The sweet scent of elephant apples hangs heavy in the still air of a tropical forest. Asian elephants have a particular fondness for these large green fruits whose big seeds the pachyderms help to disperse. Many tropical forest trees depend on large fruit-eating creatures such as elephants, tapirs, monkeys and hornbills for dispersing their sizeable seeds. Due to hunting and forest degradation, large wildlife in tropical forests across the world are in decline. Tree species dependent on these animals for seed dispersal also tend to decline, getting replaced over time by species whose seeds are dispersed by other modes. Such losses of large-seeded trees could substantially change carbon storage in tropical forests around the world, thereby altering their ability to regulate our world's climate.

Plants convert atmospheric carbon-dioxide to carbon stored in their living tissues through the process of photosynthesis. Trees in tropical forests form a large terrestrial 'carbon sink', which play a crucial role in regulating atmospheric carbon-dioxide concentrations. However, when it comes to storing carbon, not all tree species are the same. In a recent study in the journal Nature Communications, researchers find that large-seeded tree species which depend on big animals for seed dispersal, grow to greater sizes as adults and thus have higher carbon storage potential than species with smaller seeds in tropical forests worldwide. Losses of large seed dispersers can therefore reduce carbon storage by the Earth's tropical forests by decreasing the volume of vegetation biomass in these forests.

"Scientists are only just beginning to understand the numerous ways in which animals affect the carbon cycle of tropical forests, and the consequences of declines of these animals - also termed 'defaunation' - for terrestrial carbon storage ", says Anand M Osuri, a member of Mahesh Sankaran's group at NCBS and the study's lead author. "Although defaunation is a problem affecting tropical forests the world over, our understanding of its consequences for carbon storage relies heavily on patterns seen in one part of the tropics - the forests of South America", he adds.

How do losses of animals that disperse large seeds affect the carbon storage potential of tropical forests in different parts of the world? In order to address this question, scientists from the National Centre for Biological Sciences (NCBS), Nature Conservation Foundation (NCF), University of Leeds and 12 other international academic and conservation organisations examined how tree species that are dispersed by large animals differ from tree species dispersed by other modes (for example, wind or gravity) in their ability to store carbon. The researchers used a dataset of over 26000 trees belonging to 2500 species from tropical forests in the Americas, Africa, Asia and Australia to examine the effects of defaunation on aboveground carbon storage of tropical forests globally. These data were obtained from open-access datasets published online by Tropical Ecology Assessment and Monitoring (TEAM) network, Commonwealth Scientific and Industrial Research Organisation (CSIRO) and Institut Français de Pondichéry (IFP).

Adopting a simulation-based approach, the study created various 'defaunation scenarios' - different conditions in which trees dependent on large animals for seed dispersal gradually declined, and were replaced by trees with other modes of dispersal such as small animals, wind or gravity. Defaunation-driven changes in carbon storage were assessed using information on tree sizes and wood densities in the original and defaunated forest stands. This allowed the scientists to examine how shifts in tree community composition in the defaunated forests could translate to changes in carbon storage.

The results suggest that in Africa, the Americas and South Asia, where the majority of tree species are dispersed by animals, losses of large seed dispersers can reduce carbon storage by decreasing the volume of vegetation biomass. In some regions, especially the Americas, carbon losses may be heightened further due to declines of tree species with high wood density following losses of large seed dispersers. If 50% all trees dispersed by large animals were replaced over time by trees with other modes of seed dispersal, carbon storage in these forests would be reduced by 2%. This is roughly equivalent to 14 years' worth of Amazonian deforestation.

In contrast to forests in Africa, South Asia and the New World, carbon storage may be less sensitive to losses of large seed dispersers in the forests of Southeast Asia, where a number of big tree species depend on wind and gravity rather than animals for seed dispersal.

"Insights into these regional differences, which arise due to the unique evolutionary histories of trees in each region, are among the most interesting contributions of this study", says Jayashree Ratnam, one of the co-authors of this publication. "Even while tropical forests across the globe seem to look alike, they are functionally different - and we are reminded once again that a one-size-fits-all model for a biome may often hide critically important differences in form and function", she adds.

Apart from these predictions, the study also highlights an urgent need for more research on understanding the complex dynamics of defaunated forests.

"A number of factors contribute to determining which tree species will ultimately succeed in defaunated forests. Many of these factors are not understood very well at present", says Sankaran. "Long term observations and experiments are essential to predict the exact magnitudes of, and time scales over which, community-level responses to defaunation may play out", he adds.

At present, policies such as REDD+ (Reducing Emissions from Deforestation and Forest Degradation) that are aimed at mitigating terrestrial carbon emissions primarily focus on reducing carbon losses by protecting tropical forests from deforestation and logging. This study shows that in addition to protecting the forests themselves, policies that give importance to conserving their fauna could also result in significant benefits for carbon storage in tropical forests.

About the paper:

The work described in this article has been published in a paper titled "Contrasting effects of defaunation on aboveground carbon storage across the global tropics" in April 2016 in the journal Nature Communications and can be accessed here: http://www.nature.com/ncomms/2016/160425/ncomms11351/full/ncomms11351.html