Spatial variation in GHG emissions from drained organic forest soils : implications for Swedish UNFCCC reporting

Abstract: Drainage of organic soils for forestry generally increases soil CO2 and N2O emissions but decreases CH4 emissions. Under the United Nations Framework Convention on Climate Change (UNFCCC) countries provide annual national inventories of anthropogenic greenhouse gas (GHG) emissions and removals, which includes emissions of GHGs from drained organic forest soil. In the Swedish GHG inventory these emissions are estimated by multiplying default emission factors (EFs) provided by the IPCC by the area of drained organic forest soil. A soil is considered drained if there is a functioning ditch within 25 m from the centre of a plot in the Swedish National Forest Inventory (NFI), however, this corresponds to only half of the total area of productive forest on organic soils. The aim of this study was to assess GHG emissions from drained organic forest soils, and to suggest additional parameters with which emission factors could be adjusted. To assess the default EFs, a literature review was carried out in which emission data from original peer-reviewed studies were compiled and compared to the default values. To investigate the effect of distance to a ditch on GHG emissions, CO2, CH4 and N2O fluxes were measured along 52.5 m long transects from a ditch in a pine forest stand on drained organic peat soil. Water table depth (WTD) and soil moisture were also measured. The default EFs and those developed based on the literature review in this study differed markedly and emission estimates varied substantially within climate and nutrient subtypes. Using the emission factors found in the literature review in this study would result in total GHG emissions of 6.5 Mt CO2-eq., compared to 6.1 Mt CO2-eq. using the IPCC emission factors. The GHG measurements showed that average fluxes of CO2 and CH4 differed significantly between <25 m and >25 m from the ditch. CO2 fluxes at >25 m corresponded to 65% of fluxes at <25 m. CH4 fluxes were 20% higher at >25 m. No significant difference was found in N2O. CO2 was linearly correlated with both WTD and soil moisture, while CH4 showed a quadratic correlation with WTD and soil moisture. No correlation with either parameter was found in the N2O emissions. The results show that using data published within the last ten years results in EFs that differ considerably to the EFs of the IPCC, however, to develop new and robust national EFs for Sweden would require a more extensive compilation of emission data than was possible within this study. The results also show that not including drained organic forest soils farther than 25 m from a ditch may lead to substantial underestimation of GHG emissions. Using WTD or soil moisture as predictors of GHG emissions is possible. However, to do so in the inventory would require large national datasets of both WTD and GHG emissions. This requires more studies of GHG emissions from different peatland and forest types in Sweden.