Identifying septic systems in rural areas can be challenging when records are incomplete or unavailable. Texas Water Resources Institute (TWRI) researchers are exploring whether satellite imagery could provide another way to locate these systems and better understand where they are found across a watershed.
On-site sewage facilities, or OSSFs, are commonly known as septic systems and are used in rural areas that are not connected to centralized wastewater systems. These systems treat wastewater on the property, but information about their locations can be limited, particularly for older systems.
Knowing where the OSSFs are located can be important when assessing potential sources of bacteria and nutrients in a watershed, because aging or leaking systems can impact water quality. Traditionally, researchers and watershed planners have used approaches such as counting buildings in aerial imagery or using address and building footprint data to estimate where septic systems may be located.
In a recent project, OSSF Identification Using Remote Sensing and GIS, TWRI Research Specialist Ed Rhodes, Graduate Research Assistants Arifa Sultana and Sree Vidhya Kuduvaa Sankarbabu, former Program Specialist Kenneth Ferrell and Associate Director Lucas Gregory, Ph.D., tested whether satellite imagery could provide a more direct way to identify potential OSSF locations.
Looking for clues from vegetation
The project focused on the Tres Palacios Creek watershed along the Texas Gulf Coast. PlanetScope satellite imagery provided 3-meter resolution across multiple wavelengths, allowing researchers to examine vegetation changes that may be associated with septic systems.
Rather than looking for the septic systems themselves, the approach focused on vegetation patterns, Rhodes said. Water and nutrients released from an OSSF can affect vegetation around the system, potentially creating subtle differences that can be detected by analyzing satellite imagery.
Eleven vegetation indices were tested as part of the project. Researchers also used several approaches for analyzing the imagery, including methods looking for areas that remained unusually wet over time.
Results from those analyses were compared with 120 known OSSF locations identified using aerial imagery. Among the vegetation indices tested, the Normalized Difference Red Edge Red (NDRER) index consistently showed the most promise for identifying potential OSSF locations, he said.
The red-edge portion of the imagery may be particularly useful because it can capture subtle differences in vegetation. Areas receiving additional water or nutrients from an OSSF may have different vegetation characteristics than surrounding areas, creating a potential signal for remote sensing to detect.
Exploring a new way to identify OSSFs
Accuracy varied depending on the analysis method and location. Using the NDRER, about half of the known OSSF locations were identified within 30 meters, while some predicted locations were much closer to the actual systems.
Several factors affected the results, Rhodes said. Smaller parcels of land generally had more accurate predictions, while larger properties and areas with substantial tree cover presented additional challenges.
Although the approach still needs refinement, the findings provide a starting point for exploring how remote sensing could supplement existing methods for identifying septic systems.
As satellite imagery becomes increasingly detailed and accessible, it could become another tool for researchers and watershed planners working in areas where OSSF records are incomplete.
This research was funded in part by the Texas Coastal Management Program, supported by the NOAA Office for Coastal Management.
