You can see a grassy area against the backdrop of the Zugspitze and the Waxenstein. A large-scale scientific test set-up consisting of dynamic measuring chambers for climate gas measurements can be seen on the grass.

Module Q&S | Sources and Sinks of Greenhouse Gases
 

In order to accurately represent the sources and sinks of the greenhouse gases carbon dioxide (CO₂), nitrous oxide (N₂O) and methane (CH₄) using inverse modelling (Module M), the estimation of ‘a priori’ fluxes between the land surface and the atmosphere is of central importance, alongside observational data (Module B). The Sources & Sinks (S&S) module develops empirical, process-based and AI-driven modelling approaches to capture GHG fluxes across all sectors at high spatial and temporal resolution, together with a systematic assessment of uncertainties for Germany and the EU. The ongoing comparison of top-down inversion results with bottom-up Q&S modelling results contributes to the continuous refinement and improvement of national GHG reporting.  At the same time, it provides important, spatially and temporally explicit information that supports more targeted and effective mitigation of GHG emissions.

In Phase 2, the Q&S module establishes a harmonised ‘first-generation’ data stream, which is gradually integrated into operational use. In this context, agriculture, forestry, biogenic CO₂ exchange and Tier 3 approaches are refined and further developed through improved modelling approaches and input data for the National Inventory Report (NIR). A key focus of Phase 2 is the systematic identification, quantification and reduction of uncertainties – arising from input data, sampling frequency, model parameters and model structure – in order to ensure demonstrably higher accuracy for the inclusion of Tier 3 methods in the emissions inventories. Furthermore, the scope is being extended to the European domain: European greenhouse gas emission fields are being provided using consistent bottom-up approaches and EU input data.

Module Coordination Sources and Sinks (Q&S)

The Q&S module is coordinated by Ralf Kiese, Clemens Scheer and Hannes Imhof from KIT IMKIFU.

 

Phase 2 | 2026 - 2030

QS.201 | Temporal downscaling of greenhouse gas emissions from the industry, transport, energy and waste sectors (NFR sectors 1, 2, 5) for the national territory, taking account of uncertainties

An infographic by the German Environment Agency (Umweltbundesamt) displaying a map of Germany's CO2 emissions for 2021 using a 1x1 kilometer grid resolution (GRETA version 1.2.0.1). Emissions are color-coded in kilotons per square kilometer: low levels are shown in shades of blue, intermediate levels in yellow and orange, and high emission levels—particularly in urban areas and along major transport corridors—in vibrant red. A magnified section on the right side highlights the detailed spatial distribution of emissions in the Berlin-Brandenburg region.

High-quality, high-resolution a priori emissions data form an essential basis for Module M’s inversion system. In Phase 2, the Federal Environment Agency (UBA) and the Free University of Berlin (FUB) are working closely together to further refine the spatial and temporal regionalisation for the energy, industry, transport and waste sectors and to integrate reliable uncertainty estimates.

The UBA is focusing on further developing the spatial distribution of sector-specific activity and emission data, as well as their uncertainty assessment within the GRETA system.
The FUB is developing temporal uncertainty profiles, optimising the code of the open-source ARTEMIS model for operational use, and preparing a prototype module for meteorologically dependent time profiles in ICON-ART. The approaches developed are continuously coordinated with TeResE (UBA) and Module M (DWD) and will be evaluated at the end of Phase 2 in conjunction with Module V (V.201).

Further information:

Beteiligte Institutionen:

  • Umweltbundesamt (UBA)
  • FU Berlin (FUB)

Project participants:

 

 

 

QS.202 | Spatial downscaling of GHG emission uncertainties (N₂O, CO₂, CH₄) for the agriculture, land use, land-use change and forestry sectors (NFR sectors 3 & 4) in Germany

Eddy-covariance tower in a field of sunflowers

The uncertainty assessment of existing greenhouse gas inventories is insufficient for an accurate a priori estimate within the context of inverse modelling, as spatio-temporal correlations are largely ignored. This work package aims to establish improved uncertainty methods for the agriculture and land use sectors (NFR 3 & 4). To this end, the Thünen Institute for Agricultural Climate Protection is developing multivariate uncertainty approaches and Monte Carlo simulations for the inventory models, whilst the Thünen Institute for Rural Areas is processing spatio-temporal uncertainty data for the input data. In collaboration with the UBA, this refined information will be integrated into the ZSE and GRETA databases to provide Module M with significantly improved input data in terms of quality, which will be evaluated jointly with Module V at the end of Phase 2.

Participating institutions:

Project participants:

 

QS.203 | Process-based GHG emission estimates (N₂O, CO₂, CH₄) for agricultural soils, including advanced uncertainty quantification for Germany

The aim of this work package is to provide high-resolution greenhouse gas emission grids (approx. 1 x 1 km², sub-daily) for agricultural soils in Germany, as well as a comprehensive quantification of uncertainties. Building on the modelling framework developed in Phase 1 using LandscapeDNDC, improved activity and input data from the Thünen Institute will be integrated in Phase 2. The uncertainty analyses systematically take into account model parameters, input data and sampling errors, incorporating, amongst other things, TERENO and ICOS measurement data. To reduce latency, activity data such as crop rotations are projected into the future using machine learning, so that near-real-time data can be provided. The high-resolution data serve as a priori estimates for Module M, support the validation of N2O emissions in north-west Germany (WP-B.209) and, via the GRETA database (WP-Q&S.206), contribute to the evaluation of potential Tier 3 methods for national reporting.

The diagram illustrates the data flow of the LandscapeDNDC modeling system. It consists of four steps: First, heterogeneous spatial and temporal activity data (e.g., satellite imagery, sensors, soil samples) are collected. These data are then projected into a regular grid. Next, the model calculations are performed on high-performance computers (HPC). The result is a high-resolution map showing spatially and temporally detailed N₂O (nitrous oxide) emissions in kgN/ha/year, including validation and calibration steps.

Participating institutions:

Project participants:

 

 

QS.204 | Prozessbasierte THG-Emissionsschätzungen (N2O, CO2, CH4) für Waldökosysteme inklusive fortgeschrittener Unsicherheitsquantifizierung für Deutschland

This work package aims to provide high-resolution simulations (approx. 1 x 1 km², sub-daily) of greenhouse gas fluxes (CO₂, CH₄ and N₂O) in German forest ecosystems, together with a comprehensive quantification of uncertainties. Building on the methodological demonstrations from Phase 1, the KIT-IMK-IFU consortium is now integrating national remote-sensing and measurement data from the Thünen Institute on forest biomass and structure, as well as soil data and N deposition values from the UBA. Using the LandscapeDNDC process model, the forest’s net CO₂ exchange (separated into vegetation and soil carbon pools), the methane sink in soils, and indirect N₂O emissions from atmospheric nitrogen input are simulated at high resolution. The Thünen Institute also quantifies the uncertainties in biomass estimates and the impacts of future extreme events. The flux data obtained serve as a priori information for Module M and are incorporated into the comparison with the LULUCF figures from the National Inventory Report (NIR).

Participating institutions:

Project participants:

 

 

QS.205 | Operationalisation of terrestrial biogenic CO₂ fluxes (NEE)To complement the process-based approaches, the Max Planck Institute for Biogeochemistry is developing, in Phase 2, the operational provision of data-driven, AI-supported estimates of terrestrial biogenic CO₂ exchange (NEE) for use as a priori estimates for atmospheric inversions in Module M. The main objective is to achieve low latency (a few weeks to months) whilst ensuring increased accuracy and long-term consistency. To this end, automated protocols are being established for the flux simulation step and for plausibility checks. To make the system independent of individual satellite missions, a methodological approach is being developed that combines observational data from various missions within a single pipeline. The derived fluxes are continuously cross-checked against the process-based results from LandscapeDNDC, the NIR data and Module V. In addition, in collaboration with the UBA and the Thünen Institute, we are investigating the extent to which biogenic CO₂ fluxes can be consolidated and incorporated into the annual GRETA database.

Infographic on the estimation of biogenic CO₂ fluxes: It illustrates the process from local measurements at eddy covariance stations (such as Gebesee) and satellite data, through the training of machine learning algorithms, to spatiotemporal extrapolation, which produces high-resolution CO₂ flux maps (hourly, 0.05 degrees).

Participating institutions:

Project participants:

 

 

QS.206 | Adaptation and further development of GRETA and TeResE as a central system for consolidating GHG emission sectors

Following the successful demonstration in Phase 1, the GRETA and TeResE tools will be technically enhanced in Phase 2 to function as a central data provision system for ITMS. GRETA consolidates the high-resolution greenhouse gas emissions data, complete with associated uncertainties, from the various sectors and methodologies (WP-Q&S.201–205) into harmonised aggregate sectors for Module M. In doing so, it also implements the new reporting standards (Common Reporting Tables, CRT) under the Paris Agreement. Together with the temporal profiles from TeResE and ARTEMIS, the UBA ensures centralised data routing. Furthermore, GRETA serves as an interface for comparing bottom-up emission estimates with regional flux measurements in the catchment areas of selected ICOS observation towers (in collaboration with Module B).

Participating institutions:

Project participants:

  • Dr Christian Mielke (UBA)
  • Stefan Feigenspan (UBA)
  • Dr Theo Wernicke (UBA)
  • Dr Roland Fuß (TI)

 

 

QS.207 | Konsistente THG-Emissionsschätzungen (CH4, CO2, N2O) für die NFR-Sektoren 1 bis 5 für den europäischen Raum

A map of Germany showing how data from European countries has been combined with that from the National Greenhouse Gas Inventory.

As the transboundary transport of long-lived greenhouse gases plays a major role, this work package provides consistent, high-resolution a priori emission grids for the whole of Europe. The Free University of Berlin is extending its dynamic emissions model (ARTEMIS) to the European region in order to disaggregate NFR sectors 1, 2 and 5, as well as livestock farming, spatially and temporally, whilst taking uncertainties into account. In parallel, KIT IMKIFU is using LandscapeDNDC to produce methodologically consistent greenhouse gas emission grids for agricultural and forestry ecosystems in Europe, driven by high-quality activity data from the European Commission’s Joint Research Centre (JRC). The results fill gaps in existing European datasets (e.g. CAMS) and will be evaluated at the end of Phase 2 in collaboration with Module V and Module M.

Participating institutions: 

Project participants:

Phase 1 | Associated projects (Q&S_II) | 2024–2027

QS_II | Agri-For-Live: Modelling greenhouse gas emissions from agriculture, forestry and livestock farming in Germany

Logo of Agri-For-Live

The joint project Agri-For-Live associated with ITMS with the partners KIT IMK-IFU, Thünen Institute and FZJ Jülich deals with three central aspects of ITMS.

  • High temporal and spatial resolution of activity data from agriculture, forestry and land use (AFOLU),
  • Emission of greenhouse gases (GHG) from livestock farming
  • Biogeochemical modelling of GHG exchange processes in agricultural and forestry ecosystems to improve the spatial and temporal resolution of bottom-up GHG inventories.

 

For more information, please see the Agri-For-Live webpage.

Participating institutions:

QS_II | MODELPEAT: Modelling of greenhouse gas emissions from peatlands: process-based, streamlined GHG modelling

A four-part infographic on environmental and bog research.

Due to the heavy agricultural utilisation of peatlands, they have lost their original carbon storage effect, and CO2 is released on a large scale. This means that peatlands are a source of greenhouse gases that should not be underestimated and whose exact source and sink strength over the course of the year cannot yet be visualised on a large scale. However, this is of great importance for an integrated analysis such as that taking place within the framework of ITMS.

Through the further development of empirical and process-oriented modelling approaches, MODELPEAT is developing tools to provide ITMS with GHG emissions from peatland soils in unprecedented spatial and temporal resolution.

For more information, please see the MODELPEAT webpage.

 

Participating institution:

QS_II | ARTEMIS: Improving the spatio-temporal variability of Analysed and near Real-Time greenhouse gas EMISsions for North-Western Europe

A schematic diagram of a spatio-temporal dynamic emission model. It illustrates the modeling structure categorized into TIER 1, TIER 2, and TIER 3, alongside Gridder and Timer components. At the bottom, input data such as land use, calendar, activity data, and emission factors serve as the foundation. Various sectors—including landfills, domestic, agriculture, traffic, energy & industry, and transport—are integrated via dynamic parametrizations as well as technology- and sector-dependent profiles. Arrows connect the main model to external components such as Module M (ICON-LAM-ART) for 4D emissions, UBA Q&S_1 / B Source campaigns, and meteorology data from ICON-LAM/DE.

The joint project ARTEMIS associated to ITMS, which is carried out by the FU Berlin, aims to improve the understanding of the processes and mechanisms that cause the spatio-temporal variability of anthropogenic greenhouse gas emissions and to develop a modelling approach that provides high-resolution emission information for ICON-ART. ARTEMIS co-operates with the QS_I WP1 (UBA) to improve the spatial and temporal distribution of emissions from the National Inventory Report.

Special attention is paid to the inclusion of improved spatial grid methods and the parameterisation of the effects of changing meteorological conditions on emission variability in Germany for the sectors energy and industry, road transport, households, waste management and agriculture.

For more information, please see the ARTEMIS website.

Participating institution:

QS_II | Coast GEm: Greenhouse gas emissions from German coastal waters

A four-part graphic illustrating marine research in the North Sea. The top-left panel shows a person deploying a measurement device on a line into the sea from a vessel. The top-right and bottom-right panels depict a metal frame equipped with scientific instruments and cables positioned on the seafloor underwater. The bottom-left panel features a map of the German Bight displaying red measurement stations and route lines off the coasts of Emden, Bremerhaven, Bremen, Hamburg, and around Heligoland.

In the Coast-GEm project, the focus lies on the quantification of greenhouse gas sources and sinks in the coastal area of the German North Sea by using existing monitoring infrastructures in combination with cruises in the North Sea coastal zone. These are complemented by measurements in the mouth of the Elbe River, which is the largest source of nutrients to the German Bight andtehrefore has a major influence.

Coast-GEm is accompanied by the AGRIO project (2025 - 2027) funded by the Helmholtz community establishing the first continuous coupled hydrodynamic and biogeochemical model system by laterally and procedurally connecting existing models for the Elbe river-sea continuum describing all three GHG pathways.

For more information, please see the Coast GEm website.

Participating institution:

QS_II | TORCH: Towards operational uncertainty characterisation for high-resolution AI based biogenic CO2 fluxes

Scientific graphics for the Integrated Greenhouse Gas Monitoring System (ITMS). On the right is a map of Europe focusing on Germany, displaying the annual mean Net Ecosystem Exchange (NEE) in gC m⁻² d⁻¹ using a color scale. The top left graph illustrates the seasonal progression of NEE throughout the year. The bottom left shows the ITMS logo featuring a magnifying glass over a globe.

In order to realise an optimal integration with atmospheric inversions in ITMS-M, TORCH complements tasks in WP-Q&S_I.3 ("Biogenic Fluxes") regarding biogenic CO2 fluxes based on machine learning with respect to a thorough assessment of the uncertainties of the generated products.

For more information, please see the TORCH website.

Participating institution:

Phase 1 | 2022 - 2024 (completed)

Q&S.101 | GHG emissions in the industry, transport, energy and waste sectors

A map of Germany with Gridded Carbondioxide (CO2) Emission in Germany from the year 2021

The Federal Environment Agency is involved in Work Package Q&S.101 and provides spatio-temporally distributed anthropogenic emissions for the industry, transport, energy and waste sectors.

The Gridding Emission Tool for ArcGIS (Greta) is already used for the spatial distribution of air pollutants. Greta will be expanded to include greenhouse gases as part of the project. In addition, the temporal distribution of emissions will be realised by further developing the Temporal Resolution of Emission data (TeReSE) tool.

Further Information:

Project participants:

 

 

Q&S.102 | Greenhouse gas fluxes for the agriculture and forestry sectors

Eddy-covariance tower in a field of sunflowers

In work package Q&S.102, the Thünen Institute provides spatially downscaled emission fields of the greenhouse gases CO2, CH4 and N2O based on national UNFCCC reporting according to the "National Inventory Report", distinguishing between the different source categories such as livestock systems, direct soil emissions, indirect emissions and emissions due to land use change.

 

 

Participating institutions:

Project participants:

 

 

Q&S.103 | Biogenic CO2 Fluxes

A map of Europe displaying the Net Ecosystem Exchange RateDeutsch: Eine Karte von Teilen Nordwesteuropas visualisiert die „Net Ecosystem Exchange Rate“ (Netto-Ökosystem-Austauschrate) in Rosatönen, mit einer seitlichen Farbskala, die Messwerte zwischen -5 und 10 $\text{µmol m}^{-2}\text{ s}^{-1}$ anzeigt. Im unteren Bereich befindet sich ein grüner Banner mit der Aufschrift „Net Ecosystem Exchange Rate“ sowie ITMS- und ICOS-Logos.

Work package Q&S.103 aims to develop a flexible and operational workflow for estimating hourly biogenic CO2 fluxes. These estimates will be performed with high spatial and temporal resolution based on the integration of machine learning, satellite data, measurements of CO2 fluxes, and supplementary parameters at eddy covariance sites in Germany. This will serve as the basis for the inversion activities of the ITMS-M module. This work draws on the MPI-BGC group's many years of experience in connection with the FLUXCOM-Initiative.

Participating institution:

Project participants:

 

 

Q&S.104 | Biogeochemische Modellierung des Treibhausgasaustauschs in Ökosystemen

original

In work package Q&S.104, the biogeochemical process model LandscapeDNDC is used to determine the sources and sink strength of agricultural areas (grassland and arable land) and forest areas. Data and maps from the Thünen Institute and remote sensing data sets are used as input data (WP2). Results from the existing ICOS and TERENO networks are used for calibration and evaluation. The a priori estimates obtained with high spatial and temporal resolution are used directly for the inverse modelling of Module M.

Participating institution:

Project participants:

 

Web-View
Print Page
Open in new window
Estimated DIN-A4 page-width
Go to Editor View