Fires and Smoke Haze in Indonesia

Prof. Rodney Keenan charts a pathway to solutions and identifies potential contributions by Australia

Fires and smoke haze in Indonesia

Fires in rainforest and peatland occur annually in Indonesia, with a peak under drier weather conditions in September-October. Conversion of forest to grassland and repeated burning have created a more fire-prone landscape. Fires are concentrated in Riau and Jambi provinces on the island of Sumatra and Central and West Kalimantan on Borneo. The smoke and haze affecting Singapore and other Asian cities is largely from the Sumatran fires. There were 120,000 active fires to early October 2015. Haze from these fires affects people in cities and rural areas across Indonesia and across South East Asia including Singapore, Malaysia and southern Thailand. Haze has been reported to have reached the Philippine city of Cebu, 2,700 km from Sumatra. Smoke and haze has implications for human health, transport and other activities, causing deaths in young children, school closures, grounding of airline flights and cancellation of outdoor events.

The causes

These are not, primarily, forest fires. Fires are deliberately set by large companies and by smallholder farmers on cleared forest areas. Large companies use fire to prepare land for oil palm, sugar or timber plantations. Smallholders clear forest and burn to prepare land for subsistence or cash crops. Clearing land using fire is cheaper than mechanical methods for large companies and often the only option for smallholders. Fire is also used as a ‘weapon’ in land tenure conflicts between companies and communities. Larger plantation development activities and transmigration schemes have significantly increased populations in remote areas and increased the incidence of fire.

Draining of peatlands with canals and drainage channels has dried out the landscape and increased the potential spread and intensity of fire. Once fires starts in dried peat they are very difficult to extinguish. Repeated fires year after year are common in degraded peatlands and forests. Haze is often the result of smouldering peat.

The extent of fires, and spread and duration of smoke haze, are typically worse in drier parts of the year, but especially during El Niño events and/or with a positive Indian Ocean Dipole. Nevertheless, recent research indicates that extensive fires and haze events are occurring on an annual basis. Wind and weather patterns determine if the smoke moves across major cities and to populated rural areas.

Fire is also used as part of the traditional land use practice of swidden agriculture. This involves forest clearing, burning and growing of crops on a rotational basis. This cause is small-scale, managed so it does not spread into surrounding forest, and the duration of burning is short. Burning to clear land is prohibited under Indonesian law with penalties including fines and prison terms. These laws have proven difficult to enforce due to uncertain responsibilities of different levels of government, ability to assemble evidence, and the capacity and resources of local agencies. Enforcement of the burning ban is seen as ‘political suicide’. Some companies are also willing to pay the fine and continue their burning practices. Uncertainty over land tenure coupled with conflicting claims to land contributes to enforcement problems.

The consequences

The fires of 1997-98 burned up to 12 million ha on Borneo and Sumatra and destroyed up to 6 million ha of forest. Haze from the most recent fires in 2015-16 drove air pollution indicators to record levels with mean particulate matter concentrations in excess of 200 μg/m3 in Sumatra which grossly exceeds the WHO’s 24-hr air quality target (50 μg/m3). The haze has also prompted legal action in Singapore against Indonesian companies and is affecting international relations, especially within ASEAN countries.

The fires themselves influence land values, infrastructure, and damage the capacity of forests to provide ecosystem services, such as flood protection, soil regulation, siltation protection, and biodiversity habitat. The 1999 fires are estimated to have cost the equivalent of over 4% of Singapore’s GDP in terms of human health morbidity and mortality costs.

It is estimated that 63% of Indonesian greenhouse gas emissions in 2005 were from deforestation, degradation and peatland destruction. Peat fires are also an important contributor to GHG emissions at a global level with the 1997-98 fires generating carbon emissions equivalent to 13-40% of annual fossil fuel emissions at that time.

In its recent INDC to the UNFCCC, Indonesia unconditionally committed to reduce greenhouse gas emissions by 29 percent against business-as-usual by 2030, and by 41 percent with international assistance and cooperation. Avoiding peat fires is, therefore, crucial for Indonesia to meet these mitigation targets.


Solutions to the haze are not simply about monitoring fires or ‘hand wringing’, but about helping Indonesia to deliver a comprehensive and long-term set of actions that change incentives so that those instigating the fires choose alternative methods of land-clearing. It requires a two-tier approach, at one level focussing on large companies and their suppliers, and the other at addressing local and landscape-scale issues, recognising traditional uses of fire, a focus on capacity building and setting clear-lines of responsibility.

How could Australia help?

An important solution pathway is to build on successful experiences with the implementation of forest conservation and management projects, for example the Nepal Australia Community Forestry Project that ran for over 20 years from the 1970s (noting the situation differs markedly to Indonesia). The success of this Nepal Project was built on supporting local initiative, landscape level engagement involving all stakeholders and long-term investment in local capacity building and technical support. Collectively, this approach facilitated a transformation in behaviour at the local level resulting in a significant increase in forest cover and new opportunities for communities.

A Forest Conservation and Management Program in Indonesia could provide incentives through performance-based payments for measures to sustainably manage and protect forest and rehabilitate peatlands, tied to indicators for fire extent and greenhouse gas emission reduction. Investment to support livelihoods needs to be clearly linked to activities that reduce the extent of fire and associated emissions.

If Green Climate Fund support were to assist in radically reducing haze over the next five years, it would: (1) provide very large payoffs for Indonesia, and its neighbours; (2) make an important contribution to global carbon dioxide mitigation, and (3) boost Australia’s credibility in the region and ‘green credentials’ globally.

Rodney Keenan is Professor of Ecosystem and Forest Sciences at the University of Melbourne.

Read more about the Fellows of the Economics & Science Group

Back to briefings