Indonesia Country Case Study:
Impacts and Responses to the 1997-98 El Niño Event
|
Team Leader: |
 |
|
Team Members: |
|
Note: The Indonesia Country Case Study was supported by the US Office of Foreign Disaster Assistance and Office of Global Programs, National Oceanic and Atmospheric Administration as a contribution to the UNEP/NCAR/WMO/UNU/ISDR study for the UN Foundation.
Executive Summary
Introduction
The 1997-98 El Niño had significant social and economic implications for Indonesia. A large part of the country suffered from severe drought, resulting in a huge shortfall in rice production that necessitated the import of over five million metric tons of rice to ensure food availability to the economically weaker sections of the society. In the forestry sector, the effects of large-scale forest fires during 1997-98 were unprecedented, damaging more than 9.7 million ha of forest area.1 The smoke and transboundary haze from these fires affected not only Indonesia but also other Southeast Asian countries, in particular Brunei Darussalam, Malaysia and Singapore. In addition to impacts on the agriculture and forestry sectors, the 1997-98 drought and fires also significantly affected other sectors such as transportation, tourism and public health.
Given that the linkage between El Niño events and drought in Indonesia has been well established scientifically, it is important to analyze why this scientific understanding did not translate into effective countermeasures essential to mitigate the worst effects of El Niño 1997-98. This report presents an analysis of the impacts of the 1997-98 El Niño event on Indonesia and identifies lessons learned that will be helpful in dealing with future extreme climate events.
Setting
Indonesia is the largest archipelago in the world, inhabited by nearly 210 million people. Over the past three decades, the country's economy has moved from being an agriculture-based to an industrialized economy. Agriculture's share of the country's GDP gradually decreased from 47 percent in 1969 to 33 percent in 1978, 21 percent in 1988, and was around 16 percent in 1998. However, with half of the country's workforce directly dependent upon agriculture for their livelihoods, it remains an important sector of the economy. Forests and estate crops and related activities also constitute an important sector of the Indonesian economy, both in terms of the percentage share of the GDP as well as in providing employment. By 1996, some 445 logging concessions were operating on 54 million ha of forestland, of which close to 1 million ha were reported to be logged annually. In 1994, wood and wood products produced about US$5.5 billion in export revenue for Indonesia, representing about 15% of total foreign earnings and employing 700,000 people.2 Both the agriculture and forestry sectors seem to be highly susceptible to the adverse impacts of climate variability.
Indonesian Climate and Its Sensitivity to ENSO
The climate of Indonesia is generally characterized by two seasons: dry (April to September) and wet (October to March). Based on the number of dry and wet months, the country is divided into 14 agroclimatic zones, of which 11 are considered as sensitive to extreme climate variability since rain fluctuations can upset established cropping patterns.
From 1877 to 1997, 93 percent of the drought years have been linked to El Niño events. Several studies show a clear positive correlation between normalized Indonesian rainfall anomalies and the Southern Oscillation Index (SOI). While in El Niño years the onset of the monsoon is later than normal, during La Niña years the onset is earlier in most areas. Therefore, an El Niño event causes delayed planting and consequently a delayed and reduced harvest. A La Niña year offers the possibility of advancing the planting season with an early increased harvest, as well as the possibility of planting an additional crop.
Although ENSO influences the climate of the entire Indonesian archipelago, it is important to note that there are discernible regional and seasonal differences. Being in the monsoonal areas, the south and southeast regions of Indonesia (comprising South Sumatra, South Kalimantan, Java, Bali and Nusa Tengara) are relatively more sensitive to ENSO's extremes. In these regions, the dry season and transition to wet season are highly influenced by the ENSO cycle. It will be important to take these regional and seasonal sensitivities into account when developing focused plans to deal with ENSO-associated consequences.
The 1997-98 El Niño Event
Rainfall anomalies. The extent of the drought in Indonesia in 1997-98 is shown by the 1997-98 annual (March-February) rainfall percentiles calculated from 33 stations.3 Most parts of the country suffered from reduced rainfall, with 13 of the 33 stations having the lowest rainfall on record (on the 0 percentile). The only areas approaching the 50th percentile (near normal) were North Sumatra and Biak. El Niño delayed the onset of rainfall and resulted in frequent dry spells during the 1997 wet season. Both these factors caused delayed planting of wet season paddy, as enough accumulated rainfall was available only in December 1997. In Semarang, Central Java, which is one of the representative stations in the Java rice growing belt, the onset of wet season rains did not occur until 15 November (nearly a month later than normal), and the season was marked by several dry spells.
The March-May 1997 rainfall was close to or even above normal across much of the country, but dry conditions prevailing across southern and eastern parts suggested an early retreat of the monsoon. By June-August 1997, the areas with substantial rainfall deficits had spread to western regions except for North Sumatra. The drought reached its peak in September-November 1997, with all parts of the country except Sumatra having extremely low rainfall, and deficits of 400-500 mm being common. By the December 1997-February 1998 quarter, rainfall was generally close to normal except for pockets in East Kalimantan and Sulawesi, precisely the areas where large-scale forest fires occurred in early 1998.
Impacts on agriculture and food security. Rice production in Indonesia is heavily influenced by the monsoon rain patterns, which have an important bearing on agricultural performance during the main (wet) and secondary (dry) seasons. The wet season normally extends from October to March and produces 60 percent of the country's annual rice crop and half of its maize, soybean and groundnuts. The dry season covers April to September, during which the remaining annual crops are produced.
The rainfall anomalies during the wet season 1997-98 caused a decrease in area under rice cultivation by 380,000 ha (3.4% below the previous wet season). Farmers planted maize as a compensatory crop in areas where paddy could not be planted. The switching over to maize was to the extent of 266,000 ha more than the area normally cropped with maize (an 8% increase from the previous wet season). The reduced rice production, coinciding with the economic crisis which began in 1997, led to a 300 percent increase in the price of rice. The government of Indonesia imported over 5 million metric tons of rice in order to maintain price levels and to ensure the availability of food to the economically weaker sections of the population.
In addition to the reduction in the main-season rice crop, secondary-season production was affected by the following factors:
- The delayed harvesting of the main wet-season crop during 1997-98, due to the delayed onset of rains, caused a subsequent delay in the planting of the 1998 dry-season crop by one to two months, with significant production loss.
- Grasshopper infestation, which normally affects tree crops, seriously damaged 1998 dry season field crops, particularly in Lampung province, probably because of the destruction of natural grasshopper habitats by serious forest fires in Kalimantan and Sumatra. The problems of pest attack were exacerbated by ineffective control measures, as most farmers could not afford the substantially higher cost of imported chemical pesticides, the price of which was estimated to have increased five-fold over that of 1997.
- In addition to ineffective pest control during the dry season, it is estimated that fertilizer application rates were considerably lower and unbalanced, which not only reduced the physiological ability of plants to resist pest attacks but also affected grain formation. Research studies indicate that the potential yield loss due to this factor could be as high as 30 percent.
Impacts on the forestry sector. Probably the most disastrous event during the 1997-1998 El Niño, one that caught international attention, was the widespread occurrence of forest fires with associated smoke and transboundary haze. The fires were among the most severe in the previous two decades and had a significant socio-economic impact. A study commissioned by the Asian Development Bank (ADB) and BAPPENAS (1999) estimated the economic cost of the 1997-98 fires and drought to be in excess of US$9 billion. The 1997-98 fires and the resulting smoke and transboundary haze became a matter of international concern. Besides Indonesia, a number of Southeast Asian countries, in particular Brunei Darussalam, Malaysia and Singapore, were badly affected. The Philippines and Thailand also suffered though to a lesser degree. The severity and extent of the smoke haze pollution was unprecedented, affecting the health of millions of people across the region. International support had to be mobilized to suppress the fires. The experience underscored the enormity of the problem, leading the Association of Southeast Asian Nations (ASEAN) Environment Ministers to adopt a Regional Haze Action Plan (RHAP) setting out cooperative measures to combat the perils brought by forest and land fires. The scope of the pollution resulting from the fires, especially those from peat soils and cleared conversion forest, shows that the impact was an environmental problem of global dimensions. In 1997-98, the forest and land fires in Indonesia contributed 22 percent of the world's carbon dioxide production. Over 700 million metric tons of carbon dioxide were released into the atmosphere, elevating Indonesia to being one of the largest carbon polluters in the world in that year.
Responses to the 1997-98 El Niño event. Forecasts indicating the possible onset of an El Niño event were available to the Indonesian Bureau of Meteorology and Geophysics (BMG) as early as late 1996. BMG incorporated this information into a dry-season seasonal climate forecast for the entire country, which was issued to all relevant user departments at national and provincial levels in March 1997. The forecast information was communicated through the existing information network which is utilized for routine administrative functions. No urgency was attached to the timely flow of information from national to provincial to district to sub-district levels of various user departments.
In the agriculture sector, no major interventions were undertaken to manage water resources, plan appropriate agricultural inputs or minimize crop losses. In the forestry sector, following the warning of possible El Niño-induced drought, the Ministers of Environment and Forestry, and a number of provincial governors, called on everyone to be alert and to take action to prevent forest and land fires. However, these warnings were not followed up and fires began to occur in early 1997. There was little evidence of a substantial institutional fire prevention and preparedness program in place. The institutional structure to respond to early warnings and provide information and guidance to field operators about fire forecasts was highly inadequate. There were almost no procedures in place to eliminate or minimize the use of fire during dangerous periods.
Lessons Learned
The following lessons can be learned from the experience of the 1997-98 El Niño in Indonesia:
Downscaling Climate Forecasts
- Even though the teleconnections between local climate and El Niño events are considered strong and reliable for Indonesia, a global ENSO forecast is not directly usable on the ground. The global climate parameters need to be translated into relevant local weather variables (e.g., for rice cultivation, the onset and duration of rainfall in wet season, and the number of dry spells in wet season) and related outlooks for various sectors, geographic locations and seasons.
- The country's past meteorological data need to be retrieved and its observational network strengthened to support a national climate research agenda to fully understand the implications of the ENSO phenomenon for the local climate in different locations of the country.
Making Climate Forecasts Actionable
- An interdisciplinary, multi-institutional climate forecast applications research agenda needs to be developed. BMG and other climate research centers need to establish closer working relationships with a range of intermediaries (such as agricultural, forestry and irrigation research centers), and existing and potential user organizations.
- Under the climate forecast applications research agenda, there is a need for national researchers to review the impacts of and responses to previous El Niño events in order to identify those impacts that can be attributed to an El Niño event. Based on such studies, El Niño risk maps should be developed for various sectors, regions and seasons.
- Based on applications research, forecast products should be tailored to meet the needs of specific user groups. For example, for rice cultivation in a given area, instead of the usual terciles, appropriate forecast parameters may include the onset of wet season rains, the probability of accumulated (threshold) rainfall of 75 to 100 mm over a period of three weeks, the number of dry spells, and the termination of wet season rains. Thorough research may indicate that the SOI has high correlation with these parameters in some areas and relatively low correlation in others. Such inter-disciplinary research is imperative for developing robust decision-making tools at national and local levels.
- Although most critical resource sectors deal with different kinds of risks (with probabilities implicit in them) in their normal-time operations, climate forecasts stated in probabilistic terms are difficult to understand. Greater dialogue among climate forecasters, intermediaries and users will help translate probabilities into concepts more familiar to different sectors. On the one hand, this will help in making forecasts more actionable, and on the other, it will assist in making the expectations of users and intermediaries from the climate science community more realistic.
- There is a need to create a higher level of awareness of the ENSO phenomenon and its impacts among the public and policy makers. ENSO extremes are here to stay and must be viewed by planners not only as important, but also as important as the natural flow of the seasons.
- It is also important, when it comes to potential disasters, that various departments and agencies share information about potential impacts as well as potential preventive, mitigative and adaptive strategies and tactics.
Setting Up Response Systems
- Based on past experiences and climate forecast applications research, it is necessary to outline specific prevention, preparedness and response strategies that clearly identify institutional responsibilities and implementation inter-relationships.
- Communication should be strengthened between different agencies at each level (horizontal), and within each agency from national to local levels (vertical). In times of El Niño-related crises, emphasis should be placed on the speedy flow of information between and within different agencies.
- The 1997-98 fires exposed inadequacies in the forest fire management system in Indonesia. Since then, a number of national, regional and international initiatives have been underway to review and strengthen the Indonesian forest fire management structure. Wide-ranging improvements have been recommended and implemented in regulatory and legislative frameworks as well. The next El Niño event will test the implementation effectiveness of these measures.
- Given scarce resources, it may be prudent for regional governments to undertake benefit-cost assessments in order to determine the most cost-effective responses to El Niño impacts on the environment and societies.
- Although in the long run, effective management of the consequences of an El Niño event would require more effective management systems in critical resource sectors even in normal times, in the short term, special institutional arrangements may be required to deal with El Niño events. Such measures may include the establishment of high level task forces assisted by professional working groups at national and provincial levels.
- There is a need for an explicit commitment of resources to support El Niño management programs.
-
Asian Development Bank (ADB) and National Development Planning Agency (APPENAS), 1999: Causes, Extent, Impact and Cost of 1997-98 Fires and Drought. Jakarta: ADB.
[ Return to Executive Summary ] -
Sunderlin, W.D. and I.A.P. Resosudarmo, 1996: Rates and Causes of Deforestation in Indonesia: Toward a Resolution of the Ambiguities. Occasional Paper No. 9. Bogor: Center for International Forestry Research.
[ Return to Executive Summary ] -
Personal communication with D. Kirono, Ph.D. student at Monash University, Australia, working on Indonesian rainfall variability.
[ Return to Executive Summary ]
