Seasonal conditions and flux footprints control the contribution of N2O and CH4 to the full GHG balance at grassland on peat soil

Drained peatlands under intensive agricultural land use are hotspots of greenhouse gas (GHG) emissions. While management intensity and soil water status have been identified as major controlling factors, only few studies focussed on temporal dynamics and the contribution of nitrous oxide (N₂O) and methane (CH₄) to full annual GHG balances, mainly due to the lack of continuous observations in high temporal resolution. We present four years of parallel eddy-covariance (EC) and chamber GHG measurements at an intensively managed grassland site on bog peat soil. The site (DE-Okd) is part of the Integrated Carbon Observation System (ICOS) and represents common agricultural practice in Northwest Germany.

Average N₂O fluxes measured by EC were consistently higher than those obtained by chambers. Following a grassland renewal, vegetation development in chamber frames was found to be more favourable than the average growth on the entire field that is seen by the EC tower. Poor grass development was identified by vegetation indices from remote sensing data and probably led to nitrogen surplus in the soil as observed by high ammonium and nitrate concentrations in drainage ditches. These conditions likely favoured both high N₂O emissions and simultaneously high rates of nitrogen leaching. While N₂O emissions made up considerable fractions of full annual GHG balances (~5 to 31%), the contribution of CH₄ was negligible with hardly any significant fluxes detected by chambers and both seasonally varying emissions and uptake measured by EC cancelling out to non-significant shares to the overall budget.

N₂O and CH₄ emissions were strongly influenced by biometeorological factors and land management. Highest N₂O peaks were observed two days after fertilizer application coinciding with about one week after grass cutting and highlighting a well-chosen chamber sampling scheme after management events. Further, N₂O emissions were elevated during daytime under medium soil moisture and high soil temperature regimes, while CH₄ emissions were strongly correlated with soil moisture dropping to nearly zero exchange under dry conditions.

Based on chamber measurements, the overall GHG balance of the site including harvest and carbon input through organic fertilization was in the range of 20 to 25 t CO₂-equivalents ha^-1 yr^-1 in the period from 2020 to 2023 with generally higher emissions in dryer years. Replacing chamber N₂O and CH₄ by EC data for the full budget, individual annual values increased between 0.8 and 10.1 t CO₂-equivalents ha^-1 yr^-1.

We conclude that the combination of EC and chamber measurements helped identifying temporal dynamics of GHG exchange for a better understanding of ecosystem functioning and quantifying method-based uncertainties. Conventionally managed grassland on drained peat soils with high fertilizer input and 4 to 5 grass cuts per year proved to be a significant net GHG emission source. Accounting for footprint heterogeneity – for example through adequate positioning of chamber frames – is of utmost importance for robust determination of total GHG balances at site-level scale.