Zum Inhalt springen

Involved Groups and Research Projects

The Machine Learning for Hydrological and Earth Systems (ML4HES) group, led by Dr. Shijie Jiang and funded by the Carl Zeiss Foundation, is affiliated with the Max Planck Institute for Biogeochemistry. In light of growing environmental and climatic pressures, understanding the variability and causality of water, energy, and carbon cycles across spatial and temporal scales is crucial for effective resource management, ecosystem sustainability, extreme event mitigation, and climate resilience strategies. Our research focuses on integrating data and domain knowledge with hybrid and explainable machine learning methods to enhance the understanding of the interactions between climate, water, and ecosystems. The ultimate goal is to develop new tools and advance Earth sciences, guiding sustainable development and risk adaptation strategies under changing conditions. The group’s key research interests include terrestrial ecohydrological processes and interactions at different scales, the integration of physics and data with hybrid and explainable AI, coupling and feedback mechanisms between water, energy, and carbon cycles, the predictability, attribution, and impacts of climate extremes, and the influence of human activity on hydroclimatology and climate change impacts.
The primary aim of the Model-Data Integration (MDI) group, lead by Dr. Nuno Carvalhais and affiliated with the Max Planck Institute for Biogeochemistry, is to enhance our understanding of interactions in carbon and water dynamics within terrestrial ecosystems, focusing on their relationship to the broader Earth system. This is achieved by integrating observational data with modeling efforts to explore the variability and causality of these critical biogeochemical cycles. The group seeks to advance knowledge on how terrestrial ecosystems interact with climate, particularly in terms of the global carbon and water cycles. The MDI group emphasizes the development of methods to extract and transfer information from observational data to models, thereby investigating the dynamics and sensitivities of terrestrial ecosystems. By employing multiple-constraint model-data assimilation techniques that utilize in situ and Earth Observation data, the group aims to improve the representation of key processes that underpin uncertainties within the Earth system. The research spans several areas, including in situ dynamics of carbon and water fluxes to enhance parameter estimation and model selection, as well as examining terrestrial carbon and water dynamics across various scales. This includes the use of observation-driven methodologies to assess determinants of carbon sink/source strength and the coupling of carbon and water at ecosystem and watershed levels. Additionally, the group studies ecosystem carbon turnover times to diagnose the mechanisms governing land-atmosphere carbon cycle interactions. Their work breaks down ecosystem carbon estimates into components, investigating the influence of scale and environmental factors on carbon turnover dynamics. Through these efforts, the group aims to contribute significantly to Earth system science and improve future climate projections.

ABC3 group

The Atmosphere-Biosphere Coupling, Climate and Causality (ABC3) group led by Dr. Alexander Winkler, is affiliated with the Max Planck Institute for Biogeochemistry and focuses on identifying feedbacks and causal links in the exchange of carbon, water and energy fluxes between the terrestrial biosphere and atmosphere. In particular, we aim to understand how this coupled system changes under rising atmospheric CO2 concentration. We use models of varying complexity (conceptual models to fully coupled Earth system models) and combine our research with statistical analysis and modelling (machine learning to hybrid modelling) based on various Earth observation data streams.
For almost 20 years the Computer Vision Group Jena, led by Prof. Dr. Joachim Denzler from the Friedrich Schiller University Jena, has been developing methods and algorithms, which allow for automatic analysis of different kinds of sensor data including camera images, LiDAR scans and time-series data. Over the years, the group formed a particular interest in image understanding tasks like semantic segmentation and fine-grained classification, as well as in learning concepts like the life-long learning scenario. Additionally, there are always close co-operations with research institutes in Jena and partners from German industry. Currently, there is on the one hand a focus on developing computer vision and machine learning algorithms for different application areas. On the other hand, there are multiple projects centred around the topic of causal reasoning, which becomes increasingly relevant for building intelligent and explainable systems.
The GIScience Group, led by Prof. Dr. Alexander Brenning from the Friedrich Schiller University Jena is dedicated to advancing research related to the development and application of geographic information technologies and spatial models. Research of the GIScience Group focuses on environmental modeling using empirical and process-based models, the development of specialized software tools, and environmental monitoring. The developed methods and software are applied in study areas worldwide through international collaboration and in externally funded projects. Courses offered by the GIScience Group comprise training in GIS, GIScience and environmental statistics for undergraduate and graduate students of Geography as well as for the Master’s program in GIScience, which is jointly offered by the GIScience Group and the Department of Earth Observation.
The goal of the Causal Inference Group is to contribute to a data-driven understanding of complex dynamic processes. To achieve this, the group develops theory, methods and software in the field of causal inference for applications within and beyond German Aerospace Center (DLR). Causal inference, at the intersection of statistics and machine learning, is an active field of research that develops methods and algorithms for the data-driven derivation and analysis of cause-and-effect relationships. Causal inference thus goes beyond the analysis of mere statistical dependencies. In combination with machine learning, causal inference also has the potential to increase the robustness and generalisability of the learned models. In its research and development work, the group follows an application-driven approach. This means, firstly, close collaborations with the respective domain experts in the conception and implementation of applications for causal inference. Secondly, we identify the challenges that arise in these applications and address these challenges through targeted further development of our method portfolio. A special focus of the group is on time series data. Our methods are used in various areas of German Aerospace Center (DLR) and beyond. Examples include the analysis of causal relationships in the field of space weather as part of the PLASMA project and the explanation of anomalies as part of the CausalAnomalies project with a wide range of applications within German Aerospace Center (DLR), e.g. in the field of unmanned aerial vehicles and transport infrastructure.