Science of the Total Environment 557–558 (2016) 562–570

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Indonesia palm oil production without deforestation and peat conversion

by 2050
Dian Afriyanti a,⁎, Carolien Kroeze a, Asmadi Saad b
a
Environmental System Analysis Group, Wageningen University, Droevendaalsesteeg 3, 6708 PB Wageningen, The Netherlands
b
Soil Science, Agriculture Faculty Jambi University, Jl. Jambi-Muara Bulian KM. 1, Mendalo Darat, Jambi Luar Kota, Muaro Jambi, Jambi Province, Sumatra 36363, Indonesia

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• We explore scenarios for sustainable

palm oil production in Indonesia in
2050.
• Indonesia can stay a major oil palm
exporting country in our scenarios.
• Our scenarios assume no further defor-
estation or peat conversion.
• Most sustainable palm cultivation is en-
visaged in Kalimantan and Sumatra.
• Potential yields are much higher than
the current yields of plantations.

Potential production of Crude Palm Oil (CPO) in the year 2050 on potentially available suitable land in Indonesia,
and the total global demand for CPO in PalMA scenarios (PalmAM, PalmOS, PalmTG and PalmGO), representing
interpretation of Millennium Ecosystem Assessment (MA) scenarios, including Mosaic (AM), Order from
strengths (OS), Techno Garden (TG), and Global Orchestration respectively. All assume no further deforestation
and no oil palm plantations on peat soils. Dian Afriyanti, Carolien Kroeze, and Asmadi Saad.

a r t i c l e i n f o a b s t r a c t

Article history: Palm oil is a promising source of cooking oil and biodiesel. The demand for palm oil has been increasing world-
Received 27 August 2015 wide. However, concerns exist surrounding the environmental and socio-economic sustainability of palm oil pro-
Received in revised form 3 March 2016 duction. Indonesia is a major palm oil producing country. We explored scenarios for palm oil production in
Accepted 6 March 2016
Indonesia until 2050, focusing on Sumatra, Kalimantan and Papua. Our scenarios describe possible trends in
Available online xxxx
crude palm oil production in Indonesia, while considering the demand for cooking oil and biodiesel, the available
Editor: Simon Pollard land for plantations, production capacity (for crude palm oil and fresh fruit bunches) and environmentally
restricting conditions. We first assessed past developments in palm oil production. Next, we analysed scenarios
Keywords: for the future. In the past 20 years, 95% of the Indonesian oil palm production area was in Sumatra and Kaliman-
Scenarios tan and was increasingly cultivated in peatlands. Our scenarios for the future indicate that Indonesia can meet a
Crude palm oil considerable part of the global and Asian demand for palm oil, while avoiding further cultivation of peatlands and
No further deforestation forest. By 2050, 264–447 Mt crude palm oil may be needed for cooking oil and biodiesel worldwide. In Indonesia,
Avoiding further peat cultivation the area that is potentially suitable for oil palm is 17 to 26 Mha with a potential production rate of 27–38 t fresh
fruit bunches/ha, yielding 130–176 Mt crude palm oil. Thus Indonesia can meet 39–60% of the international
demand. In our scenarios this would be produced in Sumatra (21–26%), Kalimantan (12–16%), and Papua

⁎ Corresponding author at: PO Box 47 6700 AA, Wageningen, The Netherlands.

E-mail address: dian.afriyanti@wur.nl (D. Afriyanti).

http://dx.doi.org/10.1016/j.scitotenv.2016.03.032
0048-9697/© 2016 Elsevier B.V. All rights reserved.
 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570
(2%). The potential areas include the current oil palm plantation in mineral lands, but exclude the current oil palm
plantations in peatlands.
1. Introduction
Crude palm oil (CPO) production plays important roles in both
the local-global environment and socio economy. Between 2000 and
2014, exports and consumption of CPO in Indonesia have increased
from 5 to 22 Mt and from approximately 3 to 11 Mt respectively
(USDA, 2014). The production of CPO has resulted in economic im-
provement within rural areas through providing jobs for local people
(Basiron, 2002; Hirawan, 2011). CPO generates cooking oil (Corley,
2009) and biodiesel (IEA, 2011). As fossil fuel reserves become depleted,
we see an increase in petroleum prices (Tan et al., 2009), which in turn,
increases demand of CPO (IEA, 2010, 2011), an inexpensive source
(Miranowski and Rosburg, 2012). Renewable energy uses promote
the mitigation of carbon emissions (Cork et al., 2005, IEA, 2010). How-
ever, oil palm expansion has resulted in the conversion of peatlands
(Miettinen et al., 2012) and arable lands (Koh and Wilcove, 2009), in-
creased carbon dioxide (CO2) emissions (Inubushi et al., 2003, Hooijer
et al., 2006), resulted in annual fires (Hooijer et al., 2006), as well as in-
creased both subsidence and flood risk (Sumarga et al., submitted for
publication). Palm oil mill effluents (POME) are eutrophying, acidifying,
and toxic compounds, which pollute terrestrial and aquatic systems and
release greenhouse gases (Reijnders and Huijbregts, 2008). Social prob-
lems associated with conflicts over land rights (Dhiaulhaq et al., 2014)
and livelihood (low income) of smallholders due to decreasing price
of fresh fruit bunches (FFB). The negative effects of palm oil production
impact upon the marketability of CPO within the international market
(Disdier et al., 2013).
Several sustainability standards and policies promote no further
deforestation and peat conversion. The Roundtable Sustainable Palm
Oil (RSPO) provides principle, criteria and best management practices
for sustainable palm oil production (RSPO, 2014). European Energy
Directive prohibits biodiesel from peatlands (Lendle and Schaus,
2010). Similarly a moratorium agreement restricts new concessions in
forests and peatlands (stipulated in the Instruction of The President of
Indonesia No. 6/2013). The Indonesia President Decree No. 32/1990
forbids agriculture in deep peatlands (3 m depth or more). The world
research institute (WRI) has sustainability standard that oil palm can
be grown on soils containing b35 t carbon/ha, in areas with a low biodi-
versity and without conversion of forest and peatlands (Gingold et al.,
2012). Recently, the newly set President Decree No. 71/2014 stipulates
that the water table in peatlands should not exceed 40 cm below sur-
faces to prevent fires. ASEAN restricts fires based on an Agreement on
Trans-boundary Haze Pollution (AATHP), ratified on October 14, 2014
(stipulated on UU No. 26/2014).
Avoidance of deforestation and peat conversion results in uncer-
tainties related to potential areas and production capacity of CPO within
Indonesia. Miettinen et al. (2012) observed 7 Mha containing potential
areas of degraded lands in West and Central Kalimantan. While the min-
imum size of these areas is 5000 ha, while we expect that smaller-scale
areas (minimum size 1.5 ha) could be available for smallholders playing
important roles in expansion of oil palm in the future given the exten-
sion of moratorium. Conversely, Sumatra requires a trade-off between
environmental problems and palm oil production as Papua is rife with
potential areas for future expansion. Hoffman et al. (2014) simulated
potential FFB yield in Sumatra and Kalimantan through a palm simula-
tion (PALMSIM) model. The simulation implies much higher potential
yield than the actual yield. Although the model simulates yield in opti-
mal conditions (good planting materials, water, nutrients, absent pest
and diseases), improved yield is possible as several field assessments
approach the potential yields. However, limited by solar radiation,
563
© 2016 Elsevier B.V. All rights reserved.
running the model in flat and coastal areas mislead the potential yields
in peatlands. Peatlands are environmentally restricted, as well as nutri-
tionally and biophysically limited (Hoffmann et al., 2015). Our study
approaches this problem through the exclusion of peatlands. Never-
theless, it is still debatable whether increasing yield would decrease
expansions (Carrasco et al., 2014). We expect policy enforcement on
minimizing deforestation and peat conversion could play important
roles in the future.
We explore scenarios for CPO production in Indonesia until 2050,
focusing on Sumatra, Kalimantan, and Papua. Our scenarios describe
possible trends in CPO production in Indonesia, considering the demand
for cooking oil and biodiesel, the available land for plantations, produc-
tion capacity, and environmentally restricting conditions. This study is
novel in that it explores future trends under the condition that no
further deforestation takes place, and that oil palm is not grown on
peatlands. We developed our scenarios based on the Millennium
Ecosystem Assessment (MA). This resulted in a set of, so-called, PalMA
scenarios including Palm Global Orchestration (PalmGO), Palm Techno
Garden (PalmTG), Palm Order from Strengths (PalmOS), and Palm
Adaptic Mosaic (PalmAM). The MA scenarios are global scenarios
requiring further inclusion of local process complexity. PalMA scenarios
represent the low-high demand of cooking oil associated with con-
sumption level, population growth, and low-high demand of biodiesel
due to renewable energy demand interpreted by MA scenario at the
global level and downscaled to the demand of CPO in Indonesia. The
MA scenarios were interpreted in relation to South Africa and Europe
(van Vuuren et al., 2010); in Indonesia, for Jakarta Bay (Arifin, 2004)
and for palm oil production in South East Asia covering Indonesia,
Malaysia, and Thailand (Asmaryan, 2012). Our scenarios provide
a basis for broad analyses of sustainable palm oil production in
Indonesia and may facilitate future decision making. These scenarios
may predict Indonesia's future contribution to palm oil demand.
Such local interpretation of the global MA scenarios are highly rele-
vant for local stakeholders (Kok et al., 2007).
2. Methods
2.1. Study area
This study focused on CPO production in Indonesia. We specifically
focused on an area covering Sumatra, Kalimantan, and Papua. These
islands were selected because Sumatra and Kalimantan have been the
major oil palm plantation areas. Further expansion in these islands be-
comes limited in mineral lands, while expansions would further take
place in peatlands. Nevertheless, environmental problems become
more pronounced in these islands. This situation implies the necessity
of trade-off measures to promote good practices of palm oil production
and mitigation of environmental problems. Papua would provide large
areas for oil palm plantation in the future, but moratorium and environ-
mental concern may restrict oil palm cultivation in Papua.
2.2. Past trend of oil palm expansion
Our analysis of past oil palm development focused on the area and
types of plantations in the period of 1990 to 2010. We used data on
plantation area available from an online database of the Agriculture
Department (Deptan, 2014) and provided by Tropenbos (personal com-
munication, 2012). The Agriculture Department database presents the
annual area of oil palm, but does not distinguish oil palm plantations
in mineral lands from those in peatlands. We analysed the trend of
564 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570
conversion of mineral soils and peat to oil palm plantations based on
data from Tropenbos (personal communication, 2012) in Sumatra, Kali-
mantan, and Papua. The conversion trends were grouped into the pe-
riods of 1990–2000, 2000–2005, 2005–2010. Past trends of oil palm
expansion were used to predict how the areas would develop in the fu-
ture with the absence of policy enforcement.
2.3. Scenario analysis
Scenario analysis provides an exercise as to how the future would
evolve under some plausible situation. The scenario analysis in this
study aims to explore palm oil production in restricted condition (no
further deforestation and peat conversion), while CPO demand
increases for cooking oil and biodiesel. We conducted qualitative
(storylines) and quantitative scenario analysis.
2.3.1. Qualitative scenario analysis
We developed PalMA scenarios for the year 2050, describing CPO
production in Indonesia based on the MA scenarios. The original MA
scenarios, published as part of the Millennium Ecosystem Assessment,
consists of four scenarios combining two types of socio-economic devel-
opment and two approaches of environmental management. The
four MA scenarios are Global Orchestration (GO), Technogarden (TG),
Order from Strength (OS) and Adaptic Mosaic (AM). GO and TG assume
globalized socio-economic developments with either a reactive (GO) or
a proactive (TG) approach towards environmental management. OS
and AM assume regionalized socio-economic developments with either
a reactive (OS) or proactive (AM) environmental management ap-
proach (Cork et al., 2005). Our four PalMA scenarios for CPO production
are local interpretations of the four original MA scenarios. The PalMA
scenarios consist of PalmGO, PalmTG, PalmOS, and PalmAM, for the in-
terpretation of GO, TG, OS, and AM, respectively (Table 1). The PalMA
scenarios, consider socio-economic development including population
growth, level of consumption, and biodiesel requirement as well as pol-
icies about forest and peatlands. We qualitatively interpret MA scenari-
os into PalMA scenarios with respect to: (i) CPO demand for cooking oil
and biodiesel as affected by population growth, consumption level, and
biodiesel mandates; (ii) environmental management and underlying
reasons for prohibition of forest and peatlands; and (iii) mill manage-
ment considering environmental and socio-economic aspects (relations
with smallholders). The interpretation is based on our judgement on
the relevance of MA scenarios with the current debates on palm oil de-
velopment and how it would evolve/change along the scenarios. (See
Table 2.)
2.3.2. Quantitative scenario analysis
Quantitative analysis of the PalMA scenarios is based on our inter-
pretation of qualitative PalMA scenarios. Some values are consistent
with MA scenarios (population, biofuel requirement and production in
Asia) and how the implication of values to CPO production in
Indonesia. The consumption levels influence the demand of CPO for
cooking oil; we used low-medium-high consumption levels. The con-
sumption levels are based on MA scenario assumptions including con-
sumption at: (i) a low level applying for TG and AM, hence PalmTG
and PalmAM, (ii) a medium level applying for GO, hence PalmGO, and
(iii) a high level applying for OS, hence PalmOS. Further interpretation
of consumption levels of cooking oil is based on Corley (2009). We as-
sume two CPO extraction rates (oil extraction rate/OER). The first rate
we assumed was 23% OER for proactive scenarios, based on the target
of Malaysia (ETP, 2015) which is lower than the target of Indonesia
Palm Oil Research Centre (PPKS, 2014). We used the lower number to
make the scenarios become more realistic and a lower extraction rate
(20%) for reactive scenario based on the current OER from Deptan
(2014).
We calculated the demand for CPO for both cooking oil and biodiesel
by 2050. The CPO demand for cooking oil was estimated by multiplying
Table 1
The storylines of PalMA scenarios for future palm oil production in Indonesia. All four
scenarios assume that no further deforestation and peat conversion takes place.
PalMA scenarios
Palm Global Orchestration (PalmGO):
▪ CPO demands increase as cooking oil consumption and biodiesel uses increase.
▪ Cooking oil consumption increases because of population growth and high
consumption per capita.
▪ Biodiesel mandates play an important role in biodiesel demand. This is
associated with the replacement of fossil fuels to mitigate greenhouse gas
emissions.
▪ Demand for palm oil mostly comes from Europe and Asia, driving exports from
Indonesia.
▪ The environmental management is reactive, thus current policies remain but
are not strengthened: RSPO remains at the current principles and criteria, the
moratorium of forest and peatlands remains as well as the sustainability
criteria for renewable energy. Global environmental problems like global
warming and emission from peatlands prohibit cultivation on peatlands. Oil
palm is widely cultivated in areas that are not constraint by the international
concerns; but reducing other main agriculture commodities.
▪ The reactive environmental management leads to careless waste manage-
ment.
▪ Mills are located in areas of big companies where independent smallholders
have limited access. Smallholders can sell FFB through middle men who buy
and resell to mills.
Palm Techno Garden (PalmTG):
▪ CPO demands increase as cooking oil consumption and biodiesel uses increase.
▪ Cooking oil consumption increases because of population growth; despite low
consumption rates.
▪ Biodiesel mandates play an important role in biodiesel demand; this is
associated with requirements for green energy and CPO from sustainable
management.
▪ Demand for palm oil mostly comes from Europe and Asia, driving export of
CPO and its derivatives from Indonesia.
▪ The environmental management is proactive. Thus current policies are
strengthened. RSPO improves at the principle and criteria of RSPO for
peatlands management, which ban further expansion in peatlands and
reduces utilization of peatlands in existing plantations in 2010. Markets only
buy RSPO-certified palm oil which encourages the participation of Indonesia
to promote a certification mechanism and strengthen the roles of ISPO.
Nevertheless, the moratorium on forests and peatlands is stopped as it is
seen to be less effective than the certification mechanism. As well, consumer
awareness of the negative effects of oil palm cultivation on peatlands itself
improves sustainability. Oil palm is widely cultivated in suitable areas while
maintaining other main agricultural commodities.
▪ The proactive environmental management leads to environmentally friend-
ly waste management in mills. This is accomplished by reducing discharge of
nutrients and waste to water bodies through closed system POME
management, and application of solid waste like empty fruit bunches (EFB)
for organic fertilizers.
▪ Mills are located in large industrial areas. Nevertheless, good transportation
ensure shareholder accessibility to mills.
Palm Order from Strengths (PalmOS):
▪ CPO demands increase as cooking oil consumption and biodiesel use increase.
▪ Cooking oil consumption increases because of population growth and high
cooking oil consumption.
▪ Biodiesel use increases due to the depletion of fossil fuels.
▪ Demand for palm oil mostly comes from Asia, particularly China and India,
driving exports from Indonesia.
▪ The environmental management is reactive. Thus consumers of CPO
from Indonesia do not care about sustainability of the product.
Nevertheless, concerns about environmental problems may only be
associated with trans-boundary haze problems related to Indonesian peat
fires. Therefore, Indonesia bans further expansion in peatlands for oil palm
plantations.
▪ The reactive environmental management leads to careless waste manage-
ment. Thus, most waste is damped on land or water systems (EFB is not
utilized for organic matter, and POME discharges nutrients to water bodies).
▪ Mills are located in large companies, causing limited access to smallholders.
Middle men play an important role in selling FFB to mills.
Palm Adaptic Mosaic (PalmAM):
▪ CPO demands increases as cooking oil consumption and biodiesel use
increases.
 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570 565

Table 1 (continued) Table 3

International demand for cooking oil and biodiesel to be produced in Asia in the four
PalMA scenarios PalMA scenarios for 2050.
▪ Cooking oil consumption increases because of population growth; although
PalMA Demand (Mt/year) Required area if
consumption rate is low.
scenario production
▪ Biodiesel use in Asia increases as renewable energy policies penetrate in Asia. Cooking Biodiesel CPO for CPO for Total
remains at 3.6 t
▪ Demand for palm oil comes mostly from Asia, specifically China and India. oil cooking biodiesel CPO
CPO/ha (Mha)
▪ The environmental management is proactive. Thus Indonesia produces oil demand
sustainable CPO. This is caused by the awareness of Indonesia of better natural
PalmGO 202 175 253 194 447 124
resource management and sustainable production. ISPO is strengthened and
PalmTG 176 118 220 106 326 91
national policies enforce utilization of suitable areas for oil palm while
PalmOS 186 56 233 62 295 82
banning oil palm in peatlands.
PalmAM 100 78 125 139 264 73
▪ The proactive environmental management leads to better waste management.
Integrated mills are developed which are connected with sewage systems,
producing biogas and organic fertilizers.
in Asia includes biodiesel and bioethanol, of which biodiesel production
▪ Mills are located in plantations, and independent smallholders perform
small-scale mills in every district to improve benefits for smallholders. The shares 67% with bioethanol (Koçar and Civaş, 2013). Then, we assume a
government supports the establishment of smallholder mills. rate of 90% extracted biodiesel from CPO.

the associated global and Asian population as envisaged in the MA sce-
narios, by per capita cooking oil consumption levels from Corley (2009).
We assumed a rate of 80% extracted cooking oil from CPO. Biodiesel de-
mand was derived from renewable energy requirement in Asia of MA
scenarios by 2050 (Alcamo et al., 2003). The CPO demand for biodiesel
was based on renewable energy requirement from Asia based on MA
scenarios (Cork et al., 2005). We assumed renewable energy production
Table 2
Quantitative aspects of PalMA scenarios. All four scenarios assume increasing demands of
CPO for cooking oil and biodiesel as influenced by population growth, consumption levels,
and renewable energy requirements as including green energy demand (climate change
mitigation) and depletion of fossil fuels.
Palm Global Orchestration (PalmGO):
▪ By 2050, 8095 million people (Cork et al., 2005) consume 25 kg cooking
oil/person (Corley, 2009); and 384 Mt biofuel/year is needed, of which 261
Mt is produced in Asia (Cork et al., 2005).
▪ Companies assist independent smallholders to improve their yield through
best management practices. In return, independent smallholders sell FFB to
mills via ramps, the FFB collectors of companies. Consequently, much more
time is needed to transfer FFB to mills, causing low OER (20%) of current OER
calculated from ratio of FFB:CPO (Deptan, 2014).
Palm Techno Garden (PalmTG):
2.4. Potentially suitable areas for oil palm cultivation in Indonesia by 2050
We spatially identified potentially suitable areas for oil palm planta-
tions in Indonesia. ArcGIS 10.2 is used. Potential suitable areas are divided
into two categories based on environmental approaches of the scenarios
(reactive and proactive). Areas are identified from the maps of land sys-
tem, land uses, moratorium, and conservation areas. Spatial analysis
was done following three sets of suitability criteria. They are physical
properties, land uses, existence of moratorium and conservation areas.
2.4.1. Step 1: Selection of suitable areas based on physical properties
The first set of suitability criteria included physical properties (slopes,
terrains, and soil types). We identify the suitable areas from land system
maps provided by National Soil Agency (BPN) (personal communication,
2014). Suitable areas for reactive scenarios (PalmGO and PalmOS) in-
clude: areas having slopes b 30% and terrains of non-mountainous
areas, and hilly areas. Suitable areas for proactive scenarios (PalmTG
and PalmAM) include areas having slopes b30%, non-mountainous and
non-hilly. Thus, the reactive scenario differs in the uses of hilly areas
from proactive scenario. Nevertheless, the slope mostly ranges from 8
to 25%. This selection has eliminated mountainous areas which are not
suitable for oil palm in terms of soil properties and low yield due to
limiting radiation in higher altitude (Hoffmann et al., 2015).

▪ By 2050, 8821 million people (Cork et al., 2005) consume 20 kg cooking oil/-
Table 4
person (Corley, 2009); and 281 Mt biofuel/year is needed, of which 176 Mt is
Potentially suitable areas for oil palm plantations and the associated potential FFB yield in
produced in Asia.
the PalMA scenarios for 2050 for Sumatra, Kalimantan, and Papua.
▪ In plantations of smallholders and companies, high-tech nutrient
management, and effective pest and disease controls are practiced. Mills are Soil types in the Reactive scenarios Proactive scenarios
established in the areas where most raw materials of private companies and three islands a
smallholders area available, resulting in high OER (23%) with assumptions Area FFB FFB Areab FFB FFB
made from the target of Economic Transformation Program (ETP) (ETP, 2015). (Mha) yield production (Mha) yield production
(t/ha) (Mt) (t/ha) (Mt)
Palm Order from Strengths (PalmOS): Sumatra
Entisols 1.0 35 35 1.0 35 35
▪ By 2050, 5023 million people (Cork et al., 2005) consume 37 kg cooking
Inceptisols 8.3 32 265.6 6.5 33 214.5
oil/person1 (Corley, 2009), and 172 Mt biofuel/year is needed, of which 84 Mt
biofuel/year is produced in Asia (Cork et al., 2005), with the majority being Ultisols 7.7 35 269.5 1.5 35 52.5
exported to China and India. Kalimantan
▪ Private mills are accessible by smallholders, and FFB is accepted without Alfisols 0.1 27 2.7 0.1 27 2.7
quality control as long as materials are available to run mills. Middle men play Entisols 1.5 37 55.5 1.5 37 55.5
an important role in selling FFB to mills. These practices cause lower OER Inceptisols 0.6 36 21.6 0.6 36 21.6
Mollisols 0.1 32 3.2 0.1 32 3.2
(20%) (Deptan, 2014) due to long distance and storage of FFB before processed
Spodosols 0.6 38 22.8 0.6 38 22.8
in mills.
Ultisols 5.2 31 161.2 3.9 31 120.9
Palm Adaptic Mosaic (PalmAM): Papua
Alfisols 0.1 27 2.7 0.01 27 0.27
▪ The CPO production increases due to increasing CPO demand associating with Entisols 0.3 37 11.1 0.30 37 11.1
low level cooking oil consumption and biodiesel uses. By 2050, 4992 million Inceptisols 0.3 36 10.8 0.30 36 10.8
people (Cork et al., 2005) consume 20 kg cooking oil/person(Corley, 2009). Mollisols 0.01 32 0.32 n.a n.a. n.a.
Approximately, 207 Mt/year biofuel is needed, of which 116 Mt biofuel/year Ultisols 0.5 31 15.5 0.4 31 14
is produced in Asia (Cork et al., 2005). The majority is exported to Asia, Potential areas 26.3 27–38 877.5 16.8 27–38 564.9
particularly China and India, which also raises concerns about environmental a
Potential areas for oil palm plantations in reactive scenarios, this includes areas of
problems.
other agriculture commodities, this may associate with displacement effect.
▪ The close distance transport of FFB and better management causes high OER b
Potential areas for oil palm plantations in proactive scenarios, these areas exclude
(23%) (ETP, 2015).
areas for other commodities.
566 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570

Table 5 potential FFB yield with a resolution of 0.1°. The model calculated yields
Potential CPO production by 2050 in the four PalMA scenarios assuming 20% OER in affected by solar radiation parameterized with slopes, azimuth, and
reactive scenarios and 23% OER in proactive scenarios.
cloudiness index. The model figures out FFB yield under optimal condi-
Island Potential production (Mt) Potential production (% of CPO demand) tions (sufficient water and nutrients, no damages of pest and diseases as
PalmGO and PalmTG and well as uniform planting materials (Hoffman et al., 2014). Due to high
PalmOS PalmAM solar radiation in flat and coastal areas, higher yield would be simulated
CPO CPO PalmGO PalmTG PalmOS PalmAM in peatlands (Hoffmann et al., 2015). Nevertheless, our study excluded
(20% OER) (23% OER) peatlands. The model used in this study does not simulate effect of
Sumatra 114 69 26% 21% 39% 26%
water limitation on oil palm production. Water limitation can decrease
Kalimantan 53 52 12% 16% 18% 20% FFB production (traditionally called trek) and exists in dry seasons, al-
Papua 8 8 2% 2% 3% 3% though Indonesia is characterized by high rainfall. The newest version
Total 176 130 39% 40% 60% 49% of PALMSIM incorporates a water limitation effect. Many regions expe-
rienced periodic drought, but these regions are not within the scope of
the current study. Therefore, the first generation of PALMSIM is still suf-
Furthermore, we reclassified soil types (ordos) following the key to ficient to provide information on national scale (at the level of islands).
soil taxonomy of United States Department of Agriculture (USDA). The
land system map contains soil type data at the level of soil families. 2.5.2. Calculation of FFB and CPO production
After reclassification, we observed that information on peatlands is These overlays resulted in identified yields based on individual soil
not intensively available in the land system map. Thus, we overlaid types. We converted tables of attributes from maps to excel, then omitted
the identified suitable areas from the land system map with peat layers, and calculated average yields per individual soil type. Yield map
maps from Wetlands International Indonesia Program, comprising a for Papua is not available and we applied the average yield per soil type
peat map of Sumatra (Wahyunto et al., 2003), a peat map of Kalimantan in Kalimantan to calculate potential yield in Papua, as this is the nearest
(Wahyunto et al., 2004), and a peat map of Papua (Wahyunto et al., geographical position which has the most available data. Thereafter,
2006). Then we identified suitable areas excluding peatlands. This total potential FFB yields were calculated by multiplying individual soil
stage results in individual maps for each island presenting suitable type areas with the average yields. To calculate CPO production, we mul-
areas based on physical properties. tiplied the FFB yields based on soil types (resulted in Section 2.5.1) with
assumed OER (Section 2.3.2), which are a 20% and a 23% OER of FFB to
2.4.2. Step 2: Selection of areas based on land uses CPO, for reactive and proactive approaches respectively.
We overlaid the individual maps generated in step 1 with land use
maps of the National Soil Agency (BPN) (personal communication, 3. Results
2014). We identified areas complying with the second set of suitability
criteria including areas not used for rice fields (paddy), forests (natural 3.1. Past trends
and production forest), water bodies, industrial areas, and villages/set-
tlement. The land use maps of BPN do not present specific areas classi- Oil palm plantations in Indonesia have expanded from about 6.7 Mha
fied as oil palm plantations. Instead, the maps classified plantations in in 1990 to about 9 Mha in 2013. 95% of the plantations are in Sumatra
general, which could include overall plantations based on Deptan and Kalimantan (Deptan, 2014; personal communication of Tropenbos,
(2014). They are about 7.6 Mha in size, which include 7.4 Mha of land 2012). In Sumatra, oil palm plantations increased rapidly between 1990
for other commodities. These include Cocoa 1.77 Mha, Rubber and 2010, particularly between 1990 and 2005. The rate of increase de-
3.46 Mha, Coffee 1.23 Mha, Nutmeg 0.18 Mha, Nilam 0.03 Mha, Pala clined after 2005. Similarly, in Kalimantan, oil palm has rapidly expanded
0.12 Mha, Vanilla 0.02 Mha, Pinang 0.15 Mha, Lemon grass 0.02 Mha, since 2000. Oil palm plantations increased from b 1 Mha in 1990 to
Siwalan 0.03 Mha, and Sugarcane 0.45 Mha. These are cultivated in Su- 2.9 Mha in 2010. Since 2005, expansions have mostly taken places in Kali-
matra (6.2 Mha), Kalimantan (1.3 Mha), and Papua (0.1 Mha) (Deptan, mantan and in Sumatra. Oil palm plantations have also increased in Papua.
2014). Since 1990, oil palm production areas have increased from 29,000 ha in
1990 to 48,000 ha in 2000, and from 69,000 ha in 2005, and 84,000 ha in
2.4.3. Step 3: Selection of areas out of moratorium and conservation areas 2010 (personal communication of Tropenbos, 2012). (See Fig. 1).
The third set of suitable areas included areas of non-moratorium and Oil palm has increasingly been cultivated in peatlands. In Sumatra,
non-conservation areas. We overlaid maps indicative of moratorium pro- peat conversion increased over time while that of mineral land de-
vided by the National Soil Agency (BPN) (personal communication, 2014) creased (Fig. 2). In Kalimantan, the conversion of peatlands to oil palm
with the map of conservation areas provided by Forestry Department and plantation increased, with a rise in both peatland and in mineral land.
maps generated in step 2. Then, we selected areas of non-moratorium and In the period of 2000 to 2010, oil palm plantations increased from
non-conservation areas, and classified these areas as suitable areas for oil 0.6 Mha to 2.5 Mha in mineral lands. As well, oil palm plantations in
palm plantations. Finally, we selected polygons representing areas of at peatlands increased from 0.02 Mha to 0.3 Mha. In Papua, oil palm plan-
least 1.5 ha to present potentially profitable areas of plantations for small- tations were unchanged in both peatlands and mineral lands with areas
holders. Six maps of potential suitable areas for oil palm plantations in- encompassing 0.002 Mha and 0.05 Mha respectively.
clude two maps (of reactive and proactive scenario) of potential
suitable areas in Sumatra, Kalimantan, and Papua. 3.2. Future trends in palm oil production in Indonesia

2.5. Potential production of fresh fruit bunches and crude palm oil in
Indonesia by 2050
2.5.1. Overlaying the maps of potential areas and yields gathered from
PALMSIM model
We gathered maps of potential yields of PALMSIM models (personal
communication), and overlaid the maps with potential areas for oil
palm plantations in Sumatra and Kalimantan. The PALMSIM model is a
model simulating potential growth and yield. This model plotted
3.2.1. CPO demand
In the PalMA scenarios, by 2050, the global demand of CPO for
cooking oil and biodiesel ranges from 264 to 447 Mt/year. The CPO
demand for cooking oil is higher than that for biodiesel, ranging from
125 to 253 Mt/year and 62 to 194 Mt/year respectively. The larger pop-
ulation in the globalized scenarios (PalmGO PalmTG) causes a larger
demand for CPO for cooking oil, while the consumption levels are as-
sumed to medium to low level. In the regionalized scenario PalmOS, a
high consumption level creates a high demand for CPO for cooking oil;
 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570
Fig. 1. Areas of oil palm production in Indonesia according to Tropenbos (2012) (dotted
lines) and an online database of Department of Agriculture (Deptan, 2014).
although the population is smaller than in the globalized scenarios. On
the other hand, the regionalized scenario PalmOS has the lowest CPO
demand for biodiesel. In the other regionalized scenario (PalmAM),
CPO demand for biodiesel is higher than that of PalmOS. To meet
these demands, large areas of additional plantations are needed. If the
productivity remains at the current level (3.6 t/ha, in average), an area
is needed that is about 10 fold the current plantation area (Table 3).
Therefore, population and consumption levels play important roles in
CPO demand and the areas required for plantations, unless productivity
increases.
3.2.2. Potential suitable areas and potential fresh fruit bunches yield in
Indonesia
The potential suitable areas range from 17 Mha (in proactive
scenarios) to 26 Mha (in reactive scenarios). The difference associ-
ates with exclusion and inclusion of other agricultural areas in pro-
active and reactive scenarios respectively. These areas are smaller
than those are required to meet the CPO demand (Table 3), if FFB
productivity (yield per hectare) and oil extraction remains at the
current level. In the proactive scenarios (PalmTG and PalmAM),
smaller areas are available for oil palm as these scenarios aim to
maintain the commodities (7.2 Mha) and exclude hilly areas
(2 Mha). These areas cannot be seen clearly in maps. We therefore
provide maps for reactive scenarios (Figs. 3, 4, and 5). The dominant
soils are Ultisols and Inceptisols. These soils are highly productive
according to the PALMSIM model. The mineral lands can produce
565–880 Mt FFB.
Fig. 2. Area of peat land (left) and mineral soils (rights) that were converted to oil palm plantations in Sumatra, Kalimantan, and Papua during the period 1990–2010.
(Source: Tropenbos, 2012)
567
3.2.3. Potential CPO production in Indonesia
Potential CPO production is defined here as the CPO produced from
projected FFB yields in areas deemed suitable. Indonesia can produce
large amounts of FFB and CPO, meeting a considerable part of the inter-
national demand for CPO. By 2050, the Indonesian CPO production
amounts range from 130 Mt in the proactive scenarios to 176 Mt in
the reactive scenarios. The total CPO production capacity in Indonesia
ranges from 39 to 60% of the global demand in reactive scenarios, to
40 to 49% in proactive scenarios. Indonesia can supply 39–40% of the
global demand in PalmGO and PalmTG. Assuming that Indonesia merely
exports CPO to Asia, Indonesia can supply 49% of the demand for CPO in
PalmAM and 60% in PalmOS (Table 5).
4. Discussion
This study explored future trends of palm oil production in Indonesia
considering CPO demand for cooking oil and biodiesel. It remains uncer-
tain as to how Indonesia would shift towards more sustainable palm oil
development, considering environmental and social problems associat-
ed with palm oil development (Inubushi et al., 2003, Reijnders and
Huijbregts, 2008; Koh and Wilcove, 2009; Disdier et al., 2013,
Dhiaulhaq et al., 2014, Sumarga et al., submitted for publication).
This study presented global scenarios (MA) that consider the current
national conditions in Indonesia. The existing MA scenarios include
assessments at the global, regional, and local levels (Kok et al., 2007).
Interpretation of global scenarios at the local level is important. Such
downscaling were conducted by others, for instance, South Africa (van
Vuuren et al., 2010) and Jakarta Bay (Arifin, 2004). Local interpretations
provide insight in the effects of global development to local-national
development, which is applicable to the palm oil development in
Indonesia presented in this study. The scenario analysis could facilitate
decision making to explore plausible trajectories on how to develop
the palm oil sector in Indonesia under restricting conditions (no further
deforestation and peat conversion). However, it remains uncertain as to
how Indonesia would relocate existing oil palm plantations. This re-
quires further stakeholder consultation.
In this study, it is assumed that the demand for palm oil increases
due to population growth, consumption levels, and renewable energy
demand as also suggested by EIA (IEA, 2010, 2011). The increasing de-
mand would require expansion and/or intensification (to increase
yield). With the absence of policy enforcement to avoid deforestation
and peat conversion, this trend would indicate a threat to peatlands
in Indonesia. This threat is an agreement with our observation that
in the period of 2000 to 2010 (the past trend), palm oil production
in Indonesia was associated with increasing expansion of oil palm
plantations in peatlands. Similarly, Miettinen et al. (2012) expects
increasing use of peatlands for oil palm plantation in their projec-
tion up to 2020.
568 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570

Fig. 3. Potential areas for oil palm cultivation in Sumatra in the reactive scenarios.

Our scenarios explored palm oil production without further defores-
tation and peat conversion. We found this constraint can maintain
Indonesia's place within global palm oil production rates. Hence, this
strategy could enhance the role played by Indonesia in climate change
mitigation. Indonesia can expand oil palm plantations up to 17 Mha
where about 7.6 Mha are maintained for other agricultural commodities
(Cocoa, Rubber, Coffee, Nutmeg, Nilam, Pala, Vanilla, Pinang, Lemon
grass, Siwalan, and Sugarcane), or up to 26 Mha where aforementioned
commodities compete with oil palm. This assumption stands for rubber
given the current low price of latex, which increases the interest to con-
vert rubber plantations to oil palm plantations (Koh and Wilcove, 2009).
Our scenarios do not project rice demand within Indonesia as supplied
by current rice fields, nor how this would compete with oil palm. How-
ever, to avoid error, we exclude rice fields from convertible areas for oil
palm plantations. By 2050, the requirement for rice fields is projected to
be approximately 11 Mha. This area requirement would be fulfilled
through the establishment of new rice fields (sawah cetak) in Java, Sula-
wesi, and Eastern parts of Indonesia (BIN, 2012), with the rice fields cov-
ering 12 Mha (Deptan, 2014). In addition, this study is not able to
address competing uses between potential area for oil palm plantations
and industrial forest (Acacias and Eucalyptus plantations). This is result-
ed from lack of data about dynamic of changes of assigned agricultural
areas to industrial forestry purpose; although we exclude forest zone
areas.
In terms of intensification, this study found that the potential yields
are double than the current yields (18 t FFB/ha with 20% extracted CPO).
The potential yield implies the required increase yield approximately
0.9 t FFB/ha FFB annually up to 2050, while recent Indonesia palm oil

Fig. 4. Potential areas for oil palm cultivation in Kalimantan in the reactive scenarios.
 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570
Fig. 5. Potential areas for oil palm cultivation in Papua in the reactive scenarios.
productivity has been stagnant (Sheil et al., 2009). The low yield may be
associated with the use of infertile soil like peatlands, bad seedlings (not
from tenera seedlings), insufficient nutrients, pests, a prolonged dry
season, and long transport periods from field to mill (Sheil et al.,
2009). Closing the gap of the current to potential yield requires some
efforts. These include choosing suitable lands (lands indicated in this
study, for instance, see Table 4, Figs. 3, 4, and 5), replanting with good
planting materials, providing sufficient nutrients, sufficient water, and
preventing threats of both pests and diseases as suggested by PALMSIM
model. However, solar radiation, as one parameter of the model, cannot
be controlled. Sufficient nutrients can be monitored by foliar analysis;
however this requires knowledge and technology (accessible laborato-
ries) in all levels plantations (big companies and smallholders). This
measure may not yet be applicable at the smallholder level in Indonesia.
Sufficient water would be limited in a (prolonged) dry season, when pro-
duction decreases (locally referred to as trek season). Replanting with
good seedlings would be the barrier for smallholders to improve their
plantations, because of the inevitable high costs which smallholders can-
not typically afford.
Although the potential yield was observed in some trials and compa-
nies plantations, when considering the aforementioned hurdles, this
study may overestimate the yield that could potentially be obtained.
In practice, the highest attainable CPO yield was 6–8 t/ha at large scales
in leading plantation groups in Indonesia and Malaysia (Donough et al.,
2009). Meanwhile the maximum theoretical yield is 18.2 t/ha (MPOB,
2014). While maximizing yields, it is important to minimise detrimental
impacts to the surrounding environment (Huth et al., 2014). Research is
being done to reach the potential yield by organizations such as the
Indonesian Oil Palm Research Institute (PPPK). PPKS aims to increase
palm oil productivity about 9 t/ha (from 35 t FFB with 26% extractable
oil) (PPKS, 2014). Oil palm production is more responsive to best man-
agement practices than to soil types; although soil type can be a limiting
factor (Fairhurst and McLaughlin, 2009). In order to improve produc-
tion, evaluation of limiting soil factors and yield is needed (Hoffman
et al., 2014). Potential FFB yields differ with soil type, implying that
best management practices also are soil-specific. The potential suitable
soils are marginal soils, which can be managed for oil palm plantations
(Paramantan, 2013). This study assumes that smallholders produce FFB
at a comparable rate to large companies. However, the plantations of
smallholders may have much lower productivity when compared to
569
large plantations, unless smallholders are trained in best management
practices including improvement of plant materials in replanting
programmes. Some planting materials can yield between 6.9 and 8.4 t
CPO/ha/y (PPKS website).
5. Conclusion
Our analyses of past trends indicate that expansion of oil palm plan-
tations in peatlands are more likely in the absence of policies to avoid
deforestation and peat conversion, such as a moratorium. During the
past 20 years oil palm production areas have been expanding fast in
Sumatra and Kalimantan. In Sumatra, oil palm expanded in particularly
in peatland areas and the rate of expansion decreased in mineral lands;
while in Kalimantan, the rate of oil palm expansion increased in both
peatlands and mineral lands. Oil palm cultivation in Papua is not yet
developed. In 2010, oil palm plantations covered about 3 Mha peatlands
and about 5 Mha in mineral lands in Indonesia, of which most in Sumatra
and Kalimantan.
We presented scenarios that can be considered indications of the
potential to produce sustainable palm oil in Indonesia. Our scenarios as-
sume that oil palm can expand, but not on forested land, conservation
areas, peatland or rice fields. Thus our scenarios imply effective sustain-
able land use planning. Our scenarios furthermore assume that yields
are higher than they are today. This can only be reached through imple-
mentation of better varieties and best management practices.
We conclude that Indonesia can stay a major CPO producing country
while avoiding further deforestation and peatland conversion. The
demand for CPO from Indonesia could increase by 264 to 447 Mton by
2050, following the international demand for cooking oil and biodiesel.
In our scenarios, Indonesia produces 130 to 176 Mton CPO, from 565 to
880 Mton FFB, cultivated on 17 to 26 Mha of land that we consider suit-
able. Thus our scenarios indicate that Indonesia may meet 39–60% of the
international CPO demand, while avoiding further deforestation and
expansion in peatlands.
This study finds that socio-economic development and environmen-
tal management approaches could influence international demand and
CPO production in Indonesia. Nevertheless, Indonesia can maintain its
place in global palm oil production without further deforestation and
peat conversion. This requires enforcement of policies regarding zero
deforestation (moratorium) and banning of peat conversion. In the
570 D. Afriyanti et al. / Science of the Total Environment 557–558 (2016) 562–570
absence of policy enforcement, future oil palm plantations would
further expand into forest and peatlands. Moreover, given the existing
condition of smallholders, the future remains uncertain as to how
Indonesia will close the gap between potential yields and actual yields.
Our scenarios assume that yields are higher than they are today. This re-
quires implementation of improved oil palm varieties and best manage-
ment practices. As such, our scenarios presented in this study can be
considered indications of the potential to produce sustainable palm oil
in Indonesia. It could serve as a basis for discussions on how to over-
come technological barriers, awareness rising, and costs of sustainable
production.
Acknowledgements
This study was funded through the Integrated Research Fund
(INREF) of Wageningen University, and part of the SUSPENSE project.
We would like to thank Badan Pertanahan Nasional Indonesia Pro-
gramme for supplying us spatial data of land system, land use, and mora-
torium areas. We also extend our gratitude to the Wetlands International
Indonesia program for supplying us peatland maps, and the Forestry De-
partment for supplying us conservation areas maps. We thank Prof. Lars
Hein, dr. ir Maja Slingerlands for invaluable inputs throughout the
study; to Munir Hoffman. M.A, MSc for the potential yield maps; and to
Mohammad Zuhdi, MSc for his assistance for ArcGIS application.
References
Alcamo, J., van Vuuren, D., Cramer, W., Alder, J., Bennett, E., Carpenter, S., Christensen, V.,
Foley, J., Maerker, M., Masui, T., Morita, T., O'Neill, B., Peterson, G., Ringler, C.,
Rosegrant, M., Schulze, K., 2003. Changes in ecosystem services and their drivers
across the scenarios. Ecosystems and Human Well Being. Four Scenarios. Island Press.
Arifin, Z., 2004. local millennium ecosystem assessment: condition and trend of the great-
er Jakarta bay ecosystem. The Ministry of Environment, Republic of Indonesia,
Jakarta.
Asmaryan, H., 2012. Environmental consequences of palm-oil production in Southeast
Asia: a scenario analysis of past and future trends. (MSc Thesis).
Basiron, Y., 2002. Palm oil and its global supply and demand prospect. Oil Palm Ind. Econ.
J. 2 (1).
BIN, 2012. Hari Pangan Sedunia: Ancaman Krisis Dalam Kemandirian Pangan Indonesia.
The World Food Day: Threats of Crisis in Independence of Indonesia Food (Retrieved
March 2, 2015, from www.bin.go.id/wawasan/detil/141/3/25/09/2012/hari-pangan-
sedunia-ancaman-krisis-dalam-kemandirian-pangan-indonesia).
Carrasco, L.R., Larrosa, C., Milner-Gullard, E.J., Edwards, D.P., 2014. A double-edged sword
for tropical forest: contrary to expectation, high-yielding tropical crops may cause
forest loss in the tropics. Science 364 (6025).
Cork, S., Peterson, G., Petschel-Held, G., Alcamo, J., Alder, J., Bennett, E., Carr, E.R., Deane, D.,
Nelson, G.C., Ribeiro, T., Butler, C., Mendiondo, E.M., Oluoch-Kosura, W., Zurek, M.,
2005. Four scenarios. Ecosystem and Human Well-BeingScenarios. Island Press.
Corley, R.H.V., 2009. How much palm oil do we need? Environ. Sci. Pol. 12, 6.
Deptan, 2014. Pusat Data dan Informasi Pertanian. Center for agriculture data and in-
formation. (29 January 2014). from http://aplikasi.deptan.go.id/bdsp/index-e.asp.
Dhiaulhaq, A., Gritten, D., De Bruyn, T., Yasmi, Y., Zazali, A., Silalahi, M., 2014. Transforming
conflict in plantations through mediation: lessons and experiences from Sumatera,
Indonesia. Forest Policy Econ. 41, 22–30.
Disdier, A.-C.l., Marette, S.p., Millet, G., 2013. Are consumers concerned about palm oil?
Evidence from a lab experiment. Food Policy 43, 180–189.
Donough, C.R., Witt, C., Fairhurst, T.H., 2009. Yield intensification in oil palm plantations
through best management practice. Better Crops 93 (1).
ETP, 2015. EPP 4: increasing the Oil Extraction Rate (OER). Retrieved 24 June, 2015, from
http://etp.pemandu.gov.my/Palm_Oil-@-Palm_Oil_-%E2%97%98-_Rubber_-_EPP_4-;_
Increasing_the_Oil_Extraction_Rate_%28OER%29.aspx.
Fairhurst, T., McLaughlin, D., 2009. Sustainable oil palm development on degraded land in
Kalimantan.
Gingold, B., Rossenbarger, A., Muliastra, Y.I.K.D., Stolle, F., Sudana, I.M., Manessa, M.D.M.,
Murdimanto, A., Tiangga, S.B., Madusari, C.C., Douard, P., 2012. How to identify
degraded land for sustainable palm oil in Indonesia. World Research Institute and
Sekala.
Hirawan, F.B., 2011. The impact of palm oil plantations on Indonesia's rural economy. Ag-
ricultural Development, Trade and Regional Cooperation in Developing East Asia Vol.
66. ERIA, Jakarta.
Hoffman, M.P., Castaneda Vera, A., Wijk, M.T.v., Giller, K.E., Oberthur, T., Donough, C.,
Whitbread, A.M., 2014. Simulating potential growth and yield of oil palm (Elaeis
guineensis) with PALMSIM: model description, evaluation, and application. Agric.
Syst. 131, 10.
Hoffmann, M.P., Donough, C., Oberthür, T., Castaneda, A., Wijk, M.T.V., Lim, C.H., Asmono,
D., Samosir, Y., Lubis, A.P., Moses, D.S., Whitebread, A.M., 2015. Benchmarking yield
for sustainable intensification of oil palm production in Indonesia using PALMSIM.
Planter 91 (1067), 16.
Hooijer, A., Silvius, M., Wösten, H., Page, S., 2006. PEAT-CO2, Assessment of CO2 Emission
From Drained Peatland in SE Asia. Delft Hydraulic.
Huth, N.I., Banabas, M., Nelson, P.N., Webb, M., 2014. Development of an oil palm
cropping systems model: lessons learned and future direction. Environ. Model.
Softw. 62, 9.
IEA, 2010. Deploying Renewables in Southeast Asia: Trends and Potentials. M. B.
Samantha Olz, France, International Energy Agency.
IEA, 2011. Technology Roadmap: Biofuel for Transport. International Energy Agency,
France.
Inubushi, K., Furukawa, Y., Hadi, A., Purnomo, E., Tsuruta, H., 2003. Seasonal changes of
CO2, CH4 and N2O fluxes in relation to land-use change in tropical peatlands located
in coastal area of South Kalimantan. Chemosphere 52 (3), 6.
Koçar, G., Civaş, N., 2013. An overview of biofuels from energy crops: current status and
future prospects. Renew. Sust. Energ. Rev. 28, 900–916.
Koh, L.P., Wilcove, D.S., 2009. Oil palm: disinformation enables deforestation. Trends Ecol.
Evol. 24 (2), 67–68.
Kok, K., Biggs, R.O., Zurek, M., 2007. Methods for developing multiscale participatory
scenarios: insights from Southern Africa and Europe. Ecol. Soc. 13 (1), 8.
Lendle, D., Schaus, M., 2010. Sustainability criteria in the EU renewable energy directive:
consistent with WTO rules? ICTSD Project on WTO Jurisprudence and Sustainable
Development
Miettinen, J., Hoojer, A., Tollenaar, D., Page, S., Malins, C., Vernimmen, R., Shi, C., Liew, S.C.,
2012. Historical analysis and projection of oil palm plantation expansion on peatland
in Southeast Asia. Vol. 54. Indirect Effect of Biofuel Production, Washington DC.
Miranowski, J., Rosburg, A., 2012. Long-term biofuel projections under different oil price
scenarios. AgBioforum 16 (1), 9.
MPOB, 2014. Malaysian palm oil industry. Retrieved 29 January, 2014, from http://www.
palmoilworld.org/about_malaysian-industry.html.
Paramantan, S., 2013. Managing marginal soil for sustainable growth of oil palms. J. Oil
Palm Environ.
PPKS, 2014. Bagaimana cara meningkatkan produktifitas kelapa sawit? How to improve
productivity of oil palm? from http://www.iopri.org/component/content/article/
74-faq/203-bagaimana-cara-meningkatkan-produktifitas-kelapa-sawit.html
Reijnders, L., Huijbregts, M.A.J., 2008. Palm oil and the emission of carbon-based green-
house gases. J. Clean. Prod. 16, 6.
RSPO, 2014. Factsheet: roundtable on sustainable palm oil. Retrieved April 16, 2014,
2014, from http://www.rspo.org/file/RSPO_factsheet_120705_25july.pdf.
Sheil, D., Casson, A., Meijaard, E., Noordwijk, M.v., Gaskell, J., Sunderland-Groves, J., Wertz,
K., Kanninen, M., 2009. The impacts and opportunities of oil palm in Southeast Asia:
what do we know and what do we need to know? Occasional Paper No. 51. CIFOR,
Bogor, Indonesia
Sumarga, E., L. Hein, A. Hooijer and R. Vernimmen. "Hydrological and economic impacts of
oil palm cultivations in Indonesia peatlands." (submitted for publication).
Tan, K.T., Lee, K.T., Mohamed, A.R., Bhatia, S., 2009. Palm oil: addressing issues and
towards sustainable development. Renew. Sust. Energ. Rev. 13 (2), 420–427.
USDA, 2014. Indonesia palm oil exports by year. from http://www.indexmundi.com/
agriculture/?country=id&commodity=palm-oil&graph=exports.
van Vuuren, D.P., Smith, S.J., Riahi, K., 2010. Downscaling socioeconomic and emis-
sions scenarios for global environmental change research: a review. WIREs
Clim. Change 1, 12.
Wahyunto, Ritung, S., Subagjo, H., 2003. Peta Luasan Sebaran Lahan Gambut dan
Kandungan Karbon di Pulau Sumaetra 1990–2002. Maps of Area of Peatland Distribu-
tion and Carbon Content in Sumatera 1990–2002. Wetlands International-Indonesia
Programme &Wildlife Habitat Canada (WHC), Bogor, Indonesia.
Wahyunto, Ritung, S., Subagjo, H., 2004. Peta Sebaran Lahan Gambut, Luas, dan
Kandungan Karbon di Kalimantan. Map of Peatland Distribution Area and Carbon
Content in Kalimantan 2000–2002. Wetlands International-Indonesia Programme
&Wildlife Habitat Canada (WHC), Bogor, Indonesia.
Wahyunto, Heryanto, B., Bekti, H., Widiastuti, F., 2006. Peta -peta Sebaran Lahan Gambut,
Luas, dan Kandungan Karbon di Papua. Maps of Peatland Distribution, Area, and
Carbon Content in Papua. 2000–2001. Wetlands International-Indonesia Programme
&Wildlife Habitat Canada (WHC), Bogor, Indonesia.