Yield Gap Analysis in Oil Palm Production

Systems in Ghana

Tiemen Rhebergen

i
Thesis committee

Promoter
Prof. Dr K.E. Giller
Professor at Plant Production Systems
Wageningen University & Research

Co-promotors
Dr S. Zingore
Director, East and Southern Africa Program, International Plant Nutrition Institute
(IPNI)
Nairobi, Kenya

Dr A. Whitbread
Research Program Director for Innovation Systems for the Drylands (ISD),
International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)
Telangana, India

Other members
Prof. Dr E. Hoffland, Wageningen University & Research
Prof. Dr A. Gransee, K+S KALI GmbH, Kassel, Germany
Dr P.J.A. van Asten, Olam International Ltd., Kampala, Uganda
Dr J.P. Caliman, SMART Research Institute (SMARTRI), Pekanbaru, Indonesia

This research was conducted under the auspices of the C.T. de Wit Graduate School for
Production Ecology and Resource Conservation (PE&RC).

ii
Yield Gap Analysis in Oil Palm Production
Systems in Ghana

Tiemen Rhebergen

Thesis
submitted in fulfilment of the requirements for the degree of doctor
at Wageningen University
by the authority of the Rector Magnificus,
Prof. Dr A.P.J. Mol,
in the presence of the
Thesis Committee appointed by the Academic Board
to be defended in public
on Wednesday 12 June 2019
at 1:30 p.m. in the Aula.

iii
Tiemen Rhebergen
Yield Gap Analysis in Oil Palm Production Systems in Ghana
vii + 213 pages.

PhD thesis, Wageningen University, Wageningen, The Netherlands (2019)

With references, with summaries in English and Dutch

ISBN: 978-94-6343-976-3
DOI: https://doi.org/10.18174/477061

iv
 ABSTRACT
Palm oil is the world’s most important edible oil crop, and in the Ghanaian economy, oil palm (Eleais
guineensis) is the next most important perennial crop after cocoa. Since the early 2000s, the area under
oil palm in Ghana expanded but average fruit bunch (FB) yields remained low at ~7 t ha-1 yr-1 in 2014,
resulting in large yield gaps. Despite the pressing need to increase palm oil production and improve
yields, knowledge of the underlying causes of poor yields in Ghana is lacking. The objective of this
thesis was to analyse the factors that currently limit oil palm production on plantations and smallholder
farms in Ghana and to assess opportunities to increase yields with improved agronomic practices. Using
land suitability evaluation methods and yield gap analysis, a framework for yield intensification with
‘Best Management Practices’ (BMP) on existing mature oil palm plantings was developed. Data was
acquired from online databases, farm surveys, a three-year trial established on three plantations and
twenty smallholder farms across Ghana, and an irrigation and fertilizer experiment established on a
large-scale oil palm plantation in Western region, Ghana. The results showed that the oil palm industry
in Ghana is limited to expand large-scale plantations due to the lack of suitable and available large and
contiguous tracts of lands. A feasible strategy for oil palm expansion can be sought through smallholder
production, which can make use of smaller parcels of land. However, yield gaps in oil palm production
were large and ubiquitous in Ghana. Using a boundary line analysis, the water-limited yield (Yw) over
a planting cycle was estimated at 21 t ha-1 yr-1 FB, with yield gaps of 15.4 t ha-1 yr-1 FB at smallholder
farms and 9.8 t ha-1 yr-1 FB at plantations. Poor management practices, including incomplete crop
recovery (i.e. harvesting all suitable crop) and inadequate agronomic management were the main factors
contributing to these yield gaps. Productivity losses were exacerbated by low oil extraction rates (OER)
by small-scale processors of 12%, compared with 21% by large-scale processors, resulting in respective
crude palm (CPO) oil losses exceeding 5 and 3 t ha-1 yr-1 in the crop plateau yield phase. Reducing yield
gaps with appropriate BMPs can therefore make a significant contribution to closing the supply gap for
palm oil in Ghana. The implementation of BMPs at plantations and smallholders was evaluated amongst
those that increased yield in the short-term term (≤ 1 year) with ‘yield taking’ (improved crop recovery),
and in the long-term (>1 year) with ‘yield making’ (better agronomy) practices. Compared with current
management practices, average yields and yield gains with BMP were larger, and averaged 17.9 t ha-1
at plantations and 17.6 t ha-1 at smallholders. About 2.1 t ha-1 (+19%) and 4.7 t ha-1 (+89%) of the yield
increase was attributed to yield taking and 4.7 t ha-1 (+36%) and 7.6 t ha-1 (+76%) with yield making at
plantations and smallholders respectively. Important crop recovery activities included more frequent
harvesting events and improved field access (roads, paths, weeded circles) to increase harvester
efficiency and productivity, whilst analysis of fertilizer usage and leaf nutrient concentrations suggest
a more balanced approach to nutrient management could contribute considerably to yield making,
particularly at smallholder farms. The irrigation × fertilizer trial demonstrated the potential to obtain
even larger yields with adequate water and nutrient management. Average yields of 32.6 t ha-1 FB were
achieved with irrigation and fertilizer, which was 4.7 t ha-1 greater than the control (27.9 t ha-1) and
4.1 t ha-1 greater than with irrigation alone (28.5 t ha-1). Fertilizer was therefore essential for a maximum
response to irrigation. The results of this thesis demonstrate the potential to increase production in
Ghana by improving current management practices with BMPs. Particularly, large increases in
production can be sought by improving management practices at smallholder farms which consist of
the majority of growers in Ghana, but require viable integration into the oil palm supply chain to be
successful. Increasing average attainable yields to 21.0 t ha-1 FB has the potential to increase national
FB production almost three-fold from 2.5 Mt to 6.9 Mt (1.4 Mt CPO at 21% OER, worth about 1 billion
US$ at US$750 t-1 CPO-1) and reduce oil palm expansion with almost 600,000 ha of land. Yield
intensification can be achieved without the typical capital expenditure required for new plantings (e.g.
road infrastructure, planting cost), and financial returns from investments in yield intensification accrue

v
more rapidly because production starts to increase as soon as agronomic constraints are removed.
Rehabilitation of existing mature oil palm plantings may therefore be an important policy for
sustainable oil palm development in Ghana and West Africa.
Keywords: Yield intensification, crude palm oil, land expansion, land sparing, irrigation, fertilizer,
agronomic management, yield taking, yield making, West Africa.

vi
 TABLE OF CONTENTS
CHAPTER 1 - General introduction .............................................................................. 1
CHAPTER 2 - Climate, soil and land-use based land suitability evaluation for oil
palm production in Ghana ................................................................... 11
CHAPTER 3 - Yield gap analysis and entry points for improving productivity on
large oil palm plantations and smallholder farms in Ghana ................ 33
CHAPTER 4 - Closing yield gaps in oil palm production systems in Ghana through
Best Management Practices ................................................................. 59
CHAPTER 5 - The influence of water and nutrient management on oil palm yield
trends on a large-scale plantation in Ghana ......................................... 89
CHAPTER 6 - General discussion ............................................................................. 115
REFERENCES .......................................................................................................... 135
APPENDICES ........................................................................................................... 157
SUMMARY / SAMENVATTING ............................................................................ 197
LIST OF PUBLICATIONS ....................................................................................... 201
CURRICULUM VITAE ............................................................................................ 207
ACKNOWLEDGEMENTS ....................................................................................... 209
PE&RC Training and Education Certificate .............................................................. 211
FUNDING.................................................................................................................. 213

vii
 CHAPTER 1

General introduction
CHAPTER 1

1.1 Background and problem statement

Although the Gulf of Guinea, West Africa, is believed to be the centre of origin of the
oil palm (Elaeis guineensis Jacq.) (Simmonds, 1976; Zeven, 1964), current average fruit
bunch (FB) yields in the region (3.9 t ha-1 FB) are far below those achieved in the major
producing countries in Southeast Asia (18.9 t ha-1 FB) and Latin America
(16.5 t ha-1 FB) where climatic conditions are more favourable (FAO, 2017;
Paramananthan, 2003) (Fig. 1.1).
In West Africa, the oil palm’s natural habitat largely consists of swampy locations
which are unfavourable for most dicotyledonous trees. At sites usually too wet for the
establishment of large forest trees, oil palm will grow for some months in flooded
conditions, including the edges of fresh water swamps and along river banks
(Simmonds, 1976; Zeven, 1972).
At present, oil palm is grown commercially in more than 40 countries, ranging from
19°N in the Dominican Republic to 15°S in Brazil (FAO, 2017; Lim et al., 2011).
However, >90% of the area under oil palm cultivation is in 20 countries within 10°N
and S of the Equator (Lim et al., 2011; Paramananthan, 2003). In Africa, oil palm is
found between 16°N in Senegal and 15°S in Angola, and eastwards to Zanzibar and
parts of Madagascar where suitable microclimates occur (Gilbert, 2013) (Fig. 1.1).
Today, oil palm accounts for nearly 30% of global vegetable oil production and is an
important driver in the economic development of many tropical countries (Corley,
2009; Griffiths et al., 2002; Hansen et al., 2015). It is a very efficient crop in terms of
input utilization and produces the highest oil yield compared with other oil producing
crops (de Vries et al., 2010). Driven by an increasing demand for edible oil (Corley,
2009), oil palm has become one of the world’s most rapidly expanding equatorial crops
with a global area expansion from 3.6 million ha in 1961 to 18.7 million ha in 2014
(FAO, 2017; Koh and Wilcove, 2008). Southeast Asia is currently the largest producer
of palm oil, with Indonesia and Malaysia accounting for approximately 80% of global
oil palm production (FAO, 2017). However, with a continued increasing demand in
palm oil in the next decades (Corley, 2009) and limited availability of land in Southeast
Asia, most expansion of the industry is expected in Central and South America and sub-
Saharan Africa (Laurance et al., 2014). Current production increases are therefore
sought mainly through area expansion, whilst average yields remained far below its
economic potential (Donough et al., 2009). Over the next 50 years, rapid and
widespread agricultural expansion is expected to pose a serious threat to natural
ecosystems worldwide (Tilman et al., 2001).

2
 GENERAL INTRODUCTION

Fig. 1.1. Worldwide oil palm cultivation showing a) production (tonnes), b) area (ha) and c) fruit bunch
yield (t ha-1) per country in 2014 (FAO, 2017).

3
CHAPTER 1

In West Africa, the expansion of the large-scale industrial sector has regained interest
from foreign and domestic investors over the past decade (Carrere, 2010). In Ghana, for
example, oil palm evolved from its presence in swidden systems into an agricultural
crop, with the first plantations established in the nineteenth century (Gilbert, 2013). The
plantation system, however, failed to gain significant hold of the industry because
complex land tenure arrangements prevented the development of large-scale
agricultural production (Gyasi, 1996). As a result, the oil palm sector in Ghana is
dominated by independent smallholder farmers, who have been the main driver of
expansion over the past 50 years. Currently, however, large-scale plantations are
responsible for the largest expansion in the oil palm sector (Ofosu-Budu and Sarpong,
2013).
Whilst some area expansion, that meets the criteria for sustainable oil palm
development, may yet be possible, yield intensification to close ‘yield gaps’ on land
already planted to oil palm may be an important aspect in developing policy on
sustainable oil palm development in Ghana. Average yields obtained at plantations
range between 10–13 t ha-1 FB, while smallholder farmers only achieve yields around
3 t ha-1 FB (Ofosu-Budu and Sarpong, 2013). With an estimated average yield potential
ranging from 20–25 t ha-1 FB for sites suitable to grow oil palm (Fairhurst, personal
communication), large yield gaps within Ghana, but also between production systems
occur.
To avoid the negative environmental impacts associated with agricultural expansion
such as forest destruction and loss of biodiversity (Griffiths and Fairhurst, 2003;
Griffiths et al., 2002), land sparing through ecological intensification of cropping
systems instead of land expansion is commonly proposed to maintain productivity
(Phalan et al., 2011; Soliman et al., 2016). Ecological intensification seeks to increase
yield per unit of land (e.g. t ha-1 FB) and approach the “attainable yield” of farming
systems whilst meeting acceptable standards of environmental stewardship (Fairhurst
and Griffiths, 2014). It is therefore important to understand which factors explain
productivity first. Yield Gap Analysis (YGA) has been commonly used as an approach
to assess possibilities for improving land productivity (Woittiez et al., 2017).
1.2 Yield gap analysis
Yield Gap Analysis (YGA) of individual crops have been used to analyse opportunities
for increasing crop production at local to global scales, thus providing information
crucial to food security (Guilpart et al., 2017). YGA looks to which factors are
responsible for yield gaps, and how they can be improved upon by identifying and then
correcting management practices. YGA thus provides a systematic process to assess
opportunities and identify entry points to increase yields. In production ecology, four

4
 GENERAL INTRODUCTION

production levels are distinguished: i) potential yield (Yp) which is determined by

growth-defining factors (radiation, temperature, ambient CO2 concentration, and
cultivar features), ii) water-limited (Yw) and iii) nutrient-limited yield (Yn) where
growth is limited by water and/or nutrients respectively, iv) and actual yield (Ya), which
is determined by yield-reducing factors (weeds, pest & diseases) (van Ittersum et al.,
2013; van Ittersum and Rabbinge, 1997) (Fig. 1.2). In YGA, two major steps can be
identified. The first is to quantify the yield gap (Yg), which is the difference between
Yp for irrigated crops, or Yw for rainfed crops, and the actual yield (Ya). The second is
to identify the underlying causes of the yield gap, in order to identify remedial measures
with management interventions (Lobell, 2013).

Fig. 1.2. Different production levels and the contributing factors (after van Ittersum et al., 2013).

Most yield gap studies have focussed on annual crops such as wheat (Aggarwal and
Kalra, 1994; Anderson, 2010; Bell et al., 1995), cassava (Fermont et al., 2009), rice
(Laborte et al., 2012; Yang et al., 2008), and cereals in general (Neumann et al., 2010),
whilst little focus has been on perennial crops (Hoffmann et al., 2017; Woittiez et al.,
2017). In oil palm, several attempts have been made to quantify and explain yield gaps.
At plantation sites located in Malaysia and Indonesia, yield gaps ranged 5–7 t ha-1 FB
(Hoffmann et al., 2017), whilst management practices such as fertilizer dosage, length
of harvesting intervals and plant mortality explained yield gaps at independent

5
CHAPTER 1

(13.0 t ha-1 FB) and supported (9.2 t ha-1 FB) smallholder farms in eastern Sumatra,
Indonesia (Euler et al., 2016).
However, because perennial crops, like oil palm, are structurally different than annual
crops, a slightly different approach to YGA is required. For example, in annual crops,
growers can take advantage of new seeds with each growing season, whilst for perennial
crops the yield potential is fixed at the nursery and planting stage for each planting cycle
over at least 25 years (Woittiez et al., 2017). In addition, like other perennial crops (e.g.
rubber, cocoa and coffee) there is a 35–40 month time-lag in oil palm between the
occurrence of abiotic and biotic stress events and their impact on yield (Breure, 2003;
Fairhurst and Griffiths, 2014), which complicates separating and quantifying the effects
of individual factors (Woittiez et al., 2017). For example, a dry period at any time during
bunch development, as well as reduced fertilizer applications may both drastically
reduce yield. These time-lagged effects related to the complex physiology of oil palm
must be taken into account when assessing yield gaps (Hoffmann et al., 2017).
Moreover, infrequent, incomplete or incorrect harvesting practices (i.e. harvesting
unripe or overripe bunches) may directly reduce actual yields if logistical operations are
not carried out adequately (Donough et al., 2010; Fairhurst and Griffiths, 2014). These
factors, combined with differing scales of assessment (from smallholder farms of
<10 ha to large-scale oil palm plantations of 10,000 ha) make YGA in oil palm a
challenge (Hoffmann et al., 2017).
When making an analysis of yield gaps and their management, it is therefore important
to distinguish between annual and perennial crops. Opportunities to close yield gaps in
annual crops mainly occur between cropping cycles. In perennial crops, there are
opportunities to close yield gaps within a cropping cycle (with time lags between the
implementation of corrective agronomic interventions and their effect on yield) (Fig.
1.3). There may also be opportunities to close yield gaps between cropping cycles due
to the effect of improved germplasm and better crop establishment and management
during the immature growth phase.

6
 GENERAL INTRODUCTION

Fig. 1.3. Model showing the time-lagged effect of yield gap reduction in oil palm with Best
Management Practices (BMP) (after Fairhurst et al., 2019).

1.3 Reducing yield gaps with Best Management Practices (BMP)

Increases in production are crucial due to multiple interacting challenges such as the
need to adapt to a rapidly changing climate, increased competition for agricultural land
from other land uses, and societal demands for environmental stewardship (Bryan et al.,
2014). Intensification and improvement in resource use efficiency have been identified
as two of the most prospective ways to increase agricultural production (Bryan et al.,
2014). Closing the yield gap between what farmers achieve and those that can be
achieved with best practice and current technology in a given environment is a key
strategy for increasing production on existing cropland (Hochman et al., 2016).
In oil palm, the scope to increase production with sustainable intensification is
considerable (Donough et al., 2010; Hoffmann et al., 2015; Oberthür et al., 2012;
Rhebergen et al., 2014). In Indonesia, yields increased by 6.0 t ha-1 FB in South Sumatra
with “Best Management Practices (BMP)” (Griffiths and Fairhurst, 2003) and by
3.4 t ha-1 (+15%) across six commercial plantations in Indonesia (Donough et al., 2010).
BMPs integrate yield intensification with environmental goals and are defined as
“agronomic methods and techniques found to be the most cost effective and practical
means to reduce the gap between actual and site yield potential and minimize the impact
of the production system on the environment by using external inputs and production
resources efficiently” (Donough et al., 2009).
In the concept, a set of site-specific BMPs are identified and implemented in several
mature oil palm plantings. The BMPs are assessed based on agronomic, economic and

7
CHAPTER 1

environmental performance, in comparison to a parallel set of reference (or control)

plantings (REF) where current standard practices are maintained. Yield gaps between
BMP and REF can therefore be directly linked to differences in crop recovery, canopy
and nutrient management, drainage, and other management practices (Oberthür et al.,
2012). BMP implementation and evaluation is largely a trial-and-evaluation process in
which a set of agronomic practices is identified and assessed, until the most desirable
combination of BMPs for a particular site is developed. This is done in collaboration
and full participation of farmers, plantation managers and field staff. Once the BMPs
are successfully implemented at scale, they become standard practice, and the cycle of
evaluation and implementation starts over (Donough et al., 2010; Donough et al., 2009)
(Fig. 1.4).
Modelled after successful projects in Southeast Asia (Donough et al., 2010; Donough
et al., 2009; Oberthür et al., 2012), the International Plant Nutrition Institute (IPNI) and
Solidaridad’s Sustainable West Africa Palm Oil Programme (SWAPP) initiated a
project in 2013 to assess the potential to increase oil palm yields in plantations and
smallholder farms in Ghana. The main goal of both projects was to enable plantation
managers and smallholder farmers to adopt BMP to close yield gaps at plantations and
smallholder farms. BMPs were applied at three oil palm plantations (Benso Oil Palm
Plantation Ltd. (BOPP), Norpalm Ghana Ltd. and Twifo Oil Palm Plantation Ltd.
(TOPP)) and twenty smallholder sites for a duration of 4 years. This thesis largely uses
data generated from these projects.

Fig. 1.4. Conceptual framework for evaluation of BMPs in mature oil palm standings (after Witt and
Donough, 2007).

8
 GENERAL INTRODUCTION

1.4 Objectives and thesis outline

Oil palm production in Ghana is well below its potential and the factors determining its
efficiency are poorly understood. In this thesis, we developed a framework for YGA in
oil palm in order to diagnose production constraints and assess opportunities for
increasing yields with BMPs. Analysis of the factors that explain low productivity in
Ghana is an important step in identifying entry points for intensifying oil palm
production for various producers. The general hypothesis of this thesis is that Yield Gap
Analysis (YGA) helps to improve fruit bunch yields in Ghana in a systematic way with
Best Management Practices (BMP).
The objectives of this study were therefore first to review and characterize the
environment for growing oil palm in Ghana, and to determine the suitability and
availability of land to grow it with land evaluation methods (Chapter 2). We then
describe the main constraints to oil palm production in Ghana based on land
characteristics that limit productivity, and conclude by providing recommendations for
the sustainable development of the oil palm sector in Ghana.
Second, we describe the oil palm sector in Ghana, and characterize the various oil palm
production systems. We analyse why current productivity is low, and large yield gaps
exist with production ecology concepts. In order to assess entry points for increasing
productivity in oil palm, we apply Yield Gap Analysis (YGA) to quantify and partition
yield gaps between different causes (Chapter 3). We conclude with proposing remedial
measures to increase fruit bunch yields.
Third, we identify and implement Best Management Practices (BMP) to close yield
gaps in existing mature oil palm plantings (Chapter 4). We analyse the impact of “yield
taking” and “yield making” practices on yield components and identify which
(management) factors correspond most with yield. We then present a management
approach for yield gap reduction in Ghana and a framework for scaling BMPs across a
wider scale.
Finally, we investigate the interaction of nutrients and water on the yield performance
of oil palm on a large-scale oil palm plantation in Western region, Ghana (Chapter 5).
We quantify yield gaps caused by water and nutrient constraints and assess the potential
to maximize yields with site-specific management interventions. The relevance of my
findings is discussed in Chapter 6 (Fig. 1.5).
Results of this study are expected to generate knowledge on the causes behind yield
gaps in oil palm production systems in Ghana, and to provide incentives for site-specific
management on closing them. Understanding the mechanisms and variables that
determine yield can help eliminate yield differences in- and between sites and enable

9
CHAPTER 1

us improve crop response to natural variation and variation in management practices.

The ultimate goal of this study is to contribute evidence and incentives for plantation
managers and smallholder farmers across Ghana to adopt and adapt BMP as a
management tool in intensifying oil palm production.

Fig. 1.5. Conceptual overview of the chapters of the thesis.

10
 CHAPTER 2

Climate, soil and land-use based land suitability

evaluation for oil palm production in Ghana

This chapter is published as:

Rhebergen, T., Fairhurst, T., Zingore, S., Fisher, M., Oberthür, T. and Whitbread, A.,
2016. Climate, soil and land-use based land suitability evaluation for oil palm
production in Ghana. Eur. J. Agron., 81: 1–14.
CHAPTER 2

Abstract
In the past decade, oil palm (Elaeis guineensis Jacq.) has become the world’s most important oil crop.
The large demand for palm oil has resulted in a rapid expansion of oil palm cultivation across the globe.
Because of the dwindling availability of land in Southeast Asia, most expansion of the industry is
expected in Central and South America and sub-Saharan Africa, where land with suitable agro-
ecological conditions is available. Using Ghana as a case study, a method for evaluating areas that are
both suitable and available for oil palm production is presented. Our assessment used spatial data and
GIS techniques, and showed that areas with suitable climatic conditions (annual average water deficit
<400 mm) is about 20% greater than was previously identified. The observed differences are the result
of using different methods to determine suitability, and climate change. A major climatic factor limiting
suitability for oil palm production in Ghana is the annual water deficit, with the most suitable areas
located in the rainforest and semi-deciduous forest zones with higher rainfall in southern Ghana.
Opportunities for large-scale oil palm plantation development is limited, however, because of the lack
of availability of large and contiguous tracts of land that are required for commercial plantation oil palm
development. A feasible strategy for oil palm expansion is therefore smallholder production, which can
make use of smaller parcels of land. Alternatively, oil palm production in Ghana can be increased by
yield intensification on land already planted to oil palm. This can also reduce the requirement for further
land clearance for new plantations to meet the growing demand for palm oil. Such assessments will be
essential for guiding government policy makers and investors considering investments in oil palm
development.
Keywords: Ghana, oil palm productivity, water deficit, land suitability.

12
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.1 Introduction
In the past decade, oil palm (Elaeis guineensis Jacq.) has become the world’s most
important oil crop, contributing nearly 30% of the world’s edible vegetable oil
requirements (Corley, 2009; Hansen et al., 2015). The large demand for palm oil has
resulted in a rapid expansion of oil palm cultivation across the globe. Because of the
dwindling availability of suitable land in Southeast Asia, most future expansion is
expected in Central and South America and sub-Saharan Africa (SSA), where large
areas with agricultural potential are available (Laurance et al., 2014).
In West Africa (WA), the consumption of palm oil and derived products is expected to
increase as the population grows. Strong demand for vegetable oil in Africa has resulted
in the expansion of the oil palm sector in WA. This has a large economic impact in
producer countries by providing employment for millions of workers. In addition, palm
oil is now a major source of income and trade along border districts (Ofosu-Budu and
Sarpong, 2013). In the Ghanaian economy for example, oil palm is the second most
important perennial crop after cocoa (Angelucci, 2013; Ofosu-Budu and Sarpong,
2013). The yield of fruit bunches (FB) in Ghana is poor, however, and has decreased
from 6.5 t ha-1 in 1990 to 5.4 t ha-1 in 2012 (FAO, 2014a). By contrast, FB yields in the
major producing countries in Southeast Asia and Latin America, where the climate is
more favourable, are more than three times greater at 18.5–19.0 t ha-1 (FAO, 2014a). In
response to increasing local demand for palm oil and the present requirement for costly
palm oil imports (for example, Ghana imported 74,431 t palm oil at a cost of
US$ 83 million in 2011) (FAO, 2014a), the governments of some West African
countries are encouraging both national and foreign investors to plant more oil palm
(Ofosu-Budu and Sarpong, 2013). Area expansion is therefore proposed to increase
local production and to reduce imports of crude palm oil.
An analysis of the availability of suitable land for oil palm plantations and the obtainable
yields is essential information for government policy makers and investors. Some 45
years ago, van der Vossen (1969) identified areas in Ghana with suitable climatic
conditions for oil palm based on 400 and 250 mm mean annual water deficit isolines.
As in much of SSA, climatic conditions in the oil palm belt in Ghana have changed since
the 1960s (Lemoalle and de Condappa, 2012). Rainfall isohyets and water-deficit
isolines have been displaced to the south by about 1° latitude (i.e. about 110 km),
possibly due to climate change and the impact of deforestation. Furthermore, total
annual rainfall has also become more variable over the past three decades (Manzanas et
al., 2014; Owusu, 2009; Stanturf et al., 2011). Whilst total annual rainfall has remained
similar, rainfall distribution has changed, with more rainfall now occurring in what was
formerly the dry season (Abdul-Aziz et al., 2013; Owusu, 2009). At a local scale, the

13
CHAPTER 2

amount of land with climatic conditions suitable for oil palm production may have either
expanded or diminished due to changes in rainfall patterns (Gilbert, 2013).
In this chapter, we provide an up to date and more accurate assessment of land suitability
and availability for oil palm based upon geographic information systems (GIS) using
spatial data that was not available when van der Vossen (1969) developed his suitability
map. Using Ghana as a case study, we describe the agro-ecological conditions and, using
information from the literature, derive environmental parameters that define whether or
not areas are suitable for oil palm cultivation. We used these parameters together with
the spatial data on climate to develop a land suitability classification and estimate the
amount of suitable land that is available for oil palm development.
We determine the main constraints to oil palm production in Ghana based on an analysis
of land characteristics that limit productivity, and conclude by providing
recommendations for the sustainable development of the oil palm sector in Ghana.
2.2 Material and methods
2.2.1 Agro-ecological conditions in Ghana
Ghana is located in the middle of the West African coast along the Gulf of Guinea
(4.5–11.5°N, 3.5°W–1.3°E), and is bordered by the Ivory Coast to the west, Burkina
Faso to the north, and Togo to the east. The climate of Ghana is strongly influenced by
the movement of the tropical rain belt, known as the inter-tropical convergence zone
(ITCZ). The ITCZ oscillates between the northern and southern tropics in the course of
a year, transporting a dry continental air mass to the north, and a tropical, maritime air
mass to the south (Hayward and Oguntoyinbo, 1987). As a result, northern and southern
regions of Ghana have distinct climates (McSweeney et al., 2010a; McSweeney et al.,
2010b). Six distinct agro-ecological zones have been identified in Ghana, with a gradient
of increased aridity from south to north (Antwi-Agyei et al., 2012). In order of
increasing aridity they are: rainforest, semi-deciduous forest, coastal savanna, forest-
savanna transition, Guinea savanna, and Sudan savanna (Antwi-Agyei et al., 2012)
(Appendix 1). Rainfall decreases from the southwest (>2,000 mm yr-1) to the northeast
(<1,000 mm yr-1) (Environmental Protection Agency (EPA) and Ministry of
Environment, 2011). In the south, the annual mean relative humidity (RH) is ~80%
(except for some days during the ‘Harmattan’ in the dry season), while in the north RH
is lower at ~40% (Oppong-Anane, 2006). Northern Ghana has a single wet season
between May and November, while the southern regions of Ghana have two wet seasons
with long rains usually from March–July and short rains from September–November.
With the exception of the Kwahu Plateau, which runs along the southern edge of the
Volta River Basin, the topography in Ghana is relatively flat and low-lying, with more

14
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

than half the area <150 masl. Because Ghana is in the equatorial belt, mean monthly
temperatures below 25°C are seldom recorded. Mean annual temperature ranges from
26°C in the south to 29°C in the north (FAO, 2005). Minimum temperatures less than
18°C only occur in the highlands above 400 masl (e.g. on the Kwahu Plateau) (van der
Vossen, 1969). The diurnal temperature range is small in the south (5–9°C) due to
maritime influence from the Atlantic, and greater in the north (7–14°C) due to hot and
dusty air brought in from the Sahara desert during the Harmattan (McSweeney et al.,
2010a; McSweeney et al., 2010b). As a result, climatic conditions are hot and seasonally
dry along the southeast coast, hot and humid in the southwest, and hot and dry in the
north.
Little solar radiation data is available for West Africa, and Ghana, because of the lack
of solar radiation recording stations in the region (Stout, 1990). As a surrogate, sunshine
hours are used as an estimate of solar radiation because it is easier to measure and data
are more readily available. While solar radiation and sunshine hours are generally well-
correlated, the effects of the Harmattan and atmospheric pollution under West African
conditions prevents accurate recording of sunshine, even when conditions are cloudless.
As a result, much solar radiation is reflected by the atmosphere, and reaches the earth’s
surface predominantly as diffuse radiation (Corley and Tinker, 2003; Hayward and
Oguntoyinbo, 1987). Nevertheless, taking these factors into account, the total hours of
sunshine per annum in WA is considerably less than in other oil palm growing regions
such as Southeast Asia or Central and South America (Corley and Tinker, 2003; van der
Vossen, 1974). In WA, sunshine hours increase with latitude from the Guinea Coast up
to the Sahel and the margins of the Sahara, albeit not in a regular pattern. In Ghana, for
example, a large land pocket with less sunshine occurs in the west central region
(Hayward and Oguntoyinbo, 1987). In general, there are more sunshine hours during
the dry season and less during the rainy season because of the effect of cloud cover.
However, sunshine hours during the dry season are periodically reduced due to the effect
of dust in the atmosphere during the Harmattan.
Soils in the higher rainfall zones in the south of Ghana are generally strongly weathered
and highly leached with low pH, and poor soil fertility status (Swaine, 1996). Soil type
is strongly related to topography and slope position. The topography in the oil palm belt
is undulating to rolling, with slopes ranging from 2–9°. Some relatively flat areas are
found with moderate slopes (<5°) but rolling to hilly terrain (5–17°) is more common,
with low-lying and poorly drained swamps enclosed by upland areas. In the oil palm
belt, the predominant soils are free-draining Acrisols and Ferralsols (Appendix 2). These
two soil types are often found together, with Acrisols on eroded slopes of low hills,

15
CHAPTER 2

Ferralsols on nearby stable pediments and uplands. Soils in enclosed, low-lying swamps
are commonly Gleysols (FAO, 2014b).
The most common soils in the semi-deciduous forest and parts of the forest-savanna
transition zones of the oil palm belt are Orthic Acrisols. They are deep, well drained
soils with sandy clay loam texture, strongly weathered, leached, and acid (pH <5.1–6.5).
Acrisols are characterized by a distinct argillic B horizon, in which clay particles have
accumulated from the upper soil layer. Acrisols are easily eroded once the forest cover
vegetation has been removed (Adjei-Gyapong and Asiamah, 2002; Buringh, 1979;
FAO, 2014b). Xanthic Ferralsols derived from acid rocks with high quartz content are
more common in the high rainfall forest zone of Southwestern Ghana (Buringh, 1979;
FAO, 2014b; van Wambeke, 1974). Xanthic Ferralsols have a typically yellowish or
pale yellow B horizon, due to low iron content. As a result of their relatively coarse
texture and high rainfall, they are strongly leached, acid and poor in nutrients. Well-
developed soil micro-aggregates reduce the water holding capacity in these soils. Orthic
Acrisols and Xanthic Ferralsols are acid (pH<5.5) low fertility status soils containing
small amounts of plant available nutrients (particularly nitrogen (N), phosphorus (P) and
potassium (K)), with low cation exchange capacity (e.g. Ferralsols <16 cmol kg-1) and
base saturation (e.g. Acrisols <50%). Drainage is related to topography, with poorly
drained, flood prone clay to loamy sand textured Gleysols in valley bottoms, and well
drained Acrisols and Ferralsols respectively on lower and upper slopes (Annan-Afful et
al., 2004; Annan-Afful et al., 2005; Owusu-Bennoah et al., 2000).
2.2.2 Land suitability classification and approach
We used land suitability evaluation methods to identify areas that are both suitable and
available for oil palm production in Ghana. Land suitability is defined as ‘the fitness of
a given type of land for a defined use’ and land suitability evaluation (LSE) is the
‘process of assessment of land performance for a specific purpose’ (FAO, 1976). For
agriculture, LSE evaluates the ability of a piece of land to meet the agro-ecological
requirements of a given crop for maximum yield (Abdel Kawy and Abou El-Magd,
2012). LSE assesses how well all relevant land characteristics such as soil, climate, and
topography (FAO, 1976; FAO, 1985) match the requirements of a particular crop.
For the objectives of this study, computer-assisted overlay mapping is a suitable LSE
method, in which the evaluation criteria are recorded as superimposed layers
(Malczewski, 2004). These layers are then integrated in a single data layer, which can
then be used to produce a map showing land suitability classes. We used this method to
define suitable locations for oil palm in Ghana.

16
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.2.3 Suitability classes

LSE was conducted in three-steps. In the first step, we defined climatically suitable areas
for oil palm based on mean annual water deficit. Water deficit is considered to be the
most important climatic factor affecting oil palm yield (Corley and Tinker, 2003).
Severe water deficits are particularly common in WA because of seasonal droughts. In
extreme areas, such as Pobé in Benin, average annual water deficits can vary between
300–900 mm yr-1, with an average of 550 mm yr-1 (Caliman and Southworth, 1998).
The relationship between water deficit and FB yield is complex. Water stress, as well as
other environmental factors, reduces FB yield via a time-lagged effect on floral
initiation, sex differentiation, and abortion rate (Corley and Tinker, 2003). As a result,
yield reduction may only become evident several months or even years after the
incidence of drought or other stress events (Caliman and Southworth, 1998). With
prolonged water stress, vegetative as well as generative growth is impaired and, in
extreme cases, may lead to palm death (Caliman, 1992). The extents to which water
deficits affect FB yield have been reported in several studies. In WA, Hartley (1988)
found a 10% yield reduction with every 100 mm yr-1 water deficit in the year of harvest.
In contrast, Olivin (1968) and Ochs and Daniel (1976) reported a 10–20% yield
reduction with every 100 mm yr-1 water deficit, depending on the soil’s water holding
capacity. Caliman and Southworth (1998) found an 8–10% yield reduction in the first
year, and a 3–4% yield reduction in the second year with a 100 mm yr-1 water deficit.
Mean annual water deficit is thus a useful parameter in delineating and grouping areas
that are climatically similar in terms of oil palm production (Olivin, 1968; van der
Vossen, 1969). In general, areas with water deficits > 400 mm yr-1 are considered
unsuitable for oil palm production because low yields do not provide an economic return
(Olivin, 1968; van der Vossen, 1969). The critical water deficit, after which oil palm
growth and yield start to be affected is assumed to be 200 mm yr-1 (Corley and Tinker,
2003). Based upon these assumptions, and following the suitability assessment methods
of Olivin (1968) and van der Vossen (1969), we defined four categories of water deficit
as follows:
i. Optimal: areas with a mean annual water deficit <150 mm;
ii. Favourable: areas with a mean annual water deficit <250 mm;
iii. Suitable: areas with a mean annual water deficit <400 mm; and
iv. Unsuitable: areas with a mean annual water deficit >400 mm.
In the second step, we overlaid the areas that were climatically suitable with biophysical
and topographic constraints categorized as either ‘suitable’ or ‘not suitable’ (Table 2.1).
We included solar radiation, temperature, and slope because, after water deficit, they are
the most important factors that determine oil palm yields (Paramananthan, 2003).

17
CHAPTER 2

The amount of solar radiation suitable for oil palm production is not defined precisely,
however, because it is difficult to isolate its effect from other factors affecting
productivity. Oil palm grows best where solar radiation is > 16 MJ m-2, but excessive
amounts of solar radiation can affect stomatal aperture and leaf temperature, limiting
the rate of CO2 absorption and therefore photosynthesis (Corley and Tinker, 2003;
Paramananthan, 2003). Additionally, high levels of solar radiation can cause photo-
damage to plants (Kasahara et al., 2002). Of these constraints, only topography can be
modified by costly management interventions (e.g. installation of individual terraced
platforms on moderate to strongly sloping land (5°–10°) and continuous terraces on
steep terrain (10°–20°).
Table 2.1. Suitability for oil palm production based on climate and topography parameters
(Paramananthan, 2003).
Limitation Units Suitable Unsuitable
Climate
Solar radiation MJ m-2 7–21 <7, >21
Temperature °C 18–37 <18, >37
Topography
Slope ° <20 >20
In the third and final step, we integrated the most current land-use and excluded
protected areas, which include national parks, forest reserves, World Heritage sites, and
Ramsar Wetlands. Protected areas are defined according to IUCN as ‘a clearly defined
geographical space, recognized, dedicated and managed, through legal or other
effective means, to achieve the long term conservation of nature with associated
ecosystem services and cultural values’ (Dudley, 2008). Urban settlements, which were
distinguished from rural areas based on a combination of population counts (persons),
settlement points and the presence of nighttime lights, were also excluded. Urban
settlements are defined as ‘contiguous lighted cells from the nighttime lights or
approximated urban extents based on buffered settlement points for which the total
population is greater than 5,000 persons (Balk et al., 2006; CIESIN et al., 2011).
Because of national differences in distinguishing urban from rural areas, the criteria for
urban settlements are country specific (UN, 2013). In SSA, for example, the urban
population threshold ranges from settlements of 20,000 to as few as 500 inhabitants for
certain areas in South Africa and Zimbabwe (Foote et al., 1993). In Ghana, an urban
area is defined as a settlement with >5,000 inhabitants (Ajaegbu, 1979).

18
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.2.4 Data used

We used historical climate data from the WorldClim database (www.worldclim.org)
(Hijmans et al., 2005), including monthly minimum, maximum and average temperature
(°C) and average monthly rainfall (mm month-1) for the period 1950–2000. The
WorldClim data are generated through interpolation of average monthly climate data
from meteorological stations distributed throughout Ghana. Elevation data was obtained
from the SRTM 90m database (http://srtm.csi.cgiar.org) (Jarvis et al., 2008) and soil
properties, including soil texture and soil type from the ISRIC/WDC-Soils database
(http://soilgrids1km.isric.org) (ISRIC, 2013) and the Harmonized World Soil Database
v1.2 (http://www.fao.org/soils-portal/soil-survey/soil-maps-and-
databases/harmonized-world-soil-database-v12/en/) (FAO et al., 2009). Data on
protected areas and urban settlements were obtained from (http://protectedplanet.net)
(IUCN and UNEP-WCMC, 2014) and
(http://sedac.ciesin.columbia.edu/data/set/grump-v1-urban-extents/data-download)
(CIESIN et al., 2011). The data were compiled in a geographic information system using
ArcMap 10. The climatic, biophysical, topographic, and soil spatial datasets were raster
files on 30- and 3-arcsecond grids (~1 km and 100 m). Data of urban settlements and
protected areas were imported as shapefiles. Data sources are listed in Appendix 3.
2.2.5 Data analysis
We clipped all raster grids and shapefiles to the extent of Ghana’s national boundaries
and derived slope maps from elevation contours (with Z factor = 9.12 × 10-6). We
estimated solar radiation at a 30-arcsecond resolution for each month from latitude and
day of the year (Allen et al., 1998), and then converted solar radiation into water
equivalents to calculate monthly potential evapotranspiration (ETP) following Läderach
et al. (2013):
𝐸𝑇𝑃 = 0.0023 𝑥 𝑅, 𝑥 (𝑇 − 𝑡)1.2 𝑥 (𝑡3 + 17.8) Eq. 2.1
where
ETP = evapotranspiration (mm day-1);
Ra = extraterrestrial solar radiation expressed in water equivalent (mm day-1);
T-t = the monthly mean diurnal temperature range (°C); and
tm = mean air temperature (°C).
We estimated terrestrial solar radiation (Rs) (Hargreaves and Samani, 1982):
𝑅8 = 0.16 𝑥 𝑅, 𝑥 (𝑇 − 𝑡)1.2 Eq. 2.2
where

19
CHAPTER 2

Rs = terrestrial solar radiation (MJ m-2 month-1); and

Ra and T-t are as defined in equation (1).
We converted ETP to mean annual water balances (B) (Surre, 1968):
𝐵 = 𝑆<=8 + 𝑃 − 𝐸𝑇𝑃 Eq. 2.3
where
B = monthly water balance (mm);
Sres = residual soil water from the previous month (mm);
P = monthly rainfall (mm); and
ETP = monthly potential evapotranspiration (mm).
Epebinu and Nwadialo (1993) found a high correlation (multiple R = 0.980) between
available water capacity (AWC) and the soil particle size distribution in Nigeria:
𝐴𝑊𝐶 = 0.93 + 0.54 𝑥 𝑠𝑖𝑙𝑡 + 0.13 𝑥 𝑐𝑙𝑎𝑦 Eq. 2.4
We used equation (4) to estimate AWC for five layers (0–5 cm, 5–15 cm, 15–30 cm,
30–60 cm, and 60–100 cm depth) for each soil texture class. The sum estimates the
maximum amount of available soil water (ASW) in the top 100 cm of soil, which is the
zone exploited by the majority of the roots of oil palm (Nelson et al., 2006). We assumed
that excess water was lost as runoff or drainage to depth. Negative values for monthly
water balance (B) indicate a water deficit, in which case Sres for the following month is
zero. If Sres was greater than the soil’s maximum water storage capacity, it was set to the
maximum. Annual water deficit was calculated as the sum of all negative monthly water
balances for the year.
Rainfall data for the period 2010–2014 were obtained from five meteorological stations.
Three of the stations were located at the commercial plantations Benso Oil Palm
Plantation (BOPP), Twifo Oil Palm Plantation (TOPP) and Norpalm Ghana Ltd. located
in the Western and Central Regions. The other two sites are Bogoso, a smallholder oil
palm project, and the Ghana Oil Palm Research Institute (OPRI) at Kade, respectively
in the Eastern and Western Regions. BOPP and Bogoso are located within the rain forest
zone, while the other sites are found within the semi-deciduous forest zone. The five
production sites cover a wide range of rainfall distribution (mm month-1) within the oil
palm belt. For all sites, we calculated average annual water deficits using the method of
Surre (1968) and evaluated this against the results of our suitability assessment. In the
final analysis, we integrated land use information to delineate areas that were suitable
on land potentially available for oil palm production. We generated maps of land
suitability after excluding land in protected areas and urban settlements.

20
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.3 Results
2.3.1 Areas suitable for oil palm production
We estimated the area suitable (deficit <400 mm yr-1) for oil palm in Ghana to be
7,350,000 ha, or 31% of the total land area (Fig. 2.1). The area unsuitable for oil palm
production, due to poor water availability, is 16,500,000 ha. Optimal areas for oil palm
(water deficit <150 mm yr-1) are in the south of the Western Region and a smaller area
west of Koforidua in the Eastern Region. Optimal areas are estimated at 580,000 ha
(Fig. 2.1).

Fig. 2.1. Map of southern Ghana showing areas climatically suitable for oil palm production (≤400 mm
yr-1 water deficit) in steps of 50 mm water deficit, compared to suitable areas identified by van der
Vossen (1969).

Based on our assessment, the area that has climatic conditions suitable for oil palm
(deficit <400 mm yr-1) is 20% greater than that identified by van der Vossen (1969)
(Table 2.2). The additional area suitable for oil palm is found in the southwestern coast
and east of Lake Volta (Fig 2.1). Climatically favourable areas are 170% greater than
van der Vossen’s (1969) assessment (Table 2.2). The difference is found largely in the
Western and Central Regions and the southern parts of the Ashanti and Eastern Regions
(Fig. 2.1).

21
CHAPTER 2

Table 2.2. Land area falling within different suitability classes compared with the results of van der
Vossen (1969).
Size
Difference
Suitability class Climatic zones Van der Vossen (1969)
,000 ha %
Suitable (AWD <400 mm) 7,354 6,100 1,254 +21
Favourable (AWD <250 mm) 5,049 1,889 3,160 +167
Optimal (AWD <150 mm) 580 Not defined - -
Average water deficits for five production sites in Ghana ranged between 130 mm yr-1
for BOPP and 390 mm yr-1 for Norpalm for the period 2010–2014 (Appendix 4). All
five sites are located within climatically suitable areas, but only BOPP is located in the
area with optimal climate (Fig 2.1).

Fig. 2.2. Map showing areas with temperatures <18°C and slopes >10° within the climatically suitable
area for oil palm production in Ghana. Land with slopes of 10–20° are suitable for oil palm production,
but land with slopes >20° is unsuitable.

22
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

The uplifted edges of the Volta Basin lie along the northeastern edge of the semi-
deciduous forest zone. They are found at elevations >350 masl and have mean monthly
minimum temperatures below 18°C, which is the lower limit for oil palm (Table 2.1).
We removed them from the area classified as climatically suitable (Fig 2.2).
In the areas with suitable climate, estimated solar radiation was within the limits for oil
palm (7–21 MJ m-2) proposed by Paramananthan (2003) (Table 2.1). Slopes >20°, which
occur around the edges of mountainous outcrops, are not suitable for cultivation of oil
palm, and were removed from the climatically suitable areas. Hilly and undulating
terrain with slopes 10–20° occur throughout the oil palm belt, mostly in the favourable
and optimal climate zones.
When we excluded urban settlements and protected areas, the area suitable for oil palm
production reduced with 9% to 6,720,000 ha. The reduction was mostly in the optimal
area (-30%), where there are large areas of forest reserve and urban settlements (Table
2.3).
Few large, contiguous tracts of land remain available for oil palm within this zone. Areas
of land suitable and available for expansion of oil palm in Ghana are shown in Fig 2.3.

Fig. 2.3. Map of southern Ghana showing suitable and available areas with potential for oil palm
expansion, after excluding urban settlements and protected areas.

23
 Table 2.3. Reduction in the areas climatically suitable for oil palm production in Ghana after removing topographic and biophysical constraints, and urban
settlements and protected areas.
CHAPTER 2

Area
Topographic and biophysical
Suitability Climatic constraints Protected Urban Suitable and available Reduction
class zones Slope areas settlements land
Temperature
10–20° >20°
,000 ha %
Suitable 7,354 49 72 20 473 107 6,720 635 -9
Favourable 5,049 158 265 32 961 202 3,786 1,263 -25
Optimal 580 22 72 2.4 152 24 408 172 -30

24
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.4 Discussion and conclusions

2.4.1 Suitability analysis
We used a simple approach to estimate the amount of suitable and available land for oil
palm production in Ghana that can be used to guide investment decisions. The results
provide a basis for discussion of the current large-scale expansion of oil palm in Ghana
and underlines the importance of evaluating the climatic factors that determine the
current and future potential production of oil palm.
Our assessment shows a larger area to be suitable for oil palm production in Ghana than
identified by van der Vossen (1969). The observed differences are the result of different
methods used to determine suitability, but also because of climate change. We were
able to source more detailed climate, soils and topographical data than was available to
van der Vossen (1969). Newer techniques (GIS), methodologies (calculation of ETP
and AWC) and improved knowledge of the effects of environmental parameters on oil
palm production also help to explain the observed differences. To enhance, but also
verify the suitability assessment, crop modelling would be of great benefit to simulate
and analyze crop responses to the various climate conditions.
In addition, meteorological data shows that the climate in the oil palm belt has changed
over the past 40 years. Temperatures increased and there was less rainfall, which was
more variable, throughout Ghana between 1960 and 2000. In our assessment, we used
the WorldClim climate data derived from actual data for 1960–2000 (Hijmans et al.,
2005) to take account of temporal variation.
2.4.2 Climate change and climate predictions in Ghana
The onset and the duration of the dry and rainy seasons and the annual rainfall varies
from year to year and between decades (Stanturf et al., 2011). This is caused by
variations in the movements and intensity of the ITCZ, which affects the timing and
intensity of the West African monsoon. Overall, the climate in the oil palm belt in Ghana
has become drier (i.e. decline in annual rainfall) and more variable (onset and duration
of the dry seasons) since the 1960s, which were wetter than average (Manzanas et al.,
2014; Owusu, 2009; Stanturf et al., 2011). In the late 1970s and early 1980s, annual
rainfall decreased substantially but increased again from 1986–2000. During this
period, there was less rainfall and fewer rain days per month during the short rains and
at the beginning of the long rains, whilst there was more rainfall during the short dry
season (Owusu, 2009). Similarly, rainfall was greater during the main dry season, but
decreased during the short rain season between 1974–2010 in the Ashanti region
(Abdul-Aziz et al., 2013). Expert opinion confirms that rainfall has become more
unpredictable, but previously dry months have become wetter.

25
CHAPTER 2

Projected climate trends predict that mean annual rainfall will decrease further by the
year 2050, while temperatures will increase. Mean annual rainfall is expected to
decrease by ~10% and temperatures increase by 2.5°C and 2.0°C within the semi-
deciduous forest zone, and rainforest zone respectively (Environmental Protection
Agency (EPA) and Ministry of Environment, 2011). By contrast, other workers predict
that southern Ghana will become slightly wetter, especially in the coastal regions
(Läderach et al., 2013). Total annual rainfall is predicted to increase by 20–30 mm yr-1,
while the number of cumulative dry months (rainfall <100 mm month-1) is predicted to
decrease from 4 to 3 months per year by 2050. Mean annual temperatures are expected
to increase by between 1.7–2.1°C in the southern (forest) regions, while mean daily
temperature ranges are predicted to remain constant up until 2050.
The current and predicted changes in climate suggest that growing conditions for oil
palm in Ghana will become more favourable. Although annual rainfall has decreased,
it is distributed more evenly with less dry months (rainfall <100 mm month-1). Oil palm
grows best where rainfall is distributed more evenly throughout the year with no water
deficits (Corley and Tinker, 2003). An increase in temperature, on the other hand, will
increase evapotranspiration and aggravate soil-moisture conditions during periods of
drought. This might adversely affect oil palm production within the semi-deciduous and
forest zones.
Weather data from the five production sites in Ghana showed large variability in annual
rainfall and water deficits over a period of five years (Appendix 4). At the TOPP
plantation, for example, annual rainfall was largest in 2010 with 1,990 mm (annual
water deficit 100 mm yr-1), and least in 2013 with 980 mm (annual water deficit 820
mm yr-1). A more accurate assessment of inter-annual variability on the water balance
therefore needs to be taken into account to understand the dynamics of climate and its
impact on oil palm production. The current WorldClim dataset does not allow for such
an approach.
Evaluation of actual annual water deficits against the suitability assessment only
resulted in an accurate prediction for the BOPP site. To test the validity of the current
methodology, we need comprehensive climate data for sites across Ghana. These
datasets are mostly lacking or difficult to find. This is mostly because of an inadequate
number of meteorological stations for climate data collection. Also, much data that
exists is not digitized and therefore not readily available (UNDP, 2011).

26
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.4.3 Key constraints to the production of oil palm in Ghana.

Although oil palm originated in West Africa, the suboptimal amount and distribution
of rainfall are major factors limiting its production (Corley and Tinker, 2003; Quencez,
1996). Oil palm grows best where rainfall is 2,000–2,500 mm yr-1 and evenly distributed
throughout the year with no months with < 100 mm rainfall (Goh, 2000; Hartley, 1988;
Paramananthan, 2003). In Ghana, the rainforest and semi-deciduous forest zones
therefore have the most suitable conditions for oil palm cultivation (Gyasi, 1992; van
der Vossen, 1969). Based on our suitability assessment, three of the major plantations
are located in the favourable growing areas in the southwest and east of the semi-
deciduous forest zone. Only one major plantation is located in the rainforest zone within
what we have defined as the optimal growing area. The opportunities to expand oil palm
production within the optimal zone are likely limited due to competition from both
large-scale mining and smallholder cocoa production. Smallholder farms and small-
scale processing mills, on the other hand, are scattered throughout the entire oil palm
belt.
In general, water deficit is the main constraint to oil palm production in West Africa
(Caliman, 1992; Danso et al., 2008; Olivin, 1968; van der Vossen, 1969). Olivin (1968)
found that average FB yields in oil palm plantations in West Africa correlated well with
mean annual water deficit. In Ghana, Danso et al. (2008) found an almost linear inverse
relationship between FB yield and water deficit. They concluded that availability of soil
water is critical for oil palm production to be profitable.
Each 100 mm increase of water deficit reduces yields of FB by 10–15% (Corley and
Tinker, 2003; Olivin, 1968). The reduction in yield can be as much as 40–50% if the
palms were subjected to severe water stress in the preceding year (Caliman and
Southworth, 1998). This emphasizes the need to explore the frequency and intensity of
water deficits and the occurrence of drought as prerequisites to planning expansion of
the area of oil palm (Caliman, 1992). Nevertheless, smallholders in Ghana grow oil
palm in areas that we classify as unsuitable with mean annual water deficits up to
600 mm yr-1. In areas with water deficits >400 mm yr-1, no economic oil palm
production can be expected (Olivin, 1968). These areas are considered to be unsuitable
for oil palm in Ghana, except in areas where irrigation might be possible, such as the
area surrounding Lake Volta.
Rainfall in southern Ghana is distributed bimodally, with long rains March–July and
short rains September–November, separated by a short dry spell in August that is quite
consistent across years. The dry season occurs between December and February, with
less than 50 mm rainfall in January and less than 100 mm in both December and
February (Appendix 4) (van der Vossen, 1969). The main period of water deficit, which

27
CHAPTER 2

occurs between November and March when radiation is high and rainfall is low, causes
irregular distribution of palm yield. The extent of water stress within these months is
considered the main yield-limiting factor for oil palm production in Ghana (van der
Vossen, 1974). Across five sites in 2014, 50% of the annual crop from palms aged
7–15 years was harvested during the five-month period of the long rains. Only 23% was
harvested in the dry season (Fig. 2.4). This imposes both logistical and operational
problems, such as labour shortages during the long rains and underutilized mill capacity
in the dry season.

Fig. 2.4. Monthly average fruit bunch yields (t ha-1) for 7-15 year old palms at five Ghanaian oil palm
sites in 2014, and the average yield distribution (%) across all sites for 2014.

Temperature does not appear to limit land suitability for oil palm in Ghana, except at
elevations above 350 masl where monthly minima can drop below 18°C. Cold-tolerant
oil palm hybrids, such as produced by ASD de Costa Rica, are successful in higher,
cooler elevations and provide a possible option to extend the area suitable for oil palm
production (Chapman et al., 2003). Most higher elevations in Ghana are in mountainous
areas with steep (>20°) slopes, which are unsuitable for cultivation.
Solar radiation was within the limits suitable for oil palm. After rainfall, it is considered
the second most important climatic factor for oil palm. Whilst solar radiation is of
course essential for photosynthesis, the oil palm’s requirements, in terms of sunshine
hours or photosynthetic active radiation (PAR) are poorly defined (Corley and Tinker,
2003; Paramananthan, 2003). Physiological models can be used to assess the effect of

28
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

solar radiation on crop production (e.g. Hoffmann et al., 2014). Chan (1991) showed a
decrease in yield of 4.8 t FB ha-1 yr-1 with a decrease in solar radiation of 1 GJ m-2 yr-1
for Malaysia. Lamade et al. (1996) reported an even larger effect of solar radiation on
yield: only 0.73 GJ m-2 yr-1 more solar radiation in North Sumatra than in the Ivory
Coast resulted in 13.6 t FB ha-1 yr-1 more yield. The authors mention that this large
difference is likely partly explained by differences in rainfall and management practices
between the two sites. Annual solar radiation measured at three sites in Ghana,
Indonesia, and Guatemala for 2014 with similar meteorological stations is shown in Fig.
2.5.

Fig. 2.5. Annual variation in monthly averages of daily solar radiation at oil palm plantations in Eastern
region, Ghana, Central Kalimantan, Indonesia, and the Atlantic coast of Guatemala in 2014. Average
annual solar radiation (in GJ m-2 yr-1) is 5.7 for Ghana, 5.9 for Indonesia, and 4.9 for Guatemala.

Average annual solar radiation in Ghana is slightly lower than that received in
Indonesia, with periods of high solar radiation in the dry season. Periods of high
radiation coupled with a suboptimal water availability and distribution in Ghana
undoubtedly lower the yield potential of oil palm compared with other oil palm growing
regions (Hoffmann et al., 2014; van Ittersum et al., 2013). Further research is needed
on the effect of solar radiation and its interaction with water on yield. While the amount
of solar radiation sets the maximum yield potential of oil palm (Lim et al., 2011; van
Ittersum et al., 2013), it is modified by other (climatic) factors such as rainfall and

29
CHAPTER 2

management. These factors need to be taken into account as well when assessing yield
gaps. A useful parameter combining both water availability and sunshine hours is
effective sunshine defined as the number of sunshine hours during periods of moisture
sufficiency. Sparnaaij et al. (1963) found a strong positive and time-lagged correlation
between total annual effective sunshine and annual bunch yield of mature palms
28–30 months later.
Slopes steeper than 10° occur throughout the oil palm belt, mostly in the southwestern
rainforest, which is the optimal production zone. Oil palm is best planted on slopes
<12°, but is sometimes grown on slopes ≤20°. On slopes exceeding 2°, substantial
amounts of water are lost through runoff. This negatively affects the water balance and
surface water flow can also cause severe erosion (Caliman, 1992). Careful management,
including the establishment of legume cover crops and mulching with pruned fronds
and empty fruit bunches is required to avoid erosion, and to conserve water. Other
options include the installation of terraced planting and silt pits, which increase water
infiltration and reduce soil erosion (Paramananthan, 2003).
Oil palm can be cultivated successfully on a wide range of soils, provided the soils are
deep enough (>50 cm soil) and properly drained (Paramananthan, 2003;
Paramananthan, 2011). Soil physical properties are more important than chemical
properties because they are not easily changed by management and because they control
the soil’s ability to supply water to the crop. Where rainfall distribution is not uniform,
soils with high water-holding capacity will be less affected by water deficits than those
with low water holding capacity. In Benin, with rainfall less than 1,200 mm and
4–5 dry months (Hartley, 1977), oil palm production was viable because of the soils’
high ASW content. In areas with only 100 mm water deficit, yields on a soil with low
ASW can be half those on a soil with high ASW (Olivin, 1968). Soils that have high
ASW content are therefore desirable to cope with West Africa’s climates, but are
limited within the oil palm belt of Ghana.
Poor fertility soils can be managed by applying mineral fertilizers and returning crop
residues. We did not consider drainage criteria in the assessment, because most soils in
the oil palm belt are well-drained. Landscapes in Ghana’s oil palm belt are undulating
to rolling terrain, with poorly-drained swamps in enclosed valleys that are difficult to
drain. These pockets of poorly drained low-lying areas were not identified in the GIS
assessment because the resolution of the raster grids was not fine enough to allow for
an accurate analysis. The valleys have fertile soils with high ASW (Annan-Afful et al.,
2005), and therefore have a high yield potential but often need costly drainage to
prevent flooding and water-logging in periods of high rainfall (Rhebergen et al., 2014).

30
 CLIMATE, SOIL AND LAND-USE BASED LAND SUITABILITY EVALUATION FOR OIL PALM
PRODUCTION IN GHANA

2.4.4 Opportunities to increase oil palm production in Ghana

The annual deficit in crude palm oil (CPO) will likely increase from 35,000 t to
127,000 t by 2024 (MASDAR, 2011) if current production levels are maintained. To
meet the projected oil demand in Ghana, suitability mapping provides opportunities for
area expansion into the most suitable lands for higher yields. In Ghana, land that is
suitable for oil palm is found in the rainforest and semi-deciduous forest zones in
southern Ghana, known as the ‘oil-palm belt’ (Gyasi, 1992; Quencez, 1996; van der
Vossen, 1969). Suitable and available land is estimated at 6,720,000 ha. Although it is
difficult to obtain accurate data, the total area currently under oil palm in Ghana is
estimated at 330,000 ha. This is about 5% of the land area suitable and available to grow
oil palm. About 40,000 ha (~12%) of this is managed in plantations, including
smallholder and out-grower schemes linked to them. About 140,000 ha (~42%) is
cultivated by independent medium- and small-scale farmers, and the remaining 150,000
ha (~45%) is found in wild oil palm groves (MASDAR, 2011; Ofosu-Budu and
Sarpong, 2013). The minimum size for an oil palm plantation equipped with a 45-tph
mill and an average annual yield of 18 t FB ha-1 is about 10,000 ha. The establishment
of such a plantation is thus not only dependent on land suitability and availability, but
also on milling technology. Whilst area expansion is possible for oil palm plantations,
large continuous tracts of areas of suitable and available land are limited, particularly
in the optimal growing zone. This land is possibly even more fragmented by other types
of land-use that were not taken into account in the assessment. This includes for
example land under cocoa and rubber production, but also annual crops, mining, high
conservation value (HCV) areas (RSPO, 2013), and fallow land that is part of slash and
burn agriculture. Moreover, land acquisition is further complicated by complex land
tenure arrangements that prevail in southern Ghana that make it difficult for investors
to acquire land for the development of large-scale plantations (Ahiable, personal
communication). Such obstacles to land acquisition do not apply to the same extent,
however, to the acquisition of smaller parcels of land (<100 ha) by local people. Due to
the fragmented nature of available and suitable land, and increasingly complex land
tenure with larger parcels of land, a feasible strategy for expansion is smallholder
production. That is provided there are enough and efficient milling facilities available
to process the fruit.
Alternatively, production in Ghana can be increased by improving productivity
(Rhebergen et al., 2014). To identify entry points in improving yields, yield gap analysis
(YGA) is a useful tool. YGA partitions yield gaps between different causes, such as
environment and management, thus providing a systematic process to assess
opportunities in increasing yields. Under good climatic conditions in Ghana, the
maximum average attainable bunch yield is estimated at 25 t ha-1 (Rhebergen et al.,

31
CHAPTER 2

2014). With a country average bunch yield of 5.4 t ha-1, current yield gaps are mostly
the result of inadequate crop agronomic management, and poor crop recovery.
Opportunities for increasing production can therefore be sought in improving current
management practices. Yield intensification on land already planted to oil palm may
therefore be an important policy for sustainable oil palm development in Ghana and
West Africa. Adapting ‘Best Management Practices (BMPs)’ to local conditions can
identity the management practices that are responsible for yield gaps (Donough et al.,
2010). Improving agronomic management of existing palm stands shows considerable
scope for yield intensification in Ghana which can alleviate pressure for further land
clearance for new plantations.
2.5 Acknowledgements
We thank the management and staff of Benso Oil Palm Plantation (BOPP), Norpalm
Ghana Ltd., Twifo Oil Palm Plantation (TOPP), Solidaridad West Africa and the Oil
Palm Research Institute (OPRI) for their support and the provision of climate and yield
data. We also thank AgroAmerica, Guatemala and Union Sampoerna Triputra Perdana,
Indonesia for meteorological datasets from their field locations, and Munir Hoffmann,
Rob Verdooren, and Ken Giller for their valuable suggestions and comments. Funding
was provided by Canpotex International Pte Limited and K+S Kali GmbH, who did not
participate in data collection, analysis, and interpretation, nor in the writing and decision
to submit this chapter for publication.

32
 CHAPTER 3

Yield gap analysis and entry points for improving

productivity on large oil palm plantations and smallholder
farms in Ghana

This chapter is published as:

Rhebergen, T., Fairhurst, T., Whitbread, A., Giller, K.E. and Zingore, S., 2018. Yield
gap analysis and entry points for improving productivity on large oil palm plantations
and smallholder farms in Ghana. Agric. Syst., 165: 14–25.
CHAPTER 3

Abstract
Oil palm production must increase in Ghana to meet the increasing demand for palm oil and avoid
costly imports. Although maximum fruit bunch (FB) yields of >20 t ha-1 yr-1 are achievable, average
FB yields in Ghana are only 7 t ha-1 yr-1. Despite the pressing need to increase palm oil production and
improve yields, knowledge of the underlying causes of poor yields in Ghana is lacking. Closing yield
gaps in existing plantings in smallholdings and plantations offers great opportunities to increase oil
production without area expansion, thus sparing land for other uses. This study sought to understand
the magnitude and underlying causes of yield gaps in plantation and smallholder oil palm production
systems in Ghana based on a detailed characterization of management practices and yield measurements
over a two-year period. Using a boundary line analysis, the water-limited yield (Yw) over a planting
cycle was defined as about 21 t ha-1 yr-1 FB, with yield gaps of 15.4 t ha-1 yr-1 FB at smallholder farms
and 9.8 t ha-1 yr-1 FB at plantations. Poor management practices, including incomplete crop recovery
(i.e. harvesting all suitable crop) and inadequate agronomic management were the main factors
contributing to these yield gaps. Productivity losses were further exacerbated by low oil extraction rates
by small-scale processors of 12% as compared to 21% by the large-scale processors. The potential
losses in annual crude palm oil (CPO) during the crop plateau yield phase therefore exceed 5 and
3 t ha-1 yr-1 for small-scale and large-scale production systems respectively. Investment to reduce yield
gaps by appropriate agronomic and yield recovery practices across all production systems, while
improving access of smallholder producers to more efficient oil palm processing facilities, can make a
significant contribution to closing the supply gap for palm oil in Ghana. The impact of such investments
on large-scale plantations could result in a doubling of CPO production. Smallholder farmers could
benefit the most with a fourteen-fold increase in CPO production and economic gains of > 1 billion
US$.
Keywords: Yield gap analysis, crop recovery, boundary line analysis, sustainable intensification, oil
extraction rates.

34
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

3.1 Introduction
The demand for palm oil in West Africa is outstripping supply, with an annual deficit
estimated of >1 million t crude palm oil (CPO) for the Economic Community of West
African States (ECOWAS) in 2013 (FAO, 2017). Ghana had an annual CPO production
shortfall of approximately 106,000 t in 2013. Part of the deficit was compensated by
costly imports (165,000 t CPO, at a cost of US$140 million), whilst approximately
60,000 t CPO was exported (FAO, 2017). Oil palm production in Ghana must therefore
increase to meet the high demand.
There are three main stakeholders in the Ghanaian oil palm industry: (i) large industrial
plantations (≥1,000 ha) with large-scale processing mills (processing capacity
>15 t hr-1 fruit bunches (FB)) (ii) smallholder farms of up to 100 ha and (iii) small-scale
processors using semi-mechanized mills (processing capacities of < 1 t hr-1 FB) (Adjei-
Nsiah et al., 2012a). In this chapter, we define smallholder farmers as growers that
cultivate oil palm on privately owned or rented land. They are not contractually bound
to deliver their crop to a particular mill or association (MASDAR, 2011; RSPO, 2015).
Growth defining and limiting factors (e.g. radiation, planting material, climate and
nutrient supply), as well as growth reducing factors (e.g. pests and diseases) and the
quality of field management all determine the yields that can be achieved at a particular
site (van Ittersum et al., 2013). Despite maximum observed FB yields in individual
fields of 20 t ha-1 yr-1 or more, average FB yields on large plantations are estimated to
be 10–13 t ha-1 yr-1, while smallholder farmers achieve very low average FB yields of
about 3 t ha-1 yr-1 (Ofosu-Budu and Sarpong, 2013). Oil extraction rates (OER) are
lower at small-scale processors that provide services to most smallholder producers;
10–14% as opposed to 19–22% achieved at large-scale mills (Adjei-Nsiah et al., 2012a).
The yield of crude palm oil (CPO) may therefore be an order of magnitude greater in
estates compared with smallholders given the combination of larger fruit bunch yields
and higher oil extraction rates.
In response to the increasing demand for palm oil, programmes supported by the
Government of Ghana during the period 2002–2013 led to rapid expansion in the area
planted with superior tenera (i.e. dura x pisifera (DxP)) oil palm seedlings by
smallholder farmers. The area planted increased by 20,000 ha between 2004 and 2010,
(MASDAR, 2011) and total FB production increased by >110%, from 1,100,000 t in
2002 to 2,326,920 t in 2013. Over the same period, however, average FB yields
stagnated between 5.6–7.3 t ha-1 yr-1 (FAO, 2017). In 2014, average FB yields in Ghana
(7.0 t ha-1 yr-1) were slightly less than the average FB yield for West Africa
(8.2 t ha-1 yr-1), and small compared with FB yields achieved in Southeast Asia
(15.9 t ha-1 yr-1) and Latin America (12.9 t ha-1 yr-1) (FAO, 2017). Whilst several authors

35
CHAPTER 3

have attempted to quantify and explain yield gaps in oil palm (e.g. Corley and Tinker,
2016; Euler et al., 2016; Hoffmann et al., 2017; Woittiez et al., 2017), most have
focused on production systems in Southeast Asia. Despite the pressing need to increase
palm oil production, knowledge of the underlying causes of poor yields in Ghana is
lacking. By closing yield gaps in existing plantings in smallholdings and plantations,
palm oil production could be increased without area expansion thus sparing land for
other uses.
We analysed yield gaps in oil palm production systems in Ghana due to genetic,
environmental (climate and soil), and agronomic management factors. Such analysis
helps to identify opportunities and entry points for yield intensification. Recently, Euler
et al. (2016) and Hoffmann et al. (2014; 2017) applied the crop simulation model
PALMSIM to determine oil palm yield gaps in Southeast Asian production systems.
However, without further development, the PALMSIM model is not applicable to
regions such as Ghana where rainfall deficit regularly limits crop growth and yield. The
size of yield gaps can be estimated by measuring the time-lagged effect of implementing
best management practices (BMPs) that effectively eliminate constraints due to poor
agronomic management (Fairhurst and Griffiths, 2014). In this context, we define
BMPs as agronomic methods and techniques found to be the most cost-effective and
practical means to reduce the gap between actual and maximum economic yield and
minimize the impact of the production system on the environment by using external
inputs and production resources efficiently (Donough et al., 2009).
The BMP approach also provides the means to estimate maximum economic yield
(Ymey) in a particular field and to quantify yield gaps caused by crop losses (Yield Gap
4) and agronomic management (Yield Gap 3) (Fig. 3.2; Fairhurst and Griffiths, 2014).
Yield gap analysis can therefore be used to indicate the aspects of plantation
management with the greatest potential for yield improvement. The specific objectives
of this study were to: (i) describe the various oil palm production systems in Ghana and
their current levels of productivity, (ii) estimate yield gaps on oil palm plantations and
smallholder farms, and (iii) assess the underlying causes of yield gaps and identify
remedial measures.
3.2 Methods
3.2.1 Study area
Farm surveys and trials were carried out at sites selected to represent a range of
environments and production systems in the oil palm belt of southern Ghana. We
selected three large oil palm plantations located in the Western and Central regions
(Benso Oil Palm Plantation (BOPP) (5°06’47.74”N; 1°54’55.15”W), Norpalm Ghana
Ltd. (4°55’29.04”N; 1°53’31.75”W), and Twifo Oil Palm Plantation (TOPP)

36
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

(5°32’03.30”N; 1°31’40.67”W)), and 20 smallholder farms distributed across the

Western (10), Central (3), Eastern (5), and Ashanti (2) Regions (Fig. 3.1).
Rainfall distribution is bimodal in southern Ghana. Mean annual precipitation is
greatest at sites in the southwest (with annual average rainfall of 2,400 mm), and rainfall
decreases gradually towards the north. Mean annual relative humidity (RH) is high
(~80%), and mean monthly temperatures seldom drop below 25°C, with a small diurnal
range of 5–9°C. The topography is predominately undulating (2–9°), with rolling to
hilly terrain with slopes >20° at sites in the southwest. The main soil types in the region
are coarse-textured, strongly weathered and highly leached Acrisols and Ferralsols
(USDA: Ultisols and Oxisols respectively) with low pH and poor soil fertility status
(Buringh, 1979; Swaine, 1996). Four climatic zones (CZs) with varying suitability for
oil palm have been distinguished in Ghana based upon climate and soil data (Chapter
2; van der Vossen, 1969). CZs were defined according to the mean annual water deficit
(mm yr-1), which integrates relevant climate (i.e. rainfall quantity and distribution) and
soil properties (i.e. water holding capacity) in a single parameter that delineates oil palm
areas with similar water-limited yield potential (Olivin, 1968; van der Vossen, 1969).
Areas with a mean annual water deficit <150 mm were designated as optimal CZs,
whereas areas with a mean annual water deficit <250 mm were favourable, <400 mm
suitable, and >400 mm unsuitable (van der Vossen, 1969). One plantation and 7
smallholder sites were located in the optimal CZ, and two plantations and 13
smallholder sites in the favourable CZ (Fig. 3.1).

Fig. 3.1. Map showing location of oil palm plantation and smallholder trial sites and suitability zones
for oil palm cultivation in the southern regions of Ghana.

37
CHAPTER 3

3.2.2 Agronomic trials

Each commercial oil palm plantation consisted of several administrative areas called
‘divisions’ (~1,000 to 2,500 ha), which were subdivided into ‘blocks’ (~10 to 50 ha),
the smallest management unit. Three to five pairs of blocks planted with tenera palms
between 1996 and 2010 and ranging in size from 8.9–41.2 ha were selected in each of
the three plantations.
The following criteria was used to select the 20 smallholder farm sites: (i) tenera palms
≤17 years after planting (ii) farm accessible by road, (iii) farm size ≥3 ha, (iv) triangular
palm layout with palm planting distance 8.5 or 9 m, (v) willingness to maintain farm
records, and (vi) willingness to implement BMPs on the BMP treatment plot. Farm size
ranged from 3.4–292 ha, and fields were planted between 1999 and 2010. Two
accurately measured plots (1–4 ha) were delineated in each farm and BMP and REF
treatments were allocated randomly within each pair of treatment plots.
The paired treatment plots were similar in size, topography, soil type, year of planting,
and planting material and representative of the plantation division or farm. The farmer
or plantation field practices were maintained in one of the treatment plots (REF) to
document current management practices and production levels. Best management
practices were implemented in the other treatment plot (BMP) to assess the potential
for yield improvement. Production constraints related to harvesting practices,
cultivation and field upkeep, and nutrient management were identified during field
inspections in each plantation division or farm and corrective measures were then
implemented in the BMP plots in order to maximize yield. Corrective measures
included; (i) installation of harvest paths and weeded circles to provide unimpeded
access for harvest and palm upkeep, (ii) removal of unproductive fronds with corrective
pruning to provide access for harvesting, (iii) introduction of regular and complete
harvesting cycles at 7–10 day intervals to ensure complete recovery of fruit bunches
and detached fruits, (iv) improved nutrient management and soil conservation by the
application of mineral fertilizers, crop residues (empty fruit bunch (EFB) mulch) and
box-pattern frond stacking, (v) manual (slashing, uprooting) and chemical (glyphosate,
tryclopyr) removal of woody weeds in palm inter-rows and harvest paths to favour
establishment of soft weeds, grasses and legume cover plants, (vi) improvement of
drainage in swampy areas by installing ‘V’ profile field drains, and (vii) regular patrols
to monitor and then control outbreaks of pests and diseases. At all sites, planting density
at establishment was either 143 or 160 palms ha-1, with the exception of one smallholder
site, which was planted at 151 palms ha-1. Plantation blocks were planted with planting
material from the Democratic Republic of Congo (DRC) (n=7), Ghana Sumatra (n=2),
the Oil Palm Research Institute (OPRI) (n=2), and Pobé, Benin (n=1), while smallholder

38
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

farmers obtained planting material mostly from the nearest seedling distribution centres,
such as industrial plantation nurseries (BOPP, GOPDC, Norpalm, TOPP), as well as
OPRI. All trial sites consisted of mature oil palm aged 3–18 years after planting at the
start of the project.
3.2.3 Data collection
Data collection at REF plots took place in 2013 and 2014 to document current
management practices and yields. Besides BMP yield data used to derive the water-
limited yield (Yw), we present only data collected at the REF plots in the results section.
The full results for the BMP plots are reported in Chapter 4.
At the start of the project, a census was conducted to determine the number of
productive palms per ha at each site. Production inputs (labour, fertilizer, empty fruit
bunch mulch, and agro-chemicals including herbicides and insecticides) and outputs
(bunch production, number of bunches, and FB yield and yield components) were
recorded at each maintenance/application or harvest event. Effectiveness of field
management practices such as pruning, weeding, drainage, and presence of cover crops
were assessed by carrying out detailed field assessments periodically at each site. FB
production was measured using a digital scale (smallholder farms) or the mill
weighbridge (plantation treatment plots) and bunch production data was used to
determine actual yield (Ya, t ha-1 yr-1 FB). To estimate Yw, we used FB yield data
recorded over a four-year period from BMP plots.
At each site, datum points for leaf and soil sampling were marked and geo-referenced
following standard procedures for data collection in oil palm (Foster, 2003). In the
plantations, a staggered grid pattern of datum points was used (i.e. every tenth palm in
every tenth row) to give a sample palm density of 1% of the plantation block
(1–2 palms ha-1). By contrast, every fifth palm in every fifth row was selected in
smallholder treatment plots to provide a sampling density of 3–6% at each trial plot
(5–9 palms ha-1). Sampling density was greater in the smaller smallholder treatment
plots in order to produce sufficient leaf sample material for each treatment plot.
Three upper and three lower rank leaflets were sampled from each side of the rachis of
Frond 17 at a point 2/3rds of the distance between the insertion point of the first true
leaflets and the frond tip (Chapman and Gray, 1949). Leaflets from each datum point
were bulked to produce a composite sample for each treatment plot. Sampled leaflets
were cut lengthwise into three equal parts. The middle part was selected as sample
material and the midrib removed from each leaflet. Composite leaf samples were cut
into small pieces and dried in an oven at 65°C for 48 hours. Dry samples were ground
to pass a 20 mm mesh sieve and analysed for N (combustion analyser, Dumas

39
CHAPTER 3

technique), and P, K, Mg, Ca, and B by inductively coupled plasma analyser (ICP) at
Yara Laboratory, UK.
Soil was sampled from beneath the weeded circle and beneath the frond stack to a depth
of 40 cm at each datum point. Soil samples were bulked to form a composite sample
for each zone in each treatment plot. Composite soil samples were air-dried for
3–4 days, and then ground using a pestle and mortar. Samples were passed through a
2 mm sieve to remove stones, gravel and other debris. Composite soil samples were
analysed for pH (water), organic matter (Dumas), total N (Dumas), available P (Olsen),
and exchangeable cations (K+, Ca2+, Mg2+, and Na+) (1 M Ammonium nitrate) at Yara
Laboratory, UK. Leaf nutrient concentrations and soil chemical properties were
compared with critical levels taken from Fairhurst et al. (2004) and Goh et al. (2007)
respectively.
3.2.4 Yield gap analysis
van Ittersum and Rabbinge (1997) reviewed various yield gap analysis studies (i.e. the
difference between potential, water-limited, N-limited and actual yield) on annual crops
such as wheat and rice. Such studies could be used to first quantify the amount of inputs
required to reach a particular yield and then assess whether or not the amount of inputs
used was sufficient. A second step involved the identification of reasons for suboptimal
input use (e.g. lack of knowledge, risk aversion, government policy, poor economic
returns). van Ittersum et al. (2013) used yield gap analysis to produce a global yield gap
atlas (http://www.yieldgap.org) that shows the difference between actual and potential
yield and between actual and water-limited yield for the major cereals and sugarcane in
sub-Saharan Africa.
A different approach to yield gap analysis is required with perennial crops like oil palm.
First, potential yield changes as the leaf canopy develops over the period from the onset
of harvest 2–3 years after planting (YAP) to replanting at 25–30 YAP. Four phases of
growth and production, each with different requirements in terms of agronomic
management, can be distinguished during the lifespan of field planted oil palms (Ng,
1983). Following the immature growth phase (IGP), yield increases rapidly during the
steep ascent yield phase (SAYP) from years 3–7 after planting, before reaching the
plateau yield phase (PYP). The PYP extends from 8–15 YAP before yield starts to
decline in the declining yield phase (DYP), which is largely a result of stand loss and
incomplete crop recovery with older and taller palms (Goh et al., 1994). The DYP
continues until the palm stand is replanted at 25–30 YAP, by which time palms are too
tall for economic harvesting and/or replanted palms will likely provide better economic
returns (Fairhurst and Griffiths, 2014). Second, there is a time lag between the
occurrence of abiotic and biotic stress events and their impact on yield. This is because,

40
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

in the case of oil palm, flowers are produced continuously and there is a time interval
of about 40 months between floral initiation and bunch harvest (Breure, 2003). Third,
whilst moisture stress is a frequent limitation to productivity, irrigation is seldom
practised because of the scarcity of useable water during dry periods and the capital cost
of large scale irrigation systems. Fourth, there may be a significant yield gap due to
poor crop establishment that persists over the lifespan of the planting which cannot be
fully corrected by remedial agronomic interventions. Fifth, crop losses due to
incomplete crop recovery are common in oil palm because maintaining continuous
complete crop recovery is problematic in most locations, particularly due to insufficient
labour in the peak crop period. For these reasons, it is important to apportion yield gaps
between agronomic factors at crop establishment and during the period from harvest to
replanting and logistical problems relating to crop recovery.
We used a modified yield gap model developed by Fairhurst et al. (2006) and Fairhurst
and Griffiths (2014) that partitions the gap between potential yield (Yp), the yield with
no water or nutrient limitations, and actual yield (Ya) into four gaps (Fig. 3.2).

Fig. 3.2. Modified yield gap model for perennial crops used to partition potential yield (Yp) and actual
yield (Ya) into four gaps with the size of each gap depending on site specific factors.

41
CHAPTER 3

In this model, Yield Gap 1 is the difference between Yp and the yield of a well-managed
crop under rainfed conditions (Yw). Yield Gap 2 is the difference between Yw and the
maximum economic yield (Ymey), which is caused by a suboptimal palm stand,
irrespective of good management, proper nutrient and pest management and complete
crop recovery. Yield Gap 3 is the difference between Ymey and the yield limited by past
field, nutrient and pest management (Yam). Yield Gap 4 is the difference between Yam
and the actual yield (Ya) and is explained by incomplete crop recovery. Yield gap
analysis was conducted in three steps:
(i) First, the water-limited yield (Yw) for the study area in Ghana was estimated by
fitting a boundary line through the yield data for BMP treatment plots (Schnug et al.,
1996; Wairegi et al., 2010; Wang et al., 2015). After sorting the independent variable,
i.e. year after planting (YAP) in ascending order, we removed outliers identified by
using statistical methods (e.g. box-plots in SPSS Statistics 24) and empirical knowledge
on oil palm production (e.g. FB yields exceeding 30 t ha-1 yr-1 were removed based on
empirical results of oil palm production in the region). Boundary lines were then fitted
through the selected boundary points, using the model of Fermont et al. (2009):
KL
𝑦J = Eq. 3.1
JM N =OP QRO
where, 𝑦S is the maximum observed water-limited yield, 𝑥 is the YAP and 𝐾 and 𝑅 are
constants. The best boundary line (𝑦J ) was obtained by minimizing the root mean
squared error (RMSE) between the fitted boundary line and the boundary points.
(ii) Second, the actual plantation and smallholder yields (Ya) for two climatic zones
were plotted together with the estimated Yw, and calculated the yield gaps for each site
accordingly. In doing so, it is important to account for yield dynamics with palm age
and production phase (SAYP, PYP, DYP). Hence, yield gaps were first calculated as
the difference between Yw estimated with the boundary line and Ya according to
planting year and site. The average yield gap across the entire productive lifespan for
each production system and climate zone was then determined by taking the mean
across all years after planting. Actual yields (Ya) for each site were taken as the average
for 2013 and 2014, to account for as much variability as possible in climate and
management practices.
(iii) Third, the production systems were characterized in terms of production inputs and
outputs in order to identify the underlying causes of Yield Gaps 2, 3, and 4.

42
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

3.2.5 Data analysis

A nested ANOVA was used to test for significant differences in production inputs and
outputs, as well as management practices and soil/leaf data between the different
production systems. Statistical analysis of data was performed using IBM SPSS
Statistics Version 24.
3.3 Results
3.3.1 Oil palm producers in Ghana
In 1992, there was a total of about 327,600 ha under oil palm cultivation in Ghana,
which is about 5% of the total land area within the oil palm belt (6.8 million ha) (Gyasi,
1992). This value is smaller than the 349,040 ha of mature oil palm reported by FAO
(2015). The difference between the estimate of Gyasi (1992) and FAO (2015) is most
likely explained by different approaches to census and a 23-year gap between both
studies. It is not clear which value is the most accurate. About 311,000 ha (95%) was
cultivated by smallholder farmers, producing about 897,000 t or 84% of FB production.
About 16,600 ha (5%) was managed under industrial plantations that account for
167,000 t of FB (16%) (Table 3.1).
3.3.2 Water-limited yield (Yw), actual yields and yield gaps at oil palm plantations
and smallholder farms
The boundary line analysis performed on FB yield data recorded at BMP plots followed
the typical yield profile of oil palm, illustrating a SAYP at 3 years after planting (YAP)
to the point when yield peaks at about 10 YAP (RMSE = 0.061) (Fig. 3.3). The DYP is
shown for two scenarios; one in which there is no yield decline expected after the
plateau is reached, and one where management practices impair yield with increasing
palm age at YAP ≥13. The DYP in the second scenario could not be represented by the
boundary line model and was fitted to a select number of data points with linear
regression, to accurately capture a decline in yields.
The water-limited yield (Yw) over the entire production cycle (3–20 YAP) averaged
21 t ha-1 yr-1 FB. The average FB yield in the steep ascending phase was 15 t ha-1 yr-1,
26 t ha-1 yr-1 in the plateau phase and 23 t ha-1 yr-1 in the declining phase (Fig. 3.3). Ya
was less than Yw at all smallholder and plantation sites (Fig. 3.3). Actual yields (Ya) at
plantations sites averaged 10.8 t ha-1 yr-1 FB for favourable climatic zones, and
13.4 t ha-1 yr-1 FB for optimal climatic zones. At smallholder sites, average actual yields
(Ya) were 7.7 t ha-1 yr-1 FB and 7.5 t ha-1 yr-1 FB for favourable and optimal climatic
zones respectively.
Yield gaps at smallholder sites were significantly larger (P≤0.05) than at plantations for
both climatic zones. In optimal climatic zones, average yield gaps across all production

43
CHAPTER 3

phases were 15.5 t ha-1 yr-1 FB for smallholders (n=7) and 9.4 t ha-1 yr-1 FB for
plantations (n=8). In favourable climatic zones, average yield gaps were 15.2 t ha-1 yr-1
FB for smallholders (n=13) and 10.2 t ha-1 yr-1 FB for plantations (n=8).

Fig. 3.3. Actual yields (Ya) plotted against year after planting for oil palm plantations (OP) and
smallholder farms (SML) in the optimal (Opt) and favourable (Fav) climatic zones in southern Ghana.
Vertical dotted lines indicate the boundaries between production phases (steep ascent yield phase
(SAYP), plateau yield phase (PYP), declining yield phase (DYP)) in oil palm. The boundary line (Yw)
is a fitted regression line through the upper points of the BMP trial yield data, and was calculated using
the model of Fermont et al. (2009). The dashed line was fitted to a select number of data points with
linear regression to capture the decline in yield from the 12th year.

44
 Table 3.1. Estimated area, fruit bunch production and yield by sector in Ghana in 2015 based on field surveys carried out in 2015 and data reported by
FAO (2015), MASDAR (2011) and Ofosu-Budu and Sarpong (2013).

Smallholder Fruit Mill capacity

Oil extraction
Area under harvest Fruit bunch production fruit bunch Installed Operating CPO yield
Sector rate
purchases yield capacity capacity
ha % t % t t ha-1 yr-1 t hr-1 % t ha-1 yr-1
Oil palm plantations
TOPP 3,250 1 41,300 4 37,600 12.7 30 30 19.8 2.5
BOPP 4,890 2 42,300 4 54,700 8.7 20 16 20.8 1.8
Norpalm 3,760 1 42,000 4 39,000 11.2 30 25 21.0 2.3
GOPDC 4,700 1 41,400 4 67,300 8.8 60 - 23.0 2.0
Total plantations 16,600 5 167,000 16 10.1
Smallholders
Plantation out-grower schemes 18,000 5 162,000 15 9.0 - -

45
Smallholder farmers 140,000 43 420,000 39 3.0 - -
Wild oil palm groves 150,000 46 300,000 27 2.0 - -
Medium scale farms (>10ha) 3,000 1 15,000 1 5.0 - -
Total smallholders 311,000 95 897,000 84 2.9
Total 327,600 100 1,064,000 100 3.2
FAOSTAT (2014) 349,040 2,443,270 7.0
YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA
CHAPTER 3

3.3.3 Factors contributing to oil palm yield gaps

Soil chemical properties and crop nutrition
The standard nutrient management practice on plantations is to apply fertilizers over the
weeded palm circle whilst spent male flowers, pruned fronds and empty fruit bunches
are applied in the inter row space. As a result, soil pH was significantly lower in the
weeded circle zone compared with the frond stack zone at plantation sites (P≤0.05),
probably due to acidification caused by the repeated application of ammonia-based N
fertilizers (Table 3.2, Table 3.4) (Goh and Härdter, 2003). By contrast, the average soil
pH was higher (P≤0.05) in smallholder soils (5.2) compared with plantation soils (4.6),
but differences between zones were less pronounced. At both sites, soil organic carbon
(SOC) and nitrogen (N) concentrations were larger in the soil beneath the frond stack
(significant at smallholder sites (P≤0.05)) (Table 3.2). This is likely due to the
maintenance of a weed-free zone of about 1.5–2.0 m from the base of the palm trunk to
facilitate loose fruit collection, a standard practice in oil palm plantings. Because the
weeded circle is kept clean from debris, there is no replenishment in soil organic matter.
Average soil available phosphorus (P) was significantly greater in plantation soils
(38 mg kg-1) than in smallholder soils (6 mg kg-1) (P≤0.05), with a larger (P≤0.05)
concentration of available P in soils beneath weeded circles (Table 3.2), due to
application of P fertilizers within this zone. Similarly, the concentration of
exchangeable potassium (K) was greater (P≤0.05) in plantation soils
(0.296 cmol(+) kg-1), and deficient in smallholder soils (0.125 cmol(+) kg-1).
Exchangeable K was significantly larger (P≤0.05) in the soil beneath weeded circles
than in the frond stack at plantation sites (Table 3.2), again explained by past application
of K fertilizers. By contrast, the average amount of exchangeable magnesium (Mg) was
significantly smaller (P≤0.05) and less variable in plantation soils (0.58 cmol(+) kg-1)
compared with smallholder soils (0.86 cmol(+) kg-1) (P≤0.05), and also significantly
smaller (P≤0.05) in the soil beneath the weeded circle zone at plantation sites. This
suggests that soil Mg reserves are depleted to a greater extent at plantation sites, due to
greater Mg offtake in fruit bunches and insufficient replenishment of soil Mg with either
crop residues or mineral fertilizer (Table 3.2, Table 3.4). In general, differences in soil
chemical properties between zones were less pronounced at smallholder sites, where
there was little or no past application of mineral fertilizers.
Leaf N concentration was sufficient but leaf P concentration was generally deficient at
plantation and smallholder sites (Table 3.3). Leaf K concentration was adequate in the
plantations but deficient in smallholder sites. Leaf Mg concentration was significantly
smaller at the plantations (P≤0.05), but adequate in smallholder sites, where Mg
removal in fruit bunches was smaller. Leaf B concentration (mg kg-1) was deficient at

46
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

both plantation and smallholder sites, but significantly larger in plantations compared
to smallholders (P≤0.05) due to past application of B fertilizer at some sites.
Plantation establishment and agronomic management
The average palm stand (number of productive palms ha-1) at smallholder sites was
higher (141 palms ha-1), but also more variable, compared with plantation sites
(128 palms ha-1) (Table 3.4). Very dense palm stands indicate inaccurate palm point
lining and result in inter-palm competition, and reduced light interception by the palm
leaf canopy, while low density palm stands often indicate failure to remove
unproductive palms, or infill gaps in the palm stand during the immature phase
(Fairhurst and Griffiths, 2014).
Oil palm plantation sites received on average 1.84 kg palm-1 nutrients. Average
application rates were 0.59 kg palm-1 N, 0.16 kg palm-1 P, 0.93 kg palm-1 K and
0.013 kg palm-1 Mg. Nutrients N, P, K, and Mg were supplied in a range of fertilizer
products, including compounds (e.g. 10–10–30, 15–15–15) and straight fertilizers (e.g.
KCl, kieserite, rock phosphate, ammonium sulphate and urea). In addition, one
plantation site received a one-off application of 14 t ha-1 empty fruit bunch (EFB)
mulch. Fertilizers were not used by any of the smallholders.
Glyphosate was the main herbicide used for controlling weeds in paths and circles in
both plantations and smallholdings. Use of this herbicide was greater in smallholdings
(3.9 l ha-1) compared with plantations (0.9 l ha-1) (P≤0.05). Use of tryclopyr for the
control of woody plants was also greater in smallholdings (0.3 l ha-1) than in plantations
(0.1 l ha-1). Herbicide use was more effective in plantations, however, where better weed
control was achieved with smaller amounts of herbicide presumably as manufacturer’s
recommendations were followed. Furthermore, initial farm surveys showed that
herbicide use was not a common practice in smallholder farms in Ghana, and were not
used on any of the smallholder trial sites prior to the start of the project. Herbicide use
in smallholdings was most likely copied from BMP plots, which explains the high
application doses due to the farmers’ limited experience.

47
 Table 3.2. Chemical properties of soil samples taken from the weeded circle and palm inter-row (0–40 cm) in oil palm plantations and smallholder farms
in Ghana. Critical levels are taken from Goh et al. (1997). * Indicates a significant difference of the mean between oil palm plantations and smallholders at
P≤0.05.
CHAPTER 3

pH(water) Org. C N Available P (Olsen) Exch. Mg Exch. K

Sector & zone -1 n
- % mg kg cmol(+) kg-1
Oil palm plantations
Circle 4.5* 1.17 0.11 61* 0.44* 0.387* 52
Standard deviation (σ) 0.36 0.29 0.02 58 0.15 0.277
Inter-row 4.9* 1.22* 0.12* 13* 0.73 0.200* 44
Standard deviation (σ) 0.48 0.33 0.03 9 0.51 0.137
Average 4.6* 1.20* 0.11* 38* 0.58* 0.296*
Smallholders
Circle 5.3 1.21 0.12 7 0.82 0.117 52

48
Standard deviation (σ) 0.50 0.46 0.05 7 0.50 0.070
Inter-row 5.2 1.43 0.13 5 0.91 0.133 52
Standard deviation (σ) 0.39 0.53 0.05 2 0.55 0.092
Average 5.2 1.32 0.12 6 0.86 0.125
Critical level 4.0 1.2 0.12 15 0.2 0.2
 Table 3.3. Leaf nutrient concentration (leaf 17) in oil palm plantations and smallholder farms in Ghana in 2013. * Indicates a significant difference of the
mean between oil palm plantations and smallholders at P≤0.05.
N Pa Ca Mg K B TLC K Mg
Sector n
% dry matter mg kg-1 cmol kg-1 % of TLC
Oil palm plantations
Average 2.62* 0.15* 0.71 0.31* 0.96* 14* 85* 29* 30* 24
Standard deviation (σ) 0.12 0.01 0.07 0.07 0.16 3.6 5.6 4.6 5.1
Smallholders
Average 2.51 0.14 0.72 0.41 0.81 12 91 23 38 28
Standard deviation (σ) 0.17 0.01 0.06 0.05 0.21 1.9 6.7 4.8 4.5
Optimum concentrationsb 2.40–2.80 0.15–0.18 0.50–0.75 0.25–0.40 0.90–1.20 15–25
a
Average critical leaf P concentrations (calculated with Fairhurst and Mutert (1999)) for oil palm plantations and smallholders are 0.17 (σ = 0.006) and
0.16 (σ = 0.008) respectively.
b
Taken from Fairhurst et al. (2004).

49
YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA
 Table 3.4. Oil palm plantations and smallholder farms in Ghana described by yield gap and yield components.
Oil palm plantation blocks (n=16) Smallholder farms (n=20)
Yield gap Parameter Units Range Range
Mean Std. Error Mean Std. Error
Min. Max. Min. Max.
Plantation establishment
2 Palm stand productive palms ha-1 128 4.7 100 148 141 7.9 59 232
CHAPTER 3

Agronomic management
3 Fertilizer usea kg palm-1 yr-1 fertilizer 4.5 0.69 0.0 9.6 0.0 0.00 0.0 0.0
Fertilizer N kg palm-1 yr-1 N 0.59 0.09 0.00 1.10 0.00 0.00 0.00 0.00
Fertilizer P kg palm-1 yr-1 P 0.16 0.04 0.00 0.59 0.00 0.00 0.00 0.00
Fertilizer K kg palm-1 yr-1 K 0.93 0.14 0.00 1.70 0.00 0.00 0.00 0.00
Fertilizer Mg kg palm-1 yr-1 Mg 0.013 0.009 0.00 0.11 0.00 0.00 0.00 0.00
Total nutrients kg palm-1 yr-1 nutrients 1.84 0.28 0.00 3.55 0.00 0.00 0.00 0.00
Chemicals use l ha-1 yr-1 Glyphosate 0.92 0.077 0.13 1.93 3.92* 0.333 0.00 7.81
l ha-1 yr-1 Triclopyr 0.10 0.013 0.00 0.31 0.28 0.119 0.00 4.00
Crop recovery
4 Harvest cycles cycles yr-1 29 1.33 19 50 21* 1.20 10 47

50
Harvester output t man-day-1 yr-1 FB 1.6 0.14 0.4 2.8 0.7* 0.07 0.1 2.3
ha man-day-1 yr-1 1.9 0.15 0.5 4.2 1.3* 0.10 0.4 3.4
Harvesting labour man-days ha-1 cycle-1 0.75 0.07 0.3 1.9 1.13* 0.08 0.3 2.7
Field upkeep labourb man-days ha-1 yr-1 2.38 0.21 0.53 6.12 2.76 0.17 1.10 5.56
Number of pruning cycles cycles yr-1 1.3 0.09 0 3 1.2 0.07 0 2
Yield components
Bunch number bunches ha-1 yr-1 91 5.95 45 180 77 6.30 10 180
Average bunch weight kg 12.1 0.81 2.4 17.3 9.1* 0.51 3.5 16.5
Loose fruit collection t ha-1 yr-1 loose fruit 0.9 0.15 0.0 3.4 0.5 0.06 0.0 1.3
Fruit bunch yield t ha-1 yr-1 FB 12.0 0.74 2.8 18.9 7.6* 0.56 1.3 16.7
Average oil extraction rate (OER)c % 21 12
Yield of crude palm oil t ha-1 yr-1 CPO 2.5 0.9
a
Fertilizer types include: 10–10–30, 15–15–15, KCl, Kieserite, RP, SOA, and Urea. In addition, one estate block received an application of 0.7 t ha-1 empty
fruit bunches.
b
Field upkeep labour includes circle-, path-, and interline spraying, (manual) weeding, circle raking, frondstacking, pruning, construction & maintenance of
drains and footbridges, and supervision activities.
c
Reference values for OER taken from Adjei-Nsiah et al. (2012).
* Indicates a significant difference of the mean between oil palm plantations and smallholders at P≤0.05.
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

Crop recovery and yield components

On average, plantation sites were harvested more frequently (29 cycles yr-1) than
smallholder sites (21 cycles yr-1) (Table 3.4). This corresponds to harvest intervals of
approximately 13 and 17 days respectively. As a result, harvesting and field upkeep
labour is more efficiently organised at plantations, resulting in a larger harvester output
(t man-day-1 yr-1 FB) (in terms of crop recovery) as well as more ground covered by
harvesters (ha man-day-1 yr-1) (Table 3.4). At smallholder systems, more labour is spent
on harvesting (man-days ha-1 cycle-1) and field upkeep (man-day ha-1 yr-1) because of
poor field conditions where the lack of harvest paths and weeded circles impeded access
for harvest and palm upkeep (Table 3.4).
Average FB yields with standard field practices were larger at plantation sites
(12.0 t ha-1 yr-1) than at smallholder sites (9.1 t ha-1 yr-1) (Table 3.4). Larger FB yields
at plantations can partly be explained by more frequent harvest cycles and better field
access, resulting in the recovery of more bunches with a larger average bunch weight
(mainly due to more complete collection of loose fruits) than at smallholder sites (Table
3.4). Compared with oil extraction rates of >24% reported for the tenera hybrid (Ng et
al., 2003), present oil extraction rates in Ghana are very poor with 21% for large-scale
processors and 12% for small-scale processors.
3.4 Discussion
3.4.1 Oil palm production systems in Ghana, their yield gaps, causes, and
remedial measures
The water-limited yield (Yw) over the planting cycle of oil palm averaged
21 t ha-1 yr-1 FB, with yield gaps of 15.4 t ha-1 yr-1 FB and 9.8 t ha-1 yr-1 FB at smallholder
and plantation sites, respectively, showing a large potential for yield improvement.
Current yield gaps are mostly the result of incomplete crop recovery, inadequate
agronomic management, especially nutrient management, and poor plantation
establishment (Fig. 3.4).
There is considerable scope to improve production on oil palm plantations and
smallholder farms in Ghana by closing Yield Gap 4 with better crop recovery
(Rhebergen et al., 2014). Continuity in production requires tightly controlled harvesting
cycles, as well as sufficient labour for harvesting and fruit collection and field upkeep
(Fairhurst and Griffiths, 2014). Yield Gap 4 can be closed by improvements to field
access (ground cover control, weeded circles, paths, pruning) and the implementation
of three harvest cycles per month (i.e. harvesting intervals of <10 days), particularly
during the peak crop months. Shrubs and weeds may obstruct in-field access and
compete with oil palm for water and nutrients and may be eradicated by manual removal

51
CHAPTER 3

or herbicides. In addition, a large application of rock phosphate may be effective in

triggering a succession of ground cover species composition from weeds adapted to
poor soil fertility to grasses and legume cover plants that are more competitive when
soil fertility has been improved (Giller and Fairhurst, 2003). In inland valleys, crop
recovery is often obstructed by poor drainage, due to a lack of drainage outlets and field
drains, or drains that are too shallow and require desilting.

Fig. 3.4. Partitioning yield gaps at oil palm plantations and smallholder farms. Estimates for Yp and Yw
were taken from Rhebergen et al. (2014) and the boundary line approach respectively. Ymey was
estimated by adjusting Yw based on the relationship between the stand per hectare and the planting
density (Fairhurst and Griffiths, 2014) for plantation and smallholder sites. Yam was estimated by
measuring the yield improvement at BMP plots over the first 12 months in which most plots achieved
full crop recovery and yields obtained under standard field practices were taken as estimates for Ya.

Soil and leaf analysis data shows that there are significant nutrient deficiencies,
particularly in smallholder farms, that must be addressed to close Yield Gap 3
(Fig. 3.4). At smallholder sites, for example, a strong relationship between leaf K and
P concentrations and exchangeable K and available P in the soil beneath weeded circles
was found, where leaf and soil P and K concentrations are generally deficient (Fig. 3.5).
Whilst plantations apply moderate amounts of mineral fertilizers and occasionally
recycle small amounts of crop residues (Table 3.4), most smallholder farmers apply
little if any mineral fertilizer and do not recycle crop residues, resulting in a larger Yield
Gap 3 (Fig. 3.4). Smallholder farmers also lack access to empty fruit bunches sold to
plantation mills because the empty bunches are usually recycled by plantation-owned
mills to their own plantings, albeit in small amounts. Whilst some crop residues are
available at small-scale processors, most EFB and fibre at small-scale processors are
used as fuel to cook fruit bunches before they are pressed to extract oil (Osei-Amponsah
et al., 2012). Furthermore, most smallholder farmers were unaware of the benefits of
mulching with EFB.

52
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

Fig. 3.5. Relationship between leaf K and P concentrations and the amount of exchangeable K and
available P in soil beneath weeded palm circles in smallholder (black circles) and plantation sites (open
circles. The leaf and soil critical values for K and P are represented by horizontal (long dash),
respectively, vertical (short dash) lines.

At plantation sites an average total leaf cation (TLC) of 85 cmol kg-1 was reported
(Table 3.3). At this value, data from 50 fertilizer trials in Malaysia suggest leaf critical
concentrations for N, P, K and Mg of 2.72, 0.169, 1.15, and 0.22 %DM (Foster and
Prabowo, 2006). Average leaf nutrient concentrations for plantations were far below
these values, suggesting inadequate and unbalanced fertilizer use. However, optimum
or critical values for individual nutrient concentrations vary considerably, depending on
factors such as palm age, leaf number, leaflet rank, leaf age, planting material, balance
with other nutrients, environment, spacing and inter-palm competition (von Uexkull
and Fairhurst, 1991; Fairhurst and Mutert, 1999). Therefore, the critical nutrient
concentrations applicable to the region and sites under study may vary from the values
calculated here. Yield Gap 3 can be closed with a more balanced approach towards
nutrient management, taking into account the right source of nutrient applied at the right
rate, time and place as guided by the 4R Nutrient Stewardship (IPNI, 2012). In most oil
palm fertilizer trials, yields in different treatments are significantly correlated with leaf
nutrient concentrations (Foster, 2003). The use of reference leaf critical levels
established through site-specific fertilizer trials is therefore a useful way to assess
fertilizer requirements for optimal yield levels (Foster, 2003; Foster and Prabowo,
2006). Additionally, placing pruned fronds as a ‘box’ around the palm is a management
strategy to improve soil structure, increase the rate of water infiltration and prevent
erosion (Fairhurst, 1996; Gillbanks, 2003; Goh et al., 2003). Using crop residues
efficiently and addressing these root-soil dynamics, a feeding zone for oil palm can be
created in the inter-row and targeted for nutrient application.

53
CHAPTER 3

Deficiencies in management during the establishment phase that cause Yield Gap 2
were evident at both plantation and smallholder sites (Fig. 3.4). Major problems
identified at all sites were insufficient drainage and failure to correct poor drainage in
plantings where N-deficiency symptoms were evident, poor and late infilling to replace
dead seedlings, lack of removal of abnormal and unproductive palms, poor land forming
(terraces) and lack of platforms, poor establishment of legume cover plants due to low
soil fertility (in particular P) and poor management of (woody) weed growth. Yield Gap
2 can only be closed at each 20–25 year cycle of replanting, when there are opportunities
to introduce improved germplasm and use better planting techniques. Yield Gap 2 can
be reduced by careful management to ensure a complete stand of productive palms at
maturity. This includes careful land selection, preparation and clearing with minimal
soil damage (i.e. compaction, erosion, removal of topsoil), choosing good quality and
high yielding planting material, accurate planting procedures such as correct planting
density, lining and timely infilling, and proper agronomic management up to the onset
of harvest (Fairhurst and Griffiths, 2014). Establishment of (legume) cover plants and
placement of pruned fronds and mulch along the contour can help to increase water
infiltration and reduce soil erosion (Paramananthan, 2003; Rankine and Fairhurst,
1999). An annual palm census can be used to identify the number of unproductive palms
that require replacement to maintain an optimal stand of productive palms, and to reduce
Yield Gap 2.
Based on data reported by Rhebergen et al. (2014), the average potential yield (Yp) over
a planting cycle in Ghana is estimated at 31 t ha-1 yr-1 FB. Yw is smaller than Yp when
water supply is limited. Yield Gap 1 therefore arises due to the difference between Yp
and Yw and is estimated at 10 t ha-1 yr-1 FB (Fig. 3.4). This is equivalent to the response
to irrigation under similar climatic conditions in Thailand (Tittinutchanon et al., 2008).
Yield Gap 1 can be closed over a 35–40 month period by using irrigation to eliminate
water stress.
3.4.2 Constraints to increasing palm oil production in Ghana
Besides the requirements to improve FB production, fruit processing offers challenges
of its own. For example, at current levels of production and oil extraction rate, the loss
of oil in both the large and small-scale sectors are substantial. Maximum crude palm oil
(CPO) losses are 5.2 and 3.3 t ha-1 yr-1 in the plateau yield phase, and the total amount
of CPO lost over the entire planting cycle of oil palm is ~75 t ha-1 and ~50 t ha-1 in
small-scale and large-scale production systems respectively (Fig. 3.6). Therefore, to
improve oil yields, both FB yields and milling efficiency must be improved.

54
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

Fig. 3.6. Estimated CPO yield (t ha-1) at large-scale and small-scale production systems. Yw was
calculated as the product of the boundary line from Fig. 3.2 and an oil extraction rate (OER) of 24%,
whilst the curves for large-scale and small-scale production systems were estimated with a regression
through the average yields (Ya) at oil palm plantations and smallholder farms, multiplied by current oil
extraction rates achieved at large-scale (21% OER) and small-scale processors (12% OER). The dotted
area shows the total oil loss (CPO) over the production cycle of oil palm at large scale production
systems (51 t ha-1) and the dotted + dashed area at small scale production systems (75 t ha-1).

Improving production and oil extraction rates in Ghana’s oil palm sector could make a
significant contribution to closing the supply gap for palm oil in Ghana, and could lead
to greatly increased profitability for investors and farmers alike (Table 3.5). Investments
to reduce yield and oil supply gaps will benefit smallholder farmers more, with a
fourteenfold increase in CPO production and economic value (worth > 1 billion US$),
while a twofold increase in CPO production is projected at plantations (Table 3.5).
Despite the potential for increasing oil palm yields, smallholder farmers face major
challenges that include lack of knowledge on appropriate management practices, poor
infrastructure and lack access to finance (IPPA, 2010). Currently, working capital is
commonly sourced through loans through informal community arrangements
(MASDAR, 2011). Inability to purchase agricultural inputs such as fertilizers and/or
herbicides/pesticides, results in poor yields and reliance on arduous manual labour.

55
 Table 3.5. Impact of closing current yield gaps and improving oil extraction rates in Ghana on CPO production and economic value based on production
data from 2015.

Oil palm plantations Smallholders

CHAPTER 3

Parameter Units
Before After Change Multiplier Before After Change Multiplier
Area ha 16,600 16,600 - 1 311,000 311,000 - 1
Bunch yield t ha-1 yr-1 FB 10.1 20.0 10 2 2.9 20.0 17.1 7
Bunch
t FB 167,000 332,000 165,000 2 897,000 6,220,000 5,323,000 7
production
Oil extraction
% 21 24 3 1 12 24 12 2
rate
Crude palm oil t 35,070 79,680 44,610 2 107,640 1,492,800 1,385,160 14
CPO yield t ha-1 yr-1 CPO 2.1 4.8 2.7 2 0.3 4.8 4.5 14
Crude palm oil US$ t-1 yr-1 750 750 - 1 750 750 - 1

Crude palm oil US$ 2 14

56
26,302,500 59,760,000 33,457,500 80,730,000 1,119,600,000 1,038,870,000
 YIELD GAP ANALYSIS AND ENTRY POINTS FOR IMPROVING PRODUCTIVITY ON LARGE OIL
PALM PLANTATIONS AND SMALLHOLDER FARMS IN GHANA

Because of the 35–40 month time lag between management interventions and yield
improvement in oil palm, most farmers also do not invest in agricultural inputs due to
the delayed impact on yield and revenue.
Furthermore, distribution and marketing of agricultural inputs such as fertilizer is
generally poor in Ghana (Krausova and Banful, 2010). Fertilizer dealers experience
high transport costs, lack of customer demand, unreliable suppliers and lack of technical
knowledge. In addition, compared with oil palm producing countries in Southeast Asia
such as Malaysia and Indonesia, smallholder farmers in Ghana are not part of schemes
and are not well integrated into the industry. They do not receive the benefits of
plantation-outgrower schemes such as the provision of high-yielding seedlings,
agronomic inputs, credit, and advisory services (Fold and Whitefield, 2012).
Most of the smallholder farmers sell their crop to small-scale processors who, combined
contribute >80% of the total national CPO production (Table 3.1) (Adjei-Nsiah et al.,
2012a; MASDAR, 2011; Osei-Amponsah et al., 2012). However, at current oil
extraction rates of 12% and at an average milling capacity of 7 t day-1 mill-1 FB, small-
scale processors combined (n=400 (Angelucci, 2013)) can only process 560,000 t yr-1
FB, equivalent to 67,200 t CPO (Table 3.6).
Table 3.6. Annual CPO production at large and small-scale mills in Ghana.
Parameter Units Large-scale mills Small-scale processors
Milling capacity tph 20 <1
Milling hours hr month-1 550 225
Monthly capacity t month-1 11,000 210
Peak crop % 15 15
Annual crop t FB 73,333 1,400
Number of mills # 20 400
Total annual crop t FB 1,466,667 560,000
Oil extraction rate (OER) % 21 12
Crude palm oil (CPO) t 308,000 67,200
Small-scale processors are often poorly organised, use inferior processing technology,
and lack price incentives. The CPO obtained at small-scale processors is usually of poor
quality (in terms of free fatty acid, moisture, and impurity content), and does not meet
the standards required to enter local industrial and/or international markets. Most of the
CPO ends up being sold at villages or small town markets for local consumption (Adjei-
Nsiah et al., 2012a; Gilbert, 2013).
On the other hand, large-scale mills in Ghana combined (n=20) have the processing
capacity to offset the annual national CPO deficit (106,000 t). Assuming an average
milling capacity of 20 t hr-1 and a peak crop production of 15%, ~1,5 million t FB can
be processed, which is the equivalent of 308,000 t CPO (at 21% OER) (Table 3.6).

57
CHAPTER 3

Large-scale mills thus have the capacity to process the current total national FB
production (including smallholder production), but additional investments in mills
would be required were significant yield improvement achieved across all sectors.
3.5 Conclusions
Yield gaps in oil palm production are large and ubiquitous in Ghana. Improving yields
through corrective management practices offers considerable opportunities for the oil
palm sector. Water-limited yield (Yw) over the planting cycle of oil palm were estimated
at 21 t ha-1 yr-1 FB and large average yield gaps estimated at 15.4 t ha-1 yr-1 FB for
smallholder farms and 9.8 t ha-1 yr-1 FB for plantations. A simple model to partition
yield gaps and understand their causes supported the identification of where and why
yield gaps occur and how they can be closed with better management practices. Yield
gaps associated with poor establishment and poor management practices require
different interventions with different potential impact and contrasting time scale for
implementation and impact. Opportunities to close yield gaps caused by poor crop
recovery (Yield Gap 4) and inadequate agronomic management (Yield Gap 3) are
particularly large at smallholder farms. Most smallholder oil palm plantings were
severely neglected and poorly accessible, and consequently, incomplete crop recovery
was the main cause of low yields. Investments in oil palm management practices to
increase productivity and improve yield recovery on plantations and smallholder farms,
and improving access to efficient processing mills by smallholder farms offer
opportunities to substantially increase the production of crude palm oil without
increasing the area planted.
3.6 Acknowledgements
We thank the management and staff of Benso Oil Palm Plantation (BOPP), Norpalm
Ghana Ltd., Twifo Oil Palm Plantation (TOPP) and Solidaridad West Africa for their
support and the provision of data. We also thank Chris Donough, Munir Hoffmann,
Thomas Oberthür, Godfrey Taulya, Rob Verdooren, Na Wang, and the Oil Palm
Research Institute (OPRI). This work was supported by Canpotex International Pte
Limited and K+S Kali GmbH. The contribution of A. Whitbread was supported by the
CGIAR research program Water, Land and Ecosystems (WLE).

58
 CHAPTER 4

Closing yield gaps in oil palm production systems in

Ghana through Best Management Practices

This chapter is submitted as:

Rhebergen, T., Zingore, S., Giller, K.E., Adu Frimpong, C., Acheampong, K., Tetteh
Ohipeni, F., Panyin, E.K., Zutah, V. and Fairhurst, T. 2019. Closing yield gaps in oil
palm production systems in Ghana through Best Management Practices.
CHAPTER 4

Abstract
Since the early 2000s, the area under oil palm in Ghana expanded but average fruit bunch (FB) yields
remained low, resulting in large yield gaps on plantations and smallholder farms. This study assessed
the potential for reducing yield gaps through ‘Best Management Practices (BMP)’ on plantations and
smallholder farms in southern Ghana, compared with current management practices. We evaluated
BMPs that increased yields in the short-term (≤ 1 year) with ‘yield taking’ (improved crop recovery),
and in the long-term (>1 year) with ‘yield making’ (better agronomy) practices and identified the
management factors that contributed most to yield improvements. Average yields and yield gains with
BMP were larger at plantations and smallholders, compared with the REF treatment. Average FB yield
increases with BMP were 2.1 t ha-1 (+19%) and 4.7 t ha-1 (+89%) with yield taking and 4.7 t ha-1 (+36%)
and 7.6 t ha-1 (+76%) with yield making at plantations and smallholders respectively. Important crop
recovery activities included more frequent harvesting events and improved field access (roads, paths,
weeded circles) to increase harvester efficiency and productivity. Analysis of fertilizer usage and leaf
nutrient concentrations suggest a more balanced approach to nutrient management could contribute
considerably to yield making, particularly at smallholder farms. The need to develop new improved
fertilizer recommendations is therefore essential to increase yields and to sustainably intensify oil palm
production in Ghana. Increasing average attainable yields to 21.0 t ha-1 FB can increase national FB
production almost three-fold from 2.5 Mt to 6.9 Mt (1.4 Mt CPO at 21% oil extraction rate) and reduce
oil palm expansion with 597,636 ha of land.
Keywords: Yield intensification, crop recovery, fertilizer, smallholder, yield gap, land sparing.

60
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

4.1 Introduction
Oil palm (Elaeis guineensis Jacq.) is one of the world’s most rapidly expanding
equatorial crops, driven by increasing global demand for vegetable oil and biofuel
(Corley, 2009; Fitzherbert et al., 2008; Wich et al., 2014). Between 1975 and 2014, the
global land area under mature oil palm increased fourfold from 3.5 Mha to 18.7 Mha,
with most expansion in Southeast Asia, notably Indonesia (with a total area currently
under harvest of 7.4 Mha) and Malaysia (total area of 4.7 Mha) (FAO, 2017). The
growth in oil palm production has contributed to improved economic growth and rural
poverty alleviation (Corley, 2009; Edwards, 2015; Sayer et al., 2012), though much of
the area expansion has been at the expense of logged-over tropical rainforest (Danielsen
et al., 2009; Fitzherbert et al., 2008; Koh and Wilcove, 2008). Limited land availability
in Southeast Asia has led to a search for suitable land elsewhere, with most future
expansion expected in Latin America and sub-Saharan Africa (SSA) (Laurance et al.,
2014; Sayer et al., 2012). In SSA, Nigeria, Cameroon and Ghana produce the most palm
oil (7.9 Mt yr-1, 2.7 Mt yr-1, and 2.4 Mt yr-1 in 2014 respectively), while the largest
expansion in area over the past decade (2004–2014) took place in the Democratic
Republic of Congo (+118,000 ha), Cameroon (+81,000 ha), Ivory Coast (+74,000 ha)
and Ghana (+31,500 ha) (FAO, 2017).
Since expansion of oil palm cultivation is often linked to deforestation, it is suggested
that increasing yields on land already planted with oil palm or expanding production
only on degraded or abandoned lands can spare land for nature (Fairhurst and
McLaughlin, 2009; Wicke et al., 2011). Yield intensification focuses on reducing the
yield gap (Yg) between the potential (Yp for environments with adequate water supply
or Yw under rainfed conditions where water is limited) and actual yield (Ya) with
improved agronomic practices or better management (Fairhurst and Griffiths, 2014;
Fischer et al., 2014; van Ittersum et al., 2013). In oil palm, for example, ‘Best
Management Practices’ increased fruit bunch (FB) yields by 6.0 t ha-1 in South Sumatra,
Indonesia (Griffiths and Fairhurst, 2003), and by 3.4 t ha-1 (+15%) across six
commercial plantations in Indonesia (Donough et al., 2010).
Oil palm is an important economic crop that provides a major source of employment in
Ghana (Gilbert, 2013). In the early 2000s, oil palm was selected by the Ghanaian
government as a strategic crop to promote agricultural and industrial growth for poverty
reduction and rural development (Asante, 2012; Osei-Amponsah et al., 2012). As part
of the program, oil palm seedlings were distributed amongst farmers but field plantings
were poorly managed. Additionally, new investments were made in the oil palm
industry, attracting interest of foreign investors to develop large-scale plantations.
During this period, the area under oil palm expanded but the average yield remained

61
CHAPTER 4

low. The small fruit bunch yields observed in Ghana are associated with multiple
constraints that include sub-optimal climate, poor soil fertility and poor management
practices, particularly related to crop recovery (Chapter 2, Chapter 3; Rhebergen et al.,
2014). With an estimated Yw over a planting cycle of 21.0 t ha-1 FB (averaged across
favourable (areas with a mean annual water deficit <250 mm) and optimal (areas with
a mean annual water deficit <150 mm) production sites in Ghana (Chapter 2)), and a Ya
of ~11.0 t ha-1 FB on large commercial plantations and ~6.0 t ha-1 FB on smallholder
farms, it is clear that large yield gaps (Yg) exist in Ghana (Chapter 3).
Best Management Practices (BMP) are cost-effective and practical agronomic
techniques that focus on reducing yield gaps in oil palm by using production inputs and
resources efficiently (Donough et al., 2009; Griffiths and Fairhurst, 2003). BMPs aim
to increase oil palm productivity through improvements in agronomic management, as
well as increased crop recovery. Implementation of BMPs is site-specific, since they
are tailored to address the particular production constraints and biophysical conditions
of individual locations (Pauli et al., 2014). BMPs are grouped in two broad categories;
‘yield taking’ and ‘yield making’ practices. Yield taking increases yield in the short
term by improving crop recovery operations (e.g. field access, harvest intervals, oil
content), while yield making includes agronomic practices that contribute to building
large and sustainable yields in the longer term (e.g. nutrient and leaf canopy
management, higher oil extraction rates (OER)) (Fairhurst and Griffiths, 2014).
In this chapter, we quantify and evaluate the effect of BMPs compared with current
plantation and farm management practices on agronomic and yield performance of oil
palm with a focus on yield taking, and yield making practices over a 3–year period
(2013–2015). Our objectives were to assess: (i) the impact of crop recovery and
agronomic practices on yield components of oil palm and to (ii) identify the
management factors that contribute most to yield improvements.
4.2 Methodology
4.2.1 Study area
The study was conducted in the oil palm belt of southern Ghana, approximately between
latitudes 6°46’N and 4°55’N and longitude 0°47’W and 2°28’W. Within the study area,
rainfall distribution is bimodal. Mean annual precipitation is highest in the southwest
(~2,400 mm yr-1), and gradually decreases when moving north. Mean annual relative
humidity (RH) is high (~80%), and mean monthly temperatures seldom drop below
25°C, with a small diurnal range of 5–9°C. The topography is predominately undulating
to rolling (2–9°), with rolling to hilly terrain (with slopes >20°) in the southwest. Soils
are predominately strongly weathered and highly leached Acrisols and Ferralsols
(USDA: Ultisols and Oxisols respectively) with low pH and poor soil fertility status.

62
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

4.2.2 Plot selection

We selected three major plantations located in the Western and Central regions,
including Benso Oil Palm Plantation (BOPP) (5°06’47.74”N; 1°54’55.15”W), Norpalm
Ghana Ltd. (4°55’29.04”N; 1°53’31.75”W), and Twifo Oil Palm Plantations (TOPP)
(5°32’03.30”N; 1°31’40.67”W), and twenty smallholder farmers distributed across the
Western (10), Central (3), Eastern (5), and Ashanti (2) regions (see location of trial sites
in Chapter 3, Fig. 3.1). At each plantation, three to five paired management blocks were
selected. The paired blocks (n=12) were representative of a plantation and comparable
in size, topography, soil type, year of planting, and planting material (Appendix 6).
Treatments were allocated randomly within each paired plot, with BMPs implemented
in one block and current standard practices maintained as reference (REF) in the second
block (Appendix 5). At smallholder farms, we accurately measured randomly selected
paired plots of 1–4 ha. Smallholder sites were selected based on the following criteria:
(i) tenera palms ≤17 years after planting (ii) farm accessible by road, (iii) farm size
≥3 ha, (iv) triangular palm layout with palm planting distance 8.5 or 9 m, (v) willingness
to maintain farm records and to (vi) implement BMPs.
4.2.3 Best Management Practices (BMP)
BMPs were implemented in a stepwise and time-lagged process. First, we identified
agronomic constraints for each site and BMP plot by conducting field agronomic audits,
and then implemented site-specific BMPs accordingly. In the first year, we focused
entirely on rehabilitating fields to achieve full field access and crop recovery and to
close Yield Gap 4 (Fig. 3.2). Once this was accomplished, additional yield making
activities such as nutrient management to achieve the yield potential (Yw) were
implemented at each site and to close or minimize Yield Gaps 2 and 3 (Chapter 3). The
sequence of implementing BMPs is crucial to maximizing economic returns, since
closing Yield Gap 4 generates increased cash returns that can be used to implement
practices that contribute to closing Yield Gaps 3 and 4.
Field auditing and palm census
Field agronomic audits were performed to (i) identify and quantify field practices that
require improvement, (ii) evaluate efficiency of production input and resource use, and
to (iii) verify whether BMP standards were achieved. Audits were carried out in
collaboration with plantation staff and smallholder farmers to facilitate knowledge
transfer and a full understanding of the BMP concepts.
We evaluated fields for (i) harvesting practices, (ii) loose fruit and fruit bunch
collection, (iii) pruning and pruned frond management, (iv) ground cover vegetation,
including legume cover plants (LCP), (v) soil conservation, (vi) path and circle weeding

63
CHAPTER 4

and maintenance, (vii) drainage, (viii) erosion, (ix) road maintenance, (x) pests and
diseases, and (xi) fertilizer and crop residue application. Harvesting and (loose) fruit
collection was scored either 1 or 3, while all other parameters 1, 2, or 3. Fertilizer and
crop residue application was scored based on compliance with best nutrient
management practices as guided by the 4R Nutrient Stewardship (IPNI, 2012).
Parameters that were given a score of ‘1’ required immediate remedial action; a score
of ‘2’ was considered below standard but no immediate attention required, while
parameters that were scored as ‘3’ were considered to be BMP standard. Harvesting
practices were only scored as either ‘1’ or ‘3’, because a score of less than 3 implies
crop loss and therefore requires urgent attention. Field audit evaluation criteria are given
in Appendix 7 (adapted from Rankine and Fairhurst (1998)).
Field audit results were summarized by first calculating the total score for each
parameter across all BMP plots at each site, relative to the total possible score for each
parameter. We then took the average of the scores for each category (harvesting,
cultivation and upkeep, pest and disease control, nutrient management), expressed as
percentage. Field audits were done more frequently in the first year to familiarize
plantation and field staff with BMP procedures and standards and in recognizing field
constraints. The frequency was subsequently reduced towards the end of the three-year
period. Reference plots, on the other hand, were not controlled, and no attempt was
made to prevent the implementation of BMPs in REF plots.
A palm census was carried out each year to determine the palm stand per hectare (SPH)
at each site (i.e. the number of productive palms per hectare). Palm points were plotted
on an isometric map indicating the number of mature, immature, new/supply, abnormal,
and dead/removed palms, or unplantable points. On the BMP plots where the SPH was
poor (<80%), corrective action was taken to optimize the SPH by infilling clusters of
vacant planting points. Based on the field audit and palm census results, a portfolio of
site-specific BMPs was developed for each field, prioritizing certain remedial actions
above others. Implementation of BMPs commenced on different dates at each site, but
all were regularly monitored for 35–40 months to maintain high standards in field
management and maintenance.
Yield taking and yield making BMPs
In oil palm, there is a time lag of 35–40 months between the removal of agronomic
constraints and their impact on yield, which is related to the time interval between floral
initiation and bunch ripening (Breure, 2003; Fairhurst and Griffiths, 2014) (Fig. 4.1).

64
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

Fig. 4.1. Diagram on the effect of stress (red line) and elimination of stress (green line) on bunch number
in oil palm (after Woittiez et al. (2017)). The vertical bars represent three periods where bunch
production is sensitive to stress: sex determination (left), inflorescence abortion (middle), and bunch
failure (right).

The exact time lag depends on stresses imposed by unfavourable growing conditions
and poor agronomic management (e.g. poor nutrient management, pruning or drainage)
which trigger complex feedback mechanisms that reduce the sex ratio (i.e. the ratio of
female to total inflorescences) and inhibit floral initiation (Corley and Tinker, 2016;
Jones, 1997). The sex ratio determines the number of bunches that reach harvest and is
affected by both sex determination and floral abortion. Sex differentiation is believed
to occur at Leaf -29 (at approximately 6 months after floral initiation) (Corley and
Tinker, 2016), while developing inflorescences are most sensitive to abortion 4–6
months before anthesis (Broekmans, 1957). Changes in FB yield are due to changes in
one or both of the yield components, bunch number and bunch weight. In oil palm,
bunch number contributes more to yield than does bunch weight (Corley and Tinker,
2016). Fruit maturation time varies from 140–180 days (depending on genetic and
environmental factors), and starts about two weeks after anthesis. Bunch weight is
determined mostly by assimilate availability and the number of flowers that are
pollinated effectively (Woittiez et al., 2017).
BMPs need to be adapted to local conditions and aim to reduce stress in oil palm with
appropriate management and agronomic techniques so that the number of bunches that

65
CHAPTER 4

reach harvest is greater and bunches are larger. Whilst improved agronomy might
already have an effect on bunch number after only six months, increases in yield during
the first year of BMP implementation are mainly caused by improvements in crop
recovery resulting in an increase in the number of available bunches harvested, and an
increase in the recorded bunch weight (because of less loose fruit loss) (Fairhurst and
Griffiths, 2014). Once full crop recovery is achieved (and sustained), additional yield
increases in subsequent years can be attributed solely to the time-lagged effect of
improved agronomic management practices on bunch weight and bunch number. Yet
the full beneficial effects of improved agronomic management practices on yield may
be masked by periods of unfavourable climatic conditions (e.g. large water deficits and
reduced photosynthetic active radiation (PAR) caused by cloud cover) and pest and
disease incidence.
Because of the time-lag between flowering and bunch ripening, at least four years are
recommended to capture the full effect of improved agronomy on yield (Fairhurst and
Griffiths, 2014). Nevertheless, major changes in bunch number and weight are already
expected within a period of 3 years (Fig. 4.1). Rapid increases with crop recovery are
expected in year 1, whilst improved agronomy increases yields by reducing stress at
three critical phases (sex determination, floral abortion and bunch failure) in years 2
and 3. In year 4, yield usually plateaus as all inflorescences initiated in month 0 (Fig.
4.1) have differentiated into harvestable bunches. In this research, we intended to
demonstrate the potential of BMP to increase yields: our goal was not to reach the yield
plateau per se. We therefore present data measured over a period of 36 months, which
is sufficient to illustrate the process of yield intensification with BMP within a time-
frame where great changes in production are expected. Continuation into the fourth year
was not possible given funding limitations.
We partitioned BMPs between those that increase yield in the short-term (i.e. ≤ 1 year
after BMP implementation) with yield taking, and in the long-term (i.e. >1 year after
BMP implementation) with yield making. Yield taking BMPs included:
i. Frequent harvesting to ensure complete crop recovery. At plantations, harvest
intervals of 7–10 days were recommended as a balance between excessively
frequent (where bunches are scarce and the harvester’s output is poor) and
infrequent harvesting (with large amounts of uncollected loose fruits and where
many bunches rot and must be discarded). At smallholder farms, harvest
intervals of 10 days were recommended in the peak season and 14 days in the
low-crop season to compensate for labour costs when bunches are few,
ii. Minimum ripeness standard of five loose fruit (LF) before harvest to ensure
maximum oil content without excessive loose fruit collection,

66
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

iii. Unimpeded in-field access, including clear harvesting paths and footbridges to
cross drains and creeks to allow access to all palms,
iv. Clean weeded circles to allow unimpeded harvesting and collection of fruit
bunches and loose fruit, as well as for efficient uptake of ammonia-based N
fertilizers,
v. Corrective pruning, to facilitate bunch ripeness assessment and to allow
unimpeded harvesting,
vi. Harvested crop delivered to the mill within 24 hours of harvest, to reduce the
amount of free fatty acids (FFA) in the crude palm oil produced.
Yield making BMPs included:
i. Maintenance pruning, including removal of surplus fronds (i.e. old, dead,
damaged or diseased fronds) to maintain a full and healthy palm canopy,
ii. Removal of unproductive and abnormal palms and replanting to improve the
palm stand,
iii. Selective thinning where there was evidence of inter-palm competition (e.g. for
light),
iv. Installation of drains to remove excess water during the wet season and to
conserve water in the dry season,
v. Regular patrols to monitor outbreaks of pests and diseases in order to minimize
fruit loss and palm damage,
vi. Eradication of plants which compete with the palms for nutrients, sunlight and
moisture (e.g. hard grasses (e.g. Panicum maxima), woody plants (e.g. Baphia
nitida)) and replacement with soft weeds and grasses to control erosion during
heavy rains,
vii. Timeliness and accuracy of application of fertilizer and crop residues (e.g. empty
fruit bunches and pruned fronds) to match crop demand and to reduce soil
erosion.
The timing and sequence of BMP implementation is of great importance to achieve the
full beneficial returns on yield. For example, nitrogen (N) fertilizer application is
ineffective if the area is not properly drained, or if woody weeds are not first fully
eradicated.

67
CHAPTER 4

4.2.4 Measurements
Crop production
Number of bunches, fruit bunch (FB) weight (kg), loose fruits (kg), and number of
harvesters were recorded at each harvest event. Fruit bunch weight was determined at
the mill weighbridge at plantations, and with a tripod and digital scale at smallholder
sites. We derived yield components (bunches ha-1, average bunch weight (kg), t ha-1
loose fruit, t ha-1 FB), harvester productivity (t man-day-1 FB, bunches man-day-1,
ha man-day-1), harvesting labour (man-days ha-1 cycle-1) and the average harvest cycle
(cycles yr-1) from the crop production data.
Field upkeep
Field upkeep included circle and path weeding (chemical and manual), circle raking,
interline and selective weeding (chemical and manual), pruning and frond stacking,
installation and maintenance of drains and construction of footbridges, steps and silt
pits. Operations were grouped into five categories; i) access (circle and path weeding),
ii) drainage, iii) interline and selective weeding, iv) pruning and making frond-stacks
and v) other. At each field upkeep event, we recorded the number of man-days and area
covered and derived the average labour spent on each category (man-days ha-1).
Leaf analysis
Permanent datum palm points were established at each site for leaf sampling. Datum
palm points were selected following a systematic layout in a staggered grid pattern of
every tenth palm in every tenth row, providing a sampling density of 1% palms at
plantation sites, or 1–2 palms ha-1. Only healthy and productive palms were selected (if
the candidate palm was not healthy, the nearest neighbour in front of or behind the
candidate palm was selected). Because smallholder sites were considerably smaller, we
selected every fifth palm in every fifth row, providing a sampling density of 3–6 %, or
5–9 palms ha-1. The same datum palm point was sampled each year to reduce the
variability that could otherwise be attributable to sampling different palms at each
sampling event and for greater operational convenience. Leaf and rachis samples were
taken to determine palm nutritional status and to guide fertilizer recommendations.
Samples were taken at each datum palm point from a point approximately two thirds of
the distance between the insertion point of the first true leaves on the leaf petiole and
the distal end of the leaf rachis of Leaf 17 (Chapman and Gray, 1949; Fairhurst et al.,
2004). Composite leaf and rachis samples were analysed for N using a combustion
analyser (Dumas technique), and P, K, Mg, Ca, and B by inductively coupled plasma
analyser (ICP). Leaf tissue samples were taken annually.

68
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

4.2.5 Fertilizer recommendations

Site-specific fertilizer recommendations were prepared for all BMP plots. Fertilizer
recommendations followed a “4R nutrient stewardship” approach, which entails
applying the (1) right source of plant nutrients, at the (2) right rate, at the (3) right time,
and in the (4) right place (IPNI, 2012). Guidelines for ‘4R’ recommendations were as
follows:
• Right source. For plantations we used straight fertilizers, while for smallholders
we used primarily compounds for greater convenience. We selected commonly
available nutrient sources in Ghana (i.e. urea as the N source, triple super
phosphate (TSP) for phosphorus (P), potassium chloride (KCl) for potassium
(K), kieserite (magnesium sulphate) for magnesium (Mg), and borate for boron
(B). Compound fertilizers for smallholders included NPK 10–10–30, which was
supplemented with rock phosphate (RP) and urea at some sites.
• Right amount. In year 1, plantations and smallholder farmers followed their own
fertilizer programmes, whilst from year 2, and after full field access and crop
recovery was achieved, BMP fertilizer recommendations were implemented.
Fertilizer recommendations were designed to improve palm nutritional status
and reach maximum economic yield, based on (un)published information from
fertilizer trials carried out in West Africa (e.g. Danso et al., 2010; van der
Vossen, 1970) and Southeast Asia. For each plantation and smallholder BMP
plot, we first determined whether a particular nutrient was deficient by
comparing leaf nutrient concentrations from the preceding year with critical leaf
and rachis nutrient concentrations based on the results of fertilizer trials (Foster
and Prabowo, 2006; von Uexkull and Fairhurst, 1991). Where nutrient status was
assessed as ‘sufficient’ we applied only a maintenance dose. Where nutrient
status was assessed as ‘deficient’, we applied a corrective dose in addition to the
maintenance dose. Where leaf K concentration was deficient but rachis K
concentration was sufficient, we applied a corrective dose of N fertilizer, which
has been shown to increase leaf K concentration in palms with low leaf K status
but large reserves of K in the leaf rachis (Foster and Prabowo, 2006) (Appendix
8). Expert knowledge was used to adjust fertilizer rates from year 3, depending
on the response in leaf levels. For example, fertilizer rates were reduced if leaf
and/or rachis levels exceeded critical levels. If no response was observed where
nutrient status was poor, fertilizer application rates were increased. Fertilizer
recommendations (elements) varied between 0.75–1.15 and 0.60–1.15 kg
palm-1 N, 0.50–1.35 and 0.30–0.50 kg palm-1 P, 1.25–1.50 and 1.25–1.50 kg
palm-1 K, 0.01–0.02 and 0.00–0.02 kg palm-1 B for plantation and smallholder
BMP plots respectively, and 0.08–0.16 palm-1 Mg for plantation BMP plots (Mg

69
CHAPTER 4

fertilizer was not applied in smallholder BMP plots). The general

recommendation for smallholders was approximately 6 kg palm-1 NPK 10–10–
30, which delivers 0.6 kg palm-1 N, 0.26 kg palm-1 P and 1.49 kg palm-1 K.
Average recommended application rates on smallholder farms were smaller than
on plantations because of the project’s assessment of farmer’s attitude to risk
(due to the time-lagged effect of fertilizer on yield, no immediate economic
returns are expected). To lower the threshold in purchasing fertilizer products,
50% of the costs for smallholder farmers were paid for by the project in all years.
• Right time. All fertilizers were applied during the short and long rains (March-
July and September-November). Urea, TSP, RP, and KCl were applied in two
split applications to optimize nutrient recovery and compound fertilizers in three
split applications.
• Right place. All fertilizers were applied over the edge of the palm circles and the
frond stack, apart from urea, which was applied in the weeded circle to reduce
volatilization losses, and borate, which was applied in a band 1 m wide and 1 m
from the palm trunk. Compared with palm circles, root development in the frond
stack is more favoured due a larger nutrient supply and a higher water
conservation as a result of the accumulation of organic debris (Bachy, 1964;
Fairhurst, 1996; Purvis, 1956; Tailliez, 1971). Fertilizers are also less susceptible
to surface runoff and are washed into the soil under the frond stack where water
infiltration rates are greater than in soil beneath the path and the weeded circle.
In order to control noxious woody growth in smallholder and plantation BMP blocks,
we aimed to increase the competiveness of soft weeds and grasses that otherwise do not
establish under poor soil fertility, with an integrated approach to ground cover
management. We recommended a combination of manual control (slashing, uprooting),
chemical control with tryclopyr (GarlonTM) herbicide, the introduction of legume cover
plants (Pueraria phaseoloides), and improving soil fertility (to overcome acute soil
phosphorus (P) deficiency) with a one-off application of RP (0.5 t ha-1, ~65 kg ha-1 P)
at project start.
For BMP blocks at plantations, mulching with empty fruit bunches (EFB) was
recommended at 30–40 t ha-1, and repeated once fully decomposed (i.e. every two
years). If there was a shortage of EFB, we focussed application on sloping BMP blocks
to reduce erosion and to conserve soil moisture. However, the amounts of nutrients
applied were too small to include in the overall nutrient budgeting.
4.2.6 Data analysis
All agronomic data was recorded and collated in OMP, an agronomic database designed
for oil palm (Agrisoft Systems, 2018). Production data was recorded at monthly

70
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

intervals but all data was summarized on an annual basis. We first present an overview
of the agronomic management procedures in BMP and REF plots and then analyze
differences between treatments according to year with a nested UNIANOVA or paired
sample t-test where appropriate. The UNIANOVA consisted of (i) yield components,
(ii) harvest cycles, (iii) harvester productivity and harvesting labour parameters, (iv)
field upkeep labour, (v) palm stand, and (vi) leaf nutrient concentrations as dependent
variables, and production system, site and treatment as factors. The model design
consisted of a main effect Production system, a nested effect Site within Production
system, and an interaction effect of Treatment with Production system:
y = Production system + Site(Production system) + Treatment + Production system x Treatment

Production system contained two levels: i) plantations and ii) smallholder farms, each
with three (BOPP, Norpalm, TOPP) and two (SWAPP East, SWAPP West) sites and
two treatments (BMP, REF) respectively. For plantations, we also compared BMP
yields with neighbouring blocks as a ‘second’ control (of the same year of planting,
planting material, soil type), since it is likely that REF blocks did not provide an
absolute reference as they too were undergoing improvement, due to a gradual and
unavoidable adoption of BMPs by plantation management.
Despite random allocation of BMP and REF treatments within each paired plot at
plantations and smallholder farms, initial starting differences in e.g. yield, can be
expected. Thus, we also evaluated the magnitude of change achieved with each
treatment between years using a two-tailed t-test.
Second, we applied linear regression to identify which variables correlated most with
yield. We used total annual yield (t FB ha-1) as dependent variable and yield taking and
yield making parameters as independent variables. For this analysis, we investigated
only BMP components whose effect on yield was direct and not time-lagged. For yield
taking practices we investigated the number of harvest cycles, harvester productivity
(t FB man-day-1, bunches man-day-1, ha man-day-1), harvesting labour (man-days ha-1
harvest cycle-1) and field upkeep labour (man-days ha-1), and for yield making practices
we investigated leaf nutrient concentrations (%DM), the number of pruning rounds and
palm stand (productive palms ha-1). The linear regression was performed separately for
oil palm plantations and smallholder farmers across all years. We started by entering all
variables in the model and removed the least significant parameters (P<0.05) one at a
time until we had a reasonably small model. Where applicable, we tested for squared
transformations and interactions to see if this would improve the model. All statistics
were performed using IBM SPSS Statistics Version 24.

71
CHAPTER 4

4.3 Results
4.3.1 Yield components and harvesting cycles
The BMP approach allowed for a comparison of fruit bunch yields with a baseline (pre-
BMP), between treatments (BMP vs. REF) and between project years. The baseline
fruit bunch yields for plantations averaged 11.1 t ha-1 in BMP blocks and 10.4 t ha-1 in
REF blocks. No baseline data was available for smallholders since they did not have
historical harvesting records. Despite this, the average fruit bunch yield at smallholders
at project start was estimated at about 5.3 t ha-1.
The average fruit bunch yield with BMP at plantation sites increased with 2.1 t ha-1
(+19%) in year 1 to 13.2 t ha-1, compared with an increase of 1.0 t ha-1 (+10%) to
11.4 t ha-1 in REF plots. The increase in yield can largely be attributed to improvements
in crop recovery (yield taking). The difference between treatments in year 1 was
1.8 t ha-1 (+16%) (Fig. 4.2, Table 4.1). In year 2, BMP yields were 14.6 t ha-1, and
significantly (P≤0.05) larger (2.9 t ha-1; +25%) than yields obtained in REF plots, which
averaged 11.7 t ha-1. Yields increased in year 3 for both treatments (P≤0.05), with BMP
plots at 17.9 t ha-1 and REF yields at 14.5 t ha-1, a significant difference (P≤0.05) of
3.4 t ha-1 (+23%). Compared with the baseline yields, average yields with BMP
increased with 6.8 t ha-1 (+61%) over a three-year period, whilst the increase with REF
was 4.1 t ha-1 (+39%). The difference in yield between year 1 and 3 can be attributed to
yield making, and is estimated at 4.7 t ha-1 (+36%) with BMP and 3.1 t ha-1 (+27%)
with REF. The largest improvements with BMP were at the Norpalm site. Average fruit
bunch yields with BMP were 3.6 t ha-1 (+47%) greater than yields obtained with REF
in year 1, 4.7 t ha-1 (+53%) greater in year 2 and 8.3 t ha-1 (+58%) greater in year 3. The
overall increase in yield from year 1 to 3 was 11.2 t ha-1 (+100%) with BMP and
6.5 t ha-1 (+85%) with REF. At the BOPP site, only small improvements were made
with BMP, whilst at TOPP, fruit bunch yields with BMP increased consistently each
year (Appendix 9).
There were no significant differences (P≤0.05) in average fruit bunch yield between the
REF and second control plots at plantations sites for all years, indicating minimal
influence of BMP on REF treatments. However, plantation management and farmers
improved the management of REF plots as their knowledge of better practices
increased. While BMP yields increase, REF plots often increase as well, even if
fertilizer is not applied and implementation of BMPs is less rigorous than in BMP plots.
Across smallholder sites, treatment differences between the average fruit bunch yield
were significant (P≤0.05) in all years and larger than on the plantations. Average fruit
bunch yields with BMP increased with 4.7 t ha-1 (+89%) in year 1 to 10.0 t ha-1,
compared with an increase of 2.0 t ha-1 (+38%) to 7.3 t ha-1 in REF plots, a result of an

72
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

increase in crop recovery (yield taking). The difference between treatments in year 1
was 2.7 t ha-1 (+37%). In year 2, BMP yields increased (P≤0.05) with 3.1 t ha-1 (+31%)
to 13.1 t ha-1, whilst the increase on REF plots was smaller at 1.6 t ha-1 (+22%),
averaging 8.9 t ha-1. The difference between treatments in year 2 was 4.2 t ha-1 (+47%).
In year 3, both BMP and REF yields increased significantly (P≤0.05) compared with
year 2. BMP yields increased with 4.5 t ha-1 (+34%), and averaged 17.6 t ha-1, whilst
REF yields increased with 2.8 t ha-1 (+31%) to 11.7 t ha-1. The difference between
treatments in year 3 was 5.9 t ha-1 (+50%) (Fig. 4.2, Table 4.1). Compared with the
estimated baseline yield for smallholders (5.3 t ha-1), average yields with BMP
increased with 12.3 t ha-1 (+232%) over three years, whilst the increase with REF was
6.4 t ha-1 (+121%). Approximately 7.6 t ha-1 (+76%) of the increase in yield with BMP
can be attributed to the effects of yield making and 4.4 t ha-1 (+60%) with REF. The
overall increase in fruit bunch yields with BMP was greater at SWAPP West
(+9.0 t ha-1; 88%) compared with SWAPP East (+6.2 t ha-1; 64%). The largest difference
between treatments occurred in year 3 at SWAPP East (6.0 t ha-1, +61% in year 3) and
in year 2 at SWAPP West (5.8 t ha-1, +60%) (Appendix 9). The REF treatment at
SWAPP West also showed a considerable increase in yield of 6.2 t ha-1 (+85%) by the
end of year 3, indicating the adoption of BMPs by farmers as their knowledge increased.

Fig. 4.2. Twelve-month rolling fruit bunch yields at plantation blocks, and b) average cumulative
monthly yields for the first 12 project months and twelve-month rolling yields for project months 12–
36 on smallholder farms. Average yields are uncorrected and include standard deviations (SD)) for
BMP and REF treatments. The dashed vertical lines indicate 12 month periods.

Differences in yield between treatments were largest in the plateau (PYP) and declining
yield phase (DYP) at plantations and in the steep ascending (SAYP) and plateau yield
phase (PYP) on smallholder plots (Fig. 4.3). However, more samples are required for
each treatment and growth phase to understand the underlying causes. Furthermore,
yields are still in a strongly upward mode at the end of the project (Fig. 4.2), suggesting

73
CHAPTER 4

that yields in the plateau phase are greater than reported, especially in the smallholder
fields (Fig. 4.3).
Larger (increases in) yields with BMP were partly explained by more frequent
harvesting and complete crop recovery after the installation of proper access (weeded
circles and paths, pruning) in BMP plots. On the large plantations, harvesting events
(cycles yr-1) were significantly more (P≤0.05) frequent at BMP plots compared with
REF for all years, whilst at smallholder sites the difference in harvesting events was
smaller with each consecutive year (Table 4.2). Improved crop recovery resulted in a
greater number of harvested bunches (bunches ha-1) and larger average bunch weight
(kg) at BMP plots (except at plantations year 1, where ABW was higher at REF plots),
particularly at smallholder sites where access with BMP was better than in the REF
plots (Table 4.1).

Fig. 4.3. Average BMP and REF yields with standard deviations (SD) for plantation blocks and
smallholder farms according to growth phase (steep ascending yield phase (SAYP), plateau yield phase
(PYP) and declining yield phase (DYP)) for project years (Yr) 1, 2 and 3. The dashed line shows the
water-limited yield profile (Yw) of oil palm in Ghana (averaged across favourable (areas with a mean
annual water deficit <250 mm) and optimal (areas with a mean annual water deficit <150 mm)
production sites across Ghana (Chapter 2)) plotted against year after planting (YaP) (Chapter 3).

74
 Table 4.1. Averages of yield components on plantation blocks and smallholder farms for BMP and REF treatments, project years (Yr) 1, 2 and 3. The final
column indicates whether a significant interaction of Treatment with Production system was found (*for significant at P≤0.05, NS for not significant; see
Appendix 10 for model parameters), whilst a post-hoc test (LSD) shows where significant differences occur in the interaction effect.
Oil palm plantation blocks Smallholder farms
Parameter Units Yr Yr Model
BMP (n=16) REF (n=16) BMP (n=19) REF (n=19)
-1 d
Bunch number bunches ha 1 1,381 1,210 1 1,217 936d NS
d
2 1,243 1,051 2 1,265 957d NS
3 1,351 1,144 3 1,448 1,181 NS
Av. bunch weight kg 1e; g 10.3 10.4 1e, f; g 8.7 8.1 NS
2 12.8 12.4 2e 11.6 10.2 NS
3e; g 14.7 14.3c 3f; g 13.3 11.4c NS
-1
Loose fruit collection t ha loose fruit 1 0.49 0.79c 1e; g 0.47 0.41c NS
2 0.65 0.80 2e 0.89 0.66 NS
3 0.82a 1.01 3e; g 1.51a, d 0.93d *
Fruit bunch harvest t ha-1 FB 1e; g 12.8a 10.6c 1e; g 9.5a, d 6.8c, d NS

75
f; h b
2 13.9 10.9b, c 2e; h 12.3d 8.2c, d NS
e, f; g, h b
3 17.1 13.5b, c 3e; g, h 16.1d 10.8c, d NS
-1 e; g a
Total yield** t ha 1 13.2 11.4c 1e; g 10.0a, d 7.3c, d NS
f; h b
2 14.6 11.7b, c 2e; h 13.1d 8.9c, d NS
e, f; g, h b
3 17.9 14.5b, c 3e; g, h 17.6d 11.7c, d NS
a,b,c,d
Indicates a significant difference at P≤0.05 between treatment means.
e,f
Indicates a significant difference between project years for BMP treatment means at P≤0.05.
g,h
Indicates a significant difference between project years for REF treatment means at P≤0.05.
** Sum of loose fruits and bunches.
CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES
CHAPTER 4

4.3.2 Harvester productivity and harvesting labour

At plantation sites, harvesters’ output in terms of weight of bunches (t man-day-1 FB)
and number of bunches (bunches man-day-1) was lower at BMP plots for all years. On
average, harvesters also significantly (P≤0.05) covered less area (ha man-day-1) in BMP
plots, compared with REF (Table 4.2). This is most likely related to shorter harvesting
intervals at BMP plots where less crop (including loose fruits) is expected to be
harvested during each harvesting event.
At smallholder sites, harvester output (t man-day-1 FB and bunches man-day-1) was
slightly higher at BMP plots compared with REF, whilst there were no differences in
ground covered between treatments. Because the number of harvesting cycles for BMP
and REF plots were similar, the difference is likely due to better access in BMP plots
where more crop was harvested at each harvesting event (bunches ha-1) (Table 4.1,
Table 4.2). The greater bunch availability due to yield improvement therefore more than
offset the effect of shorter harvesting intervals on harvester productivity.
The labour allocated to harvesting (man-days ha-1 cycle-1) did not differ between
treatments at plantation and smallholder sites. However, harvesting labour increased
significantly (P≤0.05) from year 1 to 2 at smallholder sites for BMP and REF
treatments, which was a result of an increase in yield (Table 4.1, Table 4.2).
4.3.3 Field upkeep
Field upkeep activities were recorded at the BOPP and Norpalm plantations but not at
TOPP. On average, field upkeep labour (man-days ha-1 yr-1) did not differ between
treatments at plantation and smallholder sites for all years (Table 4.2). However,
significantly more (P≤0.05) labour was spent at BMP smallholder sites in year 1
compared with other years, particularly to provide in-field access with circle and path
weeding and interline and selective weeding. After providing access, field upkeep
activities became less intensive with each consecutive year.
At BOPP, total field upkeep labour was largest in year 1, with most labour allocated
towards providing in-field access (circle and path weeding) and access on terraces
(category ‘other’), as well as the construction of silt pits for water conservation on BMP
plots (‘other’), whilst small differences in field activities were observed between
treatments and years at Norpalm (Fig. 4.4). At smallholder sites, most labour was
allocated to providing access with weeding activities, as well as sowing legume cover
plants to improve the ground cover vegetation (‘other’), particularly in the first year
(Fig. 4.4). Total field upkeep labour activities decreased with each year at smallholder
sites.

76
 Table 4.2. Averages of management parameters on plantation blocks and smallholder farms for BMP and REF treatments, project years (Yr) 1, 2 and 3.
The final column indicates whether a significant interaction of Treatment with Production system was found (*for significant at P≤0.05, NS for not
significant; see Appendix 10 for model parameters), whilst a post-hoc test (LSD) shows where significant differences occur in the interaction effect.
Oil palm plantation blocks Smallholder farms
Parameter Units Yr Yr Model
BMP (n=16) REF (n=16) BMP (n=19) REF (n=19)
-1 a, b b, c a, d c, d
Harvest cycles cycles yr 1 43 27 1 26 21 *
2 46a, b 29b, c 2 24a 22c *
3 42a, b 26b 3 25a 23 *
Harvester productivity t man-day-1 FB 1e 1.1a, b 1.5b, c 1e 0.7a 0.6c *
2 1.3 1.5c 2 1.0 0.7c NS
3e 1.5 1.8c 3e 1.2d 0.8c, d *
bunches man-day-1 1 121 154c 1 92 80c NS
2 102 120 2 104 86 NS
3 109 133c 3 104 84c NS

77
ha man-day-1 1 1.3b 1.9b, c 1 1.0 1.2c NS
2 1.1b 1.7b, c 2 1.1 1.2c NS
3 1.1b 1.7b, c 3 0.9 0.9c *
Harvesting labour man-days ha-1 cycle-1 1 0.34a 0.34c 1e, f; g, h 0.60a 0.70c NS
2 0.34a 0.37c 2e; g 0.66a 0.72c NS
3 0.37a 0.43c 3f; h 0.67a 0.76c NS
Field upkeep labour** man-days ha-1 yr-1 1 1.73 1.80 1e, f 1.93 1.45 NS
2 1.49 1.72 2e 1.24 1.23 NS
3 1.39 1.39 3f 1.12 1.29 NS
**Only includes plantation data for BOPP and Norpalm (n=11 for BMP and n=11 for REF), since field upkeep practices were not recorded at TOPP. Labour
for field upkeep activities was taken as the average across five categories, i) access (circle & path weeding), ii) drainage, iii) interline & selective weeding,
iv) pruning & frondstacking, and v) other.
a,b,c,d
Indicates a significant difference at P≤0.05 between treatment means.
e,f
Indicates a significant difference between project years for BMP treatment means at P≤0.05.
g,h
Indicates a significant difference between project years for REF treatment means at P≤0.05.
CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES
CHAPTER 4

Only a small amount of labour was allocated to drainage at plantation and smallholder
sites. Particularly at plantation sites, where large areas were located in valley bottoms,
crop recovery activities were obstructed by poor drainage due to lack of drainage outlets
and field drains, or drains that were too shallow and required desilting.
4.3.4 Leaf nutrient concentrations and fertilizer application
There were no significant (P≤0.05) differences in leaf nutrient concentrations between
treatments at plantation and smallholder sites at project start and end (Fig. 4.5). Average
leaf nutrient concentrations for N, P, Mg, and calcium (Ca) fell within their optimum
nutrient ranges (Fairhurst and Mutert, 1999), suggesting no nutrient deficiencies.
Because of the synergism between N and P uptake, leaf P concentration was assessed
in relation to leaf N concentration (Fairhurst and Mutert, 1999; Ollagnier and Ochs,
1981). The critical leaf P concentration (calculated based on Tampubolon et al. (1990))
fell within the optimum range for P for plantations and smallholders for both treatments
and years, suggesting a balance in leaf P and N concentration. Average leaf K
concentration was within optimum range at project start at plantation sites, but deficient
at all sites at project end, whilst average leaf Mg concentration was sufficient at both
plantation and smallholder sites for all years. However, when taking into account the
relative concentrations of the leaf cations (TLC, calculated according to Foster (2003)),
the average leaf K concentration (K as % of TLC) was deficient at plantations and
smallholders for all years, whilst the average leaf Mg concentration (Mg as % of TLC)
was only sufficient at smallholder sites. Average leaf B concentrations were deficient
at project start at plantation and smallholder sites, but sufficient at project end at both
treatments, most likely due to application with B fertilizers at BMP plots (Fig. 4.5, Fig.
4.6).

78
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

Fig. 4.4. Total labour (man-days ha-1) spent on field upkeep activities on plantation and smallholder
sites for project years 1, 2 and 3.

79
 CHAPTER 4

80
Fig. 4.5. Box plots showing leaf nutrient concentrations at oil palm plantations (n=11 for BMP and n=11 REF) and smallholder farms (n=18 for BMP and
n=18 REF) at project start and end. The vertical red line shows the average leaf nutrient concentration and the blue shaded area indicates the optimum leaf
nutrient concentrations (Fairhurst and Mutert, 1999). Critical P (Crit. P) is leaf P concentration assessed in relation to Leaf N concentration. Outliers are
plotted as individual points. Significant differences (at P≤0.05) between treatment means and years are indicated with a, b, c, d, e, f, g, h.
 Fig. 4.6. Box plots showing fertilizer nutrients (elements) N, P, K, Mg, B applied at oil palm plantations (n=15 for BMP and n=15 REF) and smallholder
farms (n=18 for BMP and n=18 REF) in project years (Yr) 2 and 3. Average applied rates of application are indicated with a red vertical line, and average
recommended rates of application for BMP plots are indicated with a green vertical line. Outliers are plotted as individual points.

81
CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES
CHAPTER 4

Whilst fertilizer nutrient applications were significantly larger (P≤0.05) in the BMP
treatment for N, P, Mg (year 2 only) and B at plantation sites and N, P, K at smallholder
sites, BMP fertilizer recommendations were not accurately implemented. As a result,
large nutrient gaps between what was recommended and actually applied were
observed, particularly at plantation sites (Fig. 4.6). Failure to implement fertilizer
recommendations partly explains the large variability in leaf nutrient concentrations
(Fig. 4.5).
4.3.5 Palm stand
At plantation sites the average number of productive palms (i.e. mature, immature and
supply palms) was significantly less (P≤0.05) than the initial planting density for both
treatments in all years, whilst at smallholder sites no significant differences were found
(Fig. 4.7). Between treatments, no significant differences (P≤0.05) in the number of
productive palms were found for all years for plantation (n=16 for BMP, n=16 for REF)
and smallholder sites (n=10 for BMP and n=10 for REF). However, infilling was only
successful at Norpalm where large vacant clusters (≥3 consecutive palms) were
replanted at project start in the BMP treatment. By the end of year 3 the average palm
stand with BMP improved greatly, with most supplied palms in production, partially
closing Yield Gap 2 (Fig. 4.7). At SWAPP West smallholder sites, the number of
productive palms was close to the initial planting density for both treatments, mainly
because experimental plots were selected based on complete and productive palm
stands. Experimental plots at smallholder sites therefore contained less unproductive
palms (abnormal or dead palms) which were more common at plantation blocks. Yield
Gap 2 was therefore smaller at smallholder sites compared with plantations.
4.3.6 Yield determinants
Using linear regression across oil palm plantations, three predictors explained 69% of
the variance in yield (R2=.69, F(3,68)=51.876, P=0.000); the number of harvest cycles
(ß=0.266, P=0.000), harvester productivity (t FB man-day-1, ß=6.289, P=0.000) and
harvesting labour (man-days ha-1 harvest cycle-1), ß=20.429, P=0.000), whilst at
smallholder farmers, 83% of the variance (R2=.823, F(3,100)=160.190, P=0.000) was
explained by harvester productivity indicators t FB man-day-1 (ß=9.369, P=0.000) and
ha man-day-1 (ß=-8.148, P=0.000)) and by leaf P concentration (%DM) (ß=78.341,
P=0.000). The linear regression results suggest the importance of crop recovery
activities such as frequent harvesting events (i.e. short harvest intervals) and improving
access for a more efficient harvester productivity (i.e. more bunch weight harvested and
less area (ha) covered per harvester) at both plantations and smallholders, and the larger
importance of improved nutrition at smallholder farms.

82
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

Fig. 4.7. Palm census results expressed as proportion of productive palms (palms ha-1) in relation to the
initial planting density (palms ha-1, see Table A1 for initial planting densities) for BMP and REF
treatments at plantation blocks and smallholder farms for project year 1, 2 and 3. Error bars represent
standard deviations.

83
CHAPTER 4

4.4 Discussion
Best Management Practices (BMP) were applied successfully in a West African context
using Ghana as a case study. The BMP approach provided the means for systematic
identification and elimination of yield gaps in mature plantings. The three-year process
started with a review of current yields and agronomic standards and an estimate of the
yield potential (Yw) by age of palm and site. Field audits were a useful method to
determine the causes of yield gaps (agronomic and management factors) and to identity
the respective corrective measures (Chew and Goh, 2003; Fairhurst and Griffiths, 2014;
Goh et al., 2004). The results present a case for using BMPs as a technique for yield
intensification on land already planted with oil palm, rather than expanding oil palm
plantings into areas that can be used for other purposes.
4.4.1 Increasing yields with BMP
In this research, we present oil palm yields measured over a period of 36 months to
demonstrate the potential of BMP to increase oil palm yields. Within this time frame,
large increases in yield were gained with improvements in bunch weight and number.
Yield improvement with BMP was partitioned in two distinct phases; i) the short-term
(≤1 year) effect of improvements in operational management of crop recovery (‘yield
taking’), and ii) the longer-term (>1 year) effect of improved agronomic management
(‘yield making’) on reducing stress (Fig. 4.1). However, the full benefits of BMP on
yield may be masked by abiotic stresses, such as periods of moisture stress, which were
not assessed in this study (Fairhurst and Griffiths, 2014). In contrast, biotic stress
events, such as pests (e.g. leaf miner (Coelaenomenodera spp.), rhinoceros beetle
(Oryctes rhinoceros)) and diseases (e.g. Fusarium spp.), were regularly monitored
during the project and did not pose a significant risk.
Average fruit bunch yield increases with BMP were 2.1 t ha-1 (+19%) and 4.7 t ha-1
(+89%) with crop recovery at oil palm plantations and smallholders respectively, whilst
yield increases of 4.7 t ha-1 (+36%) and 7.6 t ha-1 (+76%) were obtained with yield
making. Average yields and yield gains with BMP were larger at plantations and
smallholders, compared with the REF treatment over the project period. At plantations,
average yields in year 3 with BMP were 17.9 t ha-1, compared with 14.5 t ha-1 for REF
(3.4 t ha-1 yield difference), whilst the overall yield gain was 6.8 t ha-1 (+61%) and
4.1 t ha-1 (+39%) for BMP and REF respectively. At smallholder farms, however, the
response with BMP was greater compared with plantations. Average yields with BMP
in year 3 was 17.6 t ha-1, compared with 11.7 t ha-1 for the REF treatment (a difference
of 5.9 t ha-1). The overall yield gain was 6.4 t ha-1 (+121%) with REF, whilst the
response with BMP was almost double, at 12.3 t ha-1 (+232%).

84
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

The larger yield response at smallholder farms was largely due to the very poor initial
field conditions, where the benefits from yield taking BMPs where greater than at
plantations. Important crop recovery activities included more frequent harvesting
events (10-day harvest intervals) and improved field access (roads, paths, weeded
circles) to provide the means for increased harvester efficiency and productivity.
Because of the time-lagged effect of improved agronomy on yield, the contribution of
individual yield making components are more difficult to quantify. Fertilizer use and
leaf analysis data (Fig. 4.5, 4.6) suggest considerable nutritional constraints that must
be addressed and implemented correctly to intensify yields. While past research has
generated important nutrient management strategies, there are still considerable
knowledge gaps which could help our understanding of the contribution of nutrients to
yield gap closure (Tiemann et al., 2018). With the oil palm sector expanding into new
frontiers, including marginal and degraded lands, more work on agronomic needs,
including nutrients, of currently used commercial planting materials as well as new
materials now being bred will be needed to determine optimal fertilizer rates to
maximize yields (Tiemann et al., 2018). Development of new improved fertilizer
recommendations is therefore essential to increase yields and to sustainably intensify
oil palm production in Ghana. The need to establish multi-factorial, multi-locational
nutrient response trials across different agroecological zones in Ghana is therefore
essential to guide future fertilizer recommendations.
To capture the full beneficial effect of BMP, a period of at least 4 years is recommended
until yields start to plateau as all initiated inflorescences have reached maturity. Whilst
average plantation yields seem to have plateaued already within the time-frame of the
project, average yields at smallholder farms are still in a strong upwards mode,
indicating that the site yield potential has likely not yet been reached at the project sites
(Fig. 4.2). The response to yield making BMP’s on smallholder farms is therefore
expected to be larger than reported in our results. Whilst the design of the project did
not allow us to explore yields beyond 36 months, we were able to monitor the phase
where the most rapid changes in yield were expected, hence allowing us to answer
questions related to short-term yield trends and their drivers.
Rehabilitating neglected or abandoned oil palm requires significant initial investments
to achieve complete crop recovery and eliminate all agronomic constraints, particularly
in the first year when additional labour is required to establish proper access to the
palms (Fig. 4.4). Once unimpeded access was achieved, the greatest additional costs
were fertilizers. However, at project start, most smallholder farmers lacked financial
inputs and were unwilling to purchase inputs. Lack of access to credit coupled with the
time-lagged response to fertilizer use in terms of yield response inhibits fertilizer use
by smallholders and is a major reason for low yields at smallholder farms (Corley and

85
CHAPTER 4

Tinker, 2016). In the BMP project, half of the major inputs were therefore financed for
smallholders, since the goal of the project was to provide evidence of the potential
increase in profit should BMPs be applied. Once farmers were convinced with the
benefits, introducing and applying BMPs became easier. However, most smallholder
farmers are not in a position to make significant investments in fertilizer and other
inputs, particularly given the four-year payback period. A key input is therefore the
provision of credit, but banks are always reluctant to lend to farmers that lack land titles.
Instead, the provision of inputs to smallholder oil palm growers is best carried out by
milling companies that advance fertilizer materials to smallholders secured against
budgeted future crop deliveries to the respective mill.
In theory, short-term improvements to economic returns at smallholder farms can be
achieved by simply improving crop recovery, which can help to finance inputs needed
for the yield making phase. Over the long term, fertilizer use is considered essential to
maximise yield and to avoid depleting the low fertility status soils found in the oil palm
belt. For example, nutrient depletion in soils under cocoa without the use of mineral
fertilizers is a major factor in the decline of cocoa production throughout Ghana
(Kongor et al., 2018; Appiah et al., 1997). For policy makers, there is an obvious trade-
off between intensifying production with mineral fertilizer use on existing plantings and
allowing continued expansion of low-yield oil palm smallholdings that results in forest
destruction.
At plantations, the implementation and maintenance of BMPs was, at times, constrained
by lack of financial resources (e.g. insufficient budget provision for fertilizer inputs),
conflicts with, or willingness to implement program recommendations (e.g. installation
of a proper drainage system), and labour constraints for field maintenance and/or
harvesting (Appendix 11). Whilst smallholder farms are generally family operated
(Mensah-Bonsu et al., 2009), plantations, on the other hand, rely solely on the use of
hired labour. Plantations therefore tend to involve higher labour costs, are often plagued
by labour disputes between workers and management and appear less adaptable to
short-term changes than small-scale diversified systems (Gyasi, 1996). Shortage of
labour during the peak crop months mean that harvesting intervals often become
extended to >20 days (i.e. <3 harvest cycles month-1), resulting in significant crop loss.
Moreover, the high turnover in labour force at plantations results in a scarcity of skilled
labour needed to maintain good standards for particular field operations such as
harvesting and pruning.
The economics of closing yield gaps indicated by field audits and ex-ante analysis
should be considered before implementing BMPs to determine whether the cost of
remedial measures will be repaid in increased productivity over the improvement cycle.

86
 CLOSING YIELD GAPS IN OIL PALM PRODUCTION SYSTEMS IN GHANA THROUGH BEST
MANAGEMENT PRACTICES

Implementing BMPs in a representative sample of blocks provides proof of present

yield gaps and the cost and time period required to close the gaps. If the BMP trial
shows that yield gaps can be closed profitably, wide-scale adoption can be implemented
(Fairhurst and Griffiths, 2014). However, an economic analysis is beyond the scope of
this work and should be dealt with in a separate analysis.
4.4.2 Yield intensification versus area expansion
Most of the increases in the production of palm oil in Ghana have been achieved through
area expansion (Chapter 3). By contrast, increasing yields with BMP offers scope to
reduce the requirement for future area expansion to meet the increasing demand for
palm oil (preferably if tenera planting material is used). At present, approximately
327,600 ha is under oil palm cultivation (16,600 ha under plantations and 311,000 ha
under smallholders) with current FB yields (11.4 t ha-1 at plantations and 7.3 t ha-1 at
smallholders) resulting in production of 189,240 t FB and 2,270,300 t FB at plantations
and smallholders respectively. With moderate BMP implementation (increasing FB
yields to 17.9 t ha-1 and 17.6 t ha-1 at plantations and smallholders respectively),
production would increase to 5,770,740 t FB, thus avoiding 448,273 ha area expansion
at present yields. However, at potential production levels (i.e. increasing FB yields to
21.0 t ha-1 at plantations and smallholders), about 597,636 ha land can be spared.
Additionally, closing yield gaps in Ghana under current land area has the potential to
increase fruit bunch production almost three-fold from 2,5 Mt to 6,9 Mt. If all crop is
processed at an oil extraction rate (OER) of 21%, approximately 1.4 Mt CPO can be
produced (worth about 1 billion US$ at US$750 t-1 CPO-1). This is more than enough
to meet Ghana’s current annual demand of 106,000 t CPO, and does not require the
need to plant additional land (Chapter 3).
Provided there is sufficient milling capacity, yield intensification can be achieved
without the typical capital expenditure required for new plantings (e.g. road
infrastructure, planting cost), and financial returns from investments in yield
intensification accrue more rapidly because production starts to increase as soon as
agronomic constraints are removed (Donough et al., 2009; Fairhurst and McLaughlin,
2009). Ghana has large swathes of land planted to tenera palms that could be
rehabilitated. Instead of investing in new plantings that are likely to up end as
abandoned plots, due to lack of know-how, rehabilitation of existing mature plantings
may instead be an important policy for sustainable oil palm development in Ghana and
West Africa.
Increasing yields does not necessarily reduce area expansion, unless supporting policies
are in place and properly enforced, but is an important step towards reducing pressure
on land and deforestation (Angelsen, 2010; Woittiez et al., 2017). Higher yields will

87
CHAPTER 4

also make the crop more profitable, and if demand is elastic, expansion is likely to be
encouraged (Corley and Tinker, 2016).
4.5 Conclusions
Best Management Practices (BMP) offer immense potential to increase yields on
mature plantings at plantations and smallholder farms in Ghana. As such, implementing
BMP’s on neglected oil palm fields is a step-wise process and should be implemented
in the correct sequence in order to eliminate all agronomic stress effectively. Most
important is to first provide unimpeded access for harvesting operations and palm
upkeep with the installation of harvest paths and weeded circles, and removal of
unproductive fronds with corrective pruning. The second step involves the introduction
of regular and complete harvesting events at 7–10 day intervals to ensure complete crop
recovery. After yield taking constraints have been completely removed, yield making
operations should focus on: i) improving nutrient management and soil conservation
with balanced mineral nutrition and mulching with crop residues, ii) manual and
chemical removal of woody weeds in palm inter-rows and harvest paths to favour the
establishment of soft weeds, grasses and legume cover plants (LCP), iii) improving
drainage in swampy areas by installing ‘V’ shaped drains, and iv) regular patrols to
monitor outbreaks of pests (e.g. leaf miner (Coelaenomenodera spp.) and rhinoceros
beetle (Oryctes rhinoceros) and diseases.
Significant yield improvements were achieved at sites representative of the Ghanaian
oil palm industry, thus proving the applicability of the BMP process, which has also
been successfully implemented in Southeast Asia and Latin America. The BMP process
allowed for a systematic elimination of yield gaps in mature plantings by i) diagnosing
agronomic constraints, ii) identifying and interpreting causal factors, and iii)
implementing steps for corrective action. Improved field access, short harvesting
intervals and balanced palm nutrition were essential to yield intensification, particularly
at smallholder farms. The success of BMPs depends on total commitment and support
from senior plantation management and smallholder farmers alike, as well as sufficient
budget provision and resources (particularly in the first year) to implement BMPs
diligently and on time.
4.6 Acknowledgements
We thank the management and staff of Benso Oil Palm Plantation (BOPP), Norpalm
Ghana Ltd., Twifo Oil Palm Plantations (TOPP) and Solidaridad West Africa for their
support and the provision of data. We also thank Chris Donough, Thomas Oberthür,
Rob Verdooren, Anthony Whitbread, and the Oil Palm Research Institute (OPRI). This
work was supported by Canpotex International Pte Limited and K+S Kali GmbH.

88
 CHAPTER 5

The influence of water and nutrient management on oil

palm yield trends on a large-scale plantation in Ghana

This chapter is accepted with modifications as:

Rhebergen, T., Fairhurst, T., Giller, K.E. and Zingore, S. 2019. The influence of water and
nutrient management on oil palm yield trends on a large-scale plantation in Ghana.
Agricultural Water Management.
CHAPTER 5

Abstract
Oil palm (Eleais guineensis Jacq.) production in West Africa is limited by poor soils and suboptimal
climate. The potential to improve fruit bunch (FB) yields with appropriate water and nutrient
management in Ghana is thus considerable. To assess the effects of water and nutrient management on
oil palm yields, a 2 x 2 factorial randomized complete block design with irrigation and fertilizer was
implemented for three years in the main oil palm production area of Western region, Ghana. Treatments
included i) zero irrigation + zero fertilizer (control), ii) zero irrigation + fertilizer, iii) irrigation + zero
fertilizer and iv) irrigation + fertilizer, all replicated four times. Fertilized treatments received annual
nutrient application rates of 2.0 kg palm-1 nitrogen (N), 0.4 kg palm-1 phosphorus (P), 2.5 kg palm-1
potassium (K), 0.3 kg palm-1 magnesium (Mg) and 0.15 kg palm-1 boron (B). Irrigation (using
sprinklers) was scheduled using tensiometers, and was given to meet the daily evaporation of oil palm
(~450 l palm-1 day-1, or 6.5 mm day-1). Average attainable yields of 32.6 t ha-1 FB were achieved with
irrigation and fertilizer, which was 4.7 t ha-1 greater than the control (27.9 t ha-1) and 4.1 t ha-1 greater
than with irrigation alone (28.5 t ha-1). Fertilizer was therefore essential for a maximum response to
irrigation. Whilst yield increases were mostly attributed to increases in bunch number, large variations
in monthly yield persisted at all treatments despite the use of irrigation, with production peaks typically
coninciding with the onset of the long rain season (March–July). Whilst a number of factors may
contribute (e.g. vapour pressure deficit), the process(es) regulating yield cycles remain unclear and
require further research.
Keywords: Irrigation, fertilizer, water deficit, yield cycle, yield gap.

90
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

5.1 Introduction
Although oil palm (Eleais guineensis Jacq.) is indigenous to West Africa, the soils and
climatic conditions in the region are considered suboptimal for its cultivation (Chapter
2). In southern Ghana, soils are predominately coarse-textured, strongly weathered and
highly leached Acrisols and Ferralsols (USDA: Ultisols and Oxisols respectively) with
low pH and poor soil fertility status (Buringh, 1979). The major climatic constraint is
the prolonged dry season of 4–5 months with mean annual water deficits up to
600 mm yr-1 (Chapter 2; van der Vossen, 1969).
A recent land suitability assessment showed that large areas within the oil palm belt of
Ghana are susceptible to annual water deficits approaching 400 mm (Chapter 2). There
is a time lag of 35–40 months between the occurrence of abiotic and biotic stress events
and yield response in oil palm, which makes it difficult to unravel the effects of
environmental factors on fruit bunch (FB) yield (Breure, 2003; Corley and Tinker,
2016). Several authors have attempted to explain yield responses in relation to the
annual water deficit in West Africa. For example, Hartley (1988) reported a 10% yield
reduction with every 100 mm of annual water deficit. Caliman and Southworth (1998)
observed that moisture stress negatively affected floral initiation (1–3%), sex
differentiation (3–4%) and abortion rate (8–10%) per 100 mm of deficit, with a
concomitant yield reduction of 8–10% per 100 mm deficit in the first year and 3–4% in
the second. A yield reduction of 40–50% was recorded when severe water stress
occurred in the previous year. In the Ivory Coast, Olivin (1968), reported a yield
reduction of 10–20% with 100 mm annual water deficit, depending on the soils’ water
storage capacity, while Ochs and Daniel (1976) reported similar yield penalties based
on observations made in the Ivory Coast, Benin and Cameroon. In Ghana, yields ranged
from 22–25 t ha-1 FB for areas with annual water deficits of 150 mm, 16–18 t ha-1 FB
for areas with annual water deficits of 250 mm, and 6–7 t ha-1 FB for areas with an
annual water deficit of 400 mm, suggesting an almost linear relationship between yield
and water deficit (Danso et al., 2008). Therefore, there are strong prospects for reducing
the negative impact of moisture stress on yields in areas affected by large annual water
deficits in Ghana.
Supplementary water supply through irrigation provides an opportunity to increase oil
palm yields in regions with sub-optimal rainfall (Corley and Hong, 1982). Prioux et al.
(1992) obtained maximum yields of 22 t ha-1 FB with irrigation in Ivory Coast, a 21%
increase in production over non-irrigated palms, while yields increased four-fold with
irrigation in a trial described by Desmarest (1967).
In addition to moisture stress, oil palm yields in Ghana are limited by poor soil fertility
and inadequate nutrient supply (Chapter 3). Analysis of soil and plant samples from

91
CHAPTER 5

trial sites in Ghana showed severe and widespread deficiencies of soil phosphorus (P)
and potassium (K) on large plantations and smallholder farms (Chapter 3). In addition,
several trials in West Africa confirmed that K is deficient on sedimentary soils, P on
basement complex-derived soils, and that Mg may be needed on acid, sandy soils
(Corley and Tinker, 2016).
The potential to improve oil palm production with appropriate water and nutrient
management in Ghana is thus substantial and could greatly increase profitability for oil
palm producers, as well as sparing land for nature. In this chapter, we evaluated the
interaction of irrigation and nutrient application on the yield of oil palm on a large-scale
oil palm plantation over three years in Western region, Ghana. The objectives were: (i)
to quantify yield gaps caused by water and nutrient constraints and (ii) to analyse the
interaction of water and nutrients on yield trends and yield components.
5.2 Methods
5.2.1 Study area
A trial was established to assess the effects of irrigation and fertilizer on oil palm yields
at the Norpalm Ghana Ltd. plantation (NGL) (4°55’29.04”N; 1°53’31.75”W, 11 masl),
which is located in the Ahanta West District of the Western region, Ghana. NGL’s
nucleus estate consists of approximately 4,500 ha distributed between the Sese (500 ha)
and Pretsea (4,000 ha) sites (Fig. 5.1).
Slopes at NGL are gentle to hilly (4–40 %) and soils are classified as Gleysols in poorly
drained valley bottoms and flood plains and Lixisols on sloping terrain (Fig. 5.1).
Rainfall at NGL is bimodal, with long rains from March-July and short rains from
September-November. Average rainfall from 2011–2016 was lower at the Sese site
(1,136 mm) than at Pretsea (1,511 mm), with considerable variability between months
and years. As a result, the average annual water deficit (calculated using the method of
Surre (1968)) over the same period was higher at Sese (608 mm) than in Pretsea
(355 mm) (Fig. 5.2).

92
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

Fig. 5.1. Location of Norpalm Ghana Ltd. and its soil type and terrain properties.

93
CHAPTER 5

Fig. 5.2. Monthly rainfall (a) and the water balance (b) at the main plantation (Prestea) and Sese for
2010–2016. Black and grey lines show the average for Prestea and Sese respectively. Water balance
values below the zero line in graph (b) are water deficits while positive values are water surplus.

5.2.2 Trial setup and management

The trial was established at the Sese plantation on February 2015 in a 34 ha block that
had been planted in 2002 (Fig. 5.1). The trial block was planted at 143 palms ha-1 using
triangular spacing, and consisted entirely of tenera (i.e. dura x pisifera (DxP)) palms of
Pobé, Benin material, which had been planted 13 years before at the time of trial
establishment. The topography was flat.
An isometric map of the trial block was produced, including the number of mature,
immature, new/supply, abnormal, and dead/removed palms, or unplantable points, and
was used to delineate the trial plots. Each plot consisted of clusters of 36 palms (6 x 6),
of which the inner 16 palms (4 x 4) were used for taking measurements (recording
palms) and the outer palms provided a guard row (Fig. 5.3). The recording palms
consisted entirely of productive and healthy palms, while the guard row included some
abnormal and immature palms. Each recording plot (4 x 4) was approximately 0.11 ha.
All plots were bounded by a trench of at least 50 cm deep to avoid nutrient and water
movement between trial plots and access from bordering palms. At least two palm rows
were also kept as a buffer between plots. The entire trial block was managed with “Best
Management Practices” (BMP) (Donough et al., 2009), to ensure that yields were not
constrained by other factors besides nutrients and water.

94
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

5.2.3 Experimental design

The trial was established at the Sese plantation division because of its proximity to a
river for irrigation. The trial was designed to quantify the bunch yield potential (Yp)
without moisture stress, and to investigate the interaction of nutrient and water supply
on oil palm yield components. A 2 x 2 factorial randomized complete block design was
used, which included four treatments: i) zero irrigation + zero fertilizer (I0F0, i.e.
control), ii) zero irrigation + fertilizer (I0F1), iii) irrigation + zero fertilizer (I1F0) and
iv) irrigation + fertilizer (I1F1). Each treatment was replicated four times, with blocks
following moisture gradients. In total 16 plots were installed (Fig. 5.3).

Fig. 5.3. Experimental layout of the irrigation x fertilizer trial at Sese, Norpalm Ghana Ltd drawn on
isometric paper. The coloured plots indicate the separate blocks (n=4), whilst the numbers indicate the
treatment (1=irrigation + zero fertilizer (I1F0), 2=irrigation + fertilizer (I1F1), 3=zero fertilizer + zero
irrigation, i.e. control (I0F0), 4=zero irrigation + fertilizer (I0F1)). The insert shows the experimental
layout within an irrigated plot. The red and yellow lines are irrigation pipes (main and laterals
respectively) and the pink dots the sprinkler locations. The green dots are the outer palms (guard row)
and the brown dots the recording palms.

Fertilizer treatments covered the main limiting nutrients: nitrogen (N), phosphorus (P),
potassium (K) and magnesium (Mg), applied as urea, triple superphosphate (TSP),
muriate of potash (MOP) and kieserite respectively. Annual application rates of
nutrients (elemental form) were 2.0 kg palm-1 N, 0.4 kg palm-1 P, 2.5 kg palm-1 K and

95
CHAPTER 5

0.3 kg palm-1 Mg. In addition, all fertilizer treatment plots received an annual
application of 0.15 kg palm-1 boron (B), to overcome B deficiency. Fertilizer application
rates were based on previous fertilizer experiments, and were drafted to ensure nutrients
were not limiting. Fertilizers were applied to all palms in each fertilizer treatment plot,
including the guard row. P, K and Mg fertilizers were broadcast over the edge of the
weeded circle into the frond stack, whilst B was applied with sodium borate in a band
1 m wide and 1 m from the palm trunk, and urea in the weeded circle to reduce
volatilization losses. All fertilizers were applied during the rainy seasons. Sodium
borate was applied in March, urea in April and August, MOP in May and September,
TSP in June and kieserite in July. Fertilizer application rates and timing followed the
same schedule each year and no mill residues (e.g. empty fruit bunches (EFB)) were
applied to the palms during the experiment.
Irrigation needs were based on the soil water potential (kPa), which was measured using
tensiometers. Tensiometers were installed in each block (n=4) at 20 cm depth, where
most of the fine roots are located (Taillez, 1971). Because the soil was relatively coarse-
textured, the critical soil water potential for irrigation was set at 20 kPa (Alam and
Rogers, 1997). Irrigated plots consisted of 12 sprinklers (Senninger ® SmoothDrive),
each with a radius of 9.3–10.2 m. Each sprinkler thus covered three palms (see insert
Fig. 5.3). At an operating pressure of 207 kPa, approximately 150 minutes of irrigation
was given to meet the daily evaporation of oil palm (~450 l palm-1 day-1, or
6.5 mm day-1) (Carr, 2011; Corley and Tinker, 2016), which was recorded during each
irrigation event. To ensure that the right quantity of water was supplied, a flow meter
was installed in-line with the irrigation discharge pipe.
In order to evaluate the effect of irrigation in eliminating soil-moisture stress, we
calculated the water balance for irrigated and non-irrigated plots following the method
of Surre (1968):
𝐵 = 𝑆<=8 + 𝑃 − 𝐸𝑇𝑃 Eq. 5.1
where B = monthly water balance (mm); Sres = residual soil water from the previous
month (mm); P = monthly rainfall (mm); and ETP = monthly potential
evapotranspiration (mm).
Scorching of palm leaves was briefly experienced in the dry season in the first year due
to tidal saline water incursions, and a corrective measure was taken to use EC meters to
monitor the salinity of the irrigation water at each irrigation event. No irrigation was
given when EC exceeded 3000 µS cm-1 to prevent scorching of the palm leaves and
ground cover (Fipps, 2003). Piezometers were installed in eight plots (two per
replicate), and monitored weekly to assess the groundwater level. Tensiometers were

96
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

removed from the field during periods of flooding, which occurred with heavy rainfall
in March, June and October 2016, and May and June 2017.
5.2.4 Data collection
Crop production
Bunch number, fruit bunch weight (kg) (including loose fruits) and number of male
flowers were recorded at harvest intervals of 7 days at each plot. Crop production data
were reported as the total for each plot (inner 16 palms). Fruit bunch weight was
measured with a tripod and digital scale and bunch production data was used to derive
yield totals (t ha-1 FB). Bunch number and the number of male flowers were used to
determine the sex ratio, i.e. the ratio of female to total inflorescences.
Leaf sampling
Palm nutritional status and fertilizer recommendations for mature oil palm are
commonly made based on leaf and rachis nutrient concentrations from frond 17, due to
their high correlation with yield (Fairhurst, 1996; Foster and Prabowo, 2006). Leaf and
rachis samples for nutrient determination were therefore taken annually at each
recording palm (inner 16 palms) at a distance of approximately two thirds between the
insertion point of the first true leaves on the leaf petiole and the distal end of frond 17
(Chapman and Gray, 1949; Fairhurst et al., 2004). Composite leaf and rachis samples
were produced for each plot and analysed at Yara Laboratory (UK) for N using a
combustion analyser (Dumas technique), and P, K, Mg, and B (only leaf) by inductively
coupled plasma analyser (ICP), with averages derived per treatment.
Physiological measurements
Physiological measurements were taken to investigate the effects of drought and water
stress on oil palm development and production. The following effects of drought or
water stress on physiology and yields of oil palm are known to occur (Corley and Hong,
1982):
i. longer immature period (Nouy et al., 1999),
ii. stomatal closure and a decrease in transpiration from the leaves (Ng, 1972; Rees,
1961),
iii. delayed leaf opening, decrease in frond production rate, and the accumulation of
unopened leaves (spears) in the crown (Nouy et al., 1999; Ochs and Daniel,
1976; Paramananthan, 2003),
iv. vegetative damage and palm death with severe drought (Maillard et al., 1974),
v. longer flowering time (i.e. the time interval between the opening of a leaf and
flowering of the inflorescence in its axil), shorter maturation time (i.e. the time
between anthesis and bunch ripening) (Nouy et al., 1999; Olivin, 1966), delayed

97
CHAPTER 5

fruit abscission (Chan, 1972), and delayed bunch ripening (Ochs and Daniel,
1976),
vi. decrease in sex ratio and increases in male inflorescence number (Corley, 1976),
vii. irregular distribution of yield with a shorter production peak (Carr, 2011),
viii. bunch and early inflorescence abortion (Paramananthan, 2003), and
ix. reduction in fruit bunch oil content (Ochs and Daniel, 1976; van der Vossen,
1974).
Physiological measurements included maturation time and number of unopened fronds,
and were carried out at weekly intervals on each individual recording palm (inner 16
palms) with averages derived per plot. Maturation time was measured as the time
interval (days) between anthesis and harvesting (ripeness). All bunches were labelled
with wire tags embedded in the flower to record the date and frond position at anthesis
and at harvest. Newly emerging bunches were monitored routinely on each palm during
each recording event. Crop production data including bunch number, and the number
of male flowers were used to determine the sex ratio and monthly yield distribution for
each treatment.
Calculation of yield gaps
We adjusted the yield gap model presented in Chapter 3 to include the nutrient-limited
yield (Yn). Because the trial was managed with BMP and only complete (productive)
palm stands were used, production level Ymey (the yield reduced by deficiencies in
plantation establishment) and Yam (the yield reduced by inadequate agronomic
management) were not included in the model. The average yield achieved with I1F1 in
year 3 served as a proxy for the potential yield (Yp), the average yield with I0F1 in year
3 provided an estimate of oil palm production assuming optimal nutrient management
and served as a proxy for the water-limited yield (Yw), while the I1F0 treatment served
as a proxy for the yield limited by nutrients only (Yn) and I0F0 for the water- and
nutrient limited (Yw+n). The difference between average yields achieved with Yp and
Yw therefore gives us an estimate of the yield gap caused by water (Yield Gap 1 (YG1))
and was calculated as:
𝑌𝐺1 = 𝑌𝑝 − 𝑌𝑤 Eq. 5.2
while nutrient-limited yield gaps (Yield Gap 2 (YG2)) were calculated as:
𝑌𝐺2 = 𝑌𝑝 − 𝑌𝑛 Eq. 5.3
The gap caused by both water and nutrient limitations (Yield Gap 3 (YG3)) was
calculated as the difference between Yp and Yw+n:
𝑌𝐺3 = 𝑌𝑝 − 𝑌𝑤 + 𝑛 Eq. 5.4

98
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

5.2.5 Data analysis

All data was summarized on an annual basis unless indicated. Differences between
treatment means were analysed with a UNIANOVA for each year, including yield
components (yield, bunch weight and bunch number) and leaf and rachis nutrient
concentrations as dependent variables, and replication and treatment as factors.
In oil palm, there are three periods where production is sensitive to water stress
(Caliman and Southworth, 1998; Dufour et al., 1988); i) 30–33 months before
harvesting, which corresponds to the sex differentiation period, ii) 19–24 months before
harvesting, which corresponds to the fast growth of the leaf supporting the
inflorescence, and iii) 7–13 months before harvesting, which corresponds to the period
of abortion. For each water-stress sensitive period, we used simple linear regression
with fruit bunch yield as the dependent variable and the water balance as the
independent variable, to determine which months were most significant in predicting
yield for each treatment. We then entered the most significant months in a multiple
linear regression model for each treatment and removed the variables that did not make
a significant (P<0.05) contribution to the regression. All statistics were performed using
IBM SPSS Statistics Version 24.
5.3 Results
5.3.1 Yield components
The pre-trial annual yield was 17.7 ha-1 FB across the whole block. In year 1, average
FB yields were 6.9 t ha-1 greater at 24.6 t ha-1 in the control treatment (I0F0), compared
with 7.0 t ha-1 greater in the I0F1 treatment (24.7 t ha-1), 6.3 t ha-1 in the I1F0
(24.0 t ha-1) and 6.6 t ha-1 larger in the I1F1 treatment (24.3 t ha-1) (Table 5.1). We
attribute the larger yields in year 1 to the fact that trial plots consisted entirely of
productive palms as well as improved crop management including more frequent and
complete harvesting events. Average FB yields increased in year 2 and 3, with the I1F1
treatment yielding significantly more than other treatments, at 32.6 t ha-1. The large FB
yield increase with I1F1 suggests that there was a significant irrigation x fertilizer
interaction. Average bunch weight (ABW, kg) decreased with each year for all
treatments, whilst bunch number per palm (BN) increased. The increase was largest for
the I1F1 treatment, where bunch number increased from 7.5 to 11.1 bunches palm-1
(+48%). Bunch number per palm depends on the frond production rate, the sex ratio,
the abortion rate and the bunch failure rate (Carr, 2011). The increase in bunch number
coincided with a decrease in the number of male flowers and an increase in the sex ratio
for all treatments (Table 5.1), whilst increases in bunch number with irrigation are
mostly due to changes in both abortion rate and sex ratio (Corley and Hong, 1982).

99
CHAPTER 5

Monthly yield components fluctuated widely for all treatments, with most notably the
clear seasonality in bunch number (Fig. 5.4). In mature palms, variation in bunch
number contibutes much more to yield cycles than does bunch weight, with sex
variation commonly regarded as the prime cause of yield variation (Corley and Tinker,
2016; Henson and Dolmat, 2004). In regions with high and regular rainfall (e.g. parts
of North Sumatra, Indonesia and Sabah, Malaysia), sex ratios tend to vary little
throughout the year, whilst in areas with a marked dry season (West Africa), sex ratios
fluctuate widely (Adam et al., 2011). Peaks in bunch number and yield, and to a lesser
extent sex ratio, typically conincided with the onset of the long rain season (March–
July, Fig. 5.2).
In year 3, the average number of bunches ha-1 month-1 was 120 for the I0F0 treatment,
124 for I0F1 and 128 and 132 for I1F0 and I1F1 respectively. Most bunches were
harvested in February with 228 bunches (16% of the annual bunch production) for the
I0F0 treatment, 273 (18%) bunches for I0F1, 226 (15%) for I1F0 and 239 (15%) for
I1F1 (Fig. 5.5). The peak months were February–May where 55% of the annual number
of bunches were harvested for the I0F0 treatment (an average of 199 bunches ha-1
month-1, opposed to an average of 81 bunches ha-1 month-1 during the other months),
53% for I0F1 (196 bunches ha-1 month-1 vs. 88 bunches ha-1 month-1), 49% (186
bunches ha-1 month-1 vs. 98 bunches ha-1 month-1) for I1F0 and 45% for I1F1 (179
bunches ha-1 month-1 vs. 108 bunches ha-1 month-1), indicating a slightly lower peak for
irrigated treatments.
The coefficient of variance (COV) for monthly fruit bunch production was 56% for the
I0F0 treatment, 55% for I0F1, 51% for I1F0, and 48% for the I1F1 treatment, suggesting
a slightly more regular annual bunch number production with irrigation. For
comparison, Prioux et al. (1992) obtained a COV of 30% with irrigation, opposed to
60% without irrigation, suggesting irrigation was essential in regulating monthly
production distribution.

100
 Fig. 5.4. The relative response of three-month rolling averages for a) average bunch weight (ABW, kg), b) bunch number (BN, # ha-1) and c) sex ratio (-)
for all treatments to the control (I0F0) (bold horizontal line). The three-month rolling averages for the control plot (I0F0) are indicated with a red solid line
on the right y-axis, showing the actual response.

101
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA
CHAPTER 5

Fig. 5.5. Annual bunch number distribution (in %) for all treatments in year 3.

5.3.2 Water balance

I1F0 and I1F1 treatments received 656 mm of water with irrigation in year 1, 711 mm
irrigation in year 2 and 496 mm irrigation in year 3. As a result, the annual water deficit
in year 1 was 735 mm and 96 mm for non-irrigated and irrigated plots respectively,
709 mm and 80 mm in year 2, and 735 and 331 mm in year 3 (Table 5.1). The deficits
were largest in the months August, September and December–March, which coincided
with dry spells.
Irrigation events were succesfully planned using tensiometers, but water deficits could
not be fully erradicated due to pump failures and periods when the irrigation water was
too saline to be used.
Simple linear regression indicated that 31 months before harvest, 20 months before
harvest and 8 months before harvest were the three most sensitive periods to water stress
in the I0F0 treatment, whilst 30 months, 19 months, and 7 months before harvest
accounted for most of the observed variation in yield for all other treatments (Fig. 5.6,
Table 5.2).

102
 Table 5.1. The annual water deficit (WD, mm) and averages with standard deviations (SD) for fruit bunch yield (FBY, t ha-1) and yield components,
including average bunch weight (ABW, kg), bunch number (BN, # palm-1), the number of male flowers (MF, # palm-1) and sex ratio determined at harvest
(SR, %) for different irrigation x fertilizer treatments after 12, 24 and 36 months. Statistics were performed with a nested UNIANOVA.
Months 0–12 Months 13–24 Months 25–36
Treatment WD FBY ABW BN MF SR WD FBY ABW BN MF SR WD FBY ABW BN MF SR
-1 -1 -1 -1 -1 -1
mm t ha kg # palm % mm t ha kg # palm % mm t ha kg # palm %
Zero 24.6 22.5 7.7 5.7 58 27.5 19.1 10.1 4.1 71 27.9 19.4 10.1 2.8 79
irrigation + (3.0)1 (1.1)2,3 (0.6)2,3 (0.7)2,3 (2.6)2,3 (3.5)a (1.3)a,2 (0.8)2 (0.2)2,4 (2.0)2,4 (2.5)c,1 (2.6)3 (0.7)3 (0.2)c,3,4 (1.3)a,3,4
zero
fertilizer
(I0F0) 735 709 735
Zero 24.7 22.4 7.8 6.5 55 31.5 20.6 10.7 4.4 71 27.9 18.9 10.5 3.2 77
irrigation + (1.6)1,2 (1.5)3 (0.9)2,3 (1.5)2,3 (4.5)2,3 (3.5)2 (1.8) (1.1)2 (0.6)2,4 (1.8)2,4 (2.6)d,1 (2.4)3 (1.5)3 (0.4)3,4 (2.5)3,4
fertilizer
(I0F1)

103
Irrigation + 24.0 21.1 8.0 6.4 56 29.1 19.0 10.7 4.2 72 28.5 18.6 10.7 3.6 75
zero (2.5)2,3 (1.5)1,2 (1.4)2,3 (0.9)2,3 (4.1)2,3 (3.9)2 (1.2)b,2 (1.0)2 (0.4)2 (1.5)2 (3.2)a,3 (1.2)1 (1.2)3 (0.9)c,3 (3.3)a,3
fertilizer
(I1F0) 96 80 331
Irrigation + 24.3 22.7 7.5 6.2 56 32.0 20.8 10.8 4.0 73 32.6 20.8 11.1 3.4 77
fertilizer (1.2)2,3 (1.2)1 (0.3)2,3 (1.9)2,3 (7.7)2,3 (3.1)a,2 (1.3)a,b,1 (1.0)2 (0.8)2 (1.8)2,4 (2.4)a,c,d,3 (0.9) (1.3)3 (0.3)3 (1.0)3,4
(I1F1)
a,b
Indicates a significant difference at P≤0.10 between treatment means.
c,d
Indicates a significant difference at P≤0.05 between treatment means.
1
Indicates a significant difference at P≤0.10 between project years.
2,3,4
Indicates a significant difference at P≤0.05 between project years.
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA
CHAPTER 5

Fig. 5.6. The relationship between three-month rolling averages for monthly fruit bunch (FB) yield
(t ha-1) and three periods of sensitivity to water stress before harvesting: a) 30–33 months before harvest
in which sex differentiation occurs, b) 19–24 months before harvest, which corresponds to the fast
growth of the leave supporting the inflorescence, and c) 7–13 months before harvest, which is the
aborting period (Caliman, 1998) for irrigated (I1F0, I1F1) and non-irrigated (I0F0, I0F1) treatments.

104
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

Additionally, multiple linear regressions suggested that the period corresponding to

bunch abortion (7–13 months before harvest) gave the least significant prediction of
yield for all treatments (Table 5.2). The main sensitive phase for drought effects was
estimated to be 30 months before bunch harvest (i.e. the period of inflorescence sex
determination) for all treatments except the control. This is in line with results reported
by Legros et al. (2009), but in contrast to results reported by Dufour et al. (1988) and
Caliman and Southworth (1998) where the period associated with bunch abortion had
the greatest impact on yield.
Repeating both analyses, including the effect of irrigation on the water balance, showed
a weaker coefficient of determination (R2), suggesting that irrigation did not influence
yield cycles in irrigated plots and that environmental factors other than rainfall may also
contribute to yield oscillations (Corley and Tinker, 2016).
5.3.3 Leaf nutrient concentrations
There were no significant differences (P≤0.05) between average leaf nutrient
concentrations for nitrogen (N), phosphorus (P), potassium (K), magnesium (Mg) and
boron (B) for all treatments and years, except in year 2 where leaf N concentration in
I1F1 and I1F0 treatments was significantly larger (P≤0.05) than the I0F0 treatment, and
where leaf K concentration in I1F1 was significantly larger (P≤0.05) than the I1F0
treatment (Fig. 5.7). All leaf nutrient concentrations increased from year 1 to 3, except
for N, where a decline was observed (significant (P≤0.05) for the I1F0 treatment). The
increase was significant (P≤0.05) for K for all treatments but I0F0, for Mg for the I0F0
treatment, and in all treatments for B. In year 3, N was within the optimal range for the
I1F1 treatment, whilst all other treatments were deficient. P was only sufficient for the
I0F0 treatment in year 3, whilst K was deficient in all treatments. All treatments were
sufficient in Mg and B, in year 3.
No significant differences (P≤0.05) between average rachis nutrient concentrations for
N, P, K and Mg were observed for all treatments and years, but for rachis N
concentration in year 3 between the I0F0 and I1F1 treatment and for rachis P
concentration in year 1 between the I1F0 and I1F1 treatment. Rachis N concentrations
declined for all treatments from year 1 to 3, whilst rachis P, K and Mg concentrations
increased for all treatments (significant (P≤0.05) for P and K, and for the I0F0 treatment
in Mg). Rachis N concentration was deficient for all treatments, whilst rachis P, K and
Mg were all sufficient in year 3.

105
 Table 5.2. Simple linear regression model parameters for all treatments for three water-stress sensitive periods, and multiple linear regression for all
treatments.
Model parameters
CHAPTER 5

Treatment Month
Simple linear regression Multiple linear regression
I0F0 -31 R2=.47, F(1,34)=29.868, P=0.000; y=2.226+0.007x R2=.71, F(2,33)=39.689, P=0.000; y=2.248+0.005(-31)+0.006(-20)
-20 R2=.48, F(1,34)=31.205, P=0.000; y=2.224+0.008x
-8 R2=.25, F(1,34)=11.395, P=0.002; y=2.237+0.006x
I0F1 -30 R2=.52, F(1,34)=36.130, P=0.000; y=2.346+0.007x R2=.70, F(2,33)=37.031, P=0.000; y=2.353+0.005(-30)+0.005(-19)
-19 R2=.42, F(1,34)=24.717, P=0.000; y=2.310+0.007x
-7 R2=.11, F(1,34)=4.127, P=0.050; y=2.305+0.003x
I1F0 -30 R2=.47, F(1,34)=30.149, P=0.000; y=2.249+0.007x R2=.69, F(2,33)=36.148, P=0.000; y=2.257+0.005(-30)+0.005(-19)
-19 R2=.46, F(1,34)=29.168, P=0.000; y=2.217+0.007x
-7 R2=.14, F(1,34)=5.528, P=0.025; y=2.215+0.004x
I1F1 -30 R2=.47, F(1,34)=30.301 P=0.000; y=2.445+0.008x R2=.69, F(2,33)=36.731, P=0.000; y=2.454+0.006(-30)+0.006(-19)
-19 R2=.47, F(1,34)=29.572 P=0.000; y=2.408+0.008x
-7 R2=.13, F(1,34)=4.878 P=0.034; y=2.404+0.004x

106
107
Fig. 5.7. Leaf (dark lines) and rachis (red lines) nutrient concentrations for irrigated (I1F0, I1F1) and non-irrigated (I0F0, I0F1) treatments for project years
1, 2 and 3. The blue shaded area indicates the optimum leaf nutrient concentrations (Fairhurst and Mutert, 1999) and the blue dashed line the critical rachis
nutrient level (Foster and Prabowo, 2006).
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA
CHAPTER 5

5.3.4 Physiological measurements

The maturation time and accumulation of unopened leaves (spears) in the crown are
direct physiological responses to water stress. Whilst no clear relationships were found
between time of maturation and the actual water balance (no time-lag included), data
suggests a shorter bunch maturation time and an increase in the accumulation of
unopened leaves in the crown following periods of water stress (particularly after the
long dry season from December–February) for the I0F1 treatment (Fig. 5.8). A shorter
maturation time is likely caused by an increase in ethylene production initiated by water
stress (Abeles et al., 1992), which could accelerate fruit ripening. It is known that
variations in maturation time and frond production rate exacerbate production
irregularities (Nouy et al., 1999). During the dry season, leaflets often accumulate as
unopened spears in the palm crown whilst the bunch maturation time is shorter. Frond
production rate therefore shows a clear seasonality in West Africa, with a minimum
during the dry season and a maximum immediately after the dry season (Corley, 1977).
Thus, after the rains have started, a flush of leaf opening is commonly observed (Carr,
2011). However, no significant differences (P≤0.10) were found for the average
maturation time between treatments for year 2.

Fig. 5.8. Three-month rolling averages for a) maturation time (the time between anthesis to ripeness in
days) for irrigated (I1F0, I1F1) and non-irrigated (I0F0, I0F1) treatments, plotted with the actual water
balance (WB) and b) the number of unopened fronds (# ha-1) for the I0F1 treatment plotted with the
water balance in reverse order. The graphs suggest a shorter bunch maturation time and an increase in
the accumulation of unopened leaves in the crown under periods of water stress, although no clear
correlations were found.

108
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

5.3.5 Yield gaps

Oil palm is in the plateau yield phase at 16 years after planting (YAP), with the nutrient-
limited yield (Yn) estimated at 28.5 t ha-1 FB, the water-limited yield (Yw) at
27.9 t ha-1 FB and the potential yield (Yp) at 32.6 t ha-1 FB. The average Yw was similar
to the average Yw in the plateau yield phase (26.6 t ha-1 FB) reported in Chapter 3, which
was determined from yield data across 23 trial sites in the oil palm belt of Ghana. Short-
term yield gaps caused by nutrients (YG2) were estimated to be 4.1 t ha-1 FB, while
yield gaps caused by water and water + nutrients (YG1, YG3) were both estimated to
be 4.7 t ha-1 FB.
5.4 Discussion
5.4.1 Water and nutrient-limited yield gaps
We investigated the short-term fruit bunch yield performance of mature oil palm under
several fertilizer and irrigation treatments. Because of the long time-lag between sex-
differentiation of the inflorescences and harvesting of ripe fruit, it is only possible to
begin to assess the effectiveness of irrigation and fertilizer after 28 months, provided it
is sustained over the whole time period (Carr, 2011). The full treatment response is
therefore only expected in the third year after trial implementation. Compared with the
control treatment (I0F0; 27.9 t ha-1 FB), the I1F1 treatment increased fruit bunch yields
by 4.7 t ha-1 while the effect of irrigation alone (I1F0) was 0.6 t ha-1, indicating that
fertilizer was critical for a maximum response to irrigation. This is similar to findings
reported by Tittinutchanon et al. (2008) where fertilizer was essential for maximizing
yields with irrigation. The benefit of adding fertilizer with irrigation was therefore
estimated at 4.1 t ha-1 FB, which is approximately 0.9 t ha-1 crude palm oil (CPO) at
21% oil extraction rate (OER).
Leaf nutrient concentrations are known to vary due to a wide range of factors, such as
rainfall, soil type and fertilization (Fairhurst and Mutert, 1999). Clear trends were found
in leaf and rachis nutrient concentrations in all treatments, including a decrease in N
concentration and an increase in all other nutrient concentrations from start to end of
the experiment. Whilst treatments did not have a clear significant effect on nutrient
concentrations by year 3, leaf nutrient concentrations for N, K and B were largest in the
I1F1 treatment and largest in the I0F0 treatment for P. By comparison, Cheng-xu et al.
(2011), found that water and nutrient stress decreased leaf N and P concentrations in oil
palm seedlings, but increased K concentration compared with irrigated treatments. Lee
et al. (2011), on the other hand, found lower leaf nutrient concentrations for N, P, K and
Mg in irrigated palms ≥5 YAP, but higher rachis nutrient concentrations, particularly
for K and B, and a larger storage of rachis P, K, S and B contents (due to a larger
increase in dry matter) (Lee et al., 2014). Lee et al. (2011) suggests that higher yields

109
CHAPTER 5

with irrigation demands larger amounts of plant nutrients to support optimal growth and
yield, which is reflected by lower leaf nutrient concentrations. Furthermore, leaching
losses of soluble nutrients K and Mg are expected to be higher with irrigation as well
(Lee et al., 2011).
Despite efforts to ensure nutrients were non-limiting with fertilization, leaf nutrient
concentrations for P and K were deficient in all treatments, whilst rachis P, K and Mg
concentrations were excessive, indicating a nutritional imbalance in the palm.
Particularly for P and K, rachis concentrations were largest with the I1F1 treatment,
supporting the hypothesis that palms grown under more favourable conditions have
higher rachis nutrient concentrations and are able to store larger amounts of nutrients
(Lee et al., 2014). Given the nutritional imbalance between the leaf and rachis, it is
therefore essential to consider the nutrients stored in the reserve tissues (i.e. rachis) as
well as the leaf while preparing fertilizer recommendations (Foster and Prabowo, 2006).
For example, to increase a deficient concentration of leaf P, if rachis P is satisfactory,
an increase in either fertilizer N or fertilizer K is generally effective, depending on
whether N or K is most deficient. To increase a deficient concentration of leaf K, if
rachis K is satisfactory, an increase in fertilizer N is generally effective (Foster and
Prabowo, 2006). Given these complex interactions, determination of the specific
nutrients and optimal application rates following a 4R Nutrient Stewardship (IPNI,
2012) approach with and without irrigation requires further research.
Because the yield in the control treatment (I0F0) improved from year 1 to year 3 (Table
5.1), it is likely that the actual yield response with irrigation and fertilizer was
underestimated. Yield improvements were most likely due to better management in trial
plots, but it is also possible that water and nutrient poaching from neighbouring plots
occurred, since oil palm roots can spread laterally up to 25 m from the trunk (Carr,
2011). The I0F0 average yield for year 1 therefore provides a better baseline for the
water- and nutrient limited yield, giving a YG3 of 8.0 t ha-1 FB. For comparison,
Tittinutchanon et al. (2008) reported fruit bunch yield increases of up to 10 t ha-1 (to
28 t ha-1) with irrigation and high rates of fertilizer in Thailand, in a climate similar to
that of Ghana. The increase in yield can be attributed to an increase in bunch number,
with little to no effect of bunch weight. Increases in bunch number were, therefore, most
likely caused by improvements in the sex ratio and a reduction in the abortion of
immature inflorescences.

110
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

5.4.2 Understanding mechanisms that determine yield fluctuations

Results from the multiple linear regression indicated that the period associated with
inflorescence sex determination was most sensitive to drought and likely to contribute
most to yield oscillations in Ghana (Fig. 5.6). However, because of the unique
vegetative structure of oil palm (stem and leaves) together with the long time-lag
between floral initiation and fruit bunch harvest, causal links between environmental
factors and yield are difficult to establish (Carr, 2011). Whilst the effects of water stress
on oil palm are well known, the extent to which the monthly yield cycles are affected
is poorly understood (Corley and Hong, 1982). The environmental factor most likely to
explain yield oscillations in regions with severe and regular annual dry seasons is the
soil water supply (Henson and Dolmat, 2004). However, because of the trial block’s
proximity to a river, it is likely that palms in all treatments had access to water in the
dry season due to a high-water table, which can modify the actual water deficit (Caliman
and Southworth, 1998). While piezometer readings indicate the existence of a water
table ≥2 m (which was the limit of measurement) at the onset of the long rains, it is
uncertain to which depth the palms extracted water exactly since no root profiles were
done. Observations from the Ivory Coast, for example, have shown that palms can
access water from the soil at a depth of at least 5 m (Caliman, 1992) and that large yields
can be obtained under drier conditions when the lack of rainfall is compensated for by
a high water table (Quencez et al., 1987). Given that all palms may have had access to
groundwater, this suggests that yield cycles were not solely regulated by the soil water
supply. In support of this, whilst irrigation ameliorated the water deficit and increased
fruit bunch yields, the use of irrigation did not affect yield cycles (Fig. 5.6). This is
consistent with findings reported by Chan et al. (1985), Foong and Lee (2000) and Kee
and Chew (1991), suggesting that yield cycles are controlled by factors other than
rainfall (Corley and Tinker, 2016).
Alternatively, Henson and Dolmat (2004) suggested that factors associated with dry
conditions other than the soil water supply, such as high temperature, high atmospheric
vapour pressure deficit and high evaporation rate may be involved in yield cycling; all
of which correlate well with rainfall. In particular, the stomata are considerably
sensitive to vapour pressure deficit; an increase in which causes the stomata to close,
even when there is a good water supply (Caliman, 1992; Carr, 2011). Stomatal closure
is therefore an important factor limiting photosynthesis and hence dry matter production
in oil palm (Dufrene et al., 1990). Smith (1989), for example, showed that high
atmospheric vapour pressure deficits may limit production even in parts of the world
where oil palm do not normally suffer from water stress. While it can be expected that
the use of sprinkler irrigation might overcome stomatal closure in dry air by wetting the
soil surface, and humidifying the atmosphere, no evidence has been provided yet

111
CHAPTER 5

(Tittinutchanon et al., 2008). It is therefore likely that atmospheric conditions were

similar in the trial, but measurements of stomatal conductance or vapour pressure deficit
are required to confirm this.
Annual cycles of meteorological variables other than rainfall may also contribute to
yield cycle patterns. Henson and Dolmat (2004) demonstrated that it was possible to
match oscillations in solar radiation and evaporation rate with annual yield using
appropriate time lags, although the relationship may not be causal. Additionally, Legros
et al. (2009), assumed that seasonal peaks of flowering in oil palm are controlled by
photoperiod response within a phytomer, even at oil palm sites close to the equator.
They further suggest these patterns are confounded with drought effects that affect fruit
bunch yield with long time-lags, but results remain to be confirmed. Using a modelling
approach, Combres et al. (2013) also suggested the involvement of photoperiod in
bunch production dynamics, whilst Dufrene et al. (1990) observed that sunlight and
vapour pressure deficit combined may explain variations in maximal yield between
west Africa and southeast Asia. It is therefore possible that present monthly yield may
be affected by multiple past stress events at different time lags before harvest, which
complicates attempts to understand the effects of climate.
In the absence of external constraints, it has been proposed that yield cycles are
regulated by endogenous mechanisms whereby current yield levels may impact future
bunch number through sex differentiation and abortion (Corley, 1977). However, it has
been argued that the time lags involved in these processes would not readily lead to
annual yield cycles, unless abortion levels were high and sex differentiation occurred at
a relatively late stage (Henson and Dolmat, 2004).
5.4.3 Managing oil palms under water stress
In West Africa, where rainfall patterns are characterized by a long and short rainy
season and a long and short dry season, irregular patterns in production are typical. For
all treatments ≥15% of the annual bunch production in year 3 was harvested in February,
with as little as ≤4% in low crop months (Fig. 5.5). Understanding local yield cycles of
oil palm is of economic importance, because the volume of fruit in the peak month
determines the size of the mill, and hence the scale of investment, needed by a plantation
(Corley and Tinker, 2016). Furthermore, irregular distribution of production has several
consequences for plantation management, including lack of milling capacity during
peak crop months and underutilization in the production trough, lack of sufficient
harvesters in the peak months resulting in extended harvest intervals and crop loss, and
fluctuations in fruit bunch oil content with in particular a sharp drop at the end of the
dry season (Caliman, 1992; Nouy et al., 1999). Forecasting future yields on a monthly
basis therefore helps plantation managers to better plan labour and milling requirements

112
THE INFLUENCE OF WATER AND NUTRIENT MANAGEMENT ON OIL PALM YIELD TRENDS ON A
LARGE-SCALE PLANTATION IN GHANA

and is also useful to marketers as forward selling of palm oil is common (Corley and
Tinker, 2016). Because climate is becoming increasingly unpredictable, the need for
simple predictive models to forecast future yields will therefore become more essential
in plantation management.
A practical way of reducing annual fluctuations in production is to plant a mixture of
planting materials with different yield patterns (Corley and Tinker, 2016) as
demonstrated in North Sumatra (Nouy et al., 1996) and Benin (Cros et al., 2013),
implying that genetic diversity partly regulates the annual bunch production profile in
oil palm, or by partially disbudding palms to shift yield peaks out of phase and reducing
yield fluctuations of the area as a whole (Corley, 1977). Management implications to
mitigate the effect of water stress on oil palm include the use of irrigation to correct
water deficits, planting in sites with a high water table, adequate drainage to remove
excess water, since prolonged flooding is known to reduce stomatal conductance and
gas exchange processes (photosynthesis and transpiration) and to kill young palms
(Lamade et al., 1998), preventing water runoff on sloping terrain (e.g. contour planting,
terraces, appropriate frondstacking techniques), improving structure in compact soils to
allow for better root development, mulching with organic materials (e.g. empty fruit
bunches) to reduce soil surface evaporation and removal of inflorescences on young oil
palms to stimulate root development and improve drought tolerance (Caliman, 1992;
Carr, 2011). Furthermore, potassium (K) is an essential macronutrient needed in large
amounts by oil palm and is directly involved in several physiological processes,
including transport of assimilates and stomatal regulation (Taulya, 2013). K is therefore
essential in avoiding water loss during water stress situations. Maintaining adequate
plant K with balanced fertilization is thus important for plant drought resistance (Wang
et al., 2013).
5.5 Conclusions
Results of the trial show the potential of irrigation and fertilizer to close yield gaps in
mature oil palm stands in Ghana. Average FB yields of 32.6 t ha-1 were obtained with a
combination of irrigation and fertilizer, while FB yields with irrigation only averaged
28.5 t ha-1 FB. The benefit of adding fertilizer with irrigation was therefore estimated
at 4.1 t ha-1 FB, which is approximately equivalent to 0.9 t ha-1 crude palm oil (CPO) at
21% oil extraction rate (OER). Fertilizer was therefore essential for a maximum
response to irrigation, but information on which nutrients are most important and
optimal application rates are lacking to ensure balanced nutrition in oil palm with
irrigation. While yield increases were mostly caused by increases in bunch number,
annual yield cycles persisted despite the use of irrigation. Irregular distribution of
production leads to lack of mill capacity and insufficient harvesters in the peak crop

113
CHAPTER 5

months. Many factors have been proposed that cause these fluctuations, both external
and endogenous, indicating a complicated interaction of several components. Yield
cycles and yield-determining process(es) therefore remain unclear and require further
research.
5.6 Acknowledgements
We thank the management and staff of Norpalm Ghana Ltd. for their support and
management of the trials. We also thank Chris Donough for his early comments on the
trial protocols , Yuval Levi (Dizengoff Ghana Ltd.), Wim Rhebergen and Paul Schofer
for their contributions in designing the irrigation system, Rob Verdooren for his
assistance with the statistics and Lotte Woittiez for her helpful comments. Special
thanks goes to Solidaridad West Africa for funding the trial.

114
 CHAPTER 6

General discussion
CHAPTER 6

6.1 Introduction
In the early 2000s, the oil palm (Elaeis guineensis Jacq.) sector in Ghana developed
extensively, partly due to a growing interest of investors to expand and develop the area
planted to oil palm. As a result, Ghana became the third largest producer of palm oil in
Sub-Saharan Africa in 2017 at 2.5 Mt yr-1 (next to Nigeria (7.8 Mt yr-1) and Cameroon
(3.1 Mt yr-1)) (FAO, 2019), with oil palm the second most important perennial crop in
the Ghanaian economy after cocoa (Ofosu-Budu and Sarpong, 2013). Despite large
increases in production the past 15 years, fruit bunch (FB) yields remained low (Fig.
6.1, FAO, 2019). With an estimated water-limited yield potential of 21 t ha-1 FB
(Chapter 3), and actual yields of 11.0 t ha-1 on plantations and 6.0 t ha-1 on smallholder
farms in Ghana (Chapter 3, 4), large yield gaps exist.

Fig. 6.1. Changes in the area (ha) under oil palm cultivation, bunch production (tonnes) and fruit bunch
yield (t ha-1) in the Ghanaian oil palm industry from 1990 to 2017 (FAO, 2019).

Although the oil palm’s origin lies in West Africa (WA), larger yields are achieved
outside its natural habitat, such as in Southeast Asia and Latin America (Fig. 1.1c).
While differences in climatic conditions (e.g. rainfall, solar radiation) and management
practices can be attributed to this yield variability, there have always been many
misconceptions on the yield potential and its key drivers in WA. Before we initiated our
work in Ghana, we had a poor understanding of the oil palm sector in the region, which
was partly due to the limited amount of published work on oil palm in WA the past two
decades. With this work, we therefore provide an up-to-date understanding of the state
of oil palm in Ghana, that can be useful to guide decisions and investments in the West
African region.

116
 GENERAL DISCUSSION

At the core of the project was obtaining a reliable estimate on the yield potential of oil
palm in Ghana. This was essential to analyse yield gaps and their drivers at plantation
and smallholder oil palm production systems, and to develop an approach for yield
intensification. Using early publications (e.g. van der Vossen (1969)) and (more recent)
production data from oil palm plantations in Ghana, we estimated the maximum yield
in the plateau phase (PYP) at 18 t ha-1 FB at project start. However, results from our
trials far exceeded expectations, with yields approaching 28 t ha-1 FB at the best
performing BMP sites. It was therefore fundamental to our research to develop a
framework for yield gap analysis in oil palm, to allow for a systematic diagnosis of
production constraints and to assess opportunities for increasing yields with Best
Management Practices (BMPs).
Using Ghana, this thesis aimed to elucidate the factors that currently limit oil palm
production in West Africa, by evaluating agronomic factors, and to adopt BMPs as a
measure for yield gap reduction. The agro-ecological conditions that determine the
suitability and availability for its cultivation are dealt with in Chapter 2, and causes for
low productivity amongst various production systems in Chapter 3. In Chapter 4, we
investigated the potential for yield intensification with improved agronomic practices,
and in Chapter 5 we assessed the effects of water and nutrient management on oil palm
yields. While detailed discussions are provided in each chapter, the main purpose of
this discussion is to integrate findings from the different chapters, and to generate
insights for sustainable intensification of the oil palm sector in a West African context.
6.2 Water and nutrients limit oil palm cultivation in Ghana
6.2.1 Water limitations
Compared with regions in Southeast Asia that receive high and regular rainfall (e.g.
parts of North Sumatra, Indonesia and Sabah, Malaysia), the climate in WA is marked
by regular dry seasons (Carr, 2011). In the oil palm belt of Ghana, for example, a short
dry spell occurs in August and the long dry season between December and February. A
key constraint to oil palm cultivation in Ghana, as in many other parts in WA, is
therefore the annual water deficit, which can result in a large Yield Gap 1 (Fig. 3.2).
This is likely to be aggravated in the future due to climate change induced droughts.
Water stress in oil palm reduces the yield potential and partly influences the distribution
of yield during the year. In countries such as Malaysia, seasonal changes are less
distinct, and yields are relatively uniform throughout the year. In contrast, in regions
where there is marked dry season, such as in Ghana, yield peaks are larger (Fig. 5.5),
which can result in several harvesting and processing implications (section 1.3).
Irrigation is often recommended as an option to mitigate the effect of water stress on oil
palm, and it is likely to become increasingly relevant due to projected increases in the

117
CHAPTER 6

frequency and intensity of droughts and due to oil palm cultivation expanding into drier
areas (Woittiez et al., 2017). While the effects of drought on oil palm are well known,
there is limited reliable data on actual yield responses to irrigation (Carr, 2011).
In the irrigation trial installed in Western region, Ghana, large yield increases with
irrigation and fertilizer were achieved, whilst seasonality in production persisted (Fig.
5.6). The average yield response to irrigation and fertilizer over a three-year period was
32.6 t ha-1 FB (Table 5.1). The yield gap caused by nutrients (YG2) was estimated at
4.1 t ha-1 FB and 4.7 t ha-1 FB for water. The response to irrigation alone was therefore
0.6 t ha-1 FB, indicating that fertilizer was essential for increasing yields with irrigation.
Whilst the scale of the trial was too small to allow a full cost-benefit analysis, assuming
high capital costs (US$2,500 ha-1) and running costs (US$150 ha-1 yr-1 (450 l palm-1
day-1 irrigation)) for sprinkler irrigation (Tittinutchanon et al., 2008; numbers corrected
for inflation), the yield increase with irrigation alone is likely not profitable, whilst
combined with fertilizer, irrigation is likely to provide considerable increase in profits.
Whilst fertilizer is essential with irrigation, fertilizer requirements supplemented with
irrigation need further investigation. Because the yield response is expected to be larger
with irrigation than without irrigation, fertilizer rates are likely to be larger as well.
Also, which nutrient(s) are critical with irrigation remains uncertain (Tittinutchanon et
al., 2008). Because the response to irrigation may differ between locations, depending
on the severity of moisture stress, and whether dry seasons occur annually (e.g. West
Africa), or at irregular intervals (e.g. Indonesia) (Carr, 2011), the costs and benefits of
different irrigation regimes (with and without fertilizer) under a range of environmental
conditions therefore need further investigation (Woittiez et al., 2017). Where irrigation
is not profitable, drought mitigation strategies, such as selecting sites with a good water
holding capacity, mulching with organic materials, and reducing plant populations can
be considered instead (Carr, 2011).
6.2.2 Nutrient limitations
A second constraint to oil palm production in Ghana is poor nutrient management. Soil
and leaf analysis data (Table 3.2, Table 3.3) showed that there are also considerable
nutritional constraints, particularly phosphorus (P) and potassium (K), that must be
addressed to close Yield Gap 3 (Fig. 3.2). This was mainly caused by low fertilizer
applications in plantations, whilst smallholder farms applied no fertilizers at all (Table
3.4). Particularly in Ghana’s drought-sensitive climate, K is likely to be critical because
it mitigates the impact of water-stress through osmotic adjustment (Taulya, 2013).
Improved K management on K-deficient soils is therefore likely to improve
productivity and enhance water use efficiency of Ghana’s rain-fed oil palm plantings.
Development of new improved fertilizer recommendations is therefore essential to

118
 GENERAL DISCUSSION

increase yields and to sustainably intensify oil palm production in Ghana. In Southeast
Asia for example, fertilizer trials have given planters useful information on the site-
specific estimates of fertilizer requirements for oil palm based on leaf nutrient values
from analysis. This information however, is currently still lacking for Ghanaian oil palm
production systems.
In this research, we assessed the nutritional status of the palm, as well as fertilizer needs,
by comparing leaf nutrient concentrations to reference leaf critical levels accessed
through literature. Whilst the use of leaf nutrient concentrations in oil palm fronds are
important diagnostic tools to assess nutrient requirements, leaf nutrient contents may
not have general applicability (Fairhurst et al., 2004). Optimum, or critical values for
individual nutrients can vary over a considerable range, depending on such factors as
the age of palms, soil moisture regime, soil type, ratio to other nutrient concentrations,
type of planting material, spacing, and inter-palm competition (Fairhurst and Mutert,
1999). For a more accurate and economical management of fertilization, it is therefore
desirable that the results of fertilizer trials are available for major soil types across the
oil palm belt so that the optimum leaf level can be defined in the light of the ecological
conditions, the nature and age of the palms and the most economic fertilizer rates (Ochs
and Olivin, 1977). In order to provide adequate fertilizer recommendations for future
oil palm production in WA, multi-factorial, multi-locational nutrient response trials
across different environments and production phases are required, as well as nutrient
omission trials to identify the nutrients that limit crop growth most. The interaction
between potassium nutrition and drought stress deserves particular focus as this
provides a mechanism for adaptation to climate change and to increase yields.
6.3 Increasing production with sustainable yield intensification
Despite persisting water and nutrient limitations, it was possible to achieve large yields
at plantations and smallholder farms through ‘Best Management Practices’ (BMPs),
with improvements in crop recovery (yield taking) and agronomy (yield making) (Table
4.1). Particularly, crop recovery was fundamental for achieving high yields, which was
achieved with improved field access at smallholder farms and more frequent harvesting
events at plantations. Because oil palm continuously produces bunches, complete crop
recovery must be maintained throughout the year to avoid crop losses. As soon as
harvest intervals extend beyond 10 days, there will be crop loss because a proportion of
the crop is either over-ripe or rotten at harvest (Fairhurst and Griffiths, 2014). However,
maintaining short harvest intervals is problematic for plantations, particularly in the
peak crop months when labour availability is short.
Most plantations in Ghana rely on the extensive use of hired labour (Gyasi, 1996),
which is usually sourced through contractors. During the lean season, the number of

119
CHAPTER 6

harvesters is reduced due to low bunch availability. Also, since harvesting-work is paid
by piece-rate, and low crop densities result in lower earnings, many workers chose to
work elsewhere instead. In contrast, more labour is required during the peak crop
months, in order to meet FB production. However, plantations often struggle to find
sufficient labour for harvesting during the peak crop months, resulting in significant
crop losses. Efficient use of labour for harvesting therefore poses many operational and
logistical challenges for plantations throughout the production cycle. A good balance
between labour requirements and expected crop losses must therefore be considered. At
the same time, seasonality in production poses other challenges, such as insufficient
milling capacity in the peak, and underutilization of the mill in the lean season due to
low crop harvests.
6.4 Labour limitations
A major reason to why BMPs were not fully implemented or sustained (Appendix 11),
was the high reliance on sufficient (skilled) labour to maintain good standards, which
was lacking at times throughout the project. Although oil palm plantations offer
abundant employment opportunities, the work is not very attractive, since the tasks are
perceived as arduous and dangerous, particularly harvesting (Ismail, 2013). The
isolation of plantation life, unattractive terms and conditions provided to workers (e.g.
low wages), and competitiveness from other employment opportunities in large-scale
agri-businesses and mining might also play a role in labour shortages.
A possible solution is to reduce labour requirements and workload through
mechanisation of certain field operations, such as harvesting and FB collection, to make
the work more attractive. Devices related to crop collection and transport (e.g.
mechanical grabber, compact transporter, mechanical buffalo, battery-powered
wheelbarrow, loose fruit collectors) and the development of lighter, powered and more
ergonomically efficient fruit cutters, increase labour productivity, and decrease the cost
of production, whilst improving the health and safety of the workforce (Anon, 2004;
Jayaselan and Ahmad, 2011; Murphy, 2014; Shuib et al., 2010; Teo, 2000). Currently,
little to no mechanisation is used in plantations in Ghana.
Although the mechanization of the oil palm sector has progressed well over the years,
there is still room for improvement (Shuib et al., 2010). R&D efforts on mechanization
should therefore be increased, particularly on the process of fruit bunch harvesting,
since this includes a number of steps that are difficult to mechanize (e.g. assessment of
FB ripeness, cutting a heavy fruit bunch and its subtending frond at a height of
10–20 m) (Murphy, 2014). Alternative efforts include redesigning the architecture of
the oil palm where advanced breeding might be used to develop high yielding palms
with greatly reduced height to facilitate access and harvesting of fruit bunches instead

120
 GENERAL DISCUSSION

(Arolu et al., 2016; Murphy, 2009). Short to dwarf palms may offer other potential
benefits as well, including higher planting densities (and thus larger yields), resulting
in more efficient land use, and a longer economic life due to a slower growth, thus
reducing costs associated with new plantings (Zulkifli et al., 2017). However, efforts to
produce palms with these traits are still ongoing.
Whilst mechanisation offers perspective for the oil palm industry in Ghana, it should be
pursued with caution. The oil palm industry provides the means to generate significant
economic and social development, by providing employment to a large number of the
rural population. Whilst oil palm plantations struggle with labour shortages, the
introduction of mechanization to reduce labour dependency could be perceived as a
threat to plantation workers. While this is true to a certain extent, more correctly,
mechanization is aimed at increasing productivity with the same number of workers, by
reducing the workload, so that the worker can work at a faster pace and cover a bigger
area (Anon, 2004; Shuib et al., 2010). The social context and sensitivity of the labour
force therefore needs careful consideration. Depending on how management
approaches the subject, local plantation workers could either embrace or dismiss
mechanization. Consultation, education and illustrations on the advantages of
mechanization take time. A well-planned and coordinated programme is therefore a
prerequisite to successful implementation of plantation mechanization (Anon, 2004).
Taking this into account, mechanization is likely to be more readily (socially) accepted.
6.5 Sustainable intensification in oil palm should focus more on increasing oil
yields
Whilst the BMP approach is designed to increase the yield of oil palm through the
production of more fruit bunches per unit of land, the final economic product is crude
palm oil (CPO). CPO yield depends on the quantity and weight of fruit bunches
harvested (and loose fruit collected), as well as the oil and kernel content of these
bunches. Yield intensification strategies should therefore not only focus on increasing
FB production, but also the oil extraction rate (OER) of FB.
Oil content is affected by a number of factors, including planting material (dura has an
OER of 16–18%, whereas the tenera hybrid has an OER of 22–30%), fertilization (Mg
fertilization increases oil content, while K fertilization (sometimes) decreases oil
content (Ochs and Ollagnier, 1977)), and is negatively correlated with rainfall, and
positively correlated with available radiation (Woittiez et al., 2017). The oil content
therefore depends on site-specific factors, and is to some extent influenced by
management practices. Research at a number of locations in Indonesia, for example,
showed a reduction in OER with BMP (up to 1%), but total oil and kernel yields were
still larger than standard estate practices due to better FB collection and reduced loose

121
CHAPTER 6

fruit losses (Donough et al., 2010; Oberthür et al., 2012). Whilst the effect of yield
taking and yield making effects on OER was inconclusive, it is likely that a lower OER
with BMP was caused by harvesting practices, where shorter harvest intervals (7 days)
resulted in the collection of harvested bunches with less loose fruits (i.e. lower average
ripeness).
Whilst a trade-off between higher FB yield with BMP or higher OER seems to exist,
oil yield gains from increased FB were larger than the effects of BMP on oil and kernel
extraction rates (Donough et al., 2010; Oberthür et al., 2012). It is therefore important
to balance oil content (i.e. lower oil content with BMP leading to higher cost of
production) and potential crop losses in the field (i.e. lower loose fruit losses with BMP
increasing recoverable oil yield and thus revenue) (Donough et al., 2010). However,
efforts to increase oil contents with management practices (such as fertilization) should
be pursued, but with caution, so that increases in FB yields are not compromised.
Alternative efforts should focus on increasing milling efficiency, particularly at small-
scale processors (described in more detail in Chapter 3), and increasing smallholder
access to improved varieties such as the tenera hybrid (Adjei-Nsiah et al., 2012a).
6.6 Seasonality in production persists despite appropriate management
interventions
Even if the focus is on increasing oil yields and milling efficiency, seasonality in
production persists (Fig. 5.5). While water supply is the most likely factor to explain
annual cycles in production, results from the irrigation trial (Chapter 5) found no effect
on yield cycles with supplemental irrigation (Chapter 5). Instead, it is argued that
seasonal peaks in flowering might be controlled by day-length (photoperiod), even near
the equator (Combres et al., 2013; Legros et al., 2009), and are confounded with drought
effects. On the other hand, Henson and Dolmat (2004), suggest that factors associated
with dry conditions such as high temperature, vapour pressure deficit and high
evaporation rate are responsible for yield oscillations. Other authors propose that yield
cycling in oil palm is partly a result of internal feedback mechanisms, whereby a current
high yield leads to a future low yield and vice versa (Breure and Corley, 1992; Corley,
1977; Corley and Teo, 1975). Whilst conclusions on the primary cause of yield cycling
may vary, efforts must continue to unravel the process(es) behind these yield cycles in
order to more efficiently plan the milling process and labour requirements. Furthermore,
multiple time-lagged stress events may affect current yields, which complicate
unravelling the interaction between various climatic variables even more.
Understanding the causes behind yield oscillations will give insight to whether this will
be possible to manipulate through management interventions and whether it would be
practical or not.

122
 GENERAL DISCUSSION

Much work on the effects of weather on FB yield has involved a search between
correlations between climatic factors and yield components (Ong, 1982a; Ong, 1982b).
However, the existence of a correlation does not necessarily indicate cause and effect,
since it is almost inevitable that correlations between monthly means of two factors that
both vary seasonally exists. Also, without prior knowledge on the time-lag and direction
of the effect, a correlation may mean nothing (Corley and Tinker, 2016). Whilst simple
linear regression showed large correlations between monthly yield and the water deficit
with appropriate time-lags at the irrigation trial (Fig. 5.6, Table 5.2), Ong (1982b) notes
that it can be misleading to study a particular climatic factor in isolation without
considering the possibility of overshadowing effects by other climatic factors. Because
climatic factors tend to be highly interrelated (White, 1979), (time-lagged) interactions
between climatic factors should therefore be considered when analysing monthly yield
oscillations.
Since controlled experiments in growth chambers are impractical in mature oil palm,
the use of crop simulation models can offer insight into understanding the interaction
between several components on yield. A challenge for oil palm (and other perennial
crops), however, is evaluating model performance against observed results due to the
lack of comprehensive yield data sets over a complete planting cycle (Hoffmann et al.,
2014). While most models seek to limit the number of input data needed, they therefore
do not fully explore the full range of environmental parameters and interactions that
could potentially affect yield oscillations in oil palm. As indicated in Chapter 2,
comprehensive climatic datasets across Ghana were also lacking, or difficult to find.
While this issue was sought to overcome by installing automatic weather stations
(Spectrum Technologies WatchDog 2000 Series) at all plantation sites and the Oil Palm
Research Institute (OPRI), we experienced multiple breakdowns, resulting in the loss
of significant data including measurements on solar and photosynthetic active radiation
(PAR), relative humidity (RH) and temperate amongst others. Coupled with in-field
measurements on stomatal conductance and vapour pressure deficit (VPD) at the
irrigation trial at Norpalm Ghana Ltd. (Chapter 5), this would have been valuable
information to understanding interactions between climatic variables on yield
oscillations in Ghana. However, these measurements were unfortunately not taken due
to budget and logistical constraints.

123
CHAPTER 6

6.7 Reflections on my PhD research

Looking back on my research, yes, I would have done things differently with the
knowledge that I now have. Most changes would not concern the setup of the trials or
BMP work, but would be mostly related to the scale of implementation. For example, I
would have liked to add a greater number of BMP sites in different stages of production
and installed more fertilizer and irrigation trials across a larger range of environments.
Unfortunately, due to budget and time constraints, as well as the local capacity needed
to run oil palm trials in good order (e.g. skilled technicians, high turnover in trial
managers), this was not possible.
Since management practices depend also on the age of the palm, having more BMP
sites in different stages of production (i.e. steep ascending yield phase, plateau yield
phase, declining yield phase) would have allowed for a more detailed yield gap analysis
across a range of agro-ecological conditions and a more accurate assessment of the
production constraints within each age group (Chapter 3, Chapter 4). This would have
been very valuable information for oil palm practitioners in Ghana. My research looked
only at yield gaps and production constraints across the entire planting cycle, since I
did not have enough replications within each production phase for a strong statistical
analysis.
The irrigation trial (Chapter 5) was installed in close proximity to a river, so the
likelihood that this affected treatment yields is quite large. The decision to install the
trial at that particular location, however, was driven more by practical considerations
and budget constraints. Preferably, the trial would have been located further from the
river to minimize its influence. Furthermore, I would have implemented the trial at the
block level, instead of embedding plots in one plantation block. This would have greatly
reduced the need to install trenches and the likelihood of water and nutrient poaching
between treatments. However, these requirements would have added significantly to the
budget, as more pipework and a more powerful pump would have been needed.
Oil palm is a crop that takes time to learn and understand. New research ideas arise
constantly, but implementing them is time-consuming. Since a PhD program is 4 years
and oil palm has a time-lagged response of 35–40 months to management interventions,
implementing new agronomic trials along the way was in most cases already too late to
ensure a significant yield response within my PhD research. A clear understanding of
the crop and research objectives is therefore required at the very beginning.

124
 GENERAL DISCUSSION

6.8 Implications of my research at the broader scale

6.8.1 Increasing production through area expansion or yield intensification?
Currently, agricultural expansion poses a serious threat to natural ecosystems
worldwide and is expected to increase over the next 50 years (Tilman et al., 2001).
During the past few decades, oil palm has become one of the most rapidly expanding
crops worldwide (Fitzherbert et al., 2008; Koh and Wilcove, 2008), and has often
become the subject of international protest due to its impact on deforestation. As
available land becomes increasingly scarce in the traditional production centres in
Southeast Asia, oil palm production is expanding rapidly across tropical regions in Latin
America and Africa (Hansen et al., 2015; Laurance et al., 2010).
Whilst expansion of oil palm plantations is often linked to deforestation, a recent study
shows that the proportion of plantations replacing forests across Indonesia’s major oil
palm producing islands of Sumatra, Kalimantan and Papua decreased from 54% during
1995–2000 to 18% during 2010–2015 (Austin et al., 2017). In another study, Fitzherbert
et al. (2008) indicates that forest conversion to oil palm accounted for at most 16% of
deforestation between 1990–2005, while Koh and Wilcove (2008) suggest that oil palm
expansion accounted for at least 50% of deforestation in Malaysia and Indonesia during
the same time period. According to Fairhurst and Härdter (2003), only 7.6% of
deforestation between 1990–2000 can be attributed to oil palm, while Vijay et al. (2016)
reveals that 45% of oil palm plantations in Southeast Asia came from areas that were
forests in 1989, 31% in South America, and 2% and 7% for Mesoamerica and Africa
respectively. The exact percentage of forest conversion to oil palm is therefore
debatable and depends largely on the (quality) of data used, the inclusion of immature
area in the analysis, as well as the history of land-use previous to oil palm (e.g. some
oil palm plantings have been established on land previously occupied by rubber and
grassland) (Fairhurst and Härdter, 2003; Wicke et al., 2011). Although the extent to
which oil palm has been a direct cause of past deforestation is difficult to quantify, its
potential to contribute to future deforestation is considerable (Fitzherbert et al., 2008).
This raises major conservation and environmental concerns.
Despite much international scrutiny, the expansion of land to increase production is
likely to continue for many years due to an increasing global demand for palm oil. It is
therefore imperative to find and implement solutions for agriculture and biological
conservation to co-exist. A valuable approach to assess strategies for increasing food
production whilst minimizing the negative consequences for biodiversity is the land
sparing-sharing model (Green et al., 2005). In land sparing, homogeneous areas of
farmland are managed to maximize yields, while separate reserves target biodiversity
conservation. In contrast, wildlife-friendly farming, or land sharing, integrates

125
CHAPTER 6

conservation and production within more heterogeneous landscapes, often resulting in

lower yields (Green et al., 2005). Baudron and Giller (2014) note that both approaches
have their benefits and disadvantages and should be assessed according to local
circumstances. For oil palm agriculture, however, a number of studies suggest land
sparing to be the best strategy in reconciling food production and biodiversity
conservation (Edwards et al., 2010; Fitzherbert et al., 2008; Phalan et al., 2009; Phalan
et al., 2011).
Compared with natural forests, oil palm plantations support little biodiversity (Butler
and Laurance, 2009; Edwards et al., 2013; Edwards et al., 2014; Fayle et al., 2010;
Fitzherbert et al., 2008; Koh and Wilcove, 2008; Mendes-Oliveira et al., 2017).
Attempts to increase biodiversity within plantations following a land sharing approach,
by retaining forest fragments within the oil palm matrix, proved ineffective in a number
of studies (Edwards et al., 2010; Yue et al., 2015). They concluded that land sparing
strategies by retaining contiguous forest would be the most effective approach for
conserving biodiversity in an oil palm landscape. Phalan (2018) further notes that most
species will have larger populations if food is produced on as small an area as possible,
while sparing as large an area of native vegetation as possible and that the potential
benefits of land sharing or intermediate strategies for wild species are more limited. In
another study, Lee et al. (2014) indicated that oil palm expansion following a high-
yielding land-sparing approach had less environmental costs than a land sharing
approach in terms of forest conversion, greenhouse gas emissions and biodiversity
losses, though the latter would be more favourable for socio-economic development by
creating more employment opportunities.
Land sparing in oil palm could in theory be attained if yields are increased. Chapter 4,
for example, shows that increasing average attainable yields at plantations and
smallholder farmers to 21 t ha-1 FB could potentially reduce oil palm expansion in
Ghana by almost 600,000 ha of land. Oil palm produces the highest oil yield and is the
most resource-use efficient amongst the world’s major oil crops (de Vries et al., 2010).
Because of its high yield potential and low production costs (Fairhurst and Griffiths,
2014), oil palm uses land more efficiently than any other oil crop, thus rendering the
crop particularly suitable for intensification (Corley, 2009). Yield intensification on
land already planted to oil palm may therefore be an important policy for sustainable
oil palm development in Ghana and West Africa.
However, intensification does not per se lead to land sparing for conservation. Most
notably, intensification might create financial incentives for oil palm expansion if yield
increases raise the opportunity costs for nature conservation, i.e. the potential income
that is lost if land is not converted to oil palm (Laurance et al., 2014; Phelps et al., 2013).

126
 GENERAL DISCUSSION

Since oil palm is a highly profitable crop, the opportunity costs of not planting oil palm
are substantial. Moreover, if the demand for palm oil is elastic, increasing yields can in
effect increase production targets, thereby adding to the willingness to acquire more
agricultural land (Green et al., 2005). For example, an increase in the supply of palm
oil as a result from their use may not reduce palm oil prices, and the overall incentive
for increasing production through cultivating more land remains. In contrast, a price
decrease may also increase land expansion in order to maintain income (Baudron and
Giller, 2014).
Inherent to land sparing, or yield intensification, is the high reliance on, and access to,
modern agricultural technologies. In Ghana, for example, large yield gaps existed
amongst smallholder farmers, due to a lack of knowledge on appropriate management
practices, poor infrastructure and lack of access to credit to purchase agricultural inputs
such as fertilizers (Chapter 3). Access to, and adoption of, modern agricultural
technologies, such as the implementation of BMPs, is therefore essential to the land
sparing approach, and needs to be managed adequately to minimize its environmental
impact (Baudron and Giller, 2014). Key priorities to the land sparing approach are
therefore to ensure improvement and adoption of agricultural technologies amongst oil
palm producers, and implementation of policies that promote more ecologically
efficient food production, while strategically optimizing the allocation of lands to
conservation and agriculture (Laurance et al., 2014).
Recognizing the challenges around the global oil palm industry, oil palm plantations
are looking for ways to produce palm oil in an environmentally and socially responsible
way. A positive example are the current operations in Gabon. Gabon, a country covered
with more than 76% in forest, with 11% of its land area protected in national parks, sees
industrial agriculture as an important part of the nation’s future, and has developed a
national land-use plan that attempts to balance oil palm production and forest
preservation. Cooperation between government, plantations and conservationists has
shown that profitable farms, thriving rural communities and healthy ecosystems can
coexist with good land-use planning (National Geographic, 2018).
Alternatively, oil palm expansion might be accommodated by targeting lands that have
already been degraded, such as anthropogenic grasslands (Fairhurst and McLaughlin,
2009). Conversion of such landscapes to oil palm has shown to have limited negative
impacts on biodiversity (Gilroy et al., 2015) and could potentially supply all the oil
required for edible purposes in 2050, without any forest clearing (Corley, 2009).

127
CHAPTER 6

6.8.2 Stronger collaboration between plantations and smallholders is required to

increase production in Ghana
Area expansion is limited for large plantations in Ghana because of the lack of large
contiguous tracts of land and complicated land tenure systems (Chapter 2). The major
factor driving area expansion in Ghana is therefore not plantation development, but
rather an expansion of the smallholder sector (Rhebergen et al., 2014). Because most
large plantations in Ghana would like to increase their processing capacity, expansion
of fruit bunch supply may therefore come from smallholder farmers, which occupy
about 95% of the total area under oil palm cultivation and/or yield improvement
strategies.
Whilst plantations in Ghana are distinguished by their large size, mono-cultural
character and systematic layout (Gyasi, 1996), the smallholder sector is largely shaped
as an unorganized mosaic of low-yielding small farms within a highly fragmented
agricultural landscape, inherent to the land-sharing approach (Phalan et al., 2009).
Moreover, compared with e.g. Malaysia, where smallholders are tightly integrated into
the industry structure such as through the FELDA scheme (Shamsul Bahrin and Lee,
1988), smallholders in Ghana are not well integrated (Fold and Whitfield, 2012). In
Ghana, some smallholders are attached to plantation schemes as nucleus-smallholders
or out-growers (5%), but most are independent farmers with private farms (90%) (Table
3.1) (Adjei-Nsiah et al., 2012a; Osei-Amponsah et al., 2012). Despite smallholder-
plantation schemes, the volume of crop is too little to fulfil the processing needs of
plantation mills, which are thus dependent on outside fruit purchases for the mill to run
efficiently. Whilst smallholder-plantation schemes receive technical advice and inputs
(e.g. high-yielding seedlings, fertilizers) on credit, in return that all crop goes to the
plantation mill, independent smallholder farmers are completely self-reliant. They are
not contractually bound to deliver their crop to a particular mill or association, and will
sell their fruits to the highest bidder. In areas where there are several estates in close
proximity, competition for fruit bunches is therefore high, which has led to price wars
between plantations as well as with local buyers for the home consumption market.
Furthermore, the uncoordinated establishment of new mills too close to existing ones
exacerbates competition for fruit bunches, demonstrating the need for better spatial
planning in the industry (Fold and Whitfield, 2012) (Fig. 6.2).

128
 GENERAL DISCUSSION

Fig. 6.2. Large-scale mills and their catchment areas (30 km radius) in southern Ghana. Only 2.5 M ha,
or 37%, of the total land area within the oil palm belt (6.8 M ha) is covered, indicating a large overlap
in catchment area and increased competition between mills.

Viable integration of smallholder farmers into the oil palm supply chain, for example
through farmers’ groups or smallholder schemes, and better integration with existing
mills has considerable potential to increase production and regulate expansion. Jelsma
et al. (2017), for instance, suggests that with a strong institutional arrangement,
smallholder farmers can participate in supply chains on advantageous conditions and
substantially increase their productivity. Organizing smallholder farmers in large
concessions (akin to smallholder schemes as in e.g. Malaysia and Indonesia) following
a more land sparing approach, could potentially open up areas for biodiversity
conservation as well, but would require good supporting policies and a significant
restructuring of land allocated to oil palm and conservation in Ghana.
As discussed in section 6.8.1, yield intensification with BMP (i.e. the land sparing
approach), offers scope to increase production in existing plantings, and potentially
reduce the requirement for future area expansion. With improved crop recovery and
agronomic management, average yields in mature plantation and smallholder palm
stands approached 18 t ha-1 FB (Table 4.1). If average attainable yields of 20 t ha-1 FB
can be achieved, plantations in Ghana would be more self-sufficient and reliance on
outside fruit purchases would greatly reduce (Table 6.1). However, yield intensification

129
CHAPTER 6

depends on the feasibility of scaling up BMPs and would require a proper economic
analysis to assess its potential.
Smallholder farmers in Ghana face major challenges in increasing yields, such as lack
of knowledge on appropriate management practices, poor operating conditions (e.g.
poor infrastructure), and lack of access to high-yielding seedlings, agronomic inputs,
credit, and extension advice. Key to successful implementation and up-scaling of
BMP’s is therefore the provision of adequate services to smallholders. Improvements
to infrastructure (particularly feeder roads to the farm) reduces the cost of production,
whilst a ‘one-stop-shop’ approach could provide provision of milling, inputs and advice
to the industry, particularly smallholders. Because large-scale mills benefit from
investments in smallholder production to secure sufficient crop supply, they could
potentially provide these services as extension agents, by advancing inputs (e.g.
fertilizers, agrochemicals, tools, quality DxP seedlings) and advice to smallholders
under credit, against budgeted future crop deliveries to the respective mill. To
effectively monitor smallholder production (e.g. budgeted production versus actual
delivery) and to identify yield gaps at scale, collection of farm data in a database (e.g.
OMP (Agrisoft Systems, 2018)) is essential. This could further assist milling companies
as extension agents in providing adequate feedback to growers for yield intensification
in Ghana. Monitoring each smallholder farm diligently furthermore reduces the risk
attached to making loans and provides the means for targeted extension work (Fig. 6.3).

Fig. 6.3. Large-scale mills can serve as one stop shops to provide services and inputs for the
development of the smallholder sector in Ghana.

Increasing fruit supply by organizing the expanding smallholder sector and

implementing yield intensification strategies (BMPs) on existing plantings is largely an
unexploited potential for the sustainable intensification of oil palm production in
Ghana, and is essential to move the industry forward in a sustainable manner. However,
suitable organizational models have to be further explored for Ghana’s complex oil
palm sector as well as pathways on how these models can be implemented. This will
require collaborative action between industry actors (e.g. plantations, smallholders) and
governments alike.

130
 GENERAL DISCUSSION

6.9 Conclusions: opportunities for the sustainable development of the oil palm
sector in Ghana
The oil palm industry in Ghana faces many challenges, mostly related to low
productivity and the unorganized expansion of the oil palm sector. As described in
sections 1.2–1.4, productivity is limited by water and nutrient constraints, as well as
labour shortages to fully implement or sustain BMPs. The industry is furthermore
challenged by limitations to large-scale plantation expansion due to the lack of land and
complicated land tenure systems. Increases in production thus have to be sought from
smallholder farms and/or yield intensification strategies. However, this is a major
challenge, since smallholder farmers are poorly organized and are not yet well
incorporated in the oil palm supply chain. Good supporting policies are therefore
required for the viable integration of smallholders in the industry.
Production increases can alternatively be sought by applying BMPs to land already
planted with oil palm. Opportunities for increasing production can therefore be sought
by improving current management practices. Adapting BMPs to local conditions is
thereby a useful strategy to identity the management practices that are responsible for
yield gaps. My research has demonstrated that yield intensification of existing palm
stands shows considerable scope in Ghana (Chapter 4), and can be achieved through
adequate nutrient management, irrigation and yield taking (crop recovery) and yield
making (agronomy) practices. Results show that average FB yields of 17.9 t ha-1 and
17.6 t ha-1 can be achieved with BMP at plantations and smallholders respectively.
Average yield increases of 16% (+1.8 t ha-1 FB) and 37% (+2.7 t ha-1 FB) were achieved
in the short-term (≤1 year) with yield taking, whilst yield making contributed to
additional increases of 36% (+4.7 t ha-1) and 76% (+7.6 t ha-1) over the long term (>1
year) at plantations and smallholders (Chapter 4). Moreover, results from the irrigation
trial demonstrated the potential of irrigation x fertilizer to increase average yields even
further, at 32.6 t ha-1 FB in the plateau yield phase, but indicated the need for more
research on the specific water and nutrient needs of oil palm across different
environments and production phases in Africa (Chapter 5).
By increasing yields with moderate BMP implementation (average FB yields of
17.9 t ha-1 and 17.6 t ha-1 at plantations and smallholders), approximately 450,000 ha of
area expansion can be reduced when compared with present yields (11.4 t ha-1 at
plantations and 7.3 t ha-1 at smallholders). At potential production levels (i.e. average
FB yields of 21.0 t ha-1 at plantations and smallholder farms), however, about
600,000 ha of land can be spared. Additionally, closing yield gaps in Ghana under
current land area (327,600 ha) has the potential to increase FB production almost three-
fold from 2,5 Mt to 6,9 Mt. If all crop is processed at an OER of 21%, about

131
CHAPTER 6

1.4 Mt CPO can be produced, which is more than enough to meet Ghana’s current
annual demand of 106,000 t CPO, and does not require the need to plant additional land.
Yield intensification strategies therefore offers considerable scope to reduce the
requirement for future area expansion, and may therefore be an important policy for
sustainable oil palm development in Ghana and West Africa.

132
 Table 6.1. Impact of yield intensification on the land area needed to accommodate production shortages at current milling capacity and potential annual
processing capacity (at a peak crop of 16%) at three large-scale plantations in Ghana.
Potential Current status Yield intensification with BMP
Milling
annual FB FB Production Area FB Area Change Production
Plantation capacity Area
capacity production yield shortfall needed yield needed in area shortfall
tph t FB ha t t ha-1 t ha t ha-1 ha ha t
BOPP 20 73,333 4,890 42,300 8.7 31,033 3,588 20 3,667 -1,223 -24,460
Norpalm 30 110,000 3,760 42,000 11.2 68,000 6,088 20 5,500 1,740 34,800
TOPP 30 110,000 3,250 41,300 12.7 68,700 5,406 20 5,500 2,250 45,000
Total 80 293,333 11,900 125,600 10.6 167,733 15,892 20 14,667 2,767 55,340

133
GENERAL DISCUSSION
 REFERENCES
Abdel Kawy, W.A.M. and Abou El-Magd, I.H., 2012. Use of satellite data and GIS for
assessing the agricultural potentiality of the soils South Farafra Oasis, Western
Desert, Egypt. Arabian J. Geosci., 6(7): 2299–2311.
Abdul-Aziz, A.R., Anokye, M., Kwame, A., Munyakazi, L. and Nsowah-Nuamah,
N.N.N., 2013. Modeling and forecasting rainfall pattern in Ghana as a seasonal
arima process: the case of Ashanti Region. Int. J. Humanit. Soc. Sci., 3(3): 224–
233.
Abeles, F.B., Morgan, P.W. and Saltveit Jr., M.E., 1992. Ethylene in Plant Biology. 2nd
ed. Academic Press, Inc., San Diego, CA, pp. 414.
Adam, H., Collin, M., Richaud, F., Beule, T., Cros, D., Omore, A., Nodichao, L., Nouy,
B. and Tregear, J.W., 2011. Environmental regulation of sex determination in oil
palm: current knowledge and insights from other species. Ann. Bot., 108(8):
1529–37.
Adjei-Gyapong, T. and Asiamah, R.D., 2002. The interim Ghana soil classification
system and its relation with the World Reference Base for soil resources,
Quatorzième réunion du Sous-Comité ouest et centre africain de corrélation des
sols pour la mise en valeur des terres. Food and Agriculture Organization of the
United Nations (FAO), Abomey, Benin, pp. 51–76.
Adjei-Nsiah, S., Sakyi-Dawson, O. and Kuyper, T.W., 2012a. Exploring Opportunities
for Enhancing Innovation in Agriculture: The Case of Oil Palm Production in
Ghana. J. Agric. Sci., 4(10): 212–223.
Adjei-Nsiah, S., Zu, A.K.S. and Nimoh, F., 2012b. Technological and Financial
Assessment of Small Scale Palm Oil Production in Kwaebibrem District, Ghana.
J. Agric. Sci., 4(7): 111–120.
Aggarwal, P.K. and Kalra, N., 1994. Analyzing the limitations set by climatic factors,
genotype, and water and nitrogen availability of wheat. 2. Climatically potential
yields and management strategies. Field Crops Res., 38(2): 93–103.
Agrisoft Systems, 2018. Software Solutions For Precision Agriculture. URL:
http://www.agrisoft-systems.com (accessed on 22.08.2018).
Ajaegbu, H.I., 1979. Urbanisation in Africa. In: R.K. Udo (Editor), Population
Education Sourcebook for Sub-Saharan Africa. Heinemann Educational Books,
Nairobi, Kenya, pp. 87–96.

135
REFERENCES

Alam, M. and Rogers, D., 1997. Tensiometer use in scheduling irrigation, Kansas State
University Agricultural Experiment Station and Cooperative Extension Service,
Manhattan, Kansas, USA, pp. 6.
Allen, R.G., Pereira, L.S., Raes, D. and Smith, M., 1998. Crop Evapotranspiration -
Guidelines for Computing Crop Water Requirements. FAO Irrigation and
Drainage Paper No. 56. FAO, Rome, Italy, pp. 281.
Anderson, W.K., 2010. Closing the gap between actual and potential yield of rainfed
wheat. The impacts of environment, management and cultivar. Field Crops Res.,
116(1–2): 14–22.
Angelsen, A., 2010. Policies for reduced deforestation and their impact on agricultural
production. Proc. Natl. Acad. Sci. U. S. A., 107(46): 19639–19644.
Angelucci, F., 2013. Analysis of incentives and disincentives for palm oil in Ghana.
Technical Notes Series. Food and Agriculture Organization of the United
Nations (FAO), Rome, Italy, pp. 34.
Annan-Afful, E., Iwashima, N., Otoo, E., Asubonteng, K.O., Kubota, D.,
Kamidohzono, A., Masunaga, T. and Wakatsuki, T., 2004. Nutrient and bulk
density characteristics of soil profiles in six land use systems along topo-
sequences in inland valley watersheds of Ashanti Region, Ghana. Soil Sci. Plant
Nutr., 50(5): 649–664.
Annan-Afful, E., Masunaga, T. and Wakatsuki, T., 2005. Soil properties along the
toposequence of an inland valley watershed under different land uses in the
Ashanti Region of Ghana. J. Plant Nutr., 28(1): 141–150.
Anon, 2004. Plantation mechanisation: Critical success factors. The Planter, 80: 417–
419.
Antwi-Agyei, P., Fraser, E.D.G., Dougill, A.J., Stringer, L.C. and Simelton, E., 2012.
Mapping the vulnerability of crop production to drought in Ghana using rainfall,
yield and socioeconomic data. Appl. Geogr., 32(2): 324–334.
Appiah, M.R., Sackey, S.T., Ofori-Frimpong, K. and Afrifa, A.A., 1997. The
consequences of cocoa production on soil fertility in Ghana: a review. Ghana J.
Agric. Sci., 30(2): 183–190.
Arolu, I.W., Rafii, M.Y., Marjuni, M., Hanafi, M.M., Sulaiman, Z., Rahim, H.A.,
Kolapo, O.K., Abidin, M.I.Z., Amiruddin, M.D., Kushairi Din, A. and Nookiah,
R., 2016. Genetic variability analysis and selection of pisifera palms for
commercial production of high yielding and dwarf oil palm planting materials.
Ind. Crops Prod., 90: 135–141.

136
 REFERENCES

Asante, E.A., 2012. DIIS Working Paper. The Case of Ghana's President's Special
Initiative on Oil Palm (PSI - Oil Palm), Danish Institute for International Studies,
DIIS, Copenhagen, Denmark, pp. 34.
Austin, K.G., Mosnier, A., Pirker, J., McCallum, I., Fritz, S. and Kasibhatla, P.S., 2017.
Shifting patterns of oil palm driven deforestation in Indonesia and implications
for zero-deforestation commitments. Land Use Policy, 69: 41–48.
Bachy, A., 1964. Tropisme racinaire du palmier à huile. Oléagineux, 19(11): 684–685.
Bachy, A., 1965. Influence de l'éclaircie naturelle sur la production du palmier à huile.
Oléagineux, 20: 575–577.
Balk, D.L., Deichmann, U., Yetman, G., Pozzi, F., Hay, S.I. and Nelson, A., 2006.
Determining Global Population Distribution: Methods, Applications and Data.
Adv. Parasitol., 62: 119–156.
Baudron, F. and Giller, K.E., 2014. Agriculture and nature: Trouble and strife? Biol.
Conserv., 170: 232–245.
Bell, M.A., Fischer, R.A., Byerlee, D. and Sayre, K., 1995. Genetic and agronomic
contributions to yield grains: a case study for wheat. Field Crops Res., 44(2–3):
55–65.
Breure, C., 2003. The Search for Yield in Oil Palm: Basic Principles. In: T. Fairhurst
and R. Härdter (Editors), The Oil Palm - Management for Large and Sustainable
Yields. International Plant Nutrition Institute (IPNI) and International Potash
Institute (IPI), Singapore, Singapore, pp. 59–98.
Breure, C.J. and Corley, R.H.V., 1992. Fruiting activity, growth and yield of oil palm.
II. Observations in untreated populations. Exp. Agric., 28: 111–121.
Broschat, T.K., 2009. Palm nutrition and fertilization. HortTechnology, 19(4): 690–
694.
Broekmans, A., 1957. Growth, flowering and yield of the oil palm in Nigeria. J.W. Afr.
Inst. Oil Palm Res., 2(7): 187–220.
Bryan, B.A., King, D. and Zhao, G., 2014. Influence of management and environment
on Australian wheat: information for sustainable intensification and closing yield
gaps. Environ. Res. Lett., 9(4): 044005.
Buringh, P., 1979. Introduction to the Study of Soils in Tropical and Subtropical
Regions. Centre for Agricultural Publishing and Documentation, Wageningen,
pp. 124.

137
REFERENCES

Butler, R.A. and Laurance, W.F., 2009. Is oil palm the next emerging threat to the
Amazon? Trop. Conserv. Sci., 2(1): 1–10.
Caliman, J.P. and Southworth, A., 1998. Effect of drought and haze on the performance
of oil palm, International Oil Palm Conference 'Commodity of the past, today,
and the future'. IOPRI/GAPKI, Bali, Indonesia, pp. 250–274.
Caliman, J.P., 1992. Oil palm and water deficit, production, adapted cropping
techniques. Oléagineux, 47(5): 205–216.
Carr, M.K.V., 2011. The Water Relations and Irrigation Requirements of Oil Palm
(Elaeis Guineensis): A Review. Exp. Agric., 47(04): 629–652.
Carrere, R., 2010. Oil palm in Africa: Past, present and future scenarios, World
Rainforest Movement, pp. 78.
Chan, 1991. Predicting oil palm yield potential based on solar radiation, Second
National Seminar on Agrometeorology. Malaysian Meteorological Service,
Petaling Jaya, Malaysia, pp. 129–139.
Chan, K.W., Corley, R.H.V. and Seth, A.K., 1972. Effects of growth regulators on fruit
abscission in oil palm, Elaeis guineensis. Ann. Appl. Biol., 71: 243–249.
Chan, K.W., Yee, C.B., Lim, K.C. and Goh, M., 1985. Effect of rainfall and irrigation
on oil palm yield production, National Conference on Soil - International Oil
Palm/Palm Oil Conferences. Conference I: Agriculture, PORIM, Bangi,
Malaysia, pp. 36–45.
Chapman, G.W. and Gray, H.M., 1949. Leaf analysis and the nutrition of the oil palm
(Elaeis guineensis Jacq.). Ann. Bot., 13(52): 415–433.
Chapman, K.R., Escobar, R. and Griffee, P., 2003. Cold tolerant or altitude adapted oil
palm hybrid development initiatives in the Asia/Pacific Region. AU. J. Technol.,
6(3): 134–138.
Cheng-xu, S., Hong-xing, C., Hong-bo, S., Xin-tao, L. and Yong, X., 2011. Growth and
physiological responses to water and nutrient stress in oil palm. Afr. J.
Biotechnol., 10(51): 10465–10471.
Chew, P. and Goh, K., 2003. Maximising palm oil yields on estates, Proceedings of the
MOSTA Seminar 9 on Recent Advances in the Oil Palm Sector: Agriculture,
Plantation Management, End Uses and Nutrition held in Sandakan, 8–9 May
2003. Malaysian Oil Scientists' and Technologists' Association (MOSTA),
Sandakan, Malaysia.

138
 REFERENCES

CIESIN, IFPRI, The World Bank and CIAT, 2011. Global Rural-Urban Mapping
Project, Version 1 (GRUMPv1): Urban Extents Grid. Palisades, NY: NASA
Socioeconomic Data and Applications Center (SEDAC) URL:
https://doi.org/10.7927/H4GH9FVG ((accessed on 09.10.2014).
Combres, J.-C., Pallas, B., Rouan, L., Mialet-Serra, I., Caliman, J.-P., Braconnier, S.,
Soulié, J.-C. and Dingkuhn, M., 2013. Simulation of inflorescence dynamics in
oil palm and estimation of environment-sensitive phenological phases: a model
based analysis. Funct. Plant Biol., 40(3): 263–279.
Corley, R.H.V. and Breure, C.J., 1981. Measurements in Oil Palm Experiments,
Unilever Plantations Group, London, pp. 16.
Corley, R.H.V. and Hong, T.K., 1982. Irrigation of Oil Palms in Malaysia. In: E.
Pushparajah and P.S. Chew (Editors), The Oil Palm in Agriculture in the Eighties
– Volume II. ISP, Kuala Lumpur, pp. 343–356.
Corley, R.H.V. and Teo, C., 1975. Disbudding of mature oil palms as a method of
controlling yield fluctuation. MARDI Res. Bull., 4(1): 1–6.
Corley, R.H.V. and Tinker, P.B., 2003. The Oil Palm. World Agriculture Series.
Blackwell Science Ltd, Oxford, UK, pp. 562.
Corley, R.H.V. and Tinker, P.B., 2016. The Oil Palm. World Agriculture Series. Wiley-
Blackwell, Oxford, UK, pp. 639.
Corley, R.H.V., 1976. Inflorescence, abortion and sex differentiation. In: R.H.V.
Corley, J.J. Hardon and B.J. Wood (Editors), Oil Palm Research. Developments
in Crop Science (1). Elsevier, Amsterdam, pp. 37–54.
Corley, R.H.V., 1977. Oil palm yield components and yield cycles. In: D.A. Earp and
W. Newall (Editors), Malaysian International Agricultural Oil Palm Conference.
The Incorporated Society of Planters, Kuala Lumpur, Malaysia, pp. 116–129.
Corley, R.H.V., 2009. How much palm oil do we need? Environ. Sci. Policy, 12(2):
134–139.
Cros, D., Flori, A., Nodichao, L., Omoré, A. and Nouy, B., 2013. Differential response
to water balance and bunch load generates diversity of bunch production profiles
among oil palm crosses (Elaeis guineensis). Trop. Plant Biol., 6: 26–36.
Danielsen, F., Beukema, H., Burgess, N.D., Parish, F., Bruhl, C.A., Donald, P.F.,
Murdiyarso, D., Phalan, B., Reijnders, L., Struebig, M. and Fitzherbert, E.B.,
2009. Biofuel plantations on forested lands: double jeopardy for biodiversity and
climate. Conserv. Biol., 23(2): 348–358.

139
REFERENCES

Danso, I., Nuertey, B., Asamoah, T., Tetteh, F., Danso, F., Afari, P., Safo, E. and Logah,
V., 2010. The effect of rock phosphate on soil nutrient dynamics, growth,
development and yield of oil palm in the semi-deciduous forest zone of Ghana.
J. Sci. Technol., 30(1): 30–44.
Danso, I., Nuertey, B.N., Andoh-Mensah, E., Osei-Bonsu, A., Asamoah, T.E.O.,
Dwarko, D.A., Okyere-Boateng, G., Marfo-Ahenkora, E. and Opoku, A., 2008.
Response of oil palm to planting density and water deficit in three climatic zones
of southern Ghana. J. Ghana Sci. Assoc., 10(2): 93–102.
de Vries, S., van de Ven, G., van Ittersum, M. and Giller, K., 2010. Resource use
efficiency and environmental performance of nine major biofuel crops,
processed by first-generation conversion techniques. Biomass Bioenerg., 34:
588–601.
Desmarest, J., 1967. Essai d'irrigation sur jeune palmeraie industrielle. Oléagineaux,
22: 441–447.
Donough, C., Witt, C. and Fairhurst, T.H., 2010. Yield intensification in oil palm using
BMP as a management tool, Proceedings of the International Oil Palm
Conference. Indonesian Oil Palm Research Institute (IOPRI), Jogyakarta,
Indonesia, pp. 1–8.
Donough, C.R., Witt, C. and Fairhurst, T.H., 2009. Yield intensification in oil palm
plantations through best management practices. Better Crops International,
93(1): 12–14.
Dubos, B., Caliman, J.P., Corrado, F., Quencez, P., Suyanto, S. and Talliez, B., 1999.
Importance of magnesium nutrition in oil palm. Plantations, recherche,
développement, 6(5): 313–325.
Dudley, N. (Editor), 2008. Guidelines for Applying Protected Area Management
Categories. IUCN, Gland, Switzerland, pp. 86.
Dufour, O., Frere, J., Caliman, J. and Hornus, P., 1988. Description of a simplified
method of production forecasting in oil palm plantations based on climatology.
Oléagineux, 43(7): 271–278.
Dufrene, E., Ochs, R. and Saugier, B., 1990. Oil palm photosynthesis and productivity
linked to climatic factors. Oléagineux, 45: 345–355.
Edwards, D.P., Hodgson, J.A., Hamer, K.C., Mitchell, S.L., Ahmad, A.H., Cornell, S.J.
and Wilcove, D.S., 2010. Wildlife-friendly oil palm plantations fail to protect
biodoversity effectively. Conserv. Lett., 3: 236–242.

140
 REFERENCES

Edwards, F.A., Edwards, D.P., Hamer, K.C. and Davies, R.G., 2013. Impacts of logging
and conversion of rainforest to oil palm on the functional diversity of birds in
Sundaland. Ibis, 155: 313–326.
Edwards, F.A., Edwards, D.P., Larsen, T.H., Hsu, W.W., Benedick, S., Chung, A.Y.C.,
Vun Khen, C., Wilcove, D.S. and Hamer, K.C., 2014. Does logging and forest
conversion to oil palm agriculture alter functional diversity in a biodiversity
hotspot? Anim. Conserv., 17: 163–173.
Edwards, R., 2015. Is plantation agriculture good for the poor? Evidence from
Indonesia's palm oil expansion, Australian National University, pp. 41.
Environmental Protection Agency (EPA) and Ministry of Environment, Sciene and
Technology, 2011. Ghana's Second National Communication to the UNFCCC.
Ministry of Environment, Science and Technology, Accra, Ghana, pp. 168.
Epebinu, O. and Nwadialo, B., 1993. Predicting soil water availability from texture and
organic matter content fort Nigerian soils. Commun. Soil Sci. Plant Anal., 24(7–
8): 633–640.
Euler, M., Hoffmann, M.P., Fathoni, Z. and Schwarze, S., 2016. Exploring yield gaps
in smallholder oil palm production systems in eastern Sumatra, Indonesia. Agric.
Syst., 146: 111–119.
Fairhurst, T., 1996. Management of nutrients for efficient use in smallholder oil palm
plantations. PhD Thesis, Wye College, London, pp. 211.
Fairhurst, T., Caliman, J.P., Härdter, R., Witt, C., 2004. Oil Palm: Nutrient Disorders
and Nutrient Management (Oil Palm Series Volume 7). Potash & Phosphate
Institute (PPI), Potash & Phosphate Institute of Canada (PPIC), International
Potash Institute (IPI), French Agricultural Research Centre for International
Development (CIRAD), and CTP Holdings, Singapore, pp. 67.
Fairhurst, T. and Griffiths, W., 2014. Oil Palm: Best Management Practices for Yield
Intensification. International Plant Nutrition Institute and Tropical Crop
Consultants Limited, Penang, Malaysia, pp. 180.
Fairhurst, T., Griffiths, W. and Rankine, I., 2019. Oil Palm – Agronomy. Tropical Crop
Consultants Limited, Wye, United Kingdom, pp. 317.
Fairhurst, T. and Härdter, R., 2003. Conclusions: From Area Expansion To Yield
Intensification. In: T. Fairhurst and R. Härdter (Editors), Oil Palm - Managament
for Large and Sustainable Yields. International Plant Nutrition Institute (IPNI)
and International Potash Institute (IPI), Singapore, pp. 333–336.

141
REFERENCES

Fairhurst, T. and McLaughlin, D., 2009. Sustainable Oil Palm Development on

Degraded Land in Kalimantan, World Wildlife Fund, Washington, USA, pp. 45.
Fairhurst, T.H., Griffiths, W. and Gfroerer-Kerstan, A., 2006. Concept and
implementation of best management practice for maximum economic yield in an
oil palm plantation in Sumatra, International Oil Palm Conference (IOPC), Bali,
Indonesia, pp. 1–16.
Fairhurst, T.H. and Mutert, E., 1999. Interpretation and management of oil palm leaf
analysis data. Better Crops, 13(1): 48–51.
FAO, 1976. A Framework for Land Valuation. FAO Soils Bulletin No. 32. FAO, Rome,
Italy, pp. 72.
FAO, 1985. Guidelines: Land Evaluation for Irrigated Agriculture. FAO Soils Bulletin
No. 55. FAO, Rome, Italy, pp. 290.
FAO, 2005. Irrigation in Africa in Figures - AQUASTAT Survey 2005. FAO Water
Reports. FAO, Rome, Italy, pp. 633.
FAO, 2014a. FAOSTAT Database. URL: http://faostat.fao.org (accessed on
18.04.2014).
FAO, 2014b. Lecture Notes on the Major Soils of the World, Set #6 Mineral Soils
conditioned by a Wet (Sub)Tropical Climate. URL:
http://www.fao.org/docrep/003/y1899e/y1899e08a.htm (accessed on
10.09.2014).
FAO, 2014c. Perennial Crops for Food Security. Proceedings of the FAO Expert
Workshop, 28–30 August 2013. FAO, Rome, Italy, pp. 390.
FAO, 2017. FAOSTAT Database. URL: http://faostat.fao.org (accessed on
07.02.2017).
FAO, 2019. FAOSTAT Database. URL: http://faostat.fao.org (accessed on
10.01.2019).
FAO, IIASA, ISRIC, ISSCAS and JRC, 2009. Harmonized World Soil Database
(version 1.1), FAO, Rome, Italy, IIASA, Laxenburg, Austria, pp. 38.
Fayle, T.M., Turner, E.C., Snaddon, J.L., Chey, V.K., Chung, A.Y.C., Eggleton, P. and
Foster, W.A., 2010. Oil palm expansion into rain forest greatly reduces ant
biodiversity in canopy, epiphytes and leaf-litter. Basic Appl. Ecol., 11: 337–345.
Fermont, A.M., van Asten, P.J.A., Tittonell, P., van Wijk, M.T. and Giller, K.E., 2009.
Closing the cassava yield gap: An analysis from smallholder farms in East
Africa. Field Crops Res., 112(1): 24–36.

142
 REFERENCES

Fipps, G., 2003. Irrigation Water Quality Standards and Salinity Management
Strategies. Texas Agricultural Extension Service, Texas A&M University
System, Texas, USA, pp. 18.
Fischer, R.A., Byerlee, D. and Edmeades, G.O., 2014. Crop yields and global food
security. Will yield increase continue to feed the world? ACIAR Monograph No.
158. Australian Centre for International Agricultural Research, Canberra,
Australia, pp. 634.
Fitzherbert, E., Struebig, M., Morel, A., Danielsen, F., Bruhl, C., Donald, P. and Phalan,
B., 2008. How will oil palm expansion affect biodiversity? Trends Ecol. Evol.,
23(10): 538–545.
Fold, N. and Whitefield, L., 2012. Developing a Palm Oil Sector: The Experiences of
Malaysia and Ghana Compared, Danish Institute for International Studies,
Copenhagen, Denmark, pp. 43.
Foong, S.F. and Lee, C.T., 2000. Increasing Oil Palm Productivity with Irrigation -
Felda's experience. In: E. Pushparajah (Editor), International Planters
Conference on Plantation Tree Crops in the New Millennium: The Way Ahead.
International Society of Planters, Kuala Lumpur, pp. 277–301.
Foote, K.A., Hill, K.H. and Martin, L.G., 1993. Demographic Change in Sub-Saharan
Africa. The National Academies Press, Washington DC, USA, pp. 396.
Foster, H., 2003. Assessment of Oil Palm Fertilizer Requirements. In: R. Härdter and
T.H. Fairhurst (Editors), Oil Palm - Management for Large and Sustainable
Yields. International Plant Nutrition Institute (IPNI) and International Potash
Institute (IPI), Singapore, pp. 231–258.
Foster, H.L. and Prabowo, N.E., 2006. Partition and transfer of nutrients in the reserve
tissues and leaves of oil palm, Workshop on Nutrient Needs in Oil Palm,
Singapore, pp. 1–18.
Gilbert, D., 2013. Oil palm and palm oil industry in Ghana: a brief history. Int. Res. J.
Plant Sci., 4(6): 158–167.
Gillbanks, R.A., 2003. Standard agronomic procedures and practices. In: T. Fairhurst
and R. Härdter (Editors), The Oil Palm – Management for Large and Sustainable
Yields. International Plant Nutrition Institute (IPNI) and International Potash
Institute (IPI), Singapore, pp. 115–150.

143
REFERENCES

Giller, K.E. and Fairhurst, T., 2003. Legume Cover Plants. In: T. Fairhurst and R.
Härdter (Editors), Oil Palm. Management for Large and Sustainable Yields.
International Plant Nutrition Institute (IPNI) and International Potash Institute
(IPI), Singapore, pp. 151–161.
Gilroy, J.J., Prescott, G.W., Cardenas, J.S., Castañeda, P.G.D.P., Sánchez, A., Rojas-
Murcia, L.E., Uribe, C.A.M., Haugaasen, T. and Edwards, D.P., 2015.
Minimizing the biodiversity impact of Neotropical oil palm developmen. Global
Change Biol., 21: 1531–1540.
Goh, K., 2000. Climatic requirements of the oil palm for high yields. In: K. Goh
(Editor), Managing Oil Palm for High Yields: Agronomic Principles. Malaysian
Society of Soil Science and Param Agricultural Surveys and Kuala Lumpur, pp.
1–17.
Goh, K., Härdter, R. and Fairhurst, T., 2003. Fertilizing for maximum return. In: T.
Fairhurst and R. Härdter (Editors), The Oil Palm – Management for Large and
Sustainable Yields. International Plant Nutrition Institute (IPNI) and
International Potash Institute (IPI), Singapore, pp. 279–306.
Goh, K.J. and Härdter, R., 2003. General Oil Palm Nutrition. In: T. Fairhurst and R.
Härdter (Editors), The Oil Palm - Management for Large and Sustainable Yields.
International Plant Nutrition Institute (IPNI) and International Potash Institute
(IPI), Singapore, pp. 191–230.
Goh, K.J., Chew, P.S., 1997. Interpretations of analytical data from soil survey reports
for manuring recommendations: some pointers. Annual Journal/Report 1997.
Royal Johore Planters' Association, pp. 25–30.
Goh, K.J., Chew, P.S., Teo, C.B. and Chee, K.H., 1994. Maximising and maintaining
oil palm yields on commercial scale in Malaysia, International Planters
Conference on Management for Enhanced Profitability in Plantations. Kuala
Lumpur, 24–26 October 1994. ISP, pp. 121–141.
Goh, K.J., Gan, H.H. and Heng, Y.C., 2004. Yield Targeting and Yield Gap Analysis
in Oil Palm Plantations. In: P.S. Chew and Y.P. Tan (Editors), Proceedings of
Agronomy and Crop Management Workshops March to August 2004.
Malaysian Oil Scientists' and Technologists' Association (MOSTA), Petaling
Jaya, pp. 381–403.
Green, R.E., Cornell, S.J., Scharlemann, J.P.W. and Balmford, A., 2005. Farming and
the fate of wild nature. Science, 307: 550–555.

144
 REFERENCES

Griffiths, W. and Fairhurst, T., 2003. Implementation of best management practices in

an oil palm rehabilitation project. Better Crops, 17(1): 16–19.
Griffiths, W., Fairhurst, T., Rankine, I., Kerstan, A. and Taylor, C., 2002. Identification
and elimination of yield gaps in oil palm. Use of OMP7 and GIS, Proceedings
of the 2002 International Oil Palm Conference and Exhibition. Indonesian Oil
Palm Research Institute (IOPRI), Bali, Indonesia, 8–12 July 2002, pp. 1–10.
Guilpart, N., Grassini, P., Sadras, V.O., Timsina, J. and Cassman, K.G., 2017.
Estimating yield gaps at the cropping system level. Field Crops Res., 206: 21–
32.
Gyasi, E.A., 1992. Emergence of a new oil palm belt in Ghana. Tijdschr. Econ. Soc.
Geogr., 83(1): 39–49.
Gyasi, E.A., 1996. The environmental impact and sustainability of plantations in Sub-
Saharan Africa: Ghana's experiences with oil-palm plantations. In: E.A. Gyasi
and J. Uitto (Editors), Sustaining the Future: Economic, Social and
Environmental Change in Sub-Saharan Africa. United Nations University Press,
Tokyo, Japan, pp. 342–357.
Hansen, S.B., Padfield, R., Syayuti, K., Evers, S., Zakariah, Z. and Mastura, S., 2015.
Trends in global palm oil sustainability research. J. Clean. Prod., 100: 140–149.
Härdter, R., 1999. A Review of Magnesium Nutrition in Oil Palm, ISPNP Branch
Seminar on Reforming the Mindset of Plantation Management into the Next
Millenium, Taipeng, Perak, Malaysia, pp. 1–13.
Hargreaves, G.H. and Samani, Z.A., 1982. Estimating potential evapotranspiration. J.
Irr. Drain. Div., 108(3): 225–230.
Hartley, C.W.S., 1977. The Oil Palm (Elaeis guineensis Jacq.). Tropical Agriculture
Series. Longman, London, UK, pp. 806.
Hartley, C.W.S., 1988. The Oil Palm. Tropical Agriculture Series. Longman Scientific
& Technical, London/New York, pp. 761.
Hayward, D.F. and Oguntoyinbo, J.S., 1987. The Climatology of West Africa.
Hutchinson, London, UK, pp. 271.
Henson, I.E. and Dolmat, M.T., 2004. Seasonal variation in yield and developmental
processes in an oil palm density trial on a peat soil: 1. yield and bunch number
components. J. Oil Palm Res., 16(2): 88–105.

145
REFERENCES

Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. and Jarvis, A., 2005. Very high
resolution interpolated climate surfaces for global land areas. Int. J. Climatol.,
25(15): 1965–1978.
Hochman, Z., Gobbett, D., Horan, H. and Navarro Garcia, J., 2016. Data rich yield gap
analysis of wheat in Australia. Field Crops Res., 197: 97–106.
Hoffmann, M.P., Castaneda Vera, A., van Wijk, M.J., Giller, K.E., Fairhurst, T.,
Oberthür, T. Donough, C., Whitbread, A.M., 2014. Simulating potential growth
and yield of oil palm (Elaeis guineensis) with PALMSIM: Model description,
evaluation and application. Agric. Syst., 131: 1–10.
Hoffmann, M.P., Donough, C., Oberthür, T., Castaneda Vera, A., van Wijk, M.T., Lim,
C.H., Asmono, D., Samosir, Y., Lubis, A.P., Moses, D.S. and Whitbread, A.M.,
2015. Benchmarking yield for sustainable intensification of oil palm production
in Indonesia using PALMSIM. The Planter, 91(1067): 81–96.
Hoffmann, M.P., Donough, C.R., Cook, S.E., Fisher, M.J., Lim, C.H., Lim, Y.L., Cock,
J., Kam, S.P., Mohanaraj, S.N., Indrasuara, K., Tittinutchanon, P. and Oberthür,
T., 2017. Yield gap analysis in oil palm: Framework development and
application in commercial operations in Southeast Asia. Agric. Syst., 151: 12–
19.
IPNI, 2012. 4R Plant Nutrition Manual: A Manual for Improving the Management of
Plant Nutrition. International Plant Nutrition Institute, Norcross, GA, USA, pp.
130.
IPPA, 2010. African Case Study: Palm Oil and Economic Development in Nigeria and
Ghana; Recommendations for the World Bank's 2010 Palm Oil Strategy, pp. 12.
Ismail, A., 2013. The effect of labour shortage in the supply and demand of palm oil in
Malaysia. Oil palm Ind. Econ. J., 13(2): 15–26.
ISRIC, 2013. SoilGrids: An Automated System for Global Soil Mapping. URL:
http://www.soilgrids.org (accessed on 02.10.2014).
IUCN and UNEP-WCMC, 2014. The World Database on Protected Areas (WDPA).
URL: http://protectedplanet.net (accessed on 09.10.2014).
Jarvis, A., Reuther, H.I., Nelson, A. and Guevara, E., 2008. Hole-filled seamless SRTM
data V4. URL: http://srtm.csi.cgiar.org (accessed on 29.08.2014).
Jayaselan, H.A.J. and Ahmad, D., 2011. Development of a mechanization selection
system for oil palm plantations with alternative planting patterns. J. Oil Palm
Res., 23: 990–998.

146
 REFERENCES

Jelsma, I., Slingerland, M., Giller, K.E. and Bijman, J., 2017. Collective action in a
smallholder oil palm production system in Indonesia: The key to sustainable and
inclusive smallholder palm oil? J. Rural Stud., 54: 198–210.
Jones, L., 1997. The effects of leaf pruning and other stresses on sex determination in
the oil palm and their representation by a computer simulation. J. Theor. Biol.,
187: 241–260.
Kasahara, M., Kagawa, T., Olkawa, K., Suetsugu, N., Miyao, M. and Wada, M., 2002.
Chloroplast avoidance movement reduces photodamage in plants. Nature,
420(6917): 829–832.
Kee, K.K. and Chew, P.S., 1991. Oil palm responses to nitrogen and drip irrigation in
a wet monsoonal climate in Peninsular Malaysia, International Palm Oil
Conference - Progress, Prospects & Challenges Towards the 21st Century -
Module I Agriculture. Palm Oil Research Institute of Malaysia (PORIM), Kuala
Lumpur, Malaysia, 9–14 September 1991, pp. 321–339.
Koh, L.P. and Wilcove, D.S., 2008. Is oil palm agriculture really destroying tropical
biodiversity? Conserv. Lett., 1(2): 60–64.
Kongor, J.E., Boeckx, P., Vermeir, P., Van de Walle, D., Baert, G., Afoakwa, E.O. and
Dewettinck, K., 2018. Assessment of soil fertility and quality for improved
cocoa production in six cocoa growing regions in Ghana. Agrofor. Syst.
https://doi.org/10.1007/s10457-018-0253-3.
Krausova, M. and Banful, A.B., 2010. Overview of the Agricultural Input Sector in
Ghana, International Food Policy Research Institute, pp. 36.
Laborte, A.G., de Bie, K., Smaling, E.M.A., Moya, P.F., Boling, A.A. and Van Ittersum,
M.K., 2012. Rice yields and yield gaps in Southeast Asia: Past trends and future
outlook. Eur. J. Agron., 36(1): 9–20.
Läderach, P., Martinez-Valle, A., Schroth, G. and Castro, N., 2013. Predicting the future
climatic suitability for cocoa farming of the world’s leading producer countries,
Ghana and Côte d’Ivoire. Clim. Change, 119(3–4): 841–854.
Lamade, E., Bonnot, F. and Poeloengan, Z., 1996. Modelling photosynthesis and
stomatal response of oil palm clones to environmental conditions in North
Sumatra (Indonesia) - implication for plant breeding. In: D. Ariffin, M.B. Wahid,
N. Rajanaidu, M.D. Tayeb, K. Paranjothy, S.C. Cheah, K.C. Chang and S.
Ravigadevi (Editors), PORIM Int. Palm Oil Congr. 'Competitiveness for the 21st
century'. Palm Oil Res. Inst. Malaysia, Kuala Lumpur, Malaysia, pp. 87–96.

147
REFERENCES

Lamade, E., Setiyo, E. and Purba, A., 1998. Gas exchange and carbon allocation of oil
palm seedlings submitted to water logging in interaction with N fertiliser
application. In: A. Jatmika, D. Bangun, D. Asmono, E.S. Sutarta, K. Pamin, P.
Guritno, S. Prawirosukarto, T. Wakyono, T. Herawan, T. Hutomo, W.
Darmosarkoro, Y.T. Adiwiganda and Z. Poeloengan (Editors), 1998
International Oil Palm Conference. Commodity of the Past, Today and the
Future. Indonesian Oil Palm Research Institute, Medan, pp. 573–584.
Laurance, W.F., Koh, L.P., Butler, R., Sodhi, N.S., Bradshaw, C.J., Neidel, J.D.,
Consunji, H. and Mateo Vega, J., 2010. Improving the performance of the
Roundtable on Sustainable Palm Oil for nature conservation. Conserv. Biol.,
24(2): 377–381.
Laurance, W.F., Sayer, J. and Cassman, K.G., 2014. Agricultural expansion and its
impacts on tropical nature. Trends Ecol. Evol., 29(2): 107–116.
Lee, C., Rahman, Z., Musa, M., Norizan, M. and Tan, C., 2011. Leaf nutrient
concentrations in oil palm as affected by genotypes, irrigation and terrain.
Journal of Oil Palm & The Environment, 2: 38–47.
Lee, C.T., Rahman, Z.A., Hanafi, M.M., Ishak, C.F., Norizan, M.S., Tan, C.C. and
Yusof, M.S.M., 2014. Rachis nutrient concentrations of different oil palm
genotypes as affected by irrigation and terrain. J. Oil Palm Res., 26(2): 146–153.
Lee, J.S.H., Garcia-Ulloa, J., Ghazoul, J., Obidzinski, K., Koh, L.P. and Jones, J., 2014.
Modelling environmental and socio-economic trade-offs associated with land-
sparing and land-sharing approaches to oil palm expansion. J. Appl. Ecol., 51(5):
1366–1377.
Legros, S., Mialet-Serra, I., Caliman, J.P., Siregar, F.A., Clement-Vidal, A. and
Dingkuhn, M., 2009. Phenology and growth adjustments of oil palm (Elaeis
guineensis) to photoperiod and climate variability. Ann. Bot., 104(6): 1171–
1182.
Lemoalle, J. and de Condappa, D., 2012. Farming systems and food production in the
Volta Basin. In: M.J. Fisher and S.E. Cook (Editors), Water, Food and Poverty
in River Basins: Defining the Limits. Routledge, Abington, UK, pp. 192–217.
Lim, K.H., Goh, K.J., Kee, K.K. and Henson, I.E., 2011. Climatic Requirements of Oil
Palm. In: K.J. Goh, S.B. Chiu and S. Paramananthan (Editors), Agronomic
Principles & Practices of Oil Palm Cultivation. Agricultural Crop Trust (ACT),
Malaysia, pp. 771.

148
 REFERENCES

Lobell, D.B., 2013. The use of satellite data for crop yield gap analysis. Field Crops
Res., 143: 56–64.
Magne, A.N., Nonga, N.E., Yemefack, M. and Robiglio, V., 2014. Profitability and
implications of cocoa intensification on carbon emissions in Southern
Cameroun. Agrofor. Syst., 88(6): 1133–1142.
Maillard, G., Daniel, C. and Ochs, R., 1974. Analyse des effets de la sécheresse sur le
palmier à huile. Oléagineaux, 42(8–9): 397–404.
Malczewski, J., 2004. GIS-based land-use suitability analysis: a critical overview. Prog.
Plann., 62(1): 3–65.
Manzanas, R., Amekudzi, L.K., Preko, K., Herrera, S. and Gutiérrez, J.M., 2014.
Precipitation variability and trends in Ghana: an intercomparison of
observational and reanalysis products. Clim. Change, 124(4): 805–819.
MASDAR, 2011. Ghana's Oil Palm Master Plan. Final Report, Ministry of Food and
Agriculture (MOFA), pp. 234.
McSweeney, C., New, M. and Lizcano, G., 2010a. UNDP Climate Change Country
Profiles: Ghana. URL: http://country-profiles.geog.ox.ac.uk/ (accessed on
09.07.2015).
McSweeney, C., New, M., Lizcano, G. and Lu, X., 2010b. The UNDP Climate Change
Country Profiles Improving the Accessibility of Observed and Projected Climate
Information for Studies of Climate Change in Developing Countries. B. Am.
Meteorol. Soc., 91: 157–166.
Mendes-Oliveira, A.C., Peres, C.A., Maues, P., Oliveira, G.L., Mineiro, I.G.B., de
Maria, S.L.S. and Lima, R.C.S., 2017. Oil palm monoculture induces drastic
erosion of an Amazonian forest mammal fauna. PLoS One, 12(11): e0187650.
Mensah-Bonsu, A., Amegashie, D.P.K. and Gyasie, S., 2009. The structure of labour
market and demand for hired labour for oil palm production in the western region
of Ghana. J. Sci. Technol., 29(3): 102–113.
Murphy, D.J., 2009. Oil palm: future prospects for yield and quality improvements.
Lipid Technol., 21(11/12): 257–260.
Murphy, D.J., 2014. The future of oil palm as a major global crop: opportunities and
challenges. J. Oil Palm Res., 26(1): 1–24.
National Geographic, 2018. Palm oil is unavoidable. Can it be sustainable? URL:
https://www.nationalgeographic.com/magazine/2018/12/palm-oil-products-
borneo-africa-environment-impact/ (accessed on 08.01.2019).

149
REFERENCES

Nelson, P.N., Banabas, M., Scotter, D.R. and Webb, M.J., 2006. Using soil water
depletion to measure spatial distribution of root activity in oil palm
(Elaeis guineensis Jacq.) plantations. Plant Soil, 286(1–2): 109–121.
Neumann, K., Verburg, P.H., Stehfest, E. and Müller, C., 2010. The yield gap of global
grain production: A spatial analysis. Agric. Syst., 103(5): 316–326.
Ng, S.K., 1972. The Oil Palm, Its Culture, Manuring and Utilisation. International
Potash Institute, Basel, Switzerland, pp. 142.
Ng, S.K., 1983. Advances in Oil Palm Nutrition, Agronomy and Productivity in
Malaysia, Palm Oil Research Institute of Malaysia (PORIM), Kuala Lumpur,
Malaysia, pp. 20.
Ng, S.K., Cheong, T., Haw, K., Ooi, H., Yee, L., Kayaroganam, P., Von Uexkull, H.
and Härdter, R., 2003. Clonal Oil Palm Production, Yield Performance, and
Nutritional Requirements. In: T. Fairhurst and R. Härdter (Editors), The Oil
Palm – Management for Large and Sustainable Yields. International Plant
Nutrition Institute (IPNI) and International Potash Institute (IPI), Singapore, pp.
99–114.
Nouy, B., Baudouin, L., Djégui, N. and Omoré, A., 1999. Oil palm under limiting water
supply conditions. Plantations, Recherche, Développement, 6(1): 31–45.
Nouy, B., Omore, A. and Potier, F., 1996. Oil palm production cycles in different
ecologies: consequences for breeding. In: D. Ariffin, W.M. Basri, R. Nookiah,
H.T. Dolmat, P. K., C. Suan-Choo, C.K. Choong and S. Ravigadevi (Editors),
Proceedings of the 1996 PORIM International Palm Oil Congress:
competitiveness for the 21st Century. Agriculture Conference. Palm Oil
Research Institute of Malaysia, Kuala Lumpur, Malaysia, pp. 62–75.
Oberthür, T., Donough, C.R., Indrasuara, K., Dolong, T. and Abdurrohim, G., 2012.
Successful intensification of oil palm plantations with best management
practices: impacts on fresh fruit bunch and oil yield. The Planter, 89: 185–216.
Ochs, R. and Daniel, C., 1976. Research on Techniques Adapted to Dry Regions. In:
R.H.V. Corley, J.J. Hardon and B.J. Wood (Editors), Oil Palm Research.
Elsevier, Amsterdam, the Netherlands, pp. 315–330.
Ochs, R. and Olivin, J., 1977. Leaf analysis for the control of nutrition in oil palm
plantations - Taking of leaf samples. Oléagineux, 32(5): 211–216.
Ochs, R. and Ollagnier, M., 1977. Effect of fertilizers on composition of lipids produced
by perennial tropical oil plants and on their yield. Oléagineux, 32(10): 409–426.

150
 REFERENCES

Ofosu-Budu, K. and Sarpong, D., 2013. Oil Palm Industry Growth in Africa: A Value
Chain and Smallholders Study for Ghana. In: A. Elbehri (Editor), Rebuilding
West Africa's Food Potential. FAO/IFAD, pp. 349–389.
Olivin, J., 1966. Pointe annuelle de production des palmeraies au Dahomey et cycle
annuel de development du palmier a huile. Oléagineaux, 18: 231–238.
Olivin, J., 1968. Etude pour la localisation d'un bloc industriel de palmiers à huile II.
Le criteria de jugement. Oléagineux, 23(8–9): 499–504.
Ollagnier, M. and Ochs, R., 1981. Le diagnostic foliaire pour le controle de la nutrition
des plantations de palmiers à huile: Prélèvement des échantillions foliares.
Oléagineaux, 32(5): 211–216.
Ong, H.T., 1982a. System approach to the climatology of oil palm .1. Identification of
rainfall and dry spell aspects. Oléagineux, 37(3): 93–105.
Ong, H.T., 1982b. System approach to the climatology of oil palm .2. Identification of
temperature and sunshine aspects. Oléagineux, 37(10): 443–453.
Oppong-Anane, K., 2006. Ghana: Country Pasture/Forage Resource Profiles. FAO.
URL: http://www.fao.org/ag/AGP/AGPC/doc/Counprof/Ghana.htm (accessed
on 11.09.2014).
Osei-Amponsah, C., Visser, L., Adjei-Nsiah, S., Struik, P.C., Sakyi-Dawson, O. and
Stomph, T.J., 2012. Processing practices of small-scale palm oil producers in the
Kwaebibirem District, Ghana: A diagnostic study. NJAS-Wagen. J. Life Sc., 60–
63: 49–56.
Owusu-Bennoah, E., Awadzi, T.W., Boateng, E., Krogh, L., Breuning-Madsen, H. and
Borggaard, O.K., 2000. Soil properties of a toposequence in the moist seim-
deciduous forest zone of Ghana. West Afr. J. Appl. Ecol., 1: 1–10.
Owusu, K., 2009. Changing Rainfall Climatology of West Africa: Implications for
Rainfed Agriculture in Ghana and Water Sharing in the Volta Basin. Doctoral
thesis, University of Florida, Florida, USA, pp. 85.
Paramananthan, S., 2003. Land Selection for Oil Palm. In: R. Härdter and T.H. Fairhurst
(Editors), Oil Palm - Managament for Large and Sustainable Yields. Potash &
Phosphate Institute/Potash & Phosphate Institute of Canada (PPI/PPIC) and
International Potash Institute (IPI), Singapore, pp. 27–58.
Paramananthan, S., 2011. Soil Requirements and Land Evaluation for Oil Palms for
High Yields. In: K.J. Goh, S.B. Chiu and S. Paramananthan (Editors),
Agronomic Principles & Practices of Oil Palm Cultivation. Agricultural Crop
Trust (ACT), Malaysia, pp. 47–88.

151
REFERENCES

Pauli, N., Donough, C., Oberthür, T., Cock, J., Verdooren, R., Rahmadsyah,
Abdurrohim, G., Indrasuara, K., Lubis, A., Dolong, T. and Pasuquin, J.M., 2014.
Changes in soil quality indicators under oil palm plantations following
application of ‘best management practices’ in a four-year field trial. Agric.,
Ecosyst. Environ., 195: 98–111.
Phalan, B., Fitzherbert, E.B., Rafflegeau, S., Struebig, M.J. and Verwilghen, A., 2009.
Conservation in oil-palm landscapes. Conserv. Biol., 23(2): 244–5; author reply
245–6.
Phalan, B., Onial, M., Balmford, A. and Green, R.E., 2011. Reconciling food
production and biodiversity conservation: land sharing and land sparing
compared. Science, 333(6047): 1289–1291.
Phalan, B.T., 2018. What have we learned from the land sparing-sharing model?
Sustainability, 10: 1–24.
Phelps, J., Carrasco, L.R., Webb, E.L., Koh, L.P. and Pascual, U., 2013. Agricultural
intensification escalates future conservation costs. PNAS, 110(19): 7601–7606.
Prioux, G., Jacquemard, J., de Franqueville, H. and Caliman, J., 1992. Oil palm
irrigation. Initial results obtained by PHCI (Ivory Coast). Oléagineux, 47(8–9):
497–509.
Purvis, C., 1956. The root system of the oil palm: its distribution, morphology and
anatomy. Journal of the West African Institute for Oil Palm Research, 1(4): 60–
82.
Quencez, P., 1996. La culture du palmier à huile en Afrique intertropicale: les
conditions du milieu physique. Ol. Corps Gras, Lipides, 39(3): 116–118.
Quencez, P., de Berchoux, C., Hamel, P., N'Guessan, B. and Dubos, B., 1987.
Valorisation des zones marécageuses par le palmier à huile. Oléagineaux, 42(2):
43–52.
Rankine, I.R. and Fairhurst, T.H., 1998. Field Handbook. Oil Palm Series Volume 3.
Mature. Potash & Phosphate Institute (PPI), Potash & Phosphate Institute of
Canada (PPIC) and 4T Consultants (4T), Singapore, pp. 135.
Rankine, I.R. and Fairhurst, T.H., 1999. Field Handbook. Oil Palm Series Volume 2.
Immature. International Plant Nutrition Institute (IPNI), PT Agrisoft Systems
Indonesia, Singapore, pp. 154.
Rees, A.R., 1961. Midday Closure of Stomata in the Oil Palm Elaeis guineensis. Jacq.
J. Exp. Bot., 12(1): 129–146.

152
 REFERENCES

Rhebergen, T., Fairhurst, T., Whitbread, A., Giller, K.E. and Zingore, S., 2018. Yield
gap analysis and entry points for improving productivity on large oil palm
plantations and smallholder farms in Ghana. Agric. Syst., 165: 14–25.
Rhebergen, T., Fairhurst, T., Zingore, S., Fisher, M., Oberthür, T. and Whitbread, A.,
2016. Climate, soil and land-use based land suitability evaluation for oil palm
production in Ghana. Eur. J. Agron., 81: 1–14.
Rhebergen, T., Hoffmann, M., Zingore, S., Oberthür, T., Acheampong, K., Dwumfour,
G., Zutah, V., Adu-Frimpong, C., Ohipeni, F. and Fairhurst, T., 2014. The
Effects of Climate, Soil and Oil Palm Management Practices on Yield in Ghana,
International Oil Palm Conference (IOPC), Bali, Indonesia, pp. 1–16.
RSPO, 2013. RSPO Principles and Criteria for the Production of Sustainable Palm Oil,
Roundtable on Sustainable Palm Oil (RSPO), Kuala Lumpur, Malaysia, pp. 58.
RSPO, 2015. Smallholders. Roundtable on Sustainable Palm Oil (RSPO). URL:
http://www.rspo.org/smallholders (accessed 22.02.2015).
Sayer, J., Ghazoul, J., Nelson, P. and Klintuni Boedhihartono, A., 2012. Oil palm
expansion transforms tropical landscapes and livelihoods. Glob. Food Sec., 1(2):
114–119.
Schnug, E., Heym, J. and Achwan, F., 1996. Establishing critical values for soil and
plant analysis by means of the boundary line development system (bolides).
Commun. Soil Sci. Plant Anal., 27(13–14): 2739–2748.
Shamsul Bahrin, T. and Lee, B.T., 1988. Three decades of evolution. FELDA, Kuala
Lumpur, Malaysia, pp. 233.
Shuib, A.R., Khalid, M.R. and Deraman, M.S., 2010. Enhancing field mechanization in
oil palm management. Oil Palm Bulletin, 61: 1–10.
Simmonds, N.W. (Editor), 1976. The Evolution of Crop Plants. Longman, London, UK,
pp. 339.
Smith, B.G., 1989. The effects of soil water and atmospheric vapour pressure deficit on
stomatal behaviour and photosynthesis in the oil palm. J. Exp. Bot., 40(6): 647–
651.
Soliman, T., Lim, F.K.S., Lee, J.S.H. and Carrasco, L.R., 2016. Closing oil palm yield
gaps among Indonesian smallholders through industry schemes, pruning,
weeding and improved seeds. Royal Society Open Science, 3(8): 160292.
Sparnaaij, L.D., Rees, A.R. and Chapas, L.C., 1963. Annual yield variation in the oil
palm. J. W. Afr. Inst. Oil Palm Res., 4: 111–125.

153
REFERENCES

Stanturf, J.A., Warren, M.L., Charnley, S., Polasky, S.C., Goodrick, S.L., Armah, F.
and Nyako, Y.A., 2011. Ghana Climate Change Vulnerability and Adaptation
Assessment, United States Agency for International Development (USAID),
Washington, USA, pp. 244.
Stout, B., 1990. Handbook of Energy for World Agriculture. Elsevier Science
Publishers, London, UK, pp. 504.
Surre, C., 1968. Les besions en eau du palmier à huile. Calcul du bilan de l'eau et ses
applications pratiques. Oléagineux, 23(3): 165–167.
Swaine, M.D., 1996. Rainfall and soil fertility as factors limiting forest species
distributions in Ghana. J. Ecol., 84(3): 419–428.
Tailliez, B., 1971. The root system of the oil palm on the San Alberto plantation in
Colombia. Oléagineux, 26(7): 435–447.
Tampubolon, F.H., Daniel, C. and Ochs, R., 1990. Oil palm responses to nitrogen and
phosphate fertilizer in Sumatra. Oléagineaux, 45(11): 475–486.
Taulya, G., 2013. East African highland bananas (Musa spp. AAA-EA) ‘worry’ more
about potassium deficiency than drought stress. Field Crops Res., 151: 45–55.
Teo, L., 2000. Mechanisation in Oil Palm Plantations: Achievement and Challenges,
Abstracts from 'Oils and Fats International Congress 2000, pp. 70–77.
Tiemann, T.T., Donough, C.R., Lim, Y.L., Härdter, R., Norton, R., Tao, H.H.,
Jaramillo, R., Satyanarayana, T., Zingore, S. and Oberthür, T., 2018. Chapter
Four - Feeding the Palm: A Review of Oil Palm Nutrition. Adv. Agron., 152:
149–243.
Tilman, D., Fargione, J., Wolff, B., D'Antonio, C., Dobson, A., Howarth, R., Schindler,
D., Schlesinger, W.H., Simberloff, D. and Swackhamer, D., 2001. Forecasting
agriculturally driven global environmental change. Science, 292(5515): 281–
284.
Tittinutchanon, P., Nakharin, C., Clendon, J.H. and Corley, R.H.V., 2008. A review of
15 years of oil palm irrigation research in southern Thailand. The Planter,
84(989): 537–546.
UN, 2013. Population Density and Urbanization. URL:
http://unstats.un.org/unsd/demographic/sconcerns/densurb/densurbmethods.ht
m (accessed on 13.05.2015).

154
 REFERENCES

UNDP, 2011. Africa Adaptaion Programme: Capacity Building Experiences.

Improving Access, Understanding and Application of Climate Data and
Information, United Nations Development Programme, New York, USA, pp. 20.
van der Vossen, H.A.M., 1969. Areas climatically suitable for optimal oil palm
production in the forest zone of Ghana. Ghana J. Agric. Sci., 2(2): 113–118.
van der Vossen, H.A.M., 1970. Nutrient status and fertilizer responses of oil palms on
different soils in the forest zone of Ghana. Ghana J. Agric. Sci., 3: 109–129.
van der Vossen, H.A.M., 1974. Towards more efficient selection for oil yield in the oil
palm (Elaeis guineensis Jacquin). Doctoral thesis, Wageningen University,
Wageningen, The Netherlands, pp. 107.
van Ittersum, M.K. and Rabbinge, R., 1997. Concepts in production ecology for
analysis and quantification of agricultural input-output combinations. Field
Crops Res., 52: 197–208.
van Ittersum, M.K., Cassman, K.G., Grassini, P., Wolf, J., Tittonell, P. and Hochman,
Z., 2013. Yield gap analysis with local to global relevance—A review. Field
Crops Res., 143: 4–17.
van Wambeke, A., 1974. Management Properties of Ferralsols. FAO Soils Bulletin 23.
FAO, Rome, Italy, pp. 137.
Vijay, V., Pimm, S.L., Jenkins, C.N. and Smith, S.J., 2016. The impacts of oil palm on
recent deforestation and biodiversity loss. PLoS One 11(7): e0159668.
https://doi.org/10.1371/journal.pone.0159668.
von Uexkull, H.R. and Fairhurst, T.H., 1991. Fertilizing for High Yield and Quality.
The Oil Palm, International Potash Institute, Bern, Swizterland, pp. 79.
Wairegi, L.W.I., van Asten, P.J.A., Tenywa, M.M. and Bekunda, M.A., 2010. Abiotic
constraints override biotic constraints in East African highland banana systems.
Field Crops Res., 117(1): 146–153.
Wang, M., Zheng, Q., Shen, Q. and Guo, S., 2013. The critical role of potassium in
plant stress response. Int. J. Mol. Sci., 14(4): 7370–90.
Wang, N., Jassogne, L., van Asten, P.J.A., Mukasa, D., Wanyama, I., Kagezi, G. and
Giller, K.E., 2015. Evaluating coffee yield gaps and important biotic, abiotic,
and management factors limiting coffee production in Uganda. Euro. J. Agron.,
63: 1–11.
White, E.J., 1979. The prediction and selection of climatological data for ecological
purposes in Great Britain. J. Appl. Ecol., 16(1): 141–160.

155
REFERENCES

Wich, Serge A., Garcia-Ulloa, J., Kühl, Hjalmar S., Humle, T., Lee, Janice S.H. and
Koh, Lian P., 2014. Will Oil Palm’s Homecoming Spell Doom for Africa’s Great
Apes? Curr. Biol., 24(14): 1659–1663.
Wicke, B., Sikkema, R., Dornburg, V. and Faaij, A., 2011. Exploring land use changes
and the role of palm oil production in Indonesia and Malaysia. Land Use Policy,
28(1): 193–206.
Witt, C. and Donough, C.R., 2007. Oil palm management: ecological yield
intensification through BMP, Poster presented at 5th Roundtable (RT5) on
Sustainable Palm Oil, Kuala Lumpur, Malaysia.
Woittiez, L.S., van Wijk, M.T., Slingerland, M., van Noordwijk, M. and Giller, K.E.,
2017. Yield gaps in oil palm: A quantitative review of contributing factors. Euro.
J. Agron., 83: 57–77.
Yang, W., Peng, S., Laza, R.C., Visperas, R.M. and Dionisio-Sese, M.L., 2008. Yield
Gap Analysis between Dry and Wet Season Rice Crop Grown under High-
Yielding Management Conditions. Agron. J., 100(5): 1390–1395.
Yue, S., Brodie, J.F., Zipkin, E.F. and Bernard, H., 2015. Oil palm plantations fail to
support mammal diversity. Ecol. Appl., 25(8): 2285–2292.
Zeven, A.C., 1964. On the Origin of the Oil Palm (Elaeis Guineensis Jacq.). GRANA,
5(1): 121-123.
Zeven, A.C., 1972. The partial and complete Domestication of the oil palm (Elaeis
guineensis). Econ. Bot., 26(3): 274–279.
Zulkifli, Y., Norziha, A., Naqiuddin, M.H., Fadila, A.M., Nor Azwani, A.B., Suzana,
M., Samsul, K.R., Ong-Abdullah, M., Singh, R., Parveez, G.K.A. and Kushairi,
A., 2017. Designing the oil palm of the future. J. Oil Palm Res., 29(4): 440–455.

156
 APPENDICES
Appendix 1. Map showing six agro-ecological zones in Ghana.

157
APPENDICES

Appendix 2. Soil units of Ghana (FAO, ISRIC).

158
 Appendix 3. Data sources.
Data Variable Unit Format Resolution Period/year Source
Protected Protected areas - Shapefile - - Protectedplanet; http://protectedplanet.net
areas Accessed on 09.10.2014
Land cover Agro-ecological - Shapefile - - (Antwi-Agyei et al., 2012)
zones
Urban Urbanized places - Raster 30 arc- 2000 NASA; http://sedac.ciesin.columbia.edu/data/set/grump-v1-
extent (places with 5000+ seconds (~1 urban-extents/data-download
inhabitants) km) Accessed on 09.10.2014
Rainfall Average monthly mm Raster 30 arc- 1950–2000 WorldClim; http://www.worldclim.org
rainfall month- seconds (~1 Accessed on 29.08.2014
1
km)
Temperature Monthly min-, max-, °C Raster 30 arc- 1950–2000 WorldClim; http://www.worldclim.org
and mean seconds (~1 Accessed on 29.08.2014
temperature km)

159
Min temp of the
coldest month
Max temp of
warmest month
Elevation M Raster 3 arc- 2008 CGIAR; http://srtm.csi.cgiar.org
ASL seconds Accessed on 29.08.2014
(~100 m)
Soil Soil texture % Silt Raster 30 arc- - ISRIC; http://soilgrids1km.isric.org
properties % seconds (~1 Accessed on 02.10.2014
Clay km)
Soil type unit Raster 30 arc- 2012 FAO; http://www.fao.org/soils-portal/soil-survey/soil-maps-
seconds (~1 and-databases/harmonized-world-soil-database-v12/en/
km) Accessed on 01.09.2014
APPENDICES
 Appendix 4. Average rainfall, rain days, water deficits (calculated using the method of Surre (1968)) and their standard deviation
(σ) at five sites in Ghana from 2010 to 2014.
Site Region and geo-reference Parameter Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
Bogoso Western region Rainfall (mm) 41 71 113 175 243 258 179 60 211 194 83 42 1,670
5°34'06.81"N σ 27 31 70 76 63 122 99 31 128 83 32 11 368
APPENDICES

2°00'58.94"W Rain days 4 8 10 12 19 16 16 14 20 20 14 4 157

σ 2 1 5 2 1 3 3 4 6 2 4 2 16
Water deficit (mm) -70 -65 -23 76 193 253 198 62 147 173 73 -35 -193
σ 80 47 84 113 120 128 98 30 163 111 60 57 174
BOPP Western region Rainfall (mm) 46 111 114 140 214 268 166 67 159 194 158 106 1,743
5°06'47.65"N σ 48 38 14 50 46 96 59 53 69 77 58 64 413
1°54'55.09"W Rain days 2 6 6 7 12 12 9 6 11 12 10 6 99
σ 2 1 1 1 2 3 2 4 3 4 1 3 11
Water deficit (mm) -58 -33 -29 -10 100 225 124 17 61 134 115 51 -131
σ 63 39 19 50 76 171 94 73 98 125 70 93 191
Norpalm Western region Rainfall (mm) 32 34 56 117 220 335 107 44 127 132 98 65 1,367
4°55'26.84"N σ 28 18 29 56 109 90 44 29 74 49 29 34 261

160
1°53'31.54"W Rain days 1 3 3 5 8 13 7 4 12 8 6 3 74
σ 1 1 2 2 3 2 2 3 7 3 1 1 6
Water deficit (mm) -88 -110 -94 -33 96 302 96 -9 10 32 0 -60 -394
σ 65 19 29 56 124 152 34 56 88 75 60 36 185
TOPP Central region Rainfall (mm) 40 41 62 138 201 290 140 84 122 225 92 44 1,479
5°32'40.10"N σ 24 30 43 97 98 112 38 62 96 66 83 27 368
1°31'12.38"W Rain days 2 3 4 6 8 11 7 6 7 11 6 4 74
σ 1 2 3 3 3 2 3 2 5 3 3 2 19
Water deficit (mm) -65 -92 -88 -12 95 235 135 60 49 159 57 -56 -314
σ 82 48 43 97 158 123 37 50 138 98 129 76 258
OPRI Eastern region Rainfall (mm) 43 70 131 162 163 211 168 72 225 250 89 50 1,634
6°2'7.15"N σ 40 46 29 70 31 78 50 54 110 120 19 42 301
0°52'17.04"W Rain days 3 7 10 9 13 16 12 11 15 18 15 6 135
σ 2 2 3 3 3 3 2 6 3 3 2 3 17
Water deficit (mm) -47 -59 -8 34 91 176 173 69 185 225 96 -4 -118
σ 85 60 41 85 50 67 70 93 159 131 37 73 153
 APPENDICES

Appendix 5. Example of the experimental design including paired BMP (blue) and REF
(red) blocks at the nucleus plantation of Twifo Oil Palm Plantation (TOPP).

161
 Appendix 6. Site and block characteristics of plantations and smallholder farms.
Soil
Planting Planting
Plantation Block size Topography group Rainfall
material density
(nucleus)/farm (FAO)
APPENDICES

Year of
Site Block/farm size Second Annual
planting BMP REF
control average
palms
ha mm period
ha-1
Plantations
BOPP 19/03A & B 2003 4,890 25.6 27.1 50.0 DRC Undulating 160 Ferralsol 1,571 2010–
(Zaire) 2015
7/01 & 8/01 2001 45.3 41.2 61.0 DRC Undulating 160 Ferralsol 1,571 2010–
(Zaire) 2015
C4/1 & 2006 38.1 31.1 55.0 DRC Undulating 160 Ferralsol 1,876 2010–
C4/2 (Zaire) 2015
SHDR* 1996 20.8 21.3 19.5 DRC Sloping 160 Ferralsol 1,571 2010–

162
(Zaire) 2015
Norpalm 2 2002 3,760 28.9 19.0 39.5 CIRAD Flat 143 Lixisol 1,136 2011–
2015
25 2010 8.9 8.9 49.3 Ghana Undulating 143 Lixisol 1,668 2010–
Sumatra 2015
92 2005 19.2 21.2 73.3 OPRI Flat 143 Lixisol 1,668 2010–
2015
TOPP 220 & 221 2005 3,250 18.0 22.0 DRC Undulating 160 Lixisol 1,488 2010–
(Zaire) 2015
214 & 224 2004 20.0 20.0 DRC Undulating 160 Lixisol 1,488 2010–
(Zaire) 2015
300 & 301 2005 22.0 21.0 DRC Undulating 160 Lixisol 1,488 2010–
(Zaire) 2015
351 & 352 2010 19.0 23.0 Ghana Undulating 160 Lixisol 1,488 2010–
Sumatra 2015
Datano 1996 10.8 10.3 OPRI Undulating 143 Lixisol n/a
Smallholders
SWAPP East Adu farm 2009 6.8 3.4 3.4 OPRI Undulating 143 n/a n/a
 Appendix 6. Site and block characteristics of plantations and smallholder farms (continued).
Plantation Block size Rainfall
Planting
(nucleus)/farm Second Soil
Year of BMP REF Planting density Annual average
Site Block/farm size control Topography group
planting material
palms (FAO)
ha mm period
ha-1

Amo palm 2005 100 2.7 2.7 TOPP Undulating 160 n/a 1,002 2014–2016
BOA farm 2005 10.8 2.4 2.4 PSI Undulating 143 n/a n/a
Central oil 2005 12 1.4 1.4 TOPP Undulating 160 n/a 1,512 2014–2016
mill
Clement 2005 48 1.4 1.4 TOPP Flat 143 n/a 1,200 2014–2016
farm
Joe farm 2007 3.4 1.3 1.3 OPRI Undulating 143 n/a 1,028 2014–2016
Juaben oil 2004 292 2.0 2.0 PSI Undulating 143 n/a 762 2014–2016
mill
Obooma 1999 66 1.4 1.4 OPRI Undulating 143 n/a 1,619 2014–2016

163
OPRI 2001 5.5 2.0 2.0 OPRI Sloping 151 n/a 1,737 2014
Oti 2010 3.8 1.3 1.3 OPRI Sloping 160 n/a n/a
SWAPP GSOPP 2006 275 3.6 3.6 BOPP Flat 160 n/a 1,643 2014–2015
West Bogoso
GSOPP 2008 120 4.0 3.8 BOPP Undulating 160 n/a 1,684 2014–2015
Chujah
GSOPP 2007 180 4.0 4.0 BOPP Undulating 160 n/a 1,579 2014–2015
Wassa 07
GSOPP 2008 70 4.0 4.0 BOPP Flat 160 n/a 1,579 2014–2015
Wassa 08
Justice 2008 13 3.0 1.1 BOPP Sloping 143 n/a n/a
Kado 1998 13 1.8 1.8 Norpalm Sloping 143 n/a n/a
Somprey 1997 15 3.9 1.6 OPRI Sloping 143 n/a n/a
Vikwam 2003 30 1.5 1.5 BOPP Sloping 143 n/a n/a
Yam farm 2007 15 3.0 1.4 OPRI Flat 151 n/a n/a
APPENDICES

* Consists of five ~4-hectare plots managed by individual farmers for each treatment. However, for the project the separate plots were
management as one single unit.
APPENDICES

Appendix 7. BMP field audit evaluation criteria grouped according to harvesting

practices, cultivation & upkeep, pests & diseases and nutrient management for mature
oil palm stands (adapted from Rankine and Fairhurst (1998)). All BMP fields at oil palm
plantations and smallholder farms were assessed to achieve full compliance with the
auditing criteria. Harvesting, loose fruit collection and fruit collection were awarded a
score of either 1 or 3, while all other parameters were scored 1, 2, or 3. Parameters
awarded with 1 required immediate action, a 2 required action, but not urgent, while a
3 meant full compliance with BMP standards. Once a block was awarded 3 for all
parameters, the block was considered BMP standard.
Category Parameter Auditing criteria
Harvesting • Full access (paths, circles, palm pruning) implemented to provide
harvesters with full access for harvest and in-field crop transport
using wheelbarrows
• Rounds maintained at 7–10 day harvest intervals
Harvesting practices

• Minimum ripeness standard of five loose fruits before bunch harvest

implemented
• No missed palms
• No evidence of crop loss
• Fronds removed at harvest according to pruning standards
Loose fruit • No evidence of loose fruit loss
collection • Loose fruit collected within 24 hours of harvest
Fruit • Same day transport of harvested crop to palm oil mill
collection • All bunches and loose fruit delivered to crop collection point
Pruning • Pruning rounds carried out according to budgeted programme (two
rounds per year)
• Remove senescent fronds on palms <3 years after planting, two
subtending fronds on palms 3–7 YAP (48–56 green fronds), one to
two subtending fronds on palms 8–15 YAP (40–48 green fronds),
and one subtending frond on palms >15 YAP (32–40 green fronds).
Ground • Eradication of woody weeds and other noxious weeds
cover • Establishment of soft weeds in between palms, the palm inter-row,
Cultivation & upkeep

vegetation and along the harvest path

Soil • Platforms installed at slopes 5–10° (9–18%)
conservation • Terraces or contour harvest paths installed at slopes 10–20° (18–
36%)
Frond • Pruned fronds stacked in a box pattern to improve soil properties,
stacking provide mulch, and to create a zone conducive to nutrient uptake
from fertilizers
• Fronds cut into two pieces. The petiole is stacked with the thorns
facing towards the ground in the palm inter-row. The remainder is
stacked at right angles to the harvest path in between the palms with
the frond base facing the inter-row.
Circle • Circle weeding implemented according to budgeted programme
weeding and
maintenance

164
 APPENDICES

Appendix 7. BMP field audit evaluation criteria grouped according to harvesting

practices, cultivation & upkeep, pests & diseases and nutrient management for mature
oil palm stands (adapted from Rankine and Fairhurst (1998)). All BMP fields at oil palm
plantations and smallholder farms were assessed to achieve full compliance with the
auditing criteria. Harvesting, loose fruit collection and fruit collection were awarded a
score of either 1 or 3, while all other parameters were scored 1, 2, or 3. Parameters
awarded with 1 required immediate action, a 2 required action, but not urgent, while a
3 meant full compliance with BMP standards. Once a block was awarded 3 for all
parameters, the block was considered BMP standard (continued).
Category Parameter Auditing criteria
Circle • Clean palm circles and free of obstructions (logs, debris, old loose
weeding and fruit)
maintenance • Supply palms properly weeded
• No volunteer oil palm seedlings
• Weed growth in line with interval between rounds of circle weeding
in work programme
Path • Path maintenance implemented according to budgeted programme
weeding and • Unimpaired wheel barrow access to every palm
maintenance • Installation of footbridges (over drains) to provide access
Cultivation & upkeep

Drainage • Adequate installation of V shaped drains

• No standing water where topography permits drainage
• Drains properly maintained (siltation removed, culverts allow
unimpeded flow of drainage water)
• Water gates fully functioning (gates, flaps) and water levels
controlled as required
• Drains blocked in the dry season with sand bags to conserve moisture
Erosion • Soil conservation measures implemented as required
Legume • Establishment of shade tolerant cover plants
cover plants
Roads • Roads maintained to allow full all weather access by vehicle to the
field
• Sufficient collection/in-field roads with a maximum carry distance of
200 m to the centre of the field
Pests and • All pest damage (rats, leaf eating insects (LEID) such as leaf miner
diseases
Pests &

diseases (Coelaenomenodera spp.) and rhinoceros beetle (Oryctes rhinoceros)

and diseases (Ganoderma, Fusarium)) reported, and control
measures implemented based on results of monitoring
Fertilizer • Fertilizer applied accurately according to given recommendations,
application following the “4R nutrient stewardship” guidelines. All fertilizers are
Nutrient management

spread over the boxed frond stack in the rain season, apart from urea,
which is spread evenly in the weeded circle, and B fertilizers, which
are applied in a band 1 m wide and 1 m from the palm trunk
Crop • Mulching with empty fruit bunches (40 t ha-1 (~300 kg palm-1))
residue applied as a mattress, one bunch deep between palms points within
application palm rows
• At short supply of EFB, prioritize application on the slopes

165
APPENDICES

Appendix 8. Example of fertilizer recommendations for year 3 for plantation and

smallholder BMP plots showing application rates for each nutrient (elements)
(kg palm-1). Application rates were based on the results of the leaf analysis and
(un)published information on fertilizer responses in West Africa and elsewhere. All
BMP plots received a maintenance dose. If leaf nutrient concentrations fell below the
pinnae and rachis critical concentrations (after von Uexküll and Fairhurst (1991) and
Foster and Prabowo (2006) respectively), plots additionally received a corrective dose
(i.e. maximum dose). Expert knowledge was used to adjust fertilizer rates for each plot,
depending on the response in leaf and rachis nutrient concentrations with previous
years.
Leaf critical
level Maintenance Corrective Maximum
Fertilizer
Nutrient dose dose dose
source Pinnae Rachis
% kg nutrient palm-1
Nitrogen (N) Urea 2.60 0.55 0.92 0.23* 1.15
Phosphorus (P) TSP 0.16 0.09 0.30 0.20 0.50
Potassium (K) KCL 0.95 1.4 1.25 0.25 1.50
Magnesium (Mg) Kieserite 0.25 0.07 0.08 0.08 0.16
Boron (B) Borate 15 - 0.01 0.01 0.02
* A corrective dose of N is applied where rachis K is sufficient and pinnae K is deficient.

166
 APPENDICES

Appendix 9. Average yields + SD for each project year for BMP and REF treatments
on plantation (BOPP, Norpalm, TOPP) and smallholder sites (SWAPP East, SWAPP
West). The grey triangles for plantation sites show average yields for the second control
plots which serve as an absolute reference for plantation performance, since REF
treatments were also undergoing improvement.

167
APPENDICES

Appendix 10. Model parameters for the interaction term of Treatment with Production
system. *Indicates a significant interaction at P≤0.05.
Parameter Units Yr Model parameters
Yield components
Bunch number bunches ha-1 1 F(1,63) = 0.320, MSE = 52570.574
2 F(1,63) = 0.330, MSE = 58454.058
3 F(1,63) = 0.062, MSE = 15909.774
Av. bunch weight kg 1 F(1,63) = 0.225, MSE = 2.946
2 F(1,63) = 0.290, MSE = 4.731
3 F(1,63) = 0.534, MSE = 8.995
Loose fruit collection t ha-1 loose fruit 1 F(1,63) = 2.166, MSE = 0.559
2 F(1,63) = 2.778, MSE = 0.615
3* F(1,63) = 5.885, MSE = 2.586
Fruit bunch harvest t ha-1 FB 1 F(1,63) = 0.067, MSE = 1.024
2 F(1,63) = 0.369, MSE = 4.349
3 F(1,63) = 0.975, MSE = 12.536
Total yield t ha-1 1 F(1,63) = 0.189, MSE = 3.164
2 F(1,63) = 0.620, MSE = 8.132
3 F(1,63) = 1.807, MSE = 26.019
Management
Harvest cycles cycles yr-1 1* F(1,63) = 9.052, MSE = 474.908
2* F(1,63) = 12.487, MSE = 907.295
3* F(1,63) = 13.903, MSE = 838.448
Harvester productivity t man-day-1 FB 1* F(1,63) = 4.726, MSE = 1.229
2 F(1,63) = 3.330, MSE = 1.023
3* F(1,63) = 8.923, MSE = 2.376
bunches man-day-1 1 F(1,63) = 3.283, MSE = 9028.540
2 F(1,63) = 1.456, MSE = 5733.746
3 F(1,63) = 3.860, MSE = 8445.528
ha man-day-1 1 F(1,63) = 3.146, MSE = 1.408
2 F(1,63) = 3.703, MSE = 1.338
3* F(1,63) = 5.954, MSE = 1.222
Harvesting labour man-days ha-1 cycle-1 1 F(1,63) = 1.329, MSE = 0.042
2 F(1,63) = 0.221, MSE = 0.009
3 F(1,63) = 0.073, MSE = 0.004
Field upkeep labour* man-days ha-1 yr-1 1 F(1,254) = 1.056, MSE = 3.956
2 F(1,254) = 0.327, MSE = 0.327
3 F(1,254) = 0.154, MSE = 0.345

168
 APPENDICES

Appendix 11. Field audit results summarized according to category (harvesting

practices, cultivation and upkeep, pests & diseases and nutrient management) for oil
palm plantation and smallholder BMP plots. Percentages were calculated as the total
score across all BMP plots at each site relative to the total possible score for each
parameter and then averaged per category. A 100 percent score means full compliance
with BMP for that particular category. The field audit evaluation criteria are given in
Appendix 7. Field audits were performed periodically (n=6) at plantation sites and only
twice at smallholders, at project start and end.

169
APPENDICES

Appendix 12. Key lessons in rehabilitating neglected mature oil palm plantings in West
Africa: a practitioner’s guide.

Thomas Fairhursta, Francis Ohipenib, Charles Frimpongb, Tiemen Rhebergenc,

Rosemary Addicob, Rob Mossd

Introduction
The oil palm sector in Ghana is diverse, consisting of a variety of stakeholders in oil
palm production and processing. The majority of growers are smallholder farmers that
cultivate oil palm on privately owned or rented land ranging in size up to 100 ha, with
the majority of the farms below 5 ha. Smallholders comprise the largest share of the
area under oil palm cultivation in Ghana (95%) and account for most of the national
fruit bunch (FB) production (60–80 %, Moss, personal communication). However, most
smallholder farms in Ghana are poorly managed with large opportunities for yield
improvement. Compared with oil palm plantations, average FB yields at smallholder
farms are much smaller (6 t/ha versus 11 t/ha in estates) and far below the yield potential
(~20 t ha-1) (Chapter 3, 4).
Farmers commonly lack knowledge on appropriate cultivation techniques, such as
harvesting and canopy, nutrient, groundcover and water management required to
achieve large FB yields. In this appendix, we summarize our experience in rehabilitating
neglected mature oil palm plantings with Best Management Practices (BMP), by
presenting a step-wise guide for smallholder farmers. Whilst the techniques described
in this booklet are based on practical experience gained from working with farmers
across the oil palm belt in Ghana, the methods presented have general applicability for
the West African region and elsewhere.
BMP’s are ‘cost-effective and practical agronomic techniques that aim to reduce yield
gaps in oil palm by using production inputs and resources efficiently’ (Donough et al.,
2009). BMPs are grouped in two broad categories; ‘yield taking’ and ‘yield making’
practices. Yield taking increases yield over the short term by improving crop recovery
operations (e.g. field access, harvest intervals), while yield making includes agronomic
practices that contribute to building large and sustainable yields over the longer term
(e.g. pruning, nutrient management).

a
Tropical Crop Consultants Ltd, 26 Oxenturn Road, Wye, Kent TN25 5BE, United Kingdom.
b
Solidaridad West Africa, 18 Okine street, East Legon, PMB KD11, Kanda, Accra, Ghana.
c
International Plant Nutrition Institute Sub-Saharan Africa Program (IPNI SSAP), ICIPE Complex, Duduville, Kasarani, Box 30772, Nairobi,
Kenya.
d
Gold Coast Fruit, South Ring Road House 6, ACP Estate, Pokuase, Greater Accra (P.O. Box 70, Adeiso, Nsawam, ER), Ghana.

170
 APPENDICES

We provide guidance on the implementation of BMPs with yield taking and yield
making to improve palm performance. Aspects related to harvesting practices,
cultivation and upkeep, and nutrient management are discussed in detail in the sections
below. We conclude with a section on the time lags in oil palm improvements and a
simple cost-benefit analysis using results from a rehabilitation project in Ghana. We
show the rapid increase in farm production and return on investment that can be
obtained by rehabilitating existing tenera palms, to demonstrate that rehabilitating
farms is usually a better medium-term financial investment than new plantings.

Photo 1. Neglected oil palm farm. Provided the palms are planted with DxP material, such fields can
be transformed into productive oil palm plantings over a period of three to four years.

171
APPENDICES

Optimal canopy management

Box-pattern frond stacking

Clean weeded circles

Accessible
harvest paths

Photo 2. Oil palm farm after rehabilitation showing improved leaf canopy, weeded circles and
harvesting paths.

Step 1: yield taking

1 Access to palms
The first step is to create field access within, and to the farm. Farms with poor access
usually produce low yields due to inability to evacuate the crop. Furthermore, bushy
weed growth competes with the palms for light, nutrients and water.
Unrestricted access is required, from the road to palm, so that the farmer can:
i. Identify ripe bunches at optimal ripeness easily,
ii. Harvest bunches from the palms efficiently,
iii. Collect all loose fruits,
iv. Transport bunches from the palm to the roadside using a wheelbarrow, and
v. Carry out field/maintenance operations, such as pruning and fertilizer
application.
In oil palm farms, access is provided by:
i. An access road from the nearest government road to the farm,
ii. Paths between alternate palm rows,
iii. Clean weeded circles around the base of each palm, and
iv. Adequate frond removal so that the harvester can access fruit bunches located in
the palm crown.

172
 APPENDICES

It may be necessary to upgrade or improve the road that provides access from the
government road to the farm. Better access roads will make it easier to negotiate for a
better price and lower transport costs for fruit bunches.
If the farm lacks good access in the field:
i. Use a hoe and cutlass to install paths (about 1 m wide) by removing all woody
growth and weeds to leave bare soil (3–5 md/ha),
ii. Use an arboricide (e.g. tryclopyr) to poison shrubs and small trees (0.5–1.0
md/ha),
iii. Use a cutlass and hoe to install weeded circles by removing all woody growth,
weeds, oil palm seedlings, rotten bunches to leave bare soil for about 1.8 m from
the base of each palm (4–6 md/ha), and
iv. Use a knapsack sprayer and herbicides to control weeds in palm circles every
three to four months (0.5–1.0 md/ha).
Unimpeded field access helps increase fruit bunch yields in the short-term by ‘yield
taking’. Yield taking is largely the result of an increase in crop recovery, which
increased average yields by almost 5 t/ha FB at smallholder farms in Ghana (Chapter
4).

Photo 3. Clean weeded circles provide the means for ripe bunch identification (detached fruits easily
visible) and rapid bunch and loose fruit collection.

173
APPENDICES

Photo 4. A proper access path has been installed in a mature oil palm plantation, but inter-row
vegetation requires improvement.

2 Harvesting
Proper harvesting standards and techniques are required to ensure complete crop
recovery and for maximum bunch oil content at harvest. The following points should
therefore be considered to achieve these goals:
i. Oil palm produces fruit bunches continuously throughout the year. Therefore, to
avoid any crop losses, harvesting intervals (i.e. the number of days between
successive harvests) must be maintained to ensure complete crop recovery.
Harvest intervals of 7–10 days are recommended as a sensible compromise
between excessively frequent harvesting (where bunches are scarce, the
harvester’s output is poor and the temptation to harvest unripe bunches is high)
and infrequent harvesting (large amounts of loose fruits require collection and
fruit bunches may rot before they are harvested).
ii. Whilst industrial mills prefer ripe bunches to maximize oil yield, artisanal mills
prefer over-ripe bunches because it easier to separate the fruits from the bunch.
As such, a minimum ripeness standard (MRS) of five loose fruits (LF) per bunch
on the ground before harvest is recommended for industrial mills and >10 LF
before harvest for artisanal mills.
iii. Once proper access has been installed (weeded circles and paths; Section 2) and
pruning has been carried out to remove surplus fronds (Section 3), harvesting

174
 APPENDICES

becomes much easier and less costly. With good field conditions, a harvester can
cut and carry 1.0–1.2 t/md compared with 0.6–0.8 t/md in a poorly maintained
farm (Table 2, Chapter 4).
iv. Loose fruits contain 40% oil, whilst bunches contain about 20% oil. It is
therefore important to collect all loose fruit at harvest. This can only be achieved
if there are clean-weeded circles around the base of each palm (Section 2).
v. At present, fruit bunches are mostly transported to the roadside using head pans
(25–30 kg/load). However, wheelbarrows (70–80 kg/load) are a less costly
means to transport fruit bunches from the field. Wheelbarrows can only be used
where proper paths have been installed (Section 2).
vi. Palms <7 years after planting (YAP) are harvested with a chisel (blade width
10–12 cm) and palms ≥ 7 YAP with a sickle when the palms have grown too tall
for chisel harvesting (>2 m from ground level to ripe bunch position).
vii. In palms ≤4 YAP, fruit bunches can be removed without cutting any fronds, to
maximise the number of green leaves retained on each palm. From four years
after planting onwards, fronds under the ripe bunch should be removed and
stacked in the inter-row (Section 4).
viii. Use a rake or scraper to collect all loose fruits from weeded circle and place them
in an empty fertiliser bag for transport to the roadside.

Photo 5. Well prepared fruit bunches ready for collection at a smallholder farm.

175
APPENDICES

Step 2: yield making

1 Pruning and leaf canopy management
Farmers sometimes misunderstand the issue of pruning – they sometimes think that
pruning ‘damages’ the palm and reduces yield. Pruning is required to maintain a full,
healthy palm canopy and an optimal leaf areae (L=5.5–6.0) to maximize the conversion
of sunlight, nutrients and water into vegetative dry matter and bunch production. To
achieve proper pruning standards, it is important to carry out the following:
i. Remove surplus fronds (Table 1),
ii. Remove dead, damaged, diseased or old fronds,
iii. Provide field access for harvesting,
iv. Prevent loose fruit becoming trapped behind frond butts, and
v. Provide the palm with the correct number of fronds to maximise yield (Table 1).
In West Africa, pruning is best carried out twice each year:
i. One cycle in January or February (i.e. before the peak crop season), and
ii. Once cycle in October or November (i.e. at the end of the main crop season).
Table A12.1. Pruning standards for oil palm according to palm age.
Number of
Growth Years after
Fronds Spirals subtending Tool
stage* planting
fronds
Remove senescent
IGP <4 - - Chisel
fronds
Chisel,
SAYP 4–7 48–56 6–7 2
sickle
PYP 8–14 40–48 5–6 1–2 Sickle
DYP >15 32–40 4–5 1 Sickle
*IGP = Immature Growth Stage, SAYP = Steep Ascending Yield Phase, PYP = Plateau Yield Phase,
DYP = Declining Yield Phase.

Consider the following points when pruning:

i. Pruning is easier when carried out with very sharp tools.
ii. Old, dead fronds and debris should be raked out from around the base of the
palm after pruning to facilitate harvesting and field maintenance operations.
iii. Rehabilitation (or corrective) pruning is very laborious (3–10 md/ha) compared
with maintenance pruning (1–2 md/ha).
iv. After pruning, cut fronds in half at the point where thorns change to leaflets, and
stack them in a three-sided “box” at the outer edge of the weeded circle (see next
section).

e
The leaf area index (L) is a dimensionless parameter, and is defined as the leaf area per unit area of land (L = leaf area (m2) / ground area
(m2))

176
 APPENDICES

Photo 6. Un-pruned palms are impossible to harvest efficiently.

Photo 7. Future yields are reduced by excessive frond removal.

177
APPENDICES

Photo 8. Correct pruning standards lead to larger yields in the future.

2 Frond stacking
Pruned fronds contain large amounts of nutrients, which benefit the palm when returned
to the soil. Pruned fronds also provide a barrier to reduce surface wash and soil erosion.
The following method for frond stacking is therefore recommended:
i. Remove the thorny petiole from the frond rachis.
ii. Stack the pruned fronds in a three-sided box-pattern around the palm, up to the
edge of the weeded circle:
a. The rachis is placed between palms, with the sharp end facing the harvest
path. The goal is to cover as much of the soil surface as possible whilst
keeping the circle and path clear.
b. The petiole is placed in the inter-row, with the thorns facing down to
prevent injuries to workers.
c. Do not place any fronds on the side of the harvesting paths, so that
workers can enter the weeded circle without any obstructions for
maintenance and harvesting operations.
iii. Do not stack fronds on top of each other in a narrow band in between palm rows!
Box-pattern frond stacking provides the following benefits:
i. Physical protection of the soil surface to minimize erosion, and reduce the
amount of applied fertilizer lost due to surface wash,
ii. Improved soil structure, water infiltration rate in soil beneath the frond stack,
iii. Oil palm roots proliferate in the soil beneath the frond stack, and

178
 APPENDICES

iv. Easy checking of plant nutrient deficiency symptoms, leaf eating insects, or other
pests and diseases on removed fronds.
By stacking fronds in a box pattern, soil properties are improved in the soil beneath the
inter-row area that is not occupied by weeded circles and paths. It therefore makes sense
to apply P, K, and Mg fertilizers over the box frond stack where the fertilizer is less
susceptible to surface wash, and there is a large amount of active fine feeder roots.

Photo 9. Box-pattern frond stacking in a mature plantation reduces surface run-off and provides an
erosion barrier.

3 Ground cover management

Many smallholder farms in West Africa are overgrown by dense woody plants (e.g.
Baphia nitida as well as volunteer palm seedlings), which should be removed to reduce
competition for nutrients, moisture and light and to provide access for harvesting and
field maintenance operations. Unsuitable groundcover vegetation should be replaced by
groundcover such as soft grasses and herbaceous weeds. Competitive woody plants in
palm inter-rows should be eliminated with an integrated approach that includes manual
control (slashing, uprooting), chemical control, and improvements to soil fertility.

179
APPENDICES

The following methods are recommended to control woody growth and to improve the
quality of the inter-row vegetation:
i. Slash woody growth to knee height, wait for regrowth (2–4 weeks). Spray woody
plants with a mixture of tryclopyr and diesel (1:5). Repeat spraying (i.e. two to
three cycles) may be required to eliminate stubborn woody plants.
ii. At farms with little woody growth, a targeted approach is more appropriate:
a. Slash the woody plant at knee height and immediately apply a mixture of
tryclopyr and diesel (1:5) to freshly cut surface of the stump with a brush.
b. Alternatively, dig out the woody plants using an old harvesting chisel.
iii. Broadcast 3–7 kg/palm reactive rock phosphate to encourage the establishment
of soft weeds, grasses and ferns (e.g. Nephrolepis biserrata) over the inter-row
space. Apply large applications of RP in several split applications.
iv. Establishment of legume cover plants in the inter-row (3 plants/palm). Pueraria
and Mucuna spp. are more common and easier to establish in West Africa than
the plant shade tolerant Calopogonium spp.

Photo 10. Woody growth (Baphia nitida) competes with the palms for nutrients and water.

180
 APPENDICES

Photo 11. An inter-row vegetation consisting of ferns and soft weeds is less likely to compete with
palms for nutrients and water, and is less costly to manage than woody growth.

4 Drainage and water management

Most oil palm farms in West Africa include low-lying areas that are poorly drained.
Low lying areas have the potential to produce large yields due to their more fertile soils
and higher water holding capacity, but require costly drainage to prevent flooding and
water-logging during periods of high rainfall.
In poorly drained areas, palms are unable to extract water and nutrients because the
roots cannot breathe, resulting in poor palm growth and reduced yields. To increase
production in poorly drained low-lying areas, a water management system is required
to remove excess water in the rain season and to conserve water in the dry season.
Before any drainage work is undertaken, it is advised to draft a drainage plan first and
mark the direction of drains in the field with coloured pegs. Drains can be installed
effectively using excavators fitted with a trapezoidal or ‘V’ bucket. V-shaped drains
with sloping sides are less prone to collapse and the narrow bottom is self-cleaning and
less prone to silting-up. The following steps are recommended:
i. Check whether the farm drain outlets provide the means to carry drainage water
away from the farm. It may be necessary to dig new outlet drains (≥1.5 m deep)
or deepen existing drains to provide effective drainage outlets. If a number of

181
APPENDICES

farms require main drains it may be less costly to employ a contractor with an
excavator and ‘V’ bucket to install the drains.
ii. The main drain should run through the lowest points in the field.
iii. Always start drain installation and rehabilitation at the outlet-end (i.e. lowest
point) of the drain.
iv. Once the main drain has been installed, install field drains (≥1 m deep) every
four palm rows and connect them to the main drain.
v. Additional sub-field drains (about 0.5 m deep) may be required in areas that are
difficult to drain.
vi. Install culverts as required (e.g. where main drains cross a road).
vii. In hilly areas, it may be necessary to install a foot-hill drain (about 0.5 m deep)
at the base of the hill and along the contour to divert run-off water into the main
drains. This lessens the requirement for drainage within valley bottoms.
viii. Accessibility in drained areas must be provided by constructing footbridges over
the drains.
ix. To conserve water in the dry season, field drains can be blocked at the end of the
wet season by using used fertilizer bags filled with sand/soil.

Photo 12. Poor drainage in low-lying areas imposes stresses on the palms and impedes access for
complete crop recovery in the rain season.

182
 APPENDICES

Photo 13. A well-drained oil palm farm will give larger yields and better access for maintenance and
harvesting operations.

Photo 14. Example of a drainage plan for an immature oil palm field, showing the main drain, field
drains and hill-foot drains.

183
APPENDICES

5 Nutrient management
In the oil palm belt in West Africa, most soils contain very small amounts of nutrients,
especially potassium (K) and phosphorus (P) (Chapter 3). The efficient use of fertilizers
and crop residues is therefore necessary to optimize FB yields and profits in oil palm.
In this section, we provide a guideline for nutrient management, following the ‘4R
nutrient stewardship approach” (IPNI, 2012), where plant nutrients are applied with the
right source (1), at the right rate (2), at the right time (3), and in the right place (4).
Right source:
i. Large amounts of nutrients are removed from the farm in harvested fruit
bunches. On low fertility status soils, it is essential to compensate for nutrient
removal by adding nutrients with mineral fertilizers.
ii. A suitable product is the compound fertilizer NPK 10–10–30+1.5MgO+0.2B,
because it contains the correct combination and balance of nutrients and trace
elements for oil palm. If compound fertilizers are not available, straight
fertilizers (which usually contain a single nutrient) can be used instead (Table
2).
Right rate, or amount:
i. To achieve healthy palm growth and targeted yields, sufficient quantities of each
essential nutrient (nitrogen (N), phosphorus (P), potassium (K), magnesium
(Mg), and boron (B)), must be provided to meet the palm demand for vegetative
growth and yield.
ii. Fertilizer rates take into account site yield potential (based on climate, palm age,
and planting material), soil fertility and the results of annual leaf nutrient
assessments (nutrient analysis and nutrient deficiency scores). For example,
larger application rates are necessary to maximise yield where rainfall is
plentiful, while smaller application rates are recommended where rainfall is low.
iii. Based on (un)published information from fertilizer trials in West Africaf,g, the
optimal amount of fertilizer for mature oil palms is 4–6 kg/palm NPK 10–10–
30+1.5MgO+0.2B, which provides 0.4–0.6 kg/palm N, 0.4–0.6 kg/palm P2O5,
1.2–1.8 kg/palm K2O, 0.06–0.09 kg/palm MgO and 0.08–0.12 kg/palm B.
iv. If resources are limited, start with 1 kg/palm NPK, and increase the application
rate each year.
v. If NPK 10–10–30+1.5MgO+0.2B is not available, a combination of straight
fertilisers can be used instead to provide the major nutrients (Table 2). For

f
Danso, I., Nuertey, B., Asamoah, T., Tetteh, F., Danso, F., Afari, P., Safo, E. and Logah, V., 2010. The effect of rock phosphate on soil
nutrient dynamics, growth, development and yield of oil palm in the semi-deciduous forest zone of Ghana. J. Sci. Technol., 30(1): 30-44.
g
van der Vossen, H.A.M., 1970. Nutrient status and fertilizer responses of oil palms on different soils in the forest zone of Ghana. Ghana J.
Agric. Sci., 3: 109–129.

184
 APPENDICES

example, 1.3 kg urea, 1.3 kg TSP and 3.0 kg KCl is equivalent to 6 kg NPK 10–
10–30.
vi. A calibrated cup or dish should be used to measure the correct amount of
fertilizer for each palm.
Table A12.2. Amount of straight fertilizer (kg palm-1) equivalent to NPK 10–10–30 compound
fertilizer.

Equivalent straight fertilizer

NPK 10–10–30
Nitrogen (N) Phosphorus (P) Potassium (K)
compound
fertilizer Ammonium Rock
Urea TSP KCl
sulphate phosphate
kg palm-1
1 0.2 0.5 0.3 0.2 0.5
2 0.4 1.0 0.7 0.4 1.0
3 0.7 1.4 1.0 0.7 1.5
4 0.9 1.9 1.3 0.9 2.0
5 1.1 2.4 1.7 1.1 2.5
6 1.3 2.9 2.0 1.3 3.0
Right time:
i. Nutrients should be applied on time, in order to provide the palms with a fairly
continuous supply of nutrients.
ii. Losses of N from ammonia-based N fertilizers (e.g. urea) can be reduced by
applying fertilizer when there is a strong likelihood of moderate rainfall (i.e.
10–20 mm/day) within the 3–4 days following application. Urea application
should therefore be timed to take place during the wet season but not in months
where very high rainfall over short periods is likely.
iii. Avoid fertilizer application during periods of very heavy rainfall (i.e.
>20 mm/day) where fertilizer losses due to leaching and surface water run-off
will be significant.
iv. Fertilizers should be applied in several split applications. For example, 6 kg/palm
NPK 10–10–30 compound fertilizer should be applied in four split applications
each of 1.5 kg/palm. In most parts of Ghana’s oil palm belt the optimal months
for fertilizer application are March–June and August–September (Table 3).
Right place:
i. It is important to apply fertilizers where nutrient uptake by the palms is most
efficient and where the risk of losses due to surface water run-off and erosion are
minimized.

185
APPENDICES

ii. Compound and P, K, and Mg fertilizers are less susceptible to surface water
runoff after application when there is a protective layer of mulch over the soil
surface.
Table A12.3. Example of a fertilizer application schedule for 4 and 6 kg/palm (in cursive, bold) NPK
10–10–30 and its equivalent in straight fertilizers. Nitrogen can be applied with straights as either urea
or SOA and phosphorus with RP or TSP.
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
Fertilizer
kg/palm
NPK 1.5 1.5 1.0 4.0
10–10–30 1.5 1.5 1.5 1.5 6.0
Equivalent in straights:
N 0.9 0.9
Urea
1.3 1.3
1.0 0.9 1.9
SOA
1.5 1.4 2.9
P 1.3 1.3
RP
1.0 1.0 2.0
0.9 0.9
TSP
1.3 1.3
K 1.0 1.0 2.0
KCL
1.5 1.5 3.0
iii. Correct placement means applying fertilizer where there is a large amount of fine
feeder roots. In mature oil palm plantations, where pruned fronds have been
stacked in a box pattern, the greatest amount of fine feeder roots is found in soil
beneath the frond stack. Fertilizer placement is recommended as follows:
a. In the first year after rehabilitation (i.e. whilst hard weeds and woody
growth are being removed from the palm inter-rows), fertilizers should be
spread evenly over the weeded circles.
b. In the second and subsequent years after rehabilitation, and once woody
growth and hard weeds in the inter-row space have been controlled,
compound and P, K, and Mg fertilizers should be broadcast over the frond
stack.
iv. Urea fertilizers should be spread evenly over the clean weeded circle.
v. To achieve an even distribution of fertilizer in the field, flick the fertilizer from
a plate using the fingers of one hand whilst holding the plate in the other hand.

186
 APPENDICES

Mill residues, such as empty fruit bunches (EFB), decanter cake and palm oil mill
effluent (POME), are an important source of nutrients for oil palm. When they are
recycled and applied to the field, fertilizer application can be reduced as follows:
i. An application of 40 t/ha (or 300 kg/palm) EFB, spread evenly between palms
within rows and in the palm inter-rows, provides the palms with sufficient P, K
and Mg for about one year. A small amount of N fertilizer (e.g. 0.5 kg/palm as
either 1 kg/palm urea or 2.5 kg/palm ammonium sulphate) should be applied to
provide an adequate supply of nitrogen.

Photo 15. Nutrient deficiencies in oil palm (N, P, K, Mg, B) can be overcome with balanced nutrient
management, such as the 4R nutrient stewardship approach.

Photo 16. Calibrating fertilizer rates using a plastic dish allows for rapid and efficient application in
the field.

187
APPENDICES

Photo 17. In mature oil palms, fertilizers should be applied to ‘target areas’, where nutrient uptake is
most effective and losses are minimized.

Photo 18. Using the flick method for fertilizer application provides even distribution over the frond
stack and better uptake of nutrients.

188
 APPENDICES

Photo 19. Recycling crop residues, such as empty fruit bunches (EFB), reduces annual fertilizer
requirements.
Time lags between stress events and their effect on yield in oil palm
The oil palm is adapted to respond to different kinds of stress. A significant stress event
(e.g. three months of drought, full leaf defoliation by leaf miner insects, lack of fertilizer
application) triggers several responses by the palms:
i. More male flowers and less female flowers are produced at sex determination,
resulting in a ‘male flower phase’ and less fruit bunches 30–33 months later.
ii. Flowers may abort, resulting in less bunches 7–13 months later.
iii. Pollinated bunches may abort, resulting in a large number of poorly formed
bunches (bunch failure) 2–5 months later.

189
APPENDICES

Photo 20. Abiotic and biotic stresses can reduce the number of bunches at sex differentiation of the
inflorescences at approximately 30–33 months before harvest.

Photo 21. Bunch failure occurs at about 2–5 months to harvest and depends on the magnitude of stresses
(e.g. drought, pest incidence) imposed on the palms.

190
 APPENDICES

In this way, the palm avoids excessive bunch production following periods of stress.
Clearly, the time lag between a single stress event and a change in production may range
from 2–40 months. In reality, stress events may occur more frequently. For example, a
drought period may be followed several months later by defoliation caused by an
outbreak of leaf miner insects. Whilst the drought stress (an abiotic stress factor) cannot
be avoided (but there are options to improve water management), the leaf miner
outbreak (a biotic stress factor) can be avoided by early pest identification (i.e. with an
effective early warning system) and prompt implementation of control measures when
pest pressure exceeds the respective threshold but before extensive leaf canopy damage
occurs.
During crop scene investigations, it is important to try to relate the present condition of
the palms (e.g. bunch production, black bunch count, sex ratio (i.e. the ratio of female
to male plus female flowers)) to past stress events (e.g. drought, delayed fertilizer
application, pest outbreaks). To do this, proper record keeping is essential.
Rehabilitating neglected oil palm farms is a step-wise process and should be
implemented in the correct sequence in order to minimize agronomic stress effectively.
Yield taking:
i. Most important is to first provide unimpeded access for harvesting operations
and palm upkeep with the installation of harvest paths and weeded circles, and
removal of unproductive fronds with corrective pruning.
ii. The second step involves the introduction of regular and complete harvesting
events at 7–10 day intervals to ensure complete crop recovery.
Yield making:
iii. After yield taking constraints have been completely removed, yield making
operations should focus on the following practices:
a. Improving nutrient management with balanced mineral nutrition.
b. Soil conservation and mulching with mill residues.
c. Maintenance of optimal leaf canopy by implementing two cycles of
pruning per year.
d. Maintenance of optimal ground cover by the manual and chemical
removal of woody weeds in palm inter-rows and harvest paths to favour
the establishment of soft weeds, grasses and legume cover plants (LCP).
An application of P fertilizer may be required to improve the
competitiveness of soft weeds, grasses and LCP.
e. Improving drainage in swampy areas by installing ‘V’ shaped drains.

191
APPENDICES

The long time-lag between flowering and bunch harvest means that complete
rehabilitation of neglected palms takes about four years (Fig. 1.3). It is possible to
measure the separate effects of yield taking and yield making:
iv. ‘Yield taking’ takes place during the short period of 3–6 months required to
achieve crop recovery. There may be an increase in bunch number due to
complete crop recovery and bunch weight due to complete loose fruit collection.
v. ‘Yield making’ practices start to impact yield after about six months, by which
time there may be a significant increase in bunch weight. An increase in bunch
number due to increased floral initiation and changes in sex ratio occur
respectively 12–30 months after agronomic constraints have been removed.
Cost-benefit analysis
Rehabilitating neglected or abandoned oil palm requires significant initial investments
to achieve complete crop recovery and to eliminate all agronomic constraints.
Thereafter, field upkeep costs usually decrease, often to amounts less than pre-BMP
costs.
The majority of the costs occur in the first year, when additional labour is required to
rehabilitate the farm with corrective pruning, installation of paths and palm circles and
the removal of woody growth (Table 4). Farmer income therefore decreases in the first
year of rehabilitation when yields have not yet started to increase significantly. Once
proper access to the palms has been established, harvesting is easier and less costly,
resulting in improved crop recovery and profitability.
In the short-term, improvements to economic returns in smallholder farms can be
achieved by simply improving crop recovery, which helps to finance inputs needed for
the yield making phase, such as fertilizers. By the third year after rehabilitation, the
benefits of investing in yield making start to become apparent, with larger returns on
investments than before starting BMP. By Year 4, profits are more than double the pre-
BMP phase (Table 4). The best farms will likely produce yields of 20–25 t/ha FB, so
profits will be even larger.
Most smallholder farmers have insufficient funds to complete rehabilitation over a large
area, particularly given the four-year full payback period. It is therefore recommended
to carry out rehabilitation in parcels, according the farmer’s access to labour and funds.

192
 APPENDICES

Table 4. Example cash flow for the rehabilitation of a neglected mature oil palm farm with BMP, using
trial data from Ghana.
Pre-BMP BMP
Parameter Units Year
0 1 2 3 4 5
Fruit bunch yield t ha-1 7 10 13 18 20 20
Revenue US$/ha 529 755 982 1,360 1,511 1,511
Material costs
Fertilizer US$/ha 0 148 197 246 246 246
Herbicide US$/ha 0 36 36 36 36 36
Garlon US$/ha 0 58 0 0 0 0
Diesel US$/ha 0 20 0 0 0 0
Total material costs US$/ha 0 262 233 283 283 283
Machine costs
Excavator US$/ha 0 76 0 76 0 0
Grader US$/ha 0 76 0 0 76 0
Total machine costs US$/ha 0 151 0 76 76 0
Labour costs
General labour US$/ha 4 225 99 102 102 102
Harvesting labour US$/ha 127 63 82 109 118 118
Total labour costs US$/ha 131 289 181 211 220 220
All costs US$/ha 131 702 414 569 578 503
Margin over costs US$/ha 398 53 568 790 932 1,008

193
APPENDICES

Conclusions
Rehabilitating farms with BMPs shows considerable scope for increasing yields and
profitability in smallholder oil palm farms (Table 4). In Ghana, average yields with
BMP approached 18 t/ha FB after 3 years; an increase of 12 t/ha (+200%) over initial
yields. These results show that rehabilitation of existing tenera plantings is a better
option than investing in new plantings. Yield intensification can be achieved without
the capital expenditure required for new plantings (e.g. road infrastructure, planting
cost), and financial returns from investments in yield intensification accrue more rapidly
because production starts to increase as soon as field management and agronomic
constraints have been removed.
Rehabilitation of neglected mature oil palm farms can be highly profitable provided the
following conditions are met:
i. Farmers have access to competitively priced fertilizers,
ii. Farmers have access to competitively priced markets for fruit bunches. This
means access to a modern palm oil mill achieving commercial extraction rates
(≥23% oil extraction rate) where part of the benefits of high extraction rates are
passed back to the smallholder in terms of FB price.
iii. Inputs are used effectively:
a. No wasted herbicide,
b. Fertilizer applied correctly,
c. Mill residues are utilized in the field, and
d. Labour is used effectively.
iv. Field conditions up to BMP standard (field access, ground cover management,
pruning, pruned frond management).
v. Full crop recovery with 7–10 day harvest intervals, and complete bunch and
loose fruit collection.
Mature oil palm farms should be rehabilitated in steps and according to the farmer’s
capacity to invest in labour, inputs and supervision.
The farmer must adjust to more intensive enterprise management, including the
following:
i. Need for financial planning,
ii. Likely requirement to borrow funds, and
iii. Need to supervise larger investments in field labour.
It will be necessary for farmers to forge a close relationship with a commercial bank by
implementing financial transactions via a bank. Once a farmer has demonstrated
significant cash flow, the bank may be willing to provide working capital.

194
 APPENDICES

It is important for governments and investors (NGOs, banks) to consider the benefits of
investing in the rehabilitation of existing oil palm plantings to improve farm
profitability and increase palm oil production whilst reducing the requirement for area
expansion to satisfy market demand for palm oil.

195
 SUMMARY
The large demand for vegetable oil has resulted in a rapid expansion of global oil palm
cultivation. Most of the current expansion is taking place in sub-Saharan Africa and
Latin America as land available for new oil palm plantings is now limited in Southeast
Asia. In Ghana, for example, oil palm (Elaeis guineensis Jacq.) experienced a
tremendous resurgence since the early 2000s, partly due to a growing interest by
investors to expand and develop the area planted to oil palm. As a result, Ghana became
the third largest producer of palm oil in sub-Saharan Africa in 2017 at 2.5 Mt yr-1 (next
to Nigeria (7.8 Mt yr-1) and Cameroon (3.1 Mt yr-1)). Oil palm is currently the second
most important perennial crop in the Ghanaian economy after cocoa. Despite large
increases in total production the past 15 years, fruit bunch (FB) yields remained low.
With an estimated yield potential (Yw) of 21 t ha-1 FB, and actual yields of
11.0 t ha-1 FB on plantations and 6.0 t ha-1 FB on smallholder farms in Ghana, large
yield gaps exist.
This thesis analysed the factors that currently limit oil palm production in Ghana and
assessed opportunities to increase yields with improved agronomic practices. Firstly, a
review and characterization of the environment for growing oil palm in Ghana was
conducted, and used as a basis to determine the suitability and availability of land for
production (Chapter 2). Detailed characterization of the oil palm sector in Ghana was
then conducted, and yield gaps and their underlying causes analysed using the concepts
of production ecology (Chapter 3). Third, the potential of ‘Best Management Practices’
(BMP) to close yield gaps in existing mature oil palm plantings was evaluated (Chapter
4). Finally, an experiment to investigate the interaction of nutrients and water on the
yield performance of oil palm was conducted on a large-scale oil palm plantation in
Western region, Ghana (Chapter 5).
In Chapter 2, land suitability evaluation (LSE) methods were used to identify areas that
are both suitable and available for oil palm production in Ghana. A major climatic factor
limiting oil palm production in Ghana is the annual water deficit, with the most suitable
areas located in the rainforest and semi-deciduous forest zones with higher rainfall in
southern Ghana. Opportunities for the large-scale oil palm sector to develop was
constrained due to the lack of availability of large and contiguous tracts of land that are
required for commercial plantation oil palm development. The results suggest that a
feasible strategy for increasing oil palm production is through smallholders, which can
make use of smaller parcels of land. Alternatively, oil palm production in Ghana can be
increased by yield intensification on land already planted to oil palm.
To guide yield intensification practices, one must first understand the magnitude and
the underlying causes of yield gaps. In Chapter 3, Yield Gap Analysis (YGA) was used

197
SUMMARY

to elucidate the factors that were responsible for yield gaps at plantations and
smallholder farms. Farm surveys were conducted to characterize plantations and
smallholder farms in terms of production inputs (labour, fertilizer, empty fruit bunch
mulch, and agro-chemicals including herbicides and insecticides) and outputs (bunch
production, number of bunches, and FB yield and yield components). To provide a
benchmark for the yield potential (Yw) in Ghana, a boundary line was fitted through
yield data obtained from well-managed mature oil palm fields. Yield gaps were
calculated as the difference between Yw and yields obtained from smallholder and
plantation, and averaged 15.4 t ha-1 FB at smallholder farms, and 9.8 t ha-1 FB at
plantations. The analysis revealed that poor management practices, including
incomplete crop recovery (i.e. not harvesting all available crop) and inadequate
agronomic management were the main factors contributing to these yield gaps.
Particularly smallholders face major challenges in increasing yields, due to a lack of
knowledge on appropriate management practices and poor operating conditions (e.g.
poor infrastructure and lack access to credit). Productivity losses were exacerbated by
low oil extraction rates (OER) by small-scale processors of 12% as compared to 21%
by the large-scale processors. The potential losses in annual CPO during the crop
plateau yield phase were estimated to be 5 and 3 t ha-1 yr-1 for small-scale and large-
scale production systems respectively. The results of this study suggest that investments
in yield gap reduction, by improving agronomy and crop recovery at plantations and
smallholders, whilst improving access of smallholder producers to more efficient oil
palm processing facilities, can make a significant contribution to closing the supply gap
for palm oil in Ghana.
The potential of closing yield gaps on plantations and smallholder farms through
implementation of ‘Best Management Practices’ (BMPs) was assessed in Chapter 4.
BMPs were evaluated against current management practices, amongst those that
increased yield in the short-term term (≤ 1 year) with ‘yield taking’ (improved crop
recovery), and in the long-term (>1 year) with ‘yield making’ (better agronomy)
practices. Experimental results indicate that yield increases with BMP are large.
Average yields with BMP increased with 2.1 t ha-1 (+19%) and 4.7 t ha-1 (+89%) with
yield taking and with 4.7 t ha-1 (+36%) and 7.6 t ha-1 (+76%) with yield making at
plantations and smallholders respectively. Improvements in crop recovery, including
more frequent harvesting events (i.e. short harvest intervals) and improved field access
(roads, paths, weeded circles) for more efficient productivity (i.e. more bunch weight
harvested and less area (ha) covered per harvester) contributed considerably to yield
increases with BMP. Longer-term yield increases were more complicated to explain,
due to a time-lagged effect of improved agronomy on yield, but analysis of fertilizer
usage and leaf nutrient concentrations suggest a more balanced approach to nutrient

198
 SUMMARY

management could contribute considerably to yield making, particularly at smallholder

farms. There is therefore need to develop improved fertilizer recommendations to
further increase yields and to sustainably intensify oil palm production in Ghana.
Chapter 5 focuses on the assessment of the interaction of nutrients and water on the
yield performance of oil palm on a large-scale oil palm plantation in Western region,
Ghana. Average yields of about 32.6 t ha-1 were achieved with irrigation and fertilizer.
Yield gaps caused by nutrients were estimated at 4.1 t ha-1 and 4.7 t ha-1 for water, whilst
the response to irrigation alone without fertilizer was 0.6 t ha-1. These results suggest
that fertilizer is essential for increasing yields with irrigation, but information on the
specific nutrients required and their optimal application rates is lacking, and requires
further investigation. Despite the use of irrigation, annual yield cycles in oil palm
persisted. This can lead to irregular distribution of production and lack of mill capacity
and insufficient harvesters in the peak crop months. Many factors have been proposed
that cause these fluctuations, but the processes behind these yield cycles remain unclear.
The research conducted in this thesis shows that there are large yield gaps in plantation
and smallholder production systems in Ghana, which are caused by sub-optimal climate
conditions (e.g. water) and poor management practices (e.g. crop recovery, nutrient
management), amongst others. The potential to increase productivity with BMP,
however, is considerable. The significant yield improvements achievable with BMPs at
sites representative of the Ghanaian oil palm industry offer scope to reduce the
requirement for future area expansion to meet the increasing demand for oil palm
cultivation. Increasing average national yields to 21.0 t ha-1 FB, for example, could
spare almost 600,000 ha of land. In addition, closing yield gaps in Ghana under current
land area (327,000 ha) has the potential to increase fruit bunch production almost three-
fold from 2,5 Mt to 6,9 Mt. If all crop is processed at an OER of 21%, approximately
1.4 Mt CPO can be produced. This is more than enough to meet Ghana’s current annual
demand of 106,000 t CPO, and does not require the need to plant additional land. Yield
intensification strategies may therefore be an important policy for sustainable oil palm
development and land use in Ghana and West Africa.

199
 SAMENVATTING
Door de grote vraag naar plantaardige olie is het oliepalm (Elaeis guineensis Jacq.)
areaal sterk toegenomen in de wereld. Het merendeel van de recente toename vindt
plaats ten zuiden van de Sahara in Afrika en Latijns-Amerika, omdat Zuidoost-Azië nog
maar slechts beperkte mogelijkheden heeft om nieuwe oliepalm aan te planten. Ghana
bijvoorbeeld, kent sinds het begin van de 21e eeuw een enorme opleving van oliepalm,
onder meer door een groeiende interesse van investeerders. Het resultaat is dat Ghana
de op twee na grootste producent van palmolie in Afrika ten zuiden van de Sahara is
met jaarlijks 2.5 megaton productie (na Nigeria (7.8 megaton per jaar) en Kameroen
(3.1 megaton per jaar)). Oliepalm is tegenwoordig het op één na belangrijkste gewas in
de Ghanese economie, na cacao. Ondanks de toename in productie van de afgelopen 15
jaar, bleven de opbrengsten van de oliepalmvruchten laag. Met een geschatte potentiële
opbrengst (Yw) van 21 ton per ha aan oliepalmvruchten en een werkelijke opbrengst op
de plantages van 11 ton per ha en 6 ton per ha bij kleine boeren, bestaan er in Ghana
grote verschillen tussen de potentiële en werkelijke opbrengsten (de zogenaamde
opbrengstverschillen of ‘yield gaps’).
In deze thesis werden de factoren die de oliepalmproductie in Ghana limiteren,
geanalyseerd. Ook werd gekeken naar mogelijkheden om de opbrengsten te verhogen
d.m.v. verbeterde landbouwkundige praktijken. Eerst werd de groeiomgeving van
oliepalm in Ghana beschreven, om vervolgens de geschiktheid en beschikbaarheid van
land voor productie te bepalen (hoofdstuk 2). Daarna werd de oliepalm sector in Ghana
bestudeerd en werden de oorzaken voor de yield gaps geanalyseerd, daarbij
gebruikmakende van de concepten van ecologische productiefactoren (hoofdstuk 3).
Ten derde volgde er een analyse van de mogelijkheden om yield gaps te dichten via
‘Best Management Practices’ (BMP) (hoofdstuk 4). Tenslotte werd een experiment
uitgevoerd om de interactie tussen nutriënten en water op de oliepalmopbrengst te
onderzoeken op een grootschalige oliepalmplantage in Western Region, Ghana
(hoofdstuk 5).
In hoofdstuk 2 werden ‘land suitability evaluation’ (LSE) methoden gebruikt om
gebieden te identificeren die zowel geschikt als beschikbaar zijn voor
oliepalmproductie in Ghana. Hieruit bleek dat de oliepalmproductie in Ghana beperkt
werd door het jaarlijkse watertekort. De meest geschikte gebieden voor
oliepalmproductie waren te vinden in het regenwoud en semi-loofboszones met hogere
regenval in zuidelijk Ghana. Mogelijkheden voor de grootschalige oliepalmproductie
om zich beter te ontwikkelen werden ook geremd door een tekort aan grote,
aaneengesloten stukken land, wat een vereiste is voor de commerciële ontwikkeling van
oliepalmplantages.

201
SAMENVATTING

Een mogelijke geschikte strategie is daarom om de oliepalmproductie te verhogen via

kleine boeren, die gebruik kunnen maken van kleinere percelen land. Een andere
mogelijkheid voor een grotere oliepalmproductie is door verhoging van de opbrengst
op land waar al oliepalm aangeplant is.
Om opbrengst verhogende praktijken uit te voeren moet men eerst de grootte van de
yield gaps en de onderliggende oorzaken ervan begrijpen. In hoofdstuk 3, werd ‘Yield
Gap Analysis’ (YGA) gebruikt om de factoren toe te lichten die verantwoordelijk waren
voor yield gaps op plantages en kleine boeren. Plantages en bedrijven van kleine boeren
werden ingedeeld volgens hun productie inputs (o.a. arbeid, kunstmest, oogstresten, en
bestrijdingsmiddelen) en outputs (o.a. palmvrucht productie, aantal palmvruchten en
palmvrucht opbrengst per ha). De opbrengstdata van goed producerende
oliepalmvelden werd gebruikt als maatstaf voor de potentiële opbrengst (Yw). Yield
gaps werden berekend als het verschil tussen Yw en werkelijke opbrengsten, en
bedroegen gemiddeld 15.4 ton per ha bij kleine boeren en 9.8 ton per ha op plantages.
De analyse liet heel wat suboptimale managementpraktijken zien. De belangrijkste
factoren die bijdroegen aan de yield gaps waren het niet volledig oogsten van het gewas
en ontoereikend landbouwkundig management. Vooral de kleine boeren staan
tegenover grote uitdagingen om de opbrengsten te verhogen vanwege een gebrek aan
kennis en passende managementpraktijken en slechte exploitatieomstandigheden
(bijvoorbeeld slechte infrastructuur en geen toegang tot kredieten). Verliezen in
productie werden verergerd door lage olie-extractie (OER) van 12% door kleinschalige
verwerkers, vergeleken met grootschalige verwerkers van 21%. De potentiële verliezen
in jaarlijkse ruwe palmolie (CPO) gedurende de plateau opbrengstfase werd geschat op
jaarlijks 5 en 3 ton per ha voor respectievelijk kleinschalige en grootschalige
productiesystemen. De resultaten van deze studie suggereren dat het tekort aan palmolie
in Ghana aangepakt kan worden door meer te investeren in het dichten van yield gaps
door verbeterde agronomie en oogstpraktijken op plantages en bij kleine boeren, alsook
om ervoor te zorgen dat de kleinere boeren betere toegang krijgen tot
oliepalmverwerkers en door de verhoging van de efficiëntie van oliepalmverwerkers
zelf.
Best Management Practices (BMPs) zijn een manier om yield gaps van plantages en
kleine boeren te dichten (hoofdstuk 4). BMPs werden vergeleken met huidige
managementpraktijken, en werden ingedeeld onder maatregelen welke de opbrengsten
verhoogden op de korte termijn (≤ 1 jaar) met ‘yield taking’ (verbeterde
oogstpraktijken) en op de lange termijn (>1 jaar) met ‘yield making’ (betere
agronomische praktijken). Resultaten uit experimenten lieten zien dat de toenames in
opbrengsten groot zijn met BMP. De gemiddelde opbrengsten met BMP stegen met
2.1 ton per ha (+ 19%) en 4.7 ton per ha (+ 89%) bij ‘yield taking’ en met

202
 SAMENVATTING

4.7 ton per ha (+ 36%) en 7.6 ton per ha (+ 76%) bij ‘yield making’ voor respectievelijk
plantages en kleine boeren. Verbeteringen in ‘crop recovery’ (oogstpraktijken) met
inbegrip van frequenter oogsten (d.w.z. kortere oogstintervallen) en verbeterde toegang
tot het veld (wegen, paden, ‘weeded circles’ (boomspiegels)) voor een efficiëntere
productiviteit (d.w.z. hoger geoogst palmvruchtgewicht en kleiner oppervlak (ha) per
oogster) droegen aanzienlijk bij tot een verhoogde opbrengst. De volledige langere
termijn impact van BMPs was moeilijker vast te stellen, omdat de verbeterde
landbouwpraktijken nog lang effect na kunnen hebben. Maar de analyse van
kunstmestgebruik en van nutriënten in bladmonsters suggereerden dat een
evenwichtigere benadering van nutriëntenmanagement aanzienlijk kan bijdragen aan
‘yield making’, in het bijzonder bij de kleine boeren. Daarom is er behoefte aan het
ontwikkelen van verbeterde aanbevelingen met betrekking tot kunstmestgebruik, om zo
de opbrengsten verder te verhogen en om de oliepalmproductie duurzaam te
intensiveren in Ghana.
In hoofdstuk 5 lag de focus op het bepalen van de interactie van nutriënten en water op
de oliepalmopbrengst op een oliepalmplantage in Western region, Ghana. De
gemiddelde opbrengsten van ongeveer 32.6 ton per ha werden bereikt met een
combinatie van irrigatie en kunstmest. Yield gaps die veroorzaakt werden door
nutriënten, werden geschat op 4.1 ton per ha en door water 4.7 ton per ha. De respons
op irrigatie alleen zonder kunstmest was 0.6 ton per ha. Deze resultaten opperden dat
kunstmest essentieel was voor de toename in opbrengsten met irrigatie, maar informatie
over de hoeveelheid aan vereiste specifieke nutriënten kon niet bevestigd worden en
vereist verder onderzoek. Ondanks het gebruik van irrigatie bleven opbrengstcycli in
oliepalm bestaan. Dit kan leiden tot onregelmatige spreiding van de productie en een
tekort aan verwerkingscapaciteit en onvoldoende oogsters in de piek gewasmaanden. In
de literatuur zijn veel factoren naar voren gebracht die deze fluctuaties veroorzaken,
maar de processen achter deze opbrengstcycli zijn nog steeds onduidelijk.
Het onderzoek, dat in deze thesis beschreven is, toont aan dat er grote yield gaps te
vinden zijn in de productiesystemen van plantages en kleine boeren in Ghana. De
oorzaak daarvan is te vinden in de suboptimale klimaatomstandigheden (bijv. water) en
slechte managementpraktijken (bijv. crop recovery, nutriëntenmanagement). De
potentie om met BMP de productiviteit te verhogen is echter aanzienlijk. Door BMPs
toe te passen zijn er significante verbeteringen in de opbrengst bereikt op locaties die
representatief waren voor de Ghanese oliepalmindustrie. Deze resultaten bieden de
mogelijkheid om te voldoen aan de toenemende vraag naar palmolie, zonder het areaal
aan oliepalm verder uit te breiden. Het vergroten van de gemiddelde nationale
opbrengsten tot bijvoorbeeld 21.0 ton per ha zou kunnen leiden tot het uitsparen van
bijna 600,000 hectare land. Bovendien heeft het sluiten van de yield gaps in Ghana op

203
SAMENVATTING

het huidige landareaal (327,000 ha) al een potentieel om de productie te verhogen met
een factor 3, van 2.5 megaton tot 6.9 megaton palmvruchten. Als de hele oogst wordt
verwerkt bij een OER van 21% dan kan ongeveer 1.4 megaton CPO worden
geproduceerd. Dit is meer dan genoeg om te voldoen aan de jaarlijkse vraag in Ghana
van 106,000 ton CPO en dan is het niet nodig om meer land aan oliepalm aan te planten.
Intensivering van opbrengststrategieën met bijhorend landbeleid is daarom een
belangrijke strategie voor duurzame oliepalmontwikkeling in Ghana en West-Afrika.

204
 LIST OF PUBLICATIONS
Peer-reviewed publications
Rhebergen, T., Fairhurst, T., Whitbread, A., Giller, K.E. and Zingore, S., 2018. Yield
gap analysis and entry points for improving productivity on large oil palm
plantations and smallholder farms in Ghana. Agricultural Systems, 165: 14–25,
https://doi.org/10.1016/j.agsy.2018.05.012.
Rhebergen, T., Fairhurst, T., Zingore, S., Fisher, M., Oberthür, T. and Whitbread, A.,
2016. Climate, soil and land-use based land suitability evaluation for oil palm
production in Ghana. European Journal of Agronomy, 81: 1–14,
https://doi.org/10.1016/j.eja.2016.08.004.
Conference proceedings
Rhebergen, T., Hoffmann, M., Zingore, S., Oberthür, T., Acheampong, K., Dwumfour,
G., Zutah, V., Adu-Frimpong, C., Ohipeni, F. and Fairhurst, T., 2014. The
Effects of Climate, Soil and Oil Palm Management Practices on Yield in Ghana,
International Oil Palm Conference (IOPC), Bali, Indonesia, pp. 1–16.
Other publications
Rhebergen, T., Fairhurst, T., Zingore, S., Fisher, M., Oberthür, T. and Whitbread, A.,
2016. Adapting Oil Palm Best Management Practices to Ghana: Opportunities
for Production Intensification. Better Crops, 100(4): 12–15.
Rhebergen, T., Donough, C.R. and Sugianto, H. 2018. Oil Palm 4R Series, Volume 1
– Leaf Sampling Unit. International Plant Nutrition Institute, Southeast Asia
Program (IPNI SEAP), pp. 22.
Rhebergen, T., Donough, C.R. and Sugianto, H. 2018. Oil Palm 4R Series, Volume 2
– Soil Sampling. International Plant Nutrition Institute, Southeast Asia Program
(IPNI SEAP), pp. 30.
Rhebergen, T., Donough, C.R. and Sugianto, H. 2018. Oil Palm 4R Series, Volume 3
– Plant Sampling. International Plant Nutrition Institute, Southeast Asia Program
(IPNI SEAP), pp. 32.
Rhebergen, T., Donough, C.R. and Sugianto, H. 2018. Oil Palm 4R Series, Volume 4
– Growth Measurements. International Plant Nutrition Institute, Southeast Asia
Program (IPNI SEAP), pp. 42.

205
 CURRICULUM VITAE
Tiemen Rhebergen was born on 13 June 1983 in
Heerenveen, the Netherlands, but spent most of his
youth in the Lowveld of Swaziland. He has
therefore always had a passion for the natural
environment, particularly in Africa. In 2006, he
graduated with a BSc, and in 2008 with an MSc in
Forest and Nature Conservation, specialisation
Ecology at Wageningen University (WU). In 2010,
he decided to pursue a career in agriculture instead.
He enrolled in an MSc in Plant Sciences (WU),
which he completed in 2012. For his MSc Thesis,
he worked on Best Management Practices in Oil
Palm Plantations in Indonesia, from which he
developed interest in oil palm and plantation
agriculture in general. In 2013, he joined the International Plant Nutrition Institute, Sub-
Saharan Africa Program (IPNI SSA) as project manager on an oil palm yield
intensification program in Ghana. Under the framework of this project, he started his
PhD program at Wageningen University in 2014, in collaboration with IPNI. Tiemen is
currently based in Nairobi, Kenya, where he continues his work on oil palm in Africa.

207
 ACKNOWLEDGEMENTS
I would like to thank the following institutions and people who have supported me
during my PhD. Without their continuing support, assistance and guidance, the work
would not have been possible.
• Benso Oil Palm Plantation Ltd. (BOPP): Sophia Akuffo, Sylvia Armoo,
Ebenezer Arthur, Samuel Avaala, Edward Dagadu, Robert Damoah, Philip
Dong, Lawrence Dzidotor, Margaret Kainyah, Kumi Kwame Kwakye, Lim Peng
Ho, Paul Sam Tetteh Narh, Kwasi Baah Ofori, Eugene Opuni-Nimako, De-graft
Quartey, Matthew Woanyah, Victor Tetteh Zutah.
• Norpalm Ghana Ltd.: Sincere Kingsley Ayew, George Banson, George Mensah
Dwumfour, Ebenezer Kwesi Fenny, Stephen Armah Kotu, Alfred Kwofie,
Charles Mate-Kole, Alfred Aseidu Mensah, Osman Nuhu, Emmanuel Oduro,
Edward Kofi Panyin.
• Twifo Oil Palm Plantation Ltd. (TOPP): Kwame Acheampong, Seth Addo-
Kumi, Emmanuel Ahiable, Joseph Anaafi, Patricia Atebawone, Emmanuel Atoo
Otoo, Mercy Boateng, Stephen Bosie, Odartey Lamptey, Joseph Mensah, David
Nunoo, Richard Oboh, Eric Ofosu, George Ohene Kena, Charles Sackey, Benard
Tetteh, Francis Yalley.
• Solidaridad West Africa: Rosemary Addico, Charles Adu-Frimpong, Stephen
Appiah, Joan Atiase, Benjamin Gabla, Isaac Gyamfi, Harrie Hendrickx, Delle
Kpebesaan, Seth Nuamah, Francis Tetteh Ohipeni, Jabulani Sithole.
• Sustainable West Africa Palm Oil Programme (SWAPP) farmers: Charles Adu,
Kwaw Afibe-Yanney (†) Kwame Amo, Kwasi Amoasi, Barima Okyere
Ampadu, Joseph Yaw Awusie, Innocent Ayle, Kwame Boampong, Oti Boateng,
Frederick Danso, John K. Fiadonu, Felix Fynn, Moro Ibrahim, Yusif Adu
Mensah, Thomas Ntiako, Godfred Nutakor, Tufour Quiaicco, Aaron Sagoe,
Nana Kwasi Somprey, Yaw Boakye Yiadom.
• The International Plant Nutrition Institute (IPNI): Chris Donough, Adrian
Johnston, Kaushik Majumdar, Samuel Njoroge, Thomas Oberthür, Shamie
Zingore.
• Wageningen University & Research Center (WUR): Ken Giller, Petra Rietberg,
Maja Slingerland, Gerrie van de Ven, Lotte Woittiez, Mink Zijlstra.
• Göttingen University: Munir Hoffmann, Anthony Whitbread.
• K+S KALI GmbH: Olivier Goujard, Andreas Gransee, Rolf Härdter, Gerald
Sternberg.
• Canpotex: Kevin Lee.
• Agrisoft Systems: Florian Decker, Max Kerstan.

209
ACKNOWLEDGEMENTS

• Tropical Crops Consultants Ltd. (TCCL): Joan Fairhurst, Thomas Fairhurst.

• Miscellaneous: Rob Moss, Wim Rhebergen, Paul Schofer, Godfrey Taulya,
Jonathan Telfer, Rob Verdooren, Herbert van der Vossen, Na Wang.
In particular, I would like to thank Ken Giller, Thomas Fairhurst, Anthony Whitbread
and Shamie Zingore for their valuable supervision. Specials thanks goes to Thomas
Fairhurst for the good times, friendship and support in the field. Thomas’ support and
guidance was key to the success of the project. Finally, I thank my family and my wife
Ingrid Coninx for her support.

210
 PE&RC Training and Education Certificate
With the training and education activities listed below the PhD
candidate has complied with the requirements set by the C.T.
de Wit Graduate School for Production Ecology and Resource
Conservation (PE&RC) which comprises of a minimum total of
32 ECTS (= 22 weeks of activities)
Review of literature (4.5 ECTS)
• Yield gap analysis in oil palm production systems in
Ghana; presented at Göttingen University (2014)
Writing of project proposal (4.5 ECTS)
• Yield gap analysis in oil palm production systems in Ghana
Post-graduate courses (6.6 ECTS)
• Training on vegetative growth measurements in mature oil palm stands; IJM
Plantations Sabah, Malaysia (2013)
• Multivariate analysis; WUR (2015)
• Introduction to R; WUR (2015)
• Geostatistics; WUR (2015)
• Spring school ISRIC WSA; ISRIC (2016)
Invited review of (unpublished) journal manuscript (1 ECTS)
• Geoderma: changes in variability and spatial structure of soil organic carbon
under tree plantations in Okomu forest southern Nigeria (2016)
Deficiency, Refresh, Brush-up courses (6 ECTS)
• Nutrient management; WUR (2015)
Competence strengthening / skills courses (5 ECTS)
• Philosophy of science and ethics; WUR (2003)
• Mentoring process on scientific writing; IPNI (2015)
PE&RC Annual meetings, seminars and the PE&RC weekend (1.2 ECTS)
• PE&RC First years weekend (2015)
• PE&RC Day (2015)
Discussion groups / local seminars / other scientific meetings (6.6 ECTS)
• Oil palm workshop with Thomas Fairhurst and Lotte Woittiez; Wye, UK (2014)

211
PE&RC Training and Education Certificate

• Seminar: trade-offs and synergies in climate change adaptation and mitigation in

coffee and cocoa systems; Göttingen, Germany (2014)
• Meeting and scientific discussion on implementing fertilizer trials in oil palm at
K+S; Kassel, Germany (2014)
• Seminars attended, and given at Göttingen, Germany (2014, 2015)
• Lunch-break meetings; WUR (2015-2018)
• Attendance of PhD defences (2015-2018)
• Regular scientific discussions with oil palm professionals, Thomas Fairhurst,
Herbert van der Vossen, Rob Verdooren (2015-2018)
• Regular discussions with Petra Rietberg & Gerrie van de Ven on the development
of an oil palm economic model; PPS-WUR (2016)
International symposia, workshops and conferences (12 ECTS)
• International Oil Palm Conference (IOPC); Bali, Indonesia (2014)
• Annual oil palm project review and planning meeting; Takoradi, Ghana (2014-
2017)
• IPNI Staff meeting; Naivasha, Kenya (2015)
• CocoaSoils research committee meeting; Utrecht, the Netherlands (2017)

212
 FUNDING
The research presented in this thesis was funded by:
• Canpotex International Pte Limited
• K + S Kali GmbH
• Solidaridad
• International Plant Nutrition Institute (IPNI)

213