Prof. Dr. Mareike Ließ

The success of structural changes within our agricultural landscapes in terms of sustainable, resource-conserving cultivation strategies depends to a large extent on knowledge of the soil landscape. Systemic process modelling of plant cultivation measures can only provide reliable forecasts for political decision-makers on the basis of site-specific, spatially continuous soil information. This includes parameter-specific modelling as well as the agglomeration of the multivariate soil parameter space into functional spatial process units. In cooperation with the Thünen Institute of Climate-Smart Agriculture, the multivariate 3D soil parameter space of the German agricultural landscape was modelled. Methods from the field of 'machine learning' and 'deep learning' were used as well as data science approaches that deal with similarity analyses involving data of mixed variable types. (in-house UFZ & Thünen, 2019 - 2023)

Adverse changes in Agriculture management can lead to a faster release of soil organic carbon (Corg) into the atmosphere compared to the time required for its sequestration. Therefore, long-term Corg-promoting management is essential. In the SOCmonit project, soil spectroscopy was used to develop a spatio-temporal monitoring method for Corg. The remote sensing and proximal soil sensing sensors and carrier platforms used in the project generate different data formats with diverse information content in relation to the target variable. Empirical modelling was then applied to relate the signal from the sensors to Corg. The various data processing and modelling steps involved were implemented, automated and documented in a comprehensible manner in an open source software toolbox. The project work was carried out in cooperation with the company RSS - Remote Sensing Solutions GmbH. (BMEL, 2018 - 2022)

Helmholtz Climate Initiative. In interdisciplinary collaboration between a wide range of Helmholtz centres and external partners, research expertise was pooled in order to develop recommendations for action for decision-makers in politics and society in the context of climate change. In the "Agriculture and Aquatic Systems" project, the selected modelling approach was used to link soil process understanding to national soil databases in order to generate spatially high-resolution soil information. The focus was on soil properties that are considered suitable indicators for Agriculture productivity - especially water storage. The project has contributed to the development of sustainable adaptation strategies for Agriculture in response to climate change. This was done at national level (Germany), taking into account regional characteristics. (Helmholtz, 2019-2021)

Towards a guideline for digital soil mapping in Ecuador. Soil landscapes with different climates, vegetation and soils were investigated in Ecuador. The soil data collection was followed by pedometric modelling of spatial soil parameter fields to assess soil hydrology and carbon storage. In addition to methodological developments such as the adaptation of statistical sampling methods to limitations in terms of accessibility, feasibility and costs as well as model adaptation through mathematical optimisation, site-specific soil hydrological pedotransfer functions were developed in order to increase the comparatively small water retention data set and thus improve the prediction performance. The knowledge transfer also anchored in the project included the description and publication of the methodology in an easily understandable language as well as teaching in computer courses. (DFG, 2013-2018)

Landform classification for functional upscaling. Site characteristics such as nutrient availability or drainage determine forest growth and management as well as Agriculture production. For the watershed of Lake Soyang in South Korea, models for the spatially continuous prediction of important chemical soil parameters were developed to evaluate ecosystem services and to provide appropriate input variables for environmental models such as LandClim, MMF, PIXGRO and SWAT. (DFG-IRTG/ University of Bayreuth, 2012-2015)

Spatial analysis of soil related environmental risks in the Soyang watershed. The aim of the project was to develop models for predicting soil physical parameter fields in order to analyse the significance of the investigated soil properties with regard to soil erosion, landslides and landslide sources and to estimate the sediment input into Soyang Lake. (DFG-IRTG, 2012-2014)

Soil as a sustainable resource for the bioeconomy. The fertility of soils forms the basis for sustainable plant production. Soils must not only guarantee marketable yields, but also provide a wide range of ecosystem services. Ten interdisciplinary project networks and the BonaRes Centre for Soil Research conduct research under the umbrella of BonaRes. The BonaRes Centre is concerned with the transfer of existing knowledge about soil functions into decision-making aids for soil management. Its main tasks are to establish the necessary data infrastructures, modelling concepts and assessment tools for sustainable soil management. (BMBF, staff 2016 - 2022)

Functional soil landscape modelling in the Andean mountain forest zone: impact of land use and natural disturbances. The project investigated soil functional aspects in the context of soil-borne landslide risk and different land uses in Ecuador. The project is a continuation of the work from FOR 816/A3. (DFG, 2010 - 2013)

Spatiotemporal dynamics of shallow landslides and their biotic and abiotic controls - an integrating synthesis. Landslides are assumed to have a significant impact on pedogenesis, landscape evolution and biodiversity in the tropical montane forests of southern Ecuador. The spatial modelling of the soil landscape and the resulting assessment of the pedogenic landslide risk due to soil mechanical failure caused by heavy rainfall and soil water accumulation shed light on a key aspect of this complex topic. (DFG, 2007 - 2010)