Groundwater level effects on greenhouse gas emissions from undisturbed peat cores

• Rewetting above −20 cm decreased CO 2 emissions, but caused strong N 2 O emissions. • CO 2 from peat oxidation showed different responses to drying and rewetting. • Production of CO 2 and N 2 O occurred mostly in the top soil. • Variable soil moisture at a −20 cm groundwater level creates a risk for N 2 O peaks. • Oxidation was estimated to contribute 66 to 79% to vertical displacement. Peat soils store a large part of the global soil carbon stock, which can potentially be lost when they are drained and taken into cultivation, resulting in CO 2 emission and land subsidence. Groundwater level (GWL) management has been proposed to mitigate peat oxidation, but may lead to increased emissions of nitrous oxide (N 2 O) and methane (CH 4 ). The aim of this experiment was to study trade-offs between greenhouse gas emissions from peat soils as a function of GWL. We incubated 1 m deep, 24 cm diameter undisturbed bare soil cores, after removal of the grass layer, from three contrasting Dutch grassland peat sites for 370 days at 16 °C. The cores were subjected to drying-wetting cycles, with the GWL varying between near the soil surface to 160 cm below the surface. We measured gas fluxes of CO 2 , N 2 O and CH 4 from the soil surface, extracted pore water for DOC and mineral nitrogen analysis, and measured soil hydraulic and shrinkage characteristics. Emissions of CO 2 increased after lowering the GWL, but showed different GWL-response curves during rewetting of the soil. On average, highest CO 2 emissions of 1.5 g C·m −2 day −1 were found at a GWL of 80 cm below the surface. However, the 0 cm GWL was the only treatment with significantly lower CO 2 emissions than other GWLs. Cumulative CO 2 emissions differed significantly between sampling sites. Emissions of N 2 O showed a different response, peaking at GWL heights above −20 cm, particularly after a recent GWL rise. Though not significantly different, the highest N 2 O emissions were measured at the 0 cm GWL treatment. We confirmed this pattern for N 2 O in un-replicated soil cores with grass sward, although emission values were lower in these cores due to the root uptake of mineral nitrogen. CH 4 emissions or −uptake remained low under any GWL. We conclude that raising the GWL is a successful strategy to reduce CO 2 emissions from peat oxidation. However, raising the GWL close to the soil surface could lead to N 2 O emissions that negate any gains in terms of global warming potential. Our results suggest that raising the GWL in peat grasslands to −20 cm creates such a risk. A constant GWL at the surface (0 cm) would be preferential for mitigating both CO 2 and N 2 O emissions, although such conditions don’t allow for agricultural grass production (mowing or grazing).