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

net/publication/237351969

Yield Intensification in Oil Palm Plantations through Best Management Practice

Article · January 2009

CITATIONS READS
64 2,432

3 authors:

Christopher Richard Donough Christian Witt

53 PUBLICATIONS   629 CITATIONS   
Bill & Melinda Gates Foundation
64 PUBLICATIONS   4,022 CITATIONS   
SEE PROFILE
SEE PROFILE

Thomas Fairhurst
Tropical Crop Consultants Limited
50 PUBLICATIONS   1,600 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Oil Palm Best Management Practices in Ghana View project

Oil Palm Field Handbooks View project

All content following this page was uploaded by Christian Witt on 17 September 2014.

The user has requested enhancement of the downloaded file.

 SOUTHEAST ASIA

Yield Intensification in Oil Palm Plantations

through Best Management Practice
By C.R. Donough, C. Witt, and T.H. Fairhurst

By comparison with the other major vegetable oil crops, oil palm occupies a small area but
contributes about one-third of the global vegetable oil supply. Production has increased
exponentially in the last 30 years, mainly through an expansion of the area planted. Mean-
while average yields have remained far below the economic yield potential. IPNI and its
partners have developed a management concept to close existing yield gaps through best
management practices (BMPs).

P
lanted area and production of palm oil have increased are clearly more rapid and larger than returns on the develop-
exponentially in Southeast Asia since the 1970s (Figure ment of new plantings for these reasons: 1) production starts
1). Indonesia and Malaysia, the largest producers, ac- to increase as soon as agronomic constraints are removed, and
count for a combined share of 85% of global palm oil production. 2) yield intensification does not require capital outlay on new
The potential productivity of oil palm is several times greater plantings and plantation infrastructure.
than that of other oil producing crops so that, provided the crop In addition, and provided BMPs are used, increasing yield
is managed properly, much less land is required to produce on existing plantings has environmental benefits because
a quantum of vegetable oil compared with other vegetable oil production is increased while sparing wilderness land from
crops. agricultural development. A further impetus for yield inten-
Potential oil yield of oil palms planted on a commercial sification is the dwindling availability of suitable land for
scale has been estimated at 10 to 11 t/ha (Breure, 2003). The further expansion of oil palm plantings. With controls on land
largest reported oil yield at an estate scale (c. 2,000 ha) in development tightening in Southeast Asia, future expansion is
Malaysia was more than 8 t/ha and leading plantation groups likely to focus on degraded land where development costs must
in Indonesia and Malaysia have achieved average oil yields allow for the amelioration of low fertility status soils.
of 6 t/ha at a larger scale (Donough et al., 2006). At an oil For oil palm plantations, inputs are usually both available
extraction rate of 22%, this would be equivalent to a bunch and affordable and estates obtain seed with high yield potential
yield of about 27 t/ha. However, average bunch yield, even in from certified seed producers. Thus, the key to improved yields
the favorable environment in Southeast Asia, remains at less is in better agronomic management and estate organization and
than 20 t/ha (Figure 1). planning. Yield improvement efforts in existing plantations
must focus on identifying and rectifying management practices
40 8 150 that contribute to the emergence of a gap between the maximum
Yield
Area 7 economic yield and actual yield (Figure 2).
125
30 Production 6
BMPs are well established and described. For example,
Fruit yield, t/ha

note the series of handbooks and pocket guides published

Area, M ha

Production, M t

100
5
20 4 75
3
50
10 2
25
1
0 0 0
1960 1970 1980 1990 2000
Year
Figure 1. Yield, area, and production of palm oil fruits from 1961
to 2006. Source: FAOSTAT 2008.
Rationale for Yield Intensification
Better Crops/Vol. 93 (2009, No. 1)

Oil palm responds rapidly to improvements in agronomic

management with short term increases in bunch weight and
longer term increases in bunch number both contributing to frond
increased yield. There is a time lag of 3 to 4 years (i.e., the
time interval between floral initiation and the production of
a ripe bunch) between the removal of agronomic constraints
and full impact on yield. For producers, the financial returns
from investments in yield intensification in existing plantations Figure 2. Management factors contributing to the gap between
actual and maximum economic yield in existing oil palm
Abbreviations and notes for this article: M t = million metric tons.
plantations (after Fairhurst et al., 2006).
12
by IPNI, available at: >www.ipni.net/seasia<. However, New Guinea (Fairhurst et al., 2006). In 2006, IPNI launched a
plantations often lack suitable methods to identify practices new initiative to promote yield intensification based on generic
that could contribute most to yield improvement. The BMP principles of its BMP concept by setting up 30 commercial
concept promoted by IPNI is more than a description of the BMP blocks in partnership with collaborating plantations
actual practices, it is a management tool to collect the neces- in Sumatra (North, South) and Kalimantan (West, Central,
sary evidence on the potential for yield improvement before and East). Collaborating partners include Bakrie Sumatera
time and costly resources are allocated to expand practices Plantations (Site 1), Permata Hijau Group (Site 2), Wilmar
within an estate (Fairhurst et al., 2009). In this article, we International Limited (Sites 3 & 4), Sampoerna Agro Group
integrate yield intensification with environmental goals and (Site 5), and REA Kaltim Plantations (Site 6).
define BMPs as follows: At each site, five pairs of blocks sized at least 25 ha were
BMPs are agronomic methods and techniques found to be selected to represent the estate. One block was designated
the most cost effective and practical means to reduce the gap as the block for BMP implementation; the other became the
between actual and site yield potential and minimize the impact reference (REF) block where current standard practices were
of the production system on the environment by using external maintained. BMPs were implemented based on the following
inputs and production resources efficiently. priorities:
Priority 1 – Crop recovery
BMP as a Management Tool • Adopt a 7-day harvesting interval
IPNI has been instrumental in developing a BMP concept • Maintain clean palm circles and clear access paths
for yield intensification in existing mature plantings (Figure • Construct harvesting platforms and harvesters’
3). In this approach, a set of site-specific BMPs are identi- bridges
fied and implemented in a representative number of full-size • Collect loose fruit in bags
management blocks in each estate to achieve crop management Priority 2 – Canopy management
objectives related to productivity, profitability, sustainability, • Maintain proper pruning
and the environment. Through this process, estates identify • Remove abnormal and diseased palms
better ways to implement BMPs for yield intensification, and • Supply vacant planting points
decisions on larger investments in BMPs are based on practi- Priority 3 – Soil, moisture, and nutrient management
cal, commercial-scale evidence. Performance indicators are • Maintain frond placement in inter-rows and between
selected to describe the complete impact of a combination of palms
BMPs on all four crop management objectives while adhering • Apply empty fruit bunches
to sustainable development goals. • Maintain fertilizer management to support large, profit-
able yields
• Construct adequate drainage
Projects at Site 5 and Site 6 have now completed at least
one year of BMP implementation, while data for longer time
periods is available from other sites (16 months for Sites 3
and 4; 22 months for Sites 1 and 2). Preliminary results are
calculated on an annual basis (Figure 4).
Prior to project implementation, fruit bunch (FB) yield was
on average 1 t/ha greater in REF compared with candidate BMP
blocks (data not shown). After 12 to 22 months of BMP imple-
mentation, yield was the same or greater in BMP compared with
REF blocks at all six sites (Figure 4a). On an individual block
basis, higher FB yield was recorded post-implementation in 24
Figure 3. Conceptual framework for the evaluation of BMPs in ma- of 25 blocks with available pre-implementation records. There
ture oil palm plantations (after Witt and Donough, 2007). was no difference in yield between BMP and REF blocks at
Site 1 where average yields were greatest amongst all project
sites. It remains to be seen whether yields greater than 30 t/ha
The evaluation of BMPs is implemented by the estate
can be achieved with BMPs at Site 1 during the 4-year evalu-
management staff, and we emphasize the importance of in-
ation period. If not, attainable yield has been reached for the
volving key decision makers and resource persons in the local
current palm stand.
Better Crops/Vol. 93 (2009, No. 1)

management team.
The net added value with BMPs was significant at four
Once a new practice is successfully implemented at larger
out of six sites ranging from US$260 to 680/year (Figure 4b)
scale, it becomes current practice and the cycle of evaluation
based on actual cost and an assumed value of US$115/t fruit
and implementation starts over again.
bunch yield. It is expected that BMP will become profitable at
Evaluation of the BMP Concept Site 6 once the full impact of BMP is expressed in yield and
The BMP concept was first developed and successfully investments in drainage and other practices made in the first
introduced in an oil palm rehabilitation project at PT Asiatic year are recovered in increased productivity.
Persada in Jambi Province in Indonesia in 2001 (Griffiths and The additional cost associated with the implementation
Fairhurst, 2003) before being implemented at larger scale in of BMPs was relatively small, ranging from US$15 to 30/ha
several other estates of CTP Holdings in Indonesia and Papua after 10 months (sites A, B). Yield advantages with BMP were
13
 Fruit bunch yield, t/ha 40 800 The project will continue until all sites have completed a

Added value with BMP,

a) BMP b) 4-year cycle of yield improvement. Now that all BMP blocks
ns
** REF ns
30 600 have entered ‘maintenance’ mode, cost differences between
**
the BMP and REF blocks will decrease, while the effects of

US$/ha
*
400
20 * other non-harvesting BMPs such as nutrient management are
200 expected to provide further yield improvements compared to
10 standard practices.
0
0 Conclusions
1 2 3 4 5 6 1 2 3 4 5 6
Encouraging yield improvements achieved through the
3.0 25
Bunch number, 1,000/ha

c) ns d) ns implementation of BMPs at sites broadly representative of

ns **
2.5
Bunch weight, 20 ** the oil palm industry underline the general applicability of
2.0 ** kg/bunch **
the BMP concept. Clearly, a short harvesting interval and full
ns 15 ns
1.5 ns *** crop recovery is a pre-requisite for closing current yield gaps
*
1.0
10 at project sites. The next step in the yield intensification pro-
5
cess will require a thorough analysis of the data at each site to
0.5
determine the site-specific requirements for wider implemen-
0.0 0 tation of the selected BMPs. The BMP concept is consistent
1 2 3 4 5 6 1 2 3 4 5 6
with Principle 4 (best practices) and Principle 8 (continuous
18 improvement) of the Principles and Criteria for sustainable
Harvester productivity,
Harvest interval, days

e) 2.0 f)ns
16 *** *** palm oil production of the Roundtable on Sustainable Palm
*** ns
14 *** *** ** ns
ns Oil (RSPO, >www.rspo.org<). It should be noted that the
t/man-day

12 1.5 **
10 *** success of a BMP project hinges on the commitment from
8 1.0 senior management to provide direction as well as sufficient
6 budget and resources, and from local estate management to
4 0.5
2
implement the BMP(s) rigorously and on time. BC
0
1 2 3 4 5 6
0.0
1 2 3 4 5 6 Author Information
Site Site Mr. Donough is Consulting Agronomist, IPNI Southeast Asia
Program; e-mail: chrisrd@pd.jaring.my. Dr. Witt is Director, IPNI
Figure 4. Performance of BMP in comparison to reference blocks Southeast Asia Program; e-mail: cwitt@ipni.net. Dr. Fairhurst is
(REF) at six project sites in Indonesia. Data represent Tropical Agriculturalist; e-mail: thfairhurst@gmail.com.
a 12-month average, but the period of measurements
ranged from 12 to 22 months. Acknowledgments
All data presented were collected by staff of collaborating plantations.
generally due to improvements in both bunch number and We thank the management teams at the estates for their cooperation,
bunch weight (Figure 4c and 4d). Yield improvement in support, and dedication to the project. We are grateful for the interest
the first year of BMP implementation was largely attributed to and commitment of the Senior Management of collaborating com-
improved crop recovery following the implementation of 7-day panies. We gratefully acknowledge funding contributions by IPNI,
harvesting intervals, i.e., every palm is visited by harvesters Canpotex International Pte Ltd, and K+S KALI GmbH.
once each week. The average harvesting interval in REF blocks
was 12 days (Figure 4e). Because of the short harvest intervals References
Breure, C.J. 2003. In Oil palm – Management for large and sustainable yields
with less bunches to harvest per round, the daily area cov- (Fairhurst & Härdter, eds). PPI/PPIC-IPI, Singapore, p. 59-98.
ered by harvesters was 15 to 35% greater within BMP blocks Donough, C.R. and C. Witt. 2007. Poster presented at RT5 Conference, November
compared to the standard practice in REF blocks. However, 2007, Kuala Lumpur.
because of the greater yield within BMP blocks, average har- Donough, C.R., C. Witt, T. Fairhurst, W. Griffiths, and A.G. Kerstan. 2006.
vester productivity based on the weight of the harvested fruit Proceedings of 5th International Planters Conference 2006, Incorporated
Society of Planters, Kuala Lumpur.
bunches per man-day was only 4 to 15% lower with BMP than Fairhurst, T.H, W. Griffiths, C.R. Donough, and C. Witt, 2009. Best Management
REF (Figure 4f) while the number of harvested bunches was Practice in Oil Palm - A Practical Guide to Ecological Yield Intensification.
10 to 20% lower in BMP (data not shown). As harvesters are IPNI, Malaysia (in press).
paid based on the number of bunches harvested, productivity Fairhurst, T.H., W. Griffiths, and A.G. Kerstan. 2006. Proceedings of 2006
International Oil Palm Conference. Optimum use of resources: Challenges
Better Crops/Vol. 93 (2009, No. 1)

targets and payments will need to be reviewed so that harvest-

and opportunities for sustainable oil palm development. Indonesian Oil
ers benefit from the higher yield under BMPs. More harvesters Palm Research Institute (IOPRI). p. 1-16.
are needed when following the BMP scheme, but the increased Griffiths, W. and T. Fairhurst. 2003. Better Crops International 17(1):16-19.
demand is not in direct proportion to the increased frequency Witt, C. and C.R. Donough. 2007. Poster presented at RT5 Conference, November
of harvesting because each harvester covers a larger area per 2007, Kuala Lumpur, Malaysia.
day when shorter harvesting intervals are maintained.

14
View publication stats