Universität Bonn

INRES Crop Science

Research

Impact assessment using crop models in response to management, environmental conditions, and climate change.

We use experiments, analyze data and develop and apply crop models to advance our understanding of growth and yield of crops across scales in response to climate change and management to support the development of sustainable crop production systems. Our main research topics are organized in four clusters.   

Cluster 1 - Data driven modeling

Uncertainties in model simulations can be related to input data, model parameters and/or model algorithms. Model uncertainties are still high for quantifying ecosystem services like reduction of greenhouse gas emissions or nitrate leaching based on complex modelling solutions at the sub-field scale. Observations from Unmanned Aerial Vehicles (UAV), tractor or satellite based sensors with high spatial and temporal resolution together with ensemble simulations have the potential to reduce model uncertainties at the sub-field scale, thus supporting spatially explicit field operations for farmers. In addition, the highly resolved crop information can assist in improving model calibration for specific site conditions. The objectives of Cluster 1 are (1) to improve analysis and interpretation of remote sensing information for the purpose of calibration of agro-ecosystem and grassland models and (2) exploring the potential of data assimilation to reduce uncertainties of sub-field scale model simulations for estimating ecosystem services and resource use efficiency. Read more ...

Cluster 2 - Diversified cropping systems

Diverse cropping systems offer multiple advantages over conventional sole crop monocultures, including higher crop yields, complementary resource use in time and space among different species, improved weed suppression, and increased biodiversity. Crop diversification can have a temporal (e.g. catch crops, wide crop rotations) or a spatial scale (e.g. crop mixtures, patchy fields on heterogeneous soils), or on both (agroforestry).

The Crop Science Group develops and applies crop growth models not only on conventional but also on diverse cropping systems including crop mixtures, large patchy fields, grassland, and agroforestry systems. Our aim is to advance modelling of a large range of diverse cropping systems including the modelling of crop and root growth, yield, resource use (water, nutrients) and ecosystem services. Best management practices and optimal spatio-temporal field arrangements for diverse cropping systems mixtures (e.g. crop species partners and proportions, row distances, seeding densities) under varying climatic and soil conditions will be identified. Furthermore, the group works on pest and disease modelling. Read more ...

Cluster 3 - Crop physiological processes

We develop and improve crop models by improving the representation of selected physiological processes to better characterize plant-environment relationships. The main focus of this cluster is on root and shoot growth, and root:shoot interactions under different nutrient applications, soil types, and soil moisture conditions for a wide range of crop species and cropping systems (conventional and diversified cropping systems). Another focus is on further improving the understanding of stomatal functions and gas fluxes exchange (photosynthesis and transpiration) as well as long-term structural changes of the canopy for cultivars of different crops such as wheat and maize in responses to drought and elevated ozone. Improved representation and calibration of leaf photosynthesis, leaf and root traits, and sink:source relationships will be considered in various modelling subroutines and advance the root:shoot model that can be used at different scales and for different cropping systems. Read more ...

Cluster T - Technical tool development

To support modelling and data analysis our group develops various technical tools. Main emphasis is on the development of the simulation framework Simplace and the implementation and impro­vement of existing models as well as the development of new models within this framework. For greater flexibility, Simplace can deal with different input and output data formats and offers interfaces to common data science languages (R, Python, Matlab, Julia). This enables us to couple Sim­place models with other frameworks or combine them with sensitivity analysis, calibration or data assimilation procedures. We provide scripts and toolchains to automatize transformation and analysis of experimental and model data from plot scale up to global scale. We also offer training seminars to improve the technical skills of the group members and other interested users and increase our knowledge of the technical and mathematical aspects of crop modelling and data analysis. Read more ...

Research projects

Wird geladen