxygen. Both states and the EPA recognize the standard's unrealistic nature; consequently, Georgia and South Carolina are negotiating with EPA Region 4 to set a more realistic standard and to reallocate waste loads. Until a new allocation exists, growth is very limited for South Carolina.While the above example illustrates one of the major issues facing one river basin, it is by far not the only issue. Initial Task Force Report Findings from the South Carolina Governor's Floodwater Commission identified a number of urgent action items including the need to build a new cyberinfrastructure to support planning and alerts from flood hazards. As part of this next-generation infrastructure, the Smart Rivers and Dam Security Task Force recommended the installation of metrological and river stage sensors in 150+ key areas where no data is available. This data would greatly increase flood forecasting and modeling accuracy and is clearly a critical need to help make South Carolina more resilient in the face of increasing flood events.
Animal Health Component
15%
Research Effort Categories
Basic
10%
Applied
15%
Developmental
75%
Goals / Objectives
Detailed water quantity data is critical to understand the interaction between the surrounding land uses and receiving waters like streams, rivers, ponds, and lakes. Intelligent River® radar and Light (or laser) Detection and Ranging (LiDAR) water level sensors provide accurate height information, using a low form-factor sensor enclosure placed above the water surface that makes it possible to monitor streams in a resilient way that is not possible with in-water sensors. Water temperature changes caused by imperviousness surface can directly impact stream biota, so a non-contact temperature sensor is included with the level measurement sensor. A series of best-in-class sensors have been integrated into the system to support water quality and soil moisture monitoring.An applied research objective is to support the policy planning and management processes of critical issues facing the basin and its stakeholders, with particular emphasis on the agricultural and forest industries. While this research is primarily focused on the aforementioned industries, a secondary research objective is to provide information the rural communities affected by watershed policy decisions. The aim is to provide quantitative and scientifically credible answers to the following key research questions:The specific objectives of this project are:To create strategies and methods to deploy reliable sensor networks to evaluate water quantity and quality across river, forest, urban, and agricultural environments.To develop a framework to evaluate water quantity and quality sensor data in the context of detailed remote sensing derived datasets.To deploy and test integrated IoT systems for water monitoring, including automated metadata, data analysis, evaluation, and visualization systems.To develop a process of deploying selected river stage systems in the highest priority locations (as already identified by SCDNR, other state agencies/universities, and informed by modeling needs) based on available funding.To estimate the quantity of water volume in receiving waters, a combination of level and flow sensors are necessary.These sensors will be temporarily installed and moved throughout the network to develop rating curves.To evaluate commercial sensor systems as well as direct and remote micro-spectrometers for water quality monitoring.
Project Methods
High-resolution spatial data layers can be leveraged to determine the best sensor placement, based on water body and land cover characterization (H. Do et al., 2012; Pollak et al., 2012; Varekar et al., 2015; Jiang et al., 2020). Successful sensor placement is dependent on understanding the spatial and biogeochemical heterogeneity both in water bodies and watersheds. The key to this concept is identifying hotspots that may represent areas of high-biogeochemical activity in the interface between land and aquatic ecosystems (McClain et al., 2003; Krause et al., 2017). High-resolution LiDAR data can be leveraged to both characterize riparian buffers (Zurqani et al., 2020) and to understand possible direct flow through these buffers (Solomons et al., 2015) to help inform sensor placement. It may be necessary in some instances to deploy mobile sensing to capture the heterogeneity of river systems (Casper et al., 2012). With the high cost of some water quality sensor-systems, it is crucial to optimize the placement of sensors to avoid unnecessary redundancy in connected river networks (Kovács et al., 2015). Upscaling and validating sensor-based water quantity and quality assessments using satellite-based remote sensing platforms has the potential to increase the overall impact of sensor network data on large-scale evaluations of our environment (e.g., dall'Amico, 2012; Cosh et al., 2016).Clemson University's Intelligent River® program has developed a number of proven technologies that will allow for rapid deployment of resilient river stage, metrological, and environmental monitoring systems to support this effort. These sensor network systems have been developed and tested with support from the SC Legislature and Federal grants, including the National Science Foundation, and represent the technological state of the art, both in terms of the sensors employed and the hardware and software systems used to send and archive real-time data. Our efforts have been focused on fresh-water streams. Furthermore, our team is leading efforts to model and monitor flooding inundation in real-time using high-resolution LiDAR-based topography data (from the South Carolina Department of Natural Resources - SCDNR) and satellite remote sensing.The draft report from the SC Governor's Floodwater Commission Smart Rivers and Dam Security Task Force recommended a significant state-supported investment in 150 river stage installations and 150 metrological stations to support modeling and visualization outputs potentially using older technology. There is a clear scientific need for the suggested stage and metrological installations, however, the current Intelligent River® systems at Clemson are arguably technologically superior at a lower cost to South Carolina.Intelligent River® System SummaryThe Intelligent River® systems are composed of a series of hardware and software innovations. They include river stage radar sensors that can safely be installed on bridges (because they weigh a few ounces), through power-optimized internet-connected data computers and rapid deployment systems. Modern all-in-one weather stations are accurate, but much less expensive to deploy. Clemson 'University's Intelligent River® technologies include internet-connected, embedded computing platforms, power, enclosure system, and cloud-based repository for long-term deployments at low cost. The small embedded computer uses the latest machine-to-machine cellular networks available in almost all areas of the state for reliable internet communications. An alternative embedded computer platform uses existing or deployed LoraWAN networks to limit power use. The system is typically powered by a battery pack and solar panel connected through a solar charge controller. The battery system is sized to allow for multiple cloudy days and uses safe and resilient lithium batteries. A custom carrier board system serves several key system functions, including interfacing sensor protocols, recording all sensor data on a removable SD card, limiting power usage between readings, monitoring solar and battery status, and critically serving as a '''watchdog' for the embedded computer by being able to restart the computer if there is a system crash. The custom circuit boards are professionally manufacturer manufactured on-demand to limit any production or deployment issues. Each system component, including the carrier boards, have been designed for resilient sensing and have been extensively field tested as part of the Intelligent River® project. Rapid system deployment is possible because the enclosure, sensors, and solar panel are mounted to one pole which can be driven into the ground in minutes (Figure 1).The data repository system, based on a modern non-structured query language (NoSQL) database and open-source management and presentation systems, integrates streamed sensor data in near real-time. The metadata system has the unique ability to track system-level information about both the sensing platform and each deployment site. This allows tracking of individual sensors and supporting hardware so that individual sensor performance and calibration can be linked and tracked through time. A custom Android application deployment manager is used on a cellphone or tablet with GPS to note all system components (and take a photo of installation), labeled with near field communications (NFC) tags, as they are deployed so that detailed information is tied to all recorded sensor data. The advanced tracking of system and site metadata provides an audit trail for each sensor reading, which can be tied to weather, specific system hardware, and other data sources. The systematic data handling described here is a key part of providing reliable, calibrated air quality data. Data from the data repository system is simultaneously sent to the additonal IoT platforms for uptime monitoring, anomaly detection, and machine learning to relate sensor readings within and between site locations, and to a GIS-based IoT visualization system is used to create data dashboards and map-based sensor visualizations.