Month 24 saw the completion of the second batch of sensor and instrument prototypes, developed under WP4 – one of two work packages devoted to the advancement of novel sensing techniques. Work package 4 focused on the development of cost-effective sensing technologies in response to the needs for marine chemistry and deep ocean physics measurements, including biogeochemistry EOVs, deep ocean physics EOVs and MSFD descriptors. The broad focus of the WP4 instrumentation is Biogeochemistry and Biology, with sensors and instruments targeting – carbonate system/ocean acidification, silicate, deep ocean CTD, ocean plastic pollution and in-situ low-level marine radioactivity, through a variety of techniques. Five partners have taken part in the development work. The work package was divided into subtasks, concentrating on different parts of the topic.

Task 4.1 focused on carbonate system/ocean acidification sensors for measuring pH and pCO2 on observing platform. Carbonate system/ocean acidification variables are critical for understanding present day variability and future climate change impacts on the ocean. COTS sensors that are presently used for measuring pH and pCO2 tend to be very expensive, large, and not user friendly in terms of visualizing data and calibrating. We pursued further development of off-the-shelf sensing elements and sensors to a TRL of 5, to enable modularity and use with various ocean observing platforms, improved data logging, visualization and operation, calibration techniques that can be carried out by a non-expert operator and lower overall costs in comparison to the state of the art. These newly developed instruments may have slightly higher uncertainty and lower accuracy compared to more expensive sensors available on the market, however this is balanced by lower cost, smaller size, and increased ease of usability. Further integration and calibration/validation work will be carried out in subsequent tasks in WP5 and WP6 of NAUTILOS. Read more about the accomplishments in this work package here.

Figure 1 Honeywell Durafet III pressure case developed in the scope of NAUTILOS

Task 4.2 concentrated on the development and certification of a new version of electrochemical in situ sensor able to detect silicate with much lower response time than the previous version of the sensor developed by the LEGOS and nke Instrumentation. We designed an original mechanical design including all the electrodes needed to first, produce the reagents required for silicate complexation and second, to detect the silicomolybdic complex formed. The new design allows to change pumping system to fill the circuit and drastically decrease the time of sampling from 13 minutes down to less than 10 seconds therefore compatible with implementation of the sensor onto ARGO float. Finally, certification of the sensor using Certified Reference Material shows good accuracy between the concentration obtained with the sensor via its calibration and the reference values given by the supplier. In terms of repeatability and recovery tests, we need to re-do the experiences as drifts of the signal or the calibration were observed, and the results obtained can be much improved.  We reached TRL 6 as the sensor has been validated in laboratory using real seawater samples. In situ deployment needs to be done (with colorimetric measurements to intercompare the data with sensor’s data) on a fixed platform to move forward to TRL 7. This deployment is already planned in the Thau pond (Mediterranean Sea) at Sete Marine Station. Finally, a demonstration will be done with the sensor implemented on PROVOR float within the WP7, T7.4 to reach TRL 8. Read more about the accomplishments in this work package here.

Figure 2 A new silicate electrochemical sensor developed in the scope of NAUTILOS

Task 4.3 focused on the development of an underwater microplastics sampler as a standard product for the study of oceanic plastic pollution. Our partner SubCtech developed an underwater microplastic sampler prototype with build in data logging and batteries for independent sampling at water depth up to 600 m. A rotary valve was developed to enable sampling of several time-based samples. A pump and flow meter are included within the system to force water through the filters and receive information on sampled water volume. This allows determination of particles per water mass.

Figure 3 A microplastics sampler developed in the scope of NAUTILOS

A first complete system prototype was equipped with four cascade filters containing of three different mesh sizes. The system is designed in a way that the number of filters and samples is scalable. We consider that the underwater microplastic sampler prototype has reached a technological readiness level of 5. A TRL of 9 will be achieved after validation and demonstration in work packages 6 and 7. Read more about the accomplishments in this work package here.

Task 4.4 concerned the development of a low-cost fluorescence microplastic sensor. Automatic monitoring of microplastic particles remains a difficult task, since it requires automatic microplastic sampling, sample treatment and microplastic detection in one system, operating in real time, on-site and without human intervention. Traditionally, microplastic data is collected by (semi-automatic) sampling and mostly manual analysis off-site in an analytical lab. The goal was achieved by designing and building a new autonomous microplastic detection system, consisting of an automatic microplastic sampler, a sample treatment chamber, and an inline particle measurement system. The sampler uses a mesh filter to collect microplastic from marine water, in the size range from 30 to 300 μm. Then the plastic particles are transferred to a reaction chamber, where they are cleaned from the biomaterials and stained with the fluorescent stain Nile red. The microplastic particles are subsequently pumped into a particle detector, where the fluorescent light emitted from each particle is recorded, and the particles are counted, one by one. The intensity, pulse duration and fluorescence color of each particle event is extracted on the integrated computer and yields insight into the particle size and the plastic type. Functionality and characterization tests have been carried out to validate and characterize the system.

Figure 4. The connection of the laser-based fluorescence detector and the staining unit during the test session at NIVAs laboratory facilities developed in the scope of NAUTILOS

The system should be further improved before the large-scale test at NIVAs field station on the Oslo fjord and final installation on the M/S Color Fantasy. Key points to improve the system will be the integration of the system to the Ferry Box to manage the sample volumes according to the amounts of organic and biological material. One of the main challenges was the connection of the volume and flow sampling unit, on the medium volume and medium flow staining camber to the low flow and low volume detector unit. Avenues for improvement are currently being explored. Read more about the accomplishments in this work package here.

Task 4.5 focused on the design, the fabrication and incorporation cost efficient CTD sensors into a customized small footprint modular underwater CTD instrument by employing technological advances of MEMS. A new CTD instrument for deep ocean measurements was developed and tested in laboratory environment under controlled settings. The development process included development of miniature 4 electrode conductivity sensor, thin film temperature Ti RTD sensor, both integrated on the same substrate and an OEM pressure sensor. Read more about the accomplishments in this work package here. Further developments and tests are underway as part of WP6 on calibration and validation.

Figure 5: A new CTD instrument developed in the scope of NAUTILOS

Task 4.6 is dedicated to the continuation of work on a concept for a machine learning (ML) based framework for online data stream analysis. Within the NAUTILOS project many different sensors will be used on different platforms and a large amount of data will be collected. Many of the sensors are continuously analyzing sea water (e.g., Ferry Boxes) and hence, are creating time series data. Since the manual analysis of large datasets is a time-consuming challenge, an automated method that identifies ‘interesting’ changes in the data stream was required. The goal was achieved by creating a successful application of change event detection and how a useful automatic sampling strategy could be derived from it. This was shown on a first dataset collected by DFKI with similar sensors to the once that will be used in the NAUTILOS demo scenarios. However, the adaptation of the method and its evaluation must be done within the demo scenarios of WP7 once the data will become available. Read more about the accomplishments in this work package here.

Task 4.7 focused on the development of a new detection radioactivity system for the deep ocean.  A new sensor was applied in laboratory for calibration for determining the radioactivity levels in aquatic systems. HCMR developed a new radioactivity sensor for the deep ocean based on a detection crystal, connected with a photomultiplier tube, preamplifier, amplifier, and power supply, together with a multichannel analyzer for data acquisition and storage. The enclosure of the system was tested for pressure tolerance. In operation, the system offers a few advantages detailed in deliverable 4.7. The new sensor has achieved TRL4 after validation of these components and the entire technology in a laboratory environment. The next steps will be to integrate the system in an existing network laboratory for operational control in the ocean, to install the system in the rosette of the R/V Aegeao for profiling in the deep sea, to install the deep underwater radioactivity system in a lander for continuous monitoring enabling the stand-alone method. Read more about the accomplishments in this work package here.