“Scientists have recently expressed interest in developing ‘early warning systems’ for desertification based on patterns observed in satellite imagery,” Corina Tarnita of Princeton University, US, told environmentalresearchweb. “The problem is that multiple mechanisms can produce very similar-looking patterns – one mechanism might suggest that an ecosystem is on the brink of collapse, whereas another might instead reflect that termites or other ecosystem engineers are helping the ecosystem persist.”

Tarnita reckons that until we find a way to distinguish between these different mechanisms in satellite imagery, there will be a need for researchers on the ground to investigate the processes responsible for producing the patterns.

Plants tend to help their neighbours by making it easier for water to infiltrate soil but compete with plants further away for water. As rainfall reduces, the amount of plant biomass drops and gaps form in the vegetation cover, followed by labyrinths, then spots of vegetation and finally, in what’s known as a “catastrophic shift”, barren desert.

Vegetation also clusters in patches at the site of termite mounds in the savannas of Africa, Australasia and South America because it’s relatively easy to obtain water and nutrients there. Each mound tends to see a halo of barren soil around it because the vegetation on the mound removes water from the surroundings.

Similar structures built by ants and burrowing mammals may also affect vegetation; the team believes the results will apply to many types of mound as long as they raise either nutrient levels, water availability or both, on the mound or nearby.

“ ‘Ecosystem engineers’ like termites can play a crucial role not only in shaping ecosystems but also in influencing their robustness, i.e. their tendency to persist and also their ability to recover, in the face of environmental perturbations like drought and climate change,” said Tarnita. “Often multiple mechanisms can be at play in shaping how ecosystems are organized and how they respond to environmental change. This means that one has to be careful when attempting to infer the process responsible for a given pattern.”

Tarnita and colleagues introduced the effect of mounds made by fungus-cultivating termites from the genus Odontotermes into models of vegetation patterning that used scale-dependent feedback to allow for plants’ benefits to their neighbours and competition with vegetation further away. The team compared their model with aerial photos taken from a height of 10 m of semi-arid savanna containing termite mounds at Kenya’s Mpala Research Centre. The savanna matched the simulated pattern, with vegetation on the mounds being denser and more evenly distributed than that off-mound, which tended to be patterned on a scale of 20 cm.

The model also showed that termite mounds enable vegetation to persist under reduced rainfall, reduce the rainfall threshold needed for recovery from desert and make desertification more gradual and so easier to anticipate and stop.

“In these vulnerable arid landscapes, termite mounds may be crucial components of a healthy, functioning ecosystem, and as such they merit conservation,” said Tarnita. Drylands cover more than 40% of Earth’s land surface and are home to more than 38% of the population.

According to Tarnita, the work builds directly on two lines of research that weren't talking to each other. “One was a large body of work on the natural history of termites, ants, and other soil-dwelling ecosystem engineers,” she said. “The other was a largely theoretical literature with roots going back to Alan Turing and even further, and more recently developed by Max Rietkerk, Johan van de Koppel and colleagues, which showed that local interactions between plants could lead to the self-organization of plants in arid environments. What we did was try to put these two bodies of work together, and see how each influenced the other.”

Now Tarnita, Rob Pringle of Princeton University and Mpala Research Centre, Kenya, and Dan Doak of the University of Colorado, US, have a three-year grant to combine theoretical and empirical approaches for studying termite-generated patterns in Africa. “This paper provides the basis for a number of specific hypotheses – our next step is to implement field experiments in Kenya that will enable us to test some of these predictions for the first time,” said Tarnita.

The researchers reported their work in Science.

Related links

Related stories