High emissions of greenhouse gases from grasslands on peat and other organic soils

Abstract

Drainage has turned peatlands from a carbon sink into one of the world's largest greenhouse gas (GHG) sources from cultivated soils. We analyzed a unique data set (12 peatlands, 48 sites and 122 annual budgets) of mainly unpublished GHG emissions from grasslands on bog and fen peat as well as other soils rich in soil organic carbon (SOC) in Germany. Emissions and environmental variables were measured with identical methods. Site-averaged GHG budgets were surprisingly variable (29.2 ± 17.4 t CO -eq. ha yr ) and partially higher than all published data and the IPCC default emission factors for GHG inventories. Generally, CO (27.7 ± 17.3 t CO ha yr ) dominated the GHG budget. Nitrous oxide (2.3 ± 2.4 kg N O-N ha yr ) and methane emissions (30.8 ± 69.8 kg CH -C ha yr ) were lower than expected except for CH emissions from nutrient-poor acidic sites. At single peatlands, CO emissions clearly increased with deeper mean water table depth (WTD), but there was no general dependency of CO on WTD for the complete data set. Thus, regionalization of CO emissions by WTD only will remain uncertain. WTD dynamics explained some of the differences between peatlands as sites which became very dry during summer showed lower emissions. We introduced the aerated nitrogen stock (N ) as a variable combining soil nitrogen stocks with WTD. CO increased with N across peatlands. Soils with comparatively low SOC concentrations showed as high CO emissions as true peat soils because N was similar. N O emissions were controlled by the WTD dynamics and the nitrogen content of the topsoil. CH emissions can be well described by WTD and ponding duration during summer. Our results can help both to improve GHG emission reporting and to prioritize and plan emission reduction measures for peat and similar soils at different scales.

Peat decomposability in managed organic soils in relation to land use, organic matter composition and temperature

“…With rates of 6.5-9.4 t C ha −1 a −1 net CO 2 fluxes from organic soils now used as croplands were on average found to be higher than from organic soils under grassland, which were estimated to vary between 1.8 and 7.3 t C ha −1 a −1 (IPCC, 2014). However, recent studies reported emission rates of 7.6 ± 2.0 t C ha −1 a −1 on organic soils managed as grassland in Germany and thus much higher rates than previously found for this type of land use (Tiemeyer et al, 2016). Drained organic soils under forest can act as both net sinks or sources of atmospheric CO 2 (Cannell et al, 1993;Minkkinen and Laine, 1998;Minkkinen et al, 1999;Wüst-Galley et al, 2016), although they are in general considered to represent a source with average net CO 2 emissions of 2.0-3.3 t C ha −1 a −1 in the temperate zone (IPCC, 2014).…”

“…Although they cover only 3 % of the earth's terrestrial surface (Tubiello et al, 2016), they store up to 30 % of the global soil organic carbon (SOC) pool (Parish et al, 2008). In Europe, more than 50 % of the former peatland area has been degraded by peat mining and conversion of land use, including drainage, to improve their suitability for agriculture or forestry (Joosten, 2010).…”

“…This transforms the former C-sink into a major source of atmospheric carbon dioxide (CO 2 ) and makes peatlands an important contributor to global climate change (Freeman et al, 2004). Around 85 % of the global annual CO 2 emission of 915 Mt CO 2 -C from drained peatlands are estimated to originate from organic soils now used as croplands (Tubiello et al, 2016). With rates of 6.5-9.4 t C ha −1 a −1 net CO 2 fluxes from organic soils now used as croplands were on average found to be higher than from organic soils under grassland, which were estimated to vary between 1.8 and 7.3 t C ha −1 a −1 (IPCC, 2014).…”

Divergent NEE balances from manual‐chamber CO2 fluxes linked to different measurement and gap‐filling strategies: A source for uncertainty of estimated terrestrial C sources and sinks?

“…This variation of NEE by the choice of gap filling is in the range of recently reported uncertainties of discontinuous manual chamber data (Moffat et al, submitted). Interestingly, this variation is also in the same order of magnitude as the uncertainty reported with IPCC emission factors from organic soils ( Blain et al, ; Drösler et al, ) and other aggregated data sets and meta‐analyses of terrestrial C sources and sinks ( Wilson et al, ; Tiemeyer et al, ). Although very speculative, we propose that the S/C/I approach is likely closer to the true NEE of our system because studies using different methodologies (chamber measurements, long‐term crop records/soil sampling, and eddy covariance measurements) from agro‐ecosystems with forage crops ( e.g ., ryegrass and lucerne) generally estimate an NEE in the range of –100 to –400 g C m −2 ( Byrne et al, ; Bolinder et al, ; Gilmanov et al, ).…”

Comparing chamber and eddy covariance based net ecosystem CO2 exchange of fen soils

“…() and Tiemeyer at al. () for permanent grasslands on fen soils in Denmark and Germany, respectively, obtained by a similar approach. NEE budgets for site UG were smaller compared to site GW, which confirms the general assumption for rewetted peatlands ( Augustin and Joosten , ).…”