"Our work is the first to explore the lag between land allocation and oil palm development," Kim Carlson of Yale University, Woods Institute for the Environment, and Stanford University, US, told environmentalresearchweb. "By compiling governmental lease records, we show that oil palm plantation extent in 2012 is only the 'tip of the Bornean bog'."

According to Carlson, roughly 80% of oil palm leases in the study region have yet to be cleared and planted; the majority of these leases are located on carbon-rich peatlands. "Thus, while oil palm has been a relatively minor direct contributor to deforestation and carbon emissions through 2008, future oil palm expansion will become a significant source of regional carbon emissions if all allocated leases are developed," she said.

Between 1989 and 2008, fire was responsible for 93% of deforestation. But by 2007–2008 oil palm directly caused 27% of total deforestation and 40% of peatland deforestation in the study area.

Carlson says that draining of peatlands for oil palm agriculture generates 'committed' long-term carbon emissions. "Protecting peatlands and forests from all forms of land-cover change, including oil palm plantation expansion but also wildfires, especially during El Niño-related droughts, is critical to reduce regional carbon emissions," she said.

Carlson, Lisa Curran and colleagues found that more than half of the new oil palm from 2001–2008 was planted in converted agroforests and agricultural fallows. Such community-managed lands have not been included in previous examinations of sources for agricultural expansion in Indonesia. Carlson believes that their conversion "may have unexpected implications for resident rural farmers".

What's more, by modelling future scenarios the team concluded that carbon conservation does not automatically generate benefits for local communities. Their results suggested that while a moratorium on oil palm expansion into forests and peatlands, combined with effective protection of remaining forests, can reduce regional carbon emissions below projected business-as-usual levels, around 30% of community landholdings will still be converted to oil palm.

"Focusing on carbon alone may actually redirect oil palm plantations onto community lands because shifting agricultural mosaics tend to have lower carbon stocks than peatlands or intact forests," said Carlson. "Palm oil sustainability assessments should include not only quantification of land sources for and carbon emissions from plantations, but must also consider the means of land acquisition from local land users. For example, local landholders should be equitably compensated for their landholdings."

Since little information on land cover, demography or administrative boundaries was available for West Kalimantan in the early 2000s, the researchers conducted extensive field surveys. "We chose to work at high spatial and temporal resolution but in a relatively small region so that we could develop a truly robust research project," explained Carlson.

The team also used satellite data. "Because the major causes of land-cover change, such as fires and logging, in Kalimantan occur on an annual basis but can 'disappear' to satellite detection as vegetation regrows, compiling an almost-annual time series of satellite imagery was critical," said Carlson. "We went as far back in time as possible given the limitation of Landsat satellite imagery (1989) because current carbon stocks are dependent on past land-use transitions. With this satellite data, a key breakthrough was the ability to discriminate shifting agricultural or 'swidden' lands – crucial for local agrarian livelihoods – from other land cover types such as logged forests or oil palm plantations."

Now the researchers plan to apply what they learned in Ketapang to the wider Kalimantan region. "Our Indonesian collaborator, Living Landscapes Indonesia, will assist by collecting province- and district-level data we'll need to ensure our work remains realistic and locally informed," said Carlson. "We also hope to explore how indirect land-use change from oil palm plantations – such as displaced smallholder agriculture or escaping fires – contributes to regional deforestation and carbon emissions."

The team, who reported the study in PNAS, would also like to assess how oil palm development affects freshwater stream ecosystems.