Drought-induced reductions in net methane emissions from an ombrotrophic peatland are enhanced across a range of experimental warming treatments

Peatlands represent a dominant source of natural CH₄ emissions from the land surface to the atmosphere and the quantitative nature of CH₄ emissions for future climatic conditions is a key unknown. The SPRUCE experimental warming by elevated CO₂ study located in northern Minnesota has been addressing this question for in situ forested peatland plots since 2016 for five different warming treatments (+0, +2.25, +4.5, +6.75 and +9 °C). Under predominantly wet conditions from 2016 through 2019 (2020 observations were not obtained due to COVID travel restrictions) with minimal reductions in peatland water table levels, CH₄ emissions showed an exponential increase across warming treatments with no apparent impact from mid-summer drying conditions nor evidence of a clear elevated CO₂ response. The CH₄ emissions were less than 1 µmol m^-2 s^-1 for ambient or low temperature treatments but ranged from 2 to 5 µmol m^-2 s^-1 under the +6.75 and +9 °C warming treatments.  Moisture and water table levels had minimal impacts on net CH₄ flux during this wet period.

The 2021 summer season, however, provided extremely low precipitation and high evapotranspiration that led to reduced average water table depths across the warming treatments to -0.34, -0.45, -0.54, -0.71 and    -0.83 m, respectively. These drought-induced drops in the water table led to aeration of the surface peat layers (acrotelm) and effectively shut off CH₄ production in the top layers of the bog. Some evidence for limited net CH₄ uptake to the bog during the driest conditions (-0.001 to -0.01µmol m^-2 s^-1) suggested that CH₄ oxidation was playing a role in the reductions of net CH₄ emissions. An empirical fitted relationship for net CH₄ flux as a function of peat temperatures at -0.2 m and water table depth was developed across all treatments and years. That fitted curve showed that net CH₄ emissions were precluded when water table levels dropped below -0.3 m.  This depth corresponds to the peat acrotelm layer containing most of the live root production and activity. The ELM_SPRUCE model was used to fuse the CH₄ data to investigate the causes of reduction in CH₄ emission. The model was able to reconstruct the dynamics of substrates and CH₄ processes under ambient and warming treatments; hydrological feedback was confirmed as warming drives water table drop, which is exacerbated by drought in the summer of 2021.This data-model integration approach suggests the roles of mechanistic models in understanding CH₄ cycling in response to warming and drought interactions in future climates.