Skip to main content

Germany-wide spatio-temporal modelling of carbon contents of agricultural soils through an integrative analysis of satellite image time series and geodata (KoBoS)


Development of an extensible "open source" process chain for scale-specific prediction of topsoil carbon contents



  • Basic parameterisation of user-specific soil samples with multi-hierarchical relief attributes and multi-temporal soil reflectance composites
  • Modelling using machine learning methods
  • Characterisation of the modelling products with global accuracy and local uncertainty measures
  • Documentation of the geodata life cycle according to FAIR principles


Climate effect under consideration

Spatial and temporal variation of the carbon storage potential of agricultural soils

Contact persons

JKI, Institute for Crop and Soil Science

Dr. Markus Möller

Marvin Dierks

Dr. Younes Garosi


Link to project page:  KoBoS


Scroll to top