Rozante et al. (2025) Long-Term Temperature and Precipitation Trends Across South America, Urban Centers, and Brazilian Biomes

Identification

• Journal: Atmosphere
• Year: 2025
• Date: 2025-11-25
• Authors: José Roberto Rozante, Gabriela Rozante, Iracema Fonseca de Albuquerque Cavalcanti
• DOI: 10.3390/atmos16121332

Research Groups

• Center for Weather Forecast and Climate Studies, National Institute for Space Research (INPE), Cachoeira Paulista, SP, Brazil
• Faculty of Architecture, Arts, Communication, and Design, Sao Paulo State University (UNESP), Bauru, SP, Brazil

Short Summary

This study analyzed long-term trends in maximum and minimum near-surface air temperatures and precipitation across South America, its urban centers, and Brazilian biomes from 1979 to 2024 using ERA5 reanalysis. It found widespread, heterogeneous warming, with Tmax increasing faster than Tmin, and a meridional precipitation dipole, confirming the recent warming is an intensification of an externally forced secular climate change signal.

Objective

• To investigate long-term trends in maximum (Tmax) and minimum (Tmin) near-surface air temperatures and precipitation across South America, its national capitals, and Brazilian biomes (Amazon, Cerrado, Atlantic Forest, Caatinga, Pantanal, and Pampa) from 1979 to 2024.
• To determine whether the observed warming trend reflects a century-scale externally forced climate change signal or is primarily influenced by internally generated interdecadal climate variability.

Study Configuration

• Spatial Scale: Continental South America, including all national capitals, and specifically the six Brazilian biomes (Amazon, Atlantic Forest, Cerrado, Caatinga, Pantanal, Pampa).
• Temporal Scale: Primary analysis: 1979–2024 (46 years). Century-scale analysis: 1940–2024 (85 years). Fire data: 2013–2024.

Methodology and Data

• Models used:
  • Seasonal-Trend decomposition based on Loess (STL) for seasonal cycle removal.
  • Ordinary Least Squares (OLS) regression for temperature trend quantification.
  • Non-parametric Mann–Kendall test combined with Theil–Sen slope estimation for precipitation trend assessment.
  • Heteroskedasticity- and autocorrelation-consistent (HAC) regression with Newey–West correction for century-scale temperature trends.
  • Bootstrap procedure based on a first-order autoregressive model AR(1) to assess internal variability.
  • Continuous wavelet transform (CWT) to identify dominant time–frequency patterns.
• Data sources:
  • ERA5 reanalysis (European Centre for Medium-Range Weather Forecasts - ECMWF) for hourly near-surface air temperature (T2m) and total precipitation (at approximately 31 km horizontal resolution).
  • Satellite-based active fire detections from the Brazilian National Institute for Space Research (INPE) TerraBrasilis platform (MODIS and VIIRS sensors).

Main Results

• Widespread and spatially heterogeneous warming across South America (1979–2024), with Tmax increasing faster than Tmin, consistent with reduced cloudiness and evaporative cooling.
• Strongest Tmax increases (>0.4 °C decade⁻¹) observed over the Amazon, Central-West and Southeast Brazil, northern Argentina, Paraguay, Bolivia, Venezuela, and the Guianas.
• Tmin warming rates exceeding 0.3 °C decade⁻¹ found in northern Colombia, northern Chile, and southeastern Bolivia.
• A meridional precipitation dipole is evident: drying across central and southern South America (e.g., northern Argentina, Paraguay, Bolivia, central Brazil, parts of the Amazon) with declines reaching over 40 mm decade⁻¹. Conversely, rainfall increases (>20 mm decade⁻¹) are observed in the northernmost sector of the continent, linked to Intertropical Convergence Zone (ITCZ) shifts.
• South American capitals show pervasive warming (Tmax increases exceeding 0.35 °C decade⁻¹ in Asunción, La Paz, Brasília, and Caracas) and a precipitation dipole (drying in central/southern capitals, increases in northern capitals).
• Among Brazilian biomes, the Pantanal is the most vulnerable, showing strong warming (Tmax ≈ 0.51 °C decade⁻¹) and the steepest rainfall decline (≈−10.45 mm decade⁻¹). The Cerrado also exhibits strong concurrent warming (Tmax ≈ 0.38 °C decade⁻¹) and significant drying (≈−7.92 mm decade⁻¹).
• Satellite-detected fire outbreaks (2013–2024) show marked upward trends in the Amazon (+3663.7 fires yr⁻¹), Pantanal (+770.0 fires yr⁻¹), and Cerrado (+640.4 fires yr⁻¹), aligning with hotter and drier climate conditions.
• Century-scale diagnostics (1940–2024) confirm that the 1979–2024 warming (0.238 °C decade⁻¹) is an intensification of an externally forced secular warming signal, lying well above the range of internal interdecadal variability (e.g., AMO, PDO), which had a trend of 0.155 °C decade⁻¹ for the full period.

Contributions

• Provides a comprehensive and spatially continuous assessment of long-term temperature (Tmax and Tmin) and precipitation trends across the entire South American continent, including urban centers and all Brazilian biomes, overcoming limitations of sparse in situ observations.
• Explicitly quantifies and explains the asymmetric warming (Tmax increasing faster than Tmin) and identifies a clear meridional precipitation dipole, offering a holistic hydroclimatic perspective.
• Identifies specific regional hotspots of compounded temperature and precipitation change, such as the Pantanal and Cerrado, linking local vulnerabilities to large-scale atmospheric circulation patterns.
• Utilizes multi-method diagnostic analyses (moving-window trends, AR(1) bootstrap, CWT) to rigorously demonstrate that the recent warming is an externally forced secular signal, not attributable to internal interdecadal climate variability.
• Integrates satellite-based fire detection data to provide empirical corroboration of the increasing fire risk associated with the observed warming and drying trends in key biomes.

Funding

• Georeferenced Information Base Project (BIG) of INPE, funded by the National Fund for Scientific and Technological Development (FNDCT), with financial collaboration from the Funding Authority for Studies and Projects (FINEP) and the Foundation for Science, Applications, and Space Technology (FUNCATE), under grant No. 01.22.0504.00.
• CNPq through project 306738/2023-6 (for Iracema Fonseca de Albuquerque Cavalcanti).

Citation

@article{Rozante2025LongTerm,
  author = {Rozante, José Roberto and Rozante, Gabriela and Cavalcanti, Iracema Fonseca de Albuquerque},
  title = {Long-Term Temperature and Precipitation Trends Across South America, Urban Centers, and Brazilian Biomes},
  journal = {Atmosphere},
  year = {2025},
  doi = {10.3390/atmos16121332},
  url = {https://doi.org/10.3390/atmos16121332}
}