Current Projects

Duration: 01/2023 - 12/2027
Funding: EU - HORIZON EUROPE (grant ID 101086179)
PI / Co-PIs: Bernhard Ahrens, Melanie Weynants
The objective of AI4SoilHealth is to co-design, create and maintain an open access European-wide digital infrastructure, compiled using state-of-the-art Artificial Intelligence (AI) methods combined with new and deep soil health understanding and measures. The AI-based data infrastructure functions as a Digital Twin to the real-World biophysical system, forming a Soil Digital Twin. This can be used for assessing and continuously monitoring Soil Health metrics by land use and/or management parcel, supporting the Commission’s objective of transitioning towards healthy soils by 2030.

Duration: 04/2024 - 03/2028
Funding: EU - HORIZON EUROPE (grant ID GA 101137682)
PI / Co-PIs: Nuno Carvalhais, Markus Reichstein

Participating department members: Alexander Winkler, Ana Bastos, Bernhard Ahrens, Christian Reimers, Greg Duveiller, Jacob Nelson, Jeran Poehls, Lazaro Alonso, Martin Jung, Nicole Börner, Sujan Koirala, Markus Reichstein more

Duration: 04/2026 - 03/2031
Funding: Carl Zeiss Foundation
PI/Co-PIs: Markus Reichstein, Nuno Carvalhais, Shijie Jiang, Alexander Winkler

GENAI-X addresses a fundamental AI challenge: achieving robust model generalizability in non-stationary environmental systems, where conditions vary across space and evolve unpredictably. Focusing on hydro-climatic extremes (floods, landslides, droughts and late-frost events) and their impacts, we advance AI methods that adapt to shifting data patterns and uncertainties. We integrate hybrid modeling, causal modeling, equation discovery, dimension reduction and uncertainty quantification to improve environmental understanding and prediction. Beyond theory, we develop AI-driven tools for assessing risks, interpreting environmental changes, and supporting decisions.

Participating department members: Markus Reichstein, Nuno Carvalhais, Shijie Jiang, Alexander Winkler more

Duration: 07/2024 - 06/2026
Funding: ESA, 4000144482/24/I-KE
PI/Co-PI’s: Fabian Gans

Participating department members: Fabian Gans, Melanie Weynants more

Duration: 03/2024 - 02/2027
Funding: BMBF (Projekt GZ: 16LW0499)
PI / Co-PI’s: Gregory Duveiller

Expected strong climate, demographic and economic changes in Central Africa threaten the sustainability of ecological, social and economic services Congo Basin Forests (CBF), the second World’s largest rainforest, provide to humanity. In addition to deforestation, anthropogenic environmental impacts will lead to dramatic changes in forest tree functional composition with potential deleterious feedbacks on carbon and water cycles among other services. Through a unique European research partnership and transnational collaborations with Central african country experts, CoForFunc aims at developing an integrated approach for the monitoring of tree functional diversity of the CBF to support biome-scale assessment of their vulnerability to expected climate change and human-induced transformations.

Participating department members: more

Duration: 05/2014 - 12/2026
PI/Co-PIs: Marion Schrumpf, Susan Trumbore

The project provides essential soil characteristics that determine ecosystem properties such as plant and soil organism abundance, diversity, and productivity under different land use and management, and information on soil processes driving belowground carbon (C) and nutrient cycles, to link management, biodiversity and soil function.

Participating department members: Marion Schrumpf more

PI /(Co)-PIs: Ana Bastos (now with University of Leipzig)

The fate of carbon emissions from fossil fuel burning and deforestation plays a critical role in determining atmospheric CO2 levels and, consequently, climate change. Earth System Models (ESMs) project a non-linear response of the natural carbon cycle, which weakens both land and ocean sinks in response to warming. This leads to a steeper rise in atmospheric CO2, but the magnitude of this amplifying carbon-climate feedback varies significantly among current ESMs, hindering accurate climate projection. CALIPSO’s vision is to drive scientific progress, empower better climate projections, and provide essential insights into the complex relationship between climate change and carbon loss. Join us on this journey towards a sustainable and climate-resilient future.

The project has been moved to the University of Leipzig. more

Duration: 04/2024 - 03/2028
PI / (Co)-PIs: Sung-Ching Lee

 C-BLUES aims to advance knowledge and understanding of blue carbon ecosystems (BCEs) such as seagrasses, tidal marshes, mangroves, macroalgae, and macroalgae mariculture via targeting three overarching objectives: 1) develop new scientific knowledge within BCEs to reduce scientific uncertainty and improve reporting of blue carbon under UNFCCC, 2) provide input to a possible revision of the 2013 IPCC Wetlands Supplement to increase inclusion of BCEs in national greenhouse gas inventories and reporting, 3) raise awareness and promote the role of BCEs for delivering global climate policy commitments in collaboration with Chinese and other international partners.

Participating department members: Sung-Ching Lee more

Duration: 01.10.2025 - 31.09.2028
Funding: German Aerospace Center (DLR)
PI/Co-PIs: Nuno Carvalhais
Cooperation with: GFZ Potsdam, UFZ, TU Dresden, DLR, Wilderness International, University of Maryland

The DE:Cal/Val project develops robust calibration and validation methodologies for satellite-derived above-ground biomass estimates from ESA's BIOMASS mission. Our approach combines multi-source datasets including in-situ measurements, airborne and spaceborne observations (TanDEM-X, GEDI), and terrestrial laser scanning (TLS) to characterize forest structure at multiple scales. This comprehensive Cal/Val framework enables accurate assessment of BIOMASS products and enhances our understanding of carbon cycling processes, ultimately supporting improved monitoring of forest carbon and water cycles under global change conditions.

Participating department members: Nuno Carvalhais, Nicole Börner, Xu Shan, Siyuan Wang more

Duration: 06/2026 - 05/2030
Financing: EU-Horizon Europe (GA 101294938)

EarthGenerator is a four-year Horizon Europe project (June 2026 – May 2030) developing a next-generation Foundation Model of the Earth system, building on and extending the WeatherGenerator. By adopting a generative AI approach, EarthGenerator integrates atmospheric, ocean, and land processes at multiple scales and modalities within a single, unified and physically consistent framework. Following the foundation-model paradigm, the model is pre-trained once and then adapts to many downstream tasks with minimal additional training, so that knowledge learned in one context can be applied to another without retraining from scratch.

Participating department members: Vitus Benson, Markus Reichstein

Duration: 07/2024 - 06/2026
PI/Co-PIs: Sujan Koirala, Jacob Nelson

EO-LINCS promotes a community effort towards an enhanced multi-mission assessment of the terrestrial carbon cycle at resolutions in space and time compatible with decision making by improving the access to the Earth Observation (EO) data for the wider carbon scientific community. This way the key questions related to scale, representativeness, consistency, reliability, as well as the applicability of the multivariate EO data and how they affect our understanding of the carbon cycle processes across spatial and temporal scales can be addressed.

Participating department members: Sujan Koirala, Jacob Nelson, Xu Shan more

Duration: 09/2023 - 08/2027
Funding: EU - HORIZON EUROPE (grant ID GA 101120237)
PI / Co-PIs: Nuno Carvalhais, Markus Reichstein

ELIAS aims at establishing Europe as a leader in Artificial Intelligence (AI) research that drives sustainable innovation and economic development.
We will create a Network of Excellence connecting researchers in academia with practitioners in the industry to differentiate Europe as a region where AI research builds towards a sustainable long-term future for our planet, contributes to a cohesive society, and respects individual preferences and rights. The BGI is part of UC5.

Participating department members: Claire Robin, Markus Reichstein more

Duration: 01/2023 - 12/2027
Funding: Carl-Zeiss-Stiftung
PI/ Co-PI’s: Markus Reichstein

The research project focuses on the integration of data and knowledge with hybrid and explainable machine learning methods to derive new insights in environmental and Earth sciences, e.g. for hydroclimatic extremes and their impact on ecosystem functions or services.

Participating department members: Shijie Jiang, Georgios Blougouras more

Duration: 01/2021 - 12/2026
PI/Co-PIs: Markus Reichstein,  Nuno Carvlahais

The ELLIS Unit Jena is a member of the ”European Laboratory for Learning and Intelligent Systems” (ELLIS) Network, and is at the forefront of using artificial intelligence (AI) to tackle global environmental crises, like climate change and biodiversity loss. By using machine learning (ML) and AI, ELLIS researchers gain a deeper understanding of the interdependencies of Earth and climate. They are able to simulate scenarios for mitigating and adapting to the impacts of climate change.

Partner: Friedrich Schiller University Jena, Max Planck Institute for Biogeochemistry, German Aerospace Center
Funding: Carl-Zeiss-Stiftung, Michael Stifel Center Jena

Participating department members: Markus Reichstein, Nuno Carvalhais, Vitus Benson, Conrad Philipp (FSU), Shijie Jiang more

Duration: 8/2024 - 7/2026
PI/Co-PI’s: Jana Wäldchen
Funding: BfN, Project: 352460010B

Methods for closing data gaps in nationwide biodiversity monitoring

Participating department members: Jana Wälchen, Michael Rzanny, Susanne Tautenhahn, Anke Bebber, Alice Fritz more

Duration: 09/2022 - 08/2027
PI/Co-PIs: Ana Bastos (now with University of Leipzig)

Forest vulnerability to compound extremes and disturbances in a changing climate

Participating department members: Ana Bastos, Yimian Ma, Franziska Müller

The project has been moved to the University of Leipzig.

Duration: 03/2026 - 02/2031
PI/Co-PIs: Alexander Winkler
Funding: ERC-Starting Grant (ID: 101222996)

PostPeak tackles the pressing questions surrounding Earth's transition into a post-peak CO2 era. The carbon cycle determines how much of the anthropogenic carbon stays in the atmosphere, and warms the planet. It will also shape the trajectory of atmospheric CO2 once emissions level off, with lingering legacy and hysteresis effects.
I hypothesize the land carbon cycle is the key modulator in this transition, and accurately predicting its dynamics is essential for understanding how the entire climate system will respond to declining CO2 emissions. Despite decades of research, the myriad factors underlying land carbon dynamics remain poorly understood, and uncertainty prevails.

Participating department members: Alexander Winkler

Duration: 08/2022 - 07/2026
PI/Co-PIs: Gregory Duveiller, Sophia Walther

Joint project ITMS Integrated Greenhouse Gas Monitoring System for Germany - Module B (demonstration phase) more

Duration: 08/2022 - 07/2026
PI/Co-PIs: Gregory Duveiller, Sophia Walther

Joint project ITMS Integrated Greenhouse Gas Monitoring System for Germany - Module QS (demonstration phase) more

Duration: 12/2023 - 11/2026
Funding: BMBF
PI/Co-PIs: Sophia Walther, Martin Jung

Joint project ITMS Integrated Greenhouse Gas Monitoring System for Germany - Characterization of uncertainty for high-resolution AI-based biogenic CO2 fluxes. more

Duration: 04/2023 - 12/2027
Funding: Max Planck Foundation
PI/ Co-PIs: Markus Reichstein

The Max-Planck Caltech Carnegie Columbia (MC³) Center aims to transform Earth system analysis and predictions, integrating advanced Earth observations and machine learning for deeper understanding and sustainable management of our planet, in particular the land. Our mission is to set new standards in Earth system modelling, foster next-generation scientists, and lead in earth sciences research and education, shaping a sustainable future through innovative, interdisciplinary collaboration.

Participating department members: Markus Reichstein, Alexander Winkler, Nuno Carvalhais, Dora Kelemen more

Duration: 09/2024 - 02/2027
PI/ Co-PI: Nuno Carvalhais
Funding: EU Horizon EUROPE (grant ID GA 101159723)

Short description: The project brings together AI experts to develop the Mediterranean Digital Twin network, and enhance the understanding and forecasting of climate extremes in the region

Participating department members: Nuno Carvalhais more

Duration: 1/2025 - 12/2029
PI / Co-PI’s: Martin Jung, Greg Duveiller, Nuno Carvalhais, Markus Reichstein
Funding: EU - Horizon Europe, Grant ID: GA 101184989

Short description:

Participating department members: Martin Jung, Greg Duveiller, Nuno Carvalhais,

Duration: 10/2021 - 10/2026
PI/Co-PIs: Markus Reichstein, Fabian Gans

The goal of the NFDI4Earth is to provide simple, efficient and open access to all relevant Earth System Data. The NFDI4Earth aims to establish common principles, rules and standards for research data management in Earth system sciences. Major implementation guidelines are the FAIR principles, which do impact the whole research data life cycle.

Participating department members: Markus Reichstein, Fabian Gans more

Duration: 06/2022 - 05/2026
PI/Co-PIs: Gregory Duveiller

The mission of Open-Earth-Monitor is to accelerate uptake of environmental information to guide current and future users in research, decision-making and citizens toward the most sustainable solutions.

Participating department members: Fabian Gans, Markus Reichstein and Jacob Nelson more

Duration: 04/2024 - 03/2030
Funding: Carl-Zeiss-Foundation
PI / CO-PI: Jana Wäldchen

The project “PollenNet” aims to use artificial intelligence and community science to accurately predict the spread of pollen.

Participating department members: Jana Wäldchen, Michael Rzanny, Anke Bebber

Duration: 1.4.2026 - 31.4.2028
Funding: Daimler and Benz Foundation
PI / Co-PIs: Sung-Ching Lee

Nature-based climate solutions are a key supplement to accelerating climate mitigation, yet we lack critical knowledge about how they can be effectively implemented in dryland and coastal ecosystems, which harbor more than half of the world's population, store vast carbon, and safeguard high biodiversity. This project addresses these gaps using advanced monitoring and modeling techniques to assess the effects of diverse water inputs and salinity changes on ecosystem metabolism of drylands and coastal areas, respectively. Establishing this scientific foundation is essential for implementing impactful and sustainable nature-based climate solutions in the two ecosystems, fostering more reliable climate mitigation strategies implemented worldwide.

Participating department members: Sung-Ching Lee, Laura Nadolski

Duration: 09/2020 - 08/2026
PI/Co-PIs: Markus Reichstein

The EU-funded USMILE project will use machine learning to improve modelling and understanding of the Earth system.

Participating department members: Markus Reichstein, Alexander Winkler, Reda El Ghawi more

Duration: 01/2024 - 12/2026
Funding: BMBF (PT Jülich) (Grant ID: 03WIR3605A)
PI / Co-PI: Jana Wäldchen

TP1: Basics of image-based automatic damage detection and citizen science. The aim of the project is to develop an AI-powered method for the automatic detection of forest pests to support forestry and landowners in identifying threats early and responding swiftly to infestations. Another key focus is the development of educational concepts and communication strategies to raise public awareness about forest health and damage.

participating department members: Jana Wäldchen, Anke Bebber more

Duration: 2/2025 - 1/2029
Funding: EU - Horizon Europe (Grant ID: GA 101187947)
PI / Co-PI: Vitus Benson, Markus Reichstein, Nuno Carvalhais

WeatherGenerator which aims to use machine learning in novel ways for weather forecasting and to model related Earth system processes. The role of MPI-BGC is to model the land ecosystems in response to weather.

Participating department members: Nuno Carvalhais, Vitus Benson, Sebastian Hoffmann, Melanie Weynants, Markus Zehner, Claire Robin, Lazaro Alonso more