3D turbidity by correlating multibeam sonar and in-situ data
Why do we need to measure turbidity?
Turbidity (or the cloudiness of water) is related to the concentration and type of suspended particles in the water column. These particles may be either of planktonic (both zoo- and phytoplankton) or sedimentological origin (resuspension of surface-bound sediments). Combined they form a cloud of suspended particulate matter (SPM) which affects light penetration in coastal waters. In order to ensure good water quality, turbidity and SPM have been monitored in the Belgian Part of the North Sea (BPNS) for decades, either in 1D (moorings, ship-based samples, tripodes on the seafloor, …) or in 2D (Acoustic Doppler Current Profiler, ADCP, transects). However, studies have indicated the very dynamic nature of SPM variability and the complex shape turbidity clouds can have. This justifies the need for 3D measurements of turbidity and SPM.
How can we effectively measure 3D turbidity?
A possible solution lies in multibeam sonars which, next to seafloor bathymetry data, are also able to deliver a 3D dataset of backscatter values in the water column. However, these 3D datasets are still scarcely used in quantitative turbidity and SPM assessments of the water column. One of the causes is that available software lacks proficient processing capabilities and it is not flexible enough to apply new innovative water column processing techniques. Another problem lies in the relationship between the acoustic return signal and the variable character of SPM, which is still insufficiently resolved. The latter problem is thereby not specific to multibeam systems, but to all acoustic echo sounder devices (e.g. ADCPs). The development of an innovative methodology and software that could tackle these problems and convert acoustic water column data to quantitative 3D turbidity and SPM information would be extremely relevant for both science and industry. Scientists would be able to derive SPM concentrations within the water column over large areas or investigate plumes generated by storm events. Industry would benefit from the advanced 3D investigation capabilities of the technique, which for example would allow for monitoring environmental impact of sediment plumes from bottom-disturbing activities, like sand dredging or bottom-trawling fishing activities.
Our research objectives
The ultimate goal and most anticipated result of the TURBEAMS project will be the development of a methodology, which leads to 3D turbidity and SPM imaging based on multibeam water column data, improving future monitoring tools. In order to reach this goal, we outline 4 research objectives.
- The first objective of the project is to determine SPM characteristics (type, size and concentration).
- The second objective is to quantify the relation between multibeam water column data and SPM-related parameters (SPM type, concentration, grain size, optical turbidity) derived from in-situ optical and acoustic sensors. For this, we will use statistical and machine learning methods (e.g. Bayesian Evidential Learning) to analyse the large amounts of data from the many different sensors and additional environmental parameters (e.g. water currents). This novel approach allows for empirically revealing the most relevant parameters and for building a predictive model of the quantified relationships while assessing uncertainties.
- The third objective of the project is to develop a new water column multibeam processing library that allows for flexible and efficient data processing and rapid visualization using python scripts. Then we want to use this library to develop user-friendly workflows that can be applied by other scientists.
- Using the results from the previous objectives, the final (and fourth) objective of the project is to analyse the small-scale, seasonal and spatio-temporal variability and vertical distribution of SPM concentration with special attention on the near-bed SPM concentration in view of evaluating SPM concentration derived from multibeam echosounding for monitoring human impacts.
Data acquisition strategy
A fit-for-purpose acquisition strategy, together with specialized and newly-developed processing techniques will be implemented to fulfil the objectives, building on the survey approach adopted within the ongoing BELSPO-funded TIMBERS project (STEREO III call). The acquisition strategy consists of an absolute calibration of the acoustic backscatter values of the shallow-water multibeam, allowing exchange and comparison of data with other calibrated multibeam systems. We will deploy multiple acoustic systems of the RV Belgica (EM2040, ME70, EK80’s and workhorse ADCP) and a wide range of readily available in-situ optical and acoustic sensors, in combination with the collection of water and plankton samples. Furthermore, we will implement a rigorous pinging strategy for the acoustic sensors, allowing simultaneous, yet non-interfering, acquisition of multi-frequency acoustic datasets. Moreover, all in-situ sensors will be tracked using the underwater navigation system of the RV Belgica (HiPAP). 3D multibeam water column data will be processed using a data processing pipeline that builds on the newly-developed processing library. The acoustic data will be converted into turbidity and SPM information based on the relation between multibeam backscatter values and in-situ sensor data.
The multibeam water column processing workflows and empirical relationships established during the TURBEAMS project may prove crucial in acquiring 3D turbidity and SPM information in marine environments and may serve as a catalysator in follow-up research proposals. Examples include monitoring turbidity changes at windmill farms or downslope turbidity flows in canyons systems, located on continental slopes.