Valuation of Peatlands as CO₂ Stores

In VALPEATS, available input data (e.g., mapping data, scientific studies, meteorological data) are combined with field survey data (e.g., drone imagery and water level measurements), analyzed and classified using AI. This ultimately enables a polygon-based delineation of prevailing water levels and vegetation types (e.g., vegetation communities or indicator species) in previously uncharacterized areas. As a result, regular condition monitoring of specific peatland areas becomes possible, providing the necessary data foundation for applications such as CO₂ certification schemes and smart farming.

To quantify the climate impacts of peatlands, the GEST model (Couwenberg et al., 2008, 2011; Joosten et al., 2013) is frequently used. GESTs (Greenhouse Gas Emission Site Types) represent site types that are defined as homogeneous with respect to their greenhouse gas (GHG) emissions and are characterized by typical vegetation communities. Vegetation serves as a proxy for estimating the water level in peatlands. Since water level is the primary factor influencing emissions, CO₂ emissions can be estimated on this basis. The challenges associated with determining GESTs include time requirements, personnel needs (specialized botanical training), spatial transitions (uncertainties), and limited scalability. From a digitalized solution for vegetation assessment, we expect:

• High-frequency monitoring for quantifying emission reductions
• Greater accuracy in vegetation mapping
• Improved scalability (larger areas covered with the same resources)
• Deployment of personnel without botanical expertise
• Consistent monitoring quality

Resource-efficient monitoring for financial products based on GHG reductions (CO₂ certificates) or biodiversity gains (e.g., agri-environmental and climate measures (AECMs), eco-points, CSRD reporting, etc.) enables transparent reporting of relevant changes. An automated digital solution also provides the foundation for monitoring the scientifically and politically targeted goal of restoring an additional 50,000 hectares of rewetted peatlands annually.

Section: Technological Development

A range of technological activities and developments are being carried out within VALPEATS:

• Drones equipped with multispectral cameras capture high-resolution images of peatlands.
• Botanical RTK GNSS surveys provide ground-truth data.
• Deep learning models identify vegetation species based on spectral signatures as well as shape and texture characteristics.
• DSMs/DTMs and indices (e.g., NDVI), together with phenological stage data, improve classification accuracy.
• In situ water level sensors measure groundwater table depth; IoT gateways automatically transmit the data.
• Hydrological data processing enables the spatial scaling of measurements and supports (near) real-time monitoring.
• Integration of data streams into an online platform, with visualization through dashboards and APIs.

With the term paludiculture (from the Latin palus – wetland), the Greifswald Mire Centre coined an umbrella term around 25 years ago for a wide range of cultivation techniques that can be practiced on wet peatlands. Examples include the cultivation of cattails (Typha) or reed (Phragmites) on fen peatlands as raw materials for insulation and construction products, as well as the cultivation of Sphagnum mosses as a substitute for peat in growing media production.

The establishment of paludiculture, the development of biomass value chains based on it, and the valorization of peatland ecosystem services are considered key factors for accelerating peatland rewetting.

The development of the components “vegetation recognition” and “water level monitoring and hydrological modeling” opens up additional opportunities that can support the establishment of paludiculture:

• The majority of paludiculture areas will likely be managed as successional paludiculture. This means that, without the targeted cultivation of specific plant species and solely through raising the water table, succession processes lead to the establishment of new peatland-typical vegetation communities. Experience shows that particularly during the first ten years, heterogeneous vegetation stands develop that are difficult to utilize efficiently. Vegetation classification provides information on the quality and composition of harvested biomass bales. This information can, for example, be used to optimize the operation of biomass-fired power plants or to determine whether a bale is better suited for energy production, animal feed, or material use.
• By combining knowledge of vegetation development with hydrological information, both successional paludiculture and cultivated paludiculture systems can be monitored to assess how vegetation stands develop and where targeted interventions are needed to optimize yields.
• The development of value chains is essential for the successful establishment of paludiculture. Information on the type and occurrence of paludiculture biomass serves as a decision-making basis for processing industries when selecting locations for their production facilities.

The paludiculture-related preliminary work carried out within VALPEATS is now being further developed in the KIMoPa project.