Rising greenhouse gas (GHG) emissions and accelerating climate change underscore the urgent need to strengthen carbon sinks in terrestrial ecosystems. Peat-rich wet grasslands are recognized as important carbon reservoirs, but decades of drainage and land-use change have often converted them into net sources of CO2. Restoring wet grasslands by raising water tables can reinstate carbon storage capacity, yet these hydrological changes also reshape soil moisture and nutrient dynamics in complex ways. Properly quantifying the net climate effect of rewetting therefore requires detailed understanding of these coupled carbon–nitrogen–water cycles. Mechanistic ecosystem models are powerful tools for predicting biogeochemical responses, but current approaches have notable gaps in the context of wet grasslands. Three principal knowledge gaps were identified: (1) Carbon, nitrogen, and water processes are typically modelled separately, and no existing model integrates all three holistically for wet grasslands. (2) Models rarely span the full spectrum of water table conditions from fully flooded to near-dry soils, omitting important saturated–unsaturated transitions. (3) Most process-based agroecosystem models treat vegetation as static and fail to account for observed shifts in plant communities as soil moisture changes. Because most existing models are designed for arable or upland systems with fixed vegetation and minimal waterlogging, they have limited applicability to wet grasslands. As a result, predictions of carbon and greenhouse gas dynamics under wetland restoration have remained highly uncertain. To address these gaps, a process-based agroecosystem model that explicitly couples hydrology (dynamic water table) with soil carbon and nitrogen cycling and vegetation growth was evaluated. Key features include oxygen-sensitive decomposition, nitrogen feedback, and water-dependent plant productivity, enabling simulation of CO2 and nitrogen fluxes under variable moisture conditions. The model was calibrated and tested using data from a long-term lysimeter experiment in the Spreewald region of Brandenburg, Germany, where water-table levels were systematically manipulated from waterlogged to nearly drained conditions. This experimental setup provided continuous measurements of soil carbon and nitrogen fluxes under each water table regime, which was used to calibrate and validate the model. Then this calibrated model was applied at the landscape scale across the entire wet grasslands of Brandenburg, simulating hydrological and biogeochemical behaviour across diverse climate conditions. Our results provide new insights into wetland biogeochemistry under rewetting scenarios. Lowering the water table (drainage) significantly enhanced aerobic decomposition and CO2 emissions, reducing net carbon storage, whereas elevated water tables slowed decomposition and increased carbon accumulation in vegetation and soils. These hydrological shifts also induced pronounced changes in plant communities over time: dry conditions favoured drought-tolerant grass species, while consistently moist conditions supported sedge-grass communities associated with peat formation. Across years, lower water tables reduced canopy development and GPP and, together with higher heterotrophic respiration, weakened net CO₂ uptake; this hydrological control was further modulated by functional composition, with wet-adapted graminoids under shallower water tables frequently maintaining higher uptake. Notably, multi-year observations revealed that vegetation shifts were often abrupt and nonlinear, rather than gradual, underscoring the importance of modelling these dynamics. Importantly, our model reproduced these coupled patterns of hydrology, vegetation, and C–N fluxes, providing confidence in its predictive accuracy. In conclusion, this thesis establishes a comprehensive modelling framework that integrates carbon, nitrogen, water, and vegetation dynamics to evaluate wet grassland function under changing hydrological and climatic conditions. By addressing long-standing limitations of process-based models most notably the omission of species turnover, it demonstrates how vegetation dynamics fundamentally reshape predictions of carbon storage and greenhouse gas fluxes. The results show that rewetting wet grasslands can enhance long-term carbon sequestration but also induce rapid and unexpected shifts in ecosystem function, highlighting the need to account for both biogeochemical and community-level responses. Together, these advances provide a robust basis for future research and a practical tool to support climate-smart rewetting and restoration strategies in peat-rich landscapes.
Valeh Khaledi (Thu,) studied this question.