Speaker
Description
Introduction
Climate change and other societal targets require strategic decisions in agriculture, at different levels of organisation and spatial context. Process-based simulations models support such decisions by making relevant processes graspable to the stakeholders that formulate a large range of different questions. Many of those questions require model simulations in a larger spatial and temporal context, which in turn generates a number of technical challenges for the model application, including the execution of a vast number of simulations in short time, and suppling relevant data to the models. This presentation gives an overview of where we currently stand in answering questions on crop yield predictions and projections and related carbon sequestration, nitrate leaching, greenhouse gas emissions and water consumption at large spatial and varying temporal scales.
I am using the MONICA agroecosystem model as an example to demonstrate different use cases for data being supplied through remote sensing and artificial intelligence, and the background of the CASSIS simulation infrastructure to run big-data projects. For the first, I will be highlighting several remote sensing activities that aim at supplying input information for (i) initialisation, (ii) driving and (iii) testing the model. These include satellite-based crop type identification (e.g. Blickensdörfer et al. 2022), detection of irrigation events (e.g. Ghazaryan et al., under review) and sowing dates (Main-Knorn et al., in preparation), and high-resolution information of soil properties and groundwater levels accessible to crops. For the second, I will be introducing the CASSIS simulation infrastructure and its unique security philosophy, implemented using Object Capabilities. The CASSIS simulation infrastructure (Berg-Mohnicke, M. and C. Nendel, 2022) supplies all required data for large-area simulations to the simulation model at a press of a button, which makes the application of models for regional to continental research questions much more comfortable.
Presented use cases include climate change outlooks on crop production for the German government, contributions to the greenhouse gas emission inventory of the Czech Republic, upscaling of rewetting scenarios for drained grassland and the quantification of irrigation water use for crop production in Brandenburg, Germany.
References
Berg-Mohnicke, M. and C. Nendel (2022): A case for object capabilities as the foundation of an environmental model and simulation infrastructure. Environ. Model. Softw. 156, Article 105471.
Blickensdörfer, L., M. Schwieder, D. Pflugmacher, C. Nendel, S. Erasmi and P. Hostert (2022): Multi-year national-scale crop type mapping with combined time series of Sentinel-1, Sentinel-2 and Landsat 8 data. Remote Sens. Environ. 269, Article 112831.
Ghazaryan, G., S. Ernst, F. Sempel, and C. Nendel (under review): Field-level irrigation mapping with integrated use of optical, thermal and radar time series in temperate regions. Int. J. Appl. Earth Obs. Geoinf.
Main-Knorn, M., L.A. Flores, G. Ghazaryan and C. Nendel (in preparation): Crop phenology assessment with multiscale and multisource time series. Remote Sens. Environ.
| Keywords | Simulation; Modelling; Remote sensing; Crop yields, Ecosystem services |
|---|
