Subsidence monitoring: using AI to identify the underlying causes
The ground in the Netherlands is sinking more quickly than sea levels are rising. For a country with a large proportion of its territory below sea level, that is not a very reassuring prospect. And that is why TNO has devised an artificial intelligence model that can identify the human activities that are the worst offenders when it comes to causing subsidence.
Why is subsidence occurring? It is largely the consequence of human activities, such as the lowering of the ground water level. Another reason is the extraction of underground stocks of water, salt, and gas. So the most prominent causes are known. But it is often a matter of guesswork what exactly the effect is of each process.
Limiting the risk of flooding
The greater the degree of subsidence, the greater the risk of flooding. More knowledge is needed if appropriate measures are to be taken – knowledge about the causes and the extent to which they compound each other. This information is essential for finding solutions that can help slow down the rate of subsidence in the Netherlands or even stop it altogether.
Think or sink
To understand what exactly is taking place in the subsurface, a significant input of artificial intelligence is needed, in addition to the human variety, so TNO has opted for a hybrid AI model. In this case, it means the AI system not only has the relevant information about the subsurface at its disposal, but also that previously acquired knowledge about the dynamics of the ground process is used. The name of this AI project is, appropriately enough, Think or Sink.
Data on the ground water level and gas extraction
In the Netherlands, TNO collects and manages data about the geological subsurface and the ground water level. The information is therefore already present in its databases. However, to carry out effective analyses, it also requires data on gas extraction – which is confidential.
The solution to that problem is federated learning. Using this method, TNO is able to gain access – subject to strict conditions – to a specific part of the database, but without actually sharing the data. The AI then learns from the data but does not store it. It means the AI system developed by TNO can access relevant information on gas extraction.
The challenge for AI
This TNO AI project is initially aimed at two areas. In the first area, the ground is sinking as a result of both the lowering of the ground water level and the extraction of natural gas. Meanwhile in the second area, it is only the former that is responsible for subsidence. The big question is whether artificial intelligence will be able to help penetrate the deeper underlying causes of subsidence.
Christopher BrewsterFunctie:Senior scientist
Christopher Brewster is a Senior Scientist in the Data Science group and Professor of the Application of Emerging Tecnologies in the Institute of Data Science, Maastricht University. His research has focussed on the application of Semantic Technologies, Open and Linked Data, interoperability architectures and Data Governance, mostly to the food and agriculture domains.
Daniël WormFunctie:Senior consultant
Judith DijkFunctie:senior research scientist
Judith is specialised in extracting information from camera images. She now applies the subject of her PhD thesis in Physics, which she obtained 18 years ago, to her work as a research scientist at TNO, including in a research programme on camera systems for the Dutch Ministry of Defence.
Looking for another expert?View all experts
Responsible decision-making between people and machines
Bias in facial recognition and accidents with self-driving cars. AI must be developed further. The fastest way to do this is in close cooperation with people.
Knowledge representation and reasoning
Correct and unambiguous information is needed when making a decision. That is why we use AI technology called "knowledge representation & reasoning".
Natural language processing
What is natural language processing (NLP) and how do we use it intelligently? Find out how we use this AI technique to gather information from textual data.
Robotics and autonomous agents
Robotics brings a future-proof industry a big step closer. For example, we are working on automatic path planning with AI techniques.
Fair machine learning
Fair machine learning is relevant to all kinds of discrimination and bias arising from the use of biased data. Read more!