Redrawing the world's peat maps

Redrawing the world's peat maps

A new study published in Global Change Biology shows that tropical peatlands are much more extensive than previously thought. The study started by estimating total tropical and subtropical wetlands and the new estimate reaches 4.7 million km2. Several other studies have estimated wetland extent in the past and there is much disagreement between them. The higher spatial detail of this study and its multi-source approach (remote sensing climate data and geomorphology) suggests that the higher estimates in previous research may be correct. It also confirms that the American continent has the largest wetland extent (45%) and Brazil, with its Amazonian inter-fluvial region, contains the largest tropical wetland area with over 800,000 km2.

The most important innovation of the study was to develop an expert system model to map tropical and subtropical peatlands. An expert system is a computer system that mimics the decision-making ability of a human. The map is based on three biophysical indices that were based on three properties of wetland formation: 1.) Areas where long-term water supply exceeded atmospheric water demand (through evapotranspiration), 2.) The location of annually or seasonally water-logged soils, and 3.) Positions on landscapes where water accumulates because of the surrounding geology. The second innovation was that the geomorphological mapping was done at a very fine scale of 232 m. This allowed detection of smaller peatlands than previous efforts.

The model used to map peatlands was then cross-validated against 5 previous large scale mapping efforts and the map was compared with extensive field data from 275 georeferenced points in known peat areas. These comparisons with the point data allowed for the determination of bias in the model and correction of this bias to produce more accurate results, particularly regarding peat depth, which is hard to estimate from satellite data.

The new map suggests unprecedented peatland extents in the tropics and subtropics on the order of 1.7 million km2. This is more than three times as much peatland as previously reported. The volumes of peat exceed 7,000 km3, which is also a three-fold increase over recent best estimates.

The previous understanding was that Asia has the largest tropical peatland area and volume. This study upends that conclusion and suggests that South America hosts the most tropical peatland both by area and by volume (ca. 44% for both). This is partly related to a few large deep peat deposits that had not been mapped previously. However, the most important newly mapped peat areas are extensive shallow peat deposits in the Amazon Basin.

This new analysis also shows that Asia has extensive peatlands, and hosts 38% of both tropical peat area and volume, with Indonesia as the main regional contributor. Indonesia also still holds the deepest peat deposits in the tropics. Africa hosts more peat than previously reported but climatic and topographic contexts leave it as the least peat forming continent, despite of some massive peat deposits such as the Cuvette Centrale in the Congo Basin.

The new results suggest that there have been large biases in our current understanding of the distribution, area, and volumes of tropical peat and their continental contributions. Previous attention has focused on the peatlands in Southeast Asia, with estimates of the extensive areas of peatlands being in Indonesia and Malaysia. This new analysis shows that countries with under-reported peatland areas include Brazil, Peru, Venezuela, Colombia, Argentina, Colombia, China, India and Bangladesh.

Why is knowing the extent and location of peatlands important? Wetlands are crucial ecosystems for a host of reasons. They regulate water flows across landscapes, storing water when it is plentiful and slowly releasing it when things get dry. They harbor unique biodiversity, including charismatic species such as orangutans, tigers, and tapirs or less charismatic species like Mauritus flexuosa – a palm that goes by many names and is of particular importance to local economies and food supply in western Amazonia.

Peatlands are a particular kind of wetland that are characterized by the accumulation of partially decomposed plant material. These ecosystems have been storing massive amounts of carbon since the Holocene epoch, and its current release through draining and degradation is depleting this very old carbon store. These new results change significantly our understanding of the importance of peatlands in the global carbon cycle.

For example, previous estimates suggested that tropical peatlands stored around 90 billion tonnes of carbon; these new estimates put that number at around 350 billion tonnes. Previously, we thought that temperate and boreal peatlands held about twice as much carbon as tropical peats. There are still a lot of uncertainties about the amount of carbon stored in peat, but this new estimate suggests that the carbon stored in tropical is as large as that stored in boreal peats. This means that we need to completely reassess how we think about tropical peatlands.

Globally around 25 million hectares of peatlands have been drained and put into production – one-third of this is in the tropics. Drainage accelerates peat decomposition leading to large emissions of carbon dioxide and methane, both greenhouse gases. The gases emitted from these lands have the impact on the atmosphere equivalent to a billion tons of carbon dioxide every year.

At the moment, the only severely threatened tropical peatlands are those found in Southeast Asia, which partly explains the current biases in tropical peat data. The widespread use of fire leads to annual haze crises that place a burden on the healthcare systems of governments in Indonesia and neighboring countries and leads to loss of agricultural productivity. In the long term, the unsustainability of cultivating these lands will lead to economic hardships for small producers and large companies as the soils subside and water tables flood the fields. This is already happening in some regions of Sumatra.

The majority of the peatlands in Latin America and Africa are in remote areas that are not under immediate threat. However, development plans and continued climate change across the tropics could change this. Areas like the Sudd in South Sudan, Loreto in the Peruvian Amazon, and parts of Amazonia in Brazil could come under significant pressure in the coming decades.

Knowing where these peatlands are and how extensive they are should facilitate wise use and appropriate conservation of these areas so that the disastrous experiences of Southeast Asia are not repeated. By extending our understanding of the extent of the world’s peatlands, this research could also enable more rigorous assessments of ecosystem services, and the benefits and trade-offs involved in land-use change.

As noted above, many of the newly identified peatland areas are in remote areas, which by their isolated nature suffer from a lack of ground data. One of the next steps will be to validate many of these newly identified areas. Additionally, integrating this map with other mapping efforts aimed at monitoring the effects of deforestation and tropical land-use change on the atmosphere is needed to understand the vulnerability of these areas and identify priority areas for improving our data and establishing conservation and wise use plans. A better assessment of peatland areas may also help with understanding rising atmospheric methane concentrations, which appear to be the result of growing tropical sources.

One of the key shortcomings of this method is that the mapping effort has been parameterized for lowland conditions and fails to capture high altitude wetlands and peatlands. Given the importance of these ecosystems in water supply in many parts of the world, like the west coast of South America through its Andean montane peatlands, extending our understanding of the distribution of these peatlands is also important.


The research was carried out by scientists at the Center for International Forestry Research (CIFOR), Wageningen University and the International Center for Tropical Agriculture (CIAT) with the support of USAID and with technical input from the World Agroforestry Center (ICRAF). The map is freely available on the CIFOR website.