Dr Chris Huntingford and Dr Jiangpeng Cui describe recent work that used process-based modelling to describe how Amazon deforestation can reshape rainfall patterns in the region...
The land surface is not passive within the climate system. It doesn't simply respond to changes in meteorological conditions or uses, such as deforestation. Instead, land changes can also feed back into weather features, influencing atmospheric moisture transport, regional energy exchanges, and ultimately rainfall patterns.
There are many reasons to understand the implications of Amazon deforestation, including, for instance, its impact on biodiversity. However, in this recently published work, our emphasis is on how deforestation is feeding back to change rainfall levels. In particular, we were interested in the impact of Amazon land use change on rainfall levels in locations substantially downstream. Can deliberate changes to the rainforest alter rainfall levels over locations many hundreds of kilometres away, and could this affect the towns where this happens?
Our findings show that land use change in one region can reshape rainfall patterns far beyond the immediate area, highlighting the interconnected nature of the climate system.
It is known that rainfall in the Southern basin of the Amazon and its surroundings has declined substantially over recent decades. In this new research, we confirm that a major part of this reduction is due to deforestation. The loss of trees reduces the amount of rainfall returned to the atmosphere by evaporation. Furthermore, the loss of trees adjusts the structure of the atmosphere. Hence, the water that is returned to the atmosphere is more easily carried out of the Southern Amazon basin and the regions around it, weakening the land–atmosphere feedbacks that normally sustain rainfall over the Southern Amazon basin.
Over the past four decades, observed rainfall in the Southern basin has declined by approximately 10%. Our headline finding is that at least 50% of this decline is attributable to the loss of rainforest cover within the basin and in upstream regions.
A novel feature of our analysis is that, at every stage, supporting calculations were made using what are known as “process-based models”. Such models describe evaporation and the tracking of atmospheric moisture by solving sets of equations on a computer. The equations capture how each effect works, and, when coupled together, we find that they explain well the link between deforestation and lower rainfall. By combining observation-based datasets with process-based modelling, we reduce uncertainties and provide a more robust attribution of the drivers of rainfall changes.
While searching for statistical correlations can aid in discovering land-atmosphere feedbacks, process understanding provides greater robustness to the science. Such an approach allows a more rigorous explanation of observed trends.
An awareness of the background processes involved in land-atmosphere feedbacks also makes it easier to transfer this knowledge into climate models. We hope researchers will undertake this mapping, as it will allow an estimate of deforestation-induced rainfall changes in the context of any additional alteration to rainfall patterns caused by climate change.
Ultimately, improved representation of these processes will support more reliable projections of Amazon resilience, thereby informing policies on both forest conservation and climate mitigation.
Related links
Full paper in Nature Communications: Historical deforestation drives strong rainfall decline across the southern Amazon basin
