Zhang et al. (2025) Volcanic eruptions disrupt ENSO teleconnections with land summer temperature

Identification

• Journal: Nature Communications
• Year: 2025
• Date: 2025-11-10
• Authors: Xu Zhang, Jinbao Li, Shang‐Ping Xie, Fei Liu, Feng Shi, Cong Gao, Han Zhang, Qianjin Dong
• DOI: 10.1038/s41467-025-64879-w

Research Groups

• Department of Geography, The University of Hong Kong, Hong Kong SAR, China
• Institute for Climate and Carbon Neutrality, The University of Hong Kong, Hong Kong SAR, China
• HKU Shenzhen Institute of Research and Innovation, Shenzhen, China
• Scripps Institution of Oceanography, University of California San Diego, La Jolla, CA, USA
• School of Atmospheric Sciences, Sun Yat-Sen University, Zhuhai, China
• State Key Laboratory of Lithospheric and Environmental Coevolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China
• School of Geography, South China Normal University, Guangzhou, China
• State Key Laboratory of Water Resources Engineering and Management, Wuhan University, Wuhan, China

Short Summary

This study uses observations to demonstrate that volcanic eruptions significantly disrupt global El Niño–Southern Oscillation (ENSO) teleconnections with land surface air temperature in boreal summer, a phenomenon largely missed by current Earth System Models. This disruption challenges the stationarity assumption in ENSO reconstructions and highlights limitations in climate modeling under external perturbations.

Objective

• To investigate how volcanic eruptions impact the robustness and stability of El Niño–Southern Oscillation (ENSO) teleconnections with global land surface air temperature during boreal summer.
• To evaluate the ability of state-of-the-art Earth System Models (ESMs) to capture this disruptive effect.
• To assess the implications of this disruption for ENSO reconstructions based on temperature-sensitive terrestrial proxies.

Study Configuration

• Spatial Scale: Global land areas, with specific analysis of tropical, mid-latitude, and high-latitude regions (e.g., Africa, Greenland, North America, Eurasia, Australia, South America).
• Temporal Scale: Observational period from 1901 to 2022; CMIP6 historical simulations from 1901 to 2014; NCEP/NCAR Reanalysis from 1948 to 2022; AMIP simulations from 1979 to 2014.

Methodology and Data

• Models used:
  • Coupled Model Intercomparison Project Phase 6 (CMIP6) historical simulations (25 models).
  • Atmospheric Model Intercomparison Project (AMIP) simulations (CMIP6 archive).
  • Principal Components Regression model (for ENSO reconstruction).
  • Random Forest model (for ENSO reconstruction validation).
• Data sources:
  • Land Surface Air Temperature (Ta): Ensemble mean of Climatic Research Unit (CRU) gridded Time Series 4.07, Goddard Institute for Space Systems (GISS) Surface Temperature Analysis, and Berkeley Earth Surface Temperatures (BEST) datasets.
  • Sea Surface Temperature (SST): Ensemble mean of Hadley Centre Global Sea Ice and SST (HadISST), Centennial in situ Observation-Based Estimates (COBE) SST, and NOAA extended reconstruction SSTs version 5 (ERSSTv5) datasets.
  • Geopotential Height: NCEP/NCAR Reanalysis 1 project.
  • Volcanic Eruption Data: Global Volcanism Program, Smithsonian Institution (14 eruptions with Volcanic Explosivity Index (VEI) ≥ 5). Also used stratospheric aerosol optical depth (SAOD) from ice core measurements and CMIP6 forcing, and volcanic stratospheric sulfur injections (VSSI) from ice-core reconstruction and satellite observations.
  • ENSO Index: December-to-February (DJF) Niño 3.4 index (averaged SST anomalies over 5°S-5°N and 170-120°W). Relative Niño 3.4 index also examined.
  • Filtering: 15-year Butterworth high-pass filter applied to all series.
  • Statistical Analysis: Pearson correlation coefficients, Student’s t-distribution for p-values, bootstrap analysis (1000 repeats), linear regression for sensitivity, empirical orthogonal function (EOF) analysis, moving-block 13-fold cross-validation.

Main Results

• During non-eruption years (NVol), the DJF Niño 3.4 index shows a strong positive correlation (r = 0.62, p < 0.01) with global land surface air temperature (Ta) in boreal summer.
• During volcanic eruption years (Vol), this global correlation drops significantly to 0.07, indicating a disruption of ENSO teleconnections. This reduction is statistically robust, falling below the 1% quantile of NVol resamples.
• Spatially, correlations between ENSO and boreal summer Ta shift in 62.27% of global land areas where Ta is significantly connected to ENSO during NVol years. For example, 27.40% of global land areas (e.g., Africa, Greenland, eastern North America) show a shift from significant positive to non-significant correlations.
• The sensitivity of boreal summer Ta to DJF Niño 3.4 decreases from 0.14 ± 0.06 °C/°C during NVol years to 0.07 ± 0.12 °C/°C during Vol years in significantly correlated regions.
• Volcanic eruptions disrupt stratospheric and tropospheric atmospheric circulations at mid-latitudes, which underpins the weakened ENSO teleconnections.
• Current CMIP6 Earth System Models fail to reproduce this disruptive effect, showing largely similar ENSO teleconnection patterns between NVol and Vol years, and underestimating observed shifts.
• ENSO reconstructions based on extratropical temperature-sensitive proxies are likely biased during Vol years, as demonstrated by reconstructions using boreal summer Ta, which contradict observed El Niño-like conditions following major eruptions (e.g., 1991 Mount Pinatubo, 1912 Novarupta, 1902 Santa Maria).
• AMIP simulations, which prescribe observed SSTs, successfully reproduce the decreased correlations in Vol years, suggesting that current ESMs have limitations in representing oceanic responses to volcanic forcing.

Contributions

• Provides novel observational evidence quantifying the disruptive effect of volcanic eruptions on global ENSO teleconnections with land summer temperature.
• Identifies a critical shortcoming in state-of-the-art CMIP6 Earth System Models, which fail to simulate this observed disruption, raising concerns about their reliability for climate projections under volcanic forcing and geoengineering scenarios.
• Challenges the fundamental stationarity assumption in paleoclimate ENSO reconstructions, particularly those relying on extratropical temperature-sensitive proxies, suggesting potential biases in our understanding of past ENSO variability.
• Emphasizes the need for improved characterization of regional climate impacts following volcanic eruptions and a better understanding of the complex interactions between ENSO and volcanic forcing in future climate models.
• Recommends prioritizing paleoclimate proxies from regions with stationary teleconnections for more accurate ENSO reconstructions.

Funding

• National Key Research and Development Program of China (2018YFA0605601, 2023YFF0804700)
• Hong Kong Research Grants Council (17317722)
• National Natural Science Foundation of China (52279024, 52261145744)

Citation

@article{Zhang2025Volcanic,
  author = {Zhang, Xu and Li, Jinbao and Xie, Shang‐Ping and Liu, Fei and Shi, Feng and Gao, Cong and Zhang, Han and Dong, Qianjin},
  title = {Volcanic eruptions disrupt ENSO teleconnections with land summer temperature},
  journal = {Nature Communications},
  year = {2025},
  doi = {10.1038/s41467-025-64879-w},
  url = {https://doi.org/10.1038/s41467-025-64879-w}
}