Climate network analysis reveals Amazon rainforest now more connected to global climate

The climate crisis is a vicious cycle. Increases in temperatures, rising sea levels, and the growing intensity of extreme weather wreak havoc on ecosystems. In turn, the damage to vital ecosystems, such as the Amazon rainforest, further exacerbates the climate crisis.

This process is quantified as connectivity, a measure of correlation between surface air temperatures in various regions. Giammarese et al. use climate network analysis to investigate the evolution of connectivity between the global climate system and the Amazon. Analyzing spatiotemporal surface air temperature data from the last seven decades, the researchers demonstrated that the connectivity between Amazon and the global climate system has changed.

An already influential part of the global climate, the Amazon is gaining longer range connectivity with the ability to spread a perturbation faster and further.

The team found that the rainforest is gaining long-range links to several highly dynamic climate system components, including the western Atlantic region, critical for the future stability of the Atlantic Meridional Overturning Circulation.

The analysis also revealed increased connectivity between the Amazon and the South Asian monsoon region, along with the climatically critical eastern and central Pacific regions.

“This procedure could be applied to studying other vulnerable regions in the climate system as well,” said author Adam Giammarese. “One may apply our methods to various problems, such as the spread of biological and social contagions, for example.”

The team introduced several methodological innovations to climate network analysis, applying new approaches to climate network analysis before, such as K-nearest neighbor graphs, Laplacian Eigenmaps, random walks, and directed graph diffusion.

Source: “Reconfiguration of Amazon’s connectivity in the climate system,” by Adam Giammarese, Jacob Brown, and Nishant Malik, Chaos (2024). The article can be accessed at https://doi.org/10.1063/5.0165861 .

This paper is part of the Nonlinear dynamics, synchronization and networks: Dedicated to Juergen Kurths’ 70th birthday Collection, learn more here .