header_bild

soil4climate

Small-scale retrieval of relevant soil parameters by a process-integrated soil sensor system and AI-based data evaluation as the basis for a climate-resilient, small-scale-specific crop production system.


Term

2023-10-01 bis 2026-09-30

Project management

  • Heike, Gerighausen


Responsible institute

Institut für Pflanzenbau und Bodenkunde


Cooperation partner

  • Hochschule Osnabrück
  • Field-Expert GmbH
  • AMAZONEN – WERKE H. Dreyer SE & Co. KG
  • Exatrek – EXA Computing GmbH


Overall objective of the project

The consequences of climate change, e.g. heavy rainfall events, heat periods, drought and dryness have negative effects on crop production. Thus, it is necessary to mitigate these negative consequences of climate change in crop production through climate-adapted production methods. For farmers, it is important to take measures to make crop production climate-resilient. Possible measures are variable sowing depth, sowing intensity, choice of varieties (drought-tolerant or high-yielding varieties), fertilization intensity, plant protection strategy or alternative site specific use. For the implementation of such measures, practical, small-scale and crop-relevant soil information is needed to support the farmer in a decision-making process for climate adaptation strategies. The challenge for farmers is to obtain and apply this small-scale and crop-relevant soil information. In soil4climate, such small-scale soil information is retrieved and translated to practice-relevant digital soil maps for cultivation. For this purpose, a geoelectric measuring system is implemented in a standard cultivator with the aim of mapping very small-scale differences during the working process (and thus without additional effort for the farmer). In addition to this data source, other relevant data sources (e.g. tractor, drone and satellite data) is explored to identify homogeneous small-scale structures in the soil. The generation of the soil maps will be both automatic, AI-assisted, and manual, allowing for an evaluation of the AI-assisted map generation. This evaluation is carried out through field tests as well as laboratory experiments in interaction with practitioners and experts in the project.


Funder

Federal Ministry of Food and Agriculture