Europe’s Renewable Energy Directive Poised to Harm Global Forests
Searchinger et al.
The sources used to prepare table 1 are as follows. Total roundwood production of
the World from FAOSTAT: http://www.fao.org/faostat/en/#data/FO; total roundwood
production in Europe (EU-28) from Eurostat: http://ec.europa.eu/eurostat/statistics-
explained/index.php/Wood_products_-_production_and_trade. (Note, the European
roundwood production number from FAOSTAT is larger at 459 million cubic meters. We do
not know the explanation for the difference. That would increase total wood harvest to 4.1
EJ.); Conversion from cubic meters to dry matter assumes 1/3 coniferous and 2/3 non-
coniferous based on FAOSTAT, and .411 tonnes DM/cubic meter coniferous and .523 tonnes
DM/cubic meter non-coniferous wood based on Table 5 in Miles and Smith1; Calculation of
energy content of wood assuming an energy content of 20 GJ/t of dry matter; Global total
primary energy consumption is taken from International Energy Agency2. Total primary
energy consumption in Europe (EU-28) is taken from the European Commission3.
Supplementary Table 1: Calculation of the share of primary energy consumption that could
be produced from current total or industrial roundwood production.
Total wood in
% of primary
Supplementary Note 1 – Smokestack Emissions Wood v. Fossil Fuels
The additional carbon released burning wood per kWh compared to coal or natural gas is
one part of the explanation of why using wood will increase greenhouse gas emissions for
decades. The figure is a function of (a) the higher carbon content of wood than fossil fuels
per kWh of potential gross energy, and (b) the lower conversion efficiencies. The figure that
wood generates smokestack emissions 50% more than burning coal and more than three
times that of natural gas are based on the numbers used in Laganière et al.4 except we use
IPCC emission factors for the net calorific value of fuel, which are more favorable to wood
than those used in Laganière. Some papers claim higher conversion efficiencies for burning
wood, up to roughly 30%, but coal and natural gas conversion efficiencies are also often
Wood pellets can burn at higher temperature and produce electricity with greater
efficiency, but their production creates more greenhouse gas emissions and involves large
process losses along the way. For example, Röder et al.6 calculate that 38% of the wood is
lost at various stages of handling, processing or use, including 18% lost to supply heat for
pellet drying if fossil fuels are not used for that process. (If fossil fuels are used, they cause
their own emissions.) These losses in wood for fuel or other added emissions will roughly
cancel out the benefits of converting wood to wood pellets depending on the precise range
of emissions and ultimate burning efficiencies. For example, Sterman et al.7 , supplementary
information, p. 27, calculate that burning wood pellets for electricity generates 183% of the
emissions of burning coal when considering the whole supply chain, ignoring losses of
terrestrial carbon for this part of the calculation, although this paper uses a relatively low
conversion efficiency for electricity production.
Supplementary Table 2: Typical Smokestack Emissions for Electricity Production.
GJ net calorific
value of fuel*
value of fuel
* IPCC, 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Volume 2 Stationary
Installations, Table 2.2 (ref. 8)
** Laganière et al.4
Note that often wood is co-fired with coal. In that case, the total burning efficiency of the mix can be
closer to that of coal, e.g., 30-31%, but the net contribution to final kWh of wood is even lower
because it lowers the conversion efficiency of the coal.
Supplementary Note 2 – GHG Increases by 2050 from Wood Bioenergy
Many papers have calculated the greenhouse gas consequences of harvesting additional
trees for bioenergy as cited in the main text. The precise results vary greatly with the type of
forest harvested, the harvest regime, the fuel use (biomass conversion efficiencies come
closer to those of fossil fuels when used for heat than for electricity), and the type of fossil
fuel displaced. Most frequently, papers report payback or parity periods, which are the
periods trees must be allowed to regrow before the net effect on greenhouse gases
(incorporating both changes in terrestrial carbon and reductions in fossil carbon emissions)
are equivalent for use of biomass and the fossil fuel alternative. These payback periods are
sometimes reported assuming a single stand harvest in one year, which, for reasons
discussed in the main text, are shorter than for a sustained harvest. In general, if the
payback period is 60 years, then the emissions after 30 years would typically be at least and
probably more than double those of fossil fuels (because the payoff of debt should be faster
in the second thirty-year period than the first).
Using Laganière et al.4, the parity point for harvest and use of otherwise standing
Canadian trees (as opposed to salvaged trees or residues) and allowing the same forests to
regrow is always longer than 90 years for two classes of forests and has a shortest period of
roughly 70 years for the shortest rotation forests even when replacing coal and using wood
for heat. All harvests even for heat have payback periods of more than 100 years when
replacing natural gas. Sixty-year payback periods would imply at least a doubling of
emissions and 90 years would imply at least a tripling of emissions by 2050. Mitchell et al.9
also examine a wide variety of forest types and uses. Although a few possible scenarios have
payback periods as fast as 30 years (which means emissions would only equal fossil fuel use
after that period), the overwhelming majority are well more than 60 years, and most are
well more than 90 years. Holtsmark10 found slightly more than a doubling of atmospheric
carbon for harvesting from boreal forests in Europe to replace coal at around 40 years.
Based on these figures, a doubling or tripling of emissions is possible, or even more,
depending on the ultimate mix of forest sources and uses.
One possible reason for even higher concern is that plantation forests provide less than
half and probably only around one third of wood harvests today11. The fastest payback times
occur assuming that wood harvests come from fast-growing plantations that would
otherwise not be harvested. This assumption of non-harvest seems highly improbable for
these highly managed forests. Yet if this wood is diverted to bioenergy use instead of wood
products, replacements need to come from slower growing, natural forests with much
longer pay-back periods.
Final energy consumption in the EU-28 in 2015 was 1,084 MTOE or 45.4 EJ. With a share
of 16.7%, renewables contributed 7.6 EJ of final energy consumption12. The three parts of
the European Union (the Commission, Parliament and Council of States) have now agreed on
a renewable energy target of 32% for 2030. That requires that the share of renewables in
final energy would have to grow from 2015 to 2030 by 15.3% compared to 2015. In
Supplementary Note 3, we explain our reasoning that additional forest biomass could
plausibly supply 40% of this increase or an additional 6.1% of Europe’s final energy in 2030
or ~2.5 EJ. After deducting the portion of Europe’s existing final energy use that is
attributable to renewables, we obtain 37.8 EJ, so this additional final energy from forest
biomass would replace 6.6% of existing final energy from non-renewable sources.
Assuming broadly (a) that this 40% would otherwise be replaced by essentially carbon
neutral solar or wind, and (b) that the greenhouse gas emissions from this energy displaced
would otherwise have emissions at the European average energy emissions rate, the RED
would otherwise save ~6.6% of Europe’s existing energy emissions. Assuming instead
sources of biomass that double or triple emissions by 2050, the RED provisions result in a
~6.6 to ~13.2 percent increase. Because these numbers make many broad assumptions, we
present them as 5% and 10% changes to avoid conveying a false sense of precision, and to
also build in an additional level of conservatism. The actual effect will depend on many
factors including the actual amount of forest biomass used, what energy sources it replaces,
and both the precise direct and indirect sources of wood (including wood harvested to
replace wood diverted to bioenergy use that would otherwise supply wood products). We
believe this estimate provides one rough, plausible estimate of the emissions consequences
of the forest biomass provisions of the RED.
Supplementary Note 3 – Share of European Wood Harvest Needed
Using data from Tables 2.1, 2.2 and 2.3 from ref. 13, total growth of final renewable
energy in the EU28 from 2005 to 2015 was 78,248 ktoe, of which 46,430 ktoe was from
some source of biomass, or 59%. Only around 52% of this biomass was wood, so increased
use of wood provided around 30% of the growth in renewable energy in this period.
However, caps on crop-based liquid biofuels in the directive at existing levels or with small
gains are likely to reduce the contribution in coming years from those sources, which leaves
more future bioenergy gains to be achieved by wood. The RED also establishes a separate
goal of a 1% increase each year in renewable energy in heat, of which the vast majority
comes currently from biomass14. One analysis found that meeting this target would likely
require an additional 51 Mtoe, or 2.13 EJ (ref. 14, Table 4-1), which itself is roughly one third
of the increased renewable energy, and based on that study’s estimate of required 236
million cubic meters of roundwood equivalent, would itself require 55% of existing European
roundwood harvest. Putting these various trends and requirements together, we find it
plausible that forest biomass would supply 40% of Europe’s renewable energy. It could be
less but it could also be much more.
Supplementary note 2 explains how the RED requires an increase in final renewable
energy by 15.3% compared to 2015, but the amount of primary biomass energy that
requires depends in part on how the biomass is used. We here assume 35% of biomass is
used for electricity at 25% efficiency and 65% is used for heat at 85% efficiency, which
generates a total conversion efficiency from primary to final energy of 64% (efficiencies of
different bioenergy conversion technologies were taken from Edwards et al.15). The quantity
also depends on how much final energy Europe actually consumes in 2030. After debating
different targets, Europe has agreed on a target of 32.5% increase in economy-wide energy
relative to 2007 although the target is not binding16, which we calculate would lead to total
EU final energy consumption of 43.25 EJ based on extrapolations at this efficiency from EU
projections of economic growth in ref. 3. An increase of 15.3% in final renewable energy
would require an increase of 6.24 EJ of final renewable energy at the 32.5% agreed upon
efficiency target. If 40% is supplied by wood at 64% efficiency that would require 3.9 EJ of
wood, equal to 101% of Europe’s total existing wood harvest of 3.85 EJ of roundwood
Supplementary Note 4 – Harvest Increases Implied by European
The European Commission made its own estimates of future renewable biomass to
meet its originally proposed targets of the RED. One estimate using the Commission’s
PRIMEs modeling shows an increase of 43 Mtoe from 2013 to 2030, or 1.8 EJ, which would
be 46% of current roundwood harvest in Europe17,18. Green-X modeling projects an increase
in the share of energy from biomass in primary energy consumption from roughly 8% in
2013 to roughly 12% in 2030. If that increase comes from wood, it implies an increase of
2.2EJ or 56% of existing roundwood harvest assuming the European economy-wide 32.5%
energy efficiency increase in 2030 compared to 2007 as discussed above. However, these
model runs assumed a 27% renewable energy target in 2030 and different energy efficiency
increases for the whole EU economy. Scaling proportionately for a 32% renewable energy
target and a 32.5% “increase” in economy-wide energy efficiency relative to 2007 would
require 15.3 EJ more than the 2015 renewable energy use of 16.7%. That is a 48.5% larger
increase in renewable energy than implied by the increases in renewable energy assumed by
the EU modeling. Although actually running the models with the 32% renewable energy
target might generate somewhat different results, scaling up the increase in biomass
proportionately to the larger increase in renewables, the biomass increase rises to 2.2 EJ for
PRIMES modeling and 3.3 EJ for Green-X modeling, If these increases come from wood, the
additional wood would equal 56-85% of existing European wood harvest.
Supplementary Note 5 – RED Sustainability Criteria
The Renewable Energy Directive contains much language setting forth restrictions on
biomass that qualifies for the directive and that are referred to as sustainability standards,
and we discuss the most important ones in the text. None of these provisions would
significantly reduce impacts on forests overall, and we outline them here. There were small
differences in the language originally proposed by the European Commission and that
enacted by the European Parliament. In the cases of original disagreement, the language
quoted here cites language agreed upon by negotiators for the Commission, the Parliament
and the European Council, which was released on June 21, 201819.
Article 26, paragraph 7, provides requirements that the use of biomass reduce
greenhouse gas emissions by at least 50% for installations in operation before 2015, and
higher for installations coming on line later. However, the methodology for calculating
greenhouse gas emission are set forth in Appendix V for liquid biofuels and Appendix VII for
factories and power plants (among other stationary installations). They explicitly exclude the
carbon released by burning the biomass (Appendix V, Part C. par. 13; Appendix VI, Part C.,
par. 13): “Emissions of CO2 from fuel in use, eu, shall be taken to be zero for biomass fuels.”
As a result, the emissions that count are only the fossil emissions and trace gases generated
in harvesting and production process, not the carbon in the biomass itself, which is what it
means for a fuel to be carbon neutral. This can also be seen in tables providing default
values for the use of biomass fuels (Annex VI C).
The appendices (V & VI for the different types of bioenergy) also provide for counting
emissions from carbon stock changes resulting from land use change. These emissions are to
be counted if conversion of land to crop production, including bioenergy crops, involves
converting forests and other high carbon lands. Emissions from land use change must be
amortized over 20 years, meaning that each mega joule of energy in fuel is assigned
greenhouse gases equal to the emissions from land use change of a hectare divided by
twenty and divided by the mega joules that are produced from biomass grown on that
hectare each year. But land use changes only apply to the direct conversion from forest to
cropland or grassland, not merely from the harvest of wood on existing forest (Annex VIII
There are provisions restricting the use of biomass from some high carbon stocks, but
they apply only to conversions from other land uses for the production of “agricultural raw
material” (Article 26, par. 70 of the Proposed RED), and therefore do not apply to the
harvest of forests.
The directive includes numerous general references to the importance of sustainability.
For example, paragraph 76 of Article 26 provides that “woody raw material should come
only from forests that are harvested in accordance with the principles of sustainable forest
management developed under international forest processes such as Forest Europe and are
implemented through national laws or the best management practices at the forest sourcing
level.” The text also states: “Operators should take the appropriate steps in order to
minimize the risk of using unsustainable forest biomass for the production of bioenergy.”
These kinds of provisions do not restrict the quantity of forest harvest, nor do they even
appear to have much implication for major harvest methods – for example, clear-cutting will
typically comply with best management practices requirements.
There are also some more specific provisions whose principal significance is that forests
be allowed and able to regenerate (Article 26, paragraph 5). The two most specific of these
are (a)(i), which requires “forest regeneration of harvested”, and (a)(v), which requires that
“harvesting maintains or improved the long-term production capacity of the forest”. This
language is designed to assure renewability. The second type of provisions quoted is
ambiguous. It is not clear if it restricts the volume of harvest at all as a forest can be
harvested without eliminating its long-term production capacity. It is unlikely to be
interpreted as holding harvesting to the incremental growth, but if it were interpreted in
that way, then it still does not substantially restrict global harvest because it allows
elimination of the national, regional or global forest carbon sink as discussed in the main
Article 26, paragraph 6 sets forth certain provisions for the source of forest biomass
related to land use management. The basic requirement is that forest biomass comes from a
country that “has submitted a Nationally Determined Contribution (NDC) to the United
Nations Framework Convention on Climate Change (UNFCCC), covering emissions and
removals from agriculture, forestry and land use which ensures that either changes in
carbon stock associated with biomass harvest are accounted towards the country's
commitment to reduce or limit greenhouse gas emissions as specified in the NDC.” As
discussed in the main text, this provision will not discourage use of wood that increases
greenhouse gas emissions both because most country commitments are quite weak and
because all the provision might require at most is a requirement that countries compensate
for lost carbon in the forest by additional mitigation elsewhere. If a country does not
participate in the Paris accord, e.g., the United States at this time, biomass can still qualify.
According to the agreed upon text, “When evidence referred to in the first subparagraph is
not available, the biofuels, bioliquids and biomass fuels produced from forest biomass shall
be taken into account for the purposes referred to in points (a), (b) and (c) of paragraph 1 if
management systems are in place at the forest sourcing level to ensure that carbon stocks
and sinks levels in the forest are maintained or strengthened over the long term.” Although
this language may require maintenance of carbon sinks in a forest, not merely carbon stocks,
a supplier will be able to meet it just by showing that the forest material harvested would
otherwise have been harvested because that cannot reduce a sink. This language might
therefore restrict or limit new harvesting in an area but will allow diversion of wood from
other uses. Doing so would therefore not restrict the global significance of the additional
The sustainability criteria include many references to biodiversity. The most specific of
these is Commission language, which provides that “areas of high conservation value,
including wetlands and peatlands, are protected,” and which the Parliamentary version
altered to apply to areas “legally designated for nature conservation purposes”. In addition,
Recital 71 provides that biomass should not originate from areas classified by the UN Food
and Agriculture Organization as “primary forests,” or are protected by national laws.
However, FAO defines primary forests as follows: “Naturally regenerated forest of native
species, where there are no clearly visible indications of human activities and the ecological
processes are not significantly disturbed.“20. This is a highly restrictive category of forests.
In addition, the language regarding biodiversity or even quantities of forest harvest
applies only to the forest area being harvested. One of the major effects of the directive is
likely to be the diversion of wood from already highly managed forests and many other
disturbed forests, which increases the global demand for wood. Nothing in the directive
restricts what happens to even the most biologically diverse forest as a result of this
displacement of wood from more managed forests.
Supplementary Note 6 – Biomass and Global Forest Carbon Sink
Pan et al.21 estimates a net, global terrestrial carbon sink of 1.04 GtC. Assuming 50%
carbon, that translates into ~2.1 GtDM, which is equal to ~230% of existing industrial wood
harvest (~0.9 Gt) as shown in Supplemental Table 1. As a result, existing industrial wood
harvest (harvests not used for traditional firewood and charcoal) could be more than tripled
(from ~0.9 Gt to ~3Gt) without eliminating the net global forest sink and would therefore be
allowable under the sustainability principle of harvesting only the annual incremental
growth of forest.
Typical approaches to sustainability suggest netting at a country not global level, which
would greatly expand this potential harvest with a rule of harvesting only incremental
growth and just up to the elimination of the carbon sink. Using country-netting, countries
with larger sinks could increase their harvests for bioenergy and still meet this sustainability
standard without having to maintain a sink to compensate for the net forest carbon losses in
other countries. As a result, the total sink just of countries with sinks exceeds the ~2.1GtDM
global, net sink from all countries. Our estimate of at least a three-fold increase in allowable,
industrial harvest is therefore conservative.
1. Miles, P. D. & Smith, W. B. Specific gravity and other properties of wood and bark for 156
tree species found in North America. (U.S. Department of Agriculture, Forest Service,
Northern Research Station, 2009).
2. International Energy Agency. World Energy Balances 2017: Overview. (2017).
3. European Commission. IMPACT ASSESSMENT Accompanying the document Proposal for
a Directive of the European Parliament and of the Council amending Directive
2012/27/EU on Energy Efficiency. (2016).
4. Laganière, J., Paré, D., Thiffault, E. & Bernier, P. Y. Range and uncertainties in estimating
delays in greenhouse gas mitigation potential of forest bioenergy sourced from Canadian
forests. GCB Bioenergy 9, 358–369 (2017).
5. Brack, D. Woody Biomass for Power and Heat Impacts on the Global Climate. (Chatham
6. Röder, M., Whittaker, C. & Thornley, P. How certain are greenhouse gas reductions from
bioenergy? Life cycle assessment and uncertainty analysis of wood pellet-to-electricity
supply chains from forest residues. Biomass and Bioenergy 79, 50–63 (2015).
7. Sterman, J. D., Siegel, L. & Rooney-Varga, J. N. Does replacing coal with wood lower CO2
emissions? Dynamic lifecycle analysis of wood bioenergy. Environ. Res. Lett. 13, 015007
8. IPCC. IPCC Guidelines for National Greenhouse Gas Inventories. (Published: IGES, 2006).
9. Mitchell, S. R., Harmon, M. E. & O’Connell, K. E. B. Carbon debt and carbon sequestration
parity in forest bioenergy production. Glob. Change Biol. Bioenergy 4, 818–827 (2012).
10. Holtsmark, B. Harvesting in boreal forests and the biofuel carbon debt. Climatic Change
112, 415–428 (2012).
11. Jürgensen, C., Kollert, W. & Lebedys, A. Assessment of industrial roundwood production
from planted forests. FAO Planted Forests and Trees Working Paper FP/48/E. (2014).
12. eurostat. Share of renewable energy in gross final energy consumption.
13. European Environment Agency. Renewable energy in Europe – 2017 Update. Recent
growth and knock-on effects. (2017).
14. Hennenberg, K., Böttcher, H., Fehrenbach, H. & Bischoff, M. Short analysis of the RED
2009, the iLUC Directive 2015 and the 2016 RED proposal regarding implications for
nature protection. (Oeko-Institute and IFEU, 2017).
15. Edwards, R., Larivé, J. F., Rickeard, D. & Weindorf, W. Well-to-Tank Report Version 4.a:
JEC Well-to-Wheels Analysis. (European Commission, Joint Research Centre, Institute for
Energy and Transport, 2014).
16. European Commission. Energy efficiency first: Commission welcomes agreement on
energy efficiency. (2018). Available at: http://europa.eu/rapid/press-
release_STATEMENT-18-3997_en.htm. (Accessed: 14th August 2018)
17. European Commission. IMPACT ASSESSMENT Sustainability of Bioenergy. (2016).
18. PricewaterhouseCoopers EU Services EESV’s consortium. Sustainable and optimal use of
biomass for energy in the EU beyond 2020. (2017).
19. Council of the European Union. Proposal for a Directive of the European Parliament and
of the Council on the promotion of the use of energy from renewable sources - Analysis of
the final compromise text with a view to agreement. Available at:
20. UN Food and Agriculture Association. Terms and Conditions FRA 2020. (UN Food and
Agriculture Association, 2018).
21. Pan, Y. et al. A Large and Persistent Carbon Sink in the World’s Forests. Science 333,