Sie sind auf Seite 1von 30

Identification and elimination of 

yield gaps in oil palm plantations 
in Indonesia
T.H. Fairhurst, W. Griffiths, C. Donough, C. Witt, 
D. McLaughlin, K.E. Giller

Tropical Crop Consultants Ltd, Wye, UK
Private address, Cairns, Australia
International Plant Nutrition Institute (IPNI), Penang, Malaysia
World Wildlife Fund (WWF), Washington DC, USA
Wageningen University, Netherlands

24 September, 2010 ©Tropical Crop Consultants Ltd 1


Outline of presentation

ƒ Context
ƒ Theory of yield gap management and best 
management practices (BMPs)
ƒ Practical implementation
ƒ Conclusions and perspective

24 September, 2010 ©Tropical Crop Consultants Ltd 2


CONTEXT ‐ Challenges and opportunities
ƒ Demand for vegetable oil to double by 2050 
(Corley, 2009)
ƒ At current yield an extra 12M ha of oil palm 
needed
ƒ Increasingly stringent environmental controls
ƒ Crop carbon footprint
ƒ Forest displacement for new development

24 September, 2010 ©Tropical Crop Consultants Ltd 3


Why focus on yield intensification?

ƒ For the grower
ƒ Maximize return on investment
ƒ Increase IRR and reduce payback period
ƒ Improved public profile
ƒ For the public and NGOs
ƒ Efficient use of land occupied by oil palm
ƒ Spare land (and rainforest) for other uses

24 September, 2010 ©Tropical Crop Consultants Ltd 4


BACKGROUND ‐ Site yield potential

ƒ In Indonesia and Malaysia ~35 t ha‐1 of fresh fruit 
bunches = 8 t ha‐1 of oil?
ƒ Derived from
ƒ Fertilizer trials
ƒ Blocks under long term best management

24 September, 2010 ©Tropical Crop Consultants Ltd 5


How to optimize three production 
phases?

Bunch yield (t ha‐1)
Plateau yield
35 Declining yield 
phase
phase
30
Steep ascent
• Shorten time to 
25
yield phase maturity and peak 
yield
20
• Prolong plateau 
15 phase
10
Low 17 t/ha 
• Reduce rate of 
5
Yield  Med 23 t/ha  decline
building  High 29 t/ha 
phase
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Year after planting (YAP)
After Ng, 1976
24 September, 2010 ©Tropical Crop Consultants Ltd 6
Measure change in frequency of yields for 
soil x palm age groups over time

Frequency (% area)

Reduce 
variability 
and increase 
yield!

0 5 10 15 20 25 30 35 40 45
‐1
Bunch yield (t ha )
24 September, 2010 ©Tropical Crop Consultants Ltd 7
Oil palm productivity is very sensitive to environmental stress
24 September, 2010 ©Tropical Crop Consultants Ltd 8
An interval of >36 months elapses between the formation 
of a flower and the production of a ripe bunch!

Source: Donough, 2008
24 September, 2010 ©Tropical Crop Consultants Ltd 9
Potential yield of a progeny under a given 
soil type and climate

Yield (% potential)
100
Yield 
potential of 
90 progeny for 
a given soil 
80 and climate

70

60

50

40

Y‐a Y‐n Y‐mey Y‐max

24 September, 2010 ©Tropical Crop Consultants Ltd 10


Yield gap 1 caused by deficiencies in 
planting technique 

Yield (% potential)
100
Yield gap 1 Yield 
potential of 
90 Maximum  progeny for 
economic  a given soil 
80 yield and climate

70

60

50

40

Y‐a Y‐n Y‐mey Y‐max

24 September, 2010 ©Tropical Crop Consultants Ltd 11


Causes of Yield Gap 1?

ƒ Poor plantation establishment
ƒ Poor nursery technique and culling
ƒ Erosion and compaction at land clearing
ƒ Incorrect planting density or inaccurate lining
ƒ Failure to replace unproductive palms
ƒ Poor gap filling at planting
ƒ Gaps due to palm death
ƒ Failure to establish legume cover plants

24 September, 2010 ©Tropical Crop Consultants Ltd 12


Yield gap 2 caused by nutrient deficiencies

Yield (% potential)
100
Yield gap 1 Yield 
Yield gap 2 potential of 
90 Maximum  progeny for 
economic  a given soil 
80 yield and climate
Yield reduced 
70 because of 
nutrient 
deficiencies
60

50

40

Y‐a Y‐n Y‐mey Y‐max

24 September, 2010 ©Tropical Crop Consultants Ltd 13


Causes of Yield Gap 2?

ƒ Nutrient constraints
ƒ Failure to take account of soil variability
ƒ Faulty leaf sampling
ƒ Insufficient field inspection to corroborate results 
of leaf analysis
ƒ Failure to use long term data trends
ƒ Failure to make spatial analysis of nutritional 
trends

24 September, 2010 ©Tropical Crop Consultants Ltd 14


Yield gap 3 caused by poor management

Yield (% potential)
100
Yield gap 1 Yield 
Yield gap 2 potential of 
90 Yield gap 3 Maximum  progeny for 
economic  a given soil 
80 yield and climate
Yield reduced 
70 Yield reduced  because of 
because of nutrient 
poor  deficiencies
60 management

50

40

Y‐a Y‐n Y‐mey Y‐max

24 September, 2010 ©Tropical Crop Consultants Ltd 15


Causes of Yield Gap 3?

ƒ Poor harvesting and management
ƒ Inadequate infrastructure (mill‐to‐palm access)
ƒ Poor round control
ƒ Poor harvest supervision
ƒ Failure to implement fertilizer and crop residue 
application programmes accurately
ƒ Human resource management

24 September, 2010 ©Tropical Crop Consultants Ltd 16


SITE ASSESSMENT ‐ Plot frequency of yields 
Number of blocks
200

150
Focus more 
attention on 
blocks with 
100 greatest scope 
for 
50
improvement

0
0 5 10 15 20 25 30 35 40 45

Yield (t ha‐1)
24 September, 2010 ©Tropical Crop Consultants Ltd 17
Spatial analysis of yield gaps: 
Are blocks with large yield gaps 
dispersed or clustered?

Source: Gfroerer, 2009
24 September, 2010 ©Tropical Crop Consultants Ltd 18
PRACTICAL IMPLEMENTATION of BMPs

ƒ Pilot phase runs for four years
ƒ Evaluation
ƒ Productivity
ƒ Cost benefit analysis
ƒ Changes required in management
ƒ Broad scale implementation (may begin after 
one year)

24 September, 2010 ©Tropical Crop Consultants Ltd 19


Fertilizer use efficiency is increased with proper spreading
24 September, 2010 ©Tropical Crop Consultants Ltd 20
Agronomic database provides 
quantitative basis for yield maximization

Source: Gfroerer, 2009
24 September, 2010 ©Tropical Crop Consultants Ltd 21
Healthy 
leaf canopy

Sufficient  Full 
ground cover access

Complete 
crop recovery

Excellent standards of management in place in a BMP block
24 September, 2010 ©Tropical Crop Consultants Ltd 22
RESULTS AND CONCLUSIONS

24 September, 2010 ©Tropical Crop Consultants Ltd 23


Comparison of BMP and non‐BMP blocks 
over five years

Yield (t ha-1) Bunch weight (kg) Bunch number


28 30 12
26 Non-BMP
24 BMP 11
25
22 10
20
18 20 9
16
14 8
15
12 7
10
8 10 6
01 02 03 04 05 01 02 03 04 05 01 02 03 04 05
Year Year Year

24 September, 2010 ©Tropical Crop Consultants Ltd 24


Results from six sites in Indonesia

ƒ Average yield increase of 3.2 t ha‐1 (range 0.4‐
30 ‐1)
6 t ha REF 
BMP 
ƒ Average increase in bunch number (114 bunch 
25
Bunch yield (t ha )
‐1

ha‐1) and bunch weight (+1 kg)
20

ƒ Less difference between sites after BMP 
15

10
implementation
5

0
N Sum 1 E Kal W Kal C Kal S Sum N Sum 2
Site Source: Donough, 2009
24 September, 2010 ©Tropical Crop Consultants Ltd 25
BMP works on degraded lands 
(anthropic savanna of Imperatacylindrica)

24 September, 2010 ©Tropical Crop Consultants Ltd 26


Plantation crops

24 September, 2010 ©Tropical Crop Consultants Ltd 27


Scaling up

ƒ Only where pilot phase reveals economically 
worthwhile yield improvement
ƒ Stepwise implementation in 1,000‐1,500 ha 
blocks
ƒ Identify all constraints and plan their removal

24 September, 2010 ©Tropical Crop Consultants Ltd 28


Conclusions

ƒ Determine potential yield for all sites
ƒ Goal of management is to minimize the gap 
between achievable and actual yield
ƒ BMP is a step‐wise process to close yield gaps
ƒ Small scale pilot phase
ƒ Scale up once evidence of gaps available
ƒ Maximum economic yield is the key to 
profitability and competitiveness
24 September, 2010 ©Tropical Crop Consultants Ltd 29
Thank you for your attention!
Tropical Crop Consultants Ltd
www.tropcropconsult.com

www.ipni.net
24 September, 2010 ©Tropical Crop Consultants Ltd 30

Das könnte Ihnen auch gefallen