SeaSense lays foundation for detailed coastal monitoring network

Recreational vessels travel thousands of kilometres through coastal waters every year and already carry sensors that measure wind, depth and water temperature. Project SeaSense wants to combine this unused data in a complementary, low-cost monitoring network. An initial long-term trial shows which measurements are scientifically useful – and where the technology still falls short.

A modern sailing yacht is effectively a floating measuring station. Most vessels are equipped with a wind meter, depth sounder and water-temperature sensor, as well as a digital network connecting these instruments. Data is continuously generated while sailing, but usually disappears as soon as the equipment is switched off.

Project SeaSense wants to put this information to use in coastal research. Together with researchers from The Hague University of Applied Sciences and TNO, the Dutch non-profit initiative has developed a system that automatically collects data from recreational and charter vessels. In a new research paper, Zoran Kovačević, Amey Vasulkar and Hanno Hildmann describe both the technical architecture and the results of the first long-term field test.

Complementing existing monitoring networks

Coastal seas are among the most heavily used and ecologically valuable parts of the ocean. At the same time, accurately monitoring them is difficult. Fixed stations only record what happens at a single location, research vessels are expensive and only available for limited periods, and satellite measurements become less reliable close to shore.

Recreational and charter vessels could complement these existing monitoring networks. They frequently pass through areas with few fixed measuring instruments and often revisit the same locations. Charter fleets, in particular, follow similar routes throughout the sailing season, allowing them to record repeated changes in the water, weather and seabed.

The main challenge is the quality of the measurements. Instruments on recreational vessels are not calibrated for scientific research. In addition, the positioning and installation of sensors differ from one vessel to another. A wind meter can be affected by the sails and mast, while a depth sounder does not always account for the distance between the sensor and the bottom of the hull.

SeaSense therefore links each measurement to information about the instrument, vessel and installation. When boats repeatedly pass the same location or sail close to an official monitoring station, researchers can identify and correct systematic deviations.

Two levels of measurement

The architecture consists of two tiers. The first primarily uses equipment already installed on board. A compact unit reads the vessel’s NMEA marine data bus and collects information including position, speed, course, wind, water depth and temperature. The unit adds its own measurements of atmospheric pressure and vessel movement.

According to the researchers, the hardware for this base module costs a few hundred euros. Because it uses existing sensors, the marginal cost of measuring an additional parameter is virtually zero.

The more advanced second tier adds a flow-through water-quality system. A pump moves seawater past calibrated sensors for pH, dissolved oxygen, electrical conductivity and temperature. This system costs approximately €1,000 and requires permanent installation in the vessel.

The data is initially stored locally and subsequently transmitted via Wi-Fi or a mobile connection. If the vessel temporarily loses connectivity at sea, the system retains the measurements until the connection is restored.

From Medemblik to the Baltic Sea

The main trial took place between June and August 2026 on board a Contest 63CS yacht. The vessel left Medemblik, travelled through the German Bight, the Kiel Canal, and the Baltic Sea to the Stockholm archipelago, and subsequently returned to Knokke.

© Metrosea2026

For validation, the researchers analysed 1,080 observations collected up to and including 10 August, with a median interval of 15.5 minutes. The entire test lasted 69 days. Performance varied considerably between the different types of sensors.

Atmospheric-pressure measurements produced the best results. The on-board sensor consistently recorded approximately 1 hectopascal too high. This deviation remained stable throughout the trial. After a single comparison with an official weather station, the offset could be corrected, reducing the root-mean-square error to 0.49 hPa. According to the researchers, this brought the measurements into compliance with the standards for synoptic weather observations.

The depth sounder also produced usable information, particularly in shallow water. At depths of up to five metres, the mean deviation was 0.37 metres, with a spread of 1.24 metres. Across 135 repeated passes over the same locations, measurements in water up to ten metres deep differed by a median of 0.45 metres. Corrections based on temperature and salinity further improved the results.

The measured water temperature deviated by an average of just 0.09 degrees Celsius from satellite data. The overall spread was larger, partly because a sensor mounted in the hull measures at a different depth from a satellite, which records the temperature at the water’s surface.

Water-quality sensors remain vulnerable

The water-quality sensors proved more sensitive. At Den Oever, the conductivity sensor clearly detected the transition from fresh IJsselmeer water to saltwater. Within 124 minutes, the measured conductivity increased from 1.5 to 46.3 millisiemens per centimetre. However, the sensor failed a few hours later, on the first day of the sea voyage.

On two crossings 42 days apart, the pH sensor recorded a similar pattern in the Elbe plume. Because no independent reference measurement was available, the researchers could not establish its absolute accuracy. The dissolved-oxygen sensor produced readings approximately one-third too low and was therefore deemed unsuitable for reporting.

Wind measurements also remain complicated. Data from the masthead showed a strong correlation with the European ERA5 weather model but exhibited a 11-degree directional bias. The sails, mast, rigging and heel of the vessel all affect airflow. The researchers therefore concluded that, in open water, a single moving wind sensor adds little information to existing models.

Proof still needed across multiple vessels

The trial shows that relatively simple measurements, such as atmospheric pressure and depth, hold the greatest potential. Water-quality sensors can reveal valuable local changes, but they must become more resistant to seawater, biofouling and prolonged use.

The research was also limited to a single yacht. Recreational sailing is seasonal and concentrated along popular routes and on days with reasonably good weather. The resulting data is therefore neither neutral nor evenly distributed.

The next step is a trial involving multiple vessels. This will allow SeaSense to investigate whether boats crossing the same area produce comparable measurements and whether deviations can be corrected automatically for each instrument. Eventually, quality-controlled depth and environmental measurements could be transmitted to European platforms such as EMODnet.

SeaSense does not promise to replace professional monitoring buoys, satellites or research vessels. Instead, the project makes use of something that already exists: thousands of boats equipped with sensors and continuously travelling through coastal waters that are otherwise difficult to monitor. If their data can be made sufficiently reliable, they could provide a new and highly detailed view of coastal seas at relatively low cost.