Digging Deeper: unexpected sensitivity of deep soil carbon to changing fire frequencies

Author Adam Pellegrini discusses the findings of his new Journal of Ecology article – “Frequent burning causes large losses of carbon from deep soil layers in a temperate savanna.” Find out more about how fire frequency can impact carbon storage and root biomass in deep soil layers.


Fires are dramatically changing in frequency and intensity, due to shifting climate and altered land use across the globe. Changes in fire can have dramatic effects, for instance altering plant cover and animal populations. So when we think about fire effects on ecosystems, we often are drawn to the striking changes aboveground. However, changes aboveground are only part of the story. Fire also causes many transformations belowground. However, the difficulty in measuring belowground processes and the slow processes influencing soils has limited our understanding of belowground fire effects.

Aerial overview of three plots of the fire manipulation experiment. The large middle area (below the wetland) is a frequently burned plot and the areas immediately to its left and right are infrequently burned plots. Photo: Jacob Miller. (Creative Commons CC BY-SA 4.0)

It has been repeatedly demonstrated that fire can cause losses of soil carbon, through direct combustion and subsequent reduction in plant biomass. However, those two mechanisms lead to a shift in soil carbon in the upper soil layers. Consequently, most studies that evaluate the effect of fire on soils only measure the topsoil, working under the assumption that changes in the deeper soil layers are inconsequential.

In our study, we dug deeper to better understand how fire changes soil throughout a one meter profile. Our study site was in a temperate oak savanna in the Midwest region of the USA at a Long Term Ecological Research Network site, Cedar Creek Ecosystem Science Reserve. Cedar Creek has an amazing burning experiment that started in 1964, which has dramatically transformed the ecosystem. This experiment has shown that a single factor (fire frequency) can transform this system from forest to savanna to grassland, in only a few decades. The fire treatments range from complete exclusion to frequent fire (burned 3 out of every 4 years), with several other treatments in between. Across the fire frequency gradient, we found that the change in deep soil (>20 cm) carbon roughly doubled the quantity of soil carbon lost due to frequent burning.

Prescribed burn on a plot that has been frequently burned since the 1960s. Photo: Susan Barrott (Creative Commons CC BY-SA 4.0).

A survey of plant root biomass revealed that there tended to be an increase in fine roots in the topsoil with more frequent burning, likely due to the colonization of grasses. However, there was a large decline in root biomass in deep soil. This suggests that changes in inputs within the deep layers are behind shifts in carbon storage. Analyses of carbon isotopes revealed that the relative balance between herbaceous and woody vegetation inputs to deep soil carbon did not change, contrasting with the large shift in their relative contributions in the topsoil. These two lines of evidence demonstrate the importance of considering fire-driven changes in plant biomass allocation to roots, and how this allocation changes with depth.

Our findings also have implications for conservation. We found that the transition from a forest to a woody savanna resulted in relatively little change in soil carbon. However, soil carbon declined with more frequent burning, causing the landscape to lose more plant biomass. This result suggests that savannas can be maintained with periodic burning without losing much carbon belowground — demonstrating that the conservation of savanna ecosystems need not be at odds with carbon storage.

An open oak savanna in a high fire frequency plot. Photo: Jacob Miller. (Creative Commons CC BY-SA 4.0)

Like many ecological studies, our results raise more questions. It is difficult to reconcile the contrasting sensitivity of deep soil carbon to fire across ecosystems — while this oak savanna was very sensitive other systems, such as certain tropical savannas, are not.  Moreover, although our findings suggest shifts in root biomass inputs are behind changes in deep soil carbon, this mechanism still needs to be directly tested.


Adam Pellegrini, Department of Plant Sciences, University of Cambridge, UK


Read the full research article in Journal of Ecology: Frequent burning causes large losses of carbon from deep soil layers in a temperate savanna

Leave a comment