Field Work from Your Desk: Autonomous Water Quality Assessment

The Challenge

Eelgrass (Zostera marina) is a valuable coastal habitat in Massachusetts, forming a complex underwater landscape that stabilizes seafloor and adjacent shoreline, filters sediments and nutrients, and provides important habitat for a variety of shallow water species. Once thriving across the state, extensive losses have been observed over the past 80 years resulting from wasting disease, increase in pollution and water quality deterioration, and other stressors. North of Boston, a 43,000-hectare watershed drains into Salem Sound, a large embayment (3,660 ha) surrounded by six densely populated coastal communities. The waters of Salem Sound receive riverine input from the Danvers River, large volumes of stormwater runoff (>60% impervious surface) and wastewater from two treatment facilities. In some areas of Salem Sound, eelgrass population is now severely diminished. Although an expansive bed thrives along the Beverly to Manchester shoreline, a 2016 report from the Massachusetts Division of Marine Fisheries (DMF) highlights some alarming statistics: eelgrass loss in Salem Harbor is down 81% since 1995.

The Opportunity

With funding from the U.S. Environmental Protection Agency (EPA), a team of scientists from the MA offices of Coastal Zone Management and MassBays National Estuary Program set out to examine the link between water quality and eelgrass health by collecting continuous, real-time data using a SeaTrac autonomous vehicle. Building upon a previously successful effort, the team outfitted the SeaTrac vessel with a multi-parameter sonde, and programmed it to autonomously navigate the Danvers River and Salem Sound, in proximity to eelgrass beds and sewer outfalls. Multiple missions would be run to establish a baseline and timeline for understanding nutrient dynamics across spatial and temporal variables.

The Mission

Equipped with the sonde, the SeaTrac vehicle collected temperature, conductivity (salinity), turbidity, dissolved oxygen, pH, and chlorophyll data. For comparison, manual grab samples were also collected at key points along the SeaTrac transect. Twice a month, from July through September, the SeaTrac vessel ran two courses along two pre-established transects: one that traveled down the Danvers River and out to the South Essex Sewage District outfall in the ocean, and another over eelgrass beds along the Manchester coastline and out to the Manchester outfall.

Results

Speed and Power: The SeaTrac ASV managed the currents and the speed / distance requirements and operated successfully in the varied conditions of Salem Sound and the Danvers River. The ASV had plenty of power to support the sensor in this application.

Data: The data collected by the YSI sensor was monitored live and saved for future analysis. “Repeatability of data collection was key and that is what SeaTrac does really well,” notes data scientist Ben Weatherill, of ACASAK Technologies. “The boat will go exactly along the same course, and its ability to loiter over outflows was incredibly helpful.”

Graphical Results: Among Wetherill’s key takeaways was the difference between continuous vs point sampling, specifically for chlorophyll. See charts below:

Comments

Gathering water quality data with an unmanned vessel is an especially efficient way to cover a large area of interest. This mission provides useful baseline data, which going forward helps us better understand the impacts of human actions and a changing climate on a heavily urbanized area. Collecting continuous data across a complete tidal cycle (or more) provides invaluable insight into short-term variabilities that may change across the large tidal ranges (9 ft in Salem Sound), characterizing embayments north of Cape Cod, which are often not captured by traditional monitoring methods. Because SeaTrac is capable of conducting several passes over the same area, it allows repeated data to be collected from the same points over time, thereby creating a continuous spatial and temporal picture (vs fixed buoys). This knowledge better informs suggested changes to local policy and regulations to preserve the eelgrass that remains.

According to Todd Callaghan (CZM): “Collecting data with unmanned systems such as SeaTrac’s increases the quality and timeliness of our efforts to understand and protect our water resources, which is very good news. We would like to see this mission continued and replicated along all of our coastlines.”

Missions