NERC-funded research reveals that deforestation reduces rainfall

Scientists from the University of Leeds and the Natural Environment Research Council (NERC) Centre for Ecology & Hydrology have discovered that air passing over extensive forests produces more rain than air passing over little vegetation. The team also found that forests can increase rainfall thousands of kilometres away.

Whilst it is well known that vegetation returns moisture into the air via evapotranspiration, the extent of this process was previously unclear due to a dearth of observational evidence. By exploiting newly available satellite observations of rainfall and vegetation, and a model capable of predicting atmospheric wind-flow patterns, the scientists were able to explore the relationship between tropical forests and rainfall.

The team, whose findings have been published in the journal Nature, combined the observational data with predictions of future deforestation to produce a long-term forecast of rainfall across the Amazon basin. Their results suggest that by the middle of the century, rainfall during the region’s dry season will have fallen by up to a fifth.

I spoke to Dr Dominick Spracklen, Associate Professor at Leeds’s School of Earth and Environment, to find out more about deforestation’s potential to impact tropical rainfall…

Could you outline the method that you used to investigate this process?
We began working in this area because there was a lot of anecdotal evidence to suggest that forests increase rainfall, but it was difficult to find scientific proof that this process was actually happening. Model-based studies have suggested that this process should have a pronounced effect on rainfall. However, until now, we have been unable to demonstrate this through observations.

In our case, we were able to use NASA data collected by the Tropical Rainfall Monitoring Mission (TRMM). The TRMM satellite has been in orbit for over a decade and has therefore been able to collect a really large dataset concerning tropical rainfall. A rain gauge only records rainfall in one location, whereas TRMM observes the whole of the tropics. Its dataset enabled us to measure exact quantities of tropical rainfall, and allowed us to check how much forest air had travelled over during the 10 previous days. Access to this information allowed us to answer the following question: If air has passed over a large expanse of forest, does it produce more rain, less rain, or the same amount?

You and your colleagues were surprised to discover that this process occurs across more than half of the tropics. Do you know why it doesn’t take place across all of the tropics?
Some tropical areas are located in close proximity to the coast. If air is coming straight from the ocean, rainfall will not be dependent on whether or not it has passed over vegetation. The volume of moisture in air coming from the ocean is going to swamp any effect that forests might have. To give a slightly different example, consider the United Kingdom. Air that has passed over the Atlantic Ocean has been picking up moisture for approximately 4,000 miles. The presence or absence of trees would not make any significant difference to rainfall.

So this process only applies to tropical forests that are further inland…
Not really. You have to be careful thinking like that. We identified some areas in southern Brazil that were quite close to the ocean. However, the prevailing winds have almost always travelled up and over the Amazon rainforest. Although these places are located close to the shoreline, they rarely encounter winds that have arrived directly from the ocean.

Did you expect to find that this process would impact rainfall thousands of kilometres away from forests?
No. I think that we were all surprised by the strength of the effect and also, by the fact that it occurred over such a large proportion of the tropics. We suspected that there might be a big effect over the Amazon and the Congo, but we were surprised to find that the process exerted influence so far away from forests. I think that this is one of the most interesting findings of our work. Forests are affecting rainfall hundreds and thousands of kilometres downwind.

And this of course raises additional concerns over the potential impact of deforestation. How would you like to see policymakers react to your findings?
That is a very interesting question. Obviously, our findings have really strong policy implications. Southern Brazil, for example, relies heavily on agriculture. Our work suggests that by maintaining its rainforests, Brazil is supporting rainfall and in turn, its economy in the south. It makes economic sense for Brazil to protect the Amazon rainforest.

There are also some interesting policy issues concerning sovereign boundaries. The Amazon rainforest is maintaining rainfall in Paraguay, Uruguay, and even in more southerly countries. Such policy dimensions extend further than science. In order to develop effective policies concerning deforestation, multilateral cooperation will be necessary.

Do you have any further research planned in this area?
Yes. We have just submitted another grant application to NERC as we would like to study Africa in a lot more detail. Towards the end of the recent study, we focused predominantly on South America as historically, this is where deforestation has been most pronounced. However, during the last few years, Brazil seems to have gained greater control over deforestation rates. Such rates in the Congo, on the other hand, are actually increasing. A lot of countries are allowing logging companies to operate within this region and we would like to examine the related environmental consequences.