unizeit Schriftzug

Infrastructures for Big Data from marine research

Data and data-driven services should be made available, merged, shared and used in a trustworthy manner. These are the aims of the digital infrastructure initiative GAIA-X, and are being implemented by the project MARISPACE-X for the particular requirements of data obtained from the sea.

Coast near Heidkate, Schleswig-Holstein
@ Felix Gros / Kilian Etter, Institut für Geowissenschaften, Uni Kiel

MARISPACE-X stands for Smart Maritime Sensor Data Space X. The aim of this project is to realise a new kind of digital cooperation in marine research. The objective is to facilitate and improve the collection and analysis of all ocean-relevant data. Maritime data is to be usable, interlinked and processed via intelligently networked objects (Internet of Underwater Things, IoUT). The latter, a network of Internet-connected devices that transmit sensor data to a cloud for central processing, is particularly challenging. "The medium of water is the limiting factor here. When the sensors are underwater, their signals cannot be transported through the water," explained Professor Matthias Renz. The computer scientist heads the working group Computational Archaeology – Data Science and coordinates the CAU part in MARISPACE-X. In order to pass on the data, the diving robots or underwater measuring devices must first come to the surface. "You don't have this obstacle in other applications. That's why we also need special methods for collecting information and enabling the sensors to communicate with each other, and the whole thing should be effortless as well as smart," emphasised Professor Renz.

Perhaps this challenge was one of the reasons why the project was successful in the funding competition "Innovative and practical applications and data spaces in the digital GAIA-X ecosystem", although originally marine sciences were not intended as a topic area at all. The sectors mentioned in the call included energy and mobility, health, geoinformation, and smart living. The winning projects are intended to demonstrate the technological and economic feasibility as well as the usability of innovative digital technologies and applications. The MARISPACE-X consortium led by cloud provider IONOS SE (1&1) is to receive in the region of 15 million euros from the Federal Ministry for Economic Affairs and Energy for three years.

First of all, we take measurements with all sorts of sensors in order to get the most comprehensive picture possible. This generates a tremendous volume of data.

Natascha Oppelt

The network is coordinated by the Kiel-based company north.io GmbH (formerly EGEOS). Other partners include the Kiel start-up TrueOcean GmbH, Kiel and Rostock Universities, GEOMAR Helmholtz Centre for Ocean Research Kiel, the Fraunhofer Institute for Computer Graphics Research (IGD), Stackable GmbH and MacArtney Germany GmbH.

Kiel University is represented by five working groups from computer science and one each from geography and geosciences. "Together with north.io, computer science is developing the basic methods for the data space and data infrastructure. Various application examples for the digital infrastructure will represent the different aspects of data collection," explained Professor Renz who, as a data science expert, is the interface between computer science and the other sciences, and is also driving the project forward within the framework of the "Digital Ocean" in the CAU’s priority research area Kiel Marine Science (KMS).

Satellite image of the Baltic coast around Kiel.
© Geographisches Institut, Uni Kiel

Satellite image of the Baltic coast around Kiel. Sensors with different spectral wavelengths are used to map seagrass (green stripes off the coast). The image of the Heidkate study area was evaluated in a preliminary study.

One application example that generates tremendous amounts of data is supervised by Kiel professor of geography Natascha Oppelt (working group Earth Observation and Modelling) and geophysicist Dr Jens Schneider von Deimling (working group Marine Geophysics). Using a combined approach of optical remote sensing and ship-bound hydroacoustics, they want to develop and validate a method with which aquatic vegetation on the Baltic Sea coast such as seaweed, brown kelp or invasive red algae can be recorded and monitored over a large area. Seaweed and brown kelp grow in shallow waters, are important habitats for fish and small organisms, and serve as biological climate protection by sequestering carbon and so reducing the carbon dioxide content in the atmosphere. "It is not yet known where in Schleswig-Holstein these species do and don't grow, which influencing factors come into play and where it would make sense to plant them," explained Professor Oppelt. “First of all, we take measurements with all sorts of sensors in order to get the most comprehensive picture possible. This generates a tremendous volume of data. It can very quickly go into the terabyte range." Following on from this, computer science is required to help extract the data that is really relevant using pattern recognition and artificial intelligence. Once a combination of sensors has been identified that maps the situation well, the next step is to refine the process to such an extent that the native and invasive aquatic vegetation on the coast can be continuously monitored and researched. Not only are the measurements complex; so too are the subsequent computing processes. "We are in the field of Big Data, and we need to move towards cloud solutions. Our data can no longer be processed on individual computers," said Professor Oppelt.

The corresponding infrastructure is being developed in MARISPACE-X. This includes a secure cloud environment and combination with edge and fog computing. Today, the cloud is already often widely used as a virtual data store for photos, videos or documents. Cloud computing refers to a virtual and scalable IT infrastructure. It includes applications, data, storage space and computing power from a virtual data centre. This consists of interconnected computers. Edge is the place where the data is generated, either through direct sensor measurements or as satellite data. Edge computing envisages carrying out part of the data processing locally, on a system close to the place of action, rather than in the cloud. If an additional processing layer is introduced between the edge and the cloud, we call it fog computing. "This infrastructure needs to be created. Because without stations that can communicate with the sensors and without a resilient communication network, even the most intelligent algorithms are useless," confirmed Professor Renz.

Author: Kerstin Nees

GAIA-X

The Federal Ministry for Economic Affairs and Energy (BMWi) launched the GAIA-X initiative in 2019. The aim is to reduce dependence on American and Chinese IT providers and market-dominant data-driven platforms. The project is now being supported by several European countries and by companies all over the world.

With GAIA-X, international stakeholders from business, science and politics are contributing towards the sustainable development of Europe's next generation data infrastructure. The aim is a secure and networked data infrastructure that meets the highest standards for digital sovereignty and promotes innovation. In an open and transparent digital ecosystem, data and services should be made available, merged, shared and used in a trustworthy manner.

GAIA-X's IT architecture is based on the principle of decentralisation. GAIA-X is the interaction between numerous individual platforms, all of which follow a common standard. The digital ecosystem is to ensure that European companies and business models can be competitive.