The Future of Oil Palm As A Major Global Crop: Opportunities and Challenges

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/286678734
The future of oil palm as a major global crop: Opportunities and challenges
Article in Journal of oil palm research · March 2014
CITATIONS READS
56 2,281
1 author:
Denis J Murphy
University of South Wales
285 PUBLICATIONS 7,138 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Toxicological Effects and Biological Responses of Plants Against the Persistent Organic Pollutants (POPs) View project
Molecular and biochemical regulation of Aflatoxin biosynthesis View project
All content following this page was uploaded by Denis J Murphy on 18 April 2016.
The user has requested enhancement of the downloaded file.

REVIEW ARTICLE
Journal of Oil Palm Research Vol. 26 (1) March 2014 p. 1- 24
THE FUTURE OF OIL PALM AS A MAJOR

GLOBAL CROP: OPPORTUNITIES AND
CHALLENGES
DENIS J MURPHY*
ABSTRACT
In recent years, the oil palm sector has witnessed a period of historically high prices with buoyant global
demand and high levels of production driven largely by economic development in major Asian countries
such as India and China. However, the oil palm sector is also confronted by many important challenges
that require attention. Such challenges include fragmentation of the industry, stagnating yields, and an
image problem that is largely due to the conversion of tropical rainforest and peatlands in a few regions in
South-east Asia. The biological and managerial tools to surmount these challenges already exist but need
more focussed application and political support. Potentially groundbreaking biological tools include the
new molecular breeding technologies, such as those made possible by the recent publication of the oil palm
genome sequence (Singh et al., 2013a, b). Two key R&D targets for the industry are:
• higher oil yield in fruits and trees; and
• higher mesocarp oleic acid composition – preferably over 65% w/w.
The more focussed use of new and traditional technologies can also help to confront pest and disease
problems, to redesign of crop architecture, and to facilitate yield and harvesting efficiency. In the medium-
term future, we can look forward to a considerable geographical extension of oil palm cultivation in a broad
zone across the tropics of Africa, Asia and the Americas. If these and other measures can be taken, increased
palm oil output could more than meet the highest projections for future vegetable oil requirements while
minimising adverse environmental consequences. Improved oil palm varieties could also considerably
increase the global market share for this highly productive tropical crop at the expense of some of the less
efficient temperate oilseed crops.
Keywords: oil yield, high oleic, genomics, breeding, management.
Date received: 19 August 2013; Sent for revision: 11 September 2013; Received in final form: 16 January 2014; Accepted: 16 January 2014.
INTRODUCTION t ha-1 for the best commercial plantations and 3-4

t ha-1 for smallholders. Palm oil is mainly used as
Oil palm is a uniquely productive tropical crop with an edible product and is an increasingly important
a potential yield capacity well in excess of 10 t of dietary component for well over one billion people
oil per hectare (t ha-1). However, current yields are worldwide. The oil palm also has many uses in the
well below this figure and are typically about 4-6 non-food sector, with examples ranging from high
value oleochemicals to more basic biomass-derived
* University of South Wales, CF37 4AT, United Kingdom. materials such as paper and plywood (Suleiman et
E-mail: denis.murphy@southwales.ac.uk al., 2012). In commercial terms, the major product
1
Journal of Oil Palm Research 26 (1) (MARCH 2014)
of the crop is the oil that accumulates in the fleshy although it is not a particularly good crop plant,
mesocarp of the fruits, which is often referred to E. oleifera contains useful genetic variation that can be
as palm oil, crude palm oil or mesocarp oil. About incorporated into breeding programmes for creation
84% w/w of the mesocarp oil is made up of palmitic of improved commercial varieties of E. guineensis.
(C16:0) and oleic (C18:1) acids and this oil has been Oil palm was first grown on a widespread scale
used as a traditional food source in parts of Africa when African trees were introduced into Sumatra
for many thousands of years. and Peninsular Malaysia as plantation crops in the
th
Palm fruits also contain a single hard seed, or early 20 century. Commercial cultivation increased
kernel, which is enriched in a different type of oil, dramatically, first in Malaysia after independence
often referred to as palm kernel oil, that makes up in 1957 and subsequently in Indonesia early in 21st
about 11% of the total oil fraction of the crop. Palm century. By 2012, Malaysia and Indonesia collectively
kernel oil is enriched in the medium chain fatty acids, produced about 57 million tonnes of oil, which made
lauric (C12:0) and myristic (C14:0) acids, which up over 85% of global palm production (Table 1) and
together make up 74% w/w of this oil. These fatty almost 40% of the entire world output of all forms of
acids make palm kernel oil a useful source of many vegetable oil (USDA, 2012).
industrial and cosmetic products such as soaps, For most of the period of its cultivation as
detergents and cleaning agents. Lauric acid is also a plantation crop, oil palm has been relatively
used in several edible applications including as a uncontroversial. However, over the past decade
shortening agent for margarine and cream products various aspects of the introduction, management,
and can even be applied as an anti-microbial agent. and end use of oil palm have come under increasing
Both types of palm fruit oil have been used scrutiny, and in some cases criticism. Many, but not
historically as fuels, most commonly for lighting in all, of the most negative comments about oil palm
traditional settings. Since 2000, increasing amounts have come from certain groups in Europe and to a
of palm oil and other vegetable oils have been lesser extent in North America. Often such comments
converted into their methyl ester derivatives in have focussed on the environmental consequences
order to produce biodiesel fuel for vehicles (Rosillo- (such as forest clearance) of increased oil palm
Calle et al., 2007). This has led to concerns about cultivation and also its increasing use as a biofuel
the diversion of some palm oil from edible uses crop. Other concerns about the high level of saturated
with possible increases in food prices and increased fatty acids in palm oil have been voiced but it is far
pressures to convert undeveloped forest and peat from clear whether palm oil saturates like palmitate
habitats to oil palm plantations (Bringezu et al., have the same negative health consequences as
2009; Johnston et al., 2009). However, to put this in more typically animal-derived saturates such as
context, in 2013 the biodiesel sector accounted for stearate. On the other hand, there is an ever growing
well under 10% of total palm oil production and demand for palm oil products, especially for food
the future growth of this sector is far from assured. use in developing countries. This demand needs to
For example, concerns about the environmental be satisfied somehow, which leads to the question:
credentials of some biofuels have recently led how can oil palm production be increased without
the European Union (EU) to reduce targets for adverse environmental consequences?
renewable fuels for transport. This reduction applies The conversion of land to agriculture has been
specifically to biofuels (such as palm oil biodiesel) a feature of human development for thousands
from food crops where the 10% target of replacing of years. This process has already resulted in the
fossil fuels has been halved to 5% (Van Noorden, mass clearance of forests for crop cultivation in
2013). The result of this policy shift will be a much of Europe and the Americas. Since the late
th
decreased demand for biodiesel in the EU, which 20 century, the continued conversion of land in
is currently the largest global user of this product. some parts of the world has come under increased
Therefore, while biodiesel has been responsible for scrutiny, particularly from environmental groups.
some of the increase in demand for palm oil, this is Such attention has particularly been focussed on
unlikely to be the case in the medium-term future. the conversion of land to oil palm in parts of South-
Instead, by far the major factor driving increased east Asia and to soyabean crops in South America.
future demand for palm oil will continue to be the There is a broad consensus that it is undesirable in
burgeoning global requirement for more edible oils, principle to mass-convert some ecologically sensitive
especially in Asia. habitats, such as pristine rainforests or some deep
The two major species of oil palm, Elaeis peat regions, to oil palm plantations. Also, if new
guineensis and Elaeis oleifera, have their centres of plantations are already being established, it is
origin respectively in West Africa and South America desirable to do this in a way that minimises the eco-
(Corley and Tinker, 2003). Due to its higher yield, the environmental side effects of the land conversion.
African oil palm, E. guineensis, is the species that is For example, in some areas the establishment of
overwhelmingly used for commercial cultivation, protected areas linked by wildlife corridors can
even in South America. However, as discussed below, enable plantations to coexist with traditional native
2
THE FUTURE OF OIL PALM AS A MAJOR GLOBAL CROP: OPPORTUNITIES AND CHALLENGES
TABLE 1. MAJOR CENTRES OF OIL PALM CULTIVATION
Rank Country Production (million tonnes)

1 Indonesia 31.0
2 Malaysia 19.0
3 Thailand 2.1
4 Colombia 1.0
5 Nigeria 0.9
6 Papua New Guinea 0.6
7 Ecuador 0.6
8 Honduras 0.4
9 Ivory Coast 0.4
10 Brazil 0.3
Others 0.7
Total 57.0
Source: Data from the United States Department of Agriculture (2012).
fauna. In other cases, replanting existing plantations the current global output of 65 million tonnes palm
with higher yielding palms can reduce the need to oil requires cultivation of only 15 million hectares,
establish new plantations. which contrasts dramatically with the 194 million
In the remainder of this article, I will assess the hectares needed to produce just 87 million tonnes oil
current status of oil palm as a major global crop from the temperate annual oilseed crops (Oil World,
and describe how both biological and management 2012). Therefore, in terms of total oil yield (kernel
approaches can contribute to addressing many of the + mesocarp oil) per hectare, oil palm is already
current concerns about this valuable and frequently more than 6.5-fold more efficient than the average
misunderstood plant. combined yields of the temperate oilseed crops.
Given the realistic prospects of further increases in
palm oil yield in the next decade, the future for oil
OIL PALM IS BY FAR THE MOST EFFICIENT palm as a global vegetable oil crop seems even more
CROP-BASED HYDROCARBON PRODUCTION promising during the coming years.
SYSTEM In addition to its high oil yield, oil palm is also
a much more efficient crop that its competitors in
Oil crops such as oil palm and the temperate oilseeds terms of the required intensity of land management,
(e.g. soyabean, rapeseed, sunflower, peanut, and harvesting and processing. For example, the annual
cotton) are renewable sources of oils that can used oilseed crops require replanting each year which
either as edible foodstuffs or as industrial feedstocks involves regular disruption of the soil structure and
to replace products that are otherwise derived non- rhizosphere by ploughing. These crops also require
renewable mineral oils. Industrial uses include a brief but intensive annual period of harvesting and
manufacture of a wide range of basic oleochemicals, processing that often must be completed in a matter
chemical intermediates, and more highly processed of days, whatever the weather. In contrast, an oil palm
finished products such as coatings, lubricants, and can be cultivated for 20-30 years without disturbing
biopolymers, plus biofuels such as biodiesel. As we the soil. Also, within a given plantation, harvesting
will now see, oil palm is by far the highest yielding and processing can take place on a continual year-
biological source of oil-based hydrocarbons and round basis within a relatively predictable climatic
is significantly more efficient than any of other regime that has far less seasonal fluctuation than in
commercial oil crop. temperate regions.
In 2012, the estimated global production of total This means that the workforce, machinery and
palm oil was almost 65 million tonnes, of which 58 other assets can be employed on a continuous basis
million tonnes was mesocarp oil and 6.8 million throughout the year on oil palm plantations, rather
tonnes was kernel oil (USDA, 2012). Typical average than for a single intensive period as is the case for
yields of palm oil on a global basis are in the region annual oil crops. To make an analogy with microbial
of 4 t ha-1. This figure far outstrips the yield of the biotechnology, oil palm husbandry resembles an
major temperate annual oilseed crops where yields efficient continuous culture system rather than
range from 0.3 to 1.2 t ha-1. This high yield means that the much less efficient batch-processing system
3
represented by annual crop husbandry. As well and Malaysia has risen steadily in response to global
as having a competitive edge over the temperate demand in recent years and this process shows no
oilseed crops, oil palm is also more productive than sign of stopping in the short- to medium-term future.
other oil-bearing tree crops such as olive or coconut, Before 2005, Malaysia was the major global palm
-1
which respectively yield oil at about 2.0 t ha and 0.3 oil producer but since then it has been overtaken
-1
t ha . by Indonesia. In the six years from 2005 to 2011,
Clearly, therefore, oil palm is far more efficient Indonesia increased oil palm production by 68%,
than all other oil crops in terms of oil yield, land largely due to land conversion to new plantations.
usage and asset deployment. This efficiency is In contrast, Malaysia had a more modest production
reflected in the pricing of palm oil, which is almost increase of 17% that reflected much lower rates of
invariably considerably cheaper than its rivals. For land conversion.
example, in November 2013, the wholesale price In order to position the oil palm industry to meet
for palm oil was about USD 826 per tonne while the the ever growing requirement for its major products,
average for soyabean, sunflower and rapeseed oils it is important to understand why these oils are such
was more than USD 1020 per tonne, representing a desirable commodities. This information will also
enable the sector to ensure that its R&D programmes
discount for palm oil of almost 20% over its major
are designed to optimise the delivery of the
competitors (MPOB website, www.mpob.gov.my).
highest possible quality of oil with the minimum
It should be noted, however, that one of the reasons
environmental footprint. In achieving this goal, it
for the higher prices of rapeseed and soyabean oils
will be essential to satisfy consumer demands for
(compared to palm oil) is their higher oleate content. oil functionality and also to adequately address
As discussed below, the development of high oleate sustainability criteria for the overall crop production
palm oil would greatly increase the value of the oil and processing systems used to generate palm oil.
and should therefore be a major R&D priority for the
industry.
Given its higher productivity and lower price, it FURTHER INCREASES IN DEMAND FOR
is not surprising that palm oil overtook soyabean oil PALM OIL ARE INEVITABLE
as the major global vegetable oil in 2007 and that it
is increasingly sought after as the edible oil of choice At present, the major drivers for continued increases
by developing countries throughout the world. As in demand for palm oil include population growth
shown in Table 2, palm oil production in Indonesia and economic development in those countries
TABLE 2. INCREASE IN PALM OIL PRODUCTION IN INDONESIA AND MALAYSIA
Oil production
Year (million tonnes)
Indonesia Malaysia
2000 8.3 11.9
2001 9.2 11.9
2002 10.3 13.2
2003 12.0 13.4
2004 13.6 15.2
2005* 15.6 15.5
2006 16.6 15.3
2007 18.0 17.6
2008 20.5 17.3
2009 22.0 17.8
2010 23.6 18.2
2011 26.2 18.2
2012 28.5 18.8
+
2013 31.0 19.4+
Note: * The year Indonesia became the largest producer. + Estimated value.
Source: Data from FAO (2013).
4
importing oil for food and, to a much lesser extent, calorific value of the food. The use of fats in cooking
the demand for biofuels in other countries such as also greatly enhances the taste and odour of foods
the EU. The major importers of palm oil for food because heated fats and oils produce a complex
are in the Indian subcontinent (India, Pakistan bouquet of attractive flavour compounds, many
and Bangladesh), China, and West Asia (Table 3). of them volatile. Fats also solubilise, and thereby
Collectively these regions import over 22 million enhance, otherwise cryptic flavours that may be
tonnes of palm oil, which is mainly used either present in non-fatty foodstuffs (Murphy, 2007). This
directly in cooking or as an ingredient in a host of explains the perennial human craving for lipidic
processed foods. These three regional blocs account foodstuffs such as vegetable oils.
for about 75% of current imports of palm oil, the vast In line with this phenomenon, it was found
majority of which (85% in 2013) was obtained from that, following rising income levels across much
the two major producing countries, namely Malaysia of the developing world in the 1990s, vegetable oil
and Indonesia. So, why is palm oil experiencing such consumption increased much faster than general
sustained increases in demand from these particular food intake. For example, during the 1990s, per
importing countries? The answer is mostly related capita vegetable oil consumption rose by 31% in
to economic and demographic factors. Mexico, 35% in South Africa, 64% in China, 65%
In terms of demography, the Indian subcontinent, in Indonesia, and 94% in India (Murphy, 2007).
China, and West Asia have all experienced rapid As people became more affluent, they switched
increases in population over the past few decades. to a more satisfying diet containing much higher
This will obviously result in higher demands for amounts of oil. Interestingly, the reverse is also true
food products in general. However, demographics and, when times are particularly hard, people tend
are only a small part of the reasons behind the current to cut back on ‘luxuries’ like fats and oils. Such an
demand for palm oil. For example, population effect was seen during the economic collapse that
increases have now started levelling off, especially followed the fall of the Soviet Union when, between
in China, but demand for palm oil has continued to 1990 and 1994, consumption of food oil in Russia
rise. This brings us to the key role of economics as a fell by 35%. As predicted by the correlation between
driver for increasing demand for vegetable oils and oil and income, food oil consumption in Russia
is related to a well-known correlation between per rose once again during the 2000s as the economy
capita income and the consumption of fats and oils recovered and average incomes increased.
in the human diet. According to a report from the United Nations
Food and Agriculture Organisation (FAQ):
Higher Incomes Drive Increased Demand for ‘… vegetable oils and oil products still have relatively
Edible Oils high income elasticities in the developing countries ...
In our earlier projections to 2010, the per capita food
The reason for the correlation between consumption of all vegetable oils, oilseeds and products
household income and fat consumption is that (expressed in oil equivalent) was projected to rise from
dietary fats are particularly desirable to people who 8.2 kg in 1988/90 to 11 kg in 2010. By 1997/99, it had
have historically subsisted mainly on starch- and grown to 9.9 kg. In the current projections, the per capita
vegetable-based diets made up of relatively dull and food demand for the developing countries as a whole rises
tasteless foodstuffs, such as boiled rice, manioc or further to 12.6 kg in 2015 and to nearly 15 kg by 2030.
potatoes. The addition of fats or oils to such a diet We have already noted earlier that average per capita food
considerably enhances nutrient content (i.e. the consumption (all food products) in developing countries
lipophilic vitamins A, D and E), and increases the may rise from the 2680 kcal of 1997/99 to 2850 kcal
in 2015 and to 2980 kcal by 2030. Vegetable oils and
TABLE 3. MAJOR PALM OIL IMPORTERS products would contribute some 45% of this increase.
This is an acceleration of the historical trend for these
Country/region Imports in 2012 (million tonnes) commodities to account for an ever-increasing part of the
Indian subcontinent* 12.6 growth in food consumption in the developing countries.
They had contributed 18% of the total increment in the
China 6.6
decade from the mid-1970s to the mid 1980s and 27% in
European Union 5.8 the subsequent decade.’ (FAO, 2003).
+
West Asia 3.2
USA 1.0 Meeting Increasing Demands for Food Oil Should
be the Top R&D Priority
Note: * India, Pakistan and Bangladesh.
+
Egypt, Iran, United Arab Emirates, Turkey, Saudi Arabia The bottom line from the FAO projections
and Iraq.
quoted above is that edible vegetable oils already
Source: Data from www.indexmundi.com account for almost half of the increased food
5
demand in developing countries and that this trend Revolution’ of the 1960s and 1970s that effectively
will be maintained in the next few decades. On a doubled or trebled food production in much of the
global basis, the annual per capita consumption of world and averted mass famine in Asia. During the
vegetable oils increased by an impressive 40% from 1980s and 1990s, more complex methods such as
11.3 kg to 15.9 kg between 1997 and 2010 (Gunstone, hybrid creation, assisted crosses and introgression of
2011). Despite the economic slump experienced wild germplasm were instrumental in enabling rice
across much of Europe and North America between yield to increase five-fold in some regions. The so-
-1
2007 and 2013, many parts of Asia have continued to called ‘miracle rice’ that yields as much as 10 t ha
experience modest to good growth rates and future has enabled China to become largely self sufficient in
prospects appear promising. grain crops and also laid the foundations of its recent
We can therefore predict with some confidence economic advancement. During the 2000s, much
that, providing Asian economies continue attention has been focussed on genomic approaches
their current steady growth, there should be a to plant breeding with the deployment of a new
corresponding increase in demand for vegetable oils, generation of technologies, such as marker-assisted
and particularly for oil palm, from this region and selection, next-generation sequencing, transgenesis
perhaps in other developing regions. In most parts of (genetic engineering or GM) and automatic
the world, palm oil is consumed as a refined product mutagenesis/selection (TILLING, TargetIng Local
with uses varying from vanaspati/ghee in India to Lesions IN Genomes) (Murphy, 2011; Soh, 2011; Xu,
margarine, cooking oils, and biscuits in Europe and 2010). These methods have great potential for oil
the United States. In contrast, in its centre of origin palm improvement, as we will now discuss.
in West Africa, virgin red palm oil (highly enriched
in carotenoids and tocopherols) is widely used in Sequencing the Oil Palm Genomes
soups and baked dishes. As the increasing demand
is overwhelmingly for edible oils, the palm oil sector Probably the most dramatic example of
should focus on maximising production of oils that technology improvement in the 21st century has
are optimised for global food markets. been in DNA sequencing where the cost per base has
In order to satisfy these major edible markets, decreased by an amazing 100 000-fold since 2000, as
which include well over one billion people per day shown in Figure 1 (Mardis, 2008; Shendure and Ji,
consuming palm oil products, there are two obvious 2008). The first plant genome to be fully sequenced
targets for breeders. The first priority should be to was the model species, Arabidopsis thaliana,
increase the oil yield both per fruit bunch and per published in 2001, while the first crop genome was
tree. Increased palm oil yield is urgently required rice, where a high quality sequence was published
to address the ever-rising global demand for edible in 2005. The sequencing of the much larger maize
oils. The second priority should be to maximise genome required a massive effort by company and
the oleic acid content and reduce the amount of public laboratories and the results were published
saturated palmitic and stearic acids. Although these in a series of papers in 2009. Other large-scale
saturated fatty acids have some advantages in terms projects are currently underway for developing
of producing solid fats such as margarines, they are country crops such as sorghum and foxtail millet
generally regarded as nutritionally less desirable and sequence data are now being publicly released
than monounsaturated oleic acid. This is one of the at an increasingly rapid pace. Advances in next
main reasons why existing high oleic acid commodity generation sequencing technologies are enabling the
vegetable oils, such as rapeseed, command a price genomes of even comparatively minor crops to be
premium over palm oil as edible feedstocks. High characterised (Edwards and Batley, 2010).
oleic oils have the additional advantage that they In some cases, a single sequencing method has
can also be used as highly versatile raw materials been used but, more commonly, several technologies
for production of a wide range of renewable and are used in combination for best results. For example,
biodegradable oleochemicals and other industrial Roche 454 technology was used to sequence the 430
products (Murphy, 2010). Mb genome of cocoa, Theobroma cacao, and the 1700
Mb genome of oil palm. In contrast, a combination of
Sanger and Roche 454 sequencing was used for the
BIOLOGICAL TECHNOLOGIES FOR OIL PALM apple and grape (500 Mb) genomes. A combination
IMPROVEMENT of Illumina Solexa and Roche 454 sequencing was
used for the genomes of polyploid cotton. Roche
The last few decades have witnessed huge 454 sequencing has been used for Miscanthus, while
advances in our understanding of plant biology Sanger, Illumina Solexa, and Roche 454 sequencing
and in the development of new technologies for are being used for banana. Illumina GAII sequencing
the manipulation of plants for human benefit. The has been used for the Brassica rapa genome, while
application of relatively straightforward breeding Sanger and Illumina Solexa technologies were used
and selection methods were behind the ‘Green for the cucumber genome. The cheapness and speed
6
1000
2.0x107
103
Cost per Mb of DNA sequence (USD)
Curated vs. uncurated proteins

800
Pubmed ‘Metagenomics’ hits

1.5x107
102
Sequencing cost 600
Uncurated proteins
Curated proteins
1.0x107
Publications
101
400
100 0.5x107
200
10-1 0 0
2000 2002 2004 2006 2008 2010 2012
Year
Source: Data from Pubmed and NCBI.
Figure 1. Decreasing DNA sequencing costs over the past decade has been mirrored by a huge increase in the proportion of uncurated proteins in public
databases. Since the early 2000s, the cost of sequencing genomes has plummeted by more than four orders of magnitude. This has created a glut of raw
data, much of which has yet to be curated in terms of definite identification of functional genes and proteins. By 2012, for each curated protein sequence
in public databases there were several million uncurated sequences.
of genome sequencing is now making it possible most effective in terms of rapid exploitation to
to sequence, not just a single reference genome for release sequence data as a public resource that can
each species, but many individual genomes in a then be mined by large numbers of researchers
population. This approach will be used to uncover across the world as a specialised form of ‘crowd
genome-wide variations that underlie some of the sourcing’.
more complex developmental and agronomic traits In the case of oil palm genome, in May 2008,
in crops such as oil palm (Cook and Varshney, 2010; the Asiatic Centre for Genome Technology (ACGT
Murphy, 2011; Paterson et al., 2010). – a subsidiary of Genting Berhad), in collaboration
Although sequencing costs have fallen with Synthetic Genomics (whose CEO, Craig Ventor,
dramatically in recent years, the initial sequencing led part of the human genome sequencing effort),
and basic annotation of a relatively large genome such announced completion of the first draft sequence.
as that of oil palm still amounted to several million However, the data were not released and no details
US dollars during the period 2007-2010 when much of the work were published in the scientific literature.
of the basic data were compiled. In the case of most A year later in May 2009, Sime Darby Plantation Sdn
large crop or animal genomes, sequencing has been Bhd announced sequencing and annotation of about
carried out by public/private consortia that include 94% of a different oil palm genome. Once again very
technology providers and potential end-users. few technical details were released and none of the
Unfortunately for researchers in general, in a few data were made available in the public domain. The
cases the same crop genomes have been sequenced Sime Darby-led project also involved an international
multiple times by different commercial groups consortium that included a bioinformatics company,
who have not released their data. The sequencing Synamatix Sdn Bhd and a US technology provider,
of the oil palm genome is an example of this latter 454 Life Sciences (a Roche company). The third oil
phenomenon. While such a duplication of effort and palm sequence announcement came in November
an unwillingness to share the data publicly may be 2009 when a consortium led by the publicly funded
based on sound commercial factors in the short- Malaysian Palm Oil Board (MPOB) and the US
term, it nevertheless arguably represents a failure private company, Orion Genomics, reported that
in strategic vision for the industry as a whole. This three genomes had been sequenced from the two oil
is because DNA sequence data on their own have palm species, E. guineensis and E. oleifera (Meerow et
little value and even relatively basic tools for their al., 2012).
assembly into a recognisable annotated genome are Happily for oil palm researchers, the MPOB/
still under development. Therefore, it is normally Orion-led consortium has now published an account
7
of their genomic sequences in the journal Nature Dussert et al., 2013; Tranbarger et al., 2011). In two
in July 2013 (Singh et al., 2013a). These sequences other studies, transcriptome data from normal and
are available for breeders and other scientists to mantled oil palm fruits have been compiled to
study and use for the benefit of crop R&D (data shed light on this abnormality which still plagues
are deposited at DDBJ/EMBL/GenBank under micropropagation of the crop (Shearman et al., 2013)
the accessions ASJS00000000 for E. guineensis and while a similar approach has been used to study
ASIR00000000 for E. oleifera). This landmark article somatic embryogenesis (Lin et al., 2009).
was accompanied by another equally significant The metabolome is the complete list of
paper that described the identification of a single metabolites found in a particular organelle, cell,
gene, called Shell, which was found to regulate or tissue under a specific set of conditions. The
the tenera trait of fruit shell thickness (Singh et al., identification of important plant metabolites in a
2013b). The tenera trait is found in hybrids between plant such as oil palm, which include carotenoids,
naturally occurring dura (thick shelled) and pisifera phenols, and fatty acyl components, used to be
(non-shelled) fruit forms of oil palm. The dura fruits a very slow process relying on bulky, expensive
have low oil yields and pisifera fruits are normally equipment that could only be operated by a few
female-sterile but the tenera hybrids are fertile, high skilled specialists. However, new lightweight
oil yielding plants that are now the basis for all devices, supplemented by robotic and informatics
commercial oil palm production in South-east Asia. approaches, now make it possible to automate the
Identification of the Shell gene will enable breeders process and even to assign accurate identities to
to use molecular markers to select suitable breeding complex mixtures of such molecules.
lines, instead of waiting three to four years or more Metabolome analysis can help uncover the
for the young plants to produce fruits for selection details of oil palm fruit or leaf development at a
via a visual phenotype. molecular level (Neoh et al., 2013; Teh et al., 2013;
Tahir et al., 2013). Metabolome studies can also
Beyond the Genome: Other Ómic Technologies indicate how plants are reacting at the molecular
level to specific stimuli, e.g. by comparing stressed
In order to move beyond gene composition and unstressed plants we can gain important
through to gene expression, protein function information about how some plants can tolerate
and their ultimate manifestations as phenotypes certain stresses while others cannot. In other cases,
in an organism, it is often necessary to analyse metabolome analysis can give useful information
structural and functional molecules, such as about molecular changes caused by the addition of
proteins, membrane lipids, and carbohydrates transgenes to plants. This kind of analysis is often
in particular plant cells or tissues. At a more used as part of the process of regulation of transgenic
detailed level, there are many thousands of smaller crops where it may be necessary to test whether a
metabolites whose composition differs greatly transgenic variety is ‘substantially equivalent’ to
according to tissue, developmental stage, and in non-transformed varieties of the same crop (Beale et
response to environmental conditions. The ability al., 2009).
to simultaneously analyse large numbers of often The proteome is defined as the expressed
complex molecules is the basis of the so-called protein complement of an organism, tissue, cell or
ómic technologies. Hence, transcriptomics is the subcellular region (such as an organelle) at a specified
analysis of transcribed genes in the form of mRNA; stage of development and/or under a particular
proteomics is the analysis of protein composition; set of environmental conditions. Perhaps the most
lipidomics is the analysis of lipid composition; important molecules in cells are the proteins, some
metabolomics is the analysis of small metabolites, of which are structural while many others act as
and so on. Several automated analytical techniques enzymes responsible for the biosynthesis of most
have been developed to separate and identify each of the other molecules in a cell. Proteins are the
of these classes of biomolecules. direct products of gene expression and the timing
The transcriptome is a comprehensive list and spatial distribution of their accumulation and
of the genes expressed in a particular tissue at a function results in the phenotype of a particular
particular stage of development and/or in response organism. However, patterns of gene expression
to particular environmental stimuli. In many ways, as measured by transcription, i.e. the formation
transcriptome sequences can be much more useful of mRNA, are not always reflected by patterns
than genome sequences because they only include of accumulation or activity of the corresponding
the particular fraction of the tens of thousands of proteins.
genes that are expressed under specific conditions. In some cases, the mRNA may not be efficient-
In the case of oil palm, the analysis of the fruit ly translated to protein. In other cases, the protein
transcriptome during oil accumulation is already might be synthesised but is then either broken down
proving very useful in identifying key genes that or remains inactive. A protein might be present in
may regulate this vital process (Bourgis et al., 2011; a cell but is inactive due to incomplete post-trans-
8
lational processing, e.g. failure to bind a ligand, or automatically using algorithms, or repetitive step-
due to inhibition, e.g. by phosphorylation. There are by-step mathematical procedures.
many examples where proteins might be present in Other algorithms are used in genome annota-
a cell but remain in an inactivated state until they are tion. This involves the identification of putative
activated by a specific stimulus. In such cases, both genes, including their promoters, regulatory ele-
transcriptome and proteome data would indicate ments, introns, exons, and mRNA/protein prod-
that the gene was active and the protein was being ucts. Other software can detect possible regions
synthesised but this would be misleading in terms of encoding small, non-coding RNA and specific repet-
function if the protein was not active. Ideally, the in- itive elements in genome sequences. Such sequences
formation in the proteome should therefore include are now known to play important roles in several
any post-translational processing undergone by each aspects of genome function in complex eukaryotes
st
protein analysed. The 1 -generation proteomics was such as higher plants and animals (López-Flores and
mainly concerned with identifying the gross protein Garrido-Ramos, 2012; van Wolfswinkel and Ketting,
composition of samples, but new-generation tech- 2011). Software is also used to drive robotic and oth-
nologies are beginning to focus on questions such er automated systems used in tasks such as mass-
as post-translational processing and the biological profiling of large populations. Advances that enable
activities of such proteins (Murphy, 2011). Despite non-specialists to use sophisticated software have
these advances, it remains a significant challenge to been facilitated by improved computing technology
identify which proteins within a given proteome are and more powerful linked networks. This has been
partially or completely functionally active. especially crucial in enabling massive amounts of
12
It is only by addressing these latter questions data, often measured in many terabytes (10 bytes),
that we can verify not only that a particular protein that are generated by some of the new technologies.
has been synthesised and is in the right location, but For example, a single 2 hr run on an Illumina GAII
also that it has the appropriate biological function. DNA sequencer can generate 10 terabytes of data.
Therefore, we can learn a lot about the actual function One potential problem here is that the vast
of a genome in a specific cell or tissue by examining amounts of raw data generated by DNA sequencers
its proteome. Like the metabolome, the proteome are beyond the ability of many laboratories, or even
in a plant sample can vary greatly according to companies, to archive. Therefore, the raw data
genotype, tissue location, developmental stage, are often immediately processed by proprietorial
and environmental conditions (Gómez-Vidal et al., software developed by instrument manufacturers
2009; Zamri, 2013). The full proteome will comprise and only the much-reduced processed data are
thousands of proteins, some of which may be saved. Even with the most advanced computing
present in high abundance while others are at very technology, the costs of storing the original raw
low levels. The analysis of low-abundance proteins data can be greater than the cost of repeating the
poses considerable difficulties for proteomics that entire sequencing run. Another challenge for future
have yet to be resolved but given the rate of progress software development is to improve the assembly of
it is likely that automated or semi-automated processed sequence data for the increasingly diverse
methods will be developed for the near-complete applications required by researchers. To address this,
description of the oil palm proteome in the not too new forms of open-source bioinformatics software,
distant future. such as SOLiD, are being developed where members
of the community can adapt and improve software
Bioinformatics tools to fit their own applications.
One of the challenges with the automatic
Bioinformatics is a relatively new discipline annotation and public release of genomic data can be
that brings together biologists, mathematicians and the lack of quality control mechanisms that ideally
computer scientists to make sense of the avalanche require the intervention of human experts. There are
of data generated by genome sequencing and now many hundreds of publicly accessible genomes
profiling programmes; and from the other ómic that have already been annotated, sometimes by
technologies described above (Kanehisa and Bork, older methods that have now been greatly improved
2003). The sheer volume of data generated by these upon. In some cases, different laboratories use
methods often makes it virtually impossible to different algorithms for annotation, which can
analyse raw results manually. For example, a next- complicate cross-genome comparisons. Ideally
generation DNA sequencer can generate thousands more generic tools should be developed and some
of sequence fragments making up millions of degree of editing or re-annotation made possible
base pair readouts per day. These fragments need as improved methods are invented. For example,
to be analysed for overlaps and then assembled in a collaboration with Fujitsu, we have recently
into ‘contigs’, or continuous sequences of many developed an improved tool called cisExpress for
fragments that will eventually be collated to make the detection of specific motifs in genome sequences
up an entire chromosome. This process is now done (Tříska et al., 2013). We have also publicly released
9
the tool at www.cisexpress.org where it is available this particular case of mis-annotation but it is likely
as a stand-alone open-source application or via a that there are similar examples in genome databases.
web interface. One study of protein sequences in public databases
In the future it will be desirable for bioinformat- found surprisingly high levels of mis-annotation
ics researchers to work closely with biologists to de- that averaged 5%-63% across the six superfamilies
velop a wider range of broadly applicable tools for (Schones et al., 2005). This is one of the many reasons
the extraction of useful information from the various why it is beneficial for genomic data to be made
‘omics databases related to oil palm. As this is pre- publicly available in such a way that the research
competitive research, both the databases and ana- community as a whole can help to improve the
lytical tools such as algorithms should ideally reside quality of the data and the accuracy of its annotation.
in the public domain as open-source products. The
magnitude of this problem is demonstrated in Fig- Marker-assisted Selection (MAS)
ure 1, which shows how the massive decrease in se-
quencing costs has resulted in an avalanche of gene Several types of genetic marker can be used
and protein sequence data in public repositories. to assist the selection of favourable traits in plant
However, there are currently more than one mil- breeding. Morphological and biochemical markers,
lion uncurated protein sequences for each curated such as fruit colour, fatty acid composition, or
sequence. Clearly, there is an urgent need to curate dwarfism, are relatively easy to observe or measure
and validate many more genes and proteins in these but many other key agronomic traits such as disease
databases. resistance are not so easily assessed in this way.
A major aspect of the usefulness of public By far the most useful class of genetic markers are
databases is the ability of the entire research those based on DNA sequences. Such markers are
community to access and curate their contents at a now being applied to almost every aspect of plant
level of detail impossible in a non-public resource and animal breeding, and also in medicine, basic
as shown in the following example. Many DNA research and even in forensic science. The use of
sequencing projects are now performed in specialist modern techniques like association genetics and
commercial laboratories where tissues from several quantitative trait loci (QTL) analysis are enabling
different sources may be being analysed at the chromosomal regions and individual genes
same time, with the obvious potential for cross involved in the regulation of important traits to be
contamination. The result of such contamination mapped and identified (Rafalski, 2010; Xu, 2010).
might be the erroneous inclusion in a genomic These methods have recently been used to map
database of a DNA sequence from a different the lipase gene involved in oil deterioration in ripe
organism. Unfortunately, this is not always easy to palm fruits (Morcillo et al., 2013) and QTL analysis of
detect because, as we are increasingly recognising, genes regulating the fatty acid composition of palm
naturally occurring horizontal gene transfer between oil (Montoya et al., 2013).
unrelated organisms is much more common than DNA-based MAS can save time and money
previously suspected (Boto, 2010; Keeling and in crop breeding programmes as follows. In order
Palmer, 2008). There are now numerous examples of to select most characters of interest, it is normally
genes originating from animals, fungi and bacteria necessary to grow up and analyse each new
being found in some plant genomes (Bock, 2009). generation of the crop before it is possible to perform
However, it can also be the case that an anomalous phenotypic selection of appropriate plants. Many
DNA sequence may be present in a genomic database traits, such as disease resistance or salt-tolerance
due to human error or contamination rather than as cannot be measured until plants have been grown,
a result of horizontal gene transfer. often to full maturity, and then tested in the field.
For example, during an analysis of a public A DNA-based molecular marker is used to identify
database, I was initially intrigued to discover a plant- a segment of genomic DNA within which allelic
like gene in the genome of a tick and the possible variation in sequence has allowed its location to be
implication that the gene had been somehow genetically mapped. In breeding programmes, such
transferred from a plant to an animal. However, markers are chosen because of their close proximity
further investigation showed that this gene was to a gene of interest so that the marker and target
identical to that of a lipid peroxygenase gene in gene are inherited together. This enables breeders to
a plant genome that was being sequenced in the use the marker as a relatively straightforward way
same laboratory as the tick genome. Therefore the of screening very large populations for the presence
‘tick’ peroxygenase gene was not another example of a target gene without needing to perform complex
of horizontal gene transfer but rather was the result phenotypic tests. Hence, MAS can be used to track
of contamination of the tick DNA by plant DNA the presence of useful characters in large segregating
followed by their mistaken inclusion in the same populations in crop-breeding programmes. Using
genomic database. Luckily, we were able to resolve molecular markers, breeders can screen many more
10
plants at a very early stage and save several years use PCR-based methods, such as sequence-tagged
of laborious work in the development of a new crop microsatellites and SNP. By using DNA markers
variety. in conjunction with other new breeding technolo-
This is especially useful for crops like oil palm gies like clonal propagation, it should be possible to
where it can take three to four years or more for a make rapid strides in the creation and cultivation of
fruit phenotype to become fully apparent. Molecular greatly improved varieties of crops like oil palm.
markers have now been developed for most of the
major commercial crops, including several tree Transgenic Technologies
species. In addition to their increasingly prominent
role in genetic improvement of crops, molecular Numerous high-tech breeding tools have been
markers are useful for many other applications successfully used for crop improvement over the
such as characterising crop genetic resources, past 50 years. Transgenic technologies have been
management of gene banks, and disease diagnosis. available since the 1880s and enable breeders to
At present, MAS systems are being developed for manipulate genomes using recombinant DNA
oil palm by several international public-private methods that are continually being improved and
partnerships (PPP) and comprehensive genetic and refined. It is important to realise that breeders
physical maps of the genome are now available. never employ transgenesis on its own; instead
Genetic maps have recently been used to localise oil it is used in combination with technologies such
palm genes involved in the regulation of important as tissue culture/regeneration, hybrid creation,
traits such as fatty acid composition (Singh et al., mutagenesis, backcrossing, and MAS. This means
2009; Montoya et al., 2013), embryogenesis and that it can be misleading to speak of a new crop
callogenesis (2013), seed coat thickness (Singh, variety as ‘transgenic’ or ‘GM’ as if it had only
2013b). been created using transgenic technologies. In
Several types of DNA-based marker have been reality, the transgenic or GM stage is just one part
developed for basic research and plant breeding and of the initial creation of new genetic diversity at the
these are constantly being refined to increase their beginning of the breeding process. After this initial
utility and decrease costs. Early markers included stage, there are many other non-GM stages required
RFLP (restriction fragment length polymorphisms), before the plants can be used in agriculture. Hence,
AFLP (amplified fragment length polymorphisms), in 2012 about 180 million hectares was reported as
and RAPD (random amplified polymorphic DNA). being planted with transgenic/GM crops around
More recently, MAS has used much cheaper and the world. It is estimated that the total global land
more informative PCR-generated markers, such area used for crop production is about 1500 million
as microsatellites and SNP (single nucleotide hectares (Thenkabail et al., 2010), which means that
polymorphisms). Microsatellites are short sequences transgenic crops already occupy 12% of total arable
of repetitive DNA, such as poly(TG), that can be area (Table 4). However, each of these crops has also
amplified using PCR: they are highly polymorphic benefited from one or more of the non-transgenic
with respect to the numbers of the repeat units technologies listed above. For example, well over
between individuals in a population. Even in three-quarters of all crops grown, including almost
relatively inbred crop species, microsatellites are all transgenic varieties, have resulted from some form
polymorphic, enabling individuals to be genotyped of conventional hybridisation and backcrossing.
separately. Detailed genetic maps based on Despite the fact that transgenesis is simply
microsatellites are available for most major crop one of several alternative strategies for variation
species. Some of the most useful markers are SNP. enhancement in breeding programmes, the resultant
SNP occur very frequently in genomes, e.g. once plants are treated very differently from almost-
every 60-120 bp in maize, and are used widely both identical non-transgenic varieties by government
for research and breeding. agencies and by sections of the general public.
The use of MAS in crop breeding was initially re- Transgenic varieties have a different legal status
stricted to a few economically important temperate and are subject to much more complex regulatory
crops that are bred and marketed by major private systems in various regions of the world, which
sector firms, but the list of MAS-enabled crops is ex- can hinder their development and uptake. Indeed
panding. As well as annual crops such as cereals and transgenic crops are even banned or heavily
legumes, MAS has been useful in perennial crops, restricted in some countries. For this reason, we
including subsistence and cash crops in develop- need to look at the development of transgene
ing countries. Examples include oil palm, coconut, technology in a different way to other technologies.
coffee, tea, cocoa, and tropical fruit trees such as As we will see below, some developments such as
bananas and mangoes. MAS technologies have also so-called ‘clean gene’ technologies are aimed more
benefited from more efficient screening methods in- at satisfying generalised public concerns rather than
cluding PCR, DNA/DNA hybridisation, and DNA addressing proven safety issues or wider aspects of
sequencing. Most MAS technologies in crops now crop improvement per se.
11
TABLE 4. MAJOR GLOBAL TRANSGENIC CROPS AND challenges. For example, it is still unclear whether
TRAITS IN 2012
biolistics or Agrobacterium-mediated gene transfer
will be the gene delivery method of choice (Izawati
Area et al., 2012; Parveez and Bahariah, 2012). The choice
Crop %
(million hectares)
of plant material is also crucial with some of the
Soyabean 83 50 best options including various types of callus cul-
Maize 55 31 ture. Then there is the issue of which gene promot-
Cotton 21 14 ers to use. Unlike most existing commercial trans-
Rapeseed/canola 8 5
genic crops where constitutive viral promoters are
used, the manipulation of palm oil composition will
Sugar beet 0.5 <1
require deployment of strong mesocarp-specific
Alfalfa 0.1 <1 or kernel-specific gene promoters, ideally sourced
Others <0.1 <1 from the oil palm genome itself. Also, in order to
Total 170.3 100 achieve a high oleic oil, it will probably be neces-
sary to down-regulate some genes while adding or
Area up-regulating other genes. Even when the primary
Trait %
(million tonnes)
transgenic plantlets have been produced they will
Herbicide tolerance 100 60 still need to be grown on for three to five years to
Insect resistance (Bt) 28 18 obtain fruits that can be screened for oil content.
Herbicide tolerance + Insect 43.7 22 Finally, these primary transformants will need to
resistance 25 be taken through several sexual generations, back-
Others <0.1 <1 crossed with existing elite lines, and then multiplied
via micropropagation before any new commercial
Total 170.3 100
transgenic varieties can be released.
Source: Data from James (2012). Global status of commercialised It is likely that this process of transforming oil
biotech/GM crops: 2012. ISAAA Brief 44. http://www.isaaa. palm will take several decades. However, it is still
org/resources/publications/briefs/44/default.asp important that such programmes are continued for
the long-term future of crop improvement. This is
because there may be traits that are not possible
There are several ways in which transgene to create using non-transgenic approaches. For
technology can be improved to make it technically example, although it may be possible to create a
easier, more efficient, wider in its scope, and better medium-high (55%-65%) oleic oil phenotype via
able to address concerns expressed by certain conventional methods, a more desirable ultra-high
sections of the public. Some technical issues and (80%-90%) oleic trait might only be possible via
areas of public concern are listed below: transgenic technology. Another example where
• in the future, it will be desirable to generate the use of transgenic methods is essential is the
transgenic crops that do not contain selection production of completely novel compounds, such
markers, such as genes for antibiotic or herbicide as biopolymers like polyhydroxyalkanoates, in
tolerance; oil palm. In this case, it is necessary to transfer
• until now transgenic plants have been created several bacterial genes to the crop. Meanwhile,
using random insertion of transgenes, which newer technologies such as RNAi, trait stacking,
can lead to variations in transgene behaviour chromosome engineering, pathway engineering and
and other unpredictable pleiotropic effects. more efficient gene cassettes will play important
In order to achieve stable and predictable roles in expanding the scope of transgenic crops in
transgene expression under a variety of field the future.
conditions, transgene introduction technologies
need improvement;
• the spread of transgenes into wild populations SOME KEY TARGETS FOR BIOLOGICAL
via cross pollination can be prevented using IMPROVEMENT OF OIL PALM
genetic use restriction technologies (GURT); and
• biocontainment strategies should be incorpo- The biology of a long-lived crop such as oil palm is a
rated into certain types of transgenic plants, e.g. complex topic with many fascinating aspects relating
expressing non-edible or pharmaceutical prod- to its ecology, physiology and agronomy. The large
ucts to prevent risk of contamination of human size and long generation time of oil palm create many
or animal food/feed chains. formidable challenges for researchers and breeders,
especially in comparison with the much smaller
The development of transgenic oil palm is still annual oilseed crops like rapeseed or soyabean. It
in its infancy and is fraught with many technical is therefore essential that biological approaches to
12
oil palm improvement focus on a limited number of Palm oil yield in the plant is primarily deter-
key target traits. As discussed below, by far the most mined by biological factors such as crop genetics and
important biological traits relate to overall oil yield the incidence of pests and diseases. It is known from
and quality, together with tolerance to pests and the study of other crops that the overall yield of a
diseases. The role of improved farming practices is key crop product such as starch, oil or protein can be
discussed under crop management in the Role of manipulated in several ways. The most direct route
Improved Management in Addressing Challenges is to select plants where their biosynthetic pathways
for Oil Palm section. are more efficient at making the desired end prod-
uct at the expense of other less desirable products.
Oil Yield In the case of oil-bearing seeds or fruits this means a
greater flux of carbon towards oil and reduced flux
The relatively poor yield of palm oil in commer- towards other less useful products such as starch or
cial and smallholder plantations compared to the fibre (Rahman et al., 2013). Genetic loci that regulate
proven potential yield of some existing varieties, much of the flux of carbon towards oil in soyabean
coupled with the lack of progress in improving this have been described (Chung et al., 2003). Recent ad-
situation over the past decade, has attracted much vances in genomics and biochemistry have identified
comment in the industry and more widely (USDA, regulatory genes such as WRI1 that are able to mas-
2012). In Malaysia, for example, oil palm yields ac- sively up-regulate oil accumulation, even in tissues
tually declined between 2008 and 2012 to levels such as leaves that do not normally accumulate high
-1
that were 9% below the 4.7 t ha achieved in 2008 levels of storage lipids (Bourgis et al., 2011; Ma et al.,
(Figure 2). In contrast, yields of most major commer- 2013). The continued scope for dramatic increases in
cial crops have experienced steady and sustained oil yield even in established oil crops has been high-
increases in the region of 1%-2% per year, over the lighted by the recent discovery of rapeseed varieties
past 20-30 years. To a considerable extent, this situa- that accumulate almost 65% w/w oil in their seeds
tion is the result of a failure to replant new or ageing compared to about 45% or less in most commercial
plantations with the best available varieties. How- varieties – a yield gain of about 44% (Hu et al., 2013).
ever, there is also immense and largely untapped Although palm fruits are considered as relatively
scope for far greater increases in oil yield via mod- enriched in oil, a freshly picked ripe fruit bunch only
ern breeding methods. If new oil palm varieties pro- contains about 21%-23% w/w oil. In contrast, mature
ducing 10-20 t ha-1 oil become widely available this oilseeds typically contain 40%-46% w/w oil while
would be an enormous incentive for the industry to some oil-rich nut crops contain as much as 75% w/w
accelerate its sluggish replanting programme. oil (Murphy, 2010). The other major oil-rich drupe
MALAYSIA: Historical Palm Oil Yield Statistics
5.00
4.75
4.50
4.25
t ha-1
4.00
3.75
3.50
3.25
3.00
92/93
93/94
94/95
95/96
96/97
97/98
98/99
99/00
00/01
01/02
02/03
03/04
04/05
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
Yield
Source: Data from USDA (2012).
Figure 2. After a period of stagnation in the mid 1990s, palm oil yields gradually rose from 1998 until 2008 but then declined from 2008 until 2013.
There are probably several causes for the recent decline, including failure to replant ageing plantations, conversion of new land in East Malaysia, and
labour issues as discussed in the main text.
13
crop is olive, where the oil content of the whole fruit in orchard plantations as medium sized bushy trees
is generally in the range of 18%-28% w/w, although that could reach 20 m in height. Fruit picking was
some varieties can accumulate as much as 34% w/w difficult and most of the biomass of the crop was in
oil. Avocados can accumulate between 5% and 30% the form of wood rather than fruits. A combination
of fruit weight as oil. This indicates that it may be of genetic selection and careful pruning has now
possible to create new varieties of oil palm that transformed these trees into much smaller, 1-3 m
accumulate much higher levels of oil – possibly in tall, vine-like plants that produce much higher fruit
the region of 30%-40%, which would be double the yields and much less wood. A further bonus is that in
current oil yield per fruit. The recent publication many cases these short plants can now be harvested
of the annotated genomic sequences of several mechanically which saves greatly in terms of labour
oil palms (Singh, 2013) and several biochemical, costs and fruit spoilage. Mechanisation of fruit
transcriptome and molecular mapping analyses of harvesting has also been successfully developed in
palm fruits (Bourgis et al., 2011; Dussert et al., 2013; citrus crops (Murphy, 2011) and several systems are
Montoya et al., 2013; Morcillo et al., 2013; Ramli et under development for date palms. The development
al., 2009) have laid the foundations for identification of high-yielding dwarf oil palm varieties could open
and manipulation of the regulatory genes that may the door to full scale mechanisation of harvesting
facilitate the creation of much higher levels of oil in in plantations. This would save costs and would
palm fruit tissues in the foreseeable future. greatly alleviate the increasingly problematic labour
One possible way of increasing palm oil yield is issues confronting the industry as discussed below
to channel more carbon towards lipid biosynthesis, in the Labour Issue section.
and less towards other less valuable end products
such as starch or lignin. At present, a typical mature Oil Composition
oil palm of around 10-15 years of age is a relatively
tall plant with the fruit bunches produced at the Palm mesocarp oil typically contains about
top of a trunk about 10 m above the ground. There 35%-40% oleic acid plus 40%-50% palmitic and
are three major problems with this form of plant about 10% linoleic acids. By far the most important
architecture. Firstly, the majority of the carbon quality trait target for oil palm breeding is a much
assimilated via photosynthesis is used to produce higher oleic acid composition in the mesocarp oil.
a lignified trunk that has relatively little economic This would open up new markets for edible palm
value. The much more valuable fruit oil therefore oil and could eventually lead to the displacement
only represents a small fraction of the total crop of existing less efficient high oleic oilseeds such as
biomass. Secondly, it is difficult and laborious to soyabean, rapeseed and sunflower. There are several
harvest the fruits from such a height and there can precedents for the creation of completely new mar-
be much loss of yield due to spoilage when fruits ket opportunities by breeding high oleic varieties
fall onto the ground. Finally, the height of the fruit of oil crops. Perhaps the most dramatic example is
bunches makes it difficult to inspect them closely for that of rapeseed, which like other brassica oilseeds,
signs of pest and diseases or to assess the degree of such as mustard and crambe, historically produced
ripening. a seed oil that consisted of >50% erucic acid. While
In many other crops, ranging from annual cereals erucic acid can be used in some food applications,
to perennial orchard fruits, the manipulation of plant several studies claimed that its consumption by rats
architecture has already resulted in greatly increased was associated with cancers and in the 1960s rape-
yield. Probably the most dramatic example of this seed oil was banned from use in the USA. This led
is the creation of semi-dwarf varieties of the major Canadian breeders to develop new rapeseed vari-
th
cereal crops such as wheat, rice and barley in the 20 eties with 60%-65% oleic acid and very low levels
century (Murphy, 2011). These dwarf varieties are of erucic acid. These new forms of high oleic rape-
typically only 20%-40% as tall as traditional varieties seed were called ‘canola’ and soon dominated the
and much of the carbon saved in producing a shorter market, creating an entirely new multi-billion dollar
stem is used instead to form more grain. The genetic export crop for Canada and a new relatively cheap
basis of the dwarf phenotype is now known to be and nutritious vegetable oil for consumers around
associated with disruption of action of the hormone the world (Murphy, 2007).
gibberellin, which normally causes stem elongation. The new canola varieties were the result of
Oil palms can also produce short phenotypes and naturally occurring mutations that inactivated
the use of modern molecular breeding approaches the fatty acid elongase system responsible for
may make it possible to select dwarf trees that have converting C18:1 oleic acid to C22:1 erucic acid. In
higher yields of oil-bearing fruits. order to isolate and characterise these mutants, many
Similar modifications of tree crop architecture thousands of seeds from diverse accessions were
have already been achieved in the case of several laboriously screened in a process that took almost a
hard-fruit species of which the most dramatic decade. A similar approach was used to screen seeds
example is apples. Traditional apples were grown of sunflower, which normally contain high levels of
14
the polyunsaturate, linoleic acid (C18:2). In a few Pest and Disease Tolerance
cases, mutated sunflower seeds were found where
inactivation of oleate desaturase genes resulted in a Oil palm has numerous pests and diseases in the
greatly reduced ability to form linoleic acid and the major centres of cultivation in South-east Asia but
accumulation instead of an oil containing 60%-75% two of the most important are the fungal pathogen,
oleic acid. In other cases, induced mutagenesis has Ganoderma boninense, and the rhinoceros beetle,
been used to create new genetic variation in seed oil Oryctes rhinoceros, which can cause yield losses well
content. For example, this mutagenesis approach in excess of 50% in affected areas (Flood and Bridge,
was used to develop high oleic versions of linseed 2000; Panchal and Bridge, 2005; Flood et al., 2010). In
where the seed-specific desaturases responsible for terms of long-term management of these and other
converting C18:1 oleic acid to C18:3 linolenic acid harmful organisms, it is generally more effective to
were inactivated by several mutations (Murphy, enable the crop to tolerate small levels of infection/
2007). infestation rather than to aim at total eradication.
More recently, similar conventional (i.e. non- The problem with selecting for complete resistance,
transgenic) breeding approaches have led to the rather than tolerance to pests and pathogens, is that
development of very high oleic oils such as rapeseed/ it leads to high selection pressures for the emergence
canola with 75%; soyabean with 83%; sunflower of new variants of the pest/pathogen that can
with 80%-90%; safflower with 75%; and olive with overcome the resistance in the crop. Over recent
75% oleate. The use of induced mutagenesis in crop decades we have seen the emergence of numerous
breeding, including fatty acid manipulation, has new strains of pest/pathogen that have overcome
recently been made much more effective by the crop resistance and therefore pose serious threats to
automated mutagenesis/selection system termed major food crops such as wheat and rice.
TILLING (Murphy, 2011; Shu, 2009; Xu, 2010). In G. boninense is a soil-borne fungus that causes
other cases, transgenic (GM) approaches have been basal stem rot and has become an increasingly serious
used by commercial companies to produce very high problem especially in areas where palm plantations
oleate and low polyunsaturate varieties of some of have been present for many decades. In the past,
the major annual oilseed crops. Examples include research on Ganoderma has been hampered by its
rapeseed/canola (89% oleate); Indian mustard genetic and morphological variability, but the use of
(73% oleate); soyabean (90% oleate); and cottonseed biochemical and molecular genetic markers has now
(78% oleate). These transgenic lines are based on greatly improved the identification and localisation
antisense or RNAi technologies and several other of the more virulent strains (Bridge et al., 2000).
gene deletion technologies with potential use in oil More recently a joint Malaysian/US programme has
palm are also under development (Murphy, 2011). started to sequence the genomes of several virulent
In principle, all of the above approaches can and non-virulent strains of Ganoderma. This should
be used to create high oleic varieties of oil palm. help greatly in the identification of genes related to
Major challenges for palm breeders include the virulence and help breeders to develop more tolerant
long generation time of this tree crop, a lack of varieties of oil palm and/or to investigate the
genetic variation for oleate content, and developing feasibility of out-competing high virulence strains
the ability to use modern high-tech approaches of Ganoderma with other lower virulence strains. The
such as TILLING and MAS. To some extent, the latter approach would greatly reduce the likelihood
genetic variation issue is being addressed by of the re-emergence of virulence in the future.
extensive searches by MPOB and various plantation In terms of combating the immediate threat
companies for new germplasm in the oil palm of Ganoderma, there have also been advances in
centres of origin in West Africa and South America. using molecular genetic technologies to diagnose
Already some accessions have been found with infections at earlier stages. Previously, by the time a
oil contents of 50%-60% oleic acid, which is very Ganoderma infection was diagnosed, it was already
promising for introgression into current commercial too far advanced to save the tree in question, and
varieties. Perhaps more interestingly, this medium it was possible that neighbouring trees had also
to high oleic material can now be analysed by become infected. It has been reported that within
modern genomic, transcriptomic and proteomic 15 years of getting into a plantation, Ganoderma can
methods to elucidate the genetic and biochemical kill as many as 80% of oil palm trees (USDA, 2012).
basis behind the accumulation of higher levels of Early detection of Ganoderma infection is therefore
oleic acid in mesocarp oil. Such knowledge would an important prerequisite to its eventual control.
provide targets for future attempts to engineer fatty Although there is still a long way to go, there are
acid and oil metabolism so that palmitate was more promising signs that improved diagnostic methods
effectively elongated to stearate, desaturated to will eventually help in the identification and
oleate, and then transferred to glycerol to form the management of this serious pathogen (Bridge et al.,
triacylglycerol-rich mesocarp oil. 2000; Panchal and Bridge, 2005).
15
The rhinoceros beetle has emerged as the major options for additional value creation. For the first
pest of oil palm in South-east Asia since the 1980s. option, vitamin-rich virgin palm oil, which is bright
Chemical insecticides can be effective but such red in colour, could be more effectively marketed
agents are expensive as they can sometimes also as a healthy vegetable oil similarly to virgin olive
affect beneficial insects, and the target organisms oil, which is green and cloudy. Such a campaign
may develop resistance as has been found with would be more effective if high oleic varieties of
many other insecticides. Among the most promising palm oil are developed. The second option would be
biocontrol strategies for this pest are the deployment to develop a more cost-effective process to recover
of two efficient pathogens of the beetle, namely these useful compounds during oil processing and
the entomophagous fungus Metarhizium anisopliae then sell them as affordable dietary supplements
and the Oryctes virus (Ramle et al., 2005). Both or for other applications. Another possibility is to
pathogens are specific to rhinoceros beetles and as manipulate the oil composition to include desirable
such will not affect other insects. The Oryctes virus fatty acids such as DHA (docosahexaenoic acid) or
appears to be endemic in the beetle population, EPA (eicosapentaenoic acid) that are currently found
and deliberate augmentation can raise its infection in some fish oils. This would involve transgenic
levels to above 75%. The Metarhizium fungal spores methods and would take well over a decade to
can be applied to areas of infestation as a spray achieve. It is also uncertain whether there would be
that is highly effective at controlling, but not totally a sufficiently strong market demand for substitute
eradicating, the beetles. The combined use of these fish oils to make such a breeding programme
and other natural pathogens of the rhinoceros beetle worthwhile.
have the potential to reduce its harmful impact on In terms of non-food markets, palm oil
the crop, while also minimising risks of resistance composition could be modified to include
development. high levels of various industrially useful fatty
With the projected increase in oil palm acids, although even in the case of more easily
replanting over the next few years, it will be manipulated annual oilseeds progress in this area
important to consider the wider release of such has been disappointingly slow over the past few
biocontrol agents into areas where the incidence of decades, so this remains a very long-term option
rhinoceros beetles is particularly high. These and for oil palm (Murphy, 2009a). Another interesting
other forms of integrated pest management are possibility is to engineer the accumulation of
being investigated as primary options in plantations biodegradable biopolymers in palm fruits (Sudesh,
across South-east Asia (Caudwell and Orrell, 1997). 2013). This technology is gradually being developed
The rapid expansion of high intensity commercial in some annual crops where the transfer of three
plantations in new regions such as West Africa and bacterial genes can enable plants to convert acetyl-
South/Central America will doubtless lead to the CoA into polyhydroxyalkanoate beads instead of
emergence of new pests and pathogens. Therefore, it triacylglycerol oils (Gumel et al., 2013). Providing
will be important for the public sector and industry these beads can be extracted from the plant tissue
to work together in developing improved methods
(which is currently proving rather challenging),
of surveillance and early detection of such threats.
they can be used to manufacture several types of
thermoplastic materials that, unlike conventional
Other Traits
petroleum-derived polymers, are both renewable
and biodegradable (Murphy, 2010).
The three trait groups discussed above (oil yield,
A major issue with all of these oil-modification
oil quality, and pest/disease tolerance) are by far
schemes is that palm fruits that have been modified
the most important priorities in terms of increasing
the overall output of the oil palm industry and in to produce new types of oil would need to strictly
expanding its market share into the lucrative high segregated after harvesting and in all of the
oleic sector that is currently dominated by the downstream processing steps. This would add
temperate oil crops. There are many other possible considerably to production costs and it is far from
targets for trait modification but in terms of global certain whether sufficiently reliable markets would
income generation they must be regarded as of be available to guarantee adequate returns on the
secondary importance to these three priority traits. high investment and operating costs on such niche
In terms of edible and health markets, virgin products. In summary, most of these additional
palm oil is already enriched in carotenoids, traits can certainly be considered as part of a long-
lycopenes, tocopherols and tocotrienols that are term wish list for oil palm R&D but they cannot be
sources of the lipophilic vitamins A and E. For most justified to the same extent as the three major priority
edible applications, these useful compounds are traits, namely oil yield, oil quality, and pest/disease
removed during processing which presents two tolerance.
16
THE ROLE OF IMPROVED MANAGEMENT IN thanks to the considerable improvements in

ADDRESSING CHALLENGES FOR OIL PALM education and economic prospects over recent
decades, very few young Malaysians now wish to
The study and manipulation of oil palm biology and do this type of heavy manual outdoor work. Instead,
genetics will continue to make possible considerable about 500 000 overseas workers, largely unskilled
advances in crop improvement as outlined above. young males from Indonesia, now provide the
However, the translation of these biological benefits basic labour force on commercial plantations. These
into increased oil yields in a given plantation or during labourers must harvest the palm fruits manually
downstream processing is a separate challenge that and also provide basic management services such
has not always received sufficient attention in the as pruning and fertilising trees and applying pest/
past. The actual amount of palm oil produced at a pathogen control measures.
national or regional level can be affected by many There are several problems with this labour
non-biological factors such as climate, soil type, force that make it increasingly difficult to ensure
and the effectiveness of crop management systems. that optimal yields of palm fruits are harvested
Of these by far the most important set of factors and efficiently delivered to processing mills. Firstly,
is the overall management of the cropping and there is an overall shortage of manpower caused
processing systems. To a great extent, deficiencies by government policy to restrict the numbers of
in these systems are responsible for the otherwise migrant labourers granted visas. Secondly, better
puzzling failure of oil palm producers to obtain economic prospects in Indonesia itself, including
yield increases comparable with other major crops, the expansion of oil palm plantations in that country,
including their oilseed competitors. A recent study have created more opportunities for employment
by the US Department of Agriculture (USDA, 2012) and have reduced incentives to travel to Malaysia for
has highlighted several areas of concern which will such relatively low paid work. Thirdly, the migrant
now be discussed. labourers are restricted to five-year visas, meaning
that they have little opportunity to become highly
The Labour Issue experienced in plantation work, especially compared
to the original Malaysian workers who often did
As shown in Figure 2, palm oil yields per hectare such jobs for decades. A high level of experience is
in Malaysia were relatively stagnant from 1992-1997, vital for tasks like recognising when a fruit bunch
rose slowly over the next decade, but then actually is ready to pick or to spot potential problems such
decreased during the period from 2008-2013. This as the initial symptoms of pest or disease outbreaks.
is of considerable concern because Malaysia is the If fruits are harvested too soon or too late the oil
longest established centre for large-scale commercial yield can be greatly reduced, and failure to pick up
oil palm production and is also the world leader pest/disease problems can result in huge losses of
in R&D into the biology and management of the productivity and increased mitigation costs in the
crop (Abrizah, 2012). Current average oil yields future.
for Malaysia as a whole are only slightly above 4 t The labour issue is especially acute in East Ma-
-1
ha . However, some plantations are able to routinely laysia where the Sarawak Oil Palm Plantation Asso-
achieve yields over 6 t ha-1, which points to a failure ciation has reported a 20%-30% manpower shortfall
in other parts of the industry either to plant the that has resulted in 15% losses due to rotting fruits.
best available varieties and/or to manage cropping Note that this region has also experienced the great-
and processing systems in a more efficient manner est recent decline in palm oil yield (Figure 3), which
(Murphy, 2007). indicates that the yield problem is mainly associated
As shown in Figure 3, there are also regional with labour issues rather than other factors. There
discrepancies of palm oil yield within Malaysia. have been calls for the plantation sector to institute
Following a period of similar oil yields from 2001- an across the board increase in wages combined with
2007, yields in the East fell from 2007-2013 while enhanced medical and educational benefits to the
those in the West stagnated. To some extent, this migrant labourers in order to retain their services.
may be due to the creation of new plantations on So far such appeals have been resisted by much of
peat soils in East Malaysia and a consequent yield the industry despite relatively high palm oil prices
lag. However, it is likely that management factors and healthy gross margins that have been estimat-
have also played a considerable role in the relative ed at 60%-80% by MPOB. Meanwhile, the govern-
underperformance in East Malaysia. One of the ment is loath to increase the number of visas for a
most important of these factors is the shortage of combination of security and political reasons. While
appropriate labour. The original labour force on oil there may be good reasons for this lack of action to
th
palm plantations in the mid-late 20 century was confront the labour issue, it does mean that labour
largely made up of Malaysian citizens. However, problems will continue to dog the industry and their
17
MALAYSIA: Regional Palm Oil Yields
4.5
t ha-1 4
3.5
2.5
92/93
93/94
94/95
95/96
96/97
97/98
98/99
99/00
00/01
01/02
02/03
03/04
04/05
05/06
06/07
07/08
08/09
09/10
10/11
11/12
West Malaysia East Malaysia
Source: Data from MPOB (2013).
Figure 3. West (peninsular) Malaysia has historically tended to have higher palm oil yields that East Malaysia but the yield gap between the two
regions has widened since 2008.
possible knock on effects on oil yield seem destined are still relatively young. In contrast, Malaysia is a
to become more serious in the coming years. more mature region where it is estimated that 26% of
One possible long-term solution to this seem- plantations are 20-28+ years old while 65% are 9-28+
ingly intractable problem is to reduce labour re- years old (USDA, 2012). Therefore, the majority of the
quirements via a combination of radically increased trees in Malaysia have already reached or will soon
mechanisation and the redesign of the architecture reach their peak productivity. Over the next decade,
of the oil palm tree (Murphy, 2009b). Already there there will be inevitabled declines that will continue
have been some advances in the redesign of cutting to exacerbate the poor national yield statistics. The
equipment for fruit harvesting. The development of obvious solution is to replace these older oil palm
lighter, powered and more ergonomically efficient with some of the newer varieties that are already
fruit cutters can increase efficiency and improve available. This replanting needs to be carried out on
health and safety of the workforce. Other devices are a continuous basis so that lower yielding older oil
being developed to facilitate fruit transport and the palm, that may also carry higher loads of pests and
collection of loose fruits that have fallen from trees. pathogens, do not accumulate in plantations. Such a
However, the uptake of many of these innovations scheme would also make it possible to continually
has been limited by relatively high purchasing and refresh the genetic pool of the plantation trees with
running costs (Shuib et al., 2010). Additionally, the the best and most recent varieties available from
process of fruit harvesting is currently determined modern breeding programmes.
by the need to assess the ripeness and then to cut In reality, however, the situation is more com-
and lower a heavy fruit bunch that is at the top of a plex and economics, in particular, has played a role
10 to 20 m tree. As discussed above, the introduction in inhibiting the desire to replant. One significant
of dwarf varieties in several other crops has facilitat- factor has been the high prices available for palm
ed mechanisation and led to reduction of the labour oil in international markets. This has meant that, in
force to a much smaller and more highly skilled many cases, even an ageing plantation with old oil
cadre of specialists. If this could be achieved with palm varieties can still produce sufficient palm oil to
oil palm there would be little for unskilled foreign generate a healthy profit for the grower, who there-
labour and instead workers would have a much less fore has little incentive to replant. Moreover, the de-
arduous and more skilled job that might even ap- cision to replant is rather costly for the grower in the
peal to some local Malaysians. short-term. First, there is the expense of removing
the old oil palm and planting new seedlings, and
Replanting with Elite Germplasm second there is a lengthy period of lost productiv-
ity. As the new seedlings will produce no fruits for
An oil palm has a productive lifetime of 25-30 the first three to four years so the plantation owner
years with oil yields gradually rising to a peak from would have no income but still would need to man-
9-18 years followed by a gradual but steady decline. age the growing young oil palm. After three to four
As Indonesia is a relatively new centre of oil palm years, the oil yield will gradually rise to reach peak
production, many of the oil palm in its plantations output by 8-10 years. This scenario makes it a costly
18
decision even for a large commercial plantation to Addressing Sustainability Criteria

embark on replanting, but for a smallholder who
may only have 5-10 ha of oil palm that are all at the It is beyond the scope of this article to address
same age the income loss from replanting can poten- sustainability criteria in detail. However, the wider
tially be disastrous. environmental impacts of all forms of agriculture,
The Malaysian government is now addressing including oil palm cultivation, are rightly of
this problem and in 2013, it committed USD 135 considerable public concern and must be taken
million to facilitate a national replanting programme into account in future planning. Probably the major
that is particularly targeted at smallholders (USDA, concern about oil palm is the ongoing expansion
2012). The government has estimated that 365 000 ha of the crop area, especially into environmentally
of mainly smallholder oil palms are 25-37 years old, sensitive regions such as pristine rainforest and deep
which means that they are well beyond their normal peat soils. As outlined above, in the medium-term
productive lifetime. The aim is to replace 100 000 ha the need for such land expansion can be considerably
of ageing oil palm per year by providing grants to reduced by the yield benefits that can be brought
smallholders for the replanting costs plus an annual about by breeding and improved management. In
allowance of USD 1884 for the first two years of zero the shorter-term, efforts should be made to avoid
productivity. It is hoped that the larger commercial environmentally sensitive areas as much as possible.
plantations will also replant a further 100 000 ha In addition, where land conversion does occur the
-1
yr so that by 2018 one million hectares will have impacts on wildlife should be mitigated by the
been replanted. If this can be achieved, and with the inclusion of refuges and transit corridors.
conservative assumption that new oil palm capable The main areas of concern in this regard are in
of 5-6 t ha-1 replace the ageing stock producing 2-3 t Indonesia and some parts of the Malaysian Borneo.
ha-1, this could result in an additional oil yield in the As these two countries produce the vast majority of
region of three million tonnes in Malaysia by 2020. global exports of palm oil, it is in their joint interest
This >10% increase in yield can be readily achieved to ensure that valid environmental concerns are
without converting any new land to oil palm and addressed, for example by encouraging growers to
only relies on currently available varieties. In reality, join certification schemes such as RSPO (Roundtable
much higher yielding varieties will be available on Sustainable Oil Palm, www.rspo.org). To a great
from breeding programmes so there is even greater extent, the larger commercial plantation companies
potential to increase Malaysia palm oil yields in the that are already focussed on international trade have
coming years. both the means and the incentive to improve their
The replanting issue that is currently such an sustainability status. However, a major additional
urgent problem in Malaysia will also eventually challenge in some regions will be to meet RSPO or
affect Indonesia, which is an even larger palm oil similar criteria while still facilitating the economic
producer. Many plantations in Indonesia were development of the estimated three million oil palm
installed during a comparatively brief period, smallholders worldwide who face real difficulties in
meaning that they will also require replacement, complying with such schemes (ZSL, 2013).
and consequent loss of income for growers, at about It is in the interests of the oil palm industry
the same time. The situation in Indonesia will be as a whole to try to address these issues together,
exacerbated by the far larger proportion of oil palm rather than on a piecemeal company-by-company
cultivated by smallholders compared to Malaysia. or region-by-region basis. For example, the poor
Smallholders account for about 40% of the oil palm image of palm oil presented by some NGO has
area in Indonesia and over the next decade as much led to an increasing trend in some parts of Europe
as 500 000 ha yr-1 will require replanting. It would for boycotts of oil palm products by both retailers
therefore be prudent for the Indonesian government and consumers. Such blanket bans show that there
to use some of the considerable revenues it is now tends to be no differentiation between ‘good’ and
receiving thanks to a buoyant palm oil market to ‘bad’ sources of palm oil and the entire sector ends
set aside funds for a large-scale national replanting up being tarred by poor practice in a few areas. It is
programme in the early 2020s. Such a scheme, which also both logistically difficult and expensive to begin
would have a guaranteed bonus in generating segregating a globally traded commodity like palm
higher oil yields, could also reduce future pressures oil on the basis of provenance unless the market is
to convert pristine habitats to oil palm plantations. prepared to offer a significant premium, which is
Failure to implement oil palm replanting will mean unlikely given the current economic circumstances.
declining yields in the coming decade and, given There has been some progress in addressing
the likelihood of continuing international demand environmental issues over the past few years but,
for palm oil, impoverished smallholders might given the very real potential for further increases in
well be encouraged to embark on a new round of palm oil production, it is highly desirable that this
ecologically undesirable land conversion. huge global industry, which is worth over USD 50
19
billion annually, should collectively redouble its (Malaysia/Indonesia) is not ideal. The expansion
efforts to address sustainability and public image of cultivation into suitable areas of West Africa and
issues as a matter of priority. South/Central America that is now underway will
create a more secure production system in the longer
term.
FUTURE PROSPECTS FOR OIL PALM Currently, the three major centres of oil palm
cultivation in Central/South America are Colombia,
In this article, we have seen that there are powerful Ecuador and Honduras with a collective annual
drivers for the continued expansion of demand output of 2 million tonnes oil. According to IIASA
for palm oil in the mediums to long-term future. estimates, these three countries and Peru have some
In 2013, a global area of about 15 million hectares modest potential to expand cultivation but by far the
produced 65 million tonnes of palm mesocarp + largest new prospective area is in Brazil. Oil palm is
kernel oils. Forecasting future levels of demand for more suitable as a crop in low elevation regions in
any commodity is always challenging but estimates the humid tropics and can even tolerate the highly
from reliable sources predict requirements of about acidic non-forest soils of Amazonia (Butler and
77 million tonnes palm oil by 2018 (FAO) and Laurance, 2009). In Brazil, an estimated 32 million
between 93 and 156 million tonnes by 2050 (Corley, hectares (excluding rainforest) are suitable for oil
2009). Therefore, it can be confidently predicted that palm production (EMBRAPA, 2010), which is more
global demand will remain high and that there will than double the entire global crop area at present. In
be sustained pressure for yield improvements and contrast to South-east Asia and parts of West Africa,
additional land conversion for decades to come. the vast majority of this possible expansion into oil
Given these projections, it would be prudent for the palm in South America would be on grassland or
sector to invest significantly right now in priority planted pasture with very little forest conversion
R&D areas like yield and oil quality and to address (Pacheco, 2012). Conversion of such land would
management-related issues such as environmental therefore have a lower impact on biodiversity and
impact, product image, tree replanting, labour other sensitive environmental indicators that the
supply and mechanisation. conversion of tropical forest.
Some of these measures could have a significant West Africa is the historical home of the
impact on output within the next five years. For commercial oil palm plant and, prior to the 1960s,
example, if the planned replanting programme Nigeria in particular was a globally dominant
in Malaysia is carried out, it could deliver an producer of palm oil. Since then, civil conflict
additional annual yield of 5 million tonnes palm oil coupled with poor investment and management of
without the need to use any more land. If some of the largely smallholder dominated industry meant
the best existing experimental breeding material, that palm oil production declined until by 2000,
which could theoretically yield 8-10 t ha-1, can be it was unable even to met local demand and the
developed for commercial planting throughout the country became a net importer of edible oils (Ugbah
sector then yield could be increased by 50% or more. and Nwawe, 2008). However, the same buoyant
This could deliver as much as 30 million tonnes international demand for palm oil that is driving
more oil per year – again without requiring further land conversion in Central/South America is now
land conversion. Further into the future, there is the fuelling increasing investment in replanting disused
prospect of additional yield gains by using modern plantations or establishing new plantations in
breeding technologies to produce fruits with a higher several parts of West and Central Africa. A great deal
oil content and dwarf oil palm that bear more fruit of this expansion will be required simply to meet
and are easier to harvest mechanically. At present, local requirements for vegetable oil that is currently
we cannot quantify the benefits of such biological imported at considerable expense from abroad. For
innovations but they could potentially deliver tens example, in 2010, Africa imported 2.4 million tonnes
of millions of tonnes of additional oil. palm oil, mostly from Malaysia, despite its potential
Although, it should be possible to produce a lot capacity to produce this amount of oil, and much
more palm oil by increasing the crop yield, we should more, locally. For example, it is estimated that a
also accept that some additional land conversion staggering 24 million hectares could be potentially
will be necessary, particularly in the short-term. used to grow oil palm in Nigeria alone (Business
Providing this is carried out in an environmentally Day, 2013).
responsible manner there are benefits from In terms of climate and agronomy, another
diversifying oil palm cultivation into other regions of promising region for new oil palm cultivation in
the world. For example, a more dispersed cropping Africa is in the Congo river basin (Persson and
area will be more resilient to threats from climatic Azar, 2010). Plantation companies are also steadily
factors or locally adapted pests and pathogens. In acquiring land in other parts of Africa (Global
this respect, the current concentration of >85% of Forecasting Service, 2011). Some idea of the scale of
global palm cultivation in one geographical area these acquisitions can be seen from the following
20
figures of land purchases or leases by overseas ACKNOWLEDGEMENT

companies – note that the list is only a sample of the
total land being acquired: Congo DR – 1 080 000 ha; I wish to thank the two anonymous referees for their
Liberia – 600 000 ha; Gabon – 205 000 ha; Nigeria useful and constructive comments on this article.
– 110 000 ha; Congo, R – 110 000 ha; Sierra Leone –
100 000 ha; Cote d’Ivoire – 55 000 ha; Cameroon –
40 000 ha (Carrere, 2010). In 2013, it was estimated that REFERENCES
at least 1 million hectares in the region was currently
available for conversion to oil palm plantations. ABRIZAH, A (2012). A bibliometric study on the
Investors include a variety of overseas oil palm worldwide research productivity of scientists in
businesses based in South-east Asia and Europe as Elaeis guineensis Jacq. and Elaeis oleifera. J. Oil Palm
well as USD-based investment companies. There are Res. Vol. 24: 1459-1472.
also ongoing multi-billion investments by overseas
companies in new oil palm infrastructure, such as BEALE, M H; WARD, J L and BAKER, J M (2009).
processing plants, port facilities and transport links. Establishing substantial equivalence: metabolomics.
For example, in 2013 the French-based SIFCA Group Methods Mol Biol., 478: 289-303.
announced plans to spend USD 417 million on
plantations and processing plants in Ghana, Nigeria BOCK, R (2009). The give-and-take of DNA:
and Liberia. horizontal gene transfer in plants. Trends Plant Sci.,
Therefore, it is clear that over the next decade, 15: 11-22.
West and Central Africa will emerge alongside
Central/South America as a major producer of palm BOTO, L (2010). Horizontal gene transfer in
oil, initially for local consumption but eventually evolution: facts and challenges. Proc. of the Royal Soc
for export to global markets. Between them, West/ B 277: 819-827.
Central Africa and Central/South America have the
capacity to convert well over 30-40 million hectares BOURGIS, F et al. (2011). Comparative transcriptome
to oil palm with a current yield potential of 130- and metabolite analysis of oil palm and date
170 million tonnes oil and much more than that if palm mesocarp that differ dramatically in carbon
the expected increases in crop yields are realised. partitioning. Proc. Natl. Acad. Sci. USA, 108: 12527-
It is highly unlikely that such a vast area, which is 12532.
more than double the present oil palm area, will
BRIDGE, P D; O’GRADY, E B; PILOTTI, C A
be converted. However, even if only a quarter of
and SANDERSON, F R (2000). Development of
this land is changed to oil palm plantations, these
molecular diagnostics for the detection of Ganoderma
regions could be producing as much as 50-60 million
isolates pathogenic to oil palm. Ganoderma Diseases
tonnes oil by 2025. This is close to the present global
of Perennial Crops (Flood, J and Bridge, P eds.). CABI
total and demonstrates that a doubling of palm oil Publishing, Wallingford. p. 225-235.
production in the next few decades is quite feasible.
BRINGEZU, S et al. (2009). Assessing biofuels:
CONCLUSION towards sustainable production and use of resources.
United Nations Environment Programme, Paris,
France. Available online: www.unep.org
The oil palm industry faces many challenges in
the future. However, the tools to surmount these
BUSINESS DAY (2013). Reactivating Nigeria’s
challenges already exist and have the potential
oil palm industry. http://businessdayonline.
to further transform this historic crop into a truly
com/2013/08/reactivating-nigerias-oil-palm-
global source of nutritious food and valuable non-
industry-2/
food products for the growing world population.
We can look forward to an extension of oil palm CARRERE, R (2010). Oil palm in Africa: past present
cultivation right across the tropics of Africa, Asia and future scenarios. World Rainforest Movement.
and the Americas with higher yielding varieties that http://wrm.org.uy/countries/Africa/Oil_Palm_
have improved oil compositions. A redesign of crop in_Africa.pdf
architecture and improved pest/disease tolerance
could also facilitate its management and processing. CAUDWELL, R W and ORRELL, I (1997). Integrated
If these and other measures can be undertaken, pest management for oil palm in Papua New Guinea.
increased palm oil output could more than meet Integr Pest Manag Rev., 2: 17-24.
the highest projections for future vegetable oil
requirements and might even displace some of the CHUNG, J; BABKA, H L; GRAEF, G L; STASWICK,
less efficient temperate oilseed crops. P E and LEE, D J (2003). The seed protein, oil, and
21
yield QTL on soyabean linkage group I. Crop Sci., 43: HU, Z-Y; HUA, W; ZHANG, L; DENG, L-B and
1053-1067. WANG, X-F (2013). Seed structure characteristics to
form ultrahigh oil content in rapeseed. PLoS ONE,
COOK, D R and VARSHNEY, R K (2010). From 8(4): e62099.
genome studies to agricultural biotechnology:
closing the gap between basic plant science and IZAWATI, et al. (2012). Transformation of oil palm
applied agriculture. Current Opinion in Plant Biology, using Agrobacterium tumefaciens. Methods Mol Biol.,
13: 115-118. 847: 163-175.
CORLEY, R H V and TINKER, P B (2003). The Oil JOHNSTON, M et al. (2009). Resetting global
Palm. Fourth edition, Blackwell, Oxford, United expectations from biofuels. Environmental Research
Kingdom. Letters, 4: 014004.
DUSSERT, et al. (2013). Comparative transcriptome Kalidas, P (2013). Pest problems of oil palm
analysis of three oil palm fruit and seed tissues that and management strategies for sustainability.
differ in oil content and fatty acid composition. Plant Agrotechnology, S11: 001.
Physiol, 162: 1337-1358.
Kanehisa, M and BORK, P (2003). Bioinformatics
EDWARDS, D and BATLEY, J (2010). Plant genome in the post-sequence era. Nature Genetics Supplement,
sequencing: applications for crop improvement. 33: 305-310.
Plant Biotechnology Journal, 8: 2-9.
Keeling, P J and Palmer, J D (2008). Horizontal
EMPRAPA (2010). Dendê. Ministério da Agricultura, gene transfer in eukaryotic evolution. Nature Reviews
Pecuária e Abastecimiento. Brasil. Genetics, 9: 605-618.
FAO (2003). World agriculture trends: towards 2015- LIN et al. (2009). Transcriptome analysis during
2013. An FAO perspective. http://www.fao.org/ somatic embryogenesis of the tropical monocot Elaeis
docrep/005/y4252e/y4252e05d.htm guineensis: evidence for conserved gene functions in
early development. Plant Mol Biol, 70: 173-192.
FLOOD, J and BRIDGE, P (2000). Ganoderma Diseases
of Perennial Crops. CABI Publishing, Wallingford. López-Flores, I and Garrido-Ramos, M A
(2012). The repetitive DNA content of eukaryotic
FLOOD, J; COOPER, R; REES, R; POTTER, U and genomes. Genome Dynamics, 7: 1-28.
HASAN, Y (2010). Some latest R&D on Ganoderma
diseases in oil palm. Presented at the IOPRI/MPOB MA, W et al. (2013). WRINKLED1, a ubiquitous
Seminar: Advances in the Controlling of Devastating regulator in oil accumulating tissues from Arabidopsis
Disease of Oil Palm (Ganoderma) in South East Asia, embryos to oil palm mesocarp. PLoS ONE, 8(7):
Jogjakarta, May 2010, http://r4d.dfid.gov.uk/ e68887.
Output/185439/
Mardis, E (2008). The impact of next-generation
GLOBAL FORECASTING SERVICE (2011). sequencing technology on genetics. Trends in
Commodities - Softs - Palm Oil, Economist Genetics, 24: 133-141.
Intelligence Unit, http://gfs.eiu.com/Article.aspx?
articleType=cfs&articleId=1178124702 Meerow, A W et al. (2012). Coconut, date and oil
palm genomics. Genomics of Tree Crops (Schnell, R J
GÓMEZ-VIDAL, S; SALINAS, J; TENA, M and and Priyadarsham, P M eds.). Springer, New York.
LOPEZ-LLORCA, L V (2009). Proteomic analysis p. 299-351.
of date palm (Phoenix dactylifera L.) responses to
endophytic colonization by entomopathogenic Montoya, C et al. (2013). Quantitative trait loci
fungi. Electrophoresis, 30: 2996-3005. (QTLs) analysis of palm oil fatty acid composition in
an interspecific pseudo-backcross from Elaeis oleifera
GUMEL, A M; ANNUAR, M S M and CHISTI, Y (H.B.K.) Cortés and oil palm (Elaeis guineensis Jacq.).
(2013). Recent advances in the production, recovery Tree Genetics & Genomes. Published online 23 June
and applications of polyhydroxyalkanoates. J. 2013, DOI 10.1007/s11295-013-0629-5.
Polymers Environ, 21: 580-605.
Morcillo, F et al. (2013). Improving palm oil
GUNSTONE, F D (2011). Supplies of vegetable oils quality through identification and mapping of the
for non-food purposes. European Journal of Lipid lipase gene causing oil deterioration. Nature Comm,
Science and Technology, 113: 3-7. 4: 2160.
22
Murphy, D J (2007). Future prospects for oil palm RAMLE, M; WAHID, B; NORMAN, K; GLARE, T R
st
in the 21 century: biological and related challenges. and JACKSON, T A (2005). The incidence and use of
Eur. J. Lipid Sci. Technol., 109: 296-306. Oryctes virus for control of rhinoceros beetle in oil
palm plantations. Malays J. Invertebr Pathol., 89: 85-
Murphy, D J (2009a). Industrial oil crops, when 90.
will they finally deliver on their promise? Inform, 20:
749-754. RAMLI, U S et al. (2009). Use of metabolic control
analysis to give quantitative information on control
Murphy, D J (2009b). Oil palm: future prospects for of lipid biosynthesis in the important oil crop, Elaeis
yield and quality improvements. Lipid Technology, guineensis. New Phytol, 184: 330-339.
21: 257-260.
ROSILLO-CALLE, F; PELKMANS, L and WALTER,
Murphy, D J (2010). Manipulation of oil crops for A (2007). A Global Overview of Vegetable Oils
industrial applications. Industrial Crops and Uses with Reference to Biodiesel. A Report for the IEA
(Singh, B P ed.). CABI Press, UK. p. 183-206. Bioenergy Task 40, International Energy Agency,
http://www.bioenergytrade.org/downloads/
Murphy, D J (2011). Plants, Biotechnology and vegetableoilstudyfinaljune18.pdf
Agriculture. CABI Press, UK.
SCHNOES, A M; BROWN, S D; DODEVSKI, I and
Neoh et al. (2013). Differential metabolite profiling BABBITT, P C (2009). Annotation error in public
of metabolites in oil palm mesocarp at different databases: misannotation of molecular function
stages of oil biosynthesis. J Agric Food Chem. in enzyme superfamilies. PLoS Comput Biol., 5(12):
e1000605.
Oil World (2012). http://www.oilworld.biz/
app.php SHEARMAN et al. (2013). Transcriptome assembly
and expression data from normal and mantled oil
Pacheco, P (2012). Soyabean and oil palm palm fruit. Dataset Papers in Biology: 670 - 926.
expansion in South America. A review of main trends
and implications. CFOR working paper 90, Center SHENDURE, J and JI, H (2008). Next-generation
for International Forestry Research www.cifor.org/ DNA sequencing. Nature Biotechnology, 26: 1135-
publications/pdf_files/WPapers/WP90Pacheco. 1145.
pdf
SHU, Q Y (2009). Induced plant mutations in the
Panchal, G and Bridge, P D (2005). Following genomics era. FAO, Rome. http://www.fao.org/
basal stem rot in young oil palm plantings. docrep/012/i0956e/i0956e00.htm
Mycopathologia, 159: 123-127.
SHUIB, A R; KHALID, M R and DERAMAN, M S
Parveez, G K and Bahariah, B (2012). Biolistic- (2010). Enhancing field mechanization in oil palm
mediated production of transgenic oil palm. Methods management. Oil Palm Bulletin No. 61: 1-10.
Mol Biol, 847: 163-175.
SINGH, R et al. (2009). Mapping quantitative trait
Paterson, A H; Freeling, M; TANG, H and loci (QTLs) for fatty acid composition in an interspe-
WANG, X (2010). Insights from the comparison of cific cross of oil palm. BMC Plant Biology, 9: 114.
plant genome sequences. Annual Review of Plant
Biology, 61: 349-372. SINGH, R et al. (2013a). Oil palm genome sequence
reveals divergence of interfertile species in Old and
Persson, U M and AZAR, C (2010). Preserving the New worlds. Nature doi:10.1038/nature12309.
world’s tropical forests - a price on carbon may not
do. Environmental Science & Technology, 44: 210-215. SINGH, R et al. (2013b). The Shell gene of the oil palm
(Elaeis guineensis) controls oil yield and encodes
Rafalski, J A (2010). Association genetics in crop a homologue of SEEDSTICK. Nature doi:10.1038/
improvement. Current Opinion in Plant Biology, 13: nature12356.
174-180.
SOH, A C (2010). Genomics and plant breeding. J,
Rahman, H; Harwood, J L and Weslake, Oil Palm Res. Vol. 23: 1019-1028.
R (2013). Increasing seed oil content in Brassica
species through breeding and biotechnology. Lipid SUDESH, K (2013). Polyhydroxyalkanoates from Palm
Technology, 25: 182-185. Oil: Biodegradable Plastics. Springer, Berlin.
23
SULAIMAN et al. (2012). The potential of oil detection in DNA sequences. Bioinformatics doi
palm trunk biomass as an alternative source for 10.1093/bioinformatics/btt366
compressed wood. BioResources, 7: 2688-2706.
UGBAH, M M and NWAWE, C N (2008). Trends in
TAHIR, N I et al. (2013). Identification of oil palm oil palm production in Nigeria. J. Food Agriculture &
(Elaeis guineensis) spear leaf metabolites using mass Environment, 6: 119-122.
spectrometry and neutral loss analysis. J. Oil Palm
Res. Vol. No. 25: 72-83. USDA (2012). Malaysia: stagnating palm oil yields
impede growth. Commodity Intelligence Report
TEH et al. (2013). Differential metabolite profiles (December 2012). http://www.pecad.fas.usda.gov/
during fruit development in high-yielding oil palm highlights/2012/12/Malaysia/
mesocarp. PLoS ONE, 8(4): e61344.
VAN NOORDEN, R (2013). EU debates U-turn on
THENKABAIL, P S; HANJRA, M A; DHEERAVATH, biofuels policy. Key vote could signal withdrawal of
V and GUMMA, M (2010). A holistic view of global support from biodiesel. Nature, 499: 13-14.
croplands and their water use for ensuring global
st
food security in the 21 century through advanced VAN WOLFSWINKEL, J C and KETTING, R F
remote sensing and non-remote sensing approaches. (2011). The role of small non-coding RNAs in
Remote Sens., 2: 211-261. genome stability and chromatin organization. J. Cell
Sci., 123: 1825-1839.
TING, N-C et al. (2013). Identification of QTLs
associated with callogenesis and embryogenesis in XU, Y (2010). Molecular Plant Breeding. CABI, Oxford,
oil palm using genetic linkage maps improved with United Kingdom.
SSR markers. PLoS ONE, 8(1): e53076.
ZAMRI, R (2010). Differential Expressions of Proteomes
TRANBARGER, T J et al. (2011). Regulatory of Oil Palm (Elaeis guineensis Jacq.) and Arabidopsis
mechanisms underlying oil palm fruit mesocarp thaliana Tissues during Callogenesis. M. Sc. thesis,
maturation, ripening, and functional specialisation Universiti Putra Malaysia.
in lipid and carotenoid metabolism. Plant Physiol,
156: 564-584. ZSL (2013). New Britain oil palm: smallholder.
Overcoming challenges in certifying smallholders.
TŘÍSKA, M; GROCUTT, D; SOUTHERN, J; MURPHY, ZSL website: http://www.sustainablepalmoil.
D J and TATARINOVA, T (2013). cisExpress: Motif org/growers-millers/growers/case-studies/new-
britain-palm-oil
24
View publication stats

The Future of Oil Palm As A Major Global Crop: Opportunities and Challenges

Uploaded by

Copyright:

Available Formats

The Future of Oil Palm As A Major Global Crop: Opportunities and Challenges

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Future of Oil Palm As A Major Global Crop: Opportunities and Challenges

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Article in Journal of oil palm research · March 2014

Molecular and biochemical regulation of Aflatoxin biosynthesis View project

The user has requested enhancement of the downloaded file.

Journal of Oil Palm Research Vol. 26 (1) March 2014 p. 1- 24

THE FUTURE OF OIL PALM AS A MAJOR

Keywords: oil yield, high oleic, genomics, breeding, management.

INTRODUCTION t ha-1 for the best commercial plantations and 3-4

TABLE 1. MAJOR CENTRES OF OIL PALM CULTIVATION

Rank Country Production (million tonnes)

Source: Data from the United States Department of Agriculture (2012).

TABLE 2. INCREASE IN PALM OIL PRODUCTION IN INDONESIA AND MALAYSIA

Source: Data from FAO (2013).

Cost per Mb of DNA sequence (USD)

Curated vs. uncurated proteins

Pubmed ‘Metagenomics’ hits

2000 2002 2004 2006 2008 2010 2012

Source: Data from Pubmed and NCBI.

MALAYSIA: Historical Palm Oil Yield Statistics

Source: Data from USDA (2012).

THE ROLE OF IMPROVED MANAGEMENT IN thanks to the considerable improvements in

MALAYSIA: Regional Palm Oil Yields

Source: Data from MPOB (2013).

decision even for a large commercial plantation to Addressing Sustainability Criteria

figures of land purchases or leases by overseas ACKNOWLEDGEMENT

View publication stats

You might also like