Indonesian history · Fire and carbon research

1997–2002

How Researchers Estimated Carbon from Indonesia’s 1997 Fires

Follow the 1997 Central Kalimantan peat fire study from Landsat images to burn-depth measurements, and see how the authors estimated a national range.

Satellite mapping identified affected area while ground measurements and carbon assumptions gave an estimate of mass.
1997
Fire season studied
2.5 million hectares
Central Kalimantan study area
0.51 metres
Mean estimated burn depth used in the study

The short answer

A satellite can show where vegetation changed after fire. In peatlands, carbon may also be released when the soil itself burns. The study of Indonesia’s 1997 fires combined satellite mapping with ground measurements of how much peat was lost. It then explored how those local observations might scale to the country.

Reading the records

How an area map became a carbon estimate

Area, depth and carbon content together determine how much peat carbon the model assigns to a fire.

How an area map became a carbon estimate.
StepEvidenceRole
AreaSatellite comparisonMaps the part of the study region that burned.
DepthGround measurementsEstimates the thickness of peat removed.
CarbonDensity and carbon assumptionsConverts the peat volume into a carbon mass.
ExtrapolationNational burned-area scenariosProduces a range across assumptions.

Chronology

A timeline of the turning points

  1. Fires during El Niño conditions

    Peat and surface vegetation burned in the study region.

  2. Burned areas mapped

    Landsat scenes allowed researchers to assess change.

  3. Carbon estimates published

    The paper reported a local estimate and a much wider extrapolated national range.

Start with the study area

Page and colleagues studied about 2.5 million hectares in Central Kalimantan. They compared imagery acquired before and after the fires to estimate what burned. They then used measured burn depth to estimate how much peat was removed. Some ground outside the region could behave differently, so the study area and the national estimate should retain their own labels.

Sources for this section: [1]

Surface vegetation and soil both matter

The authors estimated that 0.19–0.23 gigatonnes of carbon came from peat combustion within the study area. They estimated a further contribution from vegetation. The quantities are expressed as mass of carbon. They should not be copied as tonnes of carbon dioxide or as an estimate of the smoke people inhaled. Other studies examined fire damage or uncertainty using different combinations of field and satellite evidence.

Sources for this section: [1] · [2] · [3]

The range reflects a step beyond direct observations

The paper’s 0.81–2.57 gigatonne national range includes peat and surface vegetation and depends on alternative assumptions for burned land and peat loss. The broader range reports an extrapolation, not additional ground measurements across all of Indonesia. Compare scenarios with their assumptions intact when using this study alongside a later emissions estimate.

Sources for this section: [1] · [3]

Continue the chronology

Connected biographies

People in this history

Quick answers

Common questions

Why did the study need field measurements?
Satellite mapping provided an area. Burn depth was needed to estimate how much peat that area lost.
What does 0.81–2.57 gigatonnes refer to?
It is the authors’ extrapolated national estimate of carbon from burning peat and vegetation in 1997.

Primary and institutional sources

How this history was checked

Dates, legal milestones and historical claims are checked against the official and institutional records below. This article was reviewed on 19 November 2026.

  1. NaturePage and colleagues’ 2002 study of carbon released in Indonesia’s 1997 fires
  2. NatureSiegert and colleagues’ field and satellite work on drought and fire damage
  3. Global Biogeochemical CyclesStudy of peat combustion and estimation uncertainty