•Northern peatlands remain a CO2 sink of ∼0.1 Pg C year−1 until 2300 under RCP2.6•Northern peatlands become a CO2 source of ∼0.2 Pg C year−1 by 2300 under RCP8.5•CH4 emissions from northern peatlands will increase 5-fold by 2300 under RCP8.5•Modeling of peatland resilience, vegetation, and peat quality changes should be improved Intact peatlands remove carbon dioxide (CO2) from the atmosphere through photosynthesis and store the carbon in soils in waterlogged conditions, while emitting methane (CH4) to the atmosphere. The net climate impact of peatlands depends on the relative magnitude of these two greenhouse gases. Here, we assess the future CO2 and CH4 balance of northern peatlands using five large-scale, process-based peatland models. Our results suggest that under climate policies and action, northern peatlands are likely be climate neutral because the climate-warming effect of peatland CH4 emissions is offset by the cooling effect of peatland CO2 sinks. However, if action on climate change is not taken, northern peatlands could accelerate global warming because CH4 emissions are projected to increase substantially, and northern peatlands may turn from CO2 sinks to sources driven by strong warming and drying. Northern peatlands store 300–600 Pg C, of which approximately half are underlain by permafrost. Climate warming and, in some regions, soil drying from enhanced evaporation are progressively threatening this large carbon stock. Here, we assess future CO2 and CH4 fluxes from northern peatlands using five land surface models that explicitly include representation of peatland processes. Under Representative Concentration Pathways (RCP) 2.6, northern peatlands are projected to remain a net sink of CO2 and climate neutral for the next three centuries. A shift to a net CO2 source and a substantial increase in CH4 emissions are projected under RCP8.5, which could exacerbate global warming by 0.21°C (range, 0.09–0.49°C) by the year 2300. The true warming impact of peatlands might be higher owing to processes not simulated by the models and direct anthropogenic disturbance. Our study highlights the importance of understanding how future warming might trigger high carbon losses from northern peatlands. Northern peatlands store 300–600 Pg C, of which approximately half are underlain by permafrost. Climate warming and, in some regions, soil drying from enhanced evaporation are progressively threatening this large carbon stock. Here, we assess future CO2 and CH4 fluxes from northern peatlands using five land surface models that explicitly include representation of peatland processes. Under Representative Concentration Pathways (RCP) 2.6, northern peatlands are projected to remain a net sink of CO2 and climate neutral for the next three centuries. A shift to a net CO2 source and a substantial increase in CH4 emissions are projected under RCP8.5, which could exacerbate global warming by 0.21°C (range, 0.09–0.49°C) by the year 2300. The true warming impact of peatlands might be higher owing to processes not simulated by the models and direct anthropogenic disturbance. Our study highlights the importance of understanding how future warming might trigger high carbon losses from northern peatlands. Global mean surface temperatures are projected to increase by 0.3–4.8°C (relative to 1986–2005) by the end of the 21st century.1IPCCClimate Change 2013. The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, 2013Google Scholar Unabated greenhouse gas (GHG) emissions, such as those of carbon dioxide (CO2) and methane (CH4) in the RCP8.5 scenario and its extension (see experimental procedures), may result in a 3.0–12.6°C global mean warming by the year 2300.1IPCCClimate Change 2013. The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, 2013Google Scholar Given that warming is amplified in the northern mid- and high latitudes compared with the global average,2Seneviratne S.I. Donat M.G. Pitman A.J. Knutti R. Wilby R.L. Allowable CO2 emissions based on regional and impact-related climate targets.Nature. 2016; 529: 477-483Crossref PubMed Scopus (325) Google Scholar the stability of soil organic carbon (SOC) stocks of northern peatlands (300–600 Pg C)3Yu Z. Loisel J. Brosseau D.P. Beilman D.W. Hunt S.J. Global peatland dynamics since the last glacial maximum.Geophys. Res. Lett. 2010; 37: 1-5Crossref Scopus (627) Google Scholar, 4Yu Z. Northern peatland carbon stocks and dynamics: a review.Biogeosciences. 2012; 9: 4071-4085Crossref Scopus (404) Google Scholar, 5Hugelius G. Loisel J. Chadburn S. Jackson R.B. Jones M.C. MacDonald G.M. Marushchak M.E. Olefeldt D. Packalen M. Siewert M.B. et al.Large stocks of peatland carbon and nitrogen are vulnerable to permafrost thaw.Proc. Natl. Acad. Sci. U S A. 2020; 117: 20438-20446Crossref PubMed Scopus (90) Google Scholar is of particular concern. The future carbon balance of northern peatlands remains, however, poorly understood, with only a few studies attempting a quantification for the 21st century6Leifeld J. Wüst-Galley C. Page S. Intact and managed peatland soils as a source and sink of GHGs from 1850 to 2100.Nat. Clim. Change. 2019; 9: 945-947Crossref Scopus (57) Google Scholar, 7Qiu C. Zhu D. Ciais P. Guenet B. Peng S. The role of northern peatlands in the global carbon cycle for the 21st century.Glob. Ecol. Biogeogr. 2020; 29: 956-973Crossref Scopus (15) Google Scholar, 8Müller J. Joos F. Committed and projected future changes in global peatlands-continued transient model simulations since the last glacial maximum.Biogeosciences. 2021; 18: 3657-3687Crossref Scopus (1) Google Scholar, 9Stocker B.D. Roth R. Joos F. Spahni R. Steinacher M. Zaehle S. Bouwman L. Prentice I.C. Multiple greenhouse-gas feedbacks from the land biosphere under future climate change scenarios.Nat. Clim. Change. 2013; 3: 666-672Crossref Scopus (141) Google Scholar, 10Alexandrov G.A. Brovkin V.A. Kleinen T. Yu Z. The capacity of northern peatlands for long-term carbon sequestration.Biogeosciences. 2020; 17: 47-54Crossref Scopus (10) Google Scholar, 11Chaudhary N. Miller P.A. Smith B. Modelling past, present and future peatland carbon accumulation across the pan-Arctic region.Biogeosciences. 2017; 14: 4023Crossref Scopus (27) Google Scholar, 12Chaudhary N. Westermann S. Lamba S. Shurpali N. Sannel A.B.K. Schurgers G. Miller P.A. Smith B. Modelling past and future peatland carbon dynamics across the pan-Arctic.Glob. Change Biol. 2020; 26: 4119-4133Crossref PubMed Scopus (18) Google Scholar, 13Spahni R. Joos F. Stocker B.D. Steinacher M. Yu Z.C. Transient simulations of the carbon and nitrogen dynamics in northern peatlands: from the Last Glacial Maximum to the 21st century.Clim. Past. 2013; 9: 1287-1308Crossref Scopus (68) Google Scholar; even less work has been done on predicting peatland carbon fluxes beyond 2100.8Müller J. Joos F. Committed and projected future changes in global peatlands-continued transient model simulations since the last glacial maximum.Biogeosciences. 2021; 18: 3657-3687Crossref Scopus (1) Google Scholar, 9Stocker B.D. Roth R. Joos F. Spahni R. Steinacher M. Zaehle S. Bouwman L. Prentice I.C. Multiple greenhouse-gas feedbacks from the land biosphere under future climate change scenarios.Nat. Clim. Change. 2013; 3: 666-672Crossref Scopus (141) Google Scholar, 10Alexandrov G.A. Brovkin V.A. Kleinen T. Yu Z. The capacity of northern peatlands for long-term carbon sequestration.Biogeosciences. 2020; 17: 47-54Crossref Scopus (10) Google Scholar,14Loisel J. Gallego-Sala A. Amesbury M. Magnan G. Anshari G. Beilman D. Benavides J. Blewett J. Camill P. Charman D. et al.Expert assessment of future vulnerability of the global peatland carbon sink.Nat. Clim. Change. 2020; 11: 70-77Crossref Scopus (39) Google Scholar,15Gallego-Sala A.V. Charman D.J. Brewer S. Page S.E. Prentice I.C. Friedlingstein P. Moreton S. Amesbury M.J. Beilman D.W. Björck S. et al.Latitudinal limits to the predicted increase of the peatland carbon sink with warming.Nat. Clim. Change. 2018; 8: 907-913Crossref Scopus (101) Google Scholar Using evidence from the literature and expert elicitation, Loisel et al.14Loisel J. Gallego-Sala A. Amesbury M. Magnan G. Anshari G. Beilman D. Benavides J. Blewett J. Camill P. Charman D. et al.Expert assessment of future vulnerability of the global peatland carbon sink.Nat. Clim. Change. 2020; 11: 70-77Crossref Scopus (39) Google Scholar found that peatland experts anticipate peat carbon gains in northern regions owing to higher temperatures and peat carbon losses as a result of permafrost degradation through to the year 2300. Gallego-Sala et al.15Gallego-Sala A.V. Charman D.J. Brewer S. Page S.E. Prentice I.C. Friedlingstein P. Moreton S. Amesbury M.J. Beilman D.W. Björck S. et al.Latitudinal limits to the predicted increase of the peatland carbon sink with warming.Nat. Clim. Change. 2018; 8: 907-913Crossref Scopus (101) Google Scholar examined relationships between peat carbon accumulation rates of the last millennium and different climate parameters; they found a positive relationship between peat carbon accumulation rates and growing-season cumulative photosynthetically active radiation (PAR) for northern peatlands. They predicted that northern peatlands would remain a carbon sink until 2300 under both RCP2.6 and RCP8.5 scenarios, with the carbon sink capacity of Arctic peatlands predicted to increase continuously while that of temperate peatlands is predicted to decrease. However, projected future warming is much greater and faster than changes in the last millennium1IPCCClimate Change 2013. The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, 2013Google Scholar,16Marcott S.A. Shakun J.D. Clark P.U. Mix A.C. A reconstruction of regional and global temperature for the past 11,300 years.Science. 2013; 339: 1198-1201Crossref PubMed Scopus (884) Google Scholar and may be accompanied by more frequent and severe droughts.17Dai A. Increasing drought under global warming in observations and models.Nat. Clim. Change. 2013; 3: 52-58Crossref Scopus (2325) Google Scholar With drastic anticipated changes in temperatures and hydroclimate (particularly the water-table position, which is one of the most important controls on peatland carbon balance), peatlands may be pushed beyond their natural envelope, and the abiotic factors that appear to have been correlated with peat carbon accumulation rates in the last millennium cannot be expected to remain the dominant controls in the future.18Evans C.D. Peacock M. Baird A.J. Artz R.R.E. Burden A. Callaghan N. Chapman P.J. Cooper H.M. Coyle M. Craig E. et al.Overriding water table control on managed peatland greenhouse gas emissions.Nature. 2021; 593: 548-552PubMed Google Scholar, 19Swindles G.T. Morris P.J. Mullan D.J. Payne R.J. Roland T.P. Amesbury M.J. Lamentowicz M. Turner T.E. Gallego-Sala A. Sim T. et al.Widespread drying of European peatlands in recent centuries.Nat. Geosci. 2019; 12: 922-928Crossref Scopus (70) Google Scholar, 20Huang Y. Ciais P. Luo Y. Zhu D. Wang Y. Qiu C. Goll D.S. Guenet B. Makowski D. De Graaf I. et al.Tradeoff of CO2 and CH4 emissions from global peatlands under water-table drawdown.Nat. Clim. Change. 2021; 11: 618-622Crossref Scopus (5) Google Scholar Process-based models, which explicitly parameterize peatland hydrological, thermal, and biogeochemical processes, can account for the complex interactions among factors that are important in driving the dynamics of peatland ecosystems (e.g., temperature, soil moisture, vegetation type, thaw depth, CO2 fertilization). They are useful tools to explore responses of peatlands to future climate changes. Yet, peatlands are not explicitly represented in the land surface of the current generation of earth system models (ESMs), and only a few studies have considered peatland carbon-climate feedbacks using coupled peatland-climate models.9Stocker B.D. Roth R. Joos F. Spahni R. Steinacher M. Zaehle S. Bouwman L. Prentice I.C. Multiple greenhouse-gas feedbacks from the land biosphere under future climate change scenarios.Nat. Clim. Change. 2013; 3: 666-672Crossref Scopus (141) Google Scholar Previous studies based on independent offline peatland model simulations have already provided some insight into the future carbon balance of peatlands. However, it is difficult to draw robust conclusions because critical inputs (i.e., climate forcing, peatland extent, peat initiation time) were different among simulations resulting in divergent projections.7Qiu C. Zhu D. Ciais P. Guenet B. Peng S. The role of northern peatlands in the global carbon cycle for the 21st century.Glob. Ecol. Biogeogr. 2020; 29: 956-973Crossref Scopus (15) Google Scholar,8Müller J. Joos F. Committed and projected future changes in global peatlands-continued transient model simulations since the last glacial maximum.Biogeosciences. 2021; 18: 3657-3687Crossref Scopus (1) Google Scholar,11Chaudhary N. Miller P.A. Smith B. Modelling past, present and future peatland carbon accumulation across the pan-Arctic region.Biogeosciences. 2017; 14: 4023Crossref Scopus (27) Google Scholar, 12Chaudhary N. Westermann S. Lamba S. Shurpali N. Sannel A.B.K. Schurgers G. Miller P.A. Smith B. Modelling past and future peatland carbon dynamics across the pan-Arctic.Glob. Change Biol. 2020; 26: 4119-4133Crossref PubMed Scopus (18) Google Scholar, 13Spahni R. Joos F. Stocker B.D. Steinacher M. Yu Z.C. Transient simulations of the carbon and nitrogen dynamics in northern peatlands: from the Last Glacial Maximum to the 21st century.Clim. Past. 2013; 9: 1287-1308Crossref Scopus (68) Google Scholar In addition, while most previous attempts considered only changes in the rates of peatland carbon accumulation/storage or focused only on CO2 fluxes,7Qiu C. Zhu D. Ciais P. Guenet B. Peng S. The role of northern peatlands in the global carbon cycle for the 21st century.Glob. Ecol. Biogeogr. 2020; 29: 956-973Crossref Scopus (15) Google Scholar,8Müller J. Joos F. Committed and projected future changes in global peatlands-continued transient model simulations since the last glacial maximum.Biogeosciences. 2021; 18: 3657-3687Crossref Scopus (1) Google Scholar,11Chaudhary N. Miller P.A. Smith B. Modelling past, present and future peatland carbon accumulation across the pan-Arctic region.Biogeosciences. 2017; 14: 4023Crossref Scopus (27) Google Scholar, 12Chaudhary N. Westermann S. Lamba S. Shurpali N. Sannel A.B.K. Schurgers G. Miller P.A. Smith B. Modelling past and future peatland carbon dynamics across the pan-Arctic.Glob. Change Biol. 2020; 26: 4119-4133Crossref PubMed Scopus (18) Google Scholar, 13Spahni R. Joos F. Stocker B.D. Steinacher M. Yu Z.C. Transient simulations of the carbon and nitrogen dynamics in northern peatlands: from the Last Glacial Maximum to the 21st century.Clim. Past. 2013; 9: 1287-1308Crossref Scopus (68) Google Scholar, 14Loisel J. Gallego-Sala A. Amesbury M. Magnan G. Anshari G. Beilman D. Benavides J. Blewett J. Camill P. Charman D. et al.Expert assessment of future vulnerability of the global peatland carbon sink.Nat. Clim. Change. 2020; 11: 70-77Crossref Scopus (39) Google Scholar, 15Gallego-Sala A.V. Charman D.J. Brewer S. Page S.E. Prentice I.C. Friedlingstein P. Moreton S. Amesbury M.J. Beilman D.W. Björck S. et al.Latitudinal limits to the predicted increase of the peatland carbon sink with warming.Nat. Clim. Change. 2018; 8: 907-913Crossref Scopus (101) Google Scholar the relative contribution of different GHGs (i.e., CO2 versus CH4) is essential information if we are to understand the influence of peatlands on the global carbon cycle and on the Earth’s climate system.9Stocker B.D. Roth R. Joos F. Spahni R. Steinacher M. Zaehle S. Bouwman L. Prentice I.C. Multiple greenhouse-gas feedbacks from the land biosphere under future climate change scenarios.Nat. Clim. Change. 2013; 3: 666-672Crossref Scopus (141) Google Scholar Furthermore, a timescale of the next century is too short to look at if we want to elucidate the full peatland responses to warming, owing to long (decades to centuries) lags in the response of slow-turnover peatland carbon pools to climate forcing.8Müller J. Joos F. Committed and projected future changes in global peatlands-continued transient model simulations since the last glacial maximum.Biogeosciences. 2021; 18: 3657-3687Crossref Scopus (1) Google Scholar To address the above research gaps, we conducted a multi-model assessment of changes of net CO2 and CH4 fluxes from intact northern peatlands (north of 30°N latitude) using state-of-the-art large-scale peatland models, ORCHIDEE-PEAT,21Qiu C. Zhu D. Ciais P. Guenet B. Krinner G. Peng S. Aurela M. Bernhofer C. Brümmer C. Bret-Harte S. et al.ORCHIDEE-PEAT (revision 4596), a model for northern peatland CO2, water, and energy fluxes on daily to annual scales.Geosci. Model Dev. 2018; 11: 497-519Crossref Scopus (29) Google Scholar,22Qiu C. Zhu D. Ciais P. Guenet B. Peng S. Krinner G. Tootchi A. Ducharne A. Hastie A. Modelling northern peatland area and carbon dynamics since the Holocene with the ORCHIDEE-PEAT land surface model (SVN r5488).Geosci. Model Dev. 2019; 12: 2961-2982Crossref Scopus (10) Google Scholar LPJ-MPI,23Kleinen T. Brovkin V. Schuldt R.J. A dynamic model of wetland extent and peat accumulation: results for the Holocene.Biogeosciences. 2012; 9: 235-248Crossref Scopus (93) Google Scholar LPX-Bern,8Müller J. Joos F. Committed and projected future changes in global peatlands-continued transient model simulations since the last glacial maximum.Biogeosciences. 2021; 18: 3657-3687Crossref Scopus (1) Google Scholar,13Spahni R. Joos F. Stocker B.D. Steinacher M. Yu Z.C. Transient simulations of the carbon and nitrogen dynamics in northern peatlands: from the Last Glacial Maximum to the 21st century.Clim. Past. 2013; 9: 1287-1308Crossref Scopus (68) Google Scholar,24Wania R. Ross I. Prentice I.C. Integrating peatlands and permafrost into a dynamic global vegetation model: 1. Evaluation and sensitivity of physical land surface processes.Glob. Biogeochem. Cycles. 2009; 23: 1-19Google Scholar, 25Wania R. Ross I. Prentice I.C. Integrating peatlands and permafrost into a dynamic global vegetation model: 2. Evaluation and sensitivity of vegetation and carbon cycle processes.Glob. Biogeochem. Cycles. 2009; 23: 1-15Google Scholar, 26Müller J. Joos F. Global peatland area and carbon dynamics from the Last Glacial Maximum to the present – a process-based model investigation.Biogeosciences. 2020; 17: 5285-5308Crossref Scopus (4) Google Scholar, 27Stocker B.D. Spahni R. Joos F. DYPTOP: a cost-efficient TOPMODEL implementation to simulate sub-grid spatio-temporal dynamics of global wetlands and peatlands.Geosci. Model Dev. 2014; 7: 3089-3110Crossref Scopus (51) Google Scholar LPJ-GUESS,28Smith B. Warlind D. Arneth A. Hickler T. Leadley P. Siltberg J. Zaehle S. Implications of incorporating N cycling and N limitations on primary production in an individual-based dynamic vegetation model.Biogeosciences. 2014; 11: 2027-2054Crossref Scopus (293) Google Scholar,29McGuire A.D. Christensen T.R. Hayes D. Heroult A. Euskirchen E. Kimball J.S. Koven C. Lafleur P. Miller P.A. Oechel W. et al.An assessment of the carbon balance of Arctic tundra: comparisons among observations, process models, and atmospheric inversions.Biogeosciences. 2012; 9: 3185-3204Crossref Scopus (209) Google Scholar and LPJ-GUESS_dynP (“dynP” for dynamic multi-peat layers)11Chaudhary N. Miller P.A. Smith B. Modelling past, present and future peatland carbon accumulation across the pan-Arctic region.Biogeosciences. 2017; 14: 4023Crossref Scopus (27) Google Scholar,12Chaudhary N. Westermann S. Lamba S. Shurpali N. Sannel A.B.K. Schurgers G. Miller P.A. Smith B. Modelling past and future peatland carbon dynamics across the pan-Arctic.Glob. Change Biol. 2020; 26: 4119-4133Crossref PubMed Scopus (18) Google Scholar,30Chaudhary N. Miller P.A. Smith B. Modelling Holocene peatland dynamics with an individual-based dynamic vegetation model.Biogeosciences. 2017; 14: 2571-2596Crossref Scopus (14) Google Scholar (see Note S1 and Tables S1–S3 for more detailed information about each model). The models were forced with a fixed peatland extent (Figure 1)31Xu J. Morris P.J. Liu J. Holden J. PEATMAP: refining estimates of global peatland distribution based on a meta-analysis.Catena. 2018; 160: 134-140Crossref Scopus (187) Google Scholar and integrated from 10,000 years before the present to the year 2300 following a common simulation protocol (see experimental procedures). The same atmospheric CO2 concentration and bias-corrected gridded climate projections from the IPSL-CM5A-LR general circulation model (GCM)32Frieler K. Lange S. Piontek F. Reyer C.P.O. Schewe J. Warszawski L. Zhao F. Chini L. Denvil S. Emanuel K. Assessing the impacts of 1.5 C global warming–simulation protocol of the inter-sectoral impact model intercomparison project (ISIMIP2b).Geosci. Model Dev. 2017; 10: 4321-4345Crossref Scopus (221) Google Scholar for a strong climate mitigation scenario (RCP2.6) and a high-end emission scenario (RCP8.5) were used to drive all models. The global mean warming over land under RCP2.6 is projected to reach 1.6°C (1.9°C over land >30°N) by 2100 relative to 1986–2005 in the IPSL-CM5A-LR GCM, followed by a steady but small cooling trend until 2300. In contrast, a dramatic global land warming is projected for RCP8.5, with the air temperature increasing by ∼5.9°C (6.7°C over land >30°N) by 2100 and ∼14.7°C by 2300 (15.5°C over land >30°N) (see Figure S1). Simulated present-day (1986–2005) total SOC stock for the 3.2 million km2 covered by northern peatlands31Xu J. Morris P.J. Liu J. Holden J. PEATMAP: refining estimates of global peatland distribution based on a meta-analysis.Catena. 2018; 160: 134-140Crossref Scopus (187) Google Scholar ranges from 200 to 870 Pg C among models (Figure S2), which brackets previous estimates based on observations (300–600 Pg C).3Yu Z. Loisel J. Brosseau D.P. Beilman D.W. Hunt S.J. Global peatland dynamics since the last glacial maximum.Geophys. Res. Lett. 2010; 37: 1-5Crossref Scopus (627) Google Scholar, 4Yu Z. Northern peatland carbon stocks and dynamics: a review.Biogeosciences. 2012; 9: 4071-4085Crossref Scopus (404) Google Scholar, 5Hugelius G. Loisel J. Chadburn S. Jackson R.B. Jones M.C. MacDonald G.M. Marushchak M.E. Olefeldt D. Packalen M. Siewert M.B. et al.Large stocks of peatland carbon and nitrogen are vulnerable to permafrost thaw.Proc. Natl. Acad. Sci. U S A. 2020; 117: 20438-20446Crossref PubMed Scopus (90) Google Scholar The multi-model ensemble mean SOC (572 Pg C) falls close to the mean from observations. Hereafter, we report multi-model ensemble mean values, with ranges across models in parentheses, unless stated otherwise. The CO2 balance of peatlands in this study, termed the net biome production (NBP), is calculated as net primary productivity (NPP) minus heterotrophic respiration—anthropogenic disturbances of peatland and fires are not modeled. A positive NBP thus represents a CO2 flux from the atmosphere to the land. Simulated present-day NBP of all northern peatlands is 0.11 (0.01–0.22) Pg C year−1 (Figure 1A), matching previous estimates for the Northern Hemisphere (0.10–0.15 Pg C year−1).5Hugelius G. Loisel J. Chadburn S. Jackson R.B. Jones M.C. MacDonald G.M. Marushchak M.E. Olefeldt D. Packalen M. Siewert M.B. et al.Large stocks of peatland carbon and nitrogen are vulnerable to permafrost thaw.Proc. Natl. Acad. Sci. U S A. 2020; 117: 20438-20446Crossref PubMed Scopus (90) Google Scholar,15Gallego-Sala A.V. Charman D.J. Brewer S. Page S.E. Prentice I.C. Friedlingstein P. Moreton S. Amesbury M.J. Beilman D.W. Björck S. et al.Latitudinal limits to the predicted increase of the peatland carbon sink with warming.Nat. Clim. Change. 2018; 8: 907-913Crossref Scopus (101) Google Scholar,34Frolking S. Talbot J. Jones M.C. Treat C.C. Kauffman J.B. Tuittila E.-S. Roulet N.T. Peatlands in the Earth’s 21st century climate system.Environ. Rev. 2011; 19: 371-396Crossref Scopus (253) Google Scholar The projected future NBP of northern peatlands depends on the trajectory of climate change. Under RCP2.6, northern peatlands remain CO2 sinks with a relatively stable net CO2 uptake rate until 2300 (Figure 1A). By contrast, under RCP8.5, northern peatlands turn into a CO2 source within the coming 100–150 years (Figure 1A). The simulated future carbon balance of peatlands varies among subregions in response to projected strong climate warming and precipitation changes in RCP8.5. For four subregions out of the five main peat complexes of the Northern Hemisphere, i.e., continental Western Canada (CWC) (Figure 1B), Hudson Bay Lowlands (HBL) (Figure 1C), Northern Europe (NOE) (Figure 1D), and West Siberian Lowlands (WSL) (Figure 1E), all models projected a future shift from peatland CO2 sinks to CO2 sources (or to nearly carbon neutral) under RCP8.5. Two models (LPX-Bern and LPJ-GUESS), which explicitly simulated coupled peatland nitrogen and carbon cycling, projected that these peatlands will be larger CO2 sources in the future, as opposed to models in which NPP is not limited by available soil nitrogen (ORCHIDEE, LPJ-MPI and LPJ-GUESS_dynP). For the Russian Far East (RFE) (Figure 1F), where the projected increase in precipitation is the largest under RCP8.5 (110% increase in RFE precipitation by 2300 with respect to present-day versus 5%–75% increase in precipitation for the other four subregions) (Figure S1), only one model (LPJ-GUESS) predicts that this peatland complex will become a large net source of CO2 in the future. All five models project a trend toward shallower water tables for RFE peatlands (Figure S3), indicating that RFE peat remains preserved by anoxic conditions below the water table. Simulated NBP of peatlands for all five subregions is in good agreement with the empirical extrapolation of peat accumulation made by Gallego-Sala et al.15Gallego-Sala A.V. Charman D.J. Brewer S. Page S.E. Prentice I.C. Friedlingstein P. Moreton S. Amesbury M.J. Beilman D.W. Björck S. et al.Latitudinal limits to the predicted increase of the peatland carbon sink with warming.Nat. Clim. Change. 2018; 8: 907-913Crossref Scopus (101) Google Scholar under RCP2.6. However, under RCP8.5, only the simulated NBP of RFE peatlands, where the largest increase in precipitation and persistent anoxic conditions is projected, is comparable to the estimate from Gallego-Sala et al.15Gallego-Sala A.V. Charman D.J. Brewer S. Page S.E. Prentice I.C. Friedlingstein P. Moreton S. Amesbury M.J. Beilman D.W. Björck S. et al.Latitudinal limits to the predicted increase of the peatland carbon sink with warming.Nat. Clim. Change. 2018; 8: 907-913Crossref Scopus (101) Google Scholar For the other four subregions (CWC, HBL, NOE, and WSL), Gallego-Sala et al. predicted a slight increase of peatland NBP under RCP8.5, in contrast to the mechanistic models applied here. Gallego-Sala et al. considered only peat that accumulated during the last millennium and ignored the decomposition of deeper (older) peat. When the water table drops below a critical level, as projected by some of our models (Figure S3), the exposure of deeper peat to aerobic and warmer conditions results in substantial loss of carbon.18Evans C.D. Peacock M. Baird A.J. Artz R.R.E. Burden A. Callaghan N. Chapman P.J. Cooper H.M. Coyle M. Craig E. et al.Overriding water table control on managed peatland greenhouse gas emissions.Nature. 2021; 593: 548-552PubMed Google Scholar,35Dorrepaal E. Toet S. Van Logtestijn R.S.P. Swart E. Van De Weg M.J. Callaghan T.V. Aerts R. Carbon respiration from subsurface peat accelerated by climate warming in the subarctic.Nature. 2009; 460: 616-619Crossref Scopus (472) Google Scholar,36Fenner N. Freeman C. Drought-induced carbon loss in peatlands.Nat. Geosci. 2011; 4: 895-900Crossref Scopus (361) Google Scholar There is a large variation in the simulated trajectories of peatland carbon dynamics under RCP8.5 among model simulations. This variation is due to substantial differences among models in the parameterization of peatland vegetation, hydrological and thermal processes (Note S1 and Tables S1 and S2) and consequently a wide range of predicted peatland water balance terms (Figures S3 and S4), soil temperature (Figure S5), NPP (Figure S6), and carbon inputs to the soil in these simulations. Figures S7–S9 show the capability of the models to reproduce the current water-table position and NBP at the site level. Peatland development is strongly governed by local conditions, and it is therefore nearly impossible for large-scale models to exactly reproduce the development of peatland at each site. LPJ-GUESS and LPJ-GUESS_dynP better captured the interannual, among-sites variability in peatland water-table position. However, because the projected RCP8.5 climate lies far outside of past conditions for which we can validate the models, there is no way to ascertain which one of the models simulates the peatland hydrological and carbon dynamics most accurately under the extreme and long-term warming coupled to elevated CO2 concentrations of RCP8.5. To quantify the role of northern peatlands in the global carbon cycle, in this study we compare the NBP of northern peatlands with the NBP of all other northern (>30°N) and global lands (Figure 2). Estimates for NBP of all northern and global lands from land surface model (LSM) simulations are from the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP2b; https://doi.org/10.5880/PIK.2019.012),32Frieler K. Lange S. Pion