Watch video at http://environmentalresearchweb.org/cws/article/news/62412 

Peatlands are an important carbon sink. In Southeast Asia alone, which holds about an eighth of the global peatland carbon pool, these organic soils are believed to sequester some 68 million tonnes of carbon. It’s thought that draining peatlands — usually for agriculture such as growing oil palm — releases this sequestered carbon, as oxygen permeates the dry soil and triggers microbial activity and decomposition.

Measuring the carbon loss from soil is difficult, requiring collection of trace gases within a closed chamber, or tracking the soil’s subsidence or density. So scientists have been keen to find an alternative technique. One proposal is to monitor the water-table depth, which is as easy as inserting a plastic pipe into the soil and measuring the level the water comes up to.

Unfortunately, studies have failed to establish whether there is indeed a relationship between water-table depth and carbon loss. One study performed in 2013 at the International Center for Forestry Research in Indonesia and Peru found a linear relationship between the average water-table depth and the rate of carbon-dioxide emission in a Sumatran pulpwood plantation, yet no such relationship for individual (i.e. not averaged) measurements.

To make sense of the conflicting results, Kimberly Carlson of the University of Minnesota, US, together with colleagues at the Union of Concerned Scientists, also in the US, performed a meta-analysis of 12 studies that investigated the water-table depth and carbon loss in Southeast Asian peatlands. The researchers extracted the data from the studies and used them to create models relating the two parameters.

The models showed that for water-table depths ranging from 20 to 110 cm — figures typically maintained in plantations — there was indeed a positive, roughly linear relationship between the water-table depth and carbon loss. For a depth of 70cm — about average in the analysis — the annual carbon loss was roughly 20 tonnes per hectare.

"We were very surprised to see such a strong relationship," said Carlson. "One of the reasons that we were able to find a signal is that we only used studies that measured water tables and carbon losses over time periods of more than 10 months. Carbon fluxes can be highly variable…by averaging over many months to years, [any] high-frequency variation is minimized."

Carlson stressed that actual field measurements will still deviate from the model predictions because there are many other factors, such as soil fertility, rainfall and temperature, that affect the release of carbon. Nonetheless, she hopes that the predictions will help governments and companies estimate carbon loss in tropical peatlands more accurately, at least in Southeast Asia.

"A number of international companies that source products from Southeast Asian peatlands are committed to quantifying their climate impact," said Carlson. "Since they can now trace a product through the supply chain back to its source, this opens up possibilities for calculating greenhouse gas emissions from individual plantations, based on information such as plantation water-table depth."

The study is published in Environmental Research Letters (ERL) .