This document is an excerpt from the EUR-Lex website
Document 52012SC0343
COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the document Proposal for a Directive of the European Parliament and of the Council amending Directive 98/70/EC relating to the quality of petrol and diesel fuels and amending Directive 2009/28/EC on the promotion of the use of energy from renewable sources
COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the document Proposal for a Directive of the European Parliament and of the Council amending Directive 98/70/EC relating to the quality of petrol and diesel fuels and amending Directive 2009/28/EC on the promotion of the use of energy from renewable sources
COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the document Proposal for a Directive of the European Parliament and of the Council amending Directive 98/70/EC relating to the quality of petrol and diesel fuels and amending Directive 2009/28/EC on the promotion of the use of energy from renewable sources
/* SWD/2012/0343 final */
COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the document Proposal for a Directive of the European Parliament and of the Council amending Directive 98/70/EC relating to the quality of petrol and diesel fuels and amending Directive 2009/28/EC on the promotion of the use of energy from renewable sources /* SWD/2012/0343 final */
COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the document Proposal for a
Directive of the European Parliament and of the Council amending Directive 98/70/EC
relating to the quality of petrol and diesel fuels and
amending Directive 2009/28/EC on the promotion of the use of energy from
renewable sources
TABLE OF CONTENTS 1........... Section: Procedural
issues and consultation of interested parties. 5 1.1........ Background. 5 1.2........ Organisation and timing. 5 1.3........ Consultation of the
Impact Assessment Board. 6 2........... Section: Problem
definition. 7 2.1........ Introduction. 7 2.2........ Scene setter 7 2.3........ The characteristics of
Indirect Land-use Change. 10 2.4........ Modelling of indirect
land-use change emissions. 10 2.5........ Underlying drivers. 15 2.6........ Who is affected by
indirect land-use change?. 16 2.7........ How are existing policies
and legislation affecting indirect land-use change?. 17 2.8........ Baseline scenario for the
assessment of indirect land-use change. 17 2.9........ The right to act 26 3........... Section: Policy
objectives. 27 3.1........ Treaty based general objectives. 28 3.2........ General objectives. 28 3.3........ Specific and operational
objectives. 29 4........... Section - Policy
options. 30 4.1........ What are the possible
options for achieving the policy objectives?. 30 4.2........ Option A – take no action
for the time being; while continuing to monitor 30 4.3........ Option B - increase the
minimum greenhouse gas saving threshold for biofuels. 30 4.4........ Option C - introduce
additional sustainability requirements on certain categories of biofuels. 31 4.5........ Option D - attribute a
quantity of greenhouse gas emissions to biofuels reflecting the estimated
indirect land-use impact. 33 4.6........ Option E - Limit the
contribution from conventional biofuels to the Renewable Energy Directive
targets. 34 5........... Analysis of impacts. 35 5.1........ Assessment methodology. 35 5.2........ Option A - Take no action
for the time being, while continuing to monitor 37 5.3........ Option B - Increase the
minimum greenhouse gas saving threshold for biofuels. 39 5.4........ Option C – introduce
additional sustainability requirements on certain categories of biofuels. 45 5.5........ Option D - Attribute a
quantity of greenhouse gas emissions to biofuels reflecting the estimated
indirect land-use impact. 48 5.6........ Option E - Limit the
contribution from conventional biofuels to the Renewable Energy Directive
targets. 57 5.7........ Combination of option D
with C2 - Attribute a quantity of greenhouse gas emissions to biofuels
reflecting the estimated indirect land-use impact whilst providing exemptions
to those biofuels feedstocks produced under criteria covered by option C2. 60 6........... Comparison of policy
options. 62 7........... Conclusion. 64 8........... Future monitoring and
evaluation. 67 9........... Glossary. 68 10......... Annex I – Consultation
and use of external expertise. 72 11......... Annex II – The concept
of Indirect and direct land-use change emissions. 75 12......... Annex III – Model
limitations and uncertainties. 76 13......... Annex IV – Results from
estimating indirect land-use change with models. 80 14......... Annex V – The
IFPRI-MIRAGE-BioF model: assumptions and results. 83 15......... Annex VI – Fossil fuel
comparator 88 16......... Annex VII – Trends in
land-use- availability and expansion globally. 89 17......... Annex VIII –
Interactions between existing legislation and indirect land-use change. 94 18......... Annex IX – biofuels and
related industries baseline tables. 99 19......... Annex X – Impacts on
biodiversity. 103 20......... Annex XI – Monte-Carlo
analysis of indirect land-use change emissions estimates. 105 21......... Annex XII – Potential
for mitigating indirect land-use change emissions at project level 109 22......... Annex XIII – Assessment
methodology. 111 23......... Annex XIV – Possible
response scenarios to reduced biofuel availability through the introduction of
additional requirements 111 24......... Annex XV – Developing
indirect land-use change emission factors from the results of the Monte-Carlo
analysis 116 25......... Annex XVI – Potential
effects of including the estimated indirect land use change greenhouse gas
emissions in the reporting of the greenhouse gas emission savings of biofuels. 118 1. Section:
Procedural issues and consultation of interested parties 1.1. Background Directive
2009/28/EC on the promotion of the use of energy from renewable sources (the
"Renewable Energy Directive") established mandatory targets to be
achieved by 2020 for a 20% overall share of renewable energy in the EU and a
10% share for renewable energy in the transport sector. At the same time, an
amendment was adopted to Directive 98/70/EC[1]
("the Fuel Quality Directive") which introduced a mandatory target to
achieve by 2020 a 6% reduction in the greenhouse gas intensity of fuels used in
road transport. The
contribution towards these targets from biofuels[2]
is expected to be significant. Whilst both Directives (hereafter referred to as
the Directives) include sustainability criteria including minimum greenhouse
gas saving thresholds, the greenhouse gas emissions associated with indirect
land-use change are currently ignored by the legislation. However, the
Directives request the Commission to review[3] by 31
December 2010 the greenhouse gas emissions associated with indirect land-use
change and, if appropriate, propose ways to address them. The Commission
published a report on indirect land-use change on the 22 December 2010[4]. That report (i) identified a
number of uncertainties and limitations associated with the available numerical
models used to quantify indirect land-use change; (ii) acknowledged that
indirect land-use change can reduce greenhouse gas emissions savings associated
with biofuels; and (iii) indicated that if action is required, indirect land-use
change should be addressed under a precautionary approach. Most importantly, it
concluded that the Commission would prepare an Impact Assessment based on the
four options identified in the report, accompanied, if appropriate, by a legislative
proposal to amend the Directives. 1.2. Organisation and timing In order to
better understand the potential indirect land-use changes and impacts
associated with the production of biofuels, a number of analytical studies were
commissioned by different Commission Services. An inter-service working group[5] was established in 2009 and met
regularly during 2009 and 2010. Discussions in this group have provided an
important input to these studies. Following the
publication of these studies in mid-2010, the group focused on the production
of the impact assessment report, with meetings of the Impact Assessment
Steering Committee taking place in 2011 on 3 February, 17 February, 9 March, 18
March and 16 May. The Impact Assessment is relying mainly on the work of the
International Food Policy Institute (IFPRI). This work ("Global trade
and environmental impact study of the EU biofuels mandate")[6] takes into account stakeholder
feedback collected through the different consultation events outlined above,
and has used the most recent biofuel demand estimates up to 2020 as outlined by
the Member States in their national renewable energy action plans[7]. The Commission considers this
work to represent the best available science with regards to the estimated
indirect land-use change impacts associated with biofuels consumed in the EU.
The findings of the IFPRI-report were presented to stakeholders at a meeting on
18 November 2011[8].
The Commission also carried out two public
consultation exercises on approaches for dealing with indirect land-use change
in 2009 and 2010. Moreover, the Joint Research Centre organised various expert
consultation meetings with academics and experts in the field in 2009 and 2010.
Further detail on these exercises can be found in Annex I. 1.3. Consultation of the Impact Assessment Board The present Impact Assessment takes into account the
recommendations formulated by the Impact Assessment Board on 4 May 2011 and on
24 August 2011[9].
The comments from the Board were incorporate in this
Impact Assessment as follows: ·
An option evaluating imposing a limit on the
contribution of first generation biofuels has been introduced. ·
Restructuring of text and further clarifications
regarding the nature of the problem and the framing of biofuels and ILUC in a
broader context (i.e. the Climate and Energy package, global and EU overall
land use change and associated emissions, global trade, linkages with LULUCF,
etc). ·
The description of the baseline scenario was
restructured and extended to better describe the state of industry sectors
involved throughout the entire biofuel production and deployment chain i.e.
agricultural production, processing capacity (i.e. crushing of oilseeds),
production plants for biodiesel, bioethanol and advanced biofuels and
developments on the car fleet, down to Member State level where possible. ·
The presentation of options has been clarified
and the impact analysis and presentation have been restructured to assist
readability and enhance the link with the objectives. ·
More detail has been provided on the assessment
of environmental, economic and social impacts. Further work from the Joint
Research Centre on the biodiversity impacts has also been included in the
assessment. ·
A glossary of technical terms has been added. 2. Section:
Problem definition 2.1. Introduction The Directives
impose a number of sustainability criteria aimed at preventing the conversion
of land characterised by high carbon stock and high biodiversity for biofuel
production. Moreover, they also require biofuels to achieve minimum greenhouse
gas emission savings of 35% compared to fossil fuels[10]. The methodology
defined in the Directives to determine the greenhouse gas saving takes account
of emissions associated with direct land-use change, as well as emissions
coming from the production of biofuels. However, emissions associated with
indirect changes in land-use are currently not included (a figure explaining
both direct and indirect land-use change is provided in Annex II). In the context
of the mandatory targets set by the Directives to achieve the specified
greenhouse gas savings, and the 6% reduction in greenhouse gas intensity
required by the Fuel Quality Directive, the key problem addressed by this
impact assessment is whether greenhouse gas emissions associated with indirect land-use
change should be addressed, and if so in which way. As this impact assessment
is focused on the specific requirement related to greenhouse gas emissions from
indirect land-use change, it does not consider any wider environmental and social
impacts associated with the promotion of biofuels. The Commission intends to
consider these aspects in the Renewable Energy Directive's biennial reports to
the European Parliament and the Council from 2012 onwards. 2.2. Scene setter 2.2.1. The Climate and Energy
Package targets In March 2007
the EU’s leaders endorsed an integrated approach to climate and energy policy
aimed at combating climate change and increasing the EU’s energy security while
strengthening its competitiveness and transforming itself into a highly
energy-efficient, low carbon economy. As part of this process, a series of
demanding climate and energy targets to be met by 2020 were set, including, ·
a reduction in EU greenhouse gas emissions of at
least 20% below 1990 levels ·
a 20% of EU energy consumption to come from
renewable resources ·
and a 20% reduction in primary energy use
compared with projected levels, to be achieved by improving energy efficiency. Biofuels represent around 1 and 2.5
percentage points of the 20% greenhouse gas reduction and renewable energy
targets respectively. However, they are, as estimated in the National Renewable
Energy Action Plans[11],
expected to be the major contributor towards the sub-targets for 10% and 6%
renewable energy and greenhouse gas emission reductions in the transport sector
to 2020, as set by the Renewable Energy and Fuel Quality Directives
respectively. 2.2.2. Transport emissions
reductions to 2050 The EU is committed to achieving by 2050 an 80 to 95% reduction in
greenhouse gas emissions economy wide compared to 1990 levels. The recent
"A Roadmap for moving to a competitive low-carbon economy in 2050"[12] foresees that the transport
sector needs to reduce its greenhouse gas emissions by around 60% compared to
1990 levels by 2050 to ensure a comparable cost-effectiveness of greenhouse gas
emissions abatement in that sector. This objective has been confirmed in the
recently published transport white paper "Roadmap to a Single European Transport Area – Towards a competitive
and resource efficient transport system"[13]. Transport
emissions can be reduced through measures which affect i) the amount of
transport activity, ii) the energy efficiency with which that transport
is carried out and iii) the greenhouse gas intensity of the energy used
to perform the transport. Biofuels are one of the alternative energy carriers
available that offer the potential to reduce the greenhouse gas intensity of
the fuel. The use of biofuels may reduce greenhouse gas emissions provided that
direct and indirect greenhouse gas emissions are lower than those from the
fossil fuels they replace. Given the overall transport greenhouse gas reduction
goal, the degree to which one of the three levers to reduce emissions is not
deployed, the more action that will be required from the other two. 2.2.3. Global land-use and land-use
change emissions The globe has approximately 13 200 Mha of
land, of which around 1600 Mha is used for cropping[14],[15]. The IPCC special report
on renewable energy[16]
estimates that 780 Mha of land are available for bioenergy production without
irrigation worldwide, mostly consisting of unprotected grassland and woodland
found in Africa (35%), Latin America (21%), North America (16%) and Europe
(14%), having the potential to deliver bioenergy amounting to more than 4000
Mtoe. The estimated total biofuel use in the EU in 2020 (27 Mtoe) is in
comparison expected to cause a total land use change of less than 3 Mha of land
globally[17].
The IEA biofuels for Transport - Technology
Roadmap assumes that 27% of total transport fuel demand will be covered by
biofuels in 2050. The biofuel demand and resulting land requirements are shown
for comparison below. Figure 1: Demand
for biofuels (left) and resulting land demand (right) assumed in the IEA biofuel
technology roadmap (Source: IEA Technology Roadmap[18]) The production of conventional bioethanol
and biodiesel increases towards 2020, and then decreases, disappearing around
2045, while bioethanol from sugar cane increases over the whole period. Land-use
for biofuels increases from 30 Mha today[19]
to around 110 Mha in 2050, which corresponds to around 7% of current cropland. With regard to annual global emissions
(50.000 Mt CO2)[20],
annual emissions from land-use change represent around 15%[21] of the total (7500 Mt CO2).
In this context, estimated indirect land-use change emissions from EU biofuel
consumption in 2020 are likely to represent a very small share (0.1% if based
on annual estimated emissions by IFPRI-MIRAGE-BioF at 50 Mt CO2).
However, this level of emissions deserves consideration in the context of
greenhouse gas emissions savings offered by biofuels. This is discussed in more
detail in section 2.4.3 and 2.8.6. 2.2.4. Bioenergy and biofuel
production in a global context Bioenergy is the dominant renewable energy
source, amounting to 10% (1200 Mtoe) of global primary energy supply. The IPCC
estimates that the use of sustainable bioenergy will triple[22] towards 2050 in order to meet
climate change objectives[23].
Careful policy making, including considerations of indirect land-use change
impacts, is necessary to mobilise such quantities in a sustainable way. Most of today's biofuels are produced from
agricultural crops like maize, sugar cane and rapeseed. Total global production
of biofuels reached 70 Mtoe in 2008, which represents 1.7% of global oil
consumption. While less than 3% of global cropland is used for producing
biofuels, the relative importance of biofuels within certain global markets is
significant. For example, globally 16% of vegetable oils (rapeseed, soybean,
palm and sunflower oil) are used for biodiesel, 15% of maize and some 2% of
wheat is used for bioethanol[24].
Biofuels are traded globally, as can be
seen in figure 2 below, which depicts the production and trade of biofuels in
2009. Figure 2: 2009 production and trade of biofuels.
Source: IPCC (see footnote 16) The US (maize) and Brazil (sugar cane) are the main biofuel producers, producing more than the rest of the
world combined. The EU is the largest global market for biodiesel because of
the dominance of diesel in the car fleet. In 2009, the EU imported soy
biodiesel mainly from Argentina and US, and to a significantly lesser extent
palm oil from South East Asia. Bioethanol, to be blended with petrol, was imported
from Brazil. Two thirds of the biofuels consumed in the EU are currently
produced domestically, with the share of imports expected to grow towards 2020
(IFPRI-MIRAGE-BioF estimates that half of biofuels will be imported in 2020). 2.3. The characteristics of Indirect Land-use Change Most of today's biofuels are produced from
crops grown on agricultural land such as wheat and rapeseed. When agricultural
or pasture land previously destined for the food, feed and fibre markets is
diverted to the production of biofuels, the non-fuel demand will still need to
be satisfied. Although this additional demand can be met through
intensification of the original production, bringing non-agricultural land into
production elsewhere is also possible. It is in the latter case that land-use
change occurs indirectly, (hence the term indirect land-use change). While most biofuel feedstocks are being
produced in the EU, the estimated indirect land-use change emissions are mostly
expected to take place outside the EU, where the additional production is
likely to be realised at the lowest cost. In the case that this production is
realised through the use of additional land, its conversion could lead to
substantial greenhouse gas emissions being released if high carbon stock areas
such as forests are affected as a result. 2.4. Modelling of indirect land-use change emissions Estimating the
greenhouse gas impact due to indirect land-use change requires projecting
impacts into the future, which is inherently uncertain, since future
developments will not necessarily follow trends of the past. Moreover,
estimated land-use change can never be validated, as indirect land-use change
is a phenomenon that is impossible to directly observe or measure. Therefore
modelling is necessary to estimate its occurrence. No
macro-economic models used to estimate indirect land-use change emissions are
currently capable of modelling the effects of the EU sustainability criteria,
so these criteria are consequently assumed not to have any effect. As such, the
models are not able to distinguish between direct and indirect land-use
change. Nevertheless, the current estimates from the
models are considered the best approximation for estimating indirect land-use
change emissions. Several
non-economic factors influence what land-use change takes place and where it
occurs. Some of these drivers are related to political choices (land-use and
agricultural policy, land rights, etc.), others to institutional features
(proximity to infrastructure and markets, land-use legislation). Therefore
conceptual limitations will always remain. Annex III provides more details on
the various modelling approaches, and the related uncertainties and
limitations. 2.4.1. Results from modelling
indirect land-use change As set out above,
the Commission launched a number of studies in 2009, 2010 and 2011 on indirect
land-use change. Further details on the assumptions and results from these
studies are provided in Annex IV. A description of one of them in particular,
the IFPRI-MIRAGE-BioF model, which is the basis for the modelling establishing
the baseline used in this Impact Assessment, can be found in Annex V. The MIRAGE-BioF
model, developed by the ATLASS consortium, has been improved over the last
three years in consultation with the Commission to model the consumption of
biofuels used in the EU. Although a number of limitations and uncertainties
remain, this model has been found to be the most suitable one to estimate the
indirect land-use change emissions in the EU context. The IFPRI-MIRAGE-BioF
model is a general equilibrium model, which encompasses all economic sectors
and markets and their inter-actions at a global scale. The model is run in a
"baseline scenario", and a "policy scenario", where the
only difference is the EU biofuel policy. The resulting difference in land-use
change emissions is then divided by the additionally produced biofuels. In
addition, the Commission has identified a number of other sources illustrating
different (indirect) land-use change emissions from different feedstocks.
Figure 3 provides a summary of the most relevant modelling exercises undertaken[25]. It sets out calculations of
estimated indirect land-use change emissions in gCO2/MJ for a range
of different feedstocks showing the range of volumes obtained, and converted
where necessary to a 20 year timeframe[26].
Figure 3: Summary of estimated (indirect) land-use
change emissions. Source: various 2.4.2. Short and long term
developments to deal with limitations and uncertainties There are a
range of key assumptions used in the indirect land-use change models that can
have a substantial impact on the indirect land-use change estimates. A first
step in dealing with this cause of uncertainty is to understand the parameters
involved. The Commission's model comparison exercise and the literature review
have provided an indication of a limited number of key factors that are of high
importance. These include, but are not limited to: co-products, yield
developments, carbon stocks and displacement/substitution of other commodities.
These aspects have also been considered by the Low Carbon Fuel Standard Expert
Workgroups established as part of California Air Resource Board's attempts to
improve the modelling of indirect land-use change[27]. In relation to
uncertainty in the model results due to data and assumptions, an approach that
enables the parameters being considered to be varied randomly according to an
expected probability function can be used. This so-called Monte Carlo analysis
is a standard approach to dealing with uncertainty in modelling, and the method
chosen by the US Environmental Protection Agency in their attempt to estimate
indirect land-use change emissions. 2.4.3. Overall greenhouse gas
balance of using biofuels in the EU The emissions associated with the
cultivation, processing and transport of biofuels have been extensively
explored by the Commission making use of the JEC's Well to Wheel study[28], and form the basis for the
greenhouse gas intensity values established in EU legislation. These values for
land using first generation biofuels[29]
vary from around 20 g/MJ to 60 g/MJ, and do not include emissions for either
direct or indirect land-use change. The Directives require that biofuels
counted towards the targets need to save at least 35% greenhouse gas emissions
compared to average fossil fuel emissions, using a fossil fuel comparator
(FFC), which is currently set at 83.8g/MJ[30].
However, more recent research indicates that a higher number would be more
accurate[31].
Different models
assume different fossil fuels being substituted by biofuels. The literature
review finds that to some degree higher production cost is linked with higher
greenhouse gas emissions, but not systematically. For example, "deep
water" and "artic" sources are more costly for other reasons
than high energy consumption per barrel extracted crude. However, the general
picture is that more expensive crudes are connected with higher emissions. In the context
of analysing indirect land-use change, a consequential lifecycle analysis is
applied for the land resources, which implies that the global net effect
is analysed. This is why land-use changes taking place in areas where no
biofuel is produced still has an impact on the estimated indirect land-use
change emissions of biofuels. Applying the same framework to fossil fuel, it is
appropriate to compare overall emissions from biofuels to global marginal
emissions from fossil fuels not being extracted as a consequence of using
biofuels. Set out below is a simplified figure depicting the application of the
above approach[32]. Figure 4: Emissions balance of biofuels in 2020, including
estimated indirect land-use change emissions, compared to the emissions of
fossil fuels not extracted. The global marginal emissions from fossil
fuels are expected to be higher than average emissions of fossil fuels used in
the EU, the latter being reflected in the fossil fuel comparator (FFC), which
in this assessment has been assumed to be 90.3 g/MJ in 2020. As can be seen
from figure 4, the overall greenhouse gas emissions balance of the estimated
biofuel mix compared to fossil fuels is expected to be positive in 2020,
implying that the use of biofuels will save emissions also when the estimated
indirect land-use change emissions are taken into account. Irrespective of the
emissions from conventional sources of fossil fuels, if biofuels are to lower
the overall greenhouse gas emissions from transport fuels significantly and to
an increasing degree, the greenhouse gas intensity has to be reduced over time[33]. It is in this context that
indirect land-use change emissions pose a challenge. 2.5. Underlying drivers The drivers behind indirect land-use change
can be summarised as the increased demand for crops resulting from increased
biofuel use, coupled with poor land-use governance in areas with high carbon
stock land and lack of complete accounting rules and emission targets for land-use
change globally. 2.5.1. Land availability globally The basic
driver for indirect land-use change is the increased demand for agricultural
crops as a result of increasing biofuel production in a situation where
potential yield increases are limited and demands (most notably for food and
feed) are not fully elastic. Some other key factors, such as achieving maximum
profit from the production and complying with relevant legislation, are also
likely to play a role in determining how the increased demand is to be
realised. The extent to
which land availability is limited in various regions of the world is much
debated. Compared to 1981 the harvested land has significantly declined in
Europe, CIS and North America, thus suggesting that there would be low carbon
stock land available[34].
With regard to the EU, it is expected that the agricultural area will continue
to reduce by around 0.5 million hectares each year. Further details on this can
be found in Annex VII. 2.5.2. Where is agricultural land
expanding? Although it is
clear that a significant amount of land is available in certain areas of the world,
it is difficult to govern the proper use of these land areas. Recent studies
suggest that tropical forests were the primary sources of new agricultural land
in 1980-90s, with various studies highlighting a significant role for soy
production and cattle ranging, as well as palm oil, as drivers behind the
expansion of agricultural land into the Amazon and South East Asia respectively[35]. The lack of
effective protection of forests and carbon rich areas is another factor that
allows damaging indirect land-use change to take place. If conversion of carbon
rich areas such as forests and wetlands were to be limited, the risk of
damaging indirect land-use change would be lower. Further information,
including international developments in this area, can be found in Annex VII. 2.5.3. Accounting for land-use, land-use
change and forestry (LULUCF) In national greenhouse
gas inventories (which are the basis for countries' emission commitments, such
as the EU's greenhouse gas target) emissions from the burning of biofuels are
reported under the energy sector as zero. This means that emissions are not
added to the total national emissions and that it is assumed that any
greenhouse gas emissions from land (including any direct and indirect land-use
change) are captured under the "land-use, land-use change and
forestry" (LULUCF) sector of the inventory[36]. So, while incentives exist to promote the
use of bio-energy[37],
a coherent approach to climate change mitigation in the LULUCF sector via
measures in agriculture, forestry and related industries at the global level is
only in the making. The LULUCF sector has a positive and significant impact on
the EU's greenhouse gas emissions. The sector removes the equivalent of 9% of
greenhouse gases emitted in other parts of the economy[38]. Although emissions and
removals from LULUCF are reported under the UNFCCC and partially accounted
under the Kyoto Protocol, the sector was left out of the EU's climate
commitments for 2020 under the Climate and Energy Package[39] due to the recognition of serious
deficiencies in international accounting rules of emissions from this sector (accounting for emissions and removals is only mandatory for
some land use change activities including afforestation, deforestation and
reforestation). Developing countries do not account at all. Taken together, this
means that emissions from land-use changes in developed countries due to
agricultural expansion is unlikely to be fully reflected in the accounting, and
that land-use change emissions in developing countries are not accounted for. An international agreement on revised
accounting rules for LULUCF for the second commitment period under the Kyoto
Protocol post 2012 was achieved at the 17th Conference of the
Parties to the UNFCCC ("COP17") in Durban in December 2011. In
particular accounting for forest management activities, including harvested
wood products, will be mandatory and definitions for natural disturbances and
"wetland drainage and rewetting" have been established. Following this agreement the Commission
tabled a proposal on 12 March 2012 on how the LULUCF sector increasingly could
be integrated in the EU's climate policy using a step-wise approach. As a first
step, it proposes establishing robust, common accounting, monitoring and
reporting rules mandatory for forests, forest management, croplands and
grassing land as well as national LULUCF action plans (LAP). In view of the
sector's specific emissions profile, the Commission proposed a dedicated legal
framework, rather than including it in the EU Emissions Trading Scheme or the
rules created by the Effort Sharing Decision. The second step would be to formally include
LULUCF in the EU's greenhouse gas reduction target. It is proposed to take this
step when the Member States have implemented the accounting framework and it
has proven to be robust. Accounting for LULUCF would clarify the
benefits of sustainable bio-energy by better reflecting related emissions, in
particular resulting from the combustion of biomass, which is unaccounted for
at the moment. This would strengthen the incentives provided by sustainability
criteria in the context of renewable energy targets. However, implementing LULUCF accounting, monitoring and reporting in
the EU is likely to have a limited effect on the estimated indirect land-use
change emissions globally, as these take place mostly outside of the EU. An
implementation of LULUCF accounting on a global scale, combined with
commitments for reducing emissions, could significantly reduce the indirect
land-use change emissions, as converting high carbon stock land would have a
cost. 2.6. Who is affected by indirect land-use change? Climate change
is a global problem, while the socio-economic consequences of indirect land-use
change have regional and national effects affecting the global population.
Regulations to address indirect land-use change emissions in the field of
biofuels may affect local communities, biofuel feedstock producers, the biofuel
industry, Member States and third countries in various ways. These will be
incorporated in the assessment of the policy options in section 5. Although
land-use change can have a wide range of positive and negative impacts (i.e.
greenhouse gas emissions, biodiversity, economics, social issues, etc), this
report focuses on the consequences for the greenhouse gas emissions of
biofuels, as required by the Directives. The Commission will analyse wider
sustainability impacts associated with the promotion of biofuels in the
Renewable Energy Directive's biennial reports to the European Parliament and
the Council from 2012 onwards. 2.7. How are existing policies and legislation affecting
indirect land-use change? Developments driven by existing legislation
in a number of areas could have a significant impact on indirect land-use
change. These include existing EU legislation relating to biofuels (the
Renewable Energy and Fuel Quality Directives), as well as wider agricultural
(i.e. Common Agricultural Policy), environmental (i.e. biodiversity, forestry,
REDD+), trade (i.e. agricultural tariffs), developmental (i.e. investment into
agriculture) and research (i.e. agricultural research and advanced biofuels)
policies. Further details are found in Annex VIII. 2.8. Baseline scenario for the assessment of indirect land-use
change In order to be able to assess the full
impacts of the policy options being considered in this assessment, this section
aims to provide an overview of the EU biofuel and related industries, and the
estimated indirect land-use change emissions associated with the increased
feedstock demand for biofuels. 2.8.1. Overview of biofuels and
related industries The production of biofuels involves
economic activity and employment all along the supply chain; in agriculture,
logistics and at biofuels production facilities, but also in sectors that
supply to or support biofuels supply chains, and is generally more labour
intensive than fossil fuels. The expected employment related to biofuels in EU
in 2020 could be around 400,000 jobs in total[40].
2.8.1.1. EU production and
consumption - 2008 and 2020 Reported 2008 and estimated 2020
consumption figures for biofuels and other renewable energy sources (RES) in
transport are shown in table 1 below. The 2020 figures are based on the
National Renewable Energy Action Plans (NREAPS), which have been submitted by
the Member States[41].
The NREAPs are also the basis for the baseline established in this Impact
Assessment. Compared to the expected increase of biofuels, bioliquids are
expected to play a small role in contributing towards the overall 20% RES
target at around 5.5 Mtoe (4.4 Mtoe and 1.1 Mtoe going into the production of
heat and power, and electricity generation respectively). This does not
represent a significant increase compared to 2008 levels. The production of
biofuels from waste feedstocks and advanced biofuel technologies is not
expected to be significant and lower than anticipated, reaching 2.3 Mtoe
(approximately 1.5 percentage point with double counting) in 2020. It appears
that the current incentives, particularly,those set out in Article 21(2)[42] of the Renewable Energy
Directive, are not enough to spur the desired level of investment in advanced 2nd
generation biofuels. || 2008 || 2020 Total transport fuels (Mtoe) || 239 || 312 1st generation biofuels (Mtoe) || 10 || 26.5 of which biodiesel (Mtoe) || 8.2 || 19.8 of which bioethanol (Mtoe) || 1.8 || 6.7 || || 1st generation biofuel (p.p of RES-T) || 3.5 || 8.6 Biofuels from waste and 2nd generation share (p.p of RES-T) || 0 || 1.5 Renewable electricity in transport (p.p of RES-T inc) || 0.4 || 1.4 Table 1:
RES in transport 2008 and 2020. 2.8.1.2. EU agricultural production Biofuel feedstocks currently used are
typically 'first generation' and include biodiesel and bioethanol derived
mostly from crops, (i.e. cereals, sugars and oil crops) except those produced
from waste feedstocks. In
the EU, the share of the cereal production[43],[44] consumed in the
bioethanol market was around
9.4 Mt during the 2009/10 marketing year (3.2% of a total EU cereal production
at 292 Mt), with wheat being the most common feedstock used. Moreover,
estimated consumption of sugarbeet in the EU bioethanol market is about 6 Mt
(5.4% of the total EU sugar beet production at 110 Mt)[45]. Across the EU, the largest
producers of cereals and sugar beet respectively were France (70Mt and 33 Mt), Germany (50 Mt and 25 Mt), Poland (30 Mt and 11 Mt) and the UK (18 Mt and 8 Mt). With regard to biodiesel, the share of
vegetable oils destined for this market represented around 9 Mt. This equals to
38% of the estimated EU consumption of vegetable oils 23.4Mt in 2010/11[46]. Of this oils market, around
41% is imported as oil, and around 10% is imported as beans (mainly soya)[47]. EU oil crops production
concerns mainly rapeseed (20.4 Mt in 2010/2011), sunflower seed (6.7 Mt) and
soya (1.1. Mt). Across the EU, the largest producers of rapeseed were Germany (6.3 Mt), France (5.6 Mt), Poland (2.5 Mt) and UK (2 Mt). For sunflower, these were France (1.6 Mt), Bulgaria (1.3 Mt), Hungary (1.3 Mt), Romania (1 Mt) and Spain (0.8 Mt). The
EU-27 is traditionally a net exporter of cereals, but a net importer of
vegetable oils and oilseeds (despite recently achieving record production
levels of oil crops), and to a lesser extent of sugar. Current forecasts
predict that this trend, with regard to cereals and vegetable oils, will
continue to 2020. 2.8.1.3. Trade in biofuels The table below shows the current and
estimated split of biofuels across feedstocks in 2020. || 2008 (%) || Source 2008 || 2020 (%) || Source 2020* Biodiesel || 83 || || 72 || Rapeseed || 57 || Europe || 40 || Europe and imports Soya || 20 || Argentina, USA || 11 || Argentina, USA Palm oil || 4 || South East Asia || 17 || South East Asia Sunflower || 2 || Europe and imports || 4 || Europe and imports Bioethanol || 17 || || 28 || Sugar cane || 6 || Brazil || 13 || Brazil Wheat || 5 || Europe || 6 || Europe and imports Sugar beet || 3 || Europe || 5 || Europe Maize || 3 || Europe and imports || 4 || Europe and imports Table 2:
Land using biofuels currently used in the EU[48]
and estimations for 2020[49]. With regard to imports vs. domestic
production, IFPRI-MIRAGE-BioF estimates that biodiesel imports will grow from
0.75 Mtoe in 2008 to 2.5 Mtoe in 2020 (mostly from Indonesia, Malaysia and
Latin America); whereas bioethanol imports will increase from under 1 Mtoe in
2008 to around 3.5 Mtoe in 2020 (mostly from Brazil). In addition, a total of
14 Mt of feedstocks will also be imported into the EU (rapeseed, oil palm and
maize having the largest share) as a result of the additional demand[50]. Therefore, it is expected
that about half of the biofuels consumed in the EU in 2020 would be
domestically produced, with rapeseed being the main feedstock. 2.8.1.4. Biofuel installed production
capacity Production capacity in the EU, both
installed and under construction, currently stands at 24.5 Mtoe, of which 19.8
Mtoe is for biodiesel and 4.3 Mtoe for bioethanol. With regard to its
distribution across Member States, most of the EU biodiesel capacity can be
found in Germany (4.5 Mtoe), Spain (3.7 Mtoe), France (2.3 Mtoe), Italy (2.1 Mtoe) and the Netherlands (1.2 Mtoe)[51].
The smaller bioethanol capacity is distributed across France (0.9 Mtoe), Germany (0.7 Mtoe), UK (0.5 Mtoe), Spain (0.4 Mtoe) and Poland (0.4 Mtoe)[52].
Advanced biofuels installed capacity is currently negligible and limited to a
few pilot plants. Although the installed biodiesel production
capacity in Europe increased rapidly from 2006-7 onwards, it seems to have
slowed down in 2010. In some countries such as Germany, it has shown a slight
decrease in 2010, where some biodiesel facilities have been closed down,
decommissioned or retrofitted to other production processes. Moreover, it is
worth noting that due to a slow market uptake, capacity utilisation is at
around 50%, with total 2009 European production standing at 8.2 Mtoe and 1.9
Mtoe for biodiesel and bioethanol respectively. Germany and France alone accounted for over 50% of EU biofuel production in 2009. Other related industries include those
involved in the processing of the feedstocks, particularly oil crops into
vegetable oils before they are chemically treated to produce the final
biodiesel product. In this context, there are some 150 oil crops processing and
vegetable oils and fats production facilities across Europe, for which the
trade in biodiesel products will be one of their major markets. Of a total of
13Mt of vegetable oil being pressed in the EU in 2008, the main producing country
was Germany (4Mt), followed by France (1.9Mt), the Netherlands (1Mt), Spain (1Mt) and UK (0.8Mt). The main vegetable oils being produced were rapeseed oil (8Mt), followed
by soya oil (2.5Mt) and sunflower oil (2Mt)[53].
Full datasets for all Member States can be found in Annex IX. 2.8.1.5. Deployment of biofuels There
is some uncertainty regarding how much biofuel can be blended with petrol and
diesel, while maintaining associated warrantees from car manufactures. Based on
the biofuel volumes estimated by Member States for 2020, it seems that, in
volume terms, blends beyond 10% for diesel (currently at 7%) and around 15% for
petrol (currently at 10%) will be needed to achieve the Renewable Energy
targets EU-wide[54].
This is an important issue due to the long lead-times both in changing
specification of car engines, the slow turnover of cars, and the long lead-time
needed for changing fuel specifications. The use of various fuels as estimated
by JEC (JRC/EUCAR/Concawe) towards 2020, taking the turnover of vehicles into
account, is shown in the figure below[55]. Figure 5: Energy demand by fuel type in road
transport towards 2020 (source: JEC) Heavy duty and light duty vehicles are
referred to as HD and LD The use of petrol is expected to decline,
while the use of diesel increases. With regard to biodiesel blends, work is
currently underway to develop standards for B10 for cars and B30 for heavy duty
engines. There are also a number of plants currently producing hydrotreated
vegetable oils, which can be used at any blending levels. However, the latter
fuel is likely to be in demand from the aviation sector, which may limit the
supply available to the road transport sector. In
the case of bioethanol, the situation remains more challenging as the
petrol/diesel split is estimated to increase in favour of diesel cars towards
2020, and in addition bioethanol has a lower energy content. Work on bioethanol
blend standards is ongoing. While manufacturers can produce vehicles that are
compliant with EU emission standards at petrol and bioethanol blends up to 95%,
the sales of these vehicles are low in the EU. Certain countries such as Brazil and Sweden have shown that vehicles can readily be built to be compatible with higher levels
of oxygenates and alcohols at a low additional cost (around 100€ per vehicle or
lower). It is also possible to convert heavy duty vehicles to run on
bioethanol. In terms of current supply infrastructure
available for E85 in the EU, total sales were estimated to be about 100 ktoe in
2008[56].
Sweden had by far the most selling points for E85, with 1300, followed by France with 320 and Germany (100). The UK, Ireland, Hungary, Norway, Spain and the Netherlands had fewer than 20 stations in 2008. The situation in Sweden seems to be driven
by current incentives which include reduced registration charges and road
taxes, free parking in some cities and waived congestion charges for flexi-fuel
vehicles[57]. Biofuels
may also be employed in the shipping and aviation sectors. In terms of blending
requirements, they have opposite features; most ships can run on most
hydrocarbons, but the safety requirements for aviation are strict and only
certain types of biofuels can be used. Three types of biofuels are favoured to
be used in aviation engines, when blended with kerosene: Synthetic
Fischer-Tropsch (FT) based kerosene, Hydrogenated Vegetable Oils (HVO) and
Hydrogenated Pyrolysis Oils (HPO) produced from lignocellulosic biomass. At the
moment only HVO is available, but the two other biofuels are expected to be
available by 2020. So far no targets have been established for either the
shipping or aviation sector. However, the aviation sector aims to use around
2Mtoe of biofuels by 2020. 2.8.2. Indirect land-use change
greenhouse gas emissions assumed in the baseline The evaluation
of the policy options requires a baseline scenario that the policy options can
be compared against. As discussed in section 2.3, indirect land-use change can
only be estimated through modelling, and the Commission has taken the view that
although a number of limitations and uncertainties remain, the
IFPRI-MIRAGE-BioF is considered to be the best available estimation of the
baseline. Further results as well as the key assumptions of the IFPRI-MIRAGE-BioF
model are summarised in Annex V. The ATLASS
consortium has, since the publication of the Commission report on indirect
land-use change last year, conducted more analysis. The latest results provide
an estimate of overall land-use change emissions based on the additional 2020
biofuel volumes estimated by the Member States compared to 2008[58]. However, it is important to
note that since the modelling assumes that the sustainability criteria have no
effect, the baseline assumes consumption of some biofuels that might not meet the
greenhouse gas savings and the land-use criteria in 2020. It is also worth noting
that no further implementation of LULUCF accounting in the run up to 2020 is
assumed in the model. However, if complete accounting
for land-use, land-use change and forestry, in the framework of a system of
global emissions targets was implemented, this would send price signals
providing disincentives for such conversions. As
described in more detail in section 2.5.3 a move towards including LULUCF in
the EU's GHG reduction target can only be considered once Member States have
implemented the accounting framework and it has proven to be robust within the
EU. Other initiatives, such as the moratorium for peatland and primary forests
agreed between Norway and Indonesia, in the context of REDD+, might also impact
on indirect land-use change emissions[59].
Such agreements are not reflected in the modelling. 2.8.2.1. Total estimated indirect
land-use change impacts IFPRI-MIRAGE-BioF estimates the indirect
land-use change emissions from 2008 to 2020 to amount to around 500 Mt of CO2eq.
These emissions are equally divided between peatland emissions (in South East Asia), losses of biomass below ground, and changes in above ground biomass.
Emissions from peat conversion have a larger impact on the overall emissions
attributed to oil crops, particularly for palm oil, than for bioethanol crops.
The range of estimated indirect land-use change emissions is 24 g/MJ to 50
g/MJ, with an average of 38 gram CO2eq/MJ based on the assumed
biofuel mix. This estimate of indirect land-use change emissions is based on
conversion of around 1.7 Mha of land, which is taking place in a range of
regions globally, mostly in Brazil and the Commonwealth of Independent States
(CIS). With regard to the estimated biodiversity
impacts, IFPRI-MIRAGE-BioF
study showed that this new cropland is taken from pasture (42%), managed
forest (39%), primary forest (3%) and savannah and grassland (16%), which will
have biodiversity and wider environmental impacts. A qualitative estimation of these impacts made by the JRC
using the Mean Species Abundance (MSA) values provided by the Global
Biodiversity Model (GLOBIO 3)[60],[61], shows that the largest biodiversity losses
will be associated with the conversion of primary forest, and savannah and
grassland (both at 100% MSA), pasture land (70% MSA) and managed forest (50%
MSA) in order of decreasing significance. Cultivated and managed areas receive
a score of 10% MSA under the same classification[62]. Please see Annex X for more detail. || Forest managed || Forest primary || Pasture || Savannah Grassland || Other || TOTAL Brazil || 3% || 1% || 10% || 3% || 0% || 18% Central America and Caribbean || 0% || 0% || 1% || 0% || 0% || 1% China || 6% || 0% || 2% || 0% || 0% || 8% CIS || 11% || 0% || 9% || 3% || 0% || 23% EU27 || 4% || 0% || 1% || 2% || 0% || 7% Indonesia and Malaysia || 5% || 0% || 2% || 0% || 0% || 7% Latin America || 3% || 1% || 1% || 4% || 0% || 9% Rest of the OECD || 1% || 0% || 1% || 1% || 0% || 3% Rest of the World || 4% || 0% || 4% || 1% || 0% || 9% Sub-Saharan Africa || 1% || 0% || 11% || 1% || 0% || 13% USA || 1% || 0% || 0% || 1% || 0% || 2% TOTAL || 39% || 3% || 42% || 17% || 0% || 100% Table 3: Percentage land converted by type
and world region. Source: IFPRI-MIRAGE-BioF 2.8.2.2. Estimated indirect land-use
change impacts by feedstock The overall
estimated indirect land-use change emission impact is the aggregate of a set of
sub-results, such as country of origin and feedstock. It is therefore relevant
to present the crop-specific indirect land-use change emission estimates, as
these are components of the overall baseline impact. The IFPRI-MIRAGE-BioF
model has recently been combined with a Monte Carlo simulation, to provide a
better description of the probability distribution of the uncertainty
associated with the variables. More information on this analysis can be found
in Annex XI. Figure
6: Results of the Monte-Carlo analysis:
estimated indirect land-use change emissions (gCO2/MJ)- under scenario of
current trade policy. The bars indicate 1st and 99th
percentile, while the boxes are 25th and 75th
percentiles. It is important
to note that these crop specific values are estimated based on the increase of
biofuel consumption towards 2020 compared to the existing consumption in 2008.
They are therefore not representative for the around 10 Mtoe already consumed
in 2008. Indeed, it is pointed out that for rapeseed, which is the most
important feedstock used in 2008 (5.7 Mtoe out of a total of 10 Mtoe), the
average land-use change is significantly lower in the baseline. With regard to
the values for cereal crops, it should also be born in mind that some of the
assumptions on the yields of EU wheat and US/Brazil maize are considered
strongly optimistic. Although these assumptions can significantly influence the
estimated indirect land-use change emissions for these crops, they have not
been included in the Monte Carlo analysis. 2.8.2.3. Establishing a greenhouse
gas emissions baseline for biofuels in 2020 In order to
establish a greenhouse gas emissions baseline for biofuels, it is necessary to
compare the estimated indirect land-use change emissions to the expected direct
greenhouse gas savings from substituting fossil fuels in 2020. In this context,
assumptions regarding the expected improvements in greenhouse gas emissions
performance of biofuels, as well as changes in the carbon intensity of fossil
fuels to 2020 need to be made. With regard to the expected improvements in
greenhouse gas emissions performance of biofuels towards 2020, COWI[63] estimated how various
feedstocks would develop. However, those values do not take into account more
recent developments (such as the ETS proposals for ammonia and nitric acid plants
in EU), and do not cover improvements for all feedstocks. As such, these have
been adjusted by JRC to allow for comparison across all biofuels[64]. The results combined with the
estimated indirect land-use change emissions are summarised in the table below[65]. || Average estimated ILUC emissions || Direct emission savings || Total emissions Maize || 10 || -57 || -47 Sugar beet || 7 || -63 || -56 Sugar cane || 15 || -70 || -54 Wheat - Not specified || 14 || -40 || -26 Wheat - Natural gas/CHP || 14 || -56 || -43 Wheat - Straw/CHP || 14 || -68 || -55 Waste/2nd generation bioethanol - land using || 15 || -73 || -58 Waste/2nd generation bioethanol - non-land using || 0 || -81 || -81 Waste/2nd generation biodiesel - land using || 15 || -85 || -69 Waste/2nd generation biodiesel- non-land using || 0 || -81 || -81 Palm oil || 54 || -39 || 15 Palm oil with methane capture || 54 || -61 || -7 Rapeseed || 55 || -50 || 5 Soybean || 56 || -43 || 13 Sunflower || 54 || -58 || -4 Table 4:
Typical annual direct savings compared to estimated indirect land-use change
emissions per crop (gCO2/MJ). Source: ATLASS (2011), COWI and
Commission's calculations 2.8.2.4. Sensitivity of the baseline
scenario The uncertainty
of estimating indirect land-use change implies that several sensitivities
should be investigated. In the assessment of the options, sensitivity analysis
is limited to the changes of the assessment of the efficiency of the different
options by the range of estimated indirect land-use change impacts (5th
and 95th percentiles) from the Monte Carlo analysis (see figure 6).
With regard to the average estimated indirect land-use change emissions at 38
g/MJ respectively, the 5th and the 95th percentile values
of the distribution are a range of 24 g/MJ to 50 g/MJ[66]. 2.9. The right to act Articles 19(6)
and 7d(6) of the Directives particularly require the Commission to address the
issue of indirect land-use change, as explained in Chapter 1. The overall objective of the Fuel Quality and Renewable Energy
Directives is to contribute to the goal of reducing economy-wide greenhouse gas
emissions through the promotion of renewable energy sources. As a way to
achieve that, they create an EU-wide fuel market and market for renewable
energy, the Member States per se are not able to meet these challenges
individually for the following reasons: ·
the sustainability criteria of the Directives
have their legal basis in Article 114 of the Treaty: internal market. The
indirect land-use change impacts necessarily have transnational aspects which
cannot be dealt with satisfactorily by Member States, when the EU wants to
establish a functional EU-wide market for biofuels. Since there will also be
international trade in biofuels with countries outside the EU, this can also not be properly
regulated at Member State level. ·
it has to be considered whether and how the
objectives could be better achieved by action on the part of the EU: the “test
of European added value”. The rationale for European action in the field of
biofuels has already been decided with the adoption of the Fuel Quality and
Renewable Energy Directives. This stems from the transnational nature of the
identified problem and the desire to create a single market in renewable and
lower greenhouse gas intensity energies for transport. For these
reasons, the policy objectives set out in section 3 of the present Impact
Assessment report cannot be sufficiently achieved by actions of the Member
States alone, but have to be coordinated and harmonised across the EU. 2.9.1. Precautionary principle Article 191(2)
of the Treaty states that EU policy on the environment shall be based on the
precautionary principle. In view of this, the Commission noted in its December
2010 report on indirect land-use change that action should be based on the
precautionary approach. 3. Section:
Policy objectives As explained in
more detail in section 2.1, the Directives contain a number of sustainability
criteria which are aimed at preventing direct land-use change of areas of high
carbon stock and high biodiversity value for the production of biofuels, as
well as setting minimum greenhouse gas emission savings compared to fossil
fuels[67].
Although the current greenhouse gas emissions performance methodology takes
account of emissions associated with direct land-use change, as well as
emissions coming from the production of biofuels, it does not include emissions
from indirect land-use change. In response to
the requirement for the Commission
to review the impact of indirect land-use change on greenhouse gas emissions
and to propose ways to minimise them if appropriate[68], the
Commission published a report[69]
in December 2010 which (i) identified a number of uncertainties and limitations
associated with the available numerical models used to quantify indirect land-use
change; (ii) acknowledged that indirect land-use change can reduce greenhouse
gas emissions savings associated with biofuels; and (iii) indicated that if
action is required, indirect land-use change should be addressed under a
precautionary approach. In view of the above, the fundamental policy objective
for the Commission's continued work on indirect land-use change is to provide a
full response to the request laid down in the Directives[70]: The
Commission shall, by 31 December 2010, submit a report to the European
Parliament and to the Council reviewing the impact of indirect land-use change
on greenhouse gas emissions and addressing ways to minimise that impact. The
report shall, if appropriate, be accompanied, by a proposal, based on the best
available scientific evidence, containing a concrete methodology for emissions
from carbon stock changes caused by indirect land-use changes, ensuring
compliance with this Directive, in particular Article 17(2). Such a
proposal shall include the necessary safeguards to provide certainty for
investment undertaken before that methodology is applied. With respect to
installations that produced biofuels before the end of 2013, the application of
the measures referred to in the first subparagraph shall not, until 31 December
2017, lead to biofuels produced by those installations being deemed to have
failed to comply with the sustainability requirements of this Directive if they
would otherwise have done so, provided that those biofuels achieve a greenhouse
gas emission saving of at least 45 %. This shall apply to the capacities of the
installations of biofuels at the end of 2012. The Articles
and the corresponding recitals[71]
allude to the development of a calculation methodology for capturing indirect
land-use change emissions. However, due to the right of initiative, the
Commission wishes to consider other options which the Commission believes can
also meet the policy objectives in a suitable way. The Commission evaluates
four different options in this Impact Assessment, as laid down in chapter 4;
"Policy Options". To enable the
assessment of the options, it is desirable to understand the general policy
objectives and narrow these down to a description of more specific, operational
objectives. 3.1. Treaty based general objectives The EU's
policies on the promotion of the use of biofuels have always been developed in
the context of EU energy policy and EU policy aimed at protection of the
environment. The development of new and renewable
forms of energy is specifically foreseen as an objective in the Treaty and it
is clear that this goal is pursued with regard to the need to enhance security
of energy supply and that of preserving and improving the environment. The
Treaty foresees that environmental protection must be built into all policy
areas and action to reduce climate change is specifically foreseen within the
Treaty as an environmental objective. The provisions on sustainability criteria, including the
requirement to analyse indirect land-use change emissions, are based on the
functioning of the internal market provisions of the Treaty. Any legislative
proposal that addresses indirect land-use change emissions must therefore also
be based on these provisions. 3.2. General objectives The general objectives are those of the
Directives. In context of the Renewable Energy Directive, Recital 65 summarises
the general environmental objective related to the use of biofuels: Biofuel production should be sustainable.
Biofuels used for compliance with the targets laid down in this Directive, and
those that benefit from national support schemes, should therefore be required
to fulfil sustainability criteria. The content of Recital 65 is reflected in
Article 17 of the Directive, which requires biofuels to be sustainable, and in
particular Article 17(2) thereof, which in context of greenhouse gas savings
requires biofuels to save at least 35% compared to fossil fuels, increasing to
50% in 2017 and 60% in 2018 for new installations. The objectives of the Fuel Quality
Directive in relation to the use of biofuels are reflected in Recitals 9 and
10: Biofuel production should be sustainable.
Biofuels used for compliance with the greenhouse gas reduction targets laid
down in this Directive should therefore be required to fulfil sustainability
criteria.(…). Suppliers should, by 31 December 2020,
gradually reduce life cycle greenhouse gas emissions by up to 10 % per unit of
energy from fuel and energy supplied. This reduction should amount to at least
6 % by 31 December 2020, compared to the EU-average level of life cycle
greenhouse gas emissions per unit of energy from fossil fuels in 2010, obtained
through the use of biofuels, alternative fuels and reductions in flaring and
venting at production sites.(…). Recital 4 is reflected in Article 7b, which
sets identical sustainability requirements as Article 17 of the Renewable
Energy Directive. Recital 10 is reflected in article 7a which states that
Member States shall require fuel suppliers to reduce as gradually as possible
life cycle[72]
greenhouse gas emissions per unit of energy from fuel and energy supplied by up
to 10 % by 31 December 2020, compared with the fossil fuel baseline representing
2010. 3.3. Specific and operational objectives As described in section 2.1, this impact
assessment is focused on the specific requirement in the Directives related to
greenhouse gas emissions from indirect land-use change and does not consider any
wider environmental and social impacts associated with the use of biofuels. As
such, the general objectives presented above translate into the following
specific/operational objective to: Minimise the impact of indirect land-use
change on greenhouse gas emissions of biofuels, within the wider policy
objectives of the targets that by 2020 at least 10% of transport fuels are
renewable and that greenhouse gas intensity in road transport fuels is reduced
by at least 6% compared to 2010. The policy options will be evaluated in
context of the extent to which the options fulfil this specific objective. 4. Section
- Policy options 4.1. What are the possible options for achieving the policy
objectives? Further to the
option referred to in the Directives, the development of a concrete methodology
for emissions from carbon stock changes caused by indirect land-use change from
biofuels that could be included in the greenhouse gas calculation, the
Commission wishes to consider the effectiveness of a number of options aimed at
minimising indirect land-use change impacts. The policy options, including
those initially set out in the Commission's report on this topic adopted in
December 2010, are: A. take no action for the time
being, while continuing to monitor, B. increase the minimum greenhouse
gas saving threshold for biofuels, C. introduce additional
sustainability requirements on certain categories of biofuels, D. attribute a quantity of
greenhouse gas emissions to biofuels reflecting the estimated indirect land-use
impact. E. limit the contribution from
conventional biofuels to the Renewable Energy Directive targets to current
production levels. 4.2. Option A – take no action for the time being; while
continuing to monitor This option refers to the Commission's bi-annual
monitoring and reporting of impacts, including indirect land-use change, as
required by the Renewable Energy Directive[73],
the first of which is due in 2012. The option also implies continued monitoring
of the scientific developments related to estimating indirect land-use change
emissions. During the latest consultation on policy
options, option A was preferred by those stakeholders who believed that the
current state of development of the models was not appropriate to base policy
approaches upon. This included most of the industry, farmers' associations and
biofuel producing third countries. 4.3. Option B - increase the minimum greenhouse gas saving
threshold for biofuels This option consists of increasing the
current minimum greenhouse gas savings thresholds provided in the Directives[74], currently at 35% compared to
average fossil fuels. According to the Directives, this requirement is
increased to 50% in 2017, and 60% in 2018 for installations that started
production in 2017. Option B aims at a) compensating for
the estimated indirect land-use change emissions through requiring higher direct
savings, and thereby improving the overall greenhouse gas performance of the
biofuels consumed and; b) reducing indirect land-use change emissions
through raising the threshold to such a level that many of the biofuels with
estimated large indirect land-use change emissions are excluded. The increased level of the threshold should
be achievable for a set of feedstocks, while at the same time maximizing the
direct greenhouse gas savings. The level must be technically feasible for a
range of feedstocks, but require improvements beyond what is required by the
Directives today. A discussion of the exact level of the threshold is set out
in chapter 5. This option implies changing Article 17 of
the Renewable Energy Directive and Article 7b of the Fuel Quality Directive. Option B was not supported by any
particular stakeholder group during the last consultation exercise, as stakeholders
generally favoured option A or option D. 4.4. Option C - introduce additional sustainability
requirements on certain categories of biofuels This policy option consists of introducing
additional sustainability requirements aimed at mitigating the risk of indirect
land-use change emissions. As such, compliance with a number of additional
criteria could be required at national (country) level or at project/farm level
as detailed in sub-options C1 and C2 below. These options are very different in their
approach and are therefore, being treated as two separate sub-options. Both
these options require changing Article 17 of the
Renewable Energy Directive and Article 7b of the Fuel Quality Directive. 4.4.1. Option C1- country level
actions Given that indirect land-use change
emission impacts associated with biofuels manifest themselves through unwanted
direct land-use change across countries, and often as a result of inadequate
land governance, option C1 is aimed at addressing such negative effects by
improving land-use governance and protection of high carbon stock lands. More
specifically, under option C1 biofuel producing countries including Member
States, are requested to implement LULUCF methodology based (see chapter 2.5.3
for description) reporting (if not already in place) and protection of high
carbon stock land. Simultaneously, efforts to increase the supply of biofuels
with low risk of indirect land-use change emissions at a national level could
be implemented. To guarantee that option C1 reduces
indirect land-use change emissions completely the option would need to be
implemented globally, thus preventing any leakage. However, indirect land-use
change emissions could be reduced and even over-compensated if the
implementation of option C1 is successful, as it may reduce indirect land-use change
emissions from other commodities. An example of this is provided below,
where peatland emissions in Indonesia are halved. Stopping conversion of peatlands in Indonesia would have a considerable impact on the estimated indirect land-use change emissions
from all oil crops, as their production is assumed by the models to cause
indirect conversion of peatland (i.e. more than the third of the estimated
indirect land-use change emissions are from peatland). Moreover, the protection
of peatland would also limit the land-use change emissions from other
agricultural commodities (both directly and indirectly), as demand for biofuels
is only one of many drivers for increased production of vegetable oils
globally. Wetlands International[75]
reports that emissions from peatlands in Indonesia alone in 2008 amounted to
500 Mt. This implies that one year of emissions from peatland in Indonesia is ten times higher than the estimated annual indirect land-use change emissions
in 2020. However, the overall effectiveness of this option relies on whether
countries choose to comply or not, which is outside the control of the EU and
therefore it is not possible to guarantee that the results will develop in a
certain way. This is further analysed in chapter 5. During the latest consultation on policy
options, most of the industry and farmers' associations supported the use of
international action to address indirect land-use change emissions, although
not necessarily in the same terms as outlined in this sub-option. 4.4.2. Option C2- project level
actions This sub-option refers to practices that
could prevent indirect land-use change by producing feedstocks without the need
for additional land. Potential mitigation measures would include: (a)
Using land without provisioning services[76] that would be unlikely to be
taken into production in the absence of biofuel demand (i.e. typically land
that either requires some form of remediation prior to being used or where
significant barriers exist). Expanding production on unused land may lead to
direct land-use change, but the latter would be addressed by the current
sustainability criteria and therefore directed to those areas where effects are
acceptable. (b)
Increasing yields above projected future trends
which would not have happened in the absence of biofuel demand. This could lead
to the production of biofuel feedstocks without increasing the pressure on land
and therefore limiting indirect land-use change emissions. In this case, only
the additional feedstock production should be considered as meeting this
requirement. (c)
Integrating biofuels and non-bioenergy
production systems (i.e. land-used for cattle farming) in ways that lead to a
higher overall land productivity. Again this integration would be additional to
what would have happened in the absence of the biofuel demand. In order for indirect land-use change to be
prevented, it would also be necessary to ensure that the actions put in place
are additional and that would not have been implemented in the absence of these
criteria. Both the reporting and verification processes could be foreseen to be
incorporated into the existing reporting and verification systems established
under the sustainability scheme, which already requires the collection of
detailed information across the supply chain by economic operators or managers
of sustainability schemes on land-use and cultivation practices in order to
demonstrate compliance[77].
In addition, the process for verification would require that an assessment at
project level is carried out before the project is approved, although it would
seem reasonable to assume that additionality would only need to be
proven once within the legislative period. The literature[78] indicates that the potential
for producing biofuels in this way is significant, and that the share of non-EU crops would increase if
this option was to be implemented (i.e. sugarcane, soy and palm oil) as
opportunities for intensification seem to be greater in certain world regions
(i.e. Brazil, South East Asia). The full details for implementing these
criteria would need to be developed at a later date. A more detailed
description of these measures as well as a summary of their potential can be
found in Annex XII. Most NGOs supported sub-option C2 during the last consultation in
combination with option D. The rationale behind this approach is that biofuels
produced under the conditions outlined above, would in theory mitigate the risk
that their production would trigger indirect land-use change. Option D was by
these stakeholders believed to be a strong incentive for the actions described
under sub-option C2 to be implemented, which in some cases would require
significant changes in current production practices. 4.5. Option D - attribute a quantity of greenhouse gas
emissions to biofuels reflecting the estimated indirect land-use impact. This is the option referred to in the
Directives, which would require incorporating estimated indirect land-use
change emissions values in the reporting of existing greenhouse gas methodology[79],[80], including ensuring compliance
with minimum greenhouse gas thresholds[81].
A new Annex VIII would need to be added to the Renewable Energy Directive and a
new Annex V to the Fuel Quality Directive. This option implies incorporating the
estimated indirect land-use change emissions of biofuels into the emission
calculation. However, there are a number of issues that need to be described,
as the use of a factor may not be appropriate in some circumstances (i.e.
non-land using feedstocks such as algae). These various elements are set out below: ·
Evaluate the introduction of different estimates
of factors (eiluc) into the greenhouse gas calculation
representing the estimated indirect land-use change greenhouse gas emissions
and taking into account the results of the Monte Carlo analysis from the
IFPRI-MIRAGE-BioF. ·
Define the situations where eiluc
should have a value of zero because there is no displacement of agricultural
activity. This is likely to be appropriate for the following circumstances, i.) Waste and residual materials are used
as feedstocks. ii.) The feedstock does not require land
for its production (i.e. algae). iii.) The production of the feedstock has
led to 'direct land-use change', and as such an el value has been calculated in
accordance to greenhouse gas emissions methodology. When assessing this option, consideration
will be given at setting the indirect land-use change emission factors eiluc
at levels which mitigate/anticipate the risk of a possible model i)
overestimation (50%, 25%, 5% percentiles) and ii) underestimation (50%, 75%,
95% percentiles) of crop-specific indirect land-use change factors. In
addition, consideration will be given to the degree of disaggregation these
factors should be set at (i.e. feedstock specific or crop group level i.e. oil
crops, sugars and cereals). As with the current methodology for default
greenhouse gas emissions values, it would be required to consider whether a
review process[82]
needs to be established within the Directives for updating these values.
Consideration will be given as to whether safeguards for investment from the
introduction of this particular option, as stated in the Directives[83], should be introduced. Most NGOs and a few industrial stakeholders
from the non-biofuel sectors supported this option during the last consultation.
This was also the most supported option during the international scientific
expert workshop with academics and experts organised by the JRC in November
2010. 4.6. Option E - Limit the contribution from conventional
biofuels to the Renewable Energy Directive targets. This option aims at minimising the indirect land use change impacts
of biofuels by limiting the amount of conventional biofuels that can be counted
towards the Renewable Energy Directive targets to current production levels.
The risk of indirect land use change is mostly associated with conventional
biofuel feedstocks grown on high yielding agricultural land. To limit the
consumption of such biofuels therefore also limits the risk of indirect
land-use change emissions. Moreover, such a limit will require that the
remaining amount of biofuels needed to achieve the 10% Renewable Energy
Directive transport target would need to come from advanced biofuels with lower
indirect land use change risks, which will significantly improve the biofuel
mix that can be expected in 2020. This option implies changing Article 3 of
the Renewable Energy Directive. Although Option E was not included as one
of the shortlisted options by the Commission in the last consultation exercises,
options aimed at limiting the amount of conventional biofuels while increasing
the incentives for advanced biofuels were favoured by NGOs and certain
industrial stakeholders. 5. Analysis
of impacts 5.1. Assessment methodology 5.1.1. Introduction The baseline estimated indirect land-use
change greenhouse gas emissions impacts are outlined in chapter 2.
Consequently, the different policy options assessed here focus on potential
ways of minimising these impacts. The assessment focuses on the impacts
resulting from the potential changes in feedstock availability[84] following intervention. The assessment of the policy options has
been carried out according to their effectiveness, in achieving the
policy objectives in chapter 3[85]
and their likely wider environmental, economic[86]
and social impacts, as well as their consistency with other EU policies.
Further detail can be found in Annex XIII. The particular question of administrative
burden and associated costs is addressed to the extent possible for options B
and C2. In the case of option B, this is because it is not possible to know
precisely to what extent certain feedstocks will be excluded (in which case
there would be economic impacts but no administrative costs) or the extent to
which economic operators would need to report actual greenhouse gas emissions
values to meet increased requirements. With regard to option C2, the potential
certification system is still being developed and therefore some costs from
current pilots are presented as an indication only. Option C1 has administrative
costs related to the verification of LULUCF reporting. All other options are
considered to have insignificant additional administrative costs because they
are all based on the current framework for verifying compliance with existing
sustainability criteria under the Directives[87]. 5.1.2. Development of scenarios In most cases, it is expected that options
will limit the availability of qualifying biodiesel feedstocks as these
typically present both higher direct and estimated indirect land-use change emissions.
In order to compensate for the reduced availability of biodiesel and so to
still achieve the targets of the Directives, the contribution from all the
available Renewable Energy transport technologies (i.e. bioethanol, advanced
biodiesel and electric cars) is increased by similar amounts to maintain the
5.4% reductions in greenhouse gas intensity towards the Fuel Quality Directive
found in the baseline. It should be stressed that this is a purely theoretical
exercise which does not necessarily lead to realistic results. Alternatively,
but not assessed here, a higher contribution from the use of LPG and venting
and flaring reductions could contribute towards the carbon intensity reduction
targets in the Fuel Quality Directive[88]. With regard to the final biofuel mix, key
factors are expected to include the ability of the biofuel industry to improve
the greenhouse gas emissions performance of pathways associated with
conventional feedstocks (i.e. particularly oil crops) in order to comply with
higher greenhouse gas emissions standards, and the ability of technologies
using alternative feedstocks (i.e. non-land using feedstocks) to come to market
within given timescales. In those cases where the availability of conventional
biodiesel feedstocks is severely restricted, a number of car fleet
compatibility issues associated with higher bioethanol blends or higher shares
of certain feedstocks, such as palm oil, may arise. Further information on the
current situation can be found in Annex XIV. 5.1.3. Assessment limitations In order to appropriately interpret this
assessment, the following limitations/simplifications should be born in mind; ·
in reality, the indirect land-use change
emissions could be higher or lower. As discussed in chapter 2 and particularly
Annex III, there are still, and will possibly always remain, considerable
limitations and uncertainties related to estimating indirect land-use change
emissions. However, given the need to analyse the issue in a quantitative
manner, it has been necessary to use the latest results from IFPRI-MIRAGE-BioF
as the baseline. ·
the modelling of indirect land-use change
emissions assumes that the sustainability scheme has no effect. Still, in this
assessment it is assumed that it actually reduces both direct emissions and
indirect land-use change emissions through a reduction in consumption in the EU
of biofuels from a particular feedstock, according to the estimates made by
IFPRI-MIRAGE-BioF. ·
although measures considered here for biofuels
are also applicable to bioliquids, no calculations have been specifically made
using their fossil fuel comparators[89].
This is not likely to give rise to any leakage issues (i.e. where excluded
biofuels may be diverted to bioliquid market), as the assumed fossil fuel comparator
value for 2020 here (90.3 g/MJ) is very close to the highest value provided for
bioliquids. ·
the question of whether a particular feedstock
passes or not is a simple yes or no question. However, there are several
conditions that determine the greenhouse gas emissions performance associated
with different feedstocks, including potential improvements in greenhouse gas
emissions performance, and hence whether they can ultimately meet the minimum
thresholds set out in the Directives. Moreover, the potentially increased attractiveness
of e.g. the severely degraded/heavily contaminated land greenhouse gas bonus[90] is not assessed. ·
the analysis is based on the year 2020 (i.e. for
analytical purposes, as all the assumptions made are related to that year). All
indirect land-use change emissions are assumed to be mitigated when action is
taken in 2020. This is an oversimplification, as indirect land-use change
impacts are of cumulative nature and irreversible in the period to 2020[91]. Moreover, it should be noted
that although the assessment is done for only one year, the indirect land-use
change emissions included in the greenhouse gas emissions balance of biofuels
are averaged over a 20 year period in accordance with the existing methodology
in the Directives. ·
some of the options result in biofuels made from
certain feedstocks being unable to meet the sustainability criteria of the
Directives. The deficit that this might lead to would need to be covered by
other feedstocks or means to comply with the targets. A range of factors, such
as costs, technical blending possibilities, technical vehicle specifications
and infrastructure developments influence how the deficit can or cannot be
covered. None of these factors have been assessed in detail in this Impact
Assessment. Given all these uncertainties, it is
important to bear in mind that the results presented in this section should be
used with the utmost caution. 5.2. Option A - Take no action for the time being, while
continuing to monitor Option A is the "no policy
change" option, which implies that no further action to mitigate indirect
land-use change emissions is taken, while continuing to monitor the development
of key factors determining the indirect land-use change impact and the science
needed to assess the scale and nature of the phenomenon. 5.2.1. Effectiveness in reducing
greenhouse gas emissions The main
results in terms of reducing greenhouse gas emissions are shown in the table
below. Direct emissions only- biofuels only || Average emissions 2020 (g/MJ) || 35 Average emission savings 2020 [%] || 61% Total direct emissions 2020 (Mt) || 42 Indirect land-use change emissions- biofuels only || Average ILUC emissions 2020 [g/MJ] || 42 Total ILUC emissions 2020 [Mt] || 48 Total emissions- direct and indirect- biofuels only || Average emissions 2020 [g/MJ] || 77 Average savings 2020 [%] || 15% Change against baseline || Indirect land-use change emissions (Mt) || 0 Total emissions (Mt) || 0 Table 5: Effectiveness analysis of option
A Both direct and indirect land-use change
emissions are as described in the baseline in section 2.8, as no action is
taken. Estimated indirect
land-use change emissions are not reported, but even if they were to be
included in the methodology, the average reported savings for biofuels compared
to fossil fuels would be at 15%[92].
Similarly, with regard to the Fuel Quality Directive target, the total emission
savings offered by biofuels would be at 1.1%. 5.2.2. Impacts on achieving the
Renewable Energy Directive transport target The fulfilment of the 10% renewable energy
in transport target can be met, as envisaged in the NREAPs, as neither the
emission reduction thresholds nor the GHG methodology are changed. 5.2.3. Economic impacts Option A continues the existing policy
framework, leaving the sustainability scheme as currently laid down in the
Directives unchanged. The biofuel and related industries, as well as farmers
are thus ensured a stable policy framework, which enables investor certainty.
This also implies that Member States and industry can continue to follow the
submitted National Renewable Action Plans[93]
(NREAPs), and the EU installed biofuel production capacity can be utilised. Security
of supply is maintained as foreseen, with 10% of the energy used in the
transport sector being renewable, from diversified energy sources, with about
half of the total demand being met by domestic production and half by imported.
Option A
has no specific implications for trade policies and trade relations. Imports of
biofuels and biofuel feedstocks, particularly biodiesel, are expected to
increase significantly. This option has no additional implications for
technological development and innovation as both developments in greenhouse gas
emissions performance and advanced biofuel production remain as planned. No
additional impacts on car manufacturers with regard to the levels of biofuel
blending required other than those described in the baseline. 5.2.4. Social impacts EU rural development and employment
opportunities are not affected under this option, as activities in the
agricultural and industrial sectors associated with the production of biofuels
in the EU are not affected and continue to develop as planned. An estimated 400
000 jobs could thus be maintained in the sector. The expected increase in demand for
biodiesel and bioethanol to 2020 will increase the pressure on global commodity
markets, particularly for vegetable oil as the EU demand for biodiesel
represents a more significant share of the total production. As biofuel demand
is inelastic, this increased demand could impact on certain manufacturers of
food, cosmetics and daily care products who rely on the same raw materials,
particularly vegetable oils. Development objectives in third countries
are difficult to assess, as such impacts are dependent on local factors.
However, the current framework, which is continued under option A allows for a
range of crops typically grown in developing countries to be supplied to the
EU, as they typically fulfil the sustainability criteria. 5.2.5. Environmental impacts According to the models, the
estimated additional cropland requirements globally amount to 1.7 Mha, mainly
in regions of the Commonwealth of Independent States, Sub-Saharan Africa and Brazil, and would be taken from pasture (42%), managed forest
(39%), primary forest (3%) and savannah and grassland (16%). The JRC analysis
of the land cover classification used by IFPRI-MIRAGE-BioF, suggests that the largest biodiversity losses would be associated
with the conversion of primary forest, and savannah and grassland, pasture land
and managed forest to cropland in order of decreasing biodiversity value[94]. Although it is
not possible to provide any specific information on wider environmental impacts
based on the model results, adverse water, soil and air impacts would be
expected from the conversion of primary and managed forests, as well as the
conversion of grassland. 5.2.6. Other impacts As emission savings offered by biofuels
would not reflect indirect land-use change, biofuels would contribute less to the
integrated approach for CO2 in cars[95]. 5.3. Option B - Increase the minimum greenhouse gas saving
threshold for biofuels Option B follows the principles of the
already existing sustainability criteria which contain thresholds for minimum
greenhouse gas savings from biofuels compared to fossil fuels. To raise the
threshold further would exclude certain feedstocks with higher direct
emissions. The table below illustrates how increasing
the threshold affects compliance for various feedstocks. As shown in table 4 in
chapter 2.8, vegetable oils have high estimated indirect land-use change
emissions; these are highlighted in red colour and bold letters in the table
below. || Feedstock || Direct greenhouse gas emissions (2020) || Allowed with 50% threshold || Allowed with 55% threshold || Allowed with 60% threshold || Allowed with 65% threshold || Allowed with 70% threshold || || (g/MJ) || 45,4 || 40,8 || 36,3 || 31,7 || 27,2 Biodiesel || Palm oil || 51,1 || || || || || Soybean || 46,9 || || || || || Rapeseed || 40,2 || || || || || Sunflower || 32,4 || || || || || Palm oil with methane capture || 29,3 || || || || || 2G biodiesel - non-land using || 9,3 || || || || || 2G biodiesel - land using || 5,4 || || || || || Bioethanol || Wheat - Process fuel not specified || 50,3 || || || || || Wheat - Natural gas as process fuel in CHP plant || 33,5 || || || || || Corn (maize) || 32,7 || || || || || Sugar beet || 27,1 || || || || || Sugar cane || 20,2 || || || || || Wheat - Straw as process fuel in CHP plant || 21,6 || || || || || 2G bioethanol - land using || 16,7 || || || || || 2G bioethanol - non-land using || 9,0 || || || || || Table 6: Direct emissions are shown in the
first column, the allowed direct emissions as a function of the threshold is
shown in the second row (in bold). The colouring is indicating whether the
feedstock named in the left column achieves the threshold indicated in first
row. Grey colour indicates that the feedstock is not reaching the threshold
(however, the feedstock could be in compliance by improving performance by less
than 5 g/MJ), while red/dark grey colour means that the feedstock is below the
threshold by more than 5 g/MJ. 5.3.1. Level of threshold It should be noted that these figures are
estimates for 2020, and represent typical values, and not default values[96]. From the table it can be
observed that certain types of wheat bioethanol, palm oil biodiesel and soy
bean biodiesel have difficulties achieving 50% savings. No further changes are
occurring when the threshold is raised from 50% to 55%, but rapeseed[97] is likely to be excluded when
the threshold reaches 60%. At 65% corn bioethanol, wheat bioethanol produced
with natural gas and sunflower biodiesel are unable to achieve the threshold,
where only the latter has estimated high indirect land-use change emissions. At 70%, all vegetable oils are below the
threshold of required savings, while only bioethanol from sugar cane and
efficient wheat can be produced with sufficiently savings to be above the
threshold. A 70% threshold thus appears to be too ambitious, if a certain
variety of feedstocks for both bioethanol and biodiesel should be available,
whereas 60% savings could be achieved with several feedstocks, especially if
improvements and more efficient production techniques are implemented. To increase the threshold to 60% lowers the
direct emissions by almost 10 g/MJ, thus compensating for a certain amount of
indirect land-use change emissions. But more importantly it is expected to
exclude palm oil without methane capture, rapeseed oil and soybean oil from
being used, all of which the IFPRI-MIRAGE-BioF model estimate to have large
indirect land-use change impacts. To ensure a smooth transition to higher
thresholds, option B entails that the required minimum savings are raised
already in 2013, to 60% up from 35% today[98].
It should be noted that in accordance with
the intention of the Directives to safeguard investments[99] and allow time for industry to
adjust, biofuels produced at plants that were in operation by the end of 2012
would only need to comply with a threshold of 45% until the end of 2017, at
which point the 60% threshold would apply also to those installations. The
proposed level of threshold increase would exclude certain feedstocks, which
will need to be substituted. For the purpose of the analysis of impacts certain
assumptions as to how this could take place need to be made. The following
section explains this in more detail. Again, it should be noted that the
analysis has several shortcomings as explained in chapter 5.1. 5.3.2. Potential scenario –
meeting the targets of the Directives A range of possible scenarios for how to
achieve the targets under option B, can be developed. Essentially they imply
increasing the use of bioethanol, advanced biodiesel or electric vehicles. A
set of theoretical and extreme scenarios exploring the use of these options are
found in Annex XIV. The least unrealistic based on contributions from other
technologies (electric vehicles) is set out below. || Bioethanol [Mtoe] || Double counted biodiesel [Mtoe] || Electricity in road [Mtoe] Baseline || 6.7 || 1.8 || 2.1 Option B || 13.6 || 3.8 || 2.7 This would require both more (a doubling
of) bioethanol blended into the vehicle fleet, and more than doubling the
amount of 'double-counted' biodiesel produced (both are doubled compared to the
baseline – which means twice the amount of what is estimated in the NREAPs). The increased consumption of ethanol is a
question of blending possibilities, as the share of petrol consumption in the
EU is decreasing, and expected to continue to decline. It implies that the
average bioethanol blends in petrol cars would be needed to increase from 11%
to 22%. The JEC Analysis of scenarios for transport in the EU towards 2020[100], finds that in a scenario
with an E85 grade (flexi-vehicles) introduced now, and E20 in 2017, the EU
would consume 8-9 Mtoe of ethanol. If E20 is introduced in 2015 rather than in
2017 (2 years earlier), it leads to additional 0.7 Mtoe of ethanol being
consumed in 2020. This indicates how challenging the consumption of 13.6 Mtoe
of ethanol would be. The availability of double-counted
biodiesel is a question of supply, both in terms of availability of
raw-material e.g. waste oil, but also a technical question whether enough
production capacity can be cost-efficiently installed by 2020. Achieving a
supply of 3.8 Mtoe of double counted biodiesel would therefore be challenging. The scenario also needs 0.6 Mtoe of additional
electricity in road transport by 2020. This would be equivalent to deploying an
additional 2.3 million electric cars[101]
by 2020 on top of what is foreseen in the NREAPS. The NREAPs estimate that
around 2.1 Mtoe of electricity[102]
is consumed in electric vehicles in 2020, which implies around 8 million
electric cars. It should be noted that 2.1 Mtoe includes plug-in hybrids, as
well as other road vehicles using electricity. Again, this means that this
scenario is not very likely and these figures are included here for
illustrative purposes only. It is also possible that the increased
threshold would lead to additional contributions from other available
technologies in meeting the Fuel Quality Directive target (i.e. flaring and
venting reductions). An additional 0.5 percentage points reduction through such
means, would reduce the contribution illustrated in the table above by 0.7 Mtoe
of electricity in road transport (back to baseline levels) or with almost 2
Mtoe of double counting biodiesel. The shortfall could also be met by
increased use of sunflower and palm oil with methane capture, as both meet the
threshold at 60%. However, there are technical constraints as to how much palm
oil can be used, due to high CFPP (cold filter plugging point) properties,
which makes it inappropriate in colder climates. While this is true for FAME
and pure vegetable oils, palm oil that has been hydro-treated into HVO
(hydro-treated vegetable oil), does not give rise to issues with CFPP in colder
climates, but such fuels are also expected to be in demand from the aviation
sector, as it is one of the few available biofuels that jet engines can use.
Having regard to the expected capacity of HVO plants and the assumed amount of
palm oil with methane capture in the baseline, it is assumed that 2.3 Mtoe of
palm oil with methane capture is used in the EU in 2020. Although there are no
technical constraints associated with blending in more sunflower oil, its
volume is assumed not to increase and is maintained at baseline levels of 1.1
Mtoe. A total of 3.4 Mtoe of oil crops derived biodiesel is therefore consumed
under option B. Alternative levels of consumption of oil crops derived
biodiesel is analysed in the sensitivity section. 5.3.3. Effectiveness in reducing
greenhouse gas emissions Increasing the
threshold to 60% implies that overall emissions are reduced by 56 Mt compared
to the baseline. 33 Mt (70%) of the estimated indirect land-use change
emissions occurring in the baseline would be avoided in 2020. The biofuels used
are on average saving 56% compared to fossil fuels when indirect land-use
change emissions are considered. The emissions reported towards the Fuel
Quality Directive target will not reflect estimated indirect land-use change
emissions. The main results are shown in the table
below. Direct emissions- biofuels || Direct emissions 2020 (g/MJ) || 22 Direct emission savings 2020 [%] || 76% Total direct emissions 2020 (Mt) || 19 Indirect land-use change emissions – biofuels || Average ILUC emissions 2020 [g/MJ] || 18 Total ILUC emissions 2020 [Mt] || 14 Total emissions- direct and indirect- biofuels || Average emissions 2020 [g/MJ] || 40 Overall savings 2020 [%] || 56% Change against baseline || Indirect land-use change emissions (Mt) || -33 Total emissions (Mt) || -56 Table 7: Effectiveness analysis of option
B It should be noted that if rapeseed was to
improve performance sufficiently, or the deficit of rapeseed was to be filled
with other oil crops that pass the increased threshold (sunflower and palm oil
with methane capture), the reductions in estimated indirect land-use change
emissions would be reduced accordingly. This analysed in the sensitivity
section of option B. 5.3.4. Impacts on achieving the
Renewable Energy Directive transport target There is a risk
that the transport target of the Renewable Energy Directive is not achieved, if
industry does not manage to produce sufficient amounts of biofuels with at
least 60% direct greenhouse gas savings or the technological developments
needed to achieve required increased bioethanol blends and electric vehicles do
not take place. 5.3.5. Economic impacts The change in
cost compared to the baseline is expected to be moderate as a range of
feedstocks is still available. The increased use of electricity in road
transport, and 2nd generation biofuels will increase aggregate
costs, depending on how the costs of these technologies develop. Financial investment stability is affected,
as the use of conventional biodiesel feedstocks would be significantly reduced,
as well as less efficient bioethanol production pathways. This would have
significant implications for the existing EU biofuel industry that is not able
to increase its efficiency. In that case, considerable stranded investment
would result. It also implies that Member States and industry cannot continue
to follow the submitted National Renewable Action Plans (NREAPs), which may
have political implications. The raising of the threshold to 60% would
thus require industrial adjustment, and those countries with the largest
biodiesel installed capacity (i.e. Germany, Spain, France, Italy and the Netherlands) would be most affected. Moreover, activity will be reduced in related
industries such as those involved in the production of vegetable oils/crushing
of oil crops for all food/feed/biofuels markets (mainly being present in Germany, France, Spain, the Netherlands and the UK). The increased threshold is likely to
require more producers to report actual values, rather than default values, as
the default values are below the threshold[103].
This is more burdensome, as more data needs to be submitted and verified. In
the Impact Assessment on biomass sustainability[104] it was estimated that the
increased cost of using actual values rather then default values would be 10-20
% higher, i.e. not substantially more than in the baseline, particularly since
only a share of the market will need to use actual values. Impacts on
security of supply can be adversely affected if the necessary bioethanol blend
levels are not available in terms of specification of the vehicle fleet, or the
required high-saving biodiesel volumes are not supplied. Option B would also have a number of trade impacts. Imports of
conventional biodiesel feedstocks into the EU would be limited, while the
feedstocks and biofuels with higher savings are expected to increase. 5.3.6. Social impacts With regard to EU employment, the resulting
impacts would depend on whether the opportunities created through the increase
in the bioethanol and advanced biodiesel industry are able to make up for the
reduction in activity in the conventional biodiesel industry. The foreseen
grandfathering of installed capacity would be expected to help with the
transition. While rapeseed is likely to be excluded,
the adverse employment effects within the EU for famers are likely to be
limited, as farmers would respond to the shift in demand from rapeseed to
cereals. Rural development is dependent on the same set of variables, and thus
difficult to assess. The reduction in vegetable oil demand will
lower the pressure on global vegetable oil markets. Conversely, pressure on
coarse grains and sugar prices is likely to increase, although the impacts are
expected to be moderate as the demand for biofuels of these commodity groups
represents a lower share of the global markets than for oil crops. 5.3.7. Environmental impacts Adverse
biodiversity impacts are reduced as it is expected that reduced indirect
land-use change emissions are correlated with reduced conversion of bio-diverse
areas. Certain crops such as sunflower and palm oil with methane capture which
have associated high estimated indirect land-use change impacts would continue
to qualify, but the quantities are assumed to be limited. This option would
also contribute towards avoiding other environmental impacts associated with
land conversion (i.e. adverse water, soil and air impacts). 5.3.8. Other impacts As option B
follows the principles of the already existing sustainability criteria which already
contain thresholds for minimum greenhouse gas savings, the option is simple in
design and implementation. This option
would be expected to motivate technological development. Firstly, it would
provide incentives for accelerating the introduction of advanced biodiesel.
Moreover, the increased greenhouse gas emissions requirements should encourage
improvements in the performance of 1st generation feedstocks. This option
does not change the existing calculation methodology and thus it is not likely
to be challenged by the WTO. Moreover, this option does not depend on modelling
for the design of the policy measure. However, it does not distinguish between
feedstocks according to their estimated indirect land-use change impacts. The greenhouse
gas performance of the different biofuel pathways is only dependent on the
actions taken by the biofuel producers themselves. As such the option B follows
the existing methodology. As emission savings offered by biofuels
would not reflect indirect land-use change, biofuels would contribute less to the
integrated approach for CO2 in cars. As such, larger contributions
towards achieving greenhouse gas emissions savings will be needed from energy
efficiency and other available technologies. 5.3.9. Sensitivity of option B 5.3.9.1. Uncertainty related to the
use of oil crops under option B Three uncertainties are dominant regarding
the effectiveness of option B. Firstly, the
direct emissions from the various feedstocks might change due to technological
progress beyond what is assumed in this impact assessment, leading to changes
in which feedstocks are used and thus changes in the estimated indirect
land-use change emissions. This is particularly important aspect for rapeseed,
which can comply if sufficient improvements in performance are made. Rapeseed
has assumed direct emissions of around 41 g/MJ, while the threshold of 60% is
allowing maximum 36 g/MJ, i.e. 5 g/MJ lower. In the production of rapeseed,
cultivation is particularly important, although around 5 g/MJ can be saved by
using bio-methanol instead of fossil derived methanol for the processing of
vegetable oil into FAME. Reductions of up to 10-15 g/MJ compared to the 29 g/MJ
of cultivation emissions reported under the default values in the Directives
can be found in the literature[105].
In total is would therefore be theoretically possible to reduce emissions by
around 20 g/MJ. Secondly, there is uncertainty related to
the replacement feedstocks of rapeseed. Substitution could take place, using
more oil crops based biodiesel (rapeseed with improved efficiency and/or
sunflower and palm oil with methane capture which would still be expected to
pass the threshold) than what is assumed in B. Assuming that e.g. all rapeseed
passes the increased threshold, the effectiveness of option B would be reduced;
with overall emission savings being halved, from 56% to 28%. Moreover, the
annual estimated indirect land-use change emissions in 2020 would be reduced by
11Mt (23%) instead of 33Mt (69%). The uncertainty related to these aspects
could be reduced by further increasing the threshold. While estimated indirect land-use change
emissions would increase significantly, as well as the adverse environmental
impacts such as biodiversity loss associated with them. However, the negative
effect on European biodiesel industry would be considerably limited. The
targets under the Fuel Quality Directive and the Renewable Energy Directive
would be easier to meet, as more conventional biodiesel would be available for
blending. Thirdly, the increase in demand for double
counted biodiesel such as that made from waste and residues would most likely
increase the price for such oils, which in turn could lead indirectly to
increased use of virgin vegetable oils, as the relative price[106] for virgin oils would
decrease. 5.3.9.2. Sensitivity related to the
estimated indirect land-use change emissions If the indirect land-use change emissions
are smaller or larger than the estimated central value, the resulting changes
in emissions in response to option B are set out in the table below: || || Low || Central || High Indirect land-use change emissions only- biofuels || Average ILUC emissions 2020 [g/MJ] || 11 || 18 || 26 Total ILUC emissions 2020 [Mt] || 9 || 14 || 20 Total emissions- direct and indirect- biofuels only || Average emissions 2020 [g/MJ] || 33 || 40 || 48 Overall savings 2020 [%] || 64% || 56% || 47% Change against baseline || Indirect land-use change emissions (Mt) || -21 || -33 || -46 Total emissions (Mt) || -43 || -56 || -68 Option B gives significant emission savings
across the whole range of sensitivity. The baseline has 29 Mt and 66 Mt of
estimated indirect land-use change emissions for low and high estimates
respectively. For both extremes, B is reducing indirect land-use change
emissions with around 70%. Looking at the "high" scenario, it leads
to overall savings of using biofuels of 47%, which is not significantly
different than 56% as is the result in the central case[107]. 5.4. Option C – introduce additional sustainability
requirements on certain categories of biofuels Option C involves introducing additional
sustainability requirements for biofuels. As explained in section 4, option C1
builds on LULUCF reporting and on taking measures to reduce deforestation. This
operates on a national level. Option C2 places additional requirements on
producers. That option thus operates at the level of economic operators
(project level). Option C1, the requirement of LULUCF
reporting and reducing deforestation in biofuel producing countries would need
to be implemented globally to ensure optimal effectiveness by avoiding
"leakage". However, this risk could be mitigated against by ensuring
that additional measures aimed at increasing the availability of overall
agricultural production are put in place simultaneously. In addition, it would seem appropriate to
allow for the provision of exemptions from increased requirements resulting
from the combination of B and C1 for producers in countries that may not be in
compliance if they can demonstrate that their biofuels present a low risk of
indirect land-use change emissions under C2. Although preliminary work suggests
that large quantities of biofuels could be produced in this way and this option
has therefore significant potential to reduce indirect land-use change
emissions, a methodology for certification of biofuels produced under these
conditions is not yet available. As such, this option is only assessed in
combination with other options in chapter 5.6. Option C1 Measures aimed at reducing
deforestation in biofuel producing countries Severe indirect land-use change emissions
takes place as a result of cropland expanding into forests. A global reduction
in deforestation and improved land planning would therefore significantly
reduce indirect land-use change emissions. Option C1 requires governments
supplying biofuels to the EU market to reduce deforestation, by means of
implementing the following: ·
Member States and Third countries should report
emissions by sources and removals by sinks of greenhouse gases resulting from
LULUCF activities, in accordance with rules laid down for Annex I parties in
Article 4 of the UNFCCC, and following the IPCC Good practice guidance for
LULUCF reporting[108].
This should start as soon as possible, preferably by 2015. ·
Member States and Third countries should limit the rate of greenhouse gas
emissions from losses of wetlands and forests (as defined in the Directives).
This should start as soon as possible, preferably by 2015. ·
Member States and Third countries should in parallel undertake appropriate
measures aimed at increasing the availability of feedstocks that are suitable
for biofuel production without increasing pressure on agricultural land through
putting in place additional[109]
measures in line with their national circumstances. Such measures would for
example include increasing yields in a sustainable manner and intensification
of pasture land; encouraging the remediation of degraded and contaminated land;
reduction of waste of agricultural products at farm level and post harvest (particularly
for developing countries), and post consumer level (in the case of developed
countries). The assessment should be finalised by 2015, and published in an
appropriate manner. This provision shall not apply if similar practices are
already in place. The paragraph on LULUCF reporting is
necessary in order to establish the statistics of sufficient quality to assess
the fulfilment of second paragraph[110].
It is thus not requiring third countries taking on commitments under the
Kyoto Protocol. The year 2015 as year of implementing
LULUCF reporting is necessary to allow third countries to build the necessary
capacity to comply. The advantage of linking it to the methodology required
from Annex I countries, is that methodology and verification is catered for by
a competent international body, where due considerations to both practical and
political limitations needs to be taken into account. Moreover, review and
verification of data is qualified by existing institutions. 5.4.1. Effectiveness in reducing
greenhouse gas emissions Adding a
requirement on all Member States and Third countries to reduce deforestation and put measures in place to increase the availability
of agricultural commodities, could reduce emissions considerably. However there
are three uncertainties: (1)
The effectiveness of this option relies on
whether countries are choosing to be in compliance or not, something that is
outside of the control of the EU; (2)
The mitigation in terms of greenhouse gas
emissions depends on the level of implementation in each country (i.e. if
implementation goes beyond minimum requirements), and finally; (3)
To what extent will leakages take place, i.e.
biofuels exporting countries comply, while indirect effects are displaced to
other countries. This can be mitigated depending on the extent that measures
put in place to increase production are successful. In order to
understand the possible magnitude of potential impacts, additional estimations
have been done using IFPRI-MIRAGE-BioF. Assuming compliance by Indonesia,
Malaysia, Brazil and Central American countries, the overall land-use change
emissions would be reduced considerably more than the baseline indirect
land-use change emissions, i.e. not only those related to biofuels. For
example, a reduction of deforestation rate to below 0.5% of land area would
lead to 730 Mt of land-use change emissions being avoided each year according
to the model. By comparison, the total estimated annual indirect land-use
change emissions in 2020 are likely to reach 48 Mt. However, reducing the deforestation has
economic costs for these countries and access to a relatively small (in global
terms) biofuel EU market might not represent a rational choice for all third
countries. Therefore in some particular cases, if imports from certain
countries (Brazil, Indonesia, Malaysia and Central America) are not allowed,
resulting in trade distortions and potential transmittance of impacts to other
countries, IFPRI-MIRAGE-BioF estimates that the total land-use change emissions
may actually increase by 3 Mt compared to the baseline land-use change
emissions. However, this risk could be mitigated by ensuring that additional
measures aimed at increasing the availability of overall agricultural
production put in place simultaneously are successful. To combine
option C1 with other incentives through trade agreements could prove effective
in persuading countries to comply with option C1. Examples of such preferential
treatment is explained below in context of the GSP+ (Generalised System of
Preferences). 5.4.2. Impacts on achieving the
Renewable Energy Directive transport target The impacts on
meeting the Renewable Energy Directive transport target would depend on the
number of countries that would qualify with the above criteria as well as the
ability of others to increase their production should any of the major
suppliers to the EU market be disqualified. 5.4.3. Economic impacts The economic impacts of option C1 would be moderate,
although feedstock costs would be expected to increase as trade in biofuels and
biofuel feedstocks with countries that are not implementing the necessary
measures would be halted. The option might pose a risk in the context
of WTO compliance, as the measure is country wide. However, country wide trade arrangements
are not new. The FLEGT and GSP+ (Generalised System of Preferences)[111] have several similarities to
the proposed option C1. Access to EU markets for timber is dependent on
bilateral trade agreements in the context of FLEGT, and additional trade
preferences to countries committed to sustainable development and good
governance[112]
is given to countries in compliance with GSP+. Administrative costs are limited to those
related to reviewing the submitted LULUCF reports, which is done by the UNFCCC.
Review of all Annex I countries' reports is already implemented. 5.4.4. Social impacts Social impacts of option C1 depends on how
effective the measures are, as discussed above. Employment in the EU is
expected to increase if few third countries comply, since more production must
take place within the EU. If most exporting countries are in compliance,
employment will be as in the baseline. Reducing deforestation may also have positive
impacts on the short term economic and social growth of developing countries. 5.4.5. Environmental impacts Also environmental impacts of option C1 are
dependent on its effectiveness. Deforestation is one of the major reasons for
loss of biodiversity. The reduction of deforestation in biofuel producing
countries could therefore reduce such losses, with additional positive impacts
related to soil, water and air. In this context, it is important that the risk
of indirect land-use change emissions transmission to other countries is
appropriately mitigated. 5.4.6. Other impacts If option C1 was to result in increased
implementation of LULUCF reporting it could have wider positive impacts, as
this would be likely to improve awareness of carbon losses related to land-use
and land governance. The implementation of this option could add some countries
to the group of 42 Annex I countries[113].
As emission savings offered by biofuels
would not reflect indirect land-use change, biofuels would contribute less to the
integrated approach for CO2 in cars. As such, larger contributions towards
achieving greenhouse gas emissions savings will be needed from energy
efficiency and other available technologies. 5.5. Option D - Attribute a quantity of greenhouse gas
emissions to biofuels reflecting the estimated indirect land-use impact. This
option builds in the estimated indirect land-use change emissions into the
greenhouse gas emissions accounting methodology for biofuels. According to the
assumptions made in this impact assessment, these estimates (factors) are added
to the expected 2020 baseline direct emissions (cultivation, processing and
transport) as described in table 4 of chapter 2[114]. This would
require incorporating the estimated indirect land-use change emissions values
in both the reporting of greenhouse gases, as well as ensuring compliance with the
minimum greenhouse gas thresholds of the sustainability criteria. The
initial assessment of this option focuses on the impacts in terms of exclusion
of certain feedstocks. This depends on whether eiluc
(the indirect land-use change emission factors)
are set at a level which mitigates/anticipates the risk of a possible model
overestimation (50%, 25% and 5% percentiles) and underestimation (95%, 75% and
50% percentiles) of crop-specific indirect land-use change emission factors
based on the latest IFPRI-MIRAGE-BioF report. At the end of this section, the impacts
from setting these levels based on crop group (i.e. starch, oil crops and sugar
crops) are also explored. 5.5.1. Level of indirect land-use
change emission factors per feedstock The results in table 8 below show whether
feedstocks would comply with the thresholds set out in the Directives
respectively when the range of estimated indirect land-use change emissions
resulting from the Monte-Carlo analysis are considered. Table 8 is thus the
crop specific values in Annex XV added to the direct emissions. All values are
shown in grams of CO2-eq./MJ. || Feedstocks || 5th || 25th percentile || Central || 75th percentile || 95th Biodiesel || Palm oil || 98 || 102 || 105 || 108 || 111 Palm oil with methane capture || 76 || 80 || 83 || 86 || 90 Soybean || 85 || 97 || 103 || 108 || 121 Rapeseed || 68 || 85 || 95 || 106 || 121 Sunflower || 63 || 79 || 86 || 93 || 104 2G biodiesel - land using || 12 || 18 || 21 || 24 || 32 2G biodiesel - non-land using || 9 || 9 || 9 || 9 || 9 Biooethanol || Wheat Process fuel not specified || 59 || 62 || 64 || 66 || 69 Wheat Natural- gas process fuel CHP plant || 42 || 45 || 47 || 49 || 52 Corn (maize) || 39 || 41 || 43 || 44 || 46 Sugar cane || 27 || 33 || 36 || 39 || 47 Wheat Straw as process fuel in CHP plant || 30 || 33 || 35 || 37 || 40 Sugar beet || 28 || 31 || 34 || 37 || 40 2G bioethanol - land using || 23 || 29 || 32 || 35 || 43 2G bioethanol - non-land using || 9 || 9 || 9 || 9 || 9 Table 8: impacts on the availability of
feedstocks when different ILUC factors based on feedstock specific values are
computed with expected typical direct emissions in 2020. Feedstocks in red/dark
grey fail to achieve required levels of savings at 50% by more than 5g; in grey
the fail by less than 5g, and; in green/light grey those than meet the 50%
threshold. From the table above, it is possible to
draw a number of preliminary conclusions. ·
Firstly, when the estimated indirect land-use
change emissions are added to the expected direct emissions in 2020, none of
the oil crops feedstocks would be able to meet the 50% greenhouse gas emissions
savings as required by the Directives, regardless of the range of indirect land-use
change estimates used. This is because these feedstocks present relatively high
direct emissions (e.g. soybean fails even when the 50% threshold is applied to
the direct emissions, before the estimated indirect land-use change emissions
are included), and are attributed relatively high estimated indirect land-use
change emissions. ·
The contrary is true for bioethanol feedstocks.
There only the least efficient wheat to bioethanol pathway is estimated not to
achieve the required minimum savings, unless the most conservative indirect land-use
change emission values (95th) are applied, in which case corn and
sugarcane also marginally fail to pass. All other bioethanol pathways would
otherwise be expected to achieve the required savings in 2020. In light of the above, the evaluation of
option D is carried out using the central values. Table 8 above shows how certain feedstocks
would be excluded (in red) from being counted towards the Directive targets. In
addition, attributing estimated indirect land-use change emissions to
feedstocks would also lead to incentives for biofuels with lower indirect
land-use change emissions due to the accounting and reporting by fuel suppliers
under the Fuel Quality Directive which is likely to bring about a significant
price differentiation in favour of low-ILUC transport fuels because those
biofuels will contribute much more than others to the attainment of a
supplier's obligation to reduce the greenhouse gas intensity of the fuels it
supplies. This second element is more sophisticated than the exclusion showed
in table 8 above as the emission estimates provide incentives for biofuels with
low estimated indirect land-use change emissions. Although the effectiveness of
this second element cannot be assessed under the methodology being used in this
impact assessment, further discussion is included in the sensitivity section of
this option. 5.5.2. Potential scenario -
meeting the targets of the Directives A range of possible scenarios for how to
comply with option D, while meeting the targets, can be developed. As no vegetable
oils can be used, it implies increasing the use of bioethanol, advanced
biodiesel or electric vehicles. A set of theoretical and extreme scenarios
exploring the use of these options are found in Annex XIV. The least
unrealistic scenario including a contribution from other technologies (electric
vehicles) that would also achieve the Renewable Energy Directive transport
target is set out below (these figures are included here for illustrative
purposes only as it is not possible to determine what the final mixture of
technologies would look like in reality). || Bioethanol [Mtoe] || Double counted biodiesel [Mtoe] || Electricity in road [Mtoe] Baseline || 6.7 || 1.8 || 2.1 Option D || 13.6 || 9 || 2.6 The scenario described above would require a
doubling in the levels of bioethanol and almost a fivefold increase in the
production of double counted biodiesel compared to the volumes set out in the
NREAPs. The increased consumption of ethanol is
question of blending possibilities, as the share of petrol consumption in the
EU is decreasing, and expected to continue to decline. The JEC Analysis of
scenarios for transport in the EU towards 2020[115], finds that in a scenario
with an E85 grade (flexi-vehicles) introduced now, and E20 in 2017, the EU
would consume 8-9 Mtoe of ethanol. If E20 is introduced in 2015 rather than in
2017 (2 years earlier), it leads to additional 0.7 Mtoe of ethanol being
consumed in 2020. This indicates how the consumption of 13.6 Mtoe of ethanol
would be very challenging. The availability of double-counted biodiesel
is a question of supply, both in terms of availability of raw-material e.g.
waste oil, but also a technical question whether enough production capacity can
be cost-efficiently installed by 2020. Achieving a supply of 9 Mtoe of double
counted biodiesel would therefore be very challenging. The scenario also needs increased efforts
i.e. 0.6 Mtoe of additional electricity in road transport by 2020. This would
be the equivalent of deploying an additional 2.3 million electric cars[116] by 2020 on top of what is
foreseen in the NREAPs. The NREAPs estimate that around 2.1 Mtoe of electricity[117] is consumed in electric
vehicles in 2020, which implies 8 million electric cars. It should be noted
that 2.1 Mtoe includes plug-in hybrids, as well as other road vehicles using
electricity. Again, this means that this scenario is not very likely and these
figures are included here for illustrative purposes only. 5.5.3. Effectiveness in reducing
greenhouse gas emissions This is the
most effective option in reducing the estimated indirect land-use change
emissions, with 40 Mt (85%) of the estimated indirect land-use change emissions
taking place in the baseline being avoided in 2020 as a result. Overall
emissions would be reduced by 66 Mt compared to the baseline, with biofuels
saving an average of 70% emissions compared to fossil fuels. Moreover, the
reported savings towards the Fuel Quality Directive target remain at -5.4% when
indirect land-use change emissions are included. The main
results in terms of greenhouse gas emissions are shown in the table below. Direct emissions- biofuels || Direct emissions 2020 (g/MJ) || 17 Direct emission savings 2020 [%] || 82% Total direct emissions 2020 (Mt) || 16 Indirect land-use change emissions- biofuels || Average ILUC emissions 2020 [g/MJ] || 10 Total ILUC emissions 2020 [Mt] || 8 Total emissions- direct and indirect- biofuels || Average emissions 2020 [g/MJ] || 27 Overall savings 2020 [%] || 70% Change against baseline || Indirect land-use change emissions (Mt) || -40 Total emissions (Mt) || -66 Table 9: Effectiveness analysis of option
D 5.5.4. Impacts on the Renewable
Energy Directive transport target There is a high risk that the transport
target of the Renewable Energy Directive is not achieved, as it will be
challenging for industry to produce enough advanced biofuels, particularly
since all conventional biodiesel is excluded and 70-75% of the transport fuels
are expected to be diesel in 2020, the technological developments needed to
achieve required increased bioethanol blends and electric vehicles do not take
place. 5.5.5. Economic impacts The viability of existing investments would
be affected in the long run, as Member States and industry cannot continue to
follow the submitted National Renewable Action Plans (NREAPs) since no conventional
biodiesel feedstocks would be available. This would have significant
implications for the existing EU biodiesel industry, and those countries with
the largest biodiesel installed capacity (i.e. Germany, Spain, France, Italy and the Netherlands) would be most affected. Moreover, activity will be reduced
in related industries such as those involved in the production of vegetable
oils/crushing of oil crops for all food/feed/biofuels markets, mainly being
present in Germany, France, Spain, the Netherlands and the UK). As foreseen in
the Directives[118],
the introduction, if appropriate, of an indirect land-use change emission
methodology would need to consider the provision of safeguards to investment
undertaken from existing installations involved in the production chain that
would otherwise have complied with the greenhouse gas emissions requirements. On the one hand, it could be argued that
certainty for new investments would be increased since the estimated full
greenhouse gas emissions impacts are being taken into account. Bioethanol
production capacity would need to be increased significantly to make up for the
increased demand, which would create economic opportunities in this area. In
addition, there would also be significant opportunities for the second
generation biofuel industry as their better performance compared with
conventionally produced biodiesel would be made apparent. On the other hand,
the introduction of indirect land-use change emission factors, and the update
of these values, would create uncertainty regarding what an updating might
bring in terms of consequences to the industry. This process would therefore
need to be managed carefully. With regard to production costs, the change
compared to the baseline is expected to be moderate as most bioethanol feedstocks
required to make up for the missing biodiesel remain available once indirect
land-use change emission factors have been applied. However, the need for
electric vehicles and 2nd generation biofuels will increase the
aggregate cost, depending on how the costs of these technologies develop. There
are no additional administrative costs associated with this option, as
requirements under the current sustainability scheme are neither changed nor
increased. Impacts on
security of supply can be adversely affected if the necessary bioethanol blend
levels are not available in terms of specification of the vehicle fleet, or the
required high-saving biodiesel volumes are not supplied in time. In any case,
the need to accelerate developments in order to make the existing fleet to be
compatible with higher bioethanol blends will have increased costs for the
vehicle manufacturers. Again, the foreseen application of the grandfathering
clause on installed capacity would be expected to help with this transition. Similarly, this option would have a number
of trade impacts. Imports of conventional biodiesel feedstocks into the EU
would be severely limited. However, trade in second generation biodiesel
feedstocks and/or second generation biodiesel if available would increase, as
well as in bioethanol feedstocks, and/or processed bioethanol if EU processing
capacity does not increase in time. In terms of WTO compatibility, option D may lead to issues related
to its reliance on modelling for the determination of the factors. However,
similarly constructed indirect land-use change emission factors applying in US
federal and Californian Low Carbon Fuel Standard have not been challenged to
date. That said, although the abovementioned legislation leads to a
preferential treatment of feedstocks with estimated low-ILUC impacts, it does
not lead to certain biofuel feedstocks being totally excluded from receiving
financial support or counted towards mandatory targets. 5.5.6. Social impacts With regard to EU employment, the resulting
impacts would depend on whether the opportunities created through the increase
in the bioethanol and advanced biodiesel industry are able to make up for the
reduction in activity in the conventional biodiesel industry. Again, the
foreseen application of the grandfathering clause on installed capacity would
be expected to help with the transition. While rapeseed and sunflower is excluded,
the adverse employment effects within the EU for famers are likely to be
limited, as farmers would respond to the shift in demand from rapeseed to
cereals and sugar beet. Notably, the main Member States producing rapeseed
(i.e. Germany, France, Poland and United Kingdom) are also the main producers
of cereals and sugar beet. On the other hand, reduced production of oil crops
can be problematic in terms of crop rotation in some areas where sugar beet
cannot be grown because of soil requirements (i.e. rapeseed and sugar beet are
used as a break crops), and could increase the deficit in feed protein (i.e. oil
crops yield higher quality meals compared to cereals), as farmers are unlikely
to switch to protein crops (i.e. peas, beans). Rural development is dependent on the same
set of variables, and thus difficult to assess. However, the required increased
use of advanced biodiesel would be expected to have positive impacts. Pressure on
vegetable oil prices will be reduced in commodity markets as a result of
reduced biodiesel demand. Conversely, pressure on coarse grains and sugar
prices will increase, although to a lesser extent as the demand for biofuels of
these commodity groups represents a lower share of the global markets than for oil
crops. As such, the reliance of the EU on imported vegetable oils would
improve, as well as reducing the excess production of cereals and sugars. 5.5.7. Environmental impacts Based on the
results above, this option would also be expected to be the most effective in
reducing the biodiversity impacts associated with indirect land-use change. Due
to their irreversible and cumulative nature, the extent by which these impacts
are addressed would depend on when action is taken. In addition,
this option would also significantly contribute towards avoiding other
environmental impacts associated with land conversion (i.e. adverse water, soil
and air impacts). 5.5.8. Other impacts This option would be expected to motivate technological development.
Firstly, it would provide strong incentives for accelerating the introduction
of advanced biodiesel. Moreover, the increased greenhouse gas emissions requirements
should encourage improvements in the performance of 1st generation bioethanol
feedstocks. The implementation of the 6% target of the Fuel Quality Directive provides
strong price signals for improved greenhouse gas emissions performance for
biofuels as a result of the incorporation of indirect land-use change emissions
factors, as well as incentives for transport fuels and energy with low
estimated indirect land-use change emissions, as explained in section 5.5.1[119]. All biofuels
that would be counted towards these targets meet the greenhouse gas emissions
thresholds when estimated indirect land-use change emissions are included. As a
result, emission savings offered by qualifying biofuels would be coherent with
the integrated approach for CO2 in cars. Option D would
imply that the greenhouse gas calculation methodology would be modified from
the existing principles
of greenhouse gas performance being the result of the biofuel producer's
actions, to a system where the greenhouse gas performance is also dependent on
the actions of actors outside the producer's control. Adding the indirect
impact to the direct emissions, implies that if the
same principles of accounting greenhouse gas emissions as under option D, were
to be applied to all commodities, it would result in double-counting, as the
indirect emissions of biofuels are the direct emissions of another
commodity. In practical
terms it might be less relevant as greenhouse gas emissions from the majority
of commodities may never be accounted for globally. However, as discussed in
the LULUCF section: If LULUCF accounting were to be implemented globally together
with option D, indirect land-use change emissions would be accounted for both
as part of the biofuel greenhouse gas performance, and as LULUCF emissions in
the country where indirect land-use change took place. Whether or not
to estimated indirect land-use change emissions should be included in the
greenhouse gas accounting of the biofuels, also relates to the principle of
carbon-neutrality of biofuels. Currently it is assumed that tailpipe carbon
emissions are zero, as a similar amount of carbon is absorbed by the feedstocks
as they are grown. In a context of biofuels using existing cropland, where
crops would have been grown anyway, a similar amount of carbon would also have
been extracted from the atmosphere. While it is difficult to assess the overall
carbon impact of the land remaining as non-biofuel cropland, option D would to
some extent capture this. The greenhouse
gas accounting methodology has to balance the need for ensuring significant
savings from the use of biofuels with a stable and predictable policy
framework. There are broader implications of option D in that introducing
indirect land-use change emission factors for biofuels could lead to similar
factors being introduced for other types of bioenergy, especially due to use of
the same feedstocks. However, in order to ensure that biofuels and bioenergy
successfully contribute towards the EU climate objectives and related specific
greenhouse gas emissions reduction targets as set out in the " A
Roadmap for moving to a competitive low carbon economy in 2050", it is
necessary to ensure that biofuels do not lead, directly or indirectly, to a
decrease of the net greenhouse gas benefits. This could be challenging if the
indirect emissions associated with land using biofuel are not minimised. Moving
forward towards 2050, it is currently considered that indirect emissions could
be accounted for through global LULUCF accounting or through adding factors as
foreseen in option D. 5.5.9. Sensitivity of option D 5.5.9.1. Uncertainty related to the
effectiveness of option D The considerable increase in demand for
double counting biodiesel such as that made from waste and residues would most
likely increase the price for such oils, which in turn could lead indirectly to
increased use of virgin vegetable oils, as the relative price[120] for virgin oils would
decrease. 5.5.9.2. Sensitivity related to the
estimated indirect land-use change emissions If the indirect land-use change emissions
are smaller or larger than the estimated central value, the resulting changes
in emissions in response to option D4 are set out in the table below: || || Low || Central || High Indirect land-use change emissions- biofuels || Average ILUC emissions 2020 [g/MJ] || 4.4 || 10 || 17 Total ILUC emissions 2020 [Mt] || 3 || 8 || 13 Total emissions- direct and indirect- biofuels || Average emissions 2020 [g/MJ] || 21 || 27 || 34 Overall savings 2020 [%] || 77% || 70% || 62% Change against baseline || Indirect land-use change emissions (Mt) || -26 || -40 || -53 Total emissions (Mt) || -52 || -66 || -79 The introduction of D4 gives overall
savings from 62% to 77%. The main conclusion is that under all circumstances,
levels higher than 50% greenhouse gas emissions savings would be achieved[121]. Scenario D4 would correspond to levels of
an indirect land-use change emission factor being set at 50th
percentile. Applying the factors at lower or upper end of the sensitivity range
as set out in table 7, are not expected to have major impacts on the overall
situation described above (i.e. conventional oil crops would still need to be
replaced by equivalent levels of bioethanol, double counted biodiesel and
electricity in roads). 5.5.9.1. The impacts of including the
estimated indirect land use change greenhouse gas emissions in the reporting of
the greenhouse gas emission savings of biofuels. The reporting of the emissions towards the
Fuel Quality Directive would be expected to create a strong price premium for
those biofuels with the lowest estimated indirect land-use change emissions.
The reason for that is that biofuels with low estimated indirect land-use
change emissions will contribute much more than others to the attainment of a
supplier's obligation to reduce the greenhouse gas intensity of the fuels it
supplies. As it has not been possible to assess this aspect of option D with
the assessment methodology applied in this Impact Assessment, this element is
further outlined in the text and figure below. Figure 7: Greenhouse gas savings obtained under the
Fuel Quality Directive ·
fuels A, B and C offer similar greenhouse gas
emission savings when only direct emissions are taken into account; ·
when the estimated indirect land-use change
emissions are included in the reporting, the contribution to the attainment of
a supplier's obligation to reduce the greenhouse gas intensity of the fuels it
supplies differs considerably between the three fuels. Fuel A would be
completely excluded under this option as it would fail to comply with the
greenhouse gas emissions thresholds under the sustainability criteria. ·
as such, a fuel supplier would need twice the
amount of fuel B to obtain the same savings as by using one volume of fuel C.
This is therefore expected to result in a high price premium for fuel C, as it
would count twice as much as B towards these targets, hence maximising its
share over B in the final mix. Although it has not been possible to assess
the impact on the effectiveness of this aspect of option D in isolation in a
quantitative manner with the assessment methodology
applied in this Impact Assessment, some possible
outcomes are presented in Annex XVI. The findings strongly suggest that
significant amounts of further improved biofuels with low risk of indirect
land-use change emissions would be expected in the final mix as a result. 5.5.10. Exploring different levels
of disaggregation - Indirect land-use change emissions aggregated per crop
group (oil crops, starch and sugars) The results in table 10 show whether
feedstocks would comply with the thresholds set out in the Directives
respectively when the weighted per crop group average of estimated indirect land-use
change emissions resulting from the Monte-Carlo analysis are considered. Table 10
below is thus the crop group average values in Annex XV added to the direct
emissions. || Feedstocks || 5th || 25th percentile || Central || 75th percentile || 95th Biodiesel || Palm oil || 85 || 98 || 106 || 114 || 125 Palm oil with methane capture || 64 || 76 || 84 || 92 || 104 Soybean || 81 || 94 || 102 || 109 || 121 Rapeseed || 75 || 87 || 95 || 103 || 115 Sunflower || 67 || 80 || 87 || 95 || 107 2G biodiesel - land using || 10 || 16 || 18 || 22 || 28 2G biodiesel - non-land using || 9 || 9 || 9 || 9 || 9 Bioethanol || Wheat Process fuel not specified || 58 || 61 || 63 || 64 || 67 Wheat Natural- gas process fuel CHP plant || 41 || 44 || 46 || 48 || 50 Corn (maize) || 40 || 43 || 45 || 47 || 49 Sugar cane || 25 || 30 || 33 || 36 || 43 Wheat Straw as process fuel in CHP plant || 29 || 32 || 34 || 36 || 38 Sugar beet || 32 || 37 || 40 || 43 || 50 2G bioethanol - land using || 22 || 27 || 30 || 33 || 39 2G bioethanol - non-land using || 9 || 9 || 9 || 9 || 9 Table 10: impacts on the availability of
feedstocks when different ILUC factors based on feedstock specific values are
computed with expected typical direct emissions in 2020. Feedstocks in red/dark
grey fail to achieve required levels of savings at 50% by more than 5g; in grey
the fail by less than 5g, and; in green/light grey those than meet the 50%
threshold. All values are shown in grams of CO2-eq./MJ. None of
the oil crops feedstocks are either able to achieve the 50% minimum greenhouse
gas emissions thresholds. Similarly, none of the bioethanol crops other than
the least efficient wheat processes seem to have any difficulties to meet the
required thresholds until conservative indirect land-use change emission
estimates are applied. 5.6. Option E - Limit the contribution from conventional
biofuels to the Renewable Energy Directive targets. Option E involves limiting the us of
conventional biofuels from food crops by setting the maximum contribution of
such biofuels towards the 10% target of the Renewable Energy Directive to current
production levels at 5%. To ensure coherence, the equivalemt quantity in energy
content is the maximum that can contribute towards the Member States' overall
targets for renewable energy. 5.6.1. Potential scenario -
meeting the targets of the Directives A range of possible scenarios for how to
comply with option E, while meeting the targets, can be developed. The cap is
assumed to be equivalent to around 14 Mtoe. As less conventional biofuels are
likely to be used, it implies increasing the use of advanced biofuels or
electric vehicles. A scenario including contribution from other technologies (electric
vehicles) that would meet the Fuel Quality Directive target and achieve the
Renewable Energy Directive transport target is set out below (these figures are
included here for illustrative purposes only as it is not possible to determine
what the final mixture of technologies would look like in reality). || Bioethanol [Mtoe] || Double counted biofuels [Mtoe] || Electricity in road [Mtoe] Baseline || 6.7 || 1.8 || 2.1 Option E || 6.7 || 6.0 || 2.6 The scenario described above would require more
than a tripling of the production of double counted biofuels compared to the
volumes set out in the NREAPs. If combined with optoin C2, the reliance on
advanced biofuels would be reduced, which would in turn reduce costs and make
it easier to comply with the targets of the Directives. 5.6.2. Effectiveness in reducing
greenhouse gas emissions Capping the use
of conventional biofuels implies that overall emissions are reduced by 48 Mt
compared to the baseline. 27 Mt (56%) of the estimated indirect land-use change
emissions occurring in the baseline would be avoided in 2020. The biofuels used
are on average saving 44% compared to fossil fuels when indirect land-use
change emissions are considered. The emissions reported towards the Fuel
Quality Directive target will not reflect estimated indirect land-use change
emissions. The main results are shown in the table
below. Direct emissions- biofuels || Direct emissions 2020 (g/MJ) || 24 Direct emission savings 2020 [%] || 74% Total direct emissions 2020 (Mt) || 21 Indirect land-use change emissions – biofuels || Average ILUC emissions 2020 [g/MJ] || 27 Total ILUC emissions 2020 [Mt] || 20 Total emissions- direct and indirect- biofuels || Average emissions 2020 [g/MJ] || 51 Overall savings 2020 [%] || 44% Change against baseline || Indirect land-use change emissions (Mt) || -27 Total emissions (Mt) || -48 Table 11: Effectiveness analysis of option
E 5.6.3. Impacts on achieving the
Renewable Energy Directive transport target There is a risk
that the transport target of the Renewable Energy Directive is not achieved, if
the technological development required for the significant increase in
deploying advanced biofuels is not be achieved. However, this option is the
least challenging from a blending perspective as it does not require additional
levels of bioethanol, and the overall levels of biodiesel would be expected to
be lower than in the baseline scenario as a higher share of double counted
biofuels would be expected. 5.6.4. Economic impacts The change in
cost compared to the baseline is expected to be moderate as a range of
feedstocks would still available. The increased use of electricity in road
transport, and 2nd generation biofuels will increase aggregate
costs, depending on how the costs of these technologies develop. Financial investment stability is affected,
as the expected use of conventional biofuel feedstocks would be reduced by a
third in 2020. This would have implications for the existing EU biofuel
industry, although the impact is limited, as the cap would maintain today's
production levels of conventional biofuels. It also implies that Member States and industry may not be able to follow the submitted National Renewable
Action Plans (NREAPs), which may have political implications. Certainty for new investments would be
increased as it would be clear what levels of both conventional and advanced
biofuels that would be needed by 2020. In addition, there would also be very
strong opportunities for the second generation biofuel industry who would as a
result of the limit of conventional, have a garanteed share of the market. There
are no additional administrative costs associated with this option, as
requirements under the current sustainability scheme are neither changed nor
increased. Adverse impacts
on security of supply and trade may take place if the necessary volumes of
advanced biofuels are not available. In both cases, the limit of on the
conventional biofuels would ensure that today's energy security levels and
trade volumes are mantained. 5.6.5. Social impacts With regard to EU employment, current
production levels and associated jobs woud be mantained through the cap. The
extent of which the additional incentives for opportunities to be created
through the increase in the advanced biofuel industry are able to make up for
the reduction in activity in the conventional biofuel industry would determine
the overall impacts. While the share of conventional biofuels is
reduced by a third, the adverse employment effects within the EU for famers are
likely to be limited, as farmers would respond to the shift in demand from growing
conventional to growing and collecting advanced biofuel feedstocks. Rural development is dependent on the same
set of variables, and thus difficult to assess. However, the required increased
use of advanced biodiesel would be expected to have positive impacts. The reduction in food based biofuels will
lower the pressure on global food and feed markets, in particular for oil crops
where the biodiesel production represents a larger share of the global supply,
but also for cereals and sugars as bioethanol crops would also be capped. 5.6.6. Environmental impacts Adverse
biodiversity impacts are reduced as it is expected that reduced indirect
land-use change emissions are correlated with reduced conversion of bio-diverse
areas. This option would also contribute towards avoiding other environmental
impacts associated with land conversion (i.e. adverse water, soil and air
impacts) as the share of biofuels that need land for their production is
limited. 5.6.7. Other impacts This option
would be expected to motivate technological development. Firstly, it would
provide very strong incentives for accelerating the introduction of advanced
biofuels through limiting the contribution of conventional alternatives. Secondly,
it provides clear signals to industry as to what volumes of conventional and
advanced biofuels are needed to 2020. The option is
simple in design and implementation. This option
does not change the existing calculation methodology and thus it is not likely
to be challenged by the WTO. Moreover, this option does not depend on modelling
for the design of the policy measure. However, it does not distinguish between
feedstocks according to their estimated indirect land-use change impacts. As emission savings offered by biofuels
would not reflect indirect land-use change, biofuels would contribute less to the
integrated approach for CO2 in cars. As such, larger contributions
towards achieving greenhouse gas emissions savings will be needed from energy
efficiency and other available technologies. 5.7. Combination of option D with C2 - Attribute a quantity
of greenhouse gas emissions to biofuels reflecting the estimated indirect
land-use impact whilst providing exemptions to those biofuels feedstocks
produced under criteria covered by option C2. The rational behind combining option D with
option C2 is that biofuels produced under C2 requirements could be considered
to cause minimal indirect land-use change impacts which would justify such
factors not being applied. It is difficult to provide an accurate
quantification of the impacts arising from this option, as it is not possible
to estimate the actual changes in biofuels' supply compared to the baseline. And
although this approach remains promising, it is worth noting that currently
there are no biofuels in the market that have been certified to be compliant
with these rules. However, a number of pilots to develop detailed
methodological rules for the certification of biofuels produced are underway. 5.7.1. Effectiveness in reducing
greenhouse gas emissions The biofuels
that are produced respecting the criteria under option C2 would have minimal
risk of displacement effect and thus indirect land-use change emissions[122]. A successful implementation
of this option would therefore be expected to further achieve reductions beyond
those achieved by option D (40 Mt, or 85% of annual indirect land-use change
emissions in 2020). 5.7.2. Economic impacts As described in more detail under Annex XII,
a number of case studies have shown that production costs will generally not be
significantly higher under C2 requirements than current practices although
these are likely to be project specific. On the
other hand, there will be higher administrative costs associated with the
certification of feedstocks produced under C2, as additional proof of
compliance to those requirements already in place under the current system
would be required. Preliminary data from pilots carried out to date[123] suggest that the additional costs
for these requirements could be moderate (i.e. 10-15% increase from current
certification costs) if they were integrated as part of current certification
schemes as to maximise audit costs already in place. Costs would be expected to
be much greater if this certification had to be separated from current process,
as it would result in duplication of audit costs. With regard to investments, providing
exemptions to option D is likely to improve the availability of conventional
biodiesel feedstocks and so assist with the viability of existing investments
in installed biodiesel production capacity. Although this would also help
maintaining activity in related biodiesel industries such as those involved in
the production of vegetable oils/crushing of oil crops for all
food/feed/biofuels markets, costs may be incurred by these industries if their
processes need to be adjusted to deal with different feedstocks (i.e. rapeseed
crushing to soy). In addition, the foreseen application of the grandfathering
clause on installed capacity could be a useful way of helping with the
transition for establishing a robust implementation method for the
certification and verification of the additional C2 criteria. Similarly, the combination of these options would reduce the impacts
on trade and is likely to increase imports of conventional biodiesel feedstocks
into the EU. In terms of WTO compatibility, any potential issues related to its
reliance on modelling for the determination of the factors as mentioned under
option D may be reduced by the introduction of the possibility to be exempted. 5.7.3. Social impacts The social impact would be as for option D,
with the exception that the employment effects within the EU for famers will be
determined by the degree to which European production can be adjusted to these
criteria, and how the industry is adapting to the opportunities created by
exemptions created through option C2. 5.7.4. Environmental impacts The environmental impacts are expected to be
improved further than for option D. 5.7.5. Other impacts Other impacts would be as for option D. 6. Comparison
of policy options The table below summarises the main issues related to
the different options. Option A || Effectiveness || Advantages || Disadvantages Average total of GHG savings (incl. estimated ILUC) of 22%. 48 Mt of annual estimated ILUC emissions by 2020 (BAU). || Biofuel and other related industries' investment not affected. Development opportunities inside and outside the EU not affected. RED transport targets and FQD are achieved according to NREAPS. No stranded investments. || ILUC emissions and biodiversity impacts not mitigated. No incentives for further technological development (i.e. improved GHG savings and/or advanced biofuels). Pressure on vegetable oil prices. Estimated ILUC emissions not reported. Option B || Average total of GHG savings (incl. estimated ILUC) of 56%. 14 Mt of annual estimated ILUC emissions by 2020. || Simple in design and implementation, as it is coherent with existing methodology. Reported GHG performance of biofuels is only dependent on action taken by the biofuel producers themselves. Does not exclude all 1st generation biodiesel. Clear incentives for producers to improve direct GHG savings. || Risk of not achieving RED transport target and FQD from reduced biofuel availability. Most oil crops excluded, including all rapeseed which currently represents more than half of the biofuel feedstocks used in the EU. This would require industrial adjustment. The exclusion of rapeseed with a threshold of 60%, and its corresponding environmental and GHG impacts, are sensitive to technological progress in production pathways. Increased administrative costs as certain operators need to report actual direct greenhouse gas emission performance. Estimated ILUC emissions not reported. Option C1 || Estimated ILUC emissions ranging between a reduction of 740Mt to an increase of 3 Mt annual ILUC emissions. || Potentially large emissions savings if countries implement good governance for land-use, reduce deforestation, and limit land-use change emissions from other commodities. || Risk of WTO incompatibility. Potential leakage effects as the exclusion of production from certain countries can increase distortions and even in some cases lead to increased emissions. Option D || Average total of GHG savings (incl. estimated ILUC) of 70%. 8 Mt of annual estimated ILUC emissions by 2020. || Most efficient in reducing estimated ILUC emissions and other environmental impacts such as biodiversity. Strong incentives for development of bioethanol and advanced biofuels, particularly biodiesel. The option for addressing ILUC referred to in the Directives. Targets biofuels with high estimated ILUC directly, reducing demand for such feedstocks. || High risk of not achieving RED transport target and FQD from significant reduction in biofuel availability. Policy methodology dependent on actions outside the control of biofuel producers. Uncertainty for industry due to expected updates of ILUC factors. All oil crops excluded, including all rapeseed which currently represents more than half of the biofuel feedstocks used in the EU. This would require major industrial adjustment. Potential WTO compatibility issues. Options E || Average total of GHG savings (incl. estimated ILUC) of 44%. 20 Mt of annual estimated ILUC emissions by 2020. . || Efficient in reducing estimated ILUC emissions and other environmental impacts such as biodiversity. Moderate industrial adjustment required to 2020 as current production levels of conventional biofuels would be mantained. Least demanding option with regards to the technical blending compatibility of vehicles. Very strong incentives for development of advanced biofuels and clarity for future investors. No increase in administrative costs. || Some risk of not achieving RED transport target and FQD from reduced biofuel availability. Estimated ILUC emissions not reported. No difference in treatment across conventional biofuels acording to their ILUC impacts. Option D + C2 || Improved efficiency from D expected but not possible to quantify. || As for option D but in addition: Increased availability of conventional biodiesel and so lower impacts on biodiesel and related industries. Possibilities for biodiesel derived from vegetable oils to supply the EU, which increases security of supply compared to D. Provides a method under the control of biofuel producers to avoid uncertainty around ILUC factors. || As for option D but in addition: Certification methodology for production of low ILUC risk biofuels remains to be developed. Potentially costly and administratively burdensome to comply with C2. 7. Conclusion On the basis of the analytical work
presented in this Impact Assessment, it is possible to draw a number of
conclusions: (1)
the estimated indirect land-use change emissions
are, despite the better understanding and recent improvements in the science,
vulnerable to the modelling framework and the assumptions made; (2)
the use of biofuels in the EU saves emissions,
also when estimated indirect land-use change emissions are included. In
addition, the models indicate a hierarchy of biofuel types according to their
indirect land-use change impacts, these being considerably higher for typical
biodiesel feedstocks (oil crops), than for bioethanol feedstocks (cereals, and
sugar crops); (3)
given the strong reliance on conventional biodiesel,
and to a lesser extent conventional bioethanol, in projected biofuel volumes to
2020, there is a high risk that the estimated indirect land-use change
emissions will significantly reduce the expected savings from the policy if no
action is taken to mitigate indirect land-use change emissions; and; (4)
the development of advanced biofuels, using
low-value resources as straw, wood and forestry residues is slower than
previously expected, as the costs associated with producing such fuels is
higher than the alternative conventional biofuels. There are
reasonable grounds to believe that indirect land-use change emissions could
partly undermine the greenhouse gas savings offered by using biofuels. In
application of the precautionary principle, option A) is therefore discarded. Consideration
has also been given to options for introducing additional sustainability
requirements on certain categories of biofuels, including certain actions that
could be implemented at both country and project level. With regard to
country-wide sustainability criteria, the assessment showed that this option
would need to be implemented globally in order to be fully effective. In
respect of project level actions, the Impact Assessment showed that although
biofuels produced under these conditions could be effectively promoted through
being considered as exemptions to the application of ILUC factors, these
criteria are insufficiently developed at this time to be included in
legislative proposal as no certification scheme currently exists. As such,
option C) must also be discarded. With regards to a threshold increase, as
described for option B), this option would seem effective in reducing indirect
land-use change as long as it leads to the replacement of those biofuels with
estimated high indirect land-use change emissions (i.e. vegetable oils) by
those with estimated low emissions (i.e. cereals, sugars and advanced
biofuels). However, the effectiveness of a threshold increase to 60% (i.e. a
reduction of indirect land-use change emissions of 70%, from 46 Mt of CO2-eq./yr
to 14 Mt CO2-eq./yr in 2020) would be reduced by two thirds if
further improvements in the greenhouse gas balance of main vegetable oil crops
to levels which seem technologically feasible, can be achieved. As such, the
uncertainty around the effectiveness of this approach would always remain high
unless much higher thresholds are applied across the board, which would
discriminate against biofuels with low estimated indirect land-use change
emissions. This option in isolation has therefore been discarded. Option D concerns the introduction of
factors to demonstrate compliance with the sustainability criteria as well as
the reporting of greenhouse gas emissions towards emission reduction targets.
This would seem the most effective option in reducing indirect land use change
emissions (i.e. a reduction of indirect land-use change emissions of 85%, from
46 Mt of CO2-eq./yr to 8 Mt CO2-eq./yr in
2020). However, the application of this option in isolation would require major
industrial adjustment which does not seem achievable in the period to 2020.
This is because it would require a) the exclusion of all vegetable oil
biodiesel which today represents the vast majority of the market; b)
unrealistic levels of bioethanol given the current blend limits; and c)
unrealistic levels of advanced biofuels coming into the market. Moreover, the introduction of factors in the
sustainability criteria would not take into account the limits of the modelling
in the policy design. As such, the
application of this option in isolation has been discarded. The remaining option E, i.e. limiting the
amount of conventional biofuels counting towards the Renewable Energy Directive
transport target to current production levels, would also be effective in
reducing indirect land-use change (i.e. a reduction of indirect land-use change
emissions of 55%, from 46 Mt of CO2-eq./yr to 21 Mt CO2-eq./yr
in 2020). In addition, this option would require moderate industrial adjustment
as it would only exclude vegetable oil biodiesel beyond current production
levels in the run up to 2020 and would not necessarily pose a technical
challenge from a blending limit perspective, while providing a strong incentive
for increasing the share of advanced biofuels. The incentives for producing
advanced biofuels would be strong, as the amount of double counted advanced
biofuels would need to increase significantly[124]. Option E thus appears to
provide a basis of a suitable way forward. This Impact Assessment shows that a
balanced approach based on option E, accompanied by complementary elements of options
B and D and additional incentives for advanced biofuels, would be the best way
to minimise estimated indirect land-use change emissions. This is because (1)
option E avoids any additional ILUC-impacts to
happen for the period up to 2020 as it limits the use of conventional biofuels
to current production levels, while at the same time the targets for renewable
energy of the Renewable Energy Directive remain achievable; (2)
it protects existing investments, while giving a
clear message that after 2020 only advanced biofuels will be supported. This provides
the needed certainty for new investments in the sector as no further changes
would occur up to 2020; (3)
it distinguishes between feedstocks according to
their estimated indirect land-use change impacts which would be reported,
thereby providing more transparency; (4)
sustainability of biofuels remains a question of
verifiable and measureable direct emissions; (5)
the enhanced incentives and accounting for
advanced non-land using biofuels to four times the contribution of conventional
biofuels will spur development of such biofuels with zero risk of indirect
land-use change emissions, as no land is used for their production. Although it has not been possible to assess
the effectiveness of this package of measures under the current methodology, it
is expected to reduce indirect land-use change emissions significantly. As a
minimum, the package of measures will reduce indirect land use change emissions
as option E in isolation (55% by 2020). However, it is expected that the
additional incentives for advanced biofuels will lead to a further shift away
from biofuels with high estimated indirect land-use change emissions. In conclusion this combination would minimise
the risks of indirect land-use change emissions, while protecting existing
investments and, at the same time, acknowledging and taking into account in the
policy design the limits of the modelling. 8. Future
monitoring and evaluation The Commission will monitor the impacts of
indirect land-use change in the framework of its bi-annual reports referred to
in Article 23 (1) of the Renewable Energy Directive: The
Commission shall monitor the origin of biofuels and bioliquids consumed in the
Community and the impact of their production, including impact as a result of
displacement, on land-use in the Community and the main third countries of
supply. Such monitoring shall be based on Member States’ reports [..] and those
of relevant third countries, intergovernmental organisations, scientific
studies and any other relevant pieces of information. Related to this
and other monitoring and reporting requirements of the Renewable Energy
Directive, a study for the development of baseline data is being carried out[125]. The first Commission report
on the basis of this monitoring and analysis is due in 2012. In addition to ex-post assessment
of impacts, the monitoring would include the development of the scientific work
on ex-ante estimations of the effects of indirect land-use change. In the
context of the current understanding of modelling indirect land-use change
emissions, including the relative importance of the various parameter involved
in estimating indirect land-use change emissions, there is a need to monitor a
range of elements, including, but not limited to; the use of co-products, yield
increases induced by the biofuel policy, displacement of cropland (i.e. what
used to be the land-use of the cropland where biofuels are now grown) and
developments on protecting high carbon stock land. 9. Glossary Advanced biofuel technologies = biofuels
typically produced from non-food/feed feedstocks such as wastes and residues
(i.e. wheat straw, municipal waste), non-food crops (i.e. grasses, miscanthus)
and algae. Most technologies are at pilot scale or in development. Bioethanol = alcohol-based biofuel
typically produced from starch and sugar crops such as wheat and sugar beet,
and used as a petrol additive for its use in motor vehicles. Biodiesel = oil-based biofuels typically
produced from vegetable and animal fats, such as rapeseed oil and tallow, and
used as a diesel additive for its use in motor vehicles. Biofuels = liquid or gaseous fuel used for
transport purposes produced from biomass. Bioliquids = liquid fuels used for energy
purposes other than transport, including electricity, heating and cooling,
produced from biomass. These are typically produced from vegetable oils such as
palm and waste oils. Biomass = the biodegradable fraction of
products, waste and residues from biological origin from agriculture (including
vegetable and animal substances), forestry and related industries including
fisheries and aquaculture, as well as the biodegradable fraction of industrial
and municipal waste. Conventionally produced biofuels = biofuels
typically produced from land using feedstocks which are also used in other
markets (i.e. food and feed). These also include the use of certain waste and
residues which do not require complex technological processes (i.e. biodiesel
from used cooking oil or animal fat). Direct land-use change = land-use change
occurring directly, i.e. mostly referred to in the context of the
conversion of land areas to cropland. Direct emissions from biofuels = greenhouse
gas emissions associated directly with the production of biofuels. These may
include greenhouse gas emissions associated with the cultivation and harvest of
feedstocks, with the processing and production of the biofuel, its transportation,
direct land-use change. High carbon stock land = Land with large
amounts of carbon stored in biomass (trees, grass, roots etc.) and/or soil. Indirect land-use change = land-use change
occurring indirectly i.e. mostly referred to in the context of land-use
change as a result of displaced demand previously destined for food/feed/fibre
market as a result of biofuel demand. Land-use change = the conversion of land
from one use to another, e.g. from forestry to cropping. ACRONYMS ACP African, Caribbean and Pacific States AEZ Agro-environmental zones B10 and B30 Diesel blends containing 10%
and 30% biodiesel in volume. CAP Common Agricultural Policy CARB Californian Air Resources Board CEPII French: Institute for Research on
the International Economy CFPP Cold Filter Plugging Point CGE Computable General Equilibrium
models CHP Combined heat and power CIS Commonwealth of Independent States (Ex-USSR) CO2 Carbon dioxide COP Conference of Parties COWI Consultancy within Engineering,
Environmental Science and Economics CNG Compressed natural gas DDGS Dried Distillers Grains with Solubles EBB European Biodiesel Board EC European Commission EJ Exajoule (1018 joules) Epure European Bioethanol association ETS European Emissions Trading Scheme EU/EU-27 European Union E10 and E85 Petrol blends containing 10%
and 85% bioethanol in volume FAME Fatty acid methyl esther Fediol EU Oil and Protein Meal
association FFC Fossil fuel comparator FLEGT Forest Law Enforcement,
Governance and Trade FQD Fuel Quality Directive FT Fischer-Tropsch g Grams GAEC Good agricultural and environmental
condition GHG Greenhouse gas GSP+ Generalised System of Preferences GTAP Global Trade Analysis Project ha Hectare HD High Density HPO Hydrogenated Pyrolisis Oil HVO Hydrotreated Vegetable Oil H2 Hydrogen (referred
to in the context of liquid hydrogen as a fuel) IEA International Energy Agency IFPRI International Food Policy Research
Institute IPCC Intergovernmental Panel on Climate
Change JEC Consortium of JRC, EURCAR (the
European Council for Automotive R&D) and CONCAWE (the Oil Companies’
European Organisation for Environment, Health and Safety) JRC The Joint Research Centre of the
European Commission LCFS Low Carbon Fuel Standard LD Low Density LPG Liquified Petroleum Gas LULUCF Land-use, land-use change
and forestry Mha Million hectares MJ Megajoule (106 joules) MS Member States of the European Union MSA Mean Species Abundance Mt Million tonnes Mtoe Million tonnes of oil equivalent NGO Non-governmental Organisation NREAPS National Renewable Energy
Action Plans OWL Other wooded land PBL Netherlands Environmental Agency PE Partial Equilibrium models Pg Petagram (1015 grams) Ppm Parts per million RED Renewable Energy Directive REDD+ The United Nations Collaborative
Programme on Reducing Emissions from Deforestation and Forest Degradation in
Developing Countries RES Renewable Energy RSB Round Table for Sustainable
Biofuels TJ Terajoule (1012
joules) UNFCCC United Nations Framework
Convention for Climate Change UNEP United Nations Environment Programme UK United
Kingdom of Great Britain and Northern Ireland US United States of America WCMC World Conservation
Monitoring Centre WTO World Trade Organisation 10. Annex
I – Consultation and use of external expertise 10.1. Summary
of responses from indirect land-use change "pre-consultation" The Commission sought public views on possible approaches
to address indirect land-use change in a "pre-consultation" exercise
between 14 June and 31 July 2009. The approaches considered were: (a)
Extend to other commodities/countries the
restrictions on land-use change that will be imposed on biofuels consumed in
the European Union. (b)
International agreements on protecting
carbon-rich habitats. (c)
Do nothing. (d)
Increase the minimum required level of
greenhouse gas savings. (e)
Extending the use of bonuses. (f)
Additional sustainability requirements
for biofuels from crops/areas whose production is liable to lead to a high
level of damaging land-use change. (g)
Include an indirect land-use change
factor in greenhouse gas calculations for biofuels. (h)
Other policy elements that respondents
may wish to raise. A total of 71 responses were received[126], 28% from EU
Member States and third countries, 6% from public bodies, and the rest from
organisations among which 45% were from industry and businesses, 13% from
non-governmental organisations, and 8% from research institutions. Most industry, farmers' associations and third countries supported
either no action or dealing with indirect land-use change through wider policy
action (either through international action on protection of high carbon stock
land and/or extending sustainability criteria to all agricultural commodities).
Most NGOs and an industrial stakeholder from the non-biofuel sector supported
the inclusion of greenhouse gas emissions associated with indirect land-use
change within the existing legislative scheme for determining the greenhouse
gas emission for biofuels. Certain NGOs and research institutions supported to
lower the 10% target or set a maximum contribution conventional biofuels.
Member States were divided on this issue. 10.2. Analytical
work In order to base its work on the best
available scientific evidence, the Commission services launched a number of
analytical exercises and a review of existing literature on the subject of indirect
land-use change during 2009 and 2010[127],[128]. The International Food Policy Institute
(IFPRI) was commissioned to look at the "Global trade and environmental
impact study of the EU biofuels mandate". The final report was
published in October 2011[129],
and has used the most up to date biofuel demand estimates up to 2020 as
outlined by the Member States in the national renewable energy action plans[130]. In addition, the study aims
to provide a better characterisation of the uncertainty associated with the
crop specific indirect land-use change emission values. A number of other studies were launched by
several Commission services: –
"Impacts of the EU biofuel target on
agricultural markets and land-use: a comparative modelling assessment", by
the Institute for Prospective Technological Studies of the EC's Joint Research
Centre; –
"The impact of land-use change on
greenhouse gas emissions from biofuels and bioliquids–
an in-house review conducted for DG Energy; –
"Indirect land-use change from increased
biofuels demand – comparison of models and results for marginal biofuels
production from different feedstocks" by the EC's Institute for Energy of
the Joint Research Centre; –
"Biofuels- a new methodology to estimate
GHG emissions from global land-use change" by the Institute for
Environment and Sustainability and Institute for Energy of the EC's Joint
Research's Centre[131].
10.3. Summary
of responses from main indirect land-use change consultation Following the
publication of the relevant analytical work in July 2010, the Commission
launched a second public consultation exercise between 30 July and 31 October
2010. This sought views on whether this analytical work provided a good basis
for determining the significance of indirect land-use change; whether action
was required, and if so what course of action would be appropriate. It also set
out a reduced number of possible policy approaches: (a)
Take no action for the time being, while monitoring impacts including trends in certain key
parameters and, if appropriate, proposing corrective action later; (b)
Take action by encouraging greater use of
some categories of biofuel; (c)
Discourage the use of some categories of
biofuel by: –
increasing the minimum greenhouse gas saving
threshold for biofuels; –
imposing additional sustainability requirements
on certain categories of biofuel. –
attributing a quantity of greenhouse gas
emissions from indirect land-use change to all biofuels that use land. (d)
Take some other form of action. A total of 145
responses were received[132]
comprising 9% from EU Member States and third
countries, 2% from public bodies, and the rest from organisations among which
60% were from industry and businesses, 23% from non-governmental organisations,
and 6% from research institutions. Responses fell into two broad groups. Most
respondents from industry, farmers' associations and third countries considered that the
analytical work did not provide a good basis for determining the significance of
indirect land-use change. They considered that no further action specific to biofuel policy should
be taken, although many supported action on international agreements towards the
protection of land with high carbon stock. On the other hand, most NGOs and a few industrial
stakeholders from non-biofuel sectors considered that further action was needed and supported the
inclusion of the indirect land-use change emissions within the existing greenhouse gas
emission calculation. A number of other respondents recognised that action may
be needed, favouring a variety of other measures, in particular options aimed
at limiting the amount of conventional biofuels while increasing the share of
advanced biofuels, which were mainly favoured by NGOs and certain industrial
stakeholders. Member States
were divided on this. 10.4. External
expertise Following this public
consultation, in November 2010 the JRC organised an expert consultation on
behalf of the Commission, which brought together world-recognised academics and
experts in the field. This consultation aimed at discussing the main
uncertainties related to the estimation of indirect land-use change[133]. In February
2010, the JRC organized a workshop on “The Effects of increased demand for
biofuels feedstocks on world agricultural markets and areas” with the
participation of leading experts and modellers from the EU and US. The workshop
discussed the results of the JRC modelling comparison study and reasons for
differences between models results. 11. Annex
II – The concept of Indirect and direct land-use change emissions The figure below illustrates in a highly
simplified manner how both direct and indirect land-use change takes place. || In this highly simplified example we look at a global agricultural system with only grazing land and forest land. At the outset there is no biofuel production on neither of the two land types. || The introduction of the biofuel production on grazing land leads to direct land-use change and may cause a loss or an increase of soil organic carbon. If the biofuel production is introduced onto the forest land, then the direct land-use change emissions may be large because there is a loss of forest biomass. Both these direct land-use change emissions are included in the overall greenhouse gas calculation of the produced biofuel (part of the sustainability criteria). || Macro-economic effects cause an increase in the value of grazing animals (i.e. meat), as less is now being produced. This creates an incentive to increase the production of meat. This can be done through yield increases (e.g. more animals per hectare) or conversion of more land to grazing land. The latter is indirect land-use change and causes in this example a loss of forest carbon stocks, since grazing has now expanded into forest areas. There is not a one-to-one relationship between the pasture/cropland area converted to biofuel and the area converted to new pasture/cropland. This relationship depends on the relative productivity of the old vs. new pasture/cropland, markets for co-products and to what extent the macro-economic pressure induces increased productivity and changes in consumption. Figure 8: Examples of direct and indirect land-use
changes arising as a consequence of a biofuel project (pictures from IEA
Bioenergy Berndes et.al. 2010) 12. Annex
III – Model limitations and uncertainties 12.1. Economic
models to estimate indirect land-use change emissions Modelling of
indirect land-use change is usually based on an assessment of what the
situation would have been expected to be without policies promoting biofuels
and comparing it with an assessment carried out with such policies. Such an
assessment can be carried out with more or less degrees of sophistication.
Simple calculations based upon the land area that biofuel demand will, assuming
displacement of all crops previously grown, represent the theoretical uppermost
boundary of the indirect land-use impacts. However, the actual land area
required is likely to be much lower due to constraints encouraging higher
levels of inputs, higher yields, the production of co-products along with biofuels
and the fact that the higher commodity prices will have a dampening effect on
other demand for the agricultural commodities. A first approximation of these
effects can be taken into account in a spreadsheet based approach to give an
improved understanding of the indirect land-use change although there are
limitations of such an approach. To further
improve the understanding of indirect land-use change it is necessary to make
use of economic models which take account of price impacts to alter the expected
behaviour of different parts of the economy. While spreadsheet-models are
typically more transparent with regard to key assumptions, and allow for
relations between parameters to be established without having regard to the
ability of the model to solve all the equations (as in a macro-economic
models), they do not capture important knock-on effects and feedbacks between
sectors, as macro-economic models do. There are
mainly two groups of macro-economic models that try to capture various
feedbacks between economic sectors. These are "Partial Equilibrium
models" (PE) and "Computable General Equilibrium models" (CGE).
The former typically covers certain sectors of the economy, which are most
relevant for the purpose of the modelling effort (e.g. agricultural markets, or
energy markets). The CGE models cover the whole economy, although often at a
coarser resolution than the specific sectors covered by PE models. Often,
various PE models are interlinked in order to capture broader effects. The models that
have been employed to estimate changes in domestic and international crop
acreage have not been used in a regulatory context until recently. Rather they
have been used to give policymakers an idea of the likely consequences of
changes in agricultural and trade policy. As a guide to policy development and
understanding, these models have proved very valuable in facilitating such
policy agreements. It is generally
accepted that economic models offer the best prospect of understanding the
scale and nature of indirect land-use change in terms of land area as well as
other impacts. However, it is also known that in reality several non-economic
factors influence what land-use change takes place and where it occurs. Some of
these drivers are related to political choices (land-use and agricultural
policy, land rights, etc.), others to institutional features (proximity to
infrastructure and markets, land-use legislation). Therefore conceptual
limitations will always remain. Models
typically base their assumption on existing correlations, which are based on
historical trends, and are therefore not capturing potential changes in
policies that may take place in the future. They therefore have limitations and
uncertainties which are further explained in the next section and which would
affect this impact assessment by making it difficult to assess the scale of the
indirect land-use change phenomena related to biofuels, and the effectiveness
and efficiency of the policy options. Furthermore, the risk of
contra-productive policies is dependent on the certainty of the science, and to
what extent the policies are based on these findings. 12.2. Why
models differ in their results Models attempt
to describe the reality in various ways. Fundamentally they vary in structure (CGE, PE, spreadsheet model etc.) coverage (geographical, economic sectors, time-span), data
(carbon stocks, transport demand etc.) and assumptions on economic cause and
effect relations (elasticities, future projections etc.). In the context
of estimating indirect land-use change emissions, there will always be a range
of unsolved issues, which influence the results considerably. Aspects where
modelling is based on uncertain assumptions, that however are likely to improve
over time as more resources are invested in data and statistical analysis, are;
the treatment of co-products[134],
existing yields[135],
marginal yields[136],
type of land converted[137],
classification of land[138],
elasticities[139],
carbon stock values[140],
and the modelling of pasture[141].
Aspects that
are likely to still be at the centre of dispute also in the longer run are; the
drivers of deforestation and the implied causality[142], food and feed consumption[143], and the technology response
to higher prices[144]. In addition,
the Literature review found that current macro economic models[145] are incapable of capturing a
number of factors, including the conversion of forest on peat-land which can
lead to considerable carbon emissions. However, the majority of such factors
would, if captured, reduce the estimated land-use change impact. These include
the allocation of all emissions to crop expansion, whereas deforestation can be
driven simultaneously by crop expansion and logging; rate of yield improvements
in response to increased demand for biofuels[146];
structural changes[147];
and, the protein content of various feeds and co-products, which is rarely
fully reflected[148].
In addition, the effects of the binding sustainability criteria for biofuels in
the Directives have not been taken into account. The literature
review also found that the geographical origin of the feedstock could also be a
significant variable in estimating the (indirect) land-use change impact of a
specific biofuel, i.e. whether the origin of the feedstock matters, as most
feedstocks can be produced in various regions of the world. However, none of
the modelling done so far has explored this variability, which may in fact not
be possible with today's models. Furthermore,
the modelling comparison study found that current models do not capture a
number of factors, which if taken into account, would increase the estimated
land-use change impact. These factors include emissions from the conversion of
peat-lands[149].
Moreover, apart from (indirect) land-use change emissions as discussed in this
report, models do not consider at least two additional sources of increased
emissions: the emissions from yield intensification due to crop price rises,
and the extra emissions from growing crops on marginal land rather than on
existing cropland. The uncertainty
in modelling indirect land-use change led the Commission to ask the JRC to
organize a workshop with leading indirect land-use change experts to try to
explore the main uncertainties. The workshop was held in November 2010[150]. The main topics being
discussed were; cropland allocation (including amount of expansion on
peatland), emissions factors (including emissions from peatland), yield
developments (including marginal yields), as well as the influence of reduced
food consumption and how pasture land interacts with cropland. The workshop
also discussed briefly the various policy options. In the last
results from the IFPRI-MIRAGE-BioF, there have been attempts to address these
concerns. Most notably: ·
Peatland emissions, both in terms of the
fraction of expansion taking place on peatland, as well as the emissions from
drainage. ·
Food demand is more inelastic ·
More co-products from cereals are able to
replace meals from oilseeds. However, a number of issues remain. ATLASS
highlights some of them in the final report, noting that "the model has tested the limits of the CES/CET (constant elasticity of supply/transformation) framework.
Both for co-products but also for land-use allocation, this conventional
modeling approach leads to too many simplifications. For co-products, the two
level CES approach has helped to reinforce the substitution at the protein
contents between meals and DDGS. Unfortunately, it has also forced to simplify
the representation of substitution between proteins and carbo-hydrates.
Similarly for land-use, even if our multi-nested CET has helped to capture
substitution between crops, it is not flexible enough to provide the right full
substitution matrix across crops. More important from a long run perspective,
it is not designed to capture issues such as multi-cropping and crop rotation,
both important issues for land-use considerations in a dynamic approach". It is also worth noting that the elasticity
of substitution and transformation were taken from the – limited – estimates in
the literature. The same elasticity of substitution was assumed for all crops
and all countries. To what extent these elasticities,
which are based on a limited numbers of sources in the literature, are valid,
and more importantly represents the likely development towards 2020 remains an
open question. In addition, the sensitivity analysis on
the demand side has been limited (rigid food demand, changes in price
elasticities of intermediate demands). This is key for certain crops, such as
cereals, as a large share of their additional production for biofuels is
assumed not to be replaced, which would result in higher indirect land-use
change impacts if this was not the case. 12.3. How can
indirect land-use change emissions estimates be negative? It is worth
noting that one of the most recent spreadsheet models (E4tech) suggest that the
likely indirect land-use change values for wheat ethanol range from -53 to -5.1
g/MJ (-5.1 g/MJ being identified as the most likely scenario). These surprising
results are identified in the report to be the result of the large credits
given to wheat ethanol by assuming that its co-products are replacing soy being
grown in Argentina and Brazil for animal feed purposes, to which the same study
allocates high land-use change emissions of around 55 g/MJ. However, some
of the assumptions made by this report around the rate of land abandonment in
the EU, the amount of carbon lost to foregone sequestration by this land,
assumptions on the yields on this land, and how much land would come from yield
increases and not area increases have been questioned. Although negative
results are rare, they can also be the results from estimating indirect
land-use change emissions with macro-economic models, as can be seen in the Monte Carlo results shown in Annex XI where sugar beet has negative indirect land-use
change emissions at one of the extremes of the probability distribution range. 13. Annex
IV – Results from estimating indirect land-use change with models 13.1. Total
indirect land-use change emissions To understand
the overall size of total indirect land-use change emissions associated with
the additional biofuel demand, various scenarios were modelled in 2009 using
the general equilibrium IFPRI-MIRAGE-BioF model and the partial equilibrium
AGLINK-COSIMO model. Although only the IFPRI-MIRAGE-BioF model was able to
directly estimate the overall greenhouse gas emissions resulting from the
modelled land requirements, the JRC have estimated these impacts through the
application of their newly developed Spatial Allocation Model. A further run of
the IFPRI-MIRAGE-BioF model was carried out in 2011 based on the 2020 biofuel
estimates submitted by the National Renewable Energy Action Plans (results are
included in this section for comparison and are described in more detail in the
baseline section of section 2). The total
estimated land requirements from the additional demand (i.e. change between the
projected 2020 levels with the policy and those presumed in the absence of no
biofuels policy) as well as the key assumptions are summarised in the table
below. MODEL || Change in volume [%]a || Bioethanol vs Biodieselb || (I)LUC area (Mha) || Total GHGc (Mt CO2eq) || Total GHGc (JRC SAM) (Mt CO2eq) IFPRI-MIRAGE-BioF (5.6%) || 2.3 || 87/13 || 0.8-1 || 107-118 || 201-248 IFPRI- MIRAGE- BioF (8.6%) || 5.2 || 60/40 || 2.8-3 || 435-454 || 731-806d IFPRI- MIRAGE-BioF (NREAPs) || 5.7 || 28/72 || 1.7-1.9 || 500d || 421-472d AGLINK-COSIMO (7%) || 5.4 || 35/65 || 5.2 || n/a || 1092 Table 12: Summary of overall GHG impacts
from (indirect) land-use change. Source: IFPRI-MIRAGE-BIOF and JRC. a
Change in conventional biofuels demand as % of EU 2020 transport fuel
consumption. b
Estimated mix of bioethanol vs biodiesel in the additional demand. c Emissions
from Business as Usual and Free trade scenarios. d
Including peat emissions. MODEL || IFPRI-MIRAGE-BioF (5.6%) || IFPRI- MIRAGE-BioF (8.6%) || IFPRI- MIRAGE-BioF (NREAPs) || AGLINK-COSIMO (7%) Original model || 18 || 31-33 || 24-50 || n/a JRC SAM || 34-41 || 53-58 || 32-36 || 63 Table 13: Summary of average greenhouse
gas impacts from (indirect) land-use change (gCO2eq /MJ). Source:
ATLASS and JRC The differences between the
IFPRI-MIRAGE-BioF runs are driven by the different
bioethanol/biodiesel composition and overall mandate volume as described in
table 13. In addition, peat land emissions were underestimated compared to
current values. 13.2. Marginal
indirect land-use change values for feedstocks To better
understand whether indirect land-use change emissions are similar across
different biofuel pathways or differ between feedstocks, a modelling assessment
of the indirect land-use change caused by individual types of biofuels was
carried out. The IFPRI-MIRAGE-BioF model was used to determine values for
additional volumes of biofuels based on different feedstocks. The resulting
indirect land-use change values from this were as shown in table 14 below. || Ethanol || Biodiesel || Sugar Cane || Sugar Beet || Maize || Wheat || Palm oil || Rapeseed oil || Soy oil || Sunflower oil IFPRI- MIRAGE-BioF (5.6%) || 18 || 16 || 54 || 37 || 46 || 53 || 75 || 60 IFPRI- MIRAGE- BioF (NREAPS) || 14 || 7 || 10 || 14 || 54 || 54 || 56 || 52 JRC-SAM (from IFPRI-MIRAGE-BioF NREAPs || 22-26 || 5-6 || 13-14 || 16-17 || 21-45 || 43-53 || 44-53 || 51-59 Table 14: Summary of IFPRI-MIRAGE-BioF
marginal (indirect) land-use change emissions (gCO2eq /MJ). Source:
IFPRI-MIRAGE-BioF and JRC. As it can be
seen from the table, models tend to allocate different (indirect) land-use
change emissions to different feedstocks. This is one of the reasons that
average emissions in table 14 vary according to the overall feedstock
composition predicted in the final mix. At the crop level, we see much larger
differences due to the fact that some parameters have been altered between the
studies (i.e. new yields, increased displacement potential between crops,
better replacement ratio for co-products), as well as the method used to
compute the crop LUC now being based on a much larger share. In addition, it is
worth noting that the authors consider some of the new key assumptions
affecting the cereal crops to be strongly optimistic, including very high
yields for wheat in the EU, and maize in the US and Brazil. This is important
as these variations have not been included in the sensitivity analysis but play
a significant role in the estimation of indirect land-use change emissions. The difference between biodiesel crops and
ethanol crops has increased. For nearly all crops, except palm oil due to the
increase in peat emissions, the estimated indirect land-use change emissions
have been reduced, when one compares results from IFPRI-MIRAGE-BIOF from 2010
and from 2011. Estimated indirect land-use change emissions for soybean has
been cut by half while maize has been cut by five. However, the ranking between
feedstocks remains the same (sugars being the best and oilseeds the worst). 14. Annex
V – The IFPRI-MIRAGE-BioF model: assumptions and results 14.1. The
principles of the MIRAGE model The MIRAGE model was initially developed at
CEPII. This section summarizes the features of the standard version relevant
for this study. MIRAGE is a multisector, multiregion Computable General
Equilibrium (CGE) Model for trade policy analysis. The model operates in a sequential
dynamic recursive set-up: it is solved for one period, and then all variable
values, determined at the end of a period, are used as the initial values of
the next one. In order to evaluate the impact of public policies regarding
first generation biofuels, ATLASS has developed an extended version of the
global CGE MIRAGE, nicknamed IFPRI-MIRAGE-BioF, by improving the standard
version in several directions. A detailed description of this version of the
model is provided in Bouët et al. (2010)[151]
and in other studies (Al-Riffai, Dimaranan, and Laborde 2010a)[152]. The MIRAGE model relies on the Global Trade
Analysis Project (GTAP) database for global, economy-wide data. The GTAP
database combines domestic input-output matrices which provide details on the
intersectoral linkages within each region, and international datasets on
macroeconomic aggregates, bilateral trade, protection, and energy. We started
from the latest available database, GTAP 7, which describes global economic
activity for the 2004 reference year in an aggregation of 113 regions and 57
sectors (Narayanan and Walmsley, 2008). The database was then modified to
accommodate the sectoral changes made to the IFPRI-MIRAGE-BioF model.
Twenty-three new sectors were carved out of the GTAP sector aggregates -- the
liquid biofuels sectors (an ethanol sector with four feed-stock specific
sectors, and a biodiesel sector), major feedstock sectors (maize, rapeseed,
soybeans, sunflower, palm fruit and the related oils), co- and by-products of
distilling and crushing activities, the fertilizer sector, and the transport
fuels sector. This process did not consist of a simple disaggregation of parent
sectors, but required a full rescaling of agricultural production data
according to FAO statistics on quantity and prices, harmonization of prices on
substitutable homogenous goods such as biofuels or vegetable oils, and
bottom-up reconstruction of production costs for biofuel sectors and crushing
sectors for oilseeds. Goods are consumed by final consumers
(public and private agent) and firms or are exported to foreign markets. The
final consumption demand system is represented through a LES-CES that is
recalibrated each year along the baseline to reproduce consistent income and
price elasticities. Imported goods are differentiated from domestic goods
following the Armington assumption, which allows us to distinguish different
levels of market integration. The sector sub-utility function used in MIRAGE is
a nesting of four CES functions. In this study, Armington elasticities are
drawn from the GTAP 7 database and are assumed to be the same across regions.
But a high value of Armington elasticity, i.e. 10, is assumed for all
homogenous sectors (single crops, single vegetal oils, ethanol). For biodiesel,
we assume the same elasticity as that for other fossil fuels. From the supply side in each sector, the
production function is a Leontief function of value-added and intermediate
inputs: one output unit needs for its production x percent of an aggregate of
productive factors (labor, unskilled and skilled; capital; land and natural
resources) and (1 – x) percent of intermediate inputs. The intermediate inputs
function is a nested system of CES function of all goods: it means that
substitutability exists between two intermediate goods, depending on the
relative prices of these goods. This substitutability is constant and at the
same level for any pair of intermediate goods. Particular
care has been paid in the final and intermediary consumption nesting to the
substitution possibilities of similar products on the one side (vegetable oils,
oilseed meals, ethanol feedstocks) and to the rigidity relative to certain
inputs in the production chain (vegetable oil to produce biodiesel, sugar raw
products to produce refined sugar, etc). Similarly, in
the generic version of the model, value-added is a constant elasticity of
substitution (CES) function of unskilled labor, a logistic bundle of land and
intensification inputs (fertilizer for crops, feedstuff for livestock), natural
resources, and of a CES bundle of skilled labor and capital. This nesting
allows the modeler to introduce less substitutability between capital and
skilled labor than between these two and other factors. In other words, when
the relative price of unskilled labor is increased, this factor is replaced by
a combination of capital and skilled labor, which are more complementary. Moreover, the model relies on many features
specifically introduced to adequately represent the effects of biofuel
policies. In particular, it includes a detailed description of the insertion of
biofuel in the consumption chain, a modeling of binding incorporation mandates,
and a representation of co-products production for the bioethanol sector by
type of pathway (wheat, corn, sugar beet) and for the four oilseed processing
sectors that have been explicitly introduced (rapeseed, soybean, sunflower, and
palm fruit). Factor endowments are fully employed. The
only factor whose supply is constant is natural resources. Capital supply is
modified each year because of depreciation and investment. Growth rates of
labor supply are fixed exogenously. Skilled labor is the only factor that is
perfectly mobile. Installed capital and natural resources are sector specific.
New capital is allocated among sectors according to an investment function.
Unskilled labor is imperfectly mobile between agricultural and nonagricultural
sectors according to a constant elasticity of transformation (CET) function:
unskilled labor’s remuneration in agricultural activities is different from
that in nonagricultural activities. This factor is distributed between these
two series of sectors according to the ratio of remunerations. To capture the interactions between
biofuels production and land-use change, the model has specific features
focusing on a decomposition of land-use and land-use change dynamics. Land resources are differentiated between different
agro-environmental zones (AEZ). The possibility of extension in total land
supply to take into account the role of marginal land (and potential lower
yield) is also introduced. The modelling of land-use change captures both the
substitution effect involved in changing the existing land allocation to
different crops and economic uses, and the expansion effect of using more
arable land for cultivation. Land allocation decision across crops, pasture and
managed forest is based on a three level nested CET structure. Land extension
into pristine environment is based on an elastic land supply function,
depending on the cropland price and having an elasticity decreasing with the
amount of suitable agricultural land potentially available. With regard to yields projections, the
ATLASS consortium based them on the 2010 new baseline
of the Aglink-Cosimo used in the Agricultural Outlook of DG AGRI's forecast. Table 15: Yields. tonnes per Ha. 2020. Baseline. Source: IFPRI-MIRAGE-BIOF
(2011) (Sugar cane and sugar beet according to region where it grows; i.e.
sugar beet in the EU and sugar cane in Brazil) 14.2. More
results from IFPRI-MIRAGE-BioF The global effects of changing land-uses
are shown in an informative manner in table below, where the amount of
displaced land per TJ of biofuels is shown. The columns are indicating: || Explanation Column 1 || Amount of land needed to produce 1 TJ of biofuel using the specific feedstock indicated. Column 2 || Amount of land changed to energy crops as a result of using 1 TJ of the biofuel feedstock. Column 3 || Change in total amount of cropland. This amount of land will have to be converted from managed or unmanaged (natural) land Column 4 || Change in total amount of pasture Column 5 || Change in total amount of exploited land (i.e. land not previously used like primary forest) || Column 1 || Column 2 || Column 3 || Column 4 || Column 5 || Scenario Feedstock || Net Energy Crops || Net Cropland || Pasture || Net Exploited Land Biodiesel_PalmFruit || || || || || EU27 || || 0.35 || 0.08 || -0.01 || 0.03 World || 3.89 || 5.74 || 1.97 || -0.91 || 0.12 Biodiesel_Rapeseed || || || || || EU27 || 4.42 || 2.94 || 0.51 || -0.10 || 0.14 World || 10.91 || 11.72 || 3.90 || -1.39 || 0.64 Biodiesel_Soybean || || || || || EU27 || 0.14 || 0.77 || 0.10 || -0.02 || 0.03 World || 11.61 || 11.41 || 3.86 || -1.50 || 0.76 Biodiesel_Sunflower || || || || || EU27 || 4.28 || 2.53 || 0.33 || -0.06 || 0.09 World || 13.59 || 12.42 || 4.90 || -2.04 || 0.71 Ethanol_Beet || || || || || EU27 || 5.34 || 2.23 || 0.17 || -0.05 || 0.02 World || 5.75 || 2.97 || 0.41 || -0.13 || -0.13 Ethanol_Cane || || || || || EU27 || || 0.01 || 0.03 || -0.01 || 0.00 World || || 2.70 || 1.48 || -0.88 || 0.15 Ethanol_Maize || || || || || EU27 || 2.40 || 1.13 || 0.08 || -0.02 || 0.01 World || 6.52 || 3.69 || 0.88 || -0.40 || 0.00 Ethanol_Wheat || || || || || EU27 || 3.27 || 1.77 || 0.17 || -0.04 || 0.03 World || 7.64 || 4.99 || 1.39 || -0.54 || 0.10 Table 16: Global effects of changing
land-use in amount of displaced land per TJ of biofuels The table above shows that of the
additional land needed to produce 1 TJ of biofuels (column 1), only a fraction
is needed in terms of additional cropland (column 3) and out of the additional
cropland an even smaller fraction comes from natural areas (primary forests and
grassland – column 5). Note e.g. that 1 TJ of maize does not take any new
unused land into production, while sugar beet returns 0.11 ha back to natural
areas. Rapeseed and soy bean and sunflower are the crops that takes the
most new land into production, with aroud 0.8 ha per TJ of biofuel. However, it is not only the amount of land
that is important, but also what amout of carbon stock is on that particular
land. These aspects are shown in the figure below, indicating the balance
between amount of land needed by certain crops (column 3 above), and the
average carbon stock of that land. Figure
9: Amount of land converted,
and the corresponding carbon stock. Source: IFPRI-MIRAGE-BIOF The figure below is showing the relation
between cropland expansion (column 3), compared to additional exploited land
(natural land – primary forest and grassland – column 5). One can observe that
most of the expansion takes place into managed forest and pasture, since the
changes in these two land-uses explains the gap between exploited land and crop
land. Figure
10: Cropland extension versus
exploited land extension [Km2]. Source: IFPRI-MIRAGE-BIOF 15. Annex
VI – Fossil fuel comparator The Joint Research Centre (JRC) has
calculated the expected fossil fuel comparator (FFC) used in this impact
assessment. It is estimated to be 90.3 g/MJ. The main assumptions are set out
below. The extraction emissions for existing
oilfields gradually rise with time because the energy needed to extract the
crude increases. The resulting average production emissions from fields
supplying EU in 2020 are expected to reach 6.8 gCO2/MJ crude, or 7.2 g/MJ final
fuel, ignoring any potential effect of the Fuel Quality Directive. Updated figures from Concawe suggest that
the production (upstream emissions) greenhouse gas intensity (not including
transport) is 5.6 g/MJ final fuel. The 2020 gasoline value would become
87.6-5.6+7.2 = 89.2 and the 2020 diesel value becomes 89.1-5.6+7.2 = 90.7 If the diesel/gasoline split is 75:25 the
weighted average is 90.3 g/MJ final fuel. The assumed baseline for calculating the
contributions to the Fuel Quality Directive is set at 88.3 g/MJ which is a
weighted average of fossil fuels used in the EU in 2008. 16. Annex
VII – Trends in land-use- availability and expansion globally 16.1. Land-use change emissions Land-use change and the use of fossil fuels
are the main contributors to anthropogenic greenhouse gas emissions. The figure
below shows the accumulated anthropogenic carbon emissions to the atmosphere
since 1850. Land-use change emissions – primarily associated with the
conversion of forests to agricultural land – have contributed roughly one-third
during this period, with land-use change's share of the total diminishing over
the last decades. Figure 11: Accumulated anthropogenic carbon emissions to the
atmosphere since 1850. Source: IEA Bioenergy Berndes et.al. 2010 In the next
decades it is foreseen that a higher world population and standards of living
will lead to increasing demand for food, feed, energy and fibre from the
earth's ecosystems. Global agricultural production must increase by 70 percent
– almost 100 percent in developing countries – by 2050 to feed the world’s
forecast 9.1 billion people, and current levels of investment are not enough to
reach these levels. FAO estimates[153]
that net investments to agriculture must top USD 83 billion per year – up
roughly 50 percent from current levels – to meet future demand[154]. Furthermore, in developing
countries, one fifth of the increase in production will come from increase in
agricultural land and four fifths from improved productivity on existing land.
The increased use of biofuels in the EU adds to this existing demand for
agricultural commodities[155].
16.2. Trends in land availability
globally The extent to
which land availability is limited in various regions of the world is much debated.
Figure 5 below[156]
depicts the harvested area in different regions of the world. Compared to 1981
the harvested land has significantly declined in Europe, CIS and North America, thus suggesting that there would be low carbon stock land available[157]. The time-series
is divided into three segments, where distinct trends can be observed: ·
1961 – 1981 (20 years) harvested area increases
rapidly with roughly 150 million hectare (Mha) globally (on average 7.4 Mha per
year). ·
1981 – 2001 (20 years) harvested area increases
slowly with 56 Mha globally (on average 2.8 Mha per year). ·
2001 – 2008 (7 years) harvested area increases
rapidly with 95 Mha (13.6 Mha per year) It is worth to
note that the rapid increase in harvested area seen for the period 2001 – 2008
has not been continued into 2009, when harvested area actually decreased
by -1.2 Mha. However, this might be the result of the economic crises unfolding
in 2008, as farmers responded to reduced demand. Countries taking part in the
global agricultural trade had the steepest decrease with e.g. 2.6 Mha less area
harvested in North America. Figure 12: Globally harvested area from 1961 to 2009 With regard to the EU, DG AGRI has
estimated that the EU will continue to reduce agricultural area with around 0.5
million ha each year. || 2010 || 2012 || 2014 || 2016 || 2018 || 2020 Cereals || 56.3 || 57.1 || 57.4 || 57.8 || 58.0 || 58.3 of which EU-15 || 34.3 || 34.8 || 34.9 || 35.1 || 35.3 || 35.4 of which EU-12 || 22.0 || 22.3 || 22.5 || 22.7 || 22.8 || 22.9 Soft wheat || 23.0 || 23.3 || 23.4 || 23.7 || 23.8 || 24.0 Durum wheat || 2.9 || 2.9 || 2.8 || 2.8 || 2.8 || 2.8 Barley || 12.4 || 12.8 || 12.8 || 12.7 || 12.7 || 12.7 Maize || 8.1 || 8.3 || 8.5 || 8.7 || 9.0 || 9.2 Rye || 2.6 || 2.6 || 2.6 || 2.6 || 2.5 || 2.5 Other cereals || 7.3 || 7.3 || 7.3 || 7.2 || 7.2 || 7.1 Oilseeds || 10.9 || 10.9 || 11.0 || 11.0 || 11.0 || 11.1 of which EU-15 || 5.9 || 6.0 || 6.0 || 6.0 || 6.0 || 6.0 of which EU-12 || 5.0 || 5.0 || 5.0 || 5.0 || 5.0 || 5.0 Rapeseed || 6.9 || 7.0 || 7.1 || 7.1 || 7.2 || 7.3 Sunseed || 3.7 || 3.6 || 3.5 || 3.5 || 3.4 || 3.4 Soyabeans || 0.4 || 0.3 || 0.3 || 0.4 || 0.4 || 0.4 Sugar beet || 1.4 || 1.4 || 1.4 || 1.4 || 1.4 || 1.3 Protein crops || 1.1 || 1.1 || 1.0 || 1.0 || 1.0 || 1.0 Total selected arable crops || 69.8 || 70.5 || 70.8 || 71.2 || 71.4 || 71.6 Total utilized agricultural area || 188.3 || 187.2 || 186.1 || 185.0 || 183.9 || 182.8 Table 17: Area under arable crops in the EU, 2009-2020 (million
hectare) Source DG AGRI There are also
estimates of marginal land[158]
that is available for bioenergy production. Hoogwijk (2003; 2004 cited in
Hennenberg et.al (2010)) estimates that between 430 and 580 Mha of land is
marginal and can potentially be used. However, Okoinstitut in the report
"Sustainable Biomass Production from degraded lands" (Hennenberg et.al
(2010)) concluded that it is very challenging to quantify the amount of
degraded land available, while at the same time suggesting that the estimates
done by Hoogwijk are at least 10-times too high. 16.3. Trends in agricultural land
expansion Although land
is available it is not necessarily the case that the marginal supply of
agricultural crops is planted on marginal land. However, it is clear that a
significant amount of land is available in certain areas of the world, although
it is difficult to govern the proper use of these land areas. In fact, recent
studies suggest that tropical forests were the primary sources of new
agricultural land in 1980-90s (i.e. over 80% of land coming from forests across
the tropics, see figure 6 below), with various studies highlighting a
significant role of soy and cattle ranging, and palm into the expansion of
agricultural land into the Amazon and South East Asia respectively[159]. However, since 2005
deforestation rates in the Amazon have been going down significantly[160]. Figure 13: The origins of new agricultural land,
1980–2000, of which 80% took place in tropical areas. Bars show the average
proportion of land sources comprising new agricultural land in major tropical
regions (Source: Gibbs et.al. 2010). Further to the
spatial limited availability of low-carbon stock land in some areas, the lack
of effective protection of forests and carbon rich areas is another factor that
allows damaging indirect land-use change to take place. If conversion of carbon
rich areas were to be limited, the risk of damaging indirect land-use change
would be minimized. This is particularly applicable to forests and wetlands
although significant carbon stocks can be lost from other land types, in
particular grasslands. However, progress towards such a situation is slow. It
should also be noted that as long as these areas represent an important source
for increasing total agricultural production, reducing such conversions is
likely to increase global agricultural commodity prices. There are
highly relevant lessons to be learned from the long standing policy and
research work done on limiting deforestation. The Commission has taken part in
various efforts to reduce deforestation, through development cooperation, trade
policies and international negotiations, most notably through the UNFCCC, where
the discussions of the REDD mechanism (Reduction of Emissions from
Deforestation and forest Degradation) is the centrepiece. Extensive information
on activities aimed at reducing deforestation can be found in the Communication
on "Addressing the challenges of deforestation and forest degradation to
tackle climate change and biodiversity loss"[161], which notes that the most
important direct cause of forest destruction is changes in land-use to pursue
profitable alternative uses of land, such as obtaining commodities, while the
most important underlying cause of deforestation is ineffective governance,
linked to poorly enforced land-use policies and uncertain land tenure regimes. 17. Annex
VIII – Interactions between existing legislation and indirect land-use change 17.1. Existing EU legislation The Renewable Energy Directive and the Fuel
Quality Directive, are at the origin of the debate on potential indirect
land-use change caused by EU biofuels policy. So far there is no existing
regulatory legislation either at EU level or at Member State level that
effectively addresses the issue, although the legislation does include a
measure in order to increase the amount of land available for the cultivation
in form of a greenhouse gas bonus to biofuel produced on severely degraded and
heavily contaminated land. Measures which are taken to control
land-use change for biofuels directly, for example to protect forest or
grassland in the EU, will have limited impact on indirect land-use change
emissions. This is because while they may prevent this land from being
converted to agriculture, they do not limit the total demand for agricultural
commodities and the extra demand might be supplied where it is most cost effective
to do so. In some cases that is likely to be from conversion of new land to
agriculture, while in others it can come from yield increases. The
sustainability criteria adopted under the Directives hinders biofuels that come
from land where damaging land-use change has taken place, also referred to as
'direct land-use change', to be used in the EU through the requirement to
calculate the carbon stock change of the land-use change. The estimated
indirect land-use change emissions coming from the implementation of the
Directives are shown in the section laying down the baseline for this Impact
Assessment. 17.1.1. The Renewable Energy Directive The Renewable Energy Directive requires
Member States to achieve jointly a 20% renewable energy share of total energy
consumption over all sectors by 2020. Specifically in the transport sector,
Member States are required to achieve a minimum of 10% renewable energy by
2020. In line with overall EU energy policy, the aim of these targets is
threefold, i.e. to reduce greenhouse gas emissions, to promote the security of
energy supply, to promote technological development and innovation and provide
opportunities for employment and regional development, especially in rural and
isolated areas. Based on the demand estimate figures
supplied by the Member States (NREAPs), it seems that the vast majority of the
10% transport target (around 9%) will be met through the use of conventional
biofuels. In contrast, bioliquids are expected to play a small role in
contributing towards the 20% target. 17.1.2. The Fuel Quality Directive The Fuel Quality Directive requires fuel
suppliers to achieve a 6% reduction in greenhouse gas intensity of the energy
they supply by 2020 compared to a 2010 baseline. Blending of biofuels is
expected to be the main route of compliance, where volumes needed are expected
to be similar to those estimated by Member States to comply with the 10%
renewable energy target under the Renewable Energy Directive. Proper accounting
of emissions is necessary for the Fuel Quality Directive to work in the desired
way, i.e. providing incentives for fuels that have less greenhouse gas
intensity. 17.2. Other policies 17.2.1. Agricultural policies Agriculture
policies are essential in influencing the indirect land-use change impact.
There are basically three theoretical possibilities to mitigate indirect
land-use change. First, bringing back into cultivation idle low-carbon stock
land. Second, agricultural policy could target improvements in land
productivity beyond the rate which would have prevailed otherwise. Third,
agricultural policy can provide incentives for the production of biofuels with
low indirect land-use change emissions. The Common
Agricultural Policy (CAP) is currently in its final phase of almost entirely
decoupling its support to farmers from production. Farmers have to ensure that
their land remains in good agricultural and environmental condition (GAEC).
This provision aims at ensuring that agricultural land which becomes idle
remains available for production in the future. Furthermore, Member States are
required to report on changes in grassland and to take measure against, if
conversion occurs. Although the effectiveness of this requirement varied across
all Member States, this provision has helped to slow down and reverse the loss
of grassland in the EU. The
introduction of direct payments was accompanied by a reduction of guaranteed
prices to farmers. Lower agricultural prices contributed to less input use.
Lowering cereal prices in the EU to world market level also led to a strong
increase in cereal use, especially for feed, substituting imports of
feedstuffs. All in all, EU
farmers are free to decide whether, how much and for what purpose they grow
crops. The decoupling of support significantly improved their responsiveness to
changing market signals. A re-introduction of coupled support, e.g. for biofuel
crops would not be in line with WTO rules. It is through
the Rural Development Policy of the CAP where the EU offers financial support
to farmers who engage, amongst others, into increasing their competitiveness,
including by improving yields, or into taking agri-environmental commitments. 17.2.2. Environmental policies EU
environmental legislation, including the new EU biodiversity strategy, is aimed
to reduce environmental degradation, including, damaging land-use change
impacts in the EU. However, it is neither intended nor possible for
biodiversity actions alone to prevent the impacts of other policies. For that
to happen biodiversity considerations must be integrated and mainstreamed into
the development and implementation of all national an EU policies related to
natural resource management, such as agriculture, food security, forestry,
fisheries and energy, as well as spatial planning, transport, tourism trade and
development. At international level, the recently adopted post-2010 global
biodiversity Strategic Plan of the Biodiversity Convention may help reducing
the loss of carbon and biodiversity rich habitats including forests. As part of
its 2008 Communication on tropical deforestation, the Commission committed to
studying the impact of EU consumption of imported food and non-food commodities
(e.g. meat, soy beans, palm oil, and metal ores) that are likely to contribute
to deforestation. Such work could lead to considering policy options that
reduce indirect land-use change effects. With regard to forest policy in the EU, it
is the competence of Member States to implement sustainable forest management.
A number of MS have tight restrictions on deforestation. The EU currently
contains 5 % of the world's forests and EU forests have continuously expanded
for over 60 years, although recently at a lower rate. EU Forests and other
wooded land (OWL) now cover 155 million ha and 21 million ha, respectively,
together more than 42 % of EU land area[162].
Most of EU forests, including those under continuous management, have also
grown in terms of wood volume and carbon stock, thus effectively removing CO2
from the atmosphere. According to FAO data (Forest Resource
Assessment 2010), EU27 has reported a net increase of almost 7.3 million ha of
forest area (+5%) between 1990-2000, and only 2.5 (+3%) between 2000-2010. This
includes both afforestation and natural reforestation, e.g. on abandoned land.
The area of protected and protective forests in the EU has also increased
during the last decade[163]. However, the overall spending on
afforestation measures through rural development programmes under Common
Agricultural Policy in the EU has declined. According to current target
figures, Member States expect that about 0.9 Mha of new forests will be
established during the current programming period through Rural Development
Programmes. Currently, an average of 0.5 Mha of forest
burn in the EU annually with associated emissions, most affected are Spain,
Italy, Greece, France and Portugal. 17.2.3. Climate policies Climate change legislation may also help
reducing the risk of indirect land-use change. At the 16th
Conference of the Parties to the UNFCCC ("COP16" ) in 2010 it was
agreed to support developing countries to better protect their tropical forests
by establishing a global mechanism to Reducing Emissions from Deforestation and
Forest Degradation (REDD)[164].
The UNFCCC decision affirms "(...) that all Parties should collectively
aim to slow, halt and reverse forest cover and carbon loss" and therefore
encourages them to "(...) address drivers of deforestation". International accounting rules applicable
to annex 1 parties for land use, land use change and forestry (LULUCF) for the
post 2012 period were agreed at the 17th Conference of the Parties
to the UNFCCC ("COP17") in Durban in December 2011. In order to
transpose this decision into EU law, the Commission is
proposing as a first step robust, common accounting, monitoring and reporting
rules for LULUCF within the EU and compulsory LULUCF action plans in Member
States. The Commission propose that accounting for croplands and grasslands
become mandatory within the EU, while left voluntary at international level.
This will ensure that perennial energy crops grown on agricultural lands will
enter accounting[165].
As a second step it may consider how LULUCF could be taken into account in the
EU’s GHG emission reduction commitment. LULUCF
accounting can reduce land-use changes by internalising the environmental cost
of the related emissions at the national level, therefore making activities
that increase such emissions less attractive, but only once a target is agreed
for the sector in the context of a second step. However, it should be noted
that LULUCF accounting alone is not likely to be effective at controlling LUC
emissions. LULUCF accounting acts at the national level and costs are born by
the government, while land-use decisions are taken at the local level by land
managers. In the absence of dedicated policy instruments, the GHG cost of LUC,
even if internalised, will be passed on to the government budget. In addition,
LULUCF accounting provides only a price signal commensurate to the average
abatement cost at the national level. The GHG abatement cost of biofuels is
significantly higher than the prevailing carbon price, and demand for biofuels
is highly inelastic due to binding targets and mandates. Therefore, the
incentive to convert land is likely to be generally higher than the
disincentive provided by LULUCF accounting. Therefore, while accounting for
LULUCF ensures that land-use emissions are monitored the actual land-use
changes and resulting collateral environmental impacts may not be effectively
reduced. 17.2.4. Trade policies Changes
in biofuels import tariffs affect the land-use change impact of
the overall biofuels mandate because it induces changes in the composition of
the supply of biofuels, from different origins and feedstocks. While EU
tariffs are low on biodiesel imports, they are relative high on
bioethanol. Reducing these tariffs would consequently have
significant effects on bioethanol imports and production, less so for
biodiesel. Imported bioethanol is mainly produced from tropical sugarcane,
a feedstock with very high direct emission savings. Though
the indirect land-use emissions may be higher than for EU-produced
bioethanol feedstocks (wheat, sugar beet), the net emission
savings are higher for imported sugarcane ethanol. The risks
of indirect land-use changes in producing countries vary. The EU has
effective protection of carbon rich areas and land-use expansion is strictly
controlled. Land-use control systems may be less effective in developing
countries. Sustainability Impact Assessments of trade agreements can
help to detect and understand these potential negative side effects,
including on tropical deforestation. They enable the EU to
develop policies and flanking measures to enhance the sustainability and
reduce land-use emissions of biofuels exports to the EU. The 2003 EU action plan for Forest Law
Enforcement, Governance and Trade (FLEGT) sets out a process and a package of
measures to address the problem of illegal logging and related trade. The
cornerstone of the Action Plan is the establishment of voluntary FLEGT
Partnership Agreements (VPA) between the EU and timber producing countries,
aimed at stopping illegal logging. In December 2005, the Council adopted a
Regulation (No 2173/2005) that establishes a licensing scheme and a mechanism
to verify the legality of timber imports into the EU from partner countries.
The VPAs together with the recently approved EU Timber Regulation (No 995/2010)
will discourage unregulated and unsustainable exploitation of forests and thus
address one of the drivers of deforestation and forest degradation. In
addition, the negotiations provide an opportunity to challenge the legal
framework, particularly with respect to land conversion and environmental
sustainability of forest management and where the frameworks are not clear or
judged insufficient by stakeholders. 17.2.5. Development policies Under its
development policy, the EU is committed to increasing expenditure on demand-led
agricultural research, extension and innovation by 50% by 2015.[166] Focus is placed on
"ecologically efficient agricultural intensification for smallholder
farmers" that improves equitable and sustainable access to resources,
including land, water, (micro) credit and other agricultural inputs with the
aim of reducing food insecurity and poverty. The projected annual budget in
this area is a minimum of € 87 million per year between 2011 – 2013. Although
not aimed only at yield increases, development policy reduces indirect land-use
change by improving agricultural productivity, especially by stepping up
research to improve the productivity and sustainability of agriculture in
developing countries. 17.2.6. Research policies The research
into feedlots and animal diets in order to maximise the use of biofuel
co-products to feed European livestock will reduce the imports of protein rich
animal feed, notably soya (Weightman[167]
et.al. 2010), which can significantly influence the indirect land-use change.
Weightman et.al. calculate that today's diets for pigs, poultry and ruminants
can reduce land-use change in South America by 64 – 138 g/MJ of biofuel
produced from wheat in the EU. The average credit is 82 g/MJ and if higher
usage is made possible through nutritional research the credit can nearly
double (Weightman et.al. 2010). Since the inception of 7th Framework
Programme for Research ("FP7"), the Commission has issued calls for
demonstration projects that put particular emphasis on biofuel production from
lignocellulosic biomass and addresses practically all value chains from
sustainable biomass resources to a final marketable biofuel that meets the
thresholds laid down in the Directives for 2018, i.e. minimum 60% reduction
compared to fossil fuels for new installations. Sustainability is a key issue
in all calls. Under FP7 the Commission has supported projects for sustainable
biofuels in excess of €150 million. At the end of 2007, the European Commission
proposed the European Strategic Energy Technology Plan (SET-Plan)[168] targeting a strategic
approach to technology development and deployment in order to ensure the
achievement of energy objectives. Bioenergy was covered by the SET Plan and it
was accompanied by "A Technology Roadmap for Bioenergy" presenting
the fundamental roadmaps which serve as a basis for strategic planning and
decision making[169].
The main tool for the implementation of the SET Plan are the Industrial
Initiatives, public-private initiatives led by industry, aiming to accelerate
industrial research and innovation at the EU and Member States level[170]. Most relevant for biofuels
is the European Industrial Bioenergy Initiative (EIBI), which is characterised
by innovative technologies and high-risk investments aiming to bring new
technologies onto the market for the first time. The focus (related to
biofuels) is primarily on second-generation biofuel production from
lignocellulosic biomass and algae[171]. In addition, and in line with the
priorities identified by the SET Plan, at least four lignocellulose-to-biofuels
demonstration projects at pre-commercial scale are potentially eligible for
co-financing under the so-called NER 300 funding programme, which provides
financing for commercial-scale carbon capture and storage (CCS) and innovative
renewables technology demonstration projects from 300 million allowances
reserved in the new entrants reserve of the EU Emissions Trading System[172]. 18. Annex
IX – biofuels and related industries baseline tables || 2002 || 2003 || 2004 || 2005 || 2006 || 2007 || 2008 || 2009 Austria || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.2 || 0.2 || 0.3 Belgium || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.3 || 0.4 Bulgaria || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Cyprus || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Czech Republic || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 || 0.1 || 0.1 Denmark/Sweden || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 || 0.2 || 0.2 Estonia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Finland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.2 France || 0.3 || 0.3 || 0.3 || 0.4 || 0.7 || 0.8 || 1.6 || 1.8 Germany || 0.4 || 0.6 || 0.9 || 1.5 || 2.4 || 2.6 || 2.5 || 2.3 Greece || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 Hungary || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 Ireland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Italy || 0.2 || 0.2 || 0.3 || 0.4 || 0.4 || 0.3 || 0.5 || 0.7 Latvia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Lithuania || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 Luxemburg || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Malta || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Netherlands || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.3 Poland || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.2 || 0.3 Portugal || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.2 || 0.2 || 0.2 Romania || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.0 Slovakia || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.0 || 0.1 || 0.1 Slovenia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Spain || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.2 || 0.2 || 0.8 UK || 0.0 || 0.0 || 0.0 || 0.0 || 0.2 || 0.1 || 0.2 || 0.1 TOTAL || 1.0 || 1.3 || 1.7 || 2.9 || 4.4 || 5.2 || 7.0 || 8.2 Figure 14: European biodiesel production by country
(in Mtoe). Source: EBB || 2003 || 2004 || 2005 || 2006 || 2007 || 2008 || 2009 || 2010 Austria || 0.0 || 0.1 || 0.1 || 0.1 || 0.3 || 0.4 || 0.6 || 0.5 Belgium || 0.0 || 0.0 || 0.0 || 0.1 || 0.3 || 0.6 || 0.6 || 0.6 Bulgaria || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.2 || 0.4 || 0.4 Cyprus || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Czech Republic || 0.0 || 0.0 || 0.2 || 0.2 || 0.2 || 0.2 || 0.3 || 0.4 Denmark || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 || 0.1 || 0.2 Estonia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 Finland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.2 || 0.3 || 0.3 France || 0.5 || 0.5 || 0.5 || 0.7 || 0.7 || 1.8 || 2.3 || 2.3 Germany || 0.9 || 1.0 || 1.7 || 2.4 || 3.9 || 4.8 || 4.7 || 4.5 Greece || 0.0 || 0.0 || 0.0 || 0.1 || 0.4 || 0.5 || 0.6 || 0.6 Hungary || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.2 || 0.2 || 0.1 Ireland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 Italy || 0.4 || 0.4 || 0.7 || 0.8 || 1.2 || 1.4 || 1.7 || 2.1 Latvia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 Lithuania || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 Luxemburg || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Malta || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Netherlands || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.5 || 0.9 || 1.2 Poland || 0.0 || 0.0 || 0.1 || 0.1 || 0.2 || 0.4 || 0.5 || 0.6 Portugal || 0.0 || 0.0 || 0.0 || 0.1 || 0.2 || 0.4 || 0.4 || 0.4 Romania || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.3 || 0.3 Slovakia || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.2 || 0.2 || 0.1 Slovenia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 Spain || 0.0 || 0.1 || 0.1 || 0.2 || 0.5 || 1.1 || 3.3 || 3.7 Sweden || 0.0 || 0.0 || 0.0 || 0.0 || 0.2 || 0.2 || 0.2 || 0.3 UK || 0.0 || 0.0 || 0.1 || 0.4 || 0.6 || 0.7 || 0.5 || 0.5 TOTAL || 1.8 || 2.0 || 3.8 || 5.5 || 9.3 || 14.4 || 18.9 || 19.8 Figure 15: European biodiesel capacity by country (in Mtoe). Source: EBB || 2004 || 2005 || 2006 || 2007 || 2008 || 2009 Austria || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 Belgium || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 Bulgaria || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Czech Republic || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 Denmark || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Finland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 France || 0.1 || 0.1 || 0.1 || 0.3 || 0.5 || 0.6 Germany || 0.0 || 0.1 || 0.2 || 0.2 || 0.3 || 0.4 Hungary || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 || 0.1 Ireland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Italy || 0.0 || 0.0 || 0.1 || 0.0 || 0.0 || 0.0 Latvia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Lithuania || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Netherlands || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Poland || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 Romania || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Slovakia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.1 Slovenia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Spain || 0.1 || 0.2 || 0.2 || 0.2 || 0.2 || 0.2 Sweden || 0.0 || 0.1 || 0.1 || 0.1 || 0.0 || 0.1 UK || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 TOTAL || 0.3 || 0.5 || 0.8 || 0.9 || 1.4 || 1.9 Figure 16: European bioethanol production by country (in Mtoe). Source: ePure || 2005 || 2006 || 2007 || 2008 || 2009 || 2010 Austria || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 Belgium || 0.0 || 0.0 || >0.0 || 0.3 || 0.3 || 0.3 Bulgaria || 0.0 || >0.0 || >0.0 || >0.0 || >0.0 || >0.0 Czech Republic || >0.0 || >0.0 || >0.0 || 0.1 || 0.1 || 0.1 Denmark || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Finland || 0.0 || 0.0 || >0.0 || >0.0 || >0.0 || >0.0 France || 0.4 || 0.4 || 0.7 || 0.9 || 1.0 || 1.0 Germany || 0.2 || 0.4 || 0.4 || 0.6 || 0.6 || 0.6 Hungary || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 || 0.1 Ireland || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || >0.0 Italy || 0.2 || 0.2 || 0.2 || 0.2 || 0.2 || 0.2 Latvia || >0.0 || >0.0 || >0.0 || >0.0 || >0.0 || >0.0 Lithuania || >0.0 || >0.0 || >0.0 || >0.0 || >0.0 || >0.0 Netherlands || 0.0 || >0.0 || >0.0 || >0.0 || >0.0 || 0.3 Poland || 0.0 || 0.1 || 0.1 || 0.1 || 0.2 || 0.2 Romania || 0.0 || 0.0 || >0.0 || >0.0 || >0.0 || >0.0 Slovakia || 0.0 || 0.0 || 0.1 || 0.1 || 0.1 || 0.1 Slovenia || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 || 0.0 Spain || 0.1 || 0.3 || 0.3 || 0.3 || 0.3 || 0.3 Sweden || 0.1 || 0.1 || 0.1 || 0.2 || 0.2 || 0.2 UK || 0.0 || 0.0 || >0.0 || >0.0 || >0.0 || 0.2 TOTAL || 1.0 || 1.5 || 2.1 || 2.9 || 3.1 || 3.6 Figure 17: European bioethanol capacity by country (in Mtoe). Source: ePure (x 1000 tonnes) || Soya || Rape || Sunflower || Palm-kernel || Linseed || Castor || Maize germ || Grape pips || Palm || TOTAL Austria* || 0 || 136 || 30 || 0 || 1 || 0 || 0 || 0 || 0 || 167 Belgium || 21 || 356 || 0 || 0 || 136 || 0 || 49 || 0 || 0 || 562 Bulgaria ** || 0 || 0 || 147 || 0 || 0 || 0 || 6 || 0 || 0 || 153 Cyprus || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Czech Republic** || 8 || 284 || 15 || 0 || 1 || 0 || 0 || 0 || 0 || 308 Denmark || 9 || 214 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 223 Estonia** || 0 || 37 || 0 || 0 || 0 || 0 || 1 || 0 || 0 || 38 Finland || 2 || 104 || 4 || 0 || 0 || 0 || 0 || 0 || 0 || 110 France || 49 || 1329 || 472 || 0 || 0 || 0 || 49 || 0 || 0 || 1899 Germany || 643 || 3185 || 88 || 0 || 53 || 0 || 13 || 0 || 0 || 3982 Greece* || 63 || 22 || 14 || 0 || 0 || 0 || 4 || 0 || 0 || 103 Hungary* || 0 || 110 || 315 || 0 || 0 || 0 || 17 || 0 || 0 || 442 Ireland || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Italy || 259 || 18 || 157 || 0 || 6 || 0 || 50 || 12 || 0 || 502 Latvia** || 2 || 30 || 1 || 0 || 0 || 0 || 0 || 0 || 0 || 33 Lithuania** || 0 || 60 || 1 || 0 || 0 || 0 || 0 || 0 || 0 || 61 Malta || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Netherlands || 572 || 336 || 127 || 0 || 0 || 0 || 0 || 0 || 0 || 1035 Poland** || 1 || 693 || 10 || 0 || 2 || 0 || 0 || 0 || 0 || 706 Portugal* || 199 || 35 || 63 || 0 || 0 || 0 || 4 || 0 || 0 || 301 Romania** || 37 || 70 || 222 || 0 || 0 || 0 || 0 || 0 || 0 || 329 Slovakia** || 4 || 71 || 39 || 0 || 0 || 0 || 0 || 0 || 0 || 114 Slovenia** || 0 || 5 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 5 Spain || 549 || 35 || 361 || 0 || 1 || 0 || 17 || 0 || 0 || 963 Sweden* || 0 || 100 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 100 UK || 106 || 678 || 0 || 0 || 4 || 0 || 23 || 0 || 0 || 811 EU-27 || 2524 || 7914 || 2065 || 0 || 204 || 0 || 233 || 12 || 0 || 12947 Figure 18: 2008 production of crude vegetable oils and fats. Source: Fediol *estimate**Source:
Oilworld (x 1000 tonnes) || Soyabeans || Rapeseed || Sunflower-seeds || Palm kernel || Linseed || Castor || Maize germs || Grape pips || Palm || TOTAL Austria || 0 || 332 || 72 || 0 || 4 || 0 || 0 || 0 || 0 || 408 Belgium || 109 || 869 || 0 || 0 || 367 || 0 || 103 || 0 || 0 || 1448 Bulgaria** || 1 || 12 || 350 || 0 || 0 || 0 || 12 || 0 || 0 || 375 Cyprus || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Czech Republic** || 47 || 720 || 35 || 0 || 2 || 0 || 0 || 0 || 0 || 804 Denmark || 55 || 465 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 520 Estonia** || 0 || 91 || 1 || 0 || 0 || 0 || 2 || 0 || 0 || 94 Finland || 10 || 275 || 9 || 0 || 0 || 0 || 0 || 0 || 0 || 294 France || 280 || 3157 || 1060 || 0 || 0 || 0 || 102 || 0 || 0 || 4599 Germany || 3364 || 7705 || 198 || 0 || 150 || 0 || 27 || 0 || 0 || 11444 Greece || 351 || 54 || 37 || 0 || 0 || 0 || 8 || 0 || 0 || 450 Hungary* || 0 || 270 || 750 || 0 || 0 || 0 || 35 || 0 || 0 || 1055 Ireland || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Italy || 1787 || 45 || 388 || 0 || 15 || 0 || 120 || 89 || 0 || 2444 Latvia** || 13 || 74 || 2 || 0 || 0 || 0 || 0 || 0 || 0 || 89 Lithuania** || 0 || 148 || 3 || 0 || 0 || 0 || 0 || 0 || 0 || 151 Malta || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Netherlands || 2898 || 799 || 289 || 0 || 0 || 0 || 0 || 0 || 0 || 3986 Poland** || 6 || 1732 || 24 || 0 || 6 || 0 || 0 || 0 || 0 || 1768 Portugal || 1170 || 90 || 157 || 0 || 0 || 0 || 8 || 0 || 0 || 1425 Romania** || 217 || 179 || 532 || 0 || 1 || 0 || 0 || 0 || 0 || 929 Slovakia** || 22 || 180 || 92 || 0 || 1 || 0 || 0 || 0 || 0 || 295 Slovenia** || 0 || 13 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 5 Spain || 3026 || 91 || 903 || 0 || 3 || 0 || 35 || 0 || 0 || 4058 Sweden* || 0 || 250 || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 250 UK || 559 || 1654 || 0 || 0 || 10 || 0 || 48 || 0 || 0 || 2271 EU-27 || 13915 || 19205 || 4902 || 0 || 559 || 0 || 500 || 89 || 0 || 39162 Figure 19: 2008 crushing of oilseeds. Source: Fediol (x 1000 tonnes) || Wheat || Barley || Grain maize || Rye and maslin || Cereals total (including rice) || Sugar beet || Rape || Sunflower Austria || 1523 || 835 || 1891 || 195 || 5144 || 3083 || 171 || 71 Belgium || 1928 || 451 || 754 || 3 || 3221 || 4569 || 42 || 0 Bulgaria || 4000 || 815 || 1273 || 15 || 5273 || 0 || 231 || 1301 Cyprus || 15 || 40 || 0 || 0 || 57 || 0 || 0 || 0 Czech Republic || 4358 || 2003 || 890 || 178 || 7832 || 3038 || 1128 || 61 Denmark || 5996 || 3421 || 0 || 245 || 10200 || 2011 || 637 || 0 Estonia || 346 || 380 || 0 || 39 || 879 || 0 || 136 || 0 Finland || 887 || 2171 || 0 || 42 || 4261 || 559 || 140 || 0 France || 38325 || 12880 || 15300 || 130 || 70000 || 33146 || 5562 || 1676 Germany || 25190 || 12288 || 4527 || 4325 || 49748 || 25550 || 6307 || 57 Greece || 1830 || 280 || 2352 || 37 || 4820 || 902 || || 16 Hungary || 4396 || 1033 || 7543 || 75 || 13571 || 708 || 565 || 1306 Ireland || 951 || 1089 || 0 || 0 || 2384 || 45 || 29 || 0 Italy || 6341 || 1049 || 7878 || 12 || 15892 || 3308 || 51 || 280 Latvia || 1036 || 265 || 0 || 162 || 1663 || 0 || 209 || 0 Lithuania || 2100 || 858 || 24 || 208 || 3806 || 682 || 416 || 0 Luxembourg || 91 || 54 || 3 || 7 || 189 || 0 || 18 || 0 Malta || 0 || 0 || 0 || 0 || 0 || 0 || 0 || 0 Netherlands || 1402 || 310 || 245 || 11 || 2089 || 5735 || 12 || 0 Poland || 9790 || 3984 || 1707 || 3968 || 29827 || 10849 || 2497 || 4 Portugal || 110 || 76 || 594 || 18 || 1057 || 137 || 0 || 14 Romania || 5205 || 1183 || 8035 || 36 || 14934 || 685 || 572 || 1083 Slovakia || 1538 || 676 || 988 || 57 || 3330 || 899 || 387 || 187 Slovenia || 137 || 71 || 303 || 2 || 533 || 262 || 10 || 0 Spain || 4797 || 7400 || 3479 || 181 || 17833 || 4089 || 29 || 876 Sweden || 2284 || 1677 || 0 || 219 || 5249 || 2406 || 302 || 0 UK || 14379 || 6769 || 0 || 36 || 22036 || 8330 || 1951 || 2 EU-27 || 138 954 || 62057 || 57782 || 10202 || 295828 || 110992 || 21399 || 6934 Figure 20: Harvested production of some of the main
crops, in 1 000 tonnes, 2009. Source: Eurostat 19. Annex
X – Impacts on biodiversity According to the central scenario of IFPRI-MIRAGE-BioF model, the
estimated additional cropland requirements globally amount to 1.7 Mha, mainly
taking place in Commonwealth of Independent States, Sub-Saharan Africa and Brazil regions. New cropland is allocated by IFPRI-MIRAGE-BioF estimating changes in the
economic use of land, i.e. among forestry, cropland and pasture uses. The
results of the
IFPRI-MIRAGE-BioF study showed that this new cropland
is taken from pasture (42%), managed forest (39%), primary forest (3%) and
savannah and grassland (16%), which will have biodiversity and wider
environmental impacts. A qualitative estimation of the impacts to biodiversity of land-use
changes calculated by IFPRI-MIRAGE-BioF was made by the JRC[173] using the Mean Species Abundance (MSA) values provided by the
Global Biodiversity Model (GLOBIO 3)[174].
This model is built on a set of equations which link environmental drivers and
biodiversity impacts. The environmental drivers used as input for GLOBIO3 are
land-use change (including forestry), climate change, N deposition, harvesting,
energy use etc. Biodiversity is described in GLOBIO3 on the basis of the remaining
mean species abundance (MSA) of original species, relative to their abundance
in pristine or primary vegetation, which are assumed to be not disturbed by
human activities for a prolonged period. MSA is therefore considered as the
indicator for biodiversity. The following table was extracted from
Alkemade et al, 2009 and adapted to IFPRI-MIRAGE-BioF lands uses to evaluate
MSA values for the land-use transitions (and LU classes) in IFPRI-MIRAGE-BioF
scenario. IFPRI-MIRAGE-BIOF Land-use class || Sub-category || Description || MSALU Pasture || || Grasslands where wildlife is replaced by grazing livestock || 70% Managed Forest || Secondary forest || Areas originally covered with forest or woodlands, where vegetation has been removed, forest is re-growing or has a different cover and is no longer in use || 50% Agroforestry || Agricultural production intercropped with (native) trees. Trees are kept for shade or as wind shelter || 50% Primary Forest || || Minimal disturbance, where flora and fauna species abundance are near pristine || 100% Scrublands and grasslands || || Grassland or scrubland-dominated vegetation (for example, steppe, tundra, or savannah) || 100% Cultivated and managed areas || || High external input agriculture, conventional agriculture, mostly with a degree of regional specialization, irrigation-based agriculture, drainage-based agriculture*. || 10% Table 18: Land-use classes used to
determine mean species abundance (MSA) * The JRC assumes land management factor
for cropland as “medium or high input with manure” in its calculations (JRC
report n.24483). For consistency the same assumption is taken here. For example, according to the MSA values in the table, a transition
from pastureland (MSA 70%) to cropland (MSA 10%) will cause a loss of 60%of MSA
on top of the 30% already lost from the conversion from natural land to
pastureland.. An estimation of the land-use transition biodiversity loss in the
additional croplands which may result from IFPRI-MIRAGE-BIOF scenario has been
calculated with a weighted average of MSA values for IFPRI-MIRAGE-BIOF land-use
changes as: Where: MSAi = Mean Species Abundance of land-use type i %i = % of land conversion according to IFPRI-MIRAGE-BIOF results MSAca = Mean Species Abundance of cultivated and managed areas Considering that 42% of new cropland will come from pasture, 39%
from managed forest, 3% from primary forest and 16% from savannah and
scrublands, this will results in a “weighted” MSA value of 68%, and the
transition to cropland will cause 58% decrease in the MSA index in affected
areas. In addition to this there will also be indirect losses when part or all
of the pasture and managed forest that was converted is moved elsewhere into
natural areas. Considering this, total loss could build up to 90% when all
pasture and forestry is moved into natural land (gradual land-use conversion in
several steps). These results are just preliminary rough estimates, and more
research is needed to provide a qualitative and more precise assessment in
particular of the indirect effects. This result, in line with the conclusions of GLOBIO3 study, shows
that the extensive use of bioenergy crops will increase the rate of loss of
biodiversity, and often the GHG reduction from biofuels production are
insufficient to compensate for the losses due to land-use change (Van Oorschot
et al., 2010). 20. Annex
XI – Monte-Carlo analysis of indirect land-use change emissions estimates ATLASS created 1000 baselines, and then
performed simulations, using 1000 set of seven parameters. The parameters are
drawn from a log uniform distribution, centered on the default value of the
model, and the range of values is defined based on a literature review of potential
meaningful figures (see CARB 2011, for a discussion on elasticity values[175]). Key elements of the
parameter distribution are displayed in Table 16. As
shown by the ratio average/median, nearly all distributions of the sample have
a right tail-feature[176]
driven by the log uniform assumption of the probability used to build these
samples. The same set of parameters is used for assessing the indirect land-use
change emissions uncertainty of the full mandate as well as for each individual
crop. || Shifter in the share of extension occurring in primary forest || Shifter in intermediate demand price elasticity of agricultural inputs || Ratio between yield on new cropland and average yield || Elasticity of substitution between land and other factors (factor intensification) || Elasticity of substitution between key inputs (feedstuff or fertilizer) and land (input intensification) || Elasticity of transformation of land (intermediate level) || Land extension elasticity Average || DC || 0.99 || 1.18 || 0.75 || 0.07 || 0.11 || 0.30 || 0.02 DV || 0.99 || 1.18 || 0.75 || 0.07 || 0.20 || 0.30 || 0.05 Median || DC || 0.91 || 1.21 || 0.75 || 0.06 || 0.08 || 0.25 || 0.01 DV || 0.91 || 1.21 || 0.75 || 0.04 || 0.15 || 0.25 || 0.04 Maximal Value || DC || 1.81 || 1.83 || 0.99 || 0.18 || 0.29 || 0.74 || 0.04 DV || 1.81 || 1.83 || 0.99 || 0.33 || 0.59 || 0.74 || 0.17 Minimal value || DC || 0.46 || 0.47 || 0.50 || 0.01 || 0.02 || 0.09 || 0.00 DV || 0.46 || 0.47 || 0.50 || 0.01 || 0.04 || 0.09 || 0.01 Standard Deviation || DC || 0.39 || 0.41 || 0.13 || 0.05 || 0.08 || 0.18 || 0.01 DV || 0.39 || 0.41 || 0.13 || 0.06 || 0.16 || 0.18 || 0.04 Table 19: Range of parameters for Monte Carlo analysis. Source: IFPRI-Mirage-BioF Monte Carlo parametersNote: DC=Developed
countries. DV=Developing countries Before discussing the list of parameters
and their expected effects, one needs to indicate how the draws are done. A
first solution would have been to consider that the value of a parameter for
each sector (if relevant) and each region and AEZ is independent of the value
for other sectors/regions. For instance it would have implied that the value
used for the elasticity of land transformation into European AEZ are independent
or that the level of potential factor intensification in the wheat sector in
the US is uncorrelated with the level for the corn market. In such a case, we
will have drawn for each parameter a specific value for each sector/region
combination considering systematic uncertainty. This approach is not followed.
Rather ATLASS considered that the key uncertainty is not about the exact value
for a country/sector and its correlation with other regions/sectors but about
the real location of the parameters distribution in the space of potential
value and that all sectors/regions are affected in a similar way. It implies
that we consider a perfect correlation between parameter values across sectors
and across countries (or group of countries). For instance, for each draw,
ATLASS shifted the value of a parameter, e.g. land elasticity of
transformation, for all developed countries in the same direction. All
developed countries will be able to relocate land more easily among crops (or
less) at the same time. However, the distribution for each parameter is
considered from other parameters: the shifter in demand behavior is drawn
independently from the value of fertilizer intensification parameter. If
parameter values would have been uncorrelated, a high elasticity in one region
may have been compensated by a lower in another. Consequently, for each draw
the world median would have been closer to the distribution median and the
overall land-use effects would have been closer to the median value (even if
the geographical pattern of the land-use will have been much more dispersed)[177]. ATLASS have chosen full
correlation, since we the key challenge for many parameters, e.g. yield price
response, is to know the change in average magnitude and not the question about
the correlation and the heterogeneity among countries/sectors. There are still
independent draws across parameters. A large yield response can still be
combined to a strong sensitivity of cropland extension to land prices. The
combination of effects among parameters is not biased in a way that will
increase/decrease the results dispersion. ATLASS selected seven parameters to study,
most of them – except the two first of the following list – focused on the
agricultural supply response and the extensification/intensification trade-off: ·
Shifter in the share of extension occurring in
primary forest, this coefficient multiplies[178]
the initial share of land extension taken place in primary forest in the
Winrock coefficient dataset. It does not affect the economic response of the
model and only modify the carbon release by unit of exploited land expansion: a
value above one will increase the share of primary forest and the carbon
release; ·
Shifter in intermediate demand price elasticity
of agricultural inputs, this shifter multiplies the price elasticity of
intermediate demand (by non primary sectors) for agricultural commodities. In
the model, the elasticity of substitution in the intermediate consumption
nested CES structure is recalibrated accordingly. A value above one implies
that processing sectors will release more easily inputs (crops or vegetable
oils) following the biofuel demand shock, and therefore reduces the LUC effect; ·
Ratio between yield on new cropland and average
yield: this parameter gives the marginal productivity of new hectare of
cropland compared to existing one. The expected direct effect is that reduced
yield will lead to larger requirement of new land to meet the additional crop
demand and will increase indirect land-use change emissions. However, more
complex effects take place in the model. Indeed, in the dynamic baseline,
assuming lower yield on marginal land leads to more land extension[179]. Since the “managed land”
supply elasticity in the model is not constant but decreases with the ratio
between used agricultural land and total suitable land for agriculture, the
large expansion in the baseline needed to compensate the low productivity of
new land reduces the remaining amount of available land in the baseline and
decreases the price elasticity of land expansion that prevails when the biofuel
scenario takes place. Therefore, the net effect of a low/high marginal yield is
ambiguous;[180] ·
Elasticity of substitution between land and
other factors (factor intensification) . This is a core parameter in the
endogenous yield response of the model; it shows how production can increase
through additional capital/labor use by unit of land. A larger value describes
a more flexible production system that will reduce the land-use change
effect(more intensification); ·
Elasticity of substitution between key inputs
(feedstuff or fertilizer) and land (input intensification). This is the other
driver of intensification, both in crop production and livestock sector, since
it allows to substituting land to inputs (fertilizer or feed). A larger value
is associated to a lower indirect land-use change emissions (more
intensification); ·
Elasticity of transformation of land
(intermediate level) among broad categories of agricultural production. A
larger value is associated to a lower indirect land-use change emissions since
increased production of energy crops can displace other agricultural production
before requiring new cropland (more land reallocation). ·
Land extension elasticity. This parameter
describes the land supply response – extension of managed agricultural land to
pristine environments – following an increase in cropland price. Even if this
value is not constant in the model, as discussed above it evolves the ratio
between used and available land for agriculture, the change in the Monte Carlo modifies the initial value and its path of evolution. A larger elasticity
value reinforces the indirect land-use change emission effect (more extension). ·
Last, for several parameters, ATLASS assumed
more uncertainty i.e. a more dispersed distribution for developing countries
parameters; it should lead to more dispersed land-use change for crops produced
in these regions (e.g. sugar cane) than for others. The parameters involved are
the intensification parameters (fertilizers, feed, and factors) and the land
extension elasticity. Similar crops with similar initial technology (share of
fertilizer in total cost) and production location (concentrated in developed
countries or in developing countries e.g. tropical crops) are expected to
display high correlation in land-use change in the Monte Carlo simulations.
ATLASS did not implement uncertainty of the household demand behavior, neither
uncertainty on substitution among subset of inputs (animal fat versus vegetable
oil for instance). Other aspects such as carbon stocks or direct saving
coefficients from the life cycle analysis are considered as known even if their
role in overall land-use change uncertainty competition should not be neglected
(see Plevin et al, 2010[181]). 21. Annex
XII – Potential for mitigating indirect land-use change emissions at project
level A number of measures can in theory be put
in place at the production site in order to prevent indirect land-use change. A
short description of these, which are being assessed under section 5, as well as
a comment on their potential is included in this section. 21.1. Increase use of unused/degraded
land Using land without provisioning services[182] that would be unlikely to be taken into production in the absence
of biofuel demand (i.e. typically land that either requires some form of
remediation prior to being used or where significant barriers exist). Expanding
production on unused land may lead to direct land-use change, but the latter
would be addressed by the current sustainability criteria and therefore
directed to those areas where effects are acceptable. Another potential way for mitigating
against (indirect) land-use change would be to increase the production of
biofuels in areas that are not in agricultural production and would be likely
to remain the same in the absence of intervention. This could be because of the
existence of some sort of "barriers" for this land to become into
cultivation (i.e. remediation is needed prior to cultivation or regulatory
barriers). In terms of potential, a recent project[183] highlighted that a total
35Mha of Imperata grassland (a type of invasive grassland of low biodiversity
value) could be available for the cultivation of palm oil in South East Asia,
with about 4Mha being available in Indonesia alone[184]. This would present a
significant opportunity for mitigating against indirect land-use change as the
baseline predicts a very significant part of the emissions to be associated
with peatland drainage. Opportunities for intensification through this method
would seem to be more limited in densely populated/exploited regions such as
the EU, although it would be difficult to estimate what the availability may
be. Moreover, the reduction in cropland
expected in the EU from 2010 to 2020 of around 5.5 million hectares. It is however
important to note that the potential to mitigate indirect land-use change
emissions through using more land in the EU (besides institutional, political
and social factors) would depend on the opportunity costs of and yields on this
land, as well as what would the carbon content and GHG balance of the land be
if not used for biofuel production. It should be noted that the source of the
steady expansion of the forest area of the EU in the past decades, and the
resulting forest sink amounting to 10% of the EU’s total GHG emissions, was the
reduction of agricultural land-use. A reduction or reversal of this trend would
involve significant carbon costs. As can be recalled from figure 5 in chapter
2.4; there has also been abandonment of agricultural land in the CIS countries
as well as in the US. 21.2. Increasing yields Increasing yields above projected future
trends which would not have happened in the absence
of biofuel demand. This would in theory suggest that the biofuel feedstocks are
produced without increasing the pressure on land and therefore limiting
indirect land-use change emissions. In this case, only the additional feedstock
production should be considered as meeting this requirement. Although there are certain technical
ceilings beyond which yields cannot be improved due to regional characteristics
(i.e. soil, climate, water availability, etc), the baseline yield data in table
3 suggests that there could be significant potential for certain crops to
improve their yields if the right policy tools were put in place. For example,
typical yields of certain crops such as palm fruit are assumed to be over 5
times higher in Indonesia/Malaysia compared to those achieved in sub-Saharan Africa. A project investigating the potential for
achieving yield improvements in the palm oil harvest in Liberia, found that an
average increase of 2t/ha of crude palm oil would be achievable through
improved mechanisation and the introduction of high yielding palm varieties.
This seems significant as a total of 100,000 ha are under palm oil cultivation.
Other countries in the region that show a similar potential could be Democratic
Republic of Congo, parts of Guinea, Cote d’Ivoire, Benin, Nigeria and Cameroon[185]. Although opportunities for intensification
through yield increases are less readily available in regions already achieving
yields above the world's average (i.e. EU and USA), improvements would seem
possible. For example, recent research trials suggest that oil seed rape and
winter wheat yields could be improved significantly through improved agronomy
in areas of EU where the yields are already very high[186]. 21.3. Integration of biofuel
production with non-bioenergy systems Integrating biofuels with non-bioenergy
production systems in ways that would lead to
higher land productivity. This integration would need to be additional to what
would have happened in the absence of the biofuel demand. In theory, it would seem possible to
mitigate against indirect land-use change through the integration of biofuel
production with non-bioenergy systems (i.e. land used for cattle farming). This
could present particularly significant opportunities for regions, such as Brazil, where extensive ranching areas are available. It is in this context that a recent
project looked at the potential for integration of cattle farming with the
production of sugarcane. This project reported a total potential of over 140Mha
of pasture land could be freed for sugarcane production in Brazil alone through this method (total current land used for sugarcane production is
estimated to be at 8Mha). This is significant as Brazil is one of the regions
where most cropland extension is likely to take place[187]. 21.4. Production costs A number of case studies have shown that
production costs will not be significantly higher under C2 requirements than
current practices although these are likely to be project specific. For
example, a review of previous case studies exploring the feasibility of the
development of oil palm on degraded land in Indonesia suggested that barriers
to extension could be more of a cultural/social nature as costs could in fact
be relatively low. In fact, total planting costs reported were between
500-1000€/ha cheaper when previous status was Imperata grassland compared to
secondary forest and heathland. Similarly, operating costs were up to 500€/ha
cheaper on Imperata grassland[188].
22. Annex
XIII – Assessment methodology The assessment of the policy options
described in section 4 can give raise to a range of environmental, economic,
social and wider impacts. The most relevant impacts are listed in table 20
below. Effectiveness || Minimise the impact of indirect land-use change on greenhouse gas emissions of biofuels, within the wider policy objectives of the targets that by 2020 at least 10% of transport fuels are renewable and that greenhouse gas intensity in road transport is reduced by at least 6% compared to 2010. || Achievement of the 10% target of the Renewable Energy Directive Environmental || Greenhouse gas emissions balance (quantified) Biodiversity Other (water, soil, air, etc) Economic || Costs (including production and administrative as appropriate). Financial investment stability Security of supply (energy and food/feed) Trade policies Social || Employment EU rural development Third countries: development objectives Commodity markets Other || Promoting technological development and innovation Coherence with existing GHG methodology Table 20: List of key impacts being considered in this assessment. 23. Annex
XIV – Possible response scenarios to reduced biofuel availability through the
introduction of additional requirements As explained in section 5, it is expected that options will
limit the availability of qualifying biofuel feedstocks, particularly biodiesel
feedstocks as these typically present both higher direct and estimated indirect
land-use change emissions. Assumptions are then needed to develop potential
scenarios as to where the additional contributions required to meet the
legislative targets will come from. Figure
21: Possible responses to a
limited supply of biofuels Therefore, the state of play on a number of
issues, such as vehicle compatibility with higher biofuel blends and relevant
R&D developments, as well as possible developments on greenhouse gas
emissions performance of conventional biofuels need to be considered. 23.1. Option
B: Possible extreme scenarios The main scenario considered here implies
that when the threshold is raised to 60%, the least efficient palm oil, soy and
rapeseed are excluded and not replaced by an increased share of the available
conventional biodiesel feedstocks (i.e. sunflower and palm oil with methane
capture) but by other available technologies (i.e. non-biofuel, bioethanol,
etc). As such, for analytical purposes the following extreme scenarios have
been considered, B1) Targets set out in the Directives
are met through fossil fuels and/or other renewable energies, without
increasing bioethanol blends, or the use of waste/residues biodiesel or 2nd
generation biodiesel beyond what is already necessary to reach the levels
estimated in the NREAPs. B2) Targets set out in the Directives
are met through higher bioethanol blends. No development of 2nd
generation biodiesel or waste/residues biodiesel beyond what is estimated in
the NREAPs. B3) Targets set out in the Directives
are met through increased use of waste/residues biodiesel and 2nd
generation biodiesel. Other renewable energies and bioethanol blends remains as
estimated in the NREAPs. The tables below outline the required
contribution from each technology for biofuels to maintain their contribution
to the Fuel Quality Directive on the baseline (5.4%). || Bioethanol [Mtoe] || Double counted biodiesel [Mtoe] || Electricity in road [Mtoe] Baseline || 6.7 || 1.8 || 0.7 B1 || 6.7 || 1.8 || 1.9 B2 || 18.3 || 1.8 || 0.7 B3 || 6.7 || 11 || 0.7 These scenarios are included for
illustrative purposes only. None of the scenarios above are considered
realistic as only the contribution from one technology at a time is increased,
giving rise to very significant additional requirements by 2020, notably: · B1: Increased contribution of 3.9 Mtoe of electricity in road would
be required by 2020. This would be the equivalent to deploying an additional 9
million electric cars by 2020[189].
For comparison, EU annual car sales are roughly 15 million per year. · B2: Increased levels of bioethanol, with the average bioethanol
blends in petrol cars increasing from 11% to 25%, as well as an increase in
bioethanol processing capacity in the EU. · B3: Increased contribution of 9.2 Mtoe of biodiesel coming from
waste/residues and 2nd generation. 23.2. Option
D: Possible extreme scenarios Based on the preliminary analysis of the
impacts of different indirect land-use change emission factors at feedstock
level, the overarching trend seems to be the need to replace the oilseeds that
fail to qualify with either other biodiesel feedstocks available and/or
increased the volume of bioethanol. In line with the broad approach as set out
at the beginning of this Annex, a number of extreme scenarios have been
considered. D1) Targets set out in the Directives
are met through fossil fuels and/or other renewable energies, without
increasing bioethanol volumes, or the use of waste/residues biodiesel or 2nd
generation biodiesel beyond what is already necessary to reach the levels
estimated in the NREAPs. D2) Targets set out in the Directives
are met through higher bioethanol volumes. No development of 2nd
generation biodiesel or waste/residues biodiesel beyond what is estimated in
the NREAPs. D3) Targets set out in the Directives
are met through increased use of waste/residues biodiesel and 2nd
generation biodiesel. Other renewable energies and bioethanol volumes remain at
estimated levels in the NREAPs. The tables
below outline the required contribution from each technology for biofuels to
maintain their contribution to the Fuel Quality Directive on the baseline
(-5.4%). However, the
contribution from biofuel technologies towards this target is much smaller
under this option than under all others, as the estimated indirect land-use
change emissions are not only included in the greenhouse gas emissions methodology
to check whether the biofuel feedstock in question would pass or not, but also
included in the reported carbon intensity reduction (using the 50th
percentile values of sensitivity). || Bioethanol [Mtoe] || Double counted biodiesel [Mtoe] || Electricity in road [Mtoe] Baseline || 6.7 || 1.8 || 0.7 D1 || 6.7 || 1.8 || 2.3 D2 || 26 || 1.8 || 0.7 D3 || 6.7 || 16 || 0.7 These scenarios are included for
illustrative purposes only, as only the contribution from one technology at a
time is increased, giving rise to scenarios with following requirements to 2020
of either, i.e. · increased contribution of 5.4 Mtoe of electricity in road would be
required by 2020. This would be the equivalent to deploying an additional 13
million electric cars by 2020[190]
for D1. For comparison, EU annual car sales are roughly 15 million per year. · D2 requires increased levels of bioethanol, with the average
bioethanol blends in petrol cars needed to increase accordingly (it would need
to increase from 11% to 32% for this tool alone to achieve given bioethanol
volumes). Bioethanol processing capacity in the EU would also need to increase
(current levels at 4.3 Mtoe) or the total amount of bioethanol imports. · increased contribution of 14 Mtoe would be needed from biodiesel
coming from waste/residues and 2nd generation for D3. 23.3. Vehicle
compatibility limitations with increasing usage of higher biofuel volumes in
current fleet In the context of
the policy options, the issues around the compatibility of higher ethanol
volumes with current fleet, as well as higher uptake of certain biodiesel
feedstocks such as palm oil, should be considered. These issues are discussed
in more detail in the baseline section in chapter 2. 23.4. Biodiesel
from non-conventional sources There are a number
of ways in which developments in research and development could help replacing
the production of biodiesel from feedstocks with estimated high indirect
land-use change impacts. These include bringing forward commercialisation
pathways, currently at pilot stage, for producing biodiesel from non-land using
feedstocks (i.e. algae, pyrolysis oil, etc). In addition, research into the
development of biodiesel from sugars is also ongoing and at pilot stage but
could come into the market before 2020. In addition, it may be possible to
increase the contribution from waste feedstocks, such as used cooking oil, for
which no technological developments are needed. 23.5. Potential
improvements in greenhouse gas performance According
to the assessment methodology being applied, the question of compliance with the
threshold is binary (i.e. whether biofuel feedstocks are in compliance or not),
independently of how close to compliance the feedstock might be. This is an important point for further consideration as in practice
biofuel producers can put in place measures to improve their greenhouse gas
emissions performance beyond the levels assumed in 2020. Although it is difficult to establish where
such "performance ceiling" of different biofuel feedstocks may be, it
is believed that emission saving levels of around 75%-80% can be reached if
certain agricultural practices (including using organic fertiliser), using
bio-methanol for trans-etherification (for producing biodiesel from vegetable
oil), and better processing technologies available today (including processing
the feedstocks using biomass) are used. 23.6. Potential
contribution to Fuel Quality Directive targets from non-RES sources The potential for
achieving additional reductions in fossil fuel carbon intensity is high. The
main areas of opportunity are at source (up-stream savings) and where the fuel
is consumed (combustion savings). During the Fuel Quality Directive
negotiations, the Commission estimated that a total of 300 Mt CO2eq
greenhouse gas emissions associated with global oil production from both
flaring and venting was possible, and that a third of these emissions could be
avoided through alternative uses of the gas at relatively low costs. Further
contributions are also possible from capture and storage of refinery emissions
depending on the development of this technology, and the supply of alternative
fuels other than biofuels (LPG, CNG and H2)[191]. 24. Annex
XV – Developing indirect land-use change emission factors from the results of
the Monte-Carlo analysis There are two basic approaches towards
developing indirect land-use change emissions factors according to the level of
disaggregation desired. The results in table below show the range of estimated
feedstock specific indirect land-use change emissions from the Monte Carlo analysis used for the assessment. All values are shown in grams of CO2-eq./MJ. [g/MJ] || 5th || 25th percentile || Central || 75th percentile || 95th Maize || 6 || 8 || 10 || 12 || 13 Sugar beet || 1 || 4 || 7 || 10 || 13 Sugar cane[192] || 7 || 13 || 15 || 18 || 26 Wheat - Not specified || 8 || 12 || 14 || 16 || 18 Wheat - Natural gas/CHP || 8 || 12 || 14 || 16 || 18 Wheat - Straw/CHP || 8 || 12 || 14 || 16 || 18 Palm oil || 47 || 51 || 54 || 57 || 60 Palm oil with methane capture || 47 || 51 || 54 || 57 || 60 Rapeseed || 28 || 45 || 55 || 66 || 81 Soybean || 38 || 50 || 56 || 61 || 74 Sunflower || 31 || 46 || 54 || 60 || 72 Table 21:
Estimated indirect land-use change emissions per feedstock. Source:
IFPRI-MIRAGE-BioF(2011) There are a number of ways for developing
ILUC factors from these estimates. For example, one way would be to directly
incorporate these numbers into the greenhouse gas emissions performance
calculation for biofuels. In that case, no further adjustment would be
required. However, it could also be argued that average values for each
specific crop group could be developed from weighted average of the feedstock
data. These are outlined in the table below. || 5th || 25th percentile || Central || 75th percentile || 95th Maize || 7 || 10 || 12 || 14 || 16 Sugar beet || 5 || 10 || 13 || 16 || 23 Sugar cane || 5 || 10 || 13 || 16 || 23 Wheat - Not specified || 7 || 10 || 12 || 14 || 16 Wheat - Natural gas/CHP || 7 || 10 || 12 || 14 || 16 Wheat - Straw/CHP || 7 || 10 || 12 || 14 || 16 Palm oil || 34 || 47 || 55 || 63 || 74 Palm oil with methane capture || 34 || 47 || 55 || 63 || 74 Rapeseed || 34 || 47 || 55 || 63 || 74 Soybean || 34 || 47 || 55 || 63 || 74 Sunflower || 34 || 47 || 55 || 63 || 74 Table 22:
Estimated indirect land-use change emissions per feedstock. Source:
Commission's calculations based on weighted average of IFPRI-MIRAGE-BioF (2011)
crop specific values. The impacts associated with both approaches
have been considered in the assessment of option D. 25. Annex
XVI – Potential effects of including the estimated indirect land use change
greenhouse gas emissions in the reporting of the greenhouse gas emission
savings of biofuels Option D
provides incentives for low indirect land-use change emissions biofuels in two
ways, as described in chapter 5.5.1 and chapter 5.5.9.1, through; a) the
exclusion of biofuels with too low savings, ILUC included, compared to the
fossil fuels they replace and b) the incentives for biofuels with lower ILUC
due to the accounting and reporting by fuel suppliers under the Fuel Quality
Directive which is likely to bring about a significant price differentiation in
favour of low-ILUC transport fuels. This is because these biofuels will
contribute much more than others to the attainment of a supplier's obligation
to reduce the greenhouse gas intensity of the fuels it supplies. As it was only
possible to analyse the effectiveness of the first element using the
methodology in this Impact Assessment, the effectiveness of the second element
is explored further in this Annex. The discussion here focuses on the market
incentives (element b) and thus ignores the exlusion of certain feedstocks
resulting from element a. The allocation of feedstocks is therefore rather done
through cost-minimisation of fuel supplier's expenditure on fuels as required
to achieve the greenhouse gas emissions reductions mandated by the Fuel Quality
Directive. This exercise obviously depends on a range of assumptions, of which
some of them are listed in a footnote[193].
As the methodology used in this analysis is very sensitive to variation in
feedstock price (i.e. those feedstocks with lowest carbon abatement costs are
prioritised witrhout looking at any other variables and their availability is
assumed to be in most cases unlimited), conclusions presented here are only
intended as an illustration of the likely impacts. The feedstocks are limited to the list know
from chapter 5 of the Impact Assessment (i.e. with the same estimated GHG
performance once the estimated indirect land use change impacts are included),
however with 3 additions: "Improved vegetable oil1", "Improved
vegetable oil2" and "ILUC-free vegetable oil", whose assumed GHG
performance fro this exercise is included in the table below. The inclusion of
the two first categories is intended to reflect the possibility of vegetable
oil producers to improve their direct greenhouse gas emissions,
while still including the ILUC estimate in the reported emissions. The last
category reflects the potential succesfull development of "ILUC" free
biofuels, where the biofuels are produced in ways that are certified not to
lead to indirect land-use change emissions (with additional certification
costs). However, the latter catergory is not included in the mix, as no such
certification scheme has been developed yet. || Direct emissions [gCO2/MJ] || Estimated ILUC emissions [gCO2/MJ] || Total emissions [gCO2/MJ] || Estimated CO2 abatement costs of different biofuels [gCO2/€] Biodiesel - non-land using (UCO etc.) || 9.3 || 0.0 || 9.3 || 354 Sugar cane || 20.2 || 15.4 || 35.6 || 540 Sugar beet || 27.1 || 7.2 || 34.3 || 580 Wheat Straw as process fuel in CHP plant || 21.6 || 13.8 || 35.3 || 644 Corn (maize) || 32.7 || 10.1 || 42.7 || 683 Wheat Natural gas as process fuel in CHP plant || 33.5 || 13.8 || 47.3 || 755 2G ethanol - non-land using || 9.0 || 0.0 || 9.0 || 882 Wheat Process fuel not specified || 50.3 || 13.8 || 64.0 || 1124 2G ethanol - land using || 16.7 || 15.4 || 32.1 || 1272 2G biodiesel - land using || 5.4 || 15.4 || 20.8 || 1753 Improved vegetable oil 2 || 25 || 54.9 || 79.9 || 4153 Palm oil with methane capture || 29.3 || 54.0 || 83.3 || 5004 Improved vegetable oil 1 || 28 || 54.9 || 82.9 || 5443 Sunflower || 32.4 || 53.5 || 85.9 || 6699 Fossil fuels || 90.3 || 0 || 90.3 || 0 Rapeseed || 40.2 || 54.9 || 95.1 || No abatement Soybean || 46.9 || 56.3 || 103.2 || No abatement Palm oil || 51.1 || 54.0 || 105.1 || No abatement Table 23: Estimated carbon abatement costs
of biofuels when ILUC emissions are included. A rough assessment of the impacts of the
estimated costs and greenhouse gas emissions performance outlined above
suggests that the production of bioethanol, and that of advanced biodiesel and
vegetable oils with an improved greenhouse gas performance would be greatly
favoured. A number of observations can be made: ·
Firstly, advanced 2nd generation land
using biodiesel[194]
with high costs (3 times the cost of 1st generation biofuels) would seem to
become competitive for the fuel suppliers due to their good greenhouse gas
performance compared to other available sources of biodiesel. ·
The use of conventional vegetable oils would be
strongly discouraged unless their greenhouse gas performance is improved, as
otherwise they would seem to be of a higher or similar carbon intensity than
conventional fossil fuel sources. However improved vegetable oil pathways are
being used, and every gram they save is highly appreciated by the fuel
suppliers. This is highlighted by the fact that the "improved vegetable
oil2" is only saving 3 g/MJ more than "improved vegetable oil1",
while being 7% more expensive, it is still more competitive. ·
Analysing the boundary conditions, i.e. what
value would it have if one could change one of the them (the shadow price), it
is clear that the fuel costs for complying with the Fuel Quality Directive
would be greatly reduced if more ethanol could be blended in. This is because
the ethanol feedstocks are assumed to be cheaper and in general delivers higher
greenhouse gas savings, ILUC included, which would result in a more favourable
carbon abatement price compared to biodiesel alternatives. Another question is the availability of
so-called "ILUC-free" vegetable oil based biofuels[195]. In the assessment above they
are excluded from the mix as no such biofuels currently exist. Should these
become available at moderately higher costs compared to uncertified
alternatives, they would become more attractive to fossil fuel operators than
advanced biodiesel (i.e. as their price would be lower for similar carbon
abatement costs) and non-ILUC-free conventional biodiesel (i.e. as their carbon
abatement cost would be three times higher for the moderate increase in costs).
[1] Directive
2009/30/EC. [2] The
requirement in the Renewable Energy Directive also applies to bioliquids.
References to 'biofuels' in this document
should be taken as also applying to bioliquids. [3] Article
7d(6) of Directive 2009/30/EC and Article 19(6) of Directive 2009/28/EC. [4] COM(2010)
811. [5] Meetings
of this group were jointly chaired by DG ENER and DG CLIMA. Other Commission
Directorates General who were part of this group included
the Secretariat General, DG ENV, DG MOVE, DG ENTR, DG ECFIN, DG
AGRI, DG DEVCO, DG TRADE and the Joint Research Centre. [6] http://trade.ec.europa.eu/doclib/docs/2011/october/tradoc_148289.pdf. [7] http://ec.europa.eu/energy/renewables/transparency_platform/action_plan_en.htm. [8] http://ec.europa.eu/energy/renewables/studies/land_use_change_en.htm. [9] Insert reference before publication. [10] This
requirement is progressive as it increases to 50% in 2017 and 60% in 2018 for
new installations. [11] See
table 1 in chapter 2.8, introducing the baseline, for the estimates contained
in the plans. [12] Available at
http://ec.europa.eu/clima/documentation/roadmap/docs/com_2011_112_en.pdf. [13] Available at
http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2011:0144:FIN:EN:PDF. [14] WWF/Ecofys 2011 - The Energy Report, available
here: http://www.ecofys.com/com/publications/The Energy-Report-Ecofys.htm. [15] For comparison, total cropland in the EU
represents around 107 Mha. [16] Available at http://srren.ipcc-wg3.de/report. [17] This estimate takes into account productivity
increases, substitution effects and the estimated land saved by the production of
co-products of biofuels. [18] Available here:
http://www.iea.org/papers/2011/biofuels_roadmap.pdf. [19] Note that the land savings of co-products
produced from conventional biofuels are not considered in the figure. [20] IPCC Fourth Assessment Report (AR4). Available
here: www.ipcc.ch. [21] Van der Werf et.al. CO2 emissions from forest
loss, Nature Geoscience, vol 2, 2009. [22] The median value of 164 scenarios is at around
150 EJ, while the whole span is 35 EJ – 300 EJ for 2050. [23] To halt GHG concentration to below 440 ppm.
Executive summary. For reference see footnote 16. [24] Laborde, D.D. Domestic policies in a globalized
world: what you do is what I get (2011). [25] Lines illustrate ranges when Monte Carlo analysis
of the uncertainty has been carried out; shaded bar areas illustrate
maximum and minimum values from the analysis when different scenarios have been
considered. [26] The methodology set out in the Directives for
calculating land-use change prescribes that such emissions shall be divided by
20 years. [27] CARB Expert Workgroups:
http://www.arb.ca.gov/fuels/lcfs/workgroups/ewg/expertworkgroup.htm. [28] Reports available here:
http://ies.jrc.ec.europa.eu/jec-research-collaboration/downloads-jec.html. [29] A list of typical values for a range of most common biofuel pathways
is discussed in table 4. [30] In addition, the fossil fuel comparators for bioliquids are 91, 77
and 85 g/MJ depending on whether they are used for electricity production, heat
production or cogeneration. [31] JRC estimates on expected fossil fuel comparator in 2020 can be
found in Annex VI. [32] A: average direct emissions in 2020 based on the Member States
National Renewable Energy Action Plans (NREAPs, 27.2Mtoe, ¾ of biodiesel vs ¼
bioethanol). All biofuels are assumed to meet the greenhouse gas emissions
thresholds in the Directives - no changes to current sustainability scheme but
fossil fuel comparator is set at 90..3g CO2/MJ. B: possible range (5th to 95th
percentile) of estimated indirect land-use change emissions according to latest
IFPRI-MIRAGE-BioF 2011 study based on NREAPs (27.2Mtoe, ¾ of biodiesel vs ¼
bioethanol). Averaged over 20 year period according to the Directive's
greenhouse gas emissions methodology. C: Sum of average direct emissions in
2020 and estimated indirect land-use change emissions (A+B). D: marginal fossil
fuel emissions from crudes not being extracted based on the assumed 2020 fossil
fuel comparator (lower end) and high emitting oil sands from Brandt et al
(upper end). E: overall greenhouse gas emissions balance of the expected
biofuel mix in 2020 compared to fossil fuels; range comes from comparing high
indirect land-use change emissions with low fossil fuel emissions and vice
versa. [33] See the EU 2050 Roadmap for an indication of the required reduction
in transport emissions. [34] However, if it is the least fertile land that has been
recently abandoned, then its future production could be expected to show
typical yields below average, leading to either increased land requirements or
increased use of fertilisers. In addition, if the land is under a process of
managed reforestation, its reversion to agricultural production could result in
the release of carbon emissions. [35] Tropical forests were the primary sources of new agricultural land
in the 1980s and 1990s. H. K. Gibbs, A. S.Rueschb, F. Achardc, M. K. Claytond,
P. Holmgrene, N. Ramankuttyf, and J. A. Foleyg 2010. [36] 1996 IPCC Guidelines for National Greenhouse Gas Inventories (Vol.
3, Energy, p. 1.10). [37] Directive 2009/28/EC. [38] National total excluding the LULUCF sector. [39] Unlike non-CO2 greenhouse gases from agricultural
activities e.g. methane and nitrous oxide from ruminants and fertilisers. [40] EmployRES study (p. 133). http://ec.europa.eu/energy/renewables/studies/doc/renewables/2009_employ_res_report.pdf.
This figure represents estimated gross effects and does neither take into
account adjustments in other parts of the economy (i.e. reduced opportunities
in fossil fuel industry) nor adjustments for tax incentives and subsidies given
to the production of biofuels. [41] All the plans, in both English and original language
are available here: http://ec.europa.eu/energy/renewables/transparency_platform/action_plan_en.htm. [42] Biofuels made from certain feedstocks (waste, residues
and woody material) are counted double towards the 10% target of the Renewable
Energy Directive. [43] Data for cereals from: Prospects for agricultural
markets and income 2010-2020, http://ec.europa.eu/agriculture/publi/caprep/prospects2010/fullrep_en.pdf. [44] Bioethanol figures are for fuel use only. Data on
cereals, sugar beet and oil crops divided by member states, data available at
http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-ED-10-001/EN/KS-ED-10-001EN.PDF. [45] Commission's calculations based on total bioethanol
volumes from "The EU Beet and Sugar Sector: A model of environmental
sustainability " available at
http://www.cibe-europe.eu/Press/Brochure%20CIBE-CEFS%20Final_05.05.2010.pdf. [46] Oil World March 2011. [47] In comparison, global production of cereals reached
2240 million tons, vegetable oils at 141 million tons and sugars at 174 million
tons in 2009. OECD-FAO Agricultural Outlook 2010-2019. [48] Source: Progress report on Renewable Energy and
supporting material. The Communication and the accompanying staff working
documents are available here:
http://ec.europa.eu/energy/renewables/reports/reports_en.htm. [49] Source: IFPRI-MIRAGE-BioF simulations. [50] Although the model does not differentiate between
commodities according to their market uses, imports of these feedstocks are
attributed to the additional demand from biofuels. [51] Source European Biodiesel Board. Units adjusted to
Mtoe. [52] Source Epure. Part of the installed bioethanol capacity
quoted here is not only destined to the biofuel markets.Units adjusted to Mtoe. [53] Statistics from Fediol's website at www.fediol.be. [54] JEC Reference scenario:
http://ies.jrc.ec.europa.eu/uploads/jec/JEC%20Biofuels%20Programme.pdf. [55] The study can be found here:
http://ies.jrc.ec.europa.eu/about-jec. [56] On the basis of extrapolation of the known sales in
Germany (where the 100 filling stations selling E85 – 5% of the European total
of such filling stations – sold about 5.5 ktoe of the fuel. [57] EC 2010 RES progress report available at http://ec.europa.eu/energy/renewables/reports/doc/sec_2011_0130.pdf. [58] In their modelling, it was assumed that half of those
biofuels double counted under the RED were considered to come from waste and
residues, having no ILUC impact, whereas the other half was modelled as
increased bioethanol demand. [59] See press release here:
http://www.regjeringen.no/upload/MD/2011/vedlegg/klima/klima_skogprosjektet/Press_Release_Inpres_Mo_atorium_ENG.pdf. [60] GLOBIO 3 is developed by a consortium made up of UNEP
world Conservation Monitoring Centre (WCMC), UNEP/GRID-Arendal and the
Netherlands Environmental Agency (PBL). [Alkemade et al, 2009]. [61] Biodiversity is described in GLOBIO3 on the basis of
the remaining mean species abundance (MSA) of original species, relative to
their abundance in pristine or primary vegetation, which are assumed to be not
disturbed by human activities for a prolonged period. MSA is therefore
considered as the indicator for biodiversity. [62] For example, according to the MSA values in the table,
a transition from pastureland (MSA 70%) to cropland (MSA 10%) will cause a loss
of 60% of MSA on top of the 30% already lost from the conversion from natural
land to pastureland. [63] See details on assumptions in chapter 2.2 of the
report: Technical assistance for an evaluation of international schemes to
promote biomass sustainability (2009) http://ec.europa.eu/energy/renewables/bioenergy/sustainability_criteria_en.htm. [64] The fossil fuel comparator for biofuels in 2020 has
also been estimated. Please see Annex VI. [65] Values that were not included in the COWI set are based
on the typical values in the Directives, however, improved with the same
percentage as other bioethanol fuels or biodiesel fuels respectively. Although
the specific indirect land-use change emissions associated with typical land
using second generation biofuels were not modelled, these are assumed to be at
the same level as for sugarcane in the assessment of the options (i.e. high yielding
crops with no land saving co-products). [66] It is also worth noting that average emissions reached
116g/MJ under an extreme scenario not included in the sensitivity analysis where
the increased demand for biofuels did not lead to either yield increases, food
consumption reductions or intermediate consumption of agro-foods compared to
the baseline. [67] This requirement is progressive as it increases to 50%
in 2017 and 60% in 2018 for new installations. [68] Article 7d(6) of Directive 2009/30/EC and Article 19(6)
of Directive 2009/28/EC. [69] COM(2010) 811. [70] Article 19(6) of the Renewable Energy Directive and
Article 7(d)6 of the Fuel Quality Directive. [71] Recital 85 of the Renewable Energy Directive and
Recital 22 of the Fuel Quality Directive. [72] The Fuel Quality Directive defines "life cycle
greenhouse gas emissions" as the net emissions of CO2, CH4 and N2O that
can be assigned to the fuel, included any blended components, or energy
supplied. This includes all relevant stages from extraction or cultivation,
including land-use changes, transport and distribution, processing and
combustion, irrespective where those emissions occur (Article 1.2 of the Fuel
Quality Directive). [73] Article 23 of 2009/28/EC. [74] Article 17.2 of 2009/28/EC and Article 7b.2 of
2009/30/EC.
[75] Wetlands International (2010). [76] The Millennium Ecosystem Assessment distinguishes four
categories of ecosystem services: provisioning services, regulation services,
cultural services and supporting services. Provisioning services are defined as
harvestable goods such as fish, timber, bush meat, genetic material, etc. [77] Further detail can be found at http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:C:2010:160:0008:0016:EN:PDF. [78] Please see description of this option under Annex XII
for further detail. [79] Annex V C of 2009/28/EC and Annex IV C of 2009/30/EC. [80] Member States are required to report the greenhouse gas emissions
savings from biofuels under the general reporting requirements set by Article
22 of the Renewable Energy Directive and for demonstrating compliance with the
greenhouse gas emissions reduction target set by Article 7a of the Fuel Quality
Directive.
[81] The
minimum greenhouse gas savings threshold is 35%, raising to 50% in 2017. [82] Article 19.7 of 2009/28/EC and Article 7d. 6 of 2009/30/EC. [83] The Directives include a clause that would mean that biofuels should
not be regarded as failing to comply with the sustainability criteria until
2018 (as long as they achieve 45% minimum savings when direct savings are
looked at and are produced in a plant installed prior to 2013), should such
methodology be introduced. [84] Note that "availability" refers to whether
the feedstock meets the greenhouse gas saving threshold, and is not related to
physical availability of the feedstock. [85] The effectiveness criteria assesses whether policy
options are minimising the greenhouse gas impact of biofuels, while ensuring
that emissions from biofuels are below the required thresholds set out in the
Directives, as well as respecting the greenhouse gas emissions reduction target
set out in the Fuel Quality Directive. [86] As the level of uncertainty of the indirect land-use
change emission estimates included in the baseline is already high, the
introduction of further uncertain results through assumptions on costs gives
rise to counterintuitive results, and risks of misinformed assessment. The
assessment of costs is therefore limited to a qualitative level. [87] Please see article 7b of 2009/30/EC, Article 18 of
2009/28/EC and related guidance (http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:C:2010:160:0001:0007:EN:PDF).
An assessment of the administrative burden associated with the current
verification framework can be found in the impact assessment of the energy and
climate package (Chapter 6.7.1.5. How should performance be verified? – page
149). [88] In the baseline, which is based on how the MS are
planning to meet the RED targets, the fulfilment of the Fuel Quality Directive
is -5.4%. As such, it is therefore assumed that the remaining contribution
percentage point will be achieved through non-RES technologies qualifying for
the FQD such as flaring and venting. This is uncertain and should this not be
the case, even more biofuels or electric vehicles would be needed to meet the
Fuel Quality Directive. [89] See footnote 31. [90] See Annex V and Annex IV of the Renewable Energy
Directive and Fuel Quality Directive respectively. [91] It is important to consider the date of adoption and
the date by which it enters into force, as both influence the response of
economic actors. This is not taken into account. [92] If the sustainability criteria were assumed to have an
effect (by eliminating certain feedstocks with high direct emissions), the
emissions balance in the baseline is improved to 22%. [93] All NREAPs are available here: http://ec.europa.eu/energy/renewables/transparency_platform/action_plan_en.htm.
[94] Please see Annex X for more detail. [95] Commission Communication of 7.2.2007 on Results of the
review of the Community Strategy to reduce CO2 emissions from passenger cars
and light-commercial vehicles COM(2007)19, proposed a CO2 goal of 130g/km for
passenger cars, together with a "further reduction of 10g CO2/km, or
equivalent if technically necessary, by other technological improvements and by
an increased use of biofuels. This would need to be measurable, monitorable,
accountable and non-double-counting the reductions of CO2." [96] See definitions of default and typical values laid down
in Article 2 of the Renewable Energy Directive. [97] The crop specific values are estimated based on the
increase of biofuel consumption towards 2020 compared to the existing
consumption in 2008, and not necessarily representative for those in the baseline.
Indeed, IFPRI-MIRAGE-BioF points out that for rapeseed, which is the most
important feedstock used in 2008 (5.7 Mtoe out of a total of 12 Mtoe in 2020),
the estimated indirect land-use change emissions in the baseline should be
significantly lower. [98] To ensure a consistent increase of the level of
ambition, also the threshold for new installations is increased. New
installations after 2018 would be required to save at least 65%, instead of
currently planned requirement of 60%. [99] Article 19(6) and Article 7d(6) of the Renewable Energy
Directive and Fuel Quality Directive respectively. [100] EU renewable energy targets in 2020: Analysis of
scenarios for transport – JEC biofuels programme
http://ies.jrc.ec.europa.eu/uploads/jec/JECBiofuels%20Report_2011_PRINT.pdf. [101] Assumptions: Electricity consumption of an electric car
can be assumed to be approximately 0.20 kWh/km. Average annual electric
distance travelled is assumed to be 15.000km. Annual electricity consumption of
one electric car will be roughly 3000 kWh. Expressed in toe, this is 3 * 8.6 *
10-5 which gives 0.258 toe. Therefore 1 Mtoe electricity consumption by cars
implies 3.9 million cars on the road. [102] The NREAPs estimate a total of 0.7 Mtoe of renewable
electricity in road vehicles by 2020. In order to convert this figure to
overall electricity it has to be divided by the fraction of renewable energy in
the electricity mix of 2020, assumed to be 34% for the EU in the NREAPs. This
gives 2.1 Mtoe of electricity. The real figure is likely to be lower, as
countries with higher than average share of renewable energy in the electricity
mix, will use national values rather than the EU-average. [103] Annex V of the Renewable Energy Directive and Annex IV
of the Fuel Quality Directive. [104] Impact Assessment - Report from the Commission to the
Council and the European Parliament on sustainability requirements for the use
of solid and gaseous biomass sources in electricity, heating and cooling.
Available here: http://ec.europa.eu/energy/renewables/transparency_platform/doc/2010_report/sec_2010_0065_1_impact_assessment_en.pdf. [105] IEA Task 39 – Don O'Connor, Biodiesel GHG emissions,
past, present and future January 2011. [106] If the price of waste and residual oil increases, the
total cost of using virgin oil goes down, as the cost of using virgin oils is
the difference in price between virgin oil and waste/residual oil. [107] If estimated indirect land-use change emissions were to
be considered in the calculation of emissions, the reported savings to the Fuel
Quality Directive target would be -4.5% and -3.4% for overestimating and
underestimating indirect land-use change emissions respectively. [108] Available here:
http://www.ipcc-nggip.iges.or.jp/public/gpglulucf/gpglulucf.html. [109] As in the description of C2, the word "additional"
makes reference to measures and developments beyond progress under a likely
business as usual scenario. [110] The dissemination of the accounting rules of LULUCF is beneficial in
itself: improved data, more awareness in the countries in question, and possible
easier transition to commitments under a new agreement. [111] For ACP countries: African, Caribbean and Pacific Group
of States. Secretariat: http://www.acpsec.org/. [112] To benefit from GSP+ conditions, countries have to
ratify and implement 27 international conventions and undergo a rigorous
vetting and application process. GSP+ eligibility is reviewed every 3 years.
Further explanation can be found here:
http://trade.ec.europa.eu/doclib/docs/2007/september/tradoc_136097.pdf. [113] List of Annex I countries available here:
http://unfccc.int/parties_and_observers/parties/annex_i/items/2774.php. [114] In this context, it is important to note that the crop specific
values are estimated based on the increase of biofuel consumption towards 2020
compared to the existing consumption in 2008. IFPRI-MIRAGE-BioF reports that
for those feedstocks that display a strong non-linearity effect (i.e.
rapeseed), the indirect land-use change emissions in 2020 would be higher than
those observed at lower consumption levels. Due to the limitations of this
assessment (i.e. snapshot at 2020), this effect is not considered further in
this document. [115] EU renewable energy targets in 2020: Analysis of
scenarios for transport – JEC biofuels programme
http://ies.jrc.ec.europa.eu/uploads/jec/JECBiofuels%20Report_2011_PRINT.pdf. [116] Assumptions: Electricity consumption of an electric car
can be assumed to be approximately 0.2 kWh/km. Average annual electric distance
travelled is assumed to be 15.000km. Annual electricity consumption of one
electric car will be roughly 3 kWh.. Expressed in toe, this is 3 * 8.6 * 10-5
which gives 0.26 toe. Therefore 1 Mtoe electricity consumption by cars implies
3.9 million cars on the road. [117] The NREAPs estimate a total of 0.7 Mtoe of renewable
electricity in road vehicles by 2020. In order to convert this figure to
overall electricity it has to be divided by the fraction of renewable energy in
the electricity mix of 2020, assumed to be 34% for the EU in the NREAPs. This
gives 2.1 Mtoe of electricity. The real figure is likely to be lower, as
countries with higher than average share of renewable energy in the electricity
mix, will use national values rather than the EU-average. [118] See chapter 3 for further detail. [119] For example, it is worth noting that following the implementation
of the Californian Low Carbon Fuel Standard in 2010, a price differential
emerged between biofuels of different carbon intensity. [120] If the price of waste and residual oil increases, the
total cost of using virgin oil goes down, as the cost of using virgin oils is
the difference in price between virgin oil and waste/residual oil. [121] With regard to the Fuel Quality Directive carbon
intensity reductions, the 5.4% contribution achieved under the central scenario
(factors at 50th percentile) includes indirect land-use change emissions, which
are now being reported. The Fuel Quality Directive reduction under the biofuel
mix obtained would range between 4.8% and 5.9% depending on whether we have
overestimated or underestimated the "real" indirect land-use change
emissions. [122] In the case of "yield increases", only the
additional production to the average yield levels assumed in the baseline would
be considered to have met these additional sustainability criteria. [123] Preliminary results from pilots as communicated by the
Roundtable for Sustainable Biofuels. Most of the cost is based on travel and
administration in addition to the auditor time. [124] Around 2-3 % of the 10% target of double counted
advanced biofuels would be needed. This is equivalent to 6 to 9 Mtoe. For
comparison, the US RFS2 is requiring 36 billion gallons by 2022, of which at
least 16 billion gallons have to be advanced biofuels from cellulosic material.
16 billion gallons of ethanol is equivalent to around 30 Mtoe, i.e. an energy
quantity similar to what is required to reach the 10% transport target of the
Renewable Energy Directive. [125] Tender specifications available at:
http://ec.europa.eu/dgs/energy/tenders/doc/specifications/2009/s112_160619_specifications.pdf. [126] All responses are available at:
http://ec.europa.eu/energy/renewables/consultations/2009_07_31_iluc_pre_consultation_en.htm. [127] http://ec.europa.eu/energy/renewables/studies/land_use_change_en.htm. [128] http://re.jrc.ec.europa.eu/bf-tp/html/documents_main.htm. [129] http://trade.ec.europa.eu/doclib/docs/2011/october/tradoc_148289.pdf. [130] http://ec.europa.eu/energy/renewables/transparency_platform/action_plan_en.htm. [131] In 2011, the JRC carried out additional application of their Spatial
Allocation Methodology (SAM) to additional IFPRI-MIRAGE-BIOF scenarios. [132] All responses are available at
http://ec.europa.eu/energy/renewables/consultations/2010_10_31_iluc_and_biofuels_en.htm. [133] All contributions, including the workshop's report
"Critical Issues in Estimating ILUC Emissions. Outcomes of an Expert
Consultation" EU report n. JRC64429, are available via http://re.jrc.ec.europa.eu/biof/html/documents_publications.htm. [134] Most biofuel feedstocks co-produce considerable quantities
of co-products. Most models do now take this into account, although at various
ratios, greatly influencing model results. Co-products normally replace animal
feed, freeing up land that would otherwise be needed for its production. [135] Baseline yield increases are normally assumed to
continue at historic rates whereas such predictions are uncertain. [136] There is little empirical evidence on developments of
marginal yields. [137] The type of land that is converted to cropland has a
major influence, as carbon stocks vary considerably across land types. Due to
too course spatial resolution regional differences risk getting lost in the
geographical aggregations. [138] Land availability and land classification is an
essential input for land-use change modelling, however, figures and terminology
are not consistent across datasets. [139] Elasticities are often estimated on basis of data from
developed countries, while models suggest that indirect land-use change
typically takes place in developing countries. [140] Carbon stock values attributed to different vegetation
and soils vary considerably across studies, and play an essential part in
determining the indirect land-use change impact. [141] Pasture for animals covers large parts of the globe, and
offer potential supply of land for cultivation. However, how pasture is
modelled and its interlinkages with feed markets and cropland differ between
models. The assumptions has major impacts on the overall results, as pasture is
covering a large fraction of the terrestrial surface, and has relatively low
carbon stocks. [142] Drivers behind deforestation are complex, where local
authorities, land-use rights and political economy all play a role. It is not
possible to properly reflect this real world effects in the models, where
decision making is reduced to a purely rational economic question. [143] Economic models assume demand being a function of price,
with different assumptions as to how the additional demand for biofuels will
impact on food and feed commodity markets. The feedback from lowered oil-price
to potential lowered food-price also needs consideration (Recommendations from
the Food Consumption Subgroup ARB Expert Workgroup on Land-use Change –
http://www.arb.ca.gov/fuels/lcfs/workgroups/ewg/010511-final-rpt-food-consumption.pdf). [144] Most models include yields to increase as a result of factor
increase (labour, fertiliser, capital), but none reflect the possibility for
technology change in response to higher prices. [145] The literature review did not analyse spreadsheet
models, as very few were published at the time of writing. [146] Increased yields are a function of a complex set of
variables, among them increased investment and research, both of which take
place as a response to the biofuel policy. It is however difficult to capture
this effect in the models. [147] Structural changes are typically difficult to predict by
models as elasticities are based on historical data. Considerable increase in
use of land in e.g. CIS is therefore unlikely according to the models, while
such a structural change could take place both in the baseline and in the
policy scenario. [148] This is underestimating the land saved by co-products.
For example, in the EU soy meal is a key source of protein, of which around 97%
is imported. There is thus considerable scope of substitution. [149] Many models do not properly take into account the
emissions from peat oxidation following drainage process required in the
cultivation of palm oil, which could underestimate real emissions by an order
of magnitude. Although the estimated emissions from peatlands have been
adjusted upwards in recent modelling, the uncertainty as to what the value
should be remains. [150] Presentations from the workshop can be found here:
http://re.jrc.ec.europa.eu/bf-tp/html/documents_main.htm. [151] Bouët, A., Dimaranan, B. V. and Valin, H. (2010),
Modeling the global trade and environmental impacts of biofuel policies, IFPRI
Discussion Paper (01018), International Food Policy Research Institute. [152] Al-Riffai, P., Dimaranan, B. and Laborde, D. (2010),
Global Trade and Environmental Impact Study of the EU Biofuels Mandate, Final
Report for the Directorate General for Trade of the European Commission,
International Food Policy Research Institute. [153] http://www.fao.org/fileadmin/templates/wsfs/docs/expert_paper/How_to_Feed_the_World_in_2050.pdf [154] FAO Investment Centre website:
http://www.fao.org/tc/tci/whyinvestinagricultureandru/en/. [155] This is also clear from the modelling referred to in the
chapters below, where the overall land-use change in the baseline (what would
happen without a policy promoting biofuels in the EU) is 20 times larger than
then additional land-use change caused by biofuels. [156] FAO Statistics. Note that there is an important
difference between "harvested area" and "cultivated area". Double-cropping
in a field would double the amount of harvested area, while cultivated area
remains constant. [157] However, if it is the least fertile land that has been
recently abandoned, then its future production could be expected to show
typical yields below average, leading to either increased land requirements or
increased use of fertilisers. In addition, if the land is under a process of
managed reforestation, its reversion to agricultural production could result in
the release of carbon emissions. [158] Definitions of marginal land vary across studies.
Figures from the Okoinstitut report are specifically referred to as abandoned
cropland and particularly unused degraded land. [159] Tropical forests were the primary sources of new
agricultural land in the 1980s and 1990s. H. K. Gibbs, A. S. Rueschb, F.
Achardc, M. K. Claytond, P. Holmgrene, N. Ramankuttyf, and J. A. Foleyg 2010. [160] Based on official data from the Brazilian National
Institute of Space Research. [161] COM(2008) 645 - Available here:
http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2008:0645:FIN:EN:PDF.
[162] TBFRA 2000 -
http://www.unece.org/timber/fra/welcome.htm. [163] Protected forest area was 38.1 million hectares in 2005
compared with the 30.1 million hectares in 2000, forests with protective
functions increased by one million hectares to 21.6 million hectares by 2005. Forest habitat types designated as Natura 2000 sites cover over 22 million ha in 2008,
according to Eurostat. Forests undisturbed by man/, account for about 9 million
ha. [164] http://www.un-redd.org/. [165] Only perennial crops will enter, as international
accounting rules consider annual crops carbon neutral. [166] http://ec.europa.eu/development/icenter/repository/COMM_PDF_COM_2010_0127_EN.PDF. [167] Opportunities for avoidance of land-use change through
substitution of soya bean meal and cereals in European livestock diets with
bioethanol coproducts, GCB Bioenergy (2010). [168] COM(2007)723, Communication from the Commission to the
Council, the European Parliament the European Economic and Social Committee and
the Committee of the Regions, "European Strategic Energy Technology Plan
(SET-Plan), Towards a low carbon future" 2009. [169] SEC(2009)1295, Commission Staff Working Document
Accompanying document to the Communication from the Commission to the European
Parliament, the Council, the European Economic and Social Committee and the
Committee of the Regions on Investing in the Development of Low Carbon
Technologies (SET Plan) "A Technology Roadmap", 2009. [170] For an overview of the European Industrial Initiatives,
see the Commission website: European Commission, "SET-Plan, towards a
low-carbon future", available here:
http://ec.europa.eu/energy/technology/set_plan/doc/setplan_brochure.pdf. [171] Kyriakos Maniatis & Stefan Tostmann, " EU
Technology Strategy on Bioenergy: From Blue-Sky Research to Targeted Technology
Development", in Renewable Energy Law and Policy Review, Vol 1, N°2,
p169-179. [172] For more information see:
http://ec.europa.eu/clima/funding/ner300/index_en.htm. [173] Literature: EU Commission, DG ENVIRONMENT: Land-use Modelling
– Implementation. Preserving and enhancing the
environmental benefits of “land-use services". Final report, 1 April 2010.
Alkemade, R., Van Oorschot M., Miles L., Nellemann C. , Bakkenes M. and Ten
BrinkB.: GLOBIO3: A Framework to Investigate Options for Reducing Global
Terrestrial Biodiversity Loss.Ecosystems (2009) 12: 374–390. Van Oorschot
M.,Ros J. and Notenboom J.:Evaluation of the indirect effects of biofuel production
on biodiversity: assessment across spatial and temporal scales. PBL (Netherlands
Environmental Assessment Agency), Final report 27 May 2010. Campbell A.,
Doswald N.: The impacts of biofuel production on biodiversity: A review of the
current literature. United Nations Environment Programme – World Conservation
Monitoring Center. Final report, December 2009. [174] GLOBIO 3 is developed by a consortium made up of UNEP
world Conservation Monitoring Centre (WCMC), UNEP/GRID-Arendal and the
Netherlands Environmental Agency (PBL). [Alkemade et al, 2009]. [175] CARB (2011), Final Recommendations from the Elasticity
Values Subgroup, ARB LCFS Expert Workgroup, California. Available here:
http://www.arb.ca.gov/fuels/lcfs/workgroups/ewg/expertworkgroup.htm. [176] However, since some parameters
can increase indirect land-use change emissions when other can reduce it, the
right tail distribution of the parameter distribution does not involved a right
tail biased in the LUC expected distribution. [177] Indeed, assuming perfect
correlation among crops or regions does not affect the relative properties, and
comparative advantages of different crops. The geographical pattern of effects
and the feedstock mix for the overall scenario will not be subject to large
modifications in this framework. [178] Other shares (grassland, shrub)
are rescaled to be sure that the sum of share is equal to one. [179] Indeed, cropland extension in
the baseline, driven by economic and demographic growth is much larger than the
effects of the biofuels scenario studied here. [180] It also emphasises the role of
the baseline behavior in our assessment and the importance to understand that
we compute the effects of the biofuel policy as a marginal deviation from this
baseline when all other ongoing changes have already been taken into account. [181] Plevin, R. J., OHare, M., Jones, A. D., Torn, M. S. and
Gibbs, H. K. (2010), Greenhouse Gas Emissions from Biofuels Indirect Land-use
Change Are Uncertain but May Be Much Greater than Previously Estimated,
Environmental Science and Technology 44(21), 8015-8021. [182] The Millennium Ecosystem Assessment distinguishes four
categories of ecosystem services: Provisioning services, regulation services,
cultural services and supporting services. Provisioning services are defined as
harvestable goods such as fish, timber, bush meat, genetic material, etc. [183] Ecofys and Winrock. Available here:
http://www.ecofys.nl/com/publications/Responsible_Cultivation_Areas.htm. [184] Total amount of land in oil palm cultivation in Indonesia is supposed to be 4.5Mha. [185] Ecofys and Winrock 2009. Mitigating indirect impacts of
biofuel production. Case studies and Methodology. [186] Agrovista and Northeast biofuels grower network research
programme in North East England. 2011. [187] See table 3 in chapter 2 of this impact assessment. [188] Sustainable Oil Palm development on degraded land in Kalimantan (Fairhurst T, McLaughlin D, 2009). [189] Assumptions: Electricity consumption of an electric car
can be assumed to be approximately 0.13 kWh/km. Average annual electric
distance travelled is assumed to be 10,000km since they cannot be used for long
distances. Annual electricity consumption of one electric car will be roughly
1300kWh.. Expressed in toe, this is 1300 * 8.6 * 10-5 which gives 0.12 toe.
Therefore 1 Mtoe electricity consumption by cars implies 8 million cars on the
road. [190] Assumptions: Electricity consumption of an electric car
can be assumed to be approximately 0.13 kWh/km. Average annual electric
distance travelled is assumed to be 10,000km since they cannot be used for long
distances. Annual electricity consumption of one electric car will be roughly
1300kWh.. Expressed in toe, this is 1300 * 8.6 * 10-5 which gives 0.12 toe.
Therefore 1 Mtoe electricity consumption by cars implies 8 million cars on the
road. [191] Assuming life-cycle GHG emissions for CNG around
76.7gCO2eq/MJ and for LPG 73.6g CO2eq/MJ, each 1% increase in the sales of use
of natural gas and LPG across the EU as road transport fuel sales, would result
in a potential reduction of around 0.1% in greenhouse gas intensity. [192] Although the estimated land-use impacts associated with
second generation feedstocks were not specifically modelled, they have been
assumed to be equal to Brazilian sugar cane in this assessment (i.e. high yield
energy crop with non-land saving co-products). [193] Prices of ethanol and vegetable oils are mostly based on
F.O Lichts "World ethanol & biofuels report, vol. 10 no. 16 April
2012, page 335 and "Oils & Fats Int June 2012 Vol25 No5, page32. Cost
of crushing is assumed to be 20 $/ton. When costs have not been available, an
estimate has been made. Biodiesel from FAME is limited to 12.8 Mtoe due to
blending constraints (B7 – scenario 2 of the JEC study), ethanol is limited to
max 7.1 Mtoe in line with scenario 2 of the JRC study (see reference in chapter
2.8.1.5), 2nd generation non-land using biodiesel, which in this case refers to
biodiesel from residues and wastes, and improved vegetable oil2, are limited to
3 and 7 Mtoe respectively due to resource constraints, sugar cane import is
limited to 3.5 Mtoe, which is similar level as found in the IFPRI study. Fossil
fuel price is assumed to be 120 $/barrel, and finally it is assumed that
4%points of the 6% target of the FQD are achieved with the use of biofuels, the
rest being achieved through increased use of electricity in road transport and
reduction in flaring and venting emissions). [194] The amount of non-land using biodiesel from waste (i.e.
UCO and animal fat) is thought to be limited and as such has been capped at
3Mtoe. Should more of this cheaper alternative be available, it is expected to
take priority over the more expensive 2nd generation land using biodiesel
alternative.
[195] Such biofuels could be similar to the biofuels described
under option C2. They are here assumed to cost on average 30% more than other
vegetable oils.