Chambers et al. (2025) Hot droughts in the Amazon provide a window to a future hypertropical climate

Identification

• Journal: Nature
• Year: 2025
• Date: 2025-12-10
• Authors: Jeffrey Q. Chambers, Adriano José Nogueira Lima, Gilberto Pastorello, Bruno Gimenez, Lin Meng, Lee A. Dyer, Yanlei Feng, Cristina Santos da Silva, R. C. Oliveira, Anna Weber, Charles D. Koven, Robinson Negrón‐Juárez, Gustavo Spanner, Tatiana Dias Gaui, Clarissa G. Fontes, Alessandro Araùjo, Nate G. McDowell, L. Ruby Leung, Daniel Magnabosco Marra, J. M. Warren, Daisy C. Souza, Cynthia L. Wright, Kolby Jardine, Marcos Longo, Chonggang Xu, Paul V. A. Fine, Rosie A. Fisher, Javier Tomasella, Joaquim dos Santos, Níro Higuchi
• DOI: 10.1038/s41586-025-09728-y

Research Groups

• Instituto Nacional de Pesquisas da Amazônia (INPA), Brazil
• Lawrence Berkeley National Laboratory, USA
• University of California Berkeley, USA
• Vanderbilt University, USA
• University of Nevada, Reno, USA
• Massachusetts Institute of Technology, USA
• BeZero Carbon, UK
• Embrapa Amazônia Oriental, Brazil
• Pacific Northwest National Laboratory, USA
• Julius Kühn Institute (JKI) – Federal Research Centre for Cultivated Plants, Germany
• Max Planck Institute for Biogeochemistry, Germany
• Oak Ridge National Laboratory, USA
• Instituto Nacional de Pesquisas Espaciais, Brazil
• Los Alamos National Laboratory, USA
• CICERO Senter for klimaforskning, Norway

Short Summary

This study synthesizes long-term demographic, ecophysiological, and climate model data to assess the impact of "hot droughts" on central Amazon forests. It finds that hot droughts significantly increase tree mortality, particularly for fast-growing species, by pushing soil moisture below critical thresholds, and projects that these extreme conditions will become typical in a future "hypertropical" climate by 2100.

Objective

• To assess the effects of hot droughts on a central Amazon forest by synthesizing multiple datasets, and to understand how tropical forests will respond to these extreme conditions, including the mechanisms causing tree death and the projected emergence of a "hypertropical" climate.

Study Configuration

• Spatial Scale: Central Amazon forest (Manaus region, Brazil), specifically the BIONTE selective logging experiment (12 hectares of inventory plots) and the K34 tower site. Global projections for tropical forest biomes.
• Temporal Scale: Over 30 years of annually resolved forest demographic data (1990–2017, with data for 2023–2024 droughts collected); ecophysiological field measurements during the 2015 and 2023 El Niño droughts; climate model projections from 1980 to 2100.

Methodology and Data

• Models used:
  • Hierarchical Bayesian models (for tree mortality and Standardized Precipitation Evapotranspiration Index - SPEI relationships).
  • Structural Equation Models (SEM) (for causal relationships between climate, wood density, and mortality).
  • Piecewise linear regression model (for identifying soil moisture thresholds in sap flow).
  • Coupled Model Intercomparison Project Phase 6 (CMIP6) Earth System Models (ESMs) (e.g., MIROC-ES2L, BCC-CSM2-MR, GFDL-ESM4, MRI-ESM2-0, NorESM2-LM, MPI-ESM1-2-LR, CAMS-CSM1-0, GISS-E2-1-G, CESM2-WACCM, CNRM-CM6-1, CNRM-CERFACS.CNRM-CM6-1-HR, MOHC.HadGEM3-GC31-MM, NOAA-GFDL.GFDL-ESM4, NCAR.CESM2).
  • Energy Exascale Earth System Model (E3SM) (E3SMBCRD and E3SMBDRD configurations).
• Data sources:
  • Long-term (30+ years) annually resolved forest demographic data from the BIONTE selective logging experiment (tree growth, recruitment, mortality for trees ≥ 10 cm diameter at breast height).
  • Ecophysiological field measurements (2015 and 2023 droughts) from K34 tower and BIONTE eddy flux tower: sap flow (transpiration velocity), soil moisture (volumetric water content, VWC), canopy temperature, stem and soil water potential, air temperature, relative humidity, vapor pressure deficit (VPD), net solar radiation.
  • Daily precipitation and potential evapotranspiration data (1980–2016) from Embrapa meteorological station.
  • Wood density data for central Amazon species (harmonized from multiple sources).
  • CMIP6 and E3SM archives: 2-meter air temperatures, air pressure, specific humidity, mean-annual temperature, and precipitation for historical and Shared Socioeconomic Pathway (SSP) scenarios.

Main Results

• Increased Tree Mortality during Hot Droughts: A 30-year record showed higher tree mortality during intense droughts, particularly among fast-growing pioneer species with low wood density. A Bayesian model indicated that August–September SPEI values below -1.5 led to 55% higher mortality.
• Critical Soil Moisture Threshold: Ecophysiological measurements during the 2015 and 2023 droughts identified a soil moisture threshold of approximately 0.32–0.33 m³ m⁻³ soil water content (equivalent to -380 kPa soil water potential). Below this threshold, transpiration rates rapidly declined, increasing the risk of tree mortality from hydraulic failure and carbon starvation.
• Emergence of a "Hypertropical" Climate: CMIP6 analyses project that under high-emission scenarios (SSP5-8.5), a large area of tropical forest will shift to a novel "hypertropical" climate (mean annual temperature > 24 °C and precipitation > 1,300 mm per year, exceeding the 99th percentile of historical climates) by 2100.
• Elevated Future Mortality Risk: Under a hypertropical climate, temperature and moisture conditions during typical dry season months will more frequently exceed identified drought mortality thresholds. Projections show that extreme VPD events, similar to those in recent hot droughts, will become much more prevalent after 2040, with over 150 days annually by 2090–2100 under SSP5-8.5, occurring throughout the year.
• Vulnerability of Degraded Forests: Forests recovering from selective logging, with a higher abundance of early successional species, exhibited greater vulnerability to drought-induced mortality. This suggests secondary and degraded forests may be more susceptible to future hot droughts.
• Inadequate ESM Representation: Current Earth System Models (ESMs) predict weak feedbacks due to changes in tropical tree mortality and do not adequately capture the ecophysiological feedbacks associated with hydraulic constraints and elevated tree mortality under hot droughts, potentially underestimating future carbon cycle shifts.

Contributions

• Provided a unique synthesis of over 30 years of annually resolved forest demographic data, detailed ecophysiological measurements during recent extreme hot droughts (2015, 2023), and future climate model projections for a central Amazon forest.
• Identified and quantified a critical soil moisture threshold (approximately 0.32–0.33 m³ m⁻³ soil water content) below which transpiration rapidly declines, leading to increased tree mortality risk from hydraulic failure and carbon starvation across diverse species.
• Introduced and characterized the concept of an emerging "hypertropical" climate state, demonstrating its spatial and temporal emergence by the end of the century under high-emission scenarios, representing a climate with no current analogue.
• Quantified the projected increase in frequency and duration of extreme VPD events in the central Amazon, showing that conditions currently associated with severe droughts will become typical dry season conditions after 2040 and potentially year-round by 2100.
• Revealed that current Earth System Models (ESMs) likely underestimate tropical forest carbon-climate feedbacks by not fully incorporating the ecophysiological processes and elevated tree mortality driven by hydraulic constraints under hot droughts, particularly for fast-growing pioneer species.
• Demonstrated that, in addition to intense droughts, higher wet season precipitation (plausibly due to stronger convective storms under warming) is increasingly associated with higher tree mortality in recent years, highlighting a dual disturbance threat.

Funding

• Next Generation Ecosystem Experiments-Tropics (US Department of Energy, Office of Science, Office of Biological and Environmental Research)
• Madeiras da Amazônia project
• Institutos Nacionais de Ciência e Tecnologia (INCT)
• Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
• Ministério da Ciência, Tecnologia e Inovações (MCTI)
• Fundação de Amparo à Pesquisa do Estado do Amazonas (FAPEAM)
• National Science Foundation (DGE-2244337)
• ATTO Project (German Federal Ministry of Education and Research (BMBF; contracts 01LB1001A and 01LK1602A); Brazilian Ministério da Ciência, Tecnologia e Inovação (MCTI/FINEP contract 01.11.01248.00); Max Planck Society)
• European Union’s Horizon 2020 Research and Innovation program (grant agreement no. 101003536, ESM2025 — Earth System Models for the Future)

Citation

@article{Chambers2025Hot,
  author = {Chambers, Jeffrey Q. and Lima, Adriano José Nogueira and Pastorello, Gilberto and Gimenez, Bruno and Meng, Lin and Dyer, Lee A. and Feng, Yanlei and Silva, Cristina Santos da and Oliveira, R. C. and Weber, Anna and Koven, Charles D. and Negrón‐Juárez, Robinson and Spanner, Gustavo and Gaui, Tatiana Dias and Fontes, Clarissa G. and Araùjo, Alessandro and McDowell, Nate G. and Leung, L. Ruby and Marra, Daniel Magnabosco and Warren, J. M. and Souza, Daisy C. and Wright, Cynthia L. and Jardine, Kolby and Longo, Marcos and Xu, Chonggang and Fine, Paul V. A. and Fisher, Rosie A. and Tomasella, Javier and Santos, Joaquim dos and Higuchi, Níro},
  title = {Hot droughts in the Amazon provide a window to a future hypertropical climate},
  journal = {Nature},
  year = {2025},
  doi = {10.1038/s41586-025-09728-y},
  url = {https://doi.org/10.1038/s41586-025-09728-y}
}