The subject is addressed in a review by the University of Zurich’s  (left) MichaelSchmidt,  (right) Susan Trumbore from Max Planck Institute of Biogeochemistry, Jena and an international team of scientists published in Nature.

The researchers suggest ways to improve the ability to predict how soil carbon responds to climate change as well as land use and vegetation change. 

For years, scientists thought  organic matter persists in soil because some of it forms very complex molecular structures that were too difficult for organisms to break down. In the Nature review, however, Schmidt and colleagues point out how recent advances, from imaging the molecules in soils to experiments that track decomposition of specific compounds, show this view to be mistaken.

The major forms of organic matter in soils are in the forms of simple biomolecules, rather than large macromolecules.  Charred residues from fire provide a possible exception, but even these have been shown to decompose.  

If molecular structure is not causing organic molecules to persist, what is?

The team contends that the average time carbon resides in soil is a property of the interactions between organic matter and the surrounding soil ecosystem.  Factors like physical isolation, recycling, or protection of molecules by minerals or physical structures like aggregates, even unfavorable local temperature or moisture conditions, can all play a role in reducing the probability that a given molecule will decompose. 

Soils are teeming with bacteria (approximately 40m cells in a gram of soil), but typically occupy less than 1% of the available volume, and usually cluster in ‘hot spots’.  In some situations where microbial populations are sparse, for example in deep soils or far from roots, it may just require a long time for suitable conditions to arise that allow a molecule to be broken down.  In other locations, freezing temperatures may inhibit microbial action.

Rework  of assumptions needed
Currently, models used to predict how global soil carbon will respond to climate change include little mechanistic understanding and instead use simple factors like temperature dependence that indicate acceleration of decomposition in a warmer world.  This assumes t temperature is the major limitation to decomposition, where other factors may dominate.  

The decomposition-warming feedback predicts large soil carbon losses and an amplification of global warming, But the authors argue this approach is too simplistic. They make several suggestions where current improvements in understanding could be built into models, improving our ability to predict how soil carbon responds not only to climate but to land use but vegetation change.  
 
New experiments and models needed
The findings need to be used for new experiments and models Professor Schmidt explains. In doing so, it is not only the first few centimeters of the soil that should be examined, as has been the case up to now, but rather the full top two to three meters. In the Nature article, the researchers make various suggestions as to how the models for forecasting the response of soils to changes in the climate, vegetation and land use might be improved.

The new results also cast a critical light on bioengineering experiments with plants containing high amounts of lignin or plant charcoal (biochar), with which more carbon is prepsumed to be stored in the soil in the long run.