Plant-Based Contaminant Monitoring at SRNL: Background 

By Catelyn Folkert
November 25, 2024

Savannah River National Laboratory (SRNL) has successfully leveraged in situ well sensors for long-term contaminant monitoring across multiple Department of Energy sites and internationally. The Advanced Long-Term Environmental Monitoring Systems (ALTEMIS) approach uses sensors installed in groundwater wells to track and monitor contaminant movements to identify anomalies within groundwater plumes. This in turn enables forecasting of groundwater contaminant migration within a proactive monitoring framework via machine learning. The requirement of wells for ALTEMIS sensor deployment creates the need for alternative monitoring solutions in environments without them. Alternative solutions can be costly and require expertise to effectively manage. However, a collaboration between SRNL and the U.S. Army Corps of Engineers (USACE) has created a proving ground for plant-based contaminant monitoring.

During a presentation of the ALTEMIS project and associated machine learning capabilities in 2022, SRNL caught the attention of the Department of Energy Office of Fossil Energy and Carbon Management. This office is currently pursuing projects in phytoremediation, a process in which plants are used to pull contaminants from groundwater or soil. Phytoremediation is both a cost-effective and sustainable solution for environmental remediation that works as a type of organic sensor in a variety of climates. Though environmental variables such as seasons, lighting conditions, plant dormancy and weather can hinder the success of a plant-based approach, the versatility and accessibility provide significant benefits over other types of monitoring. Phytoremediation doesn’t require special equipment or a high level of technical expertise to implement and functions as a passive monitoring system comprised of plants native to the local ecosystem.

Trays of Bahia grasses undergoing experimental application of various treatments. Photo by Franz Lichtner (U.S. Army Corps of Engineers).

To integrate scientific developments across the portfolio, the Department of Energy Office of Fossil Energy and Carbon Management connected SRNL with the USACE to leverage SRNL’s expertise in long-term monitoring and machine learning. “Machine learning gives you an idea about how you need to replenish a specific species of plant or grass and what you need to do in order to control other plant species or weeds that could be interfering, said Tom Danielson, SRNL scientist.  

Since the relationship began, SRNL and the USACE have conducted experiments and created various proofs of concept to showcase the machine learning capabilities.