It is currently not possible to accurately determine the navigable depth of harbours with a high level of sedimentation using the technology of echo sounding. Kiel-based geophysicists are involved in a joint project aimed at solving this problem with sophisticated measurement technology and artificial intelligence.
Fluid silt, which is often found in harbours, broad estuaries or coastal waters, looks a lot like solid ground in echo sounding, regardless of how concentrated the sediment suspension is. It is not possible to make a clear distinction between black water, fluid silt or even solid ground using current echo sounding measurements. These differences are crucial, however, in deciding which vessels can enter the harbour and whether the harbour basin needs to be dredged.
"The port authority regularly has to measure the depth of the individual harbour basins. If there is a risk that the guaranteed water depth can no longer be met, the basin must be dredged," explained Professor Sebastian Krastel from the Institute of Geosciences. According to Krastel, dredging can be done more efficiently and in a more targeted manner by exactly characterising the sediments. This also has a direct impact on the environment and climate, as less sediment needs to be transported away – sometimes around 130 kilometres from the port – and the density of the dredged material can be increased.
Echo sounders are currently used in these areas to create an image of the harbour bed. Depending on the frequency used, these generate different reflection depths. For safety reasons, the first measurable horizon is selected as the navigable horizon. "What we don't know, however, is whether the sediment underneath is already so heavily compacted that it really needs dredging," said Krastel. The water might just be muddy, without any other physical properties, and so is not viscous.
Krastel is head of the Marine Geophysics and Hydroacoustics working group and is examining submarine natural hazards within the framework of the priority research area Kiel Marine Science at Kiel University. "We are using and developing acoustic methods to improve classification of the seabed. In deep water, for example, we do this to assess the risk of underwater landslides. We have also, however, developed many methods for shallow water."
Based on this expertise, Krastel and his team are partners of the joint project Akustischer Horizont (AHoi). The aim of the project is to develop an innovative acoustic measurement system to capture effective records of the waterway beds in harbours (and other watercourses), especially in terms of their navigability. The focus here is on classifying the sediment, in other words, calculating how compacted it is in order to assess its navigability. The project is supported by funding of almost €1.8 million from the Federal Ministry of Transport and Digital Infrastructure and is coordinated by the Fraunhofer Institute for Biomedical Engineering. Alongside Kiel University, Hamburg Port Authority and Kiel-based General Acoustics are also involved.
"The new measurement technology will still be based on echo sounding," explained Dr Philipp Held of the Kiel-based working group. Sound at different frequencies will be used to determine the different levels of compaction of the sediment. "These different types of silt are found to have frequency-dependent behaviour. To begin with, we use three different frequencies and aim to ascertain the silt’s geological characteristics and behavioural features from the signals of different frequencies," said Held. The silt’s acoustic characteristics are then examined in greater detail in experiments back in the laboratory. Using a test basin, different concentrations of harbour silt are targeted with different acoustic frequencies. This is to determine the silt’s frequency-dependent acoustic parameters. The main focus here is on the density and viscosity of the sediment. Alongside small-scale measurements, a series of trials is also to be conducted at the Port of Hamburg. According to Held: "When we have the new system, we will take measurements in the harbour and, at the same time, take samples of the silt and examine its characteristics back in the lab."
Based on these series of measurements, the plan is to develop self-learning algorithms that automatically classify measured waterway beds in sediment classes and provide a direct assessment of their navigability.
Author: Kerstin Nees