• Biodiesel;
  • Biofuels;
  • LCA;
  • Meta-analysis;
  • Palm oil


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

This work reviews and performs a meta-analysis of the recent life cycle assessment and flow analyses studies palm oil biodiesel. The best available data and information are extracted, summarized, and discussed. Most studies found palm oil biodiesel would produce positive energy balance with an energy ratio between 2.27 and 4.81, and with a net energy production of 112 GJ ha−1y−1. With the exception of a few studies, most conclude that palm oil biodiesel is a net emitter of greenhouse gases (GHG). The origin of oil palm plantation (planted area) is the foremost determinant of GHG emissions and C payback time (CPBT). Converting peatland forest results in GHG emissions up to 60 tons CO2equivalent (eq) ha−1y−1 leading to 420 years of CPBT. In contrast, converting degraded land or grassland for plantation can positively offset the system to become a net sequester of 5 tons CO2eq ha−1y−1. Few studies have discussed cradle-to-grave environmental impacts such as acidification, eutrophication, toxicity, and biodiversity, which open opportunity for further studies. Integr Environ Assess Manag 2013; 9: 134–141. © 2012 SETAC


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

Oil palm (Elaeis guinensis) is an energy crop that has potential as biodiesel feedstock for the reasons of the supply abundance and the fuel yield per hectare (Basiron 2007; Sumathi et al. 2008). Palm oil biodiesel is derived by a process called transesterification, where the crude palm oil feedstock is mixed with alcohol, usually methanol, in the presence of sodium hydroxide as a catalyst, to produce the fatty acid methyl ester or biodiesel (Fukuda et al. 2001).

Considering the prospect of palm oil biodiesel, and driven by the mandatory use of biofuels in more and more countries, major producing countries such as Indonesia and Malaysia have set up their national targets for the expansion of palm oil biodiesel production. Along with the increasing national targets for producing and using palm oil biodiesel as a renewable energy sources (e.g., Malaysian Biofuel Industries Act 2007 and Indonesian Presidential Instruction 2007), concerns over environmental impacts triggered by the massive production of palm oil are also rising.

Various tools and methods for assessing and benchmarking environmental impacts of different product systems have been developed. Life cycle assessment (LCA) provides a methodology to assess the potential environmental impacts and resources consumption throughout a product life cycle, from material acquisition to end use (ISO 2006; Guineé et al. 2011). Flow analysis methods such as substance flow analysis (SFA) and energy flow analysis (EFA), on the other hand, are life cycle approaches to analyze flows and stocks of a single substance (e.g., C) or a coherent group of substances (e.g., greenhouse gases [GHG]).

This study aims to review studies that apply LCA or life cycle approach flow analyses (EFA/SFA) methodologies of palm oil biodiesel supply chain. As a rule, only studies based on a life cycle analysis approach to estimate the environmental impacts of palm oil biodiesel supply chain are reviewed. In addition, only articles written in English and with good and reliable supplemental data and references were selected. The review begins with a discussion of main life cycle analysis features and approaches to the studies, such as scopes, functional units, assumptions, reference systems, etc., followed by an analysis, comparison, and interpretation of important data information, to identify the shortcoming of the current knowledge and future research needs.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

Of the 23 studies that were reviewed, 22 are articles published in scientific journals and 1 is a technical report. Detailed information for each of the reviewed studies is provided in the Supplemental Data. Although this study intended to cover a time period of at least 10 years, the oldest found literature was published in 2007. This reveals the fact that palm oil biodiesel has only become the subject of comprehensive LCA in recent years. It is observed, however, that there is a growing interest in such studies with the increasing interest in using palm oil as a renewable energy feedstock.

Geographical contexts and types of reviewed studies

Notably, almost all of the life cycle studies of palm oil biodiesel were conducted in Southeast Asia, which is home to the world's largest palm oil producing countries like Indonesia and Malaysia. A small portion of the studies reviewed here were set in South America and Africa. This fact show how the locally relevant renewable energy crops have become the subject of interest in developing countries, as a contrast to the previous finding (Larson 2006).

Most of the studies conducted in Southeast Asia (12) were set in Malaysia. Even though Indonesia is the world's largest palm oil producer, only a few studies (4) were set in this country. In contrast, although palm oil production in Thailand represents only 2% of the joint product of Malaysia and Indonesia, 6 studies were set in this country. In the Americas, there were 3 studies set in Brazil and Colombia. In Africa, there was 1 study conducted in Cameroon.

With respect to the type of study, a majority of the studies (17) limited the assessment to the energy and/or GHG balances without considering any further environmental impact categories. This approach is usually supported by motivations of energy efficiency and climate change mitigation of the development of renewable fuels. Only 6 studies go beyond the energy and/or GHG balances.

Scopes of reviewed studies

The palm oil biodiesel production system assessed by the reviewed studies can be divided in 5 major steps (Figure 1): A) land preparation, B) oil palm plantation, C) palm oil mill, E) conversion to biodiesel fuel, and E) use of biodiesel fuel (BDF). Depending on which unit processes are considered in the assessment, studies are classified into 4 categories:

  • 1.
    Cradle-to-grave (Cr-to-Gr): For studies that include all the steps from Step A to Step E, or from Step B to Step E in the case where it is claimed that the plantation is located on land that was cleared several decades ago (>30 y). The term “cradle-to-grave” refers to what is commonly used as “well-to-wheel” in fuels (including biofuels) studies. It is arguable whether land preparation is the “cradle” of the life cycle or just an input to the plantation. In this study, we define it as cradle, considering the importance of its impact to the whole life cycle of palm oil biodiesel (Wicke et al. 2008).
  • 2.
    Cradle-to-gate (Cr-to-Gt): For studies that begin from Step A but do not end in Step E.
  • 3.
    Gate-to-grave (Gt-to-Gr): For studies that end in Step E but do not begin from Step A, except in the case where it is claimed that the plantation is located on land that was cleared several decades ago (>30 y) and start from Step B.
  • 4.
    Gate-to-gate (Gt-to-Gt): For all studies other than those described in points 1, 2, and 3 above.
thumbnail image

Figure 1. Palm oil biodiesel production system. (FFB = fresh fruit bunch, CPO = crude palm oil, POME = palm oil mill effluent, BDF = biodiesel fuel)

Download figure to PowerPoint

Almost 60% of the studies (13 studies) were carried out using the gate-to-gate approach, and only 10% (3 studies) were carried out using the cradle-to-grave approach. The remaining 30% of the studies were carried out using the cradle-to-gate or the gate-to-grave approach.

Land use change (LUC) is excluded in most gate-to-gate studies, mainly due to the difficulty in quantifying the origin of the existing palm oil plantation (Yusoff and Hansen 2007). However, an attempt has been undertaken to account for the LUC impact from mixed agricultural land in Thailand (Siangjaeo et al. 2011). Different approaches were undertaken by respective authors in relation to incorporation of LUC in the life cycle impact assessment.

Key methodological choices

Functional units

We use 4 different approaches in defining the functional unit. We classify these approaches into 4 groups: agricultural land area, output mass, output energy, and output service (Table 1). In the area of “agricultural land” group, results of the assessment are normalized in the unit of hectare per year of plantation. In the “output mass” group, results of the assessment are normalized in the unit of kg or ton of BDF. Although there is 1 study (Siangjaeo et al. 2011) that uses related output volume (in liters of BDF), we included it in this group. In the “output energy” group, results of the assessment are normalized in the unit of MJ of BDF or CPO. In the “output service” group, results of the assessment are normalized in 100 km of transportation by light duty diesel vehicle.

Table 1. Studies by functional unit
Agricultural areaOutput massOutput energyOutput service
de Souza et al. 2010Siangjaeo et al. 2011Choo et al. 2011Pleanjai et al. 2009
Fargione et al. 2008Papong et al. 2010Hassan et al. 2011 
Germer and Suerborn 2008Schmidt et al. 2010Achten et al. 2010 
Kamahara et al. 2010Stichnothe and Schuchardt 2010Puah et al. 2010 
 Lam et al. 2009Yáñez et al., 2009 
 Pleanjai and Gheewala 2009Wicke et al. 2008 
 Yee et al. 2009  
 Reijnders and Huijbregts 2008  
 Yusoff and Hansen 2007  
Reference systems

Table 2 shows which reference systems the study is compared. Twelve of the 23 reviewed studies do not compare the assessed system to a reference system. Rapeseed is used in 3 studies, Jatropha biodiesel in 2 studies, and fossil fuel in 2 studies. Other substitutes, such as sugarcane ethanol and soybean biodiesel, are also mentioned in some studies. Studies that apply consequential LCA compare the assessed systems to different consequential scenarios. In Schmidt (2010), for instance, the scenarios include increasing agricultural area and increasing agricultural yield. In Stichnothe and Schuchardt (2010), the scenarios include dumping empty fruit bunch (EFB) and composting EFB.

Table 2. Studies by reference system
Fossil fuelRapeseed biodieselJathropa biodieselPalm biodiesel in different scenarioOther
Achten et al. 2010Papong et al. 2010Lam et al. 2009Schmidt 2010Fargione et al. 2008
Pleanjai et al. 2009Schmidt 2010Papong et al. 2010Stichnothe and Schuchardt 2010Gibbs et al. 2008
 Yee et al. 2009 Siangjaeo et al. 2010Zah et al. 2007
   Wicke et al. 2008 

Considering that palm oil biodiesel pathway delivers some co-products (e.g., palm kernel oil, glycerin), it is found that accounting for the impact of substituted products in the reference systems is still not considered in many of the existing studies with the exception of Achten et al. (2010). Furthermore, the alternative land use change impact of the substitute biodiesel is not widely adopted as a reference system except in Siangjaeo et al. (2011).


A majority of the reviewed studies use allocation by mass, usually for the palm kernel and glycerin co-products. In recent studies (Achten et al. 2010; Schmidt 2010), there are attempts to avoid allocation by using system boundary expansion and apply substitution method.

Impact category and characterization

Climate change or global warming potential associated with GHG emissions is the impact category addressed in a majority of the reviewed studies. Other impacts categories that are also frequently addressed include potentials of toxicity, eutrophication, acidification, and biodiversity loss. Various methodologies have been used to categorize the environmental impacts as well as to quantify the physical flow into each characterized impact. EcoIndicator 99 was used in Zah et al. (2007), Yusoff and Hansen (2007), and Puah et al. (2010). The Intergovernmental Panel on Climate Change guidelines is used in Siangjaeo et al. (2011), Achten et al. (2010), Germer and Sauerborn (2008), and Wicke et al. (2008) in quantification of the GWP. Other guidelines such as CML 2001 method, the Danish EDIP97 method, and Swiss UPB06 have been used by Stichnothe and Schuchardt (2010), Schmidt (2010), and Zah et al. (2007), respectively.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

Energy balance

Generally, all studies agree that palm oil biodiesel would produce positive energy balance, indicating it as a feasible feedstock for biodiesel. The results from the reviewed studies show a net energy ratio (NER) between 2.27 and 4.81. Because the reviewed studies differ in their system boundaries and normalization is complicated, averaging the value for NER is inappropriate.

Concerning the net energy production (NEP), which is the difference between energy output and energy inputs, the average value of the reviewed studies (normalized in the unit of GJ ha−1y−1) is 112 with a standard deviation of 24. Data are shown in Table 3.

Table 3. NER and NEP of selected studies
NrAuthorNERNEP (original)NEP (normalized)a
  • a

    Normalization factor: 5 ton ha−1y−1 and 38 MJ kg−1.

1Achten 20102.61  
2de Souza et al. 20102.338.40E + 04 MJ ha−1y−184 GJ ha−1y−1
3Kamahara et al. 20103.109.80E + 04 MJ ha−1y−198 GJ ha−1y−1
4Papong et al. 20102.482.40E + 01 MJ kg−1120 GJ ha−1y−1
5Lam et al. 20092.27  
6Pleanjai and Gheewala 20092.42  
7Yáñez et al., 20094.812.94E + 01 MJ kg−1147 GJ ha−1y−1
8Yee et al. 20093.532.23E + 01 MJ kg−1112 GJ ha−1y−1
Minimum 2.27 112 GJ ha−1y−1
Maximum 4.81 24 GJ ha−1y−1

The variety of NER and NEP results depends on several technical factors including yields, amount of fertilizer used, and mill efficiency. For instance, in the study of de Souza et al. (2010) less fertilizer and fuel inputs are used in the plantation stage when compared to that of Yee et al. (2009) or Pleanjai and Gheewala (2009). Moreover, pesticide use is not included in the study of Yee et al. (2009). Apart from these technical factors, methodological issues in calculation, such as system boundaries and allocation procedure, also play roles.

Increasing of NER and NEP is still possible through improvement in the processes. Kamahara et al. (2010) claim that NER can be improved up to 8.0 by combining several process improvement options (e.g., using all the biomass residue, glycerin co-product, and biogas for energy production, using glycerin, and increasing the FFB yield).

GHG emissions and C payback time

The reviewed studies use different methodologies in calculating GHG emissions. We classify these approaches into 3 groups (Figure 2):

  • 1.
    Those that emphasize exclusively C-loss caused by LUC as GHG emissions (Fargione et al. 2008; Germer and Sauerborn 2008).
  • 2.
    Those that calculate the life cycle GHG emissions but exclude C-debt from LUC. Authors in this group used 2 approaches:
    • (a)
      Subtracting CO2 assimilation in plantation from the total life cycle GHG emissions (Lam et al. 2009; Yee et al. 2009). This way, GHG emissions are total GHG emitted minus assimilated CO2 in palm plantation; and
    • (b)
      Ignoring CO2 assimilation in plantation. This way, GHG emissions are the sum of GHG emitted. (de Souza et al. 2010; Achten et al. 2010). Although Achten et al. (2010) also calculated GHG emission from LUC, the purpose was only for calculating C payback time (CPBT).
  • 3.
    Those that calculate the cradle-to-grave (LUC up to BDF production) or cradle-to-gate (LUC up to BDF use) GHG emissions by distributing the C-loss over a certain time span (25–100 y) (Wicke et al. 2008; Reijnders and Huijbregts 2008).
thumbnail image

Figure 2. GHG emissions calculations.

Download figure to PowerPoint

Table 4 depicts the summary of GHG calculation of selected reviewed studies. The methods of calculating the GHG emissions in the reviewed studies vary widely and thus result in completely different conclusions.

Table 4. GHG emissions of selected studies
In original unitsNormalized in ha−1y−1
  • a

    See approaches referred to in Figure 2.

1Fargione et al. 20081610 t CO2eq ha−1 50 y−112 t CO2eqFrom lowland forest
   3000 t CO2eq ha−1 50 y−160 t CO2eqFrom peatland forest
2Germer and Sauerborn 20081−135 t CO2eq ha−1 25 y−1−5 t CO2eqFrom grassland
   650 t CO2eq ha−1 25 y−126 t CO2eqFrom nonpeat forest
   1300 t CO2eq ha−1 25 y−152 t CO2eqFrom peatland forest
3Lam et al. 20092.a−6.9 t CO2eq t−1BDF−34 t CO2eqNo LUC considered
4Yee et al. 20092.a−1796 t CO2eq t−1BDF−9 kt CO2eqNo LUC considered
5Achten et al. 20102.b69 g CO2eq MJ−1BDF2 t CO2eqNo LUC considered
6de Souza et al. 20102.b1437 kg CO2eq ha−1 y−11 t CO2eqNo LUC considered
7Choo et al. 20112 b33 g CO2eq MJ-1BDF1 t CO2eqNo LUC considered
8Siangjaeo et al. 20113−0.60 kg CO2eq l−1BDF−3 t CO2eqFrom abandoned land
   −0.73 kg CO2eq l−1BDF−3 t CO2eqFrom other plantation
9Reijnders and Huijbregts 2008311.2 t CO2eq t−1 BDF56 t CO2eqMedian value
10Wicke et al. 20083−51 g CO2eq MJ−1BDF−2 t CO2eqFrom degraded land
   107 g CO2eq MJ−1BDF4 t CO2eqFrom nonpeat forest
   391 g CO2eq MJ−1BDF14 t CO2eqFrom peatland forest
11Hassan et al. 20113400 g CO2eq MJ−1BDF16 t CO2eqFrom primary forest
   320 g CO2eq MJ−1BDF13 t CO2eqFrom 2nd.forest
   100 g CO2eq MJ−1BDF4 t CO2eqFrom grassland
   −54 g CO2eq MJ−1BDF−2 t CO2eqFrom degraded land

Based on the data in Table 4, we distinguish these results into 2 groups:

  • 1.
    Studies that conclude that palm oil biodiesel is a net sequester of CO2 from the atmosphere. Included in this group are results from Lam et al. (2009), Yee et al. (2009), Siangjaeo et al. (2011) and some results from Germer and Sauerborn (2008) and Wicke et al. (2008). In Germer and Sauerborn (2008), Wicke et al. (2008), and Siangjaeo et al. (2011), this fact is attributed to the practice of cultivating oil palms in former degraded land, former mixed agricultural land, or former grassland. In Lam et al. (2009) and Yee et al. (2009), this fact is attributed to subtracting the assimilation in palm plantation from life cycle GHG emissions. However, Yee et al. (2009) obtained a difference of 3 orders-of-magnitude (−8980 t CO2eq ha−1y−1) to that of Lam et al. (2009), which calls for a revisiting of the methodology they used.
  • 2.
    Studies that conclude that palm oil biodiesel is a net emitter of CO2 to the atmosphere.

All other results of the reviewed studies indicate palm oil biodiesel is a net GHG emitter to the atmosphere. In cases when palm oil plantations were originally from peatland forest, the GHG emissions range between 14 and 60 t CO2eq ha−1y−1, whereas those in which plantations were originally from natural or lowland forest can be in the range of 4 to 26 t CO2eq ha−1y−1 (Fargione et al. 2008; Germer and Sauerborn 2008; Wicke et al. 2008; Hassan et al. 2011). This indicates how LUC is the most decisive factor in determining life cycle GHG emissions of palm oil biodiesel.

In studies that exclude LUC in calculating the GHG emission (Achten et al. 2010; de Souza et al. 2010; Choo et al. 2011) it is found that GHG emissions are between 1 and 2 t CO2eq ha−1y−1. In such system boundaries, the significant factors that determine GHG emissions are the amount of fertilizer inputs in plantations, diesel fuel use in traction, and treatment option of POME.

Carbon payback time is the time that a biofuel system needs to repay the initial C emission caused by LUC. By knowing C-loss in LUC, life cycle GHG emissions of the assessed biofuel, and life cycle GHG emissions of a referred fossil fuel system, CPBT is calculable. Similar with GHG emissions, the CPBT value is also variable depending on the land origin and other emission factors in its life cycle, such as fertilizer use and treatment option in mill effluent. For biodiesel from plantation originally from peatland forest, the CPBT can last between 169 to 900 years, whereas those that were originally from degraded land can only last between 0 to 10 years. Table 5 gives the CPBT results of selected studies.

Table 5. CPBT of selected studies
NrAuthorrCPBT (y)Remarks
1Achten et al. 201047–49 
2de Souza et al. 201039 
3Danielsen et al. 200875–93From nonpeat forest
  10From degraded land
4Fargione et al. 2008420From peatland forest
  86From nonpeat forest
5Gibbs et al. 2008900From peatland forest
  30–120From nonpeat forest
7Wicke et al. 2008169From peatland forest
  30From nonpeat forest
  0From degraded land

From these studies, it can be concluded that the impacts of palm oil biodiesel on GHG emissions and CPBT can be lower when oil palm is planted on degraded land, grassland, or other low C soils (Choo et al. 2011; Hansen et al. 2012). Figure 3 provides a graphical summary to show how GHG emission would differ depending on the land origin. However, this improvement is subject to the availability of degraded land, grassland, or other low C soils that could be used for plantation purposes. Other improvement potential is available in the POME treatment process, for instance by using POME to produce biogas followed by co-composting the residue with EFB (Kamahara et al. 2010).

thumbnail image

Figure 3. Life-cycle GHG emissions depending on the land origin brought in to palm oil plantation.

Download figure to PowerPoint

Eutrophication and acidification impacts

Eutrophication and acidification potentials are addressed in 5 studies. Nordic Guidelines on LCA (Lindfors et al. 1995) is used for characterization in Achten et al. (2010), thus eutrophication and acidification impact potentials are expressed in g O2eq and g SO2eq. CML 2001 method (Guineé 2002) is applied in Stichnothe and Schuchardt (2010), and thus eutrophication and acidification impact potentials are expressed in kg POmath imageeq and kg SO2eq. Schmidt (2010) uses the updated EDIP97 and quantifies eutrophication and acidification impact potentials in kg NO3eq and kg SO2eq/FU. Both in Puah et al. (2010) and Yusoff and Hansen (2007), Ecoindicator 99 (Goedkoop and Spriensma 2001) is applied and results are normalized in European normalization and weighting factors. Table 6 summarizes all eutrophication and acidification potentials from the respective studies.

Table 6. Eutrophication and acidification potentials
1Achten et al. 201031.9 < EP < 39.6 g O2eq/FU1.35 < AP < 1.57 g SO2eq/FU
2Puah et al. 20105 pt 
3Schmidt 200880.6 < EP < 337 kg NO3eq/FU13 < AP < 23.5 kg SO2eq/FU
4Stichnote and Schuchardt 2010−0.003 << EP << 1.4 kg POmath imageeq/FU1.3E-02 < AP < 1.1E-02 kg SO2eq/FU
5Yusoff and Hansen 20072.7 pt 

Because eutrophication and acidification impacts are reported less uniformly, we did not attempt to normalize the results across these studies. Studies state that the majority of the total eutrophication and acidification potential of the palm oil biodiesel is caused by agricultural and extraction phases. This is associated with NH3 emissions and N leaching. Anther contributing factor is the combustion of biodiesel during its end use. Improvement is possible when POME is used to generate biogas and further co-composted with EFB, then returned to the plantation.

Biodiversity impact

Out of 23 reviewed studies, only 2 studies address the potential impact on biodiversity loss (Danielsen et al. 2008; Schmidt 2010). In Danielsen et al. (2008), diversity of fauna and flora of oil palm plantations is compared to that of tropical forests in Southeast Asia, Africa, and South America. Their results indicate that total vertebrate species richness of oil palm plantation is less than 38% that of natural forest, and the flora of oil palm plantation is seriously impoverished compare to that of forest. In Schmidt (2010), life cycle impact of biodiversity was quantified using characterization factor in the unit of wS100. This term refers to weighted species richness on a standardized area at 100 m2 and the weighting refers to the factor for ecosystem vulnerability as described in Schmidt (2008). Schmidt (2010) estimate the impact of land transformation to oil palm plantation is 385 wS100 per ha, assuming that the transformation is from secondary forest to intensive plantation.

Despite the fewer amount of study dedicated to biodiversity assessment of palm oil, it is generally understood that oil palm plantations with uniform crops, lower canopy, and great human interruption are structurally less complex than natural forest. Fitzherbert et al. (2008) reviewed studies dealing with impacts of palm oil expansion to biodiversity and conclude that oil palm plantation strongly affect biodiversity through fragmentation, edge effects, and pollution that can only be averted by avoiding future deforestation.

Other impacts

Impacts on abiotic resources depletion potentials are discussed in 5 studies: Achten et al. (2010), Puah et al. (2010), Stichnothe and Schuchardt (2010), Yusoff and Hansen (2007), and Zah et al. (2007). Results show that nonrenewable energy demand is relatively significant due to the use of fossil fuels in methanol production, fertilizer production, and traction. However, some authors claim there is a reduction of 45% to 60% on nonrenewable energy use compared to fossil reference system (Achten et al. 2010; Zah et al. 2007).

Potential impact to toxicity is discussed in 4 studies: Stichnothe and Schuchardt (2010), Puah et al. (2010), Yusoff and Hansen (2007), Zah et al. (2007) using respective characterization methods described previously. Overall, studies found that there is no significant human health impact apart from respiratory inorganics resulting mainly from particles emitted by the boiler. Some of the studies neglect the impact of pesticides, assuming that the doses are relatively small, so that the impacts generally do not spread out of the plantation area.

Land transformation impact is discussed in Achten et al. (2010) and Schmidt (2010). Land transformation impact strongly depends on the land origin. Almost similar to the GHG and biodiversity impact, an improvement of functional and structural ecosystem quality can only be expected in cases when the plantation is converted from degraded land or a mix of traditional agriculture.

Source of uncertainties

From the life cycle methodological point of view, difference in system boundary used (e.g., cradle-to-grave vs gate-to-gate) is the major source of variability and uncertainty in the studies. Furthermore, the use of different impact assessment methodologies (e.g., EcoIndicator vs EDIP97) is another important source of variability of results as demonstrated in Schmidt (2010) in its sensitivity analysis. Another aspect is the allocation method of co-product that often plays a role in variation of results. Data quality is also a common issue because proxy data is used in some of the studies. With respect to estimation of GHG emission in the land use change, the assumption of the land origin and its C content are the important sources of variability leading to the uncertainty of the results (Wicke et al. 2008; Hassan et al. 2011).

Apart from the methodological variability, uncertainty comes from the effects of different agricultural and processing practices in the respective studies. Variability in crop yields due to geographical location and management practice are the key issues in cultivation stage. For instance, in the study of de Souza et al. (2010), less fertilizer and fuel inputs are used in the plantation stage when compared to that of Pleanjai and Gheewala (2009). Mill efficiency is also a key issue in the conversion stage.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

Studies using life cycle based methodology to account for environmental impacts of oil palm biodiesel have been increasing over the last 5 years. However, the number of studies is growing only in certain countries (i.e., Malaysia, Thailand). More studies in Indonesia are needed because the LCA studies conducted are still disproportional to the scale of oil palm expansion in that country. South America and Africa are also in need of further exploration in such studies. Throughout these studies, some significant impacts such as energy gain, potentials of global warming, eutrophication, acidification, toxicology, and biodiversity have been disclosed. Nevertheless, more intensive studies for improving data quality and reducing the uncertainty and sensitivity analysis while exploring possible impacts, which have not yet been investigated, are still needed. Comparing biodiversity impacts of land use change from mixed agricultural land or grassland to oil palm plantation need to be explored as these 2 options are likely preferable to rainforest for further oil palm plantation expansion.

Further toxicological studies should be carried out to assess the possible impact of the pesticides use in plantations. Studies in Indonesia, in particular, should address potential marine ecological toxicology, considering the nature of its archipelagic landscape. It is also important to assess water use and its impact to the groundwater resource around the plantation and mills, as it is indicated that a considerable amount of water is used for irrigation and steam in boiler operations (Yáñez Angarita et al. 2009; de Souza et al. 2010).

From the relative energy gain standpoint, it is evident that oil palm is a potential renewable feedstock for renewable fuel. It is also evident that some environmental consequences are associated with its supply chain. Some process improvements have been identified both in increasing the energy gain and in reducing the environmental impacts. However, it must be highlighted here that there exist some trade-offs in achieving these objectives. As an example, expanding oil palm plantation in degraded land or grassland and avoiding the conversion of rainforest is the most effective improvement potential in reducing GHG emissions. However, planting oil palm on degraded land is costly and does not provide initial capital such as timber extraction from rainforest. As another example, the use of POME to generate biogas or to be further co-composted with EFB and returned to the plantation is also effective in GHG emissions reduction. However, this option will eliminate the opportunity of co-combusting EFB in boiler that in turn will increase the energy input (i.e., fossil fuels) and reduce the NER as well as profitability. To address these trade-offs, future studies need to focus on consequential LCAs and optimization-based models to obtain an optimal solution.

Recognizing the contribution of natural resources and the lack of the comprehensive accounting of the role of ecosystem goods and services, it is important to conduct an in-depth analysis of the contribution of natural resources in the life cycle of palm oil biodiesel, e.g., using the ecologically based LCA approach (Zang et al. 2010). Finally, decision support analysis studies (e.g., multicriteria decision analysis [MCDA]) are very relevant to assess different competing options of biodiesel feedstock, such as Jatropha, rapeseed, and algae. An MCDA in general involves m alternatives (e.g., comparing palm oil biodiesel with different biofuel systems or different scenario within palm oil biodiesel processing) evaluated on n criteria (i.e., sustainability criteria). An operational framework to execute the MCDA in sustainable energy decision making has been presented by Wang et al. (2009).

Eventually, in view of the three dimensions of sustainability, efforts must be made to broaden existing LCA studies to account for more criteria that is relevant to the 3Ps (prosperity, people, and planet). To this end, it will be interesting to apply the integrated system modeling framework for the Life Cycle Sustainability Assessment framework developed in Halog and Manik (2011). This is also in line with the call from United Nations Environmental Program (UNEP) for life cycle sustainability assessment, which has been reported in Heijungs et al. (2010) and Guineé et al. (2011).


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

This study was sponsored by the Fulbright Program in cooperation with American Indonesian Exchange Foundation (AMINEF).


  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information
  • Achten WMJ, Van den Bempt P, Almeida J, Mathis E, Muys B. 2010. Life cycle assessment of a palm oil system with simultaneous production of biodiesel and cooking oil in Cameroon. Environ Sci Technol 44: 48094815.
  • Basiron Y. 2007. Palm oil production through sustainable plantation. Eur J Lipid Sci Technol 109: 289295.
  • Choo YM, Muhamad H, Hashim Z, Subramaniam V, Puah CW, Tan YA. 2011. Determination of GHG contributions by subsystems in the oil palm supply chain using the LCA approach. Int J LCA 16: 669681.
  • Danielsen F, Beukema H, Burgess ND, Parish F, Brühl CA, Donald PF, Mudiyarso D, Phalan B, Reijnders L, Struebig M., et al. 2008. Biofuel plantations on forested lands: Double jeopardy for biodiversity and climate. Conserv Biol 23: 348358.
  • de Souza SP, Pacca S, de Avila MT, Borges JLB. 2010. Greenhouse gas emissions and energy balance of palm oil biodiesel. Renew Energ 35: 25522561.
  • Fargione J, Hill J, Tilman D, Polasky S, Hawthorne P. 2008. Land clearing and the biofuel carbon debt. Science 319: 12351238.
  • Fitzherbert EB, Struebig MJ, Morel A, Danielsen F, Brühl CA, Donald PF, Phalan B. 2008. How will oil palm expansion affect biodiversity? Trends Ecol Evol 23: 538545.
  • Fukuda H, Kondo A, Noda H. 2001. Biodiesel fuel production by transesterification of oils. J Biosci Bioeng 92: 405416.
  • Germer J, Sauerborn J. 2008. Estimation of the impact of oil palm plantation establishment on greenhouse gas balance. Environ Dev Sustain 10: 697716.
  • Gibbs HK, Johnston M, Foley JA, Holloway T, Monfreda C, Ramankutty N, Zaks D. 2008. Carbon payback times for crop-based biofuel expansion in the tropics: The effects of changing yield and technology. Environ Res Lett 3: 3400134011.
  • Goedkoop M, Spriensma R. 2001. The EcoIndicator 99—A damage oriented method for Life Cycle Impact Assessment. BB Amersfoort: PRe Consultants.
  • Guineé JB. editor. 2002. Handbook on life cycle assessment: Operational guide to the ISO standards. Dordrecht: Kluwer Academic. 8182.
  • Guineé JB, Heijungs R, Huppes G, Zamagni A, Masoni P, Buonamici R, Ekvall T, Rydberg T. 2011. Life cycle assessment: Past, present and future. Environ Sci Technol 45: 9096.
  • Halog A, Manik Y. 2011. Advancing integrated system modeling framework for life cycle sustainability assessment. Sustainability 3: 469499.
  • Hassan MNA, Jaramillo P, Griffin WM. 2011. Life cycle GHG emissions from Malaysian oil palm bioenergy development: The impact on transportation sector's energy security. Energ Policy 39: 26152625.
  • Hansen SB, Olsen SI, Ujang Z. 2012. Greenhouse gas reductions through enhanced use of residues in the life cycle of Malaysian palm oil derived biodiesel. Bioresour Technol 104: 358366.
  • Heijungs R, Huppes G, Guineé JB. 2010. Life cycle assessment and sustainability analysis of products, materials and technologies: Toward scientific framework for sustainability life cycle analysis. Polym Degrad Stabil 95: 422428.
  • [ISO] International Organization for Standardization. 2006. ISO 14040: Environmental management—Life cycle assessment, principles and framework. Geneva: ISO.
  • Kamahara H, Hasanudin U, Widiyanto A, Tachibana R, Atsuta Y, Goto N, Daimon H, Fujie K. 2010. Improvement potential for net energy balance of biodiesel derived from palm oil: A case study from Indonesian practice. Biomass Bioenerg 30: 17.
  • Lam MK, Lee KT, Mohamed AR. 2009. Life cycle assessment for the production of biodiesel: A case study in Malaysia for palm oil versus Jatropha oil. Biofuels Bioprod Bioref 3: 601612.
  • Larson ED. 2006. A review of life-cycle analysis studies on liquid biofuel systems for the transportation sector. Energ Sustain Dev 10: 109126.
  • Lindfors LG, Christiansen K, Hoffman L, Virtanen Y, Juntilla V, Hansen OJ, Ronning A, Ekvall T, Finnveden G. 1995. Nordic guidelines on LCA. Copenhagen: Nord.
  • Papong S, Chom-In T, Noksa-nga S, Malakul P. 2010. Life cycle energy efficiency and potentials of biodiesel production from palm oil in Thailand. Energ Policy 38: 225233.
  • Pleanjai S, Gheewala SH. 2009. Full chain energy analysis of biodiesel production from palm oil in Thailand. Appl Energ 86: S209S214.
  • Pleanjai S, Gheewala SH, Garivait S. 2009. Greenhouse gas emissions from the production and use of palm methyl ester in Thailand. Int J Global Warm 1: 418430.
  • Puah CW, May CY, Ngan MA. 2010. Life cycle assessment for the production and use of palm biodiesel (Part 5). J Palm Oil Res 25: 927933.
  • Reijnders L, Huijbregts MAJ. 2008. Palm oil and the emission of carbon-based greenhouse gases. J Clean Prod 16: 477482.
  • Schmidt JH. 2008. Development of LCIA characterization factors for land use impact on biodiversity. J Clean Prod 16: 19291942.
  • Schmidt JH. 2010. Comparative life cycle assessment of rapeseed oil and palm oil. Int J LCA 15: 183197.
  • Siangjaeo S, Gheewala SH, Unnanon K, Chidthaisong A. 2011. Implications of land use change on the life cycle greenhouse gas emissions from palm biodiesel production in Thailand. Energ Sustain Dev 15: 17.
  • Stichnothe H, Schuchardt F. 2010. Comparison of different treatment options for palm oil production waste on a life cycle basis. Int J LCA 15: 907915.
  • Sumathi S, Chai SP, Mohamed AR. 2008. Utilization of oil palm as a source of renewable energy in Malaysia. Ren Sust Energy Rev 12: 24042421.
  • Wang J-J, Jing Y-Y, Zhang C-F, Zhao J-H. 2009. Review on multi-criteria decision analysis aid in sustainable energy decision-making. Renew Sustain Energ Rev 13: 22632278.
  • Wicke B, Dornburg V, Junginger M, Faaij A. 2008. Different palm oil production systems for energy purposes and their greenhouse gas implications. Biomass Bioenerg 28: 13221337.
  • Yáñez Angarita, Silva EE, Lora EE, da Costa RE, Torres EA. 2009. The energy balance in the palm oil-derived methyl ester (PME) life cycle for the cases in Brazil and Colombia. Renew Energ 34: 29052913.
  • Yee KF, Tan KT, Abdullah AZ, Lee KT. 2009. Life cycle assessment of palm biodiesel: Revealing facts and benefits for sustainability. Appl Energ 86: S189S196.
  • Yusoff S, Hansen SB. 2007. Feasibility study of performing an LCA on crude palm oil production in Malaysia. Int J LCA 12: 5056.
  • Zah R, Boni H, Gauch M, Hischier R, Lehmann M, Wager P. 2007. Life Cycle Assessment of energy products: Environmental assesment of biofuels. Bern: EMPA.
  • Zang Y, Baral A, Bakhsi BR. 2010. Accounting for ecosystem services in life cycle assessment, Part II: Toward and ecologically-based LCA. Environ Sci Technol 44: 26242631.

Supporting Information

  1. Top of page
  2. Abstract
  7. Acknowledgements
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

ieam_1362_sm_SupplTabA1.doc60KSupplementary Table A1: Summary of the reviewed study.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.