The Use of El Nino-Southern Oscillation Information for Crop
Management Decisions and Its Economic Implication: Case Study for
Rice-Base Farming System of West Java, Indonesia
Rizaldi Boer1, and Elza Surmaini2
1
Centre for Climate Risk and Opportunity Management in Southeast Asia and Pacific of
Bogor Agricultural University, Bogor-Indonesia E-Mail: rizaldiboer@gmail,com
2
Indonesian Agroclimate and Hydrology Research Institute, Bogor-Indonesia
Abstract
In The El-Nino-Southern Oscillation (ENSO) ‘warm event’ or El-Niño years, for most
cases the dry season starts earlier and the rainfall amount is often substantially reduced in
Indonesia’s region. One of indicators used for detecting the occurrence of ENSO events
is Southern Oscillation Index (SOI). Based on historical rainfall data in the study areas,
i.e. Ciparay and Bojongsoang sub-districts, Bandung District, when SOI phase prior to
this season was constantly negative (phase 1) or rapidly falling (phase 3) indicating ElNiño cycle, the chance of having rainfall far below normal in this season increased.
Therefore, the use SOI phase information prior to the season would be useful for
assisting farmers in making a satisfactory planting decision. The results of this study
indicates that farmers who switch their crop to soybean or maize for May planting if
April SOI Phase is 1 or 3 will get a higher income compared to farmers that keep
planting rice. Cumulatively, over 62 years, the net annual income difference between
farmers that use April SOI Phase information and chose soybean for May planting and
those that do not is about 1,700 USD at Ciparay and 2,350 USD at Bojongsoang.
Key words: SOI phases, dry season, drought, rice based farming system.
1. Introduction
Floods and droughts are climate hazards that commonly occur in rice-based
farming system in West Java, Indonesia (Boer and Subbiah, 2005; Ditlin 2007). In this
farming system, farmers plant rice crop twice a year; the first rice crops are planted in the
wet season (between November and February), and the second rice crops are planted in
the dry season (between March and June). The first crop is commonly affected by floods
and the second by drought. Floods commonly occurred between January and March,
while drought typically develops in May. These hazards are frequently associated with
ENSO events. In ENSO ‘warm event’ (El-Niño) years, for most cases the dry season
starts earlier and the rainfall amount is often substantially reduced (Boer and Subbiah,
2005; Kirono and Partridge, 2002; Moron et al., 2009). Under this condition most of
second rice crops will suffer from drought (Alimoeso et al., 2002; Meinke and Boer,
2002).
SOI phases have been commonly used to indicate the occurrence of El-Nino or
La-Nina. SOI consist of five phases (Stone et al., 1996) namely consistently negative
(Phase 1), consistently positive (Phase 2), rapidly falling (Phase 3), rapidly rising (Phase
4) and near zero (Phase 5). Phase 1 and 3 is commonly associated with ENSO ‘warm
event’, Phase 2 and 4 with ENSO ‘cold event’, and Phase 5 with normal condition.
Many studies have demonstrated that the SOI phases can provide skill in assessing future
rainfall probabilities during the subsequent planting season (e.g. Stone and Auliciems,
1992; Stone et al, 1996; Everingham et al. 2009; Fawcett and Stone, 2010). In Australia,
the SOI phase information has been used widely in managing climate risk to crop
production (Meinke, et al. 1996; Meinke and Stone, 1997; Meinke and Hochman, 2000;
Everingham et al., 2003; Wang et al. 2008) and also in some other countries such as
India (Selvaraju et al. 2004), Pakistan (Meinke et al. 2001), Argentina (Messina et al.
1999), Zimbabwe (Venus, 2000; Zinyengere, 2010), Canada and USA (Hill et al. 2000).
The use SOI phase information prior to the season would be useful for assisting farmers in
making a satisfactory planting decision. This study aimed to evaluate the economic benefit
of using SOI (Southern Oscillation Index) phase information in March/April to
determine a satisfactory planting decision for the second crop (April/May planting).
2. Methods
Location selected for this study was Ciparay and Bojongsoang sub-districts of
Bandung District, West Java Province, Indonesia. These two sub-districts like the rest of
West Java have monsoon rainfall (‘Monsoon’) with a monthly rainfall peak in December.
Areas with this type of rainfall are significantly affected by ENSO events (Tjasyono,
1997; Boer and Faqih, 2004; Boer and Subbiah, 2005).
Rice based farming systems and climate related hazards at the two sub-districts
were evaluated through survey. The likely crops yields across seasons were evaluated
using crop simulation models (DSSAT; Hoogenboom, 2010) with long-term, daily
rainfall records (1950-2012). As the model requires other climatic data such as solar
radiation and temperature, these two data were generated using the observed daily
rainfall data (Boer et al., 2004).
The DSSAT was validated using experimental data
collected from Sukamandi Rice Research Institute and Bogor Research Institute for Food
Crops. The experiments were conducted in a number of sites, i.e. Bogor, Geritan, Pati
and Binangun, West Java. The differences between observed and simulated yields were
assessed using statistical t-test.
In general the cropping pattern in rice based farming system in West Java is
dominated rice-rice-fallow. The first rice crop is planted in wet season (November-
February) with peak planting time in January, and the second crops in dry season
(March-June) with peak planting time in May. In El-Nino years, May planting was
normally exposed to high drought risk. This study evaluated the variability of crops
yields from year to year under three different planting arrangements.
The first
arrangement, the first crop was planted in 1st of December and the second crop in 1st of
April. The second, the first crop was planted in 1st of January and the second crop in 1st
of May. The third, the first crop was planted in 15th of January and the second crop in
15th of May. March/April SOI phase information is then used to determine whether the
second crop would (April/May planting) be continued with rice or changed with other
crops, i.e. maize or soybean or left to be fallowed.
With the above arrangement, there were four planting decisions namely: (i) RiceRice system, in which farmers will keep planting rice twice a year irrespective of SOI
phase information; (ii) Rice-Maize system, in which farmers will switch their second
crop into maize if March/April SOI Phase is 1 or 3, and keep planting rice otherwise; (iii)
Rice-Soybean system, in which farmers will switch their second crop into soybean if
March/April SOI Phase is 1 or 3, and keep planting rice otherwise; and (iv) Rice-fallow
system, in which farmers will not plant a second crop if March/April SOI Phase is 1 or 3,
but keep planting rice otherwise.
The economic benefit of using SOI phases was
assessed by cumulating the difference between gross margins, (i) the Rice-Rice system
and the Rice-Maize system, (ii) the Rice-Rice system and the Rice-Soybean system, and
(iii) the Rice-Rice system and the Rice-Fallow system. The gross margin is the
difference between revenue and cost.
3. Results and Discussion
Rice Based Farming System and Climate Related Problems in West Java. Most
of agriculture areas in the two sub-districts are irrigated, only a few are rainfed. Farmers
normally plant rice twice a year in the irrigated areas and once a year in rainfed areas.
Maize is the second main crops after rice. Other crops are cassave, sweet potato, onion,
long bean, chilli, cucumber and some other vegetable crops. Soybean is planted only by
very few farmers. At present, most farmers do not use their rice field for planting maize
or soybean. They normally plant maize in dryland throughout the year. Soybean is also
planted throughout the year but there is a big variation of planting area between years.
1. Technical irrigared lands

Rice-Rice-Fallow
2. Conventional irrigated lands or rainfed: rice-rice-fallow

Rice-Rice-Fallow

Rice-Upland-Vegetable/fallow

Rice-vegetable-vegetable/Fallow
3. Dry lands agriculture:

Maize-Maize-Vegetable

Maize-soybean-vegetable
Like other rice growing areas of West Java, flood and drought are two common
climate hazards in these sites. Based on interview with farmers, flood commonly occur
between January and April (flood risk increases when monthly rainfall in these months
increase above 300 mm) while drought could start developing in May. Drought will
affect second crops when rainy season ends earlier or rainfall in the dry season fall far
below normal. Too late planting for the second rice crop due to much delay of the onset
of wet season can also cause drought. These conditions commonly occur in El-Nino
years. Too late planting for second rice will not cause drought problem when rainfall in
the season is much above normal which is commonly occur in La-Nina years. A study at
Indramayu district of West Java province show clear connection between ENSO events
and rainfall characteristic and its association with drought occurrence (Fig. 1).
This result of the study shows that the April SOI phases can provide skill in
assessing DS rainfall probability distribution of May-June rainfall anomaly of the study
areas (Fig 2).
Probability to have above average rainfall (indicating by positive
anomaly) in May-June increases when April SOI phase is 2 or 4. Probability to have
below average rainfall (indicating by negative anomaly) in May-June increases when
April SOI phase is 1 or 3. This suggests that profitability of rice farmers could be
increased by using knowledge of atmospheric conditions obtained prior to planting to
forward estimate production level and risk.
The above results indicate that in rice based cropping system, the timing of the
first planting and rainfall condition during the second planting season are the two
primary factors that should be considered in making second planting decision. Much
delay in first planting due delay in the onset of rainy season will cause much delay in
second planting. Under these circumstances, keeping planting rice for the second crop
may be exposed to high drought risk. Fig. 3 shows that weak El-Nino that occurred in
2003 caused higher drought impact than that occurred in strong 1997 El-Nino. This was
because the second planting for rice delayed between one and two months due to late
planting of the fist rice crops caused by late onset of wet season in 2002.
In 1997, the
El-Nino developed rapidly and very strong compare other El-Nino years. This caused
the rainy season ended earlier and rainfall in May and June already disappeared before
some farmers start planting their second rice crop. Therefore, total area being affected
by drought in the strong 1997 El-Nino was lower than those in other El-Nino years.
Strong connection between Sea Surface Temperature (SST) anomaly at Pacific
Ocean and rice production in Indonesia has motivated a number of scientists to develop
rice production model from the SST anomaly.
Falcon et al (2004) found that the rice
production anomaly in Indonesia can be well predicted from August sea surface
temperature anomaly (AugSSA) over the Pacific Ocean. The equation as the following:
Paddy production = 35,311 – 1,318 AugSSTA + 1,675 t – 45.0 t2 (1); Adj R2 = 0.94
Where t is time measured in years, where 1 = 1983/84.
Economic Benefits of Using SOI Phase Information. The result of validation of
the DSSAT model used to evaluate economic benefits of using SOI phase information in
crop management decision was reasonably good (Fig 4). The differences between
simulated and observed yields were not significantly different.
Further analysis to assess the economic benefits of using SOI phase was then
done using the simulated yield. The results of the analysis show that farmers who switch
their crop to soybean or maize for May planting if April SOI Phase is 1 or 3 will get a
higher income compared to farmers that keep planting rice (Fig. 5). Cumulatively, over
62 years, the net income difference between farmers that use April SOI Phase
information in deciding a crop type for May planting and those that do not is about 1,700
USD at Ciparay and 2,350 USD at Bojongsoang if they chose soybean as the second
crops, and about 1,524 USD and 1,970 USD respectively if they chose maize. Farmers
who choose not to plant a second crop will accrue a lower income than those that plant
the crops. Furthermore, farmers who decide not to plant a second crop in April if March
SOI Phase is 1 or 3 will get a much lower income compared to farmers who plant the
second crops. This outcome suggests that the decision to planting or not to plant a
second crop in May is more crucial than in April. Farmers who make what appears to be
a proper decision for May planting based on April SOI Phase will, in the long term get
consistently a much higher benefit than farmers who keep planting rice irrespective of
the April SOI Phase.
This approach can be used to assess the economic benefit of changing
management options such as introducing new crop sequence (non-rice-rice-non-rice) if
length of rainy season shortened or changing second crop with horticulture etc. The
knowledge gained in study needs to be communicated to farmers so they can make us of
the knowledge to assist them in coping with drought hazards. There are a number of
means to communicate technologies and knowledge to farmers and each country will
have different experiences in doing this. Barriers and constraints faced by farmers for
implementing the options also need to be assessed and removed.
4. Conclusion
Yield variability at Ciparay and Bojongsoang is significantly correlated with
ENSO (El Nino Southern Oscillation). Consistent use of this information in designing
cropping management strategy would give benefit to farmers. Farmers’ who changes the
second crop from rice to non-rice crops (soybean and maize) whenever April SOI Phase
is rapidly falling or consistent negative will have higher income than farmers’ who keep
planting rice.
Since many parts of Indonesia are strongly influence by ENSO, the use of ENSO
information in designing better cropping strategies in particular season is encouraged.
Acknowledgement
Our thanks to Holger Meinke for valuable inputs. This work was supported by APN
CAPaBLE Project.
References
Alimoeso S, Boer R, Subroto SWG, Purwani ET, Sugiarto Y, Rahadiyan RMK and
Suciantini (2002). Distribution of drought prone rice growing area in Indonesia.
Directorate of Plant Protection, Ministry of Agriculture, Jakarta (in Indonesian).
Boer R and Faqih M (2004). Global climate forcing factors and rainfall variability in
West-Java. Indones J Agric Meteorol 18:1-12
Boer R, Rakiso P , Faqih A (2004). The use of climatic data generator to cope with daily
climatic data scarcity in simulation studies.
In Proceedings of the 4th
International Crop Science Congress Brisbane, Australia, 26 Sep – 1 Oct 2004,
http://www.cropscience.org.au/icsc2004/poster/2/6/index.htm#TopOfPage
Proceedings of the 4th International Crop Science Congress, 26 Sep – 1 Oct
2004, Brisbane, Australia. Published on CDROM (www.cropscience.org.au)
Boer R, Subbiah AR (2005)
Agricultural droughts in Indonesia.
In Boken VK,
Cracknell AP, Heathcote RL. Monitoring and Predicting Agriculture Drought.
Oxford University Press, p:330-344.
Ditlin. 2007. Pedoman Mitigasi Dampak Fenomena Iklim: Pengelolaan Dampak
Kejadian Iklim Ekstrim pada Tanaman Pangan. Direktorat Perlindungan
Tanaman, Departemen Pertanian, Jakarta.
Everingham YL, Zamykal D, McKinna LI (2009). Rain forecaster—a seasonal climate
forecasting tool. Proceeding Australian Society Sugar Cane Technology 31:5064.
Everingham YL,
Muchow RC,
Stone RC, Coomans DH (2003)
Using southern
oscillation index phases to forecast sugarcane yields: a case study for
northeastern Australia. Int. J Climatology 23: 1211–1218
Falcon WP, Naylor RL, Smith WL, Burke MB, McCullough EB (2004). Using Climate
Models To Improve Indonesian Food Security, Bull Indones Econ Stud 40: 355–
77
Fawcett RJB, Stone RC (2010). A comparison of two seasonal rainfall forecasting
systems for Australia. Australian Meteorological and Oceanographic Journal 60:
15-24
Hill JP, Mjelde JW, Rosenthal W, Love HA, Fuller SW (2000) Comparing the value of
Southern Oscillation index-based climate forecast methods for Canadian and US
wheat producers. Agric. For. Meteor., 100, 261–272.
Hoogenboom G, Jones JW, Wilkens PW, Porter CH, Boote KJ, Hunt LA, Singh U,
Lizaso JL, White JW, Uryasev O, Royce FS, Ogoshi R, Gijsman AJ, Tsuji GY,
Koo J (2010) Decision Support System for Agrotechnology Transfer (DSSAT)
Version 4.5 [CD-ROM]. University of Hawaii, Honolulu, Hawaii.
Kirono D, Partridge IJ. 2002. The climate and the SOI. p. 17-24. In Partridge IJ,
Ma’shum M(ed) Will It Rain?: The effect of the Southern Oscillation and El
Niño in Indonesia. Queensland Government, Department of Primary Industry,
Australia.
Meinke H, Boer R (2002) Plant growth and the SOI. In Partridge IJ, Ma’shum M (ed)
Will It Rain? The effect of the Southern Oscillation and El Niño in Indonesia.
Queensland Government, Department of Primary Industry, Australia. p. 25-28.
Meinke H, Baethgen WE, Carberry PS, Donatelli M, Hammer GL, Selvaraju R,
Stockle CO (200) Increasing profits and reducing risks in crop production using
participatory systems simulation approaches. Agric. Syst70: 493–513.
Messina C, Hansen JW, Hall AJ (1999) Land allocation conditioned on El Niño–
Southern Oscillation phases in the Pampas of Argentina. Agric. Syst., 60, 197–
212.
Moron V, Robertson A. Boer R (2009) Spatial Coherence and Seasonal Predictability of
Monsoon Onset over Indonesia. J Clim 22: 840–850.
Selvaraju R., Meinke H, Hansen J (2004). Approaches allowing smallholder farmers in
India to benefit from seasonal climate forecasting.
Proceedings of the 4th
International Crop Science Congress, 26 Sep – 1 Oct 2004, Brisbane, Australia.
Published on CDROM (www.cropscience.org.au)
Stone RC, Auliciems A (1992). SOI phase relationship with rainfall in earstern Australia.
Int J Climatology, 12:625-636
Stone RC, Hammer GL, Marcussen T (1996) Prediction of global rainfall probabilities
using phases of the Southern Osciallation Index. Nature 384:252-255.
Tjasyono B (1997) Mekanisme fisis para, selama, dan pasca El Niño. Paper disajikan
pada Workshop Kelompok Peneliti Dinamika Atmosfer, 13-14 Maret 1997.
Venus V (2000) Towards a framework for maize yield forecasting; a comparison of two
promising methods: The Case of Mashonaland West Province, Zimbabwe.
Thesis submitted to the International Institute for Aerospace Survey and Earth
Sciences in partial fulfillment of the requirements for the degree of Master of
Science in Rural Land Ecology Survey for Natural Resource Management. The
Netherland
Wang E., Xu JH, Smith CJ (2008) Value of historical climate knowledge, SOI based
seasonal climate forecasting and stored moisture at sowing in crop nitrogen
management in south eastern Australia. Agric For Meteor 148:1743-1753
Zinyengere N (2010) Maize yield prediction using seasonal weather forecasts and a crop
growth simulation model. Thesis submitted in partial fulfillment of the
requirements for the degree of Master of Science in Agricultural Meteorology,
University of Zimbabwe
80000
70000
200
150
100
40000
30000
20000
50
0
10000
0
Okt-90
Nov-90
Dec-90
Jan-91
Feb-91
Mar-91
Apr-91
May-91
Jun-91
Jul-91
Aug-91
Sep-91
60000
50000
DS rainfall far
below normal
Normal
Rainfall
Rainfall
Planting Area
80000
70000
60000
50000
40000
30000
20000
10000
0
Area (ha)
300
250
400
350
300
250
200
150
100
50
0
Oct-97
Nov-97
Dec-97
Jan-98
Feb-98
Mar-98
Apr-98
May-98
Jun-98
Jul-98
Aug-98
Sep-98
Rainfall
Planting Area
Drought Area
Rainfall (mm)
Normal
Rainfall
Area (ha)
Rainfall (mm)
400
350
DS rainfall far
above normal
Fig 1. Relationship between rainfall characteristics and drought occurren ce in El-Nino
(1990/91) and La-Nina (1997/98) years at Indramayu District of West Java
Province (Boer, 2002)
Probablity of Exceedence
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-200
Phase 1+3
Phase 2+4
Phase 5
0
200
400
Anomaly May-June Rainfall (mm)
Fig 2. Probability of exceedence of May-June rainfall anomaly in the study area by
April SOI phases 1+3, 2+4 and 5
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
Anomaly of SST 3.4 (oC)
-0.5
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
Cumulatuve Drought Area (000 ha)
3.0
2.5
2.0
1.5
1.0
0.5
0.0
-1.0
1991
1991
1994
1994
1997
1997
2002
2002
2003
25
20
15
10
5
0
2003
Fig 3. Relationship between El-Nino development and cumulative monthly drought rice
area in Bandung district.
3
7
6
Simulated Yield (t/ha)
r = 0.87
5
4
3
2
1
0
Simulated Yield (t/ha)
4
8
Simulated Yield (t/ha)
Soybean
Maize
Rice
r = 0.81
3
2
1
0
0
1
2
3
4
5
6
Observed Yield (t/ha)
7
8
r = 0.90
2
1
0
0
1
2
3
Observed Yield (t/ha)
4
0
1
2
3
Observed Yield (t/ha)
Fig 4. Comparison between simulated yields from DSSAT and observed yields for rice,
maize and soybean
1st Planting: 1 Dec, 2nd Planting: 1 April
Ciparay
0
-1000
-2000
-3000
-4000
Rice-Maize
Rice-Soybean
Rice-Fallow
2000
1000
0
-1000
-2000
-3000
-4000
Rice-Maize
Rice-Soybean
Rice-Fallow
1950
1955
1960
1965
1970
1975
1981
1987
1992
1997
2002
2007
2012
1000
Cumulative Income Difference
from Rice-Rice System (USD)
2000
Bojongsoang
3000
1950
1955
1960
1965
1970
1975
1981
1987
1992
1997
2002
2007
2012
Cumulative Income Difference
from Rice-Rice System (USD)
3000
1st Planting: 1 Jan, 2nd Planting: 1 May
1000
0
-1000
-2000
-3000
-4000
Rice-Maize
Rice-Soybean
Rice-Fallow
Cumulative Income Difference
from Rice-Rice System (USD)
2000
3000
Bojongsoang
2000
1000
0
-1000
-2000
-3000
-4000
Rice-Maize
Rice-Soybean
Rice-Fallow
1950
1955
1960
1965
1970
1975
1981
1987
1992
1997
2002
2007
2012
Ciparay
1950
1955
1960
1965
1970
1975
1981
1987
1992
1997
2002
2007
2012
Cumulative Income Difference
from Rice-Rice System (USD)
3000
1st Planting: 15 Jan, 2nd Planting: 15 May
1000
0
-1000
-2000
-3000
-4000
Rice-Maize
Rice-Soybean
Rice-Fallow
Cumulative Income Difference
from Rice-Rice System (USD)
2000
3000
Bojongsoang
2000
1000
0
-1000
-2000
-3000
-4000
Rice-Maize
Rice-Soybean
Rice-Fallow
1950
1955
1960
1965
1970
1975
1981
1987
1992
1997
2002
2007
2012
Ciparay
1950
1955
1960
1965
1970
1975
1981
1987
1992
1997
2002
2007
2012
Cumulative Income Difference
from Rice-Rice System (USD)
3000
Fig 5. Income difference between farmers that use and do not use SOI Phase
information