Stimulation of soil N cycling after two years of Free Air CO2 Enrichment increases nitrous oxide emissions in a temperate forest, UK.

Increasing atmospheric CO₂ concentrations in temperate forests may affect soil nitrogen (N) cycling processes due to the increased demand  for nitrogen availability by trees to support CO₂ uptake through photosynthesis. This in turn can affect the emission of nitrous oxide (N₂O) from the forest soil leading to a potential trade-off between the enhanced canopy CO₂ uptake and soil N₂O emission. The Birmingham Institute of Forest Research (BIFoR) established a Free-Air CO₂ Enrichment (FACE) facility in a mature oak forest in Staffordshire, UK, which became operational in 2017. In April 2018 and again in May 2019, two years after the start of fumigation with 550 ppm CO₂, we collected soil samples (0 – 15 cm depth) from the three elevated CO₂ (eCO₂) and three control plots. Soils were amended in the laboratory with 98 at % ¹⁵N-NH₄⁺ and ¹⁵N-NO₃^- . Gross N mineralisation and nitrification were estimated by the isotope dilution technique, while N₂O emission from nitrification (¹⁵N-NH₄⁺ treatment) and denitrification (¹⁵N-NO₃^- treatment) were estimated by the ¹⁵N Gas-Flux method. Additionally, C/N ratio and δ¹⁵N and δ¹³C were measured in unamended eCO₂ and control samples via EA-IRMS. Whilst gross N mineralisation and N₂O emission were only marginally higher in eCO₂ plots compared to controls after one year of fumigation, there was a significant stimulation of N cycling after the second year that led to more pronounced differences. Gross N mineralisation rates doubled in the eCO₂ plots (mean: 4.09 μg N g^-1 d^-1, P < 0.05) compared to the control plots (mean: 2.02 μg N g^-1 d^-1), while a similar twofold increase was observed for gross nitrification rates (mean eCO₂: 1.63 μg N g^-1 d^-1;  mean control: 0.70 μg N g^-1 d^-1, P < 0.05). N₂O emission from both denitrification (mean: 0.03 ng N g^-1 d^-1) and nitrification (mean: 0.02 ng N g^-1 d^-1) were generally low but of similar magnitude and more than double than in the control plots. C/N ratio was conservative between eCO₂ and control plots as a result of proportional increase of C and N contents in the eCO₂ plots. The observed stimulation in N cycling was further corroborated by the significantly enriched δ¹⁵N signal (-0.66 ‰) in eCO₂ plots compared to the controls (-1.38‰). Moreover, the eCO₂ samples had a more depleted δ¹³C signal (-28.37 ‰) compared to the controls (-27.99 ‰), as a result of the additional carbon supplied through fumigation (CO₂ δ¹³C ~ -28 ‰). Following the first year of CO₂ fumigation, there were indications of soil N limitation despite the high rates of atmospheric N deposition (22 kg N ha^-1 y^-1). However, after only 2 years of the FACE experiment there is strong evidence of a shift in key soil N processes to sustain the enhanced nutrient demands to support enhanced canopy CO₂ uptake. Further research is underway in BiFOR-FACE to elucidate whether carbon quantity and/or quality drives the stimulation of soil N cycling and what are the long-term implications of the trade-off between enhanced CO₂ sequestration and a potential increase in N₂O emissions.