Carbon emissions of tropical peatland measured for the first time: Scientific proof of emission control can assist international agreement

Research Press Release | November 29, 2013

Press Release
Key Points – Measured carbon flux (flow of carbon dioxide: CO2) in tropical peatland for the first time.

– Found carbon flux to be positive (emitting carbon), even when carbon uptake by the forest is included. Carbon emissions increase with disturbance to the environment (forest clearing, drainage), and are greatest in El Niño years (dry years).

– Demonstrated that tropical peatland carbon management is possible by scaling up using satellite data, potentially enabling the application of domestic Indonesian and international carbon offset systems (Note 1).


An enormous amount of CO2 is emitted from tropical peatland due to forest fires and other causes, and there is significant concern that this is a substantial factor in global warming. For this reason, the ability to understand and manage carbon emissions from tropical peatland is a key issue.

Hokkaido University has conducted international joint research with Indonesian research institutes to address this challenge. In order to gain an understanding of the amount of carbon emissions from tropical peatland, measurements of carbon flux were taken for three distinct tropical peat forest ecosystems in Central Kalimantan in Indonesia—undrained forest, drained forest, and drained burnt forest. Data for a four year period has been gathered from these areas and analyzed. In addition, the researchers succeeded in taking continuous measurements of carbon emissions due to microbial decomposition in the drained burnt forest area.

In each of these cases, the amount of CO2 emissions was found to show a linear increase as ground water levels fell. These findings made it possible to estimate CO2 emissions using ground water levels map, enabling to develop algorithm of water table monitoring data and satellite data to be used to scale up the measurements to wide areas. Consequently, this provides a practical technique for carbon management through water table control of tropical peatland, opening the way to meeting the reference level criteria for carbon offset systems in Indonesia and international carbon offset systems under construction such as REDD-plus (Note 2).


Indonesia’s tropical peatland accounts for more than half the area of the world’s peatland in tropical areas, and has accumulated an enormous amount of carbon. However, rapid development beginning in the 1990s has been associated with lower ground water levels and dehydration of the peatland, and there have also been oxidation of peat by microorganisms and peat fires, turning Indonesia’s tropical peatland into a massive source of carbon emissions. Annual emissions are reported to be of a similar scale to annual emissions by the whole of Japan (Note 3).

At the G20 Summit in September 2009, The President of the Republic of Indonesia, Susilo Bambang Yudhoyono, pledged a 26% reduction in Indonesia’s greenhouse gas (GHG) emissions by 2020. In November 2011, President Yudhoyono issued a Presidential Regulation on the National Action Plan for Green House Gas Reduction (RAN-GRK), requiring each of 33 provinces to formulate a regional action plan for greenhouse gas reduction (RAD-GRK) by the end of 2012. Peat was expected to account for about 40% of the reductions, so there were keen expectations for the construction of an MRV system (Note 4) that could scientifically estimate the amount of carbon accumulated in peatland and CO2 flux from peatland.

The largest bottleneck in the development of such a system was the issue of how to quantify the carbon emissions from peatland due to peat fires and microbial decomposition. Measurement techniques commonly employed so far attempted to calculate the emissions from the amount of peat subsidence, but the scientific underpinning for the techniques was weak because it is difficult to separate three elements, decomposition, compaction and swelling/shrinking by water condition, and the Indonesian government made it clear that it was keen to find a measurement technique that is both accurate and has good cost-performance.

Making a quantitative assessment of the amount of carbon emitted from peatland became one of the major aims of this project, and a start was made on constructing an integrated peatland management system in order to control carbon emissions from tropical peat with the aim of contributing to the mitigation of global warming. The tropical peatland areas selected for the research were in Central Kalimantan province, focusing on the area of the Mega Rice Project (MRP), and on peatland outside the development area.

Research Method

In order to measure the amount of carbon captured from the atmosphere or emitted to the atmosphere (CO2 flux) by or from tropical peatland (including both the peat and the forest standing on top of the peat), researchers installed observation towers in three distinctive peatland ecosystems distinguished by their extent of disturbance: undrained forest (UF), drained forest (DF), and drained burnt forest (DB). Using the micrometeorological technique of eddy covariance (Note 5), the towers were employed to take continuous measurements of net ecosystem CO2 exchange (NEE), which indicates the amount of CO2 exchanged between the atmosphere and the ecosystem. The data collected was then analyzed.

The ground water level was measured as the level of the ground water surface relative to ground level, which was used as the reference (zero). For each of the UF, DF, and DB sites, PVC tubes were driven into the peat, and the depth of the water level inside the tube was measured directly.

In addition, for the DB areas, automated chambers (tubes with a diameter of approximately 20 cm) were placed directly onto the peat surface to measure CO2 flux without including flux due to root respiration. This was done to assess the amount of carbon emissions resulting from oxidative decomposition of the peat by microorganisms.

Research Results

Measurement of CO2 flux (NEE) for the three ecosystems began in 2004, and analysis was performed on four years of data, using the data collected up to 2008. This data represented an accumulation of continuous measurement data for the carbon flux between the ecosystem and the atmosphere for three ecosystems with different extents of environmental disturbance. Environmental data was also accumulated at the same time, including data on climate, soil, and ground water levels. Through analysis of the data it was possible to demonstrate the following three points for the first time.

1) Even undrained peatland forest (UF) has positive net carbon emissions.

2) Carbon emissions increase with the extent of disturbance of the environment (DB>DF>UF).

3) Carbon emissions are greater in El Niño years (Note 6).

Expressed as net ecosystem CO2 exchange (NEE, converted to carbon equivalent, with a positive value indicating a carbon source), the average carbon emission over four years, including an El Niño year, was 174 ± 203 grams (mean ± standard difference) per year per square meter for undisturbed forest (UF), 328 ± 204 grams for drained forest (DF), and 499 ± 72 grams for drained burnt forest (DB).

This body of data was systematically analyzed, and its relationship with the hydrological environment (particularly ground water levels) was modeled. Through this process, it was found that although parameters differed according to the extent of environmental disturbance (drainage and presence or absence of forest), the net positive carbon emissions (NEE) of the tropical peat ecosystem resulting from uptake of CO2 by photosynthesis and release of CO2 by respiration (respiration of the vegetation and oxidative decomposition of the peat) showed a comparatively simple linear relationship to variation in ground water levels (see figure below).

In addition, at DB sites, the researchers succeeded in measuring just the carbon flux due to microbial decomposition in the soil of the tropical peat. In earlier attempts, carbon flux measurements for the three different ecosystems had been affected by the netting of respiration and photosynthesis, preventing measurement of carbon flux for just the degradation of the peat. The success of the current research has the potential for application to direct carbon management of the tropical peat itself.

The CO2 emissions per year per square meter (carbon equivalent) were 382 ± grams in 2004-2005, and 362 ± 74 grams in 2005-2006, showing a simple linear relationship to the ground water levels measured at the same time. It was found that CO2 emissions per year per square meter (carbon equivalent) would increase by 89 grams for each 0.1 meter drop in ground water level.

 Prospects and Further Research

In addition to monitoring CO2 flux by means of the eddy covariance technique using the observation towers, the researchers are experimenting with two methods for quantifying carbon balance over a wider area.

1) Assessing carbon balance on a regional scale (Central Kalimantan province) using the Ibuki satellite of Japan’s GOSAT project for satellite observation of greenhouse gases

2) Developing a system for quantifying atmospheric CO2 concentration by measuring incident sunlight (when fires occur)

The research team has also developed a technique for estimating ground water levels by satellite-based remote sensing, and confirmed that this enables the findings described above to be scaled up to cover a wider area of surrounding tropical peatland.

Having gained an accurate understanding of carbon flux by means of these methods, the knowledge can be combined with proposals based on the findings of other research—to prevent peat fires, to ensure reforestation after fires, and to maintain ground water levels by constructing dams. This makes it possible to assess the carbon emissions reduced by such initiatives, thereby enabling carbon management in tropical peatland.

When it becomes possible to accurately monitor the reduction of CO2 emissions from tropical peatland, there is potential for such reductions to be used in schemes based on REDD-plus and other advanced ecosystem-based carbon sequestration frameworks being considered under the Framework Convention on Climate Change (UNFCCC), or based on collaboration with Japan in the joint credit mechanism (JCM).

Project Outline

This research project was conducted as part of the SATREPS (Science and Technology Research Partnership for Sustainable Development) program run by JST and JICA. The research was performed by Hokkaido University and the National Standardization Agency of Indonesia (BSN) in collaboration with other institutions.

– Project title: Wild Fire and Carbon Management in Peat-forest in Indonesia

– Principal Investigator: Prof. Mitsuru Osaki (Graduate School of Agriculture, Hokkaido University)

– Collaborators in Japan: Ehime University, The University of Tokyo, Japan Aerospace Exploration Agency (JAXA), Japan Space Systems, Mitsubishi Research Institute, Hokkaido Institute of Hydrology and Climatology (HIHC), etc.

– Collaborators in Indonesia: National Standardization Agency of Indonesia (BSN), University of PalangkaRaya (UNPAR), Indonesian Institute of Sciences (LIPI), State Ministry of Research and Technology (RISTEK), Indonesian National Institute of Aeronautics and Space (LAPAN), Forest Research and Development Agency (FORDA), Agency for the Assessment and Application of Technology (BPPT)

– Research period: October 2008-March 2014

Inquiries Concerning the research:

Takashi Hirano, Professor, Research Faculty of Agriculture, Hokkaido University


Web: http://www.

Concerning the SATREPS program (JST):

Research Partnership for Sustainable Development Division

Japan Science and Technology Agency


Concerning the SATREPS program (JICA) (Note 7):

Office of Media and Public Relations

Japan International Cooperation Agency




Global Change Biology (2012.8.27)

Global Change Biology (2013.10.10)