Fire vs . fossil fuel : all CO 2 emissions are not created equal

Introduction Conclusions References Tables Figures


Introduction
Fire is a conspicuous disturbance in most terrestrial ecosystems, with considerable impacts on vegetation itself, carbon cycling, land-atmosphere exchanges, and climate in general (Bonan, 2008;Running, 2008;Bowman et al., 2009).Fire currently affects around 300-500 Mha yr −1 , leading to gross emissions of 1.5-3 Pg C yr −1 from the direct combustion of vegetation and soil-litter (Kloster et al., 2010;Mieville et al., 2010;Thonicke et al., 2010;van der Werf et al., 2010;Randerson et al., 2012;Giglio et al., 2013;Li et al., 2014).The potential for modifications in the current fire regime to modulate climate change stimulated the explicit representation of fire in the Lund-Potsdam-Jena (LPJ) Dynamic Global Vegetation Model (DGVM; Thonicke et al., 2001), and later on into various other similar process-based models of climate-vegetation interactions.These efforts have paved the way to studies that projected an increase in fire frequency and gross CO 2 emissions over the 21st century (Scholze et al., 2006;Pechony and Shindell, 2010;Kloster et al., 2012).
The net effect of fire on carbon cycling has however received less attention than the consequences from future climate-fire feedbacks.In their seminal study, Seiler and Crutzen (1980) concluded that net biospheric emissions, coming mostly from fire, could range between ±2 Pg C yr −1 by adding the effects of vegetation regrowth and other processes to their estimate of 2-4 Pg C yr −1 for gross fire emissions.The net effect of fire on terrestrial carbon storage has then apparently been left unaddressed for more than three decades, until Ward et al. ( 2012) suggested a fire-caused net reduction of ∼ 500 Pg C in pre-industrial global land carbon.They also found that this reduction could currently be slightly lower (around 425 Pg C) due to offsetting effects between fire and land-use and land cover changes (LULCC), but could increase to about 550-650 Pg C by the end of this century due to a climate-driven increase in fire activity.More recently, Li et al. (2014) concluded that net fire emissions were equal to 1.0 Pg C yr −1 on average during the 20th century, compared to gross emissions of 1.9 Pg C yr −1 on average over the same period.The latter two studies were however Figures performed in offline terrestrial models and were therefore unable to account for various climate-fire feedbacks, including fire-induced CO 2 fertilization and the impact of changes in surface albedo on temperature.To date, the only study dedicated to fire in a coupled climate-carbon model with interactive vegetation dealt primarily with the consequences of major changes in future fire regime, but also found that net CO 2 emissions following changes in fire regime quickly became much smaller than gross emissions and progressively decreased over time (Landry et al., 2015a).
The dearth of studies dedicated to fire in coupled climate-carbon models has led to potentially inaccurate methods for estimating the climatic effects of fire CO 2 emissions.Indeed, previous studies had to rely upon results from simulations of fossil fuel emissions in order to estimate the fate of fire-emitted CO 2 (Randerson et al., 2006;O'Halloran et al., 2012).A convenient way to proceed consists of combining fire-caused land-atmosphere CO 2 exchanges based on empirical or offline modelling data with a fossil fuel-derived impulse response function (IRF).IRFs give the proportion of a single pulse of CO 2 emissions that remain airborne as a function of time (t, in years) and are usually expressed as a sum of three decaying exponentials with a constant term (Joos et al., 2013): where {a i } are unitless and {τ i } are in years.Such IRF-based approaches have also been used in other contexts, for example to quantify the fate of atmospheric CO 2 anomalies resulting from boreal peatlands forestation (Lohila et al., 2010), boreal forest biofuels (Bright et al., 2011), and other disturbances like insect outbreaks and hurricanes (O'Halloran et al., 2012).Yet in the case of fire at least, estimating carbon and temperature effects based on simulations of fossil fuel emissions appears questionable due to the major differences involved.First, fossil fuel emissions entail a net transfer of CO 2 from geological reservoirs to the much more active atmospheric, oceanic, and terrestrial carbon pools, whereas fire simply redistributes the carbon al-Introduction

Conclusions References
Tables Figures

Back Close
Full ready existing in these three pools.Second, contrary to fossil fuel emissions, fire directly triggers a strong vegetation regrowth response and substantial modifications to land-atmosphere exchanges of energy through altered surface albedo and sensible/latent heat partitioning (Bremer and Ham, 1999;van der Werf et al., 2003;Amiro et al., 2006;Goulden et al., 2011).
In this study, we used a coupled climate-carbon model with interactive vegetation to advance the current knowledge regarding the effects of fire CO 2 emissions on the global carbon cycle and temperature.We focussed on non-deforestation fires that allow the different vegetation types to compete and grow back in the recently burned area, because they constitute the bulk of global burned area and gross emissions (van der Werf et al., 2010) and have been much less represented in climate models than the LULCC events associated with deforestation fires.Our main objective is to compare the long-term effects of fire CO 2 emissions to corresponding levels of fossil fuel CO 2 emissions, for single fire pulses and stable fire regimes.A second objective is to quantify the differences between gross and net fire CO 2 emissions over 1000 years following major changes in fire frequency.To facilitate the interpretation of results, we performed all simulations against a background climate corresponding to pre-industrial conditions.

Modelling of fire and fossil fuel effects
We used the University of Victoria Earth System Climate Model (UVic ESCM) version 2.9 to study the climatic effects of fire and fossil fuel CO 2 emissions.The UVic ESCM computes at a resolution of 3.6 • × 1.8 • (longitude × latitude) the exchanges of carbon, energy, and water among the land, atmosphere, and ocean (Weaver et al., 2001;Eby et al., 2009).The land module consists of a simplified version of the MOSES land surface scheme (Meissner et al., 2003) coupled to the TRIFFID DGVM (Cox, 2001).Introduction

Conclusions References
Tables Figures

Back Close
Full broadleaf tree, needleleaf tree, C 3 grass, C 4 grass, and shrub, accounting for the dynamics of different carbon pools for vegetation (leaves, stem, and roots) and soil-litter.The UVic ESCM computes the atmospheric energy and moisture balance with dynamical feedbacks, and its ocean module represents three-dimensional circulation, sea ice dynamics and thermodynamics, inorganic carbon, and ecosystem/biogeochemical exchanges (Weaver et al., 2001;Ewen et al., 2004;Schmittner et al., 2008;Eby et al., 2009).The UVic ESCM can account for various types of prescribed forcings, including the emissions of CO 2 , other greenhouse gases, and sulphate aerosols, land cover changes, volcanic aerosols, and land ice (Weaver et al., 2001;Matthews et al., 2004).In this study, we also used the UVic ESCM fire module developed by Landry et al. (2015a).In each grid cell, this module estimated the gross CO 2 emissions coming from combustion as the product of prescribed burned area (see Sect. 2.2), fuel density (simulated by the UVic ESCM), and PFT-specific combustion fractions for the different fuel types (Table 1).The carbon contained in the vegetation killed by fire but not combusted was transferred to the soil-litter pool, where it decomposed and released additional CO 2 at a rate that depended upon the simulated soil temperature and moisture.Since we were interested in non-deforestation fires, the different PFTs could compete and grow back in the recently burned area, giving rise to a regrowth CO 2 flux influenced by the climatecarbon feedbacks simulated by the UVic ESCM (e.g., fire-induced CO 2 fertilization and temperature changes).The model further accounted for the post-fire changes in land surface exchanges due to the modified vegetation cover, including the increase in land surface albedo (α L , unitless).In all simulations, we included only the CO 2 -related effects of fire and fossil fuel combustion, and not the associated aerosols and non-CO 2 greenhouse gases.Similarly, we did not include here the short-term albedo decrease due to surface blackening.Figures

Back Close
Full

Prescribed burned area
We based the prescribed burned area on the January 2001 to December 2012 monthly data from version 4 of the Global Fire Emissions Database (GFED4), which was derived from satellite observations (Giglio et al., 2013).We then simplified the GFED4 dataset in order to retain its most essential features only.Each grid cell from the UVic ESCM was labelled as a "fire cell" if it had been affected by fire at least once over the 2001-2012 period according to GFED4.The main simplification here was that the burned area fraction was set equal across all the UVic ESCM fire cells, with the specific burned area fraction value varying across fire simulations (see Sect. 2.3).The use of this binary distribution of burned area fractions (i.e., the same value for all fire cells and zero for all other cells) was necessary in order to reach the target fire CO 2 emissions while ensuring that the burned area fractions were proportional for all fire cells across the different fire simulations.(Given that the actual burned area fractions are already relatively close to 100 % in various regions (Giglio et al., 2013), upscaling the original GFED4 data would not have resulted in the same relative changes for all fire cells.) Fire happened one time per year in each of the UVic ESCM fire cells, during the month of highest burned area according to the mean 2001-2012 value from GFED4 data.

Simulation design
We started with an equilibrium run of the climate system for the year 1750, using the prescribed forcings from Eby et al. (2013) for solar radiation, atmospheric CO 2 (fixed at 277 ppmv), non-CO 2 greenhouse gases, land cover changes, land ice, and volcanic aerosols.Five groups of transient simulations then branched off from this equilibrated climate, in addition to a control transient simulation; in all cases, the forcings from year 1750 were maintained, except that the climate and carbon cycle were free to respond to the effects of the fire and fossil fuel experiments.
First, we performed three simulations that each consisted of a single year of fire activity, followed by a return towards the pre-fire equilibrium conditions.The resulting Introduction

Conclusions References
Tables Figures

Back Close
Full fire pulses had sizes of 20, 100, and 200 Pg C, based on their gross emissions (i.e., the carbon released from combustion only).We obtained these fire CO 2 pulses by adjusting the single-year burned area fraction across all fire cells and designate these simulations as Fire20P, Fire100P, and Fire200P.Second, we performed another set of fire experiments similar to the previous ones, except that the same burned area fractions were maintained year after year.We designate these stable fire regime as Fire20S, Fire100S, and Fire200S, corresponding to the previous fire pulse experiments of 20, 100, and 200 Pg C, respectively.Third, we injected fossil fuel CO 2 pulses of 20, 100, and 200 Pg C into the atmosphere over a single year.The purpose of this set of three simulations was to compare the effects from fossil fuel CO 2 emissions vs. the same amount (and timing) of gross fire emissions.We designate these simulations as FF20P-G, FF100P-G, and FF200P-G.
Fourth, we wanted to compare the effects from fossil fuel CO 2 emissions vs. the same amount (and timing) of net fire emissions following each fire pulse.In addition to the CO 2 released by combustion, net fire emissions included post-fire vegetation regrowth, decomposition of the vegetation that was killed but not combusted, and climate-carbon feedbacks.Each year, we computed the net fire emissions (land to atmosphere) as the annual change in total land carbon for the control simulation, minus the annual change in total land carbon following the fire pulse (Fire20P, Fire100P, or Fire200P).We then injected into the atmosphere yearly fossil fuel CO 2 emissions that were equal to these net fire emissions, including when they were negative.We designate these simulations as FF20P-N, FF100P-N, and FF200P-N.
Fifth, we performed a set of three fossil fuel experiments in which the yearly fossil fuel CO 2 emissions were this time equal to the net emissions from the Fire20S, Fire100S, and Fire200S stable fire regimes.We designate this last set of simulations as FF20S-N,

Assessment of the UVic ESCM fire module
The burned area fractions (unitless) in the fire cells for the 20, 100, and 200 Pg C pulses were approximately equal to 0.09, 0.45, and 0.88, respectively.Since the 200 Pg C pulse led to the burning of almost all the area within the fire cells, we used the results of this simulation to assess the post-fire simulated responses for changes in PFT cover, total biomass, and α L in different ecosystem types (Fig. 1).In northern forests, the succession among the different PFTs (Fig. 1a) agreed with observation-based trajectories (Rogers et al., 2013), while the impacts on biomass (Fig. 1c) and α L (Fig. 1e) were consistent with field observations (Amiro et al., 2006;Goulden et al., 2011).The overall slower return to pre-fire conditions compared to observations came from the lasting climatic effects from the extreme 200 Pg C fire pulse (see Sect. 3.2).As expected (van der Werf et al., 2003;Ward et al., 2012), the return to pre-fire conditions was much faster in savannas (Fig. 1b, d, and f).Note that the very small increase in total biomass soon after the fire pulse (Fig. 1d) and the associated marginal decrease in α L (Fig. 1f; not visible) likely came from the CO 2 fertilization effect caused by the long-lasting atmospheric CO 2 anomaly (see Sect. 3.2).Additional simulations performed by Landry et al. (2015a) further established the realism of results from the UVic ESCM fire module.First, they obtained gross fire CO 2 emissions of 2.2 Pg C yr −1 for the current fire regime, comparable to previous studies (Kloster et al., 2010;Mieville et al., 2010;Thonicke et al., 2010;van der Werf et al., 2010;Randerson et al., 2012;Li et al., 2014).The splitting of these gross emissions between vegetation (0. ( Ward et al., 2012) (note that we did not include surface blackening in the current study).

Single fire pulse
The atmosphere, ocean, and land carbon pools responded as previously reported (Archer et al., 2009;Eby et al., 2009Eby et al., , 2013;;Joos et al., 2013) to the fossil fuel CO 2 pulses (Fig. 2a).Part of the CO 2 injected into the atmosphere progressively became absorbed by the land and ocean, so that 1000 years after the pulses, 60 % of the additional CO 2 was found into the ocean and the remaining 40 % was divided almost equally between the land and atmosphere.The limited absolute difference among the pulse magnitudes studied here (i.e., 180 Pg C) explains why the responses were almost identical in the three cases, contrary to what has been found for a larger range of pulse magnitudes (Archer et al., 2009;Eby et al., 2009;Joos et al., 2013).Fire effects (Fig. 2b) differed substantially from the fossil fuel pulse results.This time the CO 2 injected into the atmosphere came from the land, resulting in decreased land carbon rather than increased land carbon as in the case of fossil fuel.Instead of leading to long-lasting changes, the fire pulses were followed by a gradual return towards the initial equilibrium conditions.These two features illustrate a fundamental distinction between fossil fuel and fire: fossil fuel emissions represent a near-permanent addition of CO 2 to the active (i.e., non-geological) carbon cycling pools, whereas fire pulses temporarily reshuffle the carbon already existing in these pools.Moreover, the responses varied noticeably among the three fire pulses.Finally, fractional changes greater than 1.0 were observed for the atmosphere and land shortly after the pulses because the net emissions (i.e., including the decomposition of the uncombusted vegetation killed by fire) were initially higher than the gross emissions upon which the magnitude of the pulses were defined.systematically higher for fossil fuel than for fire after about a century.Consequently, the IRF parameters differed considerably among the three fire pulses, as well as with the fossil fuel pulses (Table 2).Even if the airborne fraction behaviour was more complex for fire than for fossil fuel (Fig. 2c), the goodness of fit between the IRFs and the corresponding data was similar for both types of pulses (Table 2).Note that the physical meaning of the fire IRF parameters should not be over-emphasized, as the fit of a sum of exponential functions to data is notoriously sensitive to noise (de Groen and de Moor, 1987).These differences in the effects from fire vs. fossil fuel emissions on the carbon cycle then affected the global mean atmospheric surface temperature (T s , in K), as shown in Fig. 3a.Fossil fuel CO 2 emission pulses caused relatively stable increases in T s over millennial timescales (Matthews and Caldeira, 2008;Eby et al., 2009).In the case of fire pulses, the return of atmospheric CO 2 towards pre-fire levels (Fig. 2b) resulted in smaller warming of much shorter duration.Atmospheric CO 2 even decreased below the control level ∼ 400-500 years after the pulses, which contributed to the observed long-term net cooling effect particularly visible for Fire200P.This slight decrease in atmospheric CO 2 came from the long time needed before the ocean returned to the atmosphere all the carbon absorbed following the fire pulses.
Albedo was also involved in the diverging effects of fire vs. fossil fuel on T s (Fig. 3b).Fossil fuel-induced CO 2 fertilization slightly decreased α L (Matthews, 2007) over the whole simulation period, whereas fire noticeably increased α L for decades to centuries.Note that contrary to the situation illustrated in Fig. 1a, in some northern grid cells tree cover had not fully recovered yet to pre-fire levels 1000 years after the 200 Pg C fire pulse.This lasting increase in α L contributed to the net cooling following the fire pulses.
All previous outcomes illustrate that the effects from fire vs. fossil fuel CO 2 emissions differ fundamentally for identical pulse magnitude defined in terms of gross (i.e., combustion only) fire emissions.Now, what if fossil fuel emissions were instead set equal to the net land-to-atmosphere emissions from fire?In this case, the impacts on land carbon remained opposite because emissions came from the land for fire but not Introduction

Conclusions References
Tables Figures

Back Close
Full for fossil fuel; for the atmosphere, however, the CO 2 anomalies were much more similar (Fig. 4a vs. Fig.2b).Yet a closer look at the results reveals that the atmospheric anomalies were not actually equal (Fig. 4b).During the first ∼ 250 years, these anomalies were systematically lower for fossil fuel because the vegetation absorbed a portion of the emitted CO 2 , whereas for fire the net emissions already accounted, by definition, for vegetation regrowth and all climate-carbon feedbacks.As a result, the ocean absorbed more carbon for fire than for fossil fuel emissions (Fig. 4a vs. Fig.2b).
Based on atmospheric CO 2 alone, one would thus expect T s to be higher for fire than for fossil fuels, yet the opposite was in fact observed (Fig. 4c) due to the opposite impacts on α L (Fig. 4d).Note that in the long term, these ∆T s were however much smaller than when fossil fuel emissions were equal to gross fire emissions (Fig. 3a).The fact that atmospheric CO 2 anomalies became slightly lower for fire than for fossil fuel after about 250 years (Fig. 4b; not visible) can be explained by long-lasting impacts on ocean carbon cycling: compared with fossil fuel, the ocean absorbed substantially more carbon in the initial decades after the fire pulses, and then took more time to outgas this carbon when the atmosphere-ocean fluxes shifted sign during the return towards the initial equilibrium conditions.

Stable fire regime
The previous results provide relevant information regarding fundamental differences between fire and fossil fuel CO 2 emissions, but were based on single pulses of fire activity.We now turn to stable fire regimes for which the burned area fraction was maintained year after year, instead of being applied only once as in the pulse experiments.Figure 5 shows that the resulting gross and net emissions had qualitatively similar behaviours for the three stable regimes.Both the gross and net yearly emissions decreased quickly after an initial spike.The yearly net emissions progressively stabilized close to zero, although their mean value was still positive towards the end of the simulations as indicated by the slight positive slope of the cumulative net emissions.The yearly gross emissions, on the other hand, stabilized around much higher values 15196 Introduction

Conclusions References
Tables Figures

Back Close
Full because vegetation and soil-litter kept being combusted each year.Contrary to net emissions, the cumulative gross emissions thus increased almost linearly ∼ 50 years after the onset of fire activity and onwards (results not shown).
Gross emissions thus appear highly inadequate to assess the cumulative impacts of fire regime shifts.Indeed, yearly gross emissions towards the end of the simulations were higher for Fire100S than for Fire200S, even though the outcome was obviously the opposite for the cumulative net emissions (Table 3).The lower land carbon density caused by more frequent fires has previously been observed to result in a "saturation effect" of gross emissions (Landry et al., 2015a); here, this effect was so large that gross emissions ended up being lower for Fire200S than for Fire100S about 50 years after the onset of fire activity.A similar saturation effect clearly affected the cumulative net emissions, which were only twice as large for Fire200S compared to Fire20S, whereas the equilibrium yearly burned area was 12 times larger for Fire200S vs. Fire20S (Table 3).This slightly supra-linear scaling in burned area (e.g., 12 times instead of 10 times larger for Fire200S vs. Fire20S) among stable fire regimes was caused by fire-induced changes in vegetation composition.The input prescribed burned area in each fire cell (see Sect. 2.2) actually corresponds to a gross value that is reduced to account for the PFT-specific unburned islands occurring within burn perimeters (Kloster et al., 2010;van der Werf et al., 2010).More frequent fires led to increases in grass cover at the expense of trees and shrubs, thereby increasing the net burned area.The cumulative gross emissions at the end of the simulations were around 8, 23, and 21 Eg C for Fire20S, Fire100S, and Fire200S, respectively.These values were much higher than the corresponding cumulative net emissions (Fig. 5 and Table 3) -and, for the two most severe regimes, were in fact even higher than the estimated fossil fuel total resource base (Stewart and Weaver, 2012).The injection of such amounts of fossil fuel CO 2 into the atmosphere would obviously result in much more severe impacts on the carbon cycle and temperature (Matthews and Caldeira, 2008;Archer et al., 2009;Eby et al., 2009;Joos et al., 2013) than were observed for the three stable fire regimes.Introduction

Conclusions References
Tables Figures

Back Close
Full Even for fossil fuel emissions that were equal to the net emissions from stable fire regimes, the effects differed once again.Figure 6a shows the distribution of net cumulative emissions (i.e., from year 0 until the specific year considered) from fossil fuel among the active carbon pools.This splitting was similar to the one following a single fossil fuel pulse (Fig. 2a), except that the maximum land uptake was proportionally lower and the ocean took a little longer to become the main carbon sink.For fire (Fig. 6b), land carbon rather decreased (with a fractional change equal to −1.0 as the net emissions were, by definition, equal to the total change in land carbon) and the uptake of carbon by the ocean had to be substantially higher than for fossil fuel.
The airborne fraction of the net emissions from stable fire regimes was initially higher than for the same amount of emissions from fossil fuel, but the anomalies in atmospheric CO 2 progressively became more similar (Fig. 7a).This should have caused T s to be higher for fire than for fossil fuel, yet once again the opposite was observed (Fig. 7b).Cumulative fossil fuel CO 2 emissions led to T s increases that were relatively stable over thousands of years (Matthews and Caldeira, 2008;Eby et al., 2009).For fire, on the other hand, the initial increase in T s after the onset of fire activity was followed ∼ 50-100 years later by a gradual decrease in T s .As was the case for the pulse simulations (see Sect. 3.2), this opposite effect of fire vs. fossil fuel emissions on T s was related to changes in land albedo, which substantially increased for fire due to changes in vegetation cover, but slightly decreased for fossil fuel due to CO 2 fertilization (Fig. 7c).

Fundamental differences between fire and fossil fuel
In this study, we have shown a consistent pattern of fundamental differences between the carbon cycle and climate effects of CO 2 emitted by fire as compared to fossil fuel combustion.First, the sources of CO 2 emissions are qualitatively distinct: fire sim-Introduction

Conclusions References
Tables Figures

Back Close
Full ply reshuffles carbon among the active pools, whereas fossil fuel combustion entails a net carbon transfer from the geological to the active pools over millennial time scales (Archer et al., 2009;Eby et al., 2009).Second, the terrestrial pools (vegetation plus soil-litter) cannot respond in the same way to the atmospheric CO 2 anomalies created by fire vs. fossil fuel emissions.The only direct effect (i.e., excluding climate change) of fossil fuel emissions on land carbon storage occurs through the CO 2 fertilization effect.Fire, on the other hand, gives rise to a much more dynamic land carbon response.The combustion of land carbon and the further decomposition of killed but uncombusted vegetation constitute not only sources of fire emissions, but also decrease the amount of vegetation that can instantaneously be fertilized by the fire-induced increase in atmospheric CO 2 .Subsequently, however, vegetation regrowth and the associated soil-litter build up in the burned patches act as strong carbon sinks.Third, these contrasting effects on terrestrial vegetation mean opposing changes in land albedo: fire-induced decrease in vegetation cover increases α L , whereas fossil fuel-induced CO 2 fertilization decreases α L through dynamic vegetation changes like increased shrub and tree cover in tundra (Matthews, 2007) and generally higher leaf and stem area index for the vegetation already in place (Bala et al., 2013).This divergence in α L responses implies unequal T s changes, which then feed back to affect the carbon cycle itself.Therefore, the effects on carbon cycling and temperature are incongruent even when fossil fuel emissions are equal to the net emissions from fire.
Other variables than carbon pools and α L were affected by these different changes in T s and amplified them.Sea ice area, for example, often diverged noticeably between corresponding fossil fuel and fire simulations.For FF100P-G and FF200P-G, there was a small (∼ 2 % and ∼ 4 %, respectively) but permanent decrease in global sea ice area that did not occur in the corresponding fire simulations.For FF100P-N and FF200P-N, sea ice area also decreased a little for a few centuries at least before gradually returning toward initial levels.(For FF20P-G and FF20P-N, the changes in global sea ice area were indistinguishable from internal variability.)For fire pulses, on the other hand, the substantial ∆α L -based cooling over the Northern Hemisphere due to extensive land Introduction

Conclusions References
Tables Figures

Back Close
Full masses slightly increased Arctic sea ice area; note that ∆α L had a much smaller absolute influence on Antarctic sea ice, for which the changes were highly variable spatially.Such transfer of α L -induced cooling to the surrounding ocean has also been observed following deforestation simulations, along with an additional decrease in atmospheric temperature over most latitudes resulting from the lower ocean temperature (Davin and de Noblet-Ducoudré, 2010).In our simulations of stable fire regimes and the corresponding fossil fuel experiments, changes in sea ice area were much larger due to higher net CO 2 emissions.For fossil fuel, sea ice area was permanently reduced in all simulations.For fire, the ∆α L -based cooling was not strong enough this time to prevent major losses of both Arctic and Antarctic sea ice, because the atmospheric CO 2 anomalies were larger and longer-lasting than following a single fire pulse.However, the increase in α L helped maintaining lower temperatures for the stable fire regimes than for the corresponding fossil fuel simulations, and global sea ice area progressively recovered to the control level, albeit with spatial differences between the Arctic and Antarctic that matched the hemispherical changes in atmospheric temperature.
These fundamental differences imply that fire impacts cannot be accurately estimated from simulations of fossil fuel emissions in climate models.We already illustrated the validity of this claim for fossil fuel emissions that were equal to the net emissions from fire, for single fire pulses (Fig. 4) and stable fire regimes (Fig. 7).Here, we further assess two other adjustments based on approaches that have been used in previous studies of single fire events.The first approach consists of performing an offline estimate of the land-atmosphere CO 2 fluxes triggered by fire, and then estimating the oceanic uptake of the remaining atmospheric CO 2 anomaly based on atmosphereocean exchanges following the injection of a fossil fuel pulse in a climate model (Randerson et al., 2006).We reproduced this approach by combining results from the UVic ESCM simulations of fire (land-atmosphere CO 2 fluxes) and fossil fuel (atmosphereocean CO 2 fluxes) pulses.As shown in Fig. 8a, this approach substantially underestimated the fire-caused atmospheric CO 2 anomalies compared to the actual results from the UVic ESCM.The second approach consists of performing an offline estimate of the Introduction

Conclusions References
Tables Figures

Back Close
Full yearly land-atmosphere CO 2 fluxes triggered by fire, and then applying an IRF obtained from fossil fuel simulations to each of these yearly land-atmosphere CO 2 fluxes (O'Halloran et al., 2012).We reproduced this approach by combining the UVic ESCM land-atmosphere CO 2 fluxes from fire simulations with the appropriate fossil fuel IRF from Table 2, depending upon the magnitude of the fire pulse.The bias for this second approach was initially even more negative, but decreased quickly following vegetation regrowth and ended up being slightly positive (Fig. 8b).In fact, the results from this second approach were very similar to the ones obtained from fossil fuel emissions that were equal to net fire emissions (Fig. 4b).Note that our assessment of these two fossil fuel-based adjustments was conservative, because the UVic ESCM results we used for the land-atmosphere CO 2 fluxes actually accounted for climate-fire feedbacks in a much more comprehensive way that offline simulations could do.

Study limitations
The outcomes of our study should be interpreted with four caveats in mind.First, we developed idealized fire regimes in order to obtain substantial fire impacts while facilitating the comparison of results across the different magnitudes of pulses or stable regimes.Our fire regimes were therefore more severe than the current situation on Earth, as seen with our equilibrium results of ≥ 0.9 Gha yr −1 for burned area and ≥ 7.3 Pg C yr −1 for gross emissions under stable regimes (Table 3), vs. current values of 0.3-0.5 Gha yr −1 and 1.5-3 Pg C yr −1 , respectively (Kloster et al., 2010;Mieville et al., 2010;Thonicke et al., 2010;van der Werf et al., 2010;Randerson et al., 2012;Giglio et al., 2013;Li et al., 2014).Moreover, our "equal" spatial fire patterns (i.e., same burned area fraction in each fire cell) gave much more weight to fires in extra-tropical regions compared with the current fire distribution (Giglio et al., 2013).Despite the differences in vegetation regrowth and fire-caused changes in albedo among regions, the impacts on atmospheric CO 2 and T s did not seem overly sensitive to changes in the distribution of burned area fraction among fire cells following a single fire pulse (Fig. 9).Introduction

Conclusions References
Tables Figures

Back Close
Full Second, we neglected all non-CO 2 emissions from fire and fossil fuel.Accounting for fire non-CO 2 emissions would reduce the albedo cooling effect, due to the shortterm post-fire surface blackening caused by char.On the other hand, explicitly tracking all the patches created by individual fire events, instead of representing their average grid-level effect as we did here, would increase the simulated albedo cooling effect over boreal forests at least (Landry et al., 2015b), although the impact would likely be minor for the Fire200P and Fire200S simulations in which the burned area fraction was close to 90 % in each fire cell.Furthermore, the fire-caused emissions of aerosols and non-CO 2 greenhouse gases into the atmosphere would have a much stronger impact on T s than changes in surface albedo; however, the magnitude and even the sign of the climatic effect from these non-CO 2 atmospheric emissions remain highly uncertain (Jacobson, 2004(Jacobson, , 2014;;Jones et al., 2007;Unger et al., 2010;Ward et al., 2012;Landry et al., 2015a).Future studies on the differences in the carbon cycling and temperature impacts between fire and fossil fuel would nevertheless benefit from combining the effects of non-CO 2 emissions with climate-carbon feedbacks in climate models including interactive vegetation.
Third, the UVic ESCM does not currently simulate the non-trivial exchanges of carbon between land and ocean (Regnier et al., 2013) or between inland waters and the atmosphere (Raymond et al., 2013), which are also impacted by fire.For example, the land-to-ocean flux of all particulate and dissolved pyrogenic carbon could be as high as ∼ 50-100 Tg C yr −1 (Bird et al., 2015).More research is therefore needed to accurately represent the highly variable and poorly quantified fate of such exchanges of pyrogenic carbon in climate models; meanwhile, their influence on our results is speculative, but is unlikely to challenge the main outcomes we obtained.
Fourth, our study addressed only non-deforestation fires after which the natural vegetation is free to recover.One might argue that our stable fire regimes are similar to deforestation fires because, over large spatial scales, both fire types decrease terrestrial carbon storage and vegetation cover.However, our non-deforestation fires affected equally all fire cells, whereas deforestation fires are deemed exclusive to tropical re-Introduction

Conclusions References
Tables Figures

Back Close
Full gions (van der Werf et al., 2010).Given that fire-induced changes in terrestrial carbon density and albedo vary substantially among regions, we caution against the direct extrapolation of our results to deforestation fires.In fact, when neglecting non-CO 2 emissions, deforestation fires are conceptually more similar to other sources of LULCC than to non-deforestation fires.Note that previous global-scale climatic studies of LULCC (see Pongratz et al., 2014 for an extensive list) have represented all LULCC sources in the same way.Yet the variations in delayed CO 2 fluxes between fire and other LULCC sources matter for carbon cycling (Ramankutty et al., 2007;Houghton et al., 2012) and, as mentioned previously, non-CO 2 emissions could have a dominant impact on the climate.Consequently, studies dedicated to deforestation fires that specifically represent their delayed CO 2 fluxes and go beyond CO 2 emissions would allow for a more refined understanding of their climatic impacts.

Conclusions
The main purpose of this study was to illustrate the fundamental differences in the effects from fire vs. fossil fuel CO 2 emissions on the global carbon cycle and temperature.
To do so, we simulated fire pulses and stable fire regimes of various magnitudes, as well as the corresponding fossil fuel emissions.The main outcomes we obtained were the following.
-The carbon sink stemming from vegetation regrowth led to widely diverging longterm impacts on the carbon cycle and temperature when fossil fuel emissions were equal to the gross emissions (i.e., based on combustion only) from a fire pulse, with the opposing changes in land surface albedo further compounding these discrepancies (Figs. 2 and 3, and Table 2).Side-by-side comparisons of gross fire CO 2 emissions to fossil fuel emissions are thus misleading and should be avoided.Introduction

Conclusions References
Tables Figures

Back Close
Full -The impacts still differed, although much less severely, when fossil fuel emissions were equal to the net emissions following a fire pulse (Fig. 4).These results point towards the existence of irreconcilable disparities between the effects from fire vs. fossil fuel, a claim that was also supported by the shortcomings of two other possible adjustments aiming to estimate fire effects based on simulations of fossil fuel emissions in climate models (Fig. 8).
-Obvious differences also arose when fossil fuel emissions were equal to the net emissions caused by stable fire regimes, particularly for land carbon, oceanic carbon, surface temperature, and land surface albedo (Figs. 6 and 7).
Our results also shed light on the evolution of gross vs.net fire emissions following fire regime changes.While non-zero gross emissions were maintained indefinitely following a stable fire regime change, most of the net emissions actually occurred relatively quickly after the regime shift and net emissions progressively decreased to almost zero (Fig. 5).These results illustrate how inadequate it would be to represent the effects of fire regime changes by fossil fuel CO 2 emissions equal to gross fire emissions.
Furthermore, a higher increase in fire frequency could result in lower equilibrium gross emissions due to the fire-induced decrease in the amount of fuel available (Table 3).
Changes in gross emissions offered therefore a poor indicator of fire impacts on the carbon cycle.

FF100S
7 Pg C yr −1 ) and soil-litter (1.5 Pg C yr −1 ) also agreed with GFEDbased estimates (van der Werf et al., 2010).Second, the differences in α L between the current fire regime and a no-fire world simulated by Landry et al. (2015a) led to a global radiative forcing of −0.11 W m −2 without the effect of surface blackening and −0.07 W m −2 with surface blackening, in agreement with observation-based estimates Discussion Paper | Discussion Paper | Discussion Paper |

Figure
Figure 2c compares the airborne fraction of the CO 2 pulses from fossil fuel vs. fire.All results were similar during ∼ 25 years following the pulses, and for up to ∼ 50 years for Fire100P and the different fossil fuel pulses.However, the airborne fraction became 15194 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Author contributions.J.-S.Landry and H. D. Matthews designed the study, J.-S.Landry modified the UVic ESCM with advice from H. D. Matthews, J.-S.Landry performed the simula-Discussion Paper | Discussion Paper | Discussion Paper | tions and analyzed the results, J.-S.Landry prepared the manuscript with contributions from HDiscussion Paper | Discussion Paper | Discussion Paper | ing albedo and greenhouse gas fluxes on radiative forcing, J. Geophys.Res., 115, G4011, doi:10.1029/2010JG001327,2010.15188 Matthews, H. D.: Implications of CO 2 fertilization for future climate change in a coupled climatecarbon model, Glob.Change Biol., 13, 1068-1078, 2007.15195, 15199 Matthews, H. D. and Caldeira, K.: Stabilizing climate requires near-zero emissions, Geophys.Discussion Paper | Discussion Paper | Discussion Paper | trace gas emissions: results from a process-based model, Biogeosciences Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |

Figure 1 .
Figure 1.Changes due to the 200 Pg C fire pulse happening on year zero; each data point gives the mean value over 50 years (25 years before and 25 years after).Results are for a forested grid cell in North America (centered on 53.1 • N, 124.2 • W; panels a, c, and e) and a savanna grid cell in Africa (centered on 13.5 • N, 12.6 • E; panels b, d, and f).(a, b) Fractional cover of the different plant functional types.(c, d) Total biomass.(e, f) Land surface albedo.

Figure 2 .Figure 3 .Figure 4 .Figure 5 .Figure 6 .Figure 7 .Figure 8 .Figure 9 .
Figure 2. Changes in global carbon stocks resulting from the pulse experiments, expressed as fractions of each pulse magnitude.(a) Fossil fuel pulses, which were set equal to gross fire emissions.(b) Fire pulses.The fractions were sometimes greater than 1.0 for the atmosphere and land, because pulses were defined based on direct combustion only.(c) Results for atmospheric carbon only (i.e., airborne fraction); for fossil fuel, only FF100P-G is illustrated as the results were almost equal for the FF20P-G and FF200P-G cases (see panel a).

Table 2 .
Comparison of the IRF for fossil fuel and fire pulses of 20, 100, and 200 Pg C. For fire, the pulses correspond to the emissions from direct combustion only.The first two years were discarded from the IRF estimation, because the atmospheric CO 2 anomaly sometimes reached its maximum value in the third year for fire.The sum of the four a i was constrained to 1.0 for fossil fuel, but not for fire.See Eq. (1) for the seven parameters (a i are unitless, τ i are in years).R 2 : coefficient of determination; MBE: mean bias error; RMSE: root mean square error.× 10 −6 −5.2 × 10 −6 9.8 × 10 −6 −5.2 × 10 −6 −2.9 × 10 −6 −5.0 × 10 −6

Table 3 .
Burned area and emissions * for the three stable fire regimes.