Linearity between temperature peak and bioenergy CO 2 emission rates

Transcription

1 Linearity between temperature peak and bioenergy CO 2 emission rates Francesco Cherubini 1, Thomas Gasser 2,3, Ryan M. Bright 1, Philippe Ciais 3, Anders Hammer Strømman 1 1 Department of Energy and Process Engineering, Norwegian University of Science and Technology (NTNU), Trondheim Norway 2 Centre International de Recherche sur l'environnement et le Développement, CNRS-PontsParisTech-EHESS- AgroParisTech-CIRAD, Nogent-sur-Marne France 3 Laboratoire des Sciences du Climat et de l'environnement, Institut Pierre-Simon Laplace, CEA-CNRS-UVSQ, Gif-sur-Yvette France NATURE CLIMATE CHANGE 1

2 OSCAR v2.1 OSCAR v2.1 is a compact global carbon cycle-climate model 1-3. The oceanic carbon cycle follows the representation of a mix-layer with impulse response functions as described in ref. 4. The sensitivity of the carbonate oceanic chemistry to atmospheric CO 2 and temperature change can be emulated by 3 different non-linear functions: one from ref. 5, and two from ref. 6. The transportation of surface water to deep ocean can be modeled by 4 different impulse response functions 4. The ocean compartment thus has 12 possible parameterizations. The terrestrial carbon cycle is broadly the same as the one of the original OSCAR model 3, albeit it is here aggregated into 9 regions 7. Its preindustrial equilibrium is calibrated on upto-date simulations made with the complex model ORCHIDEE 8. The regional sensitivities of net primary productivity (NPP) to CO 2 and climate are calibrated on idealized (i.e. +1% CO 2 per year) simulations of the IPSL-CM5 model 9, and the global sensitivity of heterotrophic respiration (R h ) follows a "Q 10 " law with Q 10 = Global sensitivities of NPP and R h can also be rescaled to those of one of the eleven models of the C 4 MIP 11, or to the average of the C 4 MIP sensitivities. The sensitivity of NPP to CO 2 is modeled through either a logarithmic or a hyperbolic relationship 12. This yields 26 possibilities of parameterization for the terrestrial carbon-cycle. In the latest OSCAR version, other greenhouse gases and reactive species were also added, following the characteristics of the MAGICC6 model 13. OSCAR computes the radiative forcing (RF) from a perturbation in the atmospheric CO 2 concentration ( C), relative to a reference concentration (C 0 ), using the equation RF = α c ln[(c 0 + C)/C 0 ] (ref. 14 ), where α c = 5.35 W m -2 and C 0 is the atmospheric CO 2 concentration at preindustrial times (278 ppm). To estimate global mean surface temperature change, the model can use 28 different impulse response functions (e.g., see 15 ), whose parameters were calibrated on the 27 Earth system models that made the 4x CO 2 2 NATURE CLIMATE CHANGE

3 SUPPLEMENTARY INFORMATION simulation of the CMIP5 16. The remaining parameterization is based on the average of the CMIP5 models. For instance, the dynamics of the response calibrated on the average is formulated as follows: f(t) = exp(-t/63.10) exp(-t/2.774); and it has a climate sensitivity of about 0.72 K W -1 m 2. In this specific study, the OSCAR v2.1 model thus has a potential of 8736 combinations of parameters for the carbon-climate system: 12 from the ocean component, 26 from the terrestrial component, and 28 to estimate the global mean temperature change. The 1000 simulations used in this paper to estimate the uncertainty range and the ensemble means are drawn among these, with equal-probability. Supplementary Figure S1 compares the responses to a CO 2 emission pulse from OSCAR v2.1 with those from other models. NATURE CLIMATE CHANGE 3

4 a b Figure S1: Impulse response functions (IRF) from a CO 2 emission pulse at year 0. Figure S1a shows the fraction of CO 2 in the air following a pulse emission, Figure S1b the change in average global surface temperature. The red solid line shows the ensemble mean and the colored area represents the results within one standard deviation. The responses are computed under a constant climate following the standard protocol defined in ref. 17. The CO 2 atmospheric concentration response is compared with the multimodel mean Joos et al., 2013 from ref 17 (used in the 5 th IPCC Assessment Report 18 ) and the IRF in the 4 th IPCC Assessment Report 19 Forster et al., The temperature response is compared with the multimodel mean Joos et al., 2013 from ref. 17 and that from ref 15 Boucher and Reddy, NATURE CLIMATE CHANGE

5 SUPPLEMENTARY INFORMATION Time Short, 6 yrs Medium, 55 yrs Long, 103 yrs Normalized cumulative CO 2 flux [unitless] CumNEE, long turnover time CumNEE, medium turnover time CumNEE, short turnover time Title, years Figure S2: Net Ecosystem Exchange (NEE) chronosequences of the post-harvest ecosystem carbon response fed into OSCAR v.2.1. Table: normalized mean instantaneous NEE fluxes. Graph: normalized cumulative NEE (CumNEE) fluxes, including the CO 2 emission pulse at year zero from combustion. Each chronosequence is normalized to the size of the original pulse so to achieve the carbon neutrality along the prescribed turnover times. As indicated by the CumNEE flux, the harvest event will occur when the net cumulative emissions reach zero. These chronosequences are representative of three different CO 2 flux rates and time spans dictating the replinishment of the biomass resource. Negative values of NEE means net CO 2 removal from the air, while positive values correspond to emissions. The NEE profile for the short turnover time is based on an eucalyptus plantation from ref. 20, case BL 1 (with all the bark, understorey, slash and litter retained on site and assuming an average resource turnover time of 6 years). The NEE chronosequence for the medium turnover time is adapted from the so-called Wisconsin chronosequence in ref. 21 after averaging observed values at each time step of the chronosequence. This case is representative of a cold temperate forest plantation dominated by aspen (Populus tremuloides). In this area, all stands less than 100 years old are regenerated following commercial harvest with forestry statistics reporting an average turnover time of 55 years 22. The NEE profile for the long turnover time is representative of a hemiboreal forest plantation dominated by pine (Pinus sylvestris) from ref. 23 (case 12, turnover time 103 years). Values in the table are from a linear interpolation of the data. NATURE CLIMATE CHANGE 5

6 a b Figure S3: Comparison of the IRFs from OSCAR v2.1 following a CO 2 emission pulse from fossil fuels or bioenergy. Solid lines show the ensemble means and the colored areas represent the respective results within one standard deviation. These responses are computed following the protocol for simulations under a constant climate 17. Figure S3a is the CO 2 atmospheric concentration response, and using a simple first order decay model the atmospheric lifetimes of the perturbations can be approximated to 2.2, 17, and 27 years for the short, medium and long turnover time, respectively. Figure S3b shows the temperature response. While an emission of fossil carbon causes a permanent warming, the temperature maximum from bioenergy is reached within a few decades and is limited in time, as the temperature change gradually decreases thanks to the additional CO 2 uptake by the biomass re-growth. In the bioenergy cases, the emission pulse is associated with the following ecosystem carbon response which is modeled as a serie of small negative pulses whose cumulative value over the turnover time equals the initial pulse. Case specific net carbon losses or gains at the end of the biomass turnover time would proportionally shift the responses upwards or downwards, respectively, as shown elsewhere 24,25. 6 NATURE CLIMATE CHANGE

7 SUPPLEMENTARY INFORMATION Figure S4: Temperature response (ensemble means only) to a CO 2 emission pulse in 2100 considering the changes in atmospheric CO 2 conentration along the 21 st century according to the four RCP scenarios. Colored areas represent the range between the 2 extreme RCP scenarios, i.e. RCP2.6 and RCP8.5. The other two intermediate scenarios, i.e. RCP4.5 and RCP6.0, generally fall within this range. In these simulations, the protocol 17 is extended considering the specific RCPs from 2010 up to 2100, with the stabilization of CO 2 background concentration and other anthropogenic forcings to the 2100 level. After that, a pulse emission of 100 GtonC is added to the atmosphere in the year The figure shows that at increasing CO 2 atmospheric concentrations a CO 2 emission pulse will become gradually less effective on the average global temperature. Under the most pessimistic scenario (i.e., RCP8.5), the addition of CO 2 to the atmosphere has less effect on average global surface temperature than under the lowest forcing scenario (RCP2.6), especially in the short term. Over time the temperature decreases more quickly under RCP2.6 because of the lower level of saturation of the terrestrial and ocean sinks that uptake the CO 2 excess. NATURE CLIMATE CHANGE 7

8 Figure S5. Model simulations (ensemble means only) of the CCR as a function of time under varying CO 2 background concentration levels in CCR is measured from 2100 after letting CO 2 concentration change during the 21 st century according to the four RCP scenarios. Colored areas represent the range between the 2 extreme RCPs, i.e. RCP2.6 and RCP8.5. In all simulations CCR values decrease with the increasing background CO 2 concentration, and hence with the level of cumulative CO 2 emissions associated with each RCP. The sensitivity of CCR to time in fossil experiments becomes noticeable in the RCP8.5 case, where cumulative emissions are higher than 2 TtC, which is the upper limit identified for the CCR time constancy 26,27. In the bioenergy simulations, variations among the RCPs can be mainly observed in the first decades and then the results tend to converge over time. 8 NATURE CLIMATE CHANGE

9 SUPPLEMENTARY INFORMATION a 6 4 CO2 from fossil energy CO2 from bioenergy, short CO2 from bioenergy, medium CO2 from bioenergy, long R² = b 6 4 CO2 from fossil energy CO2 from bioenergy, short CO2 from bioenergy, medium CO2 from bioenergy, long R² = R² = R² = R² = R² = R² = Cumulative emissions (TtC) 0 R² = Maximum emission rate (TtC/year) c E N2O (left axis) N2O (left axis) R² = d CH4 (right axis) CH4 (right axis) R² = E E E R² = E E-03 ΔT peak ( C) R² = E E E-03 ΔT peak ( C) E E Cumulative emissions (Mtons) 0.0E Maximum emission rate (Mtons/year) 0.0E+00 e f Figure S6: Global temperature peak rise versus cumulative emissions (graphs on the left hand side) and maximum emission rates (graphs on the right hand side) for CO 2 from fossil fuels and bioenergy (a, b), CH 4 and N 2O (c, d), and the other GHGs selected in this study (e, f). Results (ensemble means only) are for emission trajectories characterized by 500 combinations of maximum emission rates between 0 and 20 GtC per year and resulting in cumulative emissions between 0 and 4 TtC. Values of the R 2 of each linear fit (intercept set equal to zero) are also shown. The linear correlation between ΔT peak and E max (b) can be used to build a simple relationship between maximum CO 2 emission rates from bioenergy and temperature peak (units in C per GtonC/year): ΔT peak = E max for the short turnover time, ΔT peak = E max for the medium turnover time, and ΔT peak = E max for the long turnover time. NATURE CLIMATE CHANGE 9

10 1.00 CH 4 (12.4) CO 2 from bioenergy, medium (17) HFC-125 (28.2) HFC-41 (2.8) CO 2 from bioenergy, long (27) CO 2 from bioenergy, short (2.2) R 2 ( T peak, E max ) CFC-113 (85) CFC-12 (100) N 2 O (121) 0.85 Fossil CO R 2 ( T peak, E) Figure S7: Correlation (R 2 ) between temperature peak and cumulative emissions ( T peak, E) and between temperature peak and maximum emission rates ( T peak, E max) for the selected well-mixed GHGs. Correlations are derived from the simulations in Figure S6. The dotted red line marks the equal correlation. Following the same approach used in ref. 28, GHGs can be distinguished into shorter- and long-lived. CO 2 emissions from bioenergy show characteristics similar to shorter-lived species. 10 NATURE CLIMATE CHANGE

11 SUPPLEMENTARY INFORMATION a b N2O (left axis) CH4 (right axis) 1.2E E E E E E Maximum emission rate (Mt ons per year) 0.0E+00 c CFC-12 HFC-125 CFC-113 HFC-41 (right axis) 7.0E E E E E E E-05 d CFC-12 HFC-125 CFC-113 HFC-41 (right axis) 1.8E E E Cumulative emissions (Mtons) 0.0E E Maximum emission rate (Mtons per year) Figure S8: Temperature peak (ΔT peak) from idealized emission trajectories of non-co 2 GHGs as a function of cumulative emissions (ΣE) or emission rates (E max). Graphs on the left hand side (a, c) show the dependency of the temperature peak on different levels of cumulative emissions under a constant maximum peak rate (10 Mtons/year for CH 4 and N 2O and 0.01 Mtons/year for the other GHGs). Graphs on the right hand side (b, d) show the dependency of the temperature peak on maximum emission rates under constant cumulative emissions (1000 Mtons for CH 4 and N 2O and 1 Mton for the other GHGs). These data are used to compute the normalized temperature ranges of the non-co 2 gases shown in Figure 3c of the main text. NATURE CLIMATE CHANGE 11

12 References 1 Ciais, P. et al. Attributing the increase in atmospheric CO2 to emitters and absorbers. Nature Clim. Change 3, (2013). 2 Gasser, T. & Ciais, P. A theoretical framework for the net land-to-atmosphere CO2 flux and its implications in the definition of "emissions from land-use change". Earth Syst. Dynam. 4, (2013). 3 Gitz, V. & Ciais, P. Amplifying effects of land-use change on future atmospheric CO2 levels. Global Biogeochemical Cycles 17, 1024 (2003). 4 Joos, F. et al. An Efficient and Accurate Representation of Complex Oceanic and Biospheric Models of Anthropogenic Carbon Uptake. Tellus 48B, (1996). 5 Joos, F. et al. Global warming feedbacks on terrestrial carbon uptake under the Intergovernmental Panel on Climate Change (IPCC) emission scenarios. Global Biogeochemical Cycles 15, (2001). 6 Harman, I. N., et al. SCCM - the Simple Carbon-Climate Model: Technical documentation. Centre for Australian Weather and Climatie Research, Bureau of Meteorology and CSIRO. Melbourne, Australia (2011). 7 Houghton, R. A. The annual net flux of carbon to the atmosphere from changes in land use Tellus B 51, 16 (1999). 8 Krinner, G. et al. A dynamic global vegetation model for studies of the coupled atmosphere-biosphere system. Global Biogeochemical Cycles 19, GB1015 (2005). 9 Dufresne, J. L. et al. Climate change projections using the IPSL-CM5 Earth System Model: from CMIP3 to CMIP5. Climate Dynamics 40, (2013). 10 Mahecha, M. D. et al. Global Convergence in the Temperature Sensitivity of Respiration at Ecosystem Level. Science 329, (2010). 11 Friedlingstein, P. et al. Climate Carbon Cycle Feedback Analysis: Results from the C4MIP Model Intercomparison. Journal of Climate 19, (2006). 12 Friedlingstein, P. et al. On the contribution of CO2 fertilization to the missing biospheric sink. Global Biogeochem. Cycles 9, (1995). 13 Meinshausen, M., Raper, S. C. B. & Wigley, T. M. L. Emulating coupled atmosphereocean and carbon cycle models with a simpler model, MAGICC6 Part 1: Model description and calibration. Atmos. Chem. Phys. 11, (2011). 14 Myhre, G., Highwood, E. J., Shine, K. P. & Stordal, F. New estimates of radiative forcing due to well mixed greenhouse gases. Geophysical Research Letters 25, (1998). 15 Boucher, O. & Reddy, M. S. Climate trade-off between black carbon and carbon dioxide emissions. Energ Policy 36, (2008). 12 NATURE CLIMATE CHANGE

13 SUPPLEMENTARY INFORMATION 16 Taylor, K. E., Stouffer, R. J. & Meehl, G. A. An Overview of CMIP5 and the Experiment Design. Bulletin of the American Meteorological Society 93, (2011). 17 Joos, F. et al. Carbon dioxide and climate impulse response functions for the computation of greenhouse gas metrics: a multi-model analysis. Atmos. Chem. Phys. 13, (2013). 18 Myhre, G., et al. Anthropogenic and Natural Radiative Forcing. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA (2013). 19 Forster, P. et al. Changes in Atmospheric Constituents and in Radiative Forcing. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA (2007). 20 Gonçalves, J. L. M. et al. Soil fertility and growth of Eucalyptus grandis in Brazil under different residue management practices. Southern Hemisphere Forestry Journal 69, (2007). 21 Amiro, B. D. et al. Ecosystem carbon dioxide fluxes after disturbance in forests of North America. J. Geophys. Res. 115, G00K02 (2010). 22 Hansen, M. H., Perry, C. H., Brand, G. & McRoberts, R. Wisconsin s Forests, 2004: Statistics and Quality Assurance. USDA Forest Service, Delaware, US (2008). 23 Magnani, F. et al. The human footprint in the carbon cycle of temperate and boreal forests. Nature 447, (2007). 24 Cherubini, F., Strømman, A. H. & Hertwich, E. Effects of boreal forest management practices on the climate impact of CO2 emissions from bioenergy. Ecological Modelling 223, (2011). 25 Guest, G., Bright, R. M., Cherubini, F. & Strømman, A. H. Consistent quantification of climate impacts due to biogenic carbon storage across a range of bio-product systems. Environ Impact Asses 43, (2013). 26 Gillett, N. P., Arora, V. K., Matthews, D. & Allen, M. R. Constraining the Ratio of Global Warming to Cumulative CO2 Emissions Using CMIP5 Simulations*. Journal of Climate 26, (2013). 27 Matthews, H. D., Gillett, N. P., Stott, P. A. & Zickfeld, K. The proportionality of global warming to cumulative carbon emissions. Nature 459, (2009). NATURE CLIMATE CHANGE 13

14 28 Smith, S. M. et al. Equivalence of greenhouse-gas emissions for peak temperature limits. Nature Clim. Change 2 (2012). 14 NATURE CLIMATE CHANGE