Soils are under pressure from current farming practices, and the challenges are only increasing with the growing demand for food production. Threats to soil health include loss of organic matter, loss of biodiversity, soil compaction from large machinery and, not least, loss of soil itself due to erosion.

"We are at a crossroad; we need to do something if we are to preserve European and global soil resources. It is a paradox that on the one hand soil is part of the solution to reduce greenhouse gas emissions, and at the same time 60-70% of Europe's soil is not doing well. It is therefore imperative that we have better monitoring of soil quality," says Professor Mogens H. Greve from the Department of Agroecology.

Ambitious climate goals cannot be achieved without increased focus on soil

Similar to the ambitious Danish climate targets, the European Commission has set several concrete goals to prevent further soil degradation and to combat climate change. The targets include zero emissions and the planting 3 billion trees by 2030.

"By 2028, all land managers should have access to verified data on CO2 emissions and removals. In addition, the ambition is for agriculture to have an increased focus on carbon storage and thereby contribute to meeting the 2030 target of a reduction in climate impact of 310 million tonnes CO2eq for the entire soil sector in the EU," explains Mogens H. Greve.

To succeed in improving soil health across Europe and meet the ambitious targets, methods are needed to measure and assess progress towards healthier soils. This is where the Horizon Europe project AI4SoilHealth comes into play. The aim is to support the EU's Soil Health & Food Mission to achieve the goals set out in the EU Soil Strategy 2030.

"We will develop and maintain a digital infrastructure with free access across Europe. The aim is that the infrastructure will be used to assess and continuously monitor soil health in relation to its management," says Professor Lis W. de Jonge, also from the Department of Agroecology at Aarhus University.

Artificial intelligence rhymes with soil

The kind of artificial intelligence the researchers will be working with includes software and big data solutions that can improve and automate decision-making systems. Artificial intelligence can be used for many things in soil science. Soil analysis, remote sensing, and soil mapping are just a few examples.

"In this project, we will use artificial intelligence more systematically to assess soil datasets and measurements, so that we can build an automated data-driven decision support system for European soils. We propose to develop an indicator framework that can provide new and untapped opportunities for monitoring soil properties. AI4SoilHealth will use a pan-European soil monitoring system called LUCAS along with remote sensing and soil observations to improve soil health," says Lis W. de Jonge.

The expectation is that AI4SoilHealth will create:

• a method to calculate a soil health index
• a toolbox for rapid soil health assessment
• AI4SoilHealth data for the whole of Europe
• Soil health monitoring
• An AI4SoilHealth app

"Our goal with AI4SoilHealth is to create an effective soil health certification system that can be used by farmers, landowners and not least policy makers in the context of the new Green Deal for Europe," says Mogens H. Greve.