Abstract
The Paris Agreement advances forest management as one of the pathways to halt climate warming through carbon dioxide (CO2) emission reduction1. The climate benefits from carbon sequestration from forest management may, however, be reinforced, counteracted, or even offset by concurrent management-induced changes in surface albedo, surface roughness, biogenic volatile organic compound emissions, transpiration, and sensible heat flux2–4. Forest management could, thus, offset CO2 emissions without halting global temperature rise. It remains, therefore, to be confirmed that sustainable forest management portfolios for the end of the 21st-century for Europe would comply with the Paris Agreement, i.e., reduce the growth rate of atmospheric CO2, reduce the radiative imbalance at the top of the atmosphere, and neither increase the near-surface air temperature nor decrease precipitation. Here we show that a spatially-optimized portfolio that maximises the carbon sink through carbon sequestration, wood use and product and energy substitution, reduces the growth rate of atmospheric CO2 but does not meet any of the other criteria. The portfolios that maximise the carbon sink or forest albedo pass only one, albeit different, criterion. Managing the European forests with the objective to reduce near-surface air temperature, on the other hand, will also reduce the atmospheric CO2 growth rate, thus meeting two out of four criteria. Our results demonstrate that if present-day forest cover is sustained, the additional climate benefits through forest management would be modest and local rather than global. Based on these findings we argue that if adaptation would require large-scale changes in species composition and silvicultural systems over Europe5,6, these changes could be implemented with little unintended climate effects.
Following the Paris Agreement, the European Union and its 28 Member States committed to a 40% domestic reduction in greenhouse gas emissions by 2030 compared to 1990. About seventy five percent of this reduction is expected to come from emission reductions, and the remaining 25% from land use, land use change and forestry7. The commitment to reduce the domestic greenhouse gas emissions through forestry is in turn reflected in national strategies for energy, climate change, and forestry8–10 of several European countries. These strategies typically focus on enhancing forestry-based sinks and reservoirs and developing neutral or negative emissions approaches based on woody biomass. Furthermore, European forest owners who reported to have experienced climate change, indicated that this experience influenced their management decisions11. Hence, climate change and the Paris Agreement are already shaping forest management decisions. Despite the fact that it is explicitly mentioned in both the Kyoto Protocol12 and the Paris Agreement1 little is known about the climate effects of forest management including the effects of human-induced tree species changes and silvicultural systems3,13,14.
This study searches for spatially-explicit forest management portfolios for Europe that comply with the Paris Agreement up to the turn of the 21st-century. Compliance requires that forest management jointly reduces the growth rate of atmospheric CO2 (Art. 4 and 5) and the radiative imbalance at the top of the atmosphere (Art. 2). Furthermore, forest management compliant with the Paris Agreement should neither increase the near-surface air temperature (hereafter referred to as air temperature) nor decrease precipitation since changing the climate of the terrestrial biosphere would make adaptation to climate change (Art. 7) even more difficult (see Methods “Operationalizing the Paris Agreement”).
Simulation experiments which combine vegetation modelling, climate modelling, vegetation-climate feedbacks, and life cycle analysis were used to quantify the CO2 emissions, radiative imbalance at the top of the atmosphere, near-surface air temperature, and precipitation of three spatially-explicit forest management portfolios in Europe. Each portfolio came with its own objective: maximise the forest carbon sink, maximise forest albedo, or reduce near-surface air temperature.
All portfolios started from the same 2010 species and age-class distribution. Once an individual forest reached maturity, six scenarios were explored: (i) refrain from harvesting; (ii) harvest, replant the same species and apply the same silvicultural strategy as before; (iii) harvest, replant the same species, and thin prior to the final felling; (iv) harvest, change to the most common deciduous species in that region and thin prior to the final felling; (v) harvest, change to the most common deciduous species in that region and manage it as a coppice; and (vi) harvest, change to the most common conifer species in that region and thin prior to the final felling. Subsequently, portfolios were constructed by selecting the best-performing management scenario –out of six– for each of the three objectives and for each grid cell in the European domain.
Contrary to previous land-use simulation experiments, our portfolios simulate a realistic rate of change for tree species distribution and silvicultural systems because changes were only implemented following a harvest or stand-replacing mortality. Management changes were, thus, dictated by forest growth and human choices within natural constraints, rather than through externally prescribed harvest volumes or through strictly natural succession.
A management portfolio that maximises the carbon sink15,16 reflects the widely-held view that the net climate effect of forest management is dominated by decreasing the growth rate of atmospheric CO2 through forest-based carbon sequestration, carbon storage in wood products, and material and energy substitution. Implementing the sink-maximising portfolio would –compared to business-as-usual– require converting 475,000 km2 of deciduous forest in central and southern Europe into coniferous forest whereas 266,000 km2 of previously coniferous forests in northern and central Europe would have to be converted to deciduous forests (Fig. 1; Extended Data Table 1; see “Drivers of changes in forest management”).
Figure 1.
Surface areas (x 10,000 km2) under forest management by the year 2100 for portfolios that target maximising the carbon sink, continue present-day management, and reduce the near-surface air temperature. Forest management distinguishes between tree species composition and silvicultural systems. The inset presents the mean values for all of Europe. Regional difference are shown for three geographical regions, each shown in a different shade of grey.
A sink-maximising portfolio would come with a 12 % lower wood harvest but could offset an additional 8.1 Pg C of fossil fuel emissions (Table 1) between 2010 and 2100 compared with a business-as-usual management portfolio that continues the present-day forest management portfolio into the future. This increase in the projected carbon savings is similar to estimates by the forestry sector16, and could be achieved by optimising the balance between forest-based sequestration (8.2 Pg C) on the one hand and product-based sinks and substitution (-0.3 Pg C), energy-based substitution (0.2 Pg C), and savings in the exploitation and production emissions (0.05 Pg C) on the other. Accounting for ocean uptake of atmospheric CO2 (see Methods "Life cycle analysis") results in a cumulated net reduction of the atmospheric CO2 concentration of 4.3 Pg C in 2100, which translates into a 2 ppm decrease in atmospheric CO2 compared with business-as-usual (Table 1). Owing to the changes in tree species and silvicultural systems required to realize this 2 ppm draw-down, the ~0.002 W m-2 decrease in the radiative imbalance at the top of the atmosphere from the stronger carbon sink17 is neutralized by unintended but unavoidable changes in surface albedo (-0.001) and cloud cover (-0.1%). The carbon-based portfolio has a small negative effect on precipitation (-2 mm) and no effect on air temperature (Table 1).
Table 1.
Biogeochemical and biophysical effects in 2100 for four different forest management portfolios over Europe. The business as usual simulation which served as a control, was repeated three times with slightly different initial atmospheric conditions (see Methods “Equilibrium climate for the management portfolios”). The variability between these three repetitions was considered the minimal model noise of the climate model. The reported noise was taken to be the definition of one standard deviation. TOA denotes the radiative imbalance at the top of the atmosphere.
| Variable name (units) | Business as usual (BAU) | Maximise carbon sink | Maximise albedo | Reduce near-surface temperature |
|---|---|---|---|---|
| Global average TOA (W m-2) | 4.31 ± 0.01 | 4.31 | 4.33 | 4.32 |
| Δ2100-2010 CO2 sink & avoided emissions (Pg C) | 4.7 | 12.8 | 5.0 | 8.1 |
| Δ2100-2010 net cumulated atmospheric CO2 (Pg C) | 2.7 | 7.0 | 2.8 | 4.5 |
| Atmospheric CO2 (ppm) | 934.6 | 932.6 | 934.6 | 933.8 |
| Near surface temperature (K) | 283.84 ± <0.001 | 283.84 | 283.83 | 283.81 |
| Annual precipitation (mm) | 734.7 ± 0.1 | 732.6 | 730.0 | 730.9 |
| Summer precipitation (mm) | 166.1 ± 0.1 | 165.2 | 163.7 | 165.0 |
| Wood harvest (Tg C y-1) | 203.2 | 179.5 | 144.5 | 151.6 |
| Surface albedo (-) | 0.113 ± <0.0001 | 0.113 | 0.128 | 0.126 |
| Evapotranspiration (mm) | 555.5 ± 0.1 | 552.8 | 546.4 | 549.2 |
| Latent heat (W m-2) | 44.35 ± <0.01 | 44.13 | 43.60 | 43.82 |
| Sensible heat (W m-2) | 26.67 ± <0.01 | 26.82 | 27.28 | 27.00 |
| Total cloud cover (%) | 46.8 ± <0. 1 | 46.7 | 46.7 | 46.6 |
A temperature-based portfolio reflects the idea that management-induced changes in surface properties may redistribute the heat away from the surface resulting in a local cooling of the land surface18 that can be beneficial for organisms living there. Implementing such a portfolio requires converting 493,000 km2 of coniferous forests to deciduous forests (of which 65% would be in Scandinavia) and coppicing an additional 600,000 km2 of deciduous forests (Fig. 1; Extended Data Table 1; “Description of the changes in forest management”). Such changes in forest management would, however, reduce the wood harvest by 25 % compared to business as usual (Table 1). By 2100 these changes would result in a cumulative net reduction of the atmospheric CO2 concentration of 1.8 Pg, which is equivalent to a 0.9 ppm reduction of atmospheric CO2 compared with business as usual (Table 1).
The combined biogeochemical and biophysical effects of this portfolio come without a significant effect on the radiative imbalance at the top of the atmosphere but could contribute to a 0.3 K cooling over Scandinavia, while having much less effect on temperature over the rest of Europe (Fig. 2A). Following a large-scale transition to deciduous species, cooling of the air temperature was projected to occur in winter and spring only (Extended Data Fig. 1). In spring, air temperature cooling from an increase in surface albedo due to decreased snow masking by deciduous canopies would be partly compensated by warming from a decrease in turbulent fluxes due to the absence of leaves until bud break later in spring (Fig. 2B). The simulation experiment thus confirms the role of transpiration in determining air temperature, even at high latitudes19.
Figure 2.
Changes in, and main drivers of, near-surface air temperature (ΔTa; K) in February and March by the turn of the 21st-century for a forest management portfolio that reduces the near-surface air temperature. (a) Spatially explicit changes in near-surface air temperature (K) in February and March. (b) Drivers of the changes in springtime near-surface air temperature for 0.5 degree latitudinal bands. In subplot (a) temperature changes less than 1.96 times the standard deviations are shown in white. Where, the standard deviation represents the minimal noise of LMDzORCAN (see Methods “Equilibrium climate for the management portfolios”). The change in near-surface air temperature (Ta) due to changes in atmospheric emissivity (ε) is written as ΔTa|ε. By analogy ΔTa|G is the change in air temperature due to change in the ground heat flux, ΔTa|LE+H due to changes in turbulent fluxes, ΔTa|Rsi due to changes in shortwave incoming radiation which in this simulation experiment is a proxy for cloud cover, ΔTa|α due to changes in surface albedo, and ΔTa|circ due to changes in atmospheric circulation.
A portfolio that maximises the albedo20 reflects the view that managing the forest albedo would reduce the radiative imbalance at the top of the atmosphere while maintaining the forest carbon sink. Our simulations confirm that an albedo-maximising portfolio would decrease wood harvest by 30 % and realize cumulated net emission savings of up to 2.8 Pg C which is comparable to the savings expected from the business-as-usual portfolio. However, the increase in surface albedo that can be realized through the albedo-based portfolio (+0.015) would be compensated by decreases in cloud cover (-0.1%) and, therefore, come without a significant effect on the radiative imbalance at the top of the atmosphere and a small negative effect on air temperature (-0.01 K; Table 1).
Furthermore, all portfolios reduced the mean annual precipitation by 2.1 to 4.7 mm compared to the business as usual portfolio. Reductions were evenly spread across the seasons and consistent with the decrease in cloud cover and evapotranspiration (Table 1). Hence, none of the tested forest management portfolios meet all four criteria set for compliance with the Paris Agreement. Maximising the carbon sink, and maximising the forest albedo both meet one out of four criteria. Managing the European forests with the objective to reduce air temperature results in reducing air temperature and the CO2 growth rate, thus meeting two of the four criteria.
To our knowledge, this study is the first to quantify the capacity of forest management to comply with the Paris Agreement while addressing both biogeochemical and biophysical effects; hence, its results could not be compared to previous reports. The small temperature effects, compared to those found in global afforestation and deforestation studies21–24, are thought to be the consequence of a realistic 90-year long period over which management changes were implemented, and the limited global land area for which portfolios were tested, i.e., ~7% of the global total of managed forest14. Although a global implementation of carbon-based forest management is likely to enhance the carbon sink of the forest sector globally15, the combined biogeochemical and biophysical effects cannot be extrapolated from Europe to the global scale, due to biome-specific land-atmosphere interactions25,26. A global implementation of locally optimised forest management portfolios would lead to larger areas with near-surface cooling. Given that air temperature cooling was found to quickly saturate with the fractional change in species composition (Extended Data Fig. 2), the magnitude of the cooling is not expected to change substantially following a large-scale implementation, unless ocean feedbacks19,22, cloud feedbacks through species-specific biogenic volatile organic compound emissions27, and changes in the North Atlantic Oscillation28, which were not fully accounted for in this study, are among the key drivers.
Our results demonstrate –based on a single model– that in the absence of carbon capture and storage the additional climate benefits through sustainable forest management will be modest and local rather than global. Hence, we suggest that the primary role of forest management in Europe in the coming decades is not in protecting the climate but in adapting the forest cover to future climate5 in order to sustain the provision of wood, as well as ecological, social, and cultural services29 while avoiding positive climate feedbacks from fire, wind, pests and drought disturbances30. Even if adaptation would require large-scale changes in species composition and silvicultural system over Europe5,6, our results imply that these changes themselves are likely to have little impact on the climate.
Extended Data
Extended Data Figure 1.
Drivers of the changes in mean bimonthly near-surface air temperature (ΔTa; K) for 0.5 degree latitudinal bands. The change in near-surface air temperature (Ta) due to changes in atmospheric emissivity ε is written as ΔTa|ε. By analogy ΔTa|G is the change in air temperature due to change in the ground heat flux, ΔTa|LE+H due to changes in turbulent fluxes, ΔTa|Rsi due to changes in shortwave incoming radiation (which in this simulation experiment is a proxy for cloud cover), ΔTa|α due to changes in surface albedo, and ΔTa|circ due to changes in atmospheric circulation. Although all the components contribute to the near-surface air temperature, changes in emissivity always result in a cooling and changes in shortwave incoming radiation always result in warming. Consequently, emissivity and incoming shortwave radiation cannot explain the seasonal variation in the changes in near-surface air temperature. The other components are in some months positively correlated with near-surface air temperature whereas they are negatively correlated for other months, excluding them from being the main driver of changes in near-surface air temperature. Suggesting the net effect is the outcome of the interplay between the different components.
Extended Data Figure 2.
Relationship between changes in springtime near-surface air temperature (K) and changes in fractional cover of deciduous forest (km2) for 0.5 degree latitudinal bands over Europe. Locations where the tree species were maintained between 2010 and 2100 (i.e. Δ deciduous area on the X-axis equals zero) could experience similar air temperature changes as neighbouring locations where one species was converted into another, especially in Scandinavia, suggesting advection of heat and moisture. Nevertheless, at lower latitudes the spatial scale of this advection was limited to a few pixels (e.g., Fig. 2A) corresponding to a range of 50 to 200 km. Furthermore, the temperature effect quickly saturated with the fractional cover change and showed a strong dependence on geographical location91. Whether the apparent geographical dependency was the outcome of climatic differences and/or differences between northern and southern European deciduous species could not be established by the experimental setup used in this study.
Extended Data Figure 3.
Setup of the simulation experiments following the description in “Simulation experiment”. Simulations with ORCHIDEE-CAN are shown in black, and simulations with LMDzORCAN are shown in red. Blue boxes show intermediate calculations making use of simulation results (see Methods “Spatially optimised management portfolios” and “Equilibrium climate for the management portfolios”). The labels of the different simulations shown in this figure are the same labels as those used to run and archive the simulations. Note that in the main text the results of BBESTT2M were presented as “reduced near-surface air temperature”, BESTALBEDO as “maximise surface albedo”, BWORSTALBEDO as “minimise surface albedo, BBESTLCA as “maximise C-sink”, BWORSTLCA as “minimise C-sink”, and BWAC as “business as usual”. BWAC, BWAC-P1 and BWAC-P2 were used to calculate the minimal model noise as explained in Methods “Simulation experiment”.
Extended Data Figure 4.
Optimized forest management portfolio’s for Europe to maximize the surface albedo under (a) RCP8.5 and (b) RCP4.5. The numbers in the legend relate to (i) refrain from harvesting; (ii) harvest, replant the same species and apply the same silvicultural strategy as before; (iii) harvest, replant the same species, and thin prior to the final felling; (iv) harvest, change to the most common deciduous species in that region and thin prior to the final felling; (v) harvest, change to the most common deciduous species in that region and manage it as a coppice; and (vi) harvest, change to the most common conifer species in that region and thin prior to the final felling.
Extended Data Table 1.
Changes in surface area (km2) by 2100 for six different forest management portfolios over Europe. Note that the total surface area of forests was held constant at 2,000,000 km2 between 2010 and 2100, for reasons described in “Simulation experiment”.
| Change in surface area (km2) | Business as usual (BAU) | Maximise carbon sink | Maximise albedo | Minimise carbon sink | Minimise albedo | Reduce near-surface temperature |
|---|---|---|---|---|---|---|
| Deciduous to conifers | 0 | 475,000 | 30,000 | 6,000 | 516,000 | 41,000 |
| Conifers to deciduous | 0 | 266,000 | 590,000 | 236,000 | 26,000 | 534,000 |
| Net increase conifers | 0 | 209,000 | -560,000 | -230,000 | 490,000 | -493,000 |
| Net increase thin and fell | 0 | -280,000 | -330,000 | -390,000 | -230,000 | -680,000 |
| Net increase coppice | 0 | -20,000 | 130,000 | -130,000 | -210,000 | 600,000 |
| Net increase unmanaged | 0 | 300,000 | 200,000 | 520,000 | 440,000 | 80,000 |
Extended Data Table 2.
Biogeochemical and biophysical effects in 2100 for two additional –compared to Table 1– forest management portfolios over Europe.
| Variable name (units) | Minimise carbon sink | Minimise albedo |
|---|---|---|
| TOA (W m-2) | 4.32 | 4.32 |
| Δ2100-2010 CO2 sink & avoided emissions (Pg C) | 0.7 | 10.5 |
| Δ2100-2010 net cumulated atmospheric CO2 (Pg C) | 0.5 | 5.7 |
| Atmospheric CO2 (ppm) | 935.7 | 933.2 |
| Near surface temperature (K) | 283.85 | 283.86 |
| Annual precipitation (mm) | 733.1 | 734.2 |
| Summer precipitation (mm) | 164.0 | 165.4 |
| Wood harvest (Tg C y-1) | 122.9 | 176.2 |
| Surface albedo (-) | 0.119 | 0.107 |
| Evapotranspiration (mm) | 550.0 | 553.9 |
| Latent heat (W m-2) | 43.90 | 44.23 |
| Sensible heat (W m-2) | 27.12 | 26.81 |
| Total cloud cover (%) | 46.8 | 46.8 |
Extended Data Table 3.
Key characteristics of the individual model runs in the simulation experiment. The model runs are presented in the same order as Extended Data Fig. 3. For each model run the time period is described by its start year, end year and the length of the simulation in years (Years) together with the simulation used to initialize key characteristics of the biosphere (Initial state). The atmospheric CO2, CH4, N2O, CFC11, and CFC12 concentrations at the end of the simulation were reported and their values were extracted from Refs. 86 and 95. For the portfolio model runs atmospheric CO2 concentrations were adjusted for the simulated carbon sink after discounting for ocean uptake as outlined in “Atmospheric composition” and “Life cycle analysis”. Climate and other forcing agents including sea surface temperature, sea ice extent, and atmospheric aerosol concentrations were retrieved from the RCP8.5 simulation with the IPSL-CM5 model59 as part of the AR5 model inter-comparison. In this study, forest management consisted of two activities: species changes (Species) and a silvicultural treatment (Silviculture). For historical model runs a forest management reconstruction was used47, and a single year indicates the reconstruction for that specific year was used. For future simulations, species distribution and/or silvicultural management was either fixed to the 2010 distribution or was changed towards deciduous or conifers for the species and/or conservation, high-stand, or coppice for the silvicultural system (see Methods “Forest cover and forest silvicultural reconstruction”, “Future species”, and “Future silviculture”). The labels of the different simulations shown in this table are the same labels as those used to run and archive the simulations. Note that in the main text the results of BBESTT2M were presented as “reduced near-surface air temperature”, BESTALBEDO as “maximise surface albedo”, BWORSTALBEDO as “minimise surface albedo, BBESTLCA as “maximise C-sink”, BWORSTLCA as “minimise C-sink”, and BWAC as “business as usual”. BWAC, BWAC-P1 and BWAC-P2 were used to calculate the minimal model noise as explained in “Simulation experiment”.
| Simulation label | Period | Years | Initial state | Climate | CO2 (ppm) | CH4 (ppb) | N2O (ppb) | CFC11 (ppt) | CFC12 (ppt) | Other | Species | Silviculture |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SPIN-UP | ||||||||||||
| SPIN1 | 1600/1600 | 260 | n.a. | 1901/1920 | 277.9 | n.a. | n.a. | n.a. | n.a. | n.a. | 1600 | 1600 |
| SPIN2 | 1600/1600 | 40 | SPIN1 | 1901/1920 | 277.9 | n.a. | n.a. | n.a. | n.a. | n.a. | 1600 | 1600 |
| TRANSIENT SIMULATION | ||||||||||||
| TRANS1 | 1601/1900 | 300 | SPIN2 | 1901/1930 | 295.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Recon. | Recon. |
| TRANS2 | 1901/2010 | 110 | TRANS1 | 1901/2010 | 395.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Recon. | Recon. |
| FOREST MANAGEMENT SCENARIOS | ||||||||||||
| CWAC | 2011/2100 | 90 | TRANS2 | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | 2010 | 2010 |
| CWA1 | 2011/2100 | 90 | TRANS2 | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | 2010 | Conser. |
| CWA2 | 2011/2100 | 90 | TRANS2 | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | 2010 | Thin&F. |
| CWC2 | 2011/2100 | 90 | TRANS2 | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Coni. | Thin&F. |
| CWD2 | 2011/2100 | 90 | TRANS2 | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Deci. | Thin&F. |
| CWD3 | 2011/2100 | 90 | TRANS2 | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Deci. | Coppice |
| BWAC | 2101/2101 | 10 | CWAC | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BWA1 | 2101/2101 | 10 | CWA1 | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | Conser. |
| BWA2 | 2101/2101 | 10 | CWA2 | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | Thin&F. |
| BWC2 | 2101/2101 | 10 | CWC2 | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | Coni. | Thin&F. |
| BWD2 | 2101/2101 | 10 | CWD2 | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | Deci. | Thin&F. |
| BWD3 | 2101/2101 | 10 | CWD3 | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | Deci. | Coppice |
| EQUILIBRIUM CLIMATE FOR THE MANAGEMENT PORTFOLIOS | ||||||||||||
| CBESTT2M | 2011/2100 | 90 | Optimised | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Optim. | Optim. |
| CBESTLCA | 2011/2100 | 90 | Optimised | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Optim. | Optim. |
| CWORSTLCA | 2011/2100 | 90 | Optimised | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Optim. | Optim. |
| CBESTALBEDO | 2011/2100 | 90 | Optimised | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Optim. | Optim. |
| CWORSTALBEDO | 2011/2100 | 90 | Optimised | RCP8.5 | 935.8 | n.a. | n.a. | n.a. | n.a. | n.a. | Optim. | Optim. |
| BWAC | 2101/2101 | 20 | CWAC | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BWAC-P1 | 2101/2101 | 20 | CWAC | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BWAC-P2 | 2101/2101 | 20 | CWAC | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BBESTT2M | 2101/2101 | 20 | CBESTT2M | n.a. | 933.8 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BBESTLCA | 2101/2101 | 20 | CBESTLCA | n.a. | 932.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BWORSTLCA | 2101/2101 | 20 | CWORSTLCA | n.a. | 935.7 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BBESTALBEDO | 2101/2101 | 20 | CBESTALBEDO | n.a. | 934.6 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
| BWORSTALBEDO | 2101/2101 | 20 | CWORSTALBEDO | n.a. | 933.2 | 3751 | 435 | 26 | 167 | RCP8.5 | 2010 | 2010 |
Extended Data Table 4.
Emission coefficients, conversion factors, and assumptions used in the European wide life cycle analysis.
| Component | Unit | Value | Source |
|---|---|---|---|
| Carbon in biomass | g g-1 | 0.5 | Assumed |
| Transport distance roundwood | Km | 100 | Assumed |
| Transport distance fuelwood | Km | 50 | Assumed |
| Transport by truck | kg CO2 tkm-1 | 1.12 | Ref. 100 |
| Mechanized harvest | kg CO2 ha-1 | 233 | Ref. 101 |
| Mechanized planting | kg CO2 ha-1 | 93 | Ref. 101 |
| Mechanized thinning | kg CO2 ha-1 | 69 | Ref. 101 |
| Product substitution | kg CO2 kg-1 CO2 | 1.1 | Ref. 101 |
| Energy density of biomass | GJ t-1 | 19.3 | Ref. 102 |
| Conversion efficiency | % | 34 | Ref. 103 |
| Energy from biomass-based electricity | GJ, t-1 | 6.6 | Energy density of biomass multiplied with the conversion efficiency |
| Emissions from biomass-based electricity | kg CO2 kg-1 CO2 | 1.05 | Assuming that drying consumes 0.05 kg CO2 kg-1 CO2 and burning or gasifying woody biomass produces 1 kg CO2 kg-1 CO2 |
Extended Data Table 5.
Country based CO2 emission factors (g CO2 eq kWh-1) for the current non-renewable electricity mix of energy production based on ref. 100.
| Country | Emission factor |
|---|---|
| Albania, Belarus, Kosovo, Macedonia, Moldova, | 1020 |
| Montenegro & Ukraine | |
| Andorra, France & Monaco | 810 |
| Austria & Liechtenstein | 777 |
| Belgium | 687 |
| Bosnia & Herzegovina | 1017 |
| Bulgaria | 981 |
| Croatia | 812 |
| Cyprus, Iceland & Malta | 868 |
| Czech Republic | 1010 |
| Denmark | 904 |
| Estonia | 1014 |
| Finland | 853 |
| Germany | 927 |
| Greece | 894 |
| Hungary | 780 |
| Ireland | 766 |
| Italy | 744 |
| Latvia | 615 |
| Lithuania | 591 |
| Luxembourg | 614 |
| Netherlands | 748 |
| Norway | 641 |
| Poland | 1000 |
| Portugal | 840 |
| Romania | 907 |
| Serbia | 1012 |
| Slovakia | 842 |
| Slovenia | 982 |
| Spain | 797 |
| Sweden | 857 |
| Switzerland | 628 |
| United Kingdom | 854 |
Supplementary Material
Extended Data Information is linked to the online version of the paper at www.nature.com/nature.
Acknowledgments
M.J.M., K.N., J.R., Y.C., J.O. and S.L. were funded through ERC starting grant 242564 (DOFOCO) and A.V. was funded through ADEME (BiCaFF). S.L. and G.M. were in part funded through an Amsterdam Academic Alliance (AAA) fellowship. S.L. dedicates this study to the mentorship of E.-D. Schulze, I.A. Janssens, and P. Ciais. The ORCHIDEE and LMDZ project teams as well as the Centre de Calcul Recherche et Technologie (CCRT) provided the run environment without which the type of coupled land-atmosphere simulations conducted in this study would not be possible.
Footnotes
Author contributions
S.L and M.J.M. designed the study. M.J.M., J.O., J.R., Y.C., K.N., A.V. and S.L. developed, parametrized and validated ORCHIDEE-CAN. G.M., M.J.M., J.G. and S.L. conducted the simulation experiment. S.N.D. developed the life cycle analysis. G.M., Y.C. and S.L. analysed the data. G.M., M.J.M., J.O., J.R., Y.C., K.N., A.V., A.S.L. and S.L. contributed to the interpretation of the results.
Author information
Reprints and permissions information is available at www.nature.com/reprints. Readers are welcome to comment on the online version of the paper. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The authors declare no competing financial interests.
Code availability
The code and the run environment are open source and distributed under the CeCILL (CEA CNRS INRIA Logiciel Libre) license. The code of ORCHIDEE-CAN r2290 and r3069 can be accessed from http://dx.doi.org/10.14768/06337394-73A9-407C-9997-0E380DAC5595 and http://dx.doi.org/10.14768/06337394-73A9-407C-9997-0E380DAC5596 respectively. Access to the run environment and LMDzORCAN are restricted to registered users. Requests can be sent to the corresponding author. The post-processing code used to estimate the life cycle sinks and emissions of the forestry sector (see “Life cycle analysis”), search for the optimised management portfolios (see “Management optimisation”), and decompose the near-surface air temperature into its main drivers (see “Decomposing near-surface air temperature”) can be accessed from http://dx.doi.org/10.5281/zenodo.1284533.
Data availability
Figure 1, Figure 2, Table 1, Extended Data Figure 1, Extended Data Figure 2, Extended Data Figure 4, Extended Data Table 1, and Extended Data Table 2 are based on a simulation experiment for which the output files (~7.4 Tb) will be provided upon reasonable request. The data files that were used to set the boundary conditions of ORCHIDEE-CAN and LMDzORCAN (~70 Gb) will be provided upon reasonable request.
References
- 1.UNFCCC. Adoption of the Paris Agreement; FCCC/CP/20.2015. [Google Scholar]
- 2.Pielke RA, et al. Land use/land cover changes and climate: Modeling analysis and observational evidence. Wiley Interdiscip Rev Clim Chang. 2011;2:828–850. [Google Scholar]
- 3.Naudts K, et al. Europe’s forest management did not mitigate climate warming. Science. 2016;351:597–600. doi: 10.1126/science.aad7270. [DOI] [PubMed] [Google Scholar]
- 4.Jackson RB. Trading water for carbon with biological carbon sequestration. Science. 2005;310:1944–1947. doi: 10.1126/science.1119282. [DOI] [PubMed] [Google Scholar]
- 5.Lindner M, et al. Climate change impacts, adaptive capacity, and vulnerability of European forest ecosystems. For Ecol Manage. 2010;259:698–709. [Google Scholar]
- 6.Hanewinkel M, Cullmann DA, Schelhaas M-J, Nabuurs G-J, Zimmermann NE. Climate change may cause severe loss in the economic value of European forest land. Nat Clim Change. 2012;3:203–207. [Google Scholar]
- 7.Grassi G, et al. The key role of forests in meeting climate targets requires science for credible mitigation. Nat Clim Change. 2017;7:220–226. [Google Scholar]
- 8.EC. National Action Plans. 2017 Available at: https://ec.europa.eu/energy/en/topics/renewable-energy/national-action-plans.
- 9.Ministry of Economic Affairs and Employment. Government report on the National Energy and Climate Strategy for 2030. 2017. [Google Scholar]
- 10.Legifrance. Décret n° 2017-155 du 8 février 2017 portant approbation du programme national de la forêt et du bois. 2016 2015 Available at: https://www.legifrance.gouv.fr/eli/decret/2017/2/8/AGRT1632600D/jo/texte.
- 11.Blennow K, Persson J, Tomé M, Hanewinkel M. Climate Change: Believing and Seeing Implies Adapting. PLoS One. 2011;7:e50182. doi: 10.1371/journal.pone.0050182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.UNFCCC. Kyoto protocol to the United Nations framework convention on climate change. 1998 [Google Scholar]
- 13.Erb K-H, et al. Land management: data availability and process understanding for global change studies. Global Change Biol. 2017;23:512–533. doi: 10.1111/gcb.13443. [DOI] [PubMed] [Google Scholar]
- 14.Luyssaert S, et al. Land management and land-cover change have impacts of similar magnitude on surface temperature. Nat Clim Change. 2014;4:389–393. [Google Scholar]
- 15.Canadell JG, Raupach MR. Managing Forests for Climate Change Mitigation. Science. 2008;320:1456–1457. doi: 10.1126/science.1155458. [DOI] [PubMed] [Google Scholar]
- 16.Nabuurs G, et al. A new role for forests and the forest sector in the EU post-2020 climate targets. From Science to Policy 2. European Forest Institute; 2015. [Google Scholar]
- 17.Myhre G, Highwood EJ, Shine KP, Stordal F. New estimates of radiative forcing due to well mixed greenhouse gases. Geophys Res Lett. 1998;25:2715–2718. [Google Scholar]
- 18.Davin EL, Seneviratne SI, Ciais P, Olioso A, Wang T. Preferential cooling of hot extremes from cropland albedo management. Proc Natl Acad Sci USA. 2014;111:9757–9761. doi: 10.1073/pnas.1317323111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Swann AL, Fung IY, Levis S, Bonan GB, Doney SC. Changes in Arctic vegetation amplify high-latitude warming through the greenhouse effect. Proc Natl Acad Sci USA. 2010;107:1295–1300. doi: 10.1073/pnas.0913846107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lutz DA, Howarth RB. Valuing albedo as an ecosystem service: Implications for forest management. Clim Change. 2014;124:53–63. [Google Scholar]
- 21.Arora VK, Montenegro A. Small temperature benefits provided by realistic afforestation efforts. Nat Geosci. 2011;4:514–518. [Google Scholar]
- 22.Davin EL, de Noblet-Ducoudré N. Climatic impact of global-scale deforestation: Radiative versus nonradiative processes. J Clim. 2010;23:97–112. [Google Scholar]
- 23.Bathiany S, Claussen M, Brovkin V, Raddatz T, Gayler V. Combined biogeophysical and biogeochemical effects of large-scale forest cover changes in the MPI earth system model. Biogeosciences. 2010;7:1383–1399. [Google Scholar]
- 24.Bala G, et al. Combined climate and carbon-cycle effects of large-scale deforestation. Proc Natl Acad Sci USA. 2007;104:6550–6555. doi: 10.1073/pnas.0608998104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Beringer J, Chapin FS, Thompson CC, McGuire AD. Surface energy exchanges along a tundra-forest transition and feedbacks to climate. Agric For Meteorol. 2005;131:143–161. [Google Scholar]
- 26.Alkama R, Cescatti A. Biophysical climate impacts of recent changes in global forest cover. Science. 2016;351:600–604. doi: 10.1126/science.aac8083. [DOI] [PubMed] [Google Scholar]
- 27.Unger N. Human land-use-driven reduction of forest volatiles cools global climate. Nat Clim Change. 2014;4:907–910. [Google Scholar]
- 28.Hurrell JW. Decadal trends in the North Atlantic oscillation: regional temperatures and precipitation. Science. 1995;269:676–679. doi: 10.1126/science.269.5224.676. [DOI] [PubMed] [Google Scholar]
- 29.Costanza R, et al. The value of the world ’ s ecosystem services and natural capital. Nature. 1997;387:253–260. [Google Scholar]
- 30.Seidl R, Schelhaas M-J, Rammer W, Verkerk PJ. Increasing forest disturbances in Europe and their impact on carbon storage. Nat Clim Change. 2014;4:806–810. doi: 10.1038/nclimate2318. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.