Analysis shows declining correlations between carbon and water cycles in the Amazon, suggesting increased drought impacts.
Abstract Tropical rainforest carbon and water cycles are typically assumed to be tightly coupled - defined here as exhibiting strong positive correlations - since plants exchange water for carbon through stomatal conductance1, 2. However, our analysis of 40 years of remote sensing-based estimates of gross primary production (GPP) and evapotranspiration (ET) across the Amazon Basin reveals that this coupling is both spatially and temporally non-stationary. We find that 64% of the Basin exhibits weakly positive (R<0.3) or negative (R<0) correlations between GPP and ET, particularly in regions with lowest water deficits (WD; R=0.08±0.23). These weakly positive or negative correlations are largely driven by declines in R during the dry season, implying that seasonal environmental, physiological, or phenological controls on GPP and ET diverge in areas with minimal water stress. Moreover, the strength of GPP-ET correlations has declined since 1982 coinciding with rising vapor pressure deficit (VPD) across the basin. This weakening is likely due to rising CO2 and VPD suppressing canopy conductance (Gc), thereby reducing GPP without limiting the VPD-driven increase in ET. If this decoupling trend continues, it could amplify drought impacts by increasing water loss relative to carbon gain, ultimately reducing the Amazon’s role as a carbon sink.
No takes yet. Share an insight, caveat, or question.
Worden et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: