Zhang et al. (2026) Anthropogenic warming of China and constrained future projection: updated investigation based on urbanization-bias adjusted observations and CMIP6 models

Identification

• Journal: Climatic Change
• Year: 2026
• Date: 2026-04-01
• Authors: Jintao Zhang, Guoyu Ren, Qinglong You, Kangmin WEN
• DOI: 10.1007/s10584-026-04170-z

Research Groups

• Yunnan Key Laboratory of Plateau Geographical Process and Environmental Changes, Faculty of Geography, Yunnan Normal University, Kunming, China
• Department of Atmospheric Science, School of Environmental Studies, China University of Geosciences, Wuhan, China
• National Climate Center, China Meteorological Administration, Beijing, China
• Department of Atmospheric and Oceanic Sciences, Fudan University, Shanghai, China
• Fuzhou Meteorological Bureau, Fuzhou, China
• Fujian Meteorological Science Institute, Fuzhou, China

Short Summary

This study quantifies anthropogenic contributions to China's regional warming since the 1960s using urbanization-bias adjusted observations and CMIP6 models, finding that greenhouse gases are the dominant driver and that previous projections likely overestimated future warming due to unadjusted observational biases.

Objective

• To conduct a detection and attribution analysis of annual and seasonal mean Surface Air Temperature (SAT) over China from 1961 to 2015, and subsequently project SAT changes in the 21st century, utilizing urbanization-bias adjusted observations and CMIP6 models.

Study Configuration

• Spatial Scale: China (country-averaged, eastern (105°E–east) and western (105°E–west) regions), aggregated to 2°×2° grid cells.
• Temporal Scale:
  • Historical analysis period: 1961–2015.
  • Baseline for anomalies: 1961–1990.
  • Future projection period: 2016–2100.
  • Mid-century projection period: 2041–2060.
  • Late-century projection period: 2081–2100.
  • Baseline for future changes: 1995–2014.
  • Optimal fingerprinting analysis used 11 non-overlapping five-year intervals.

Methodology and Data

• Models used: Coupled Model Intercomparison Project Phase 6 (CMIP6) Global Climate Models (GCMs).
• Data sources:
  • Observational data: Urbanization-bias adjusted monthly mean Surface Air Temperature (SAT) dataset from national meteorological stations across China (1961–2015); homogenized (urbanization bias unadjusted) SAT dataset for comparison.
  • Model data:
    • CMIP6 historical simulations (ALL forcing: 1961–2014, extended to 2015 with SSP2-4.5).
    • CMIP6 single-forcing historical simulations (hist-GHG, hist-aer, hist-nat: 1961–2015).
    • CMIP6 Shared Socioeconomic Pathway (SSP) 2-4.5 scenario simulations (2016–2100).
    • CMIP6 preindustrial control simulations (CTL) from 48 models.
  • Methodology:
    • Linear trend analysis using least-squares regression.
    • Regularized optimal fingerprinting (Ribes et al. 2013) for detection and attribution, employing single-signal, two-signal (anthropogenic and natural), and three-signal (greenhouse gases, anthropogenic aerosols, and natural) configurations.
    • Allen-Stott-Kettleborough (ASK) method for observation-constrained future projections.
    • Multi-model ensemble mean (MME) for simulated climate responses.
    • Area-weighted averaging for regional mean anomalies.

Main Results

• The urbanization-bias adjusted national-averaged annual mean SAT in China increased by 1.20 °C during 1961–2015.
• Anthropogenic forcing (ANT) signals are robustly detected on both annual and seasonal scales, and are separable from natural forcing (NAT).
• Greenhouse gases (GHG) are identified as the dominant driver of observed regional background warming across all seasons.
• A detectable cooling effect from anthropogenic aerosols (AER) is confined to summer mean SAT.
• Observation-constrained future projections under the SSP2-4.5 scenario indicate China's background climate will warm by 1.60 °C (2041–2060) and 2.51 °C (2081–2100) relative to 1995–2014 levels.
• This constrained warming magnitude is 14% lower annually and 13–22% lower seasonally compared to projections derived from unadjusted observations.
• Using unadjusted observations overestimated GHG-induced warming by 0.22 °C annually and 0.13–0.41 °C seasonally for the 1961–2015 period.
• Urbanization bias in observations does not significantly alter the detectability conclusions of warming trends but leads to a systematic overestimation of future warming magnitudes.

Contributions

• This study represents an early attempt at the detection, attribution, and observation-constrained projection of actual regional background climate change over China using urbanization-bias adjusted observations.
• It quantifies the significant overestimation of future warming projections over China in previous studies that did not account for urbanization bias in their observational baselines.
• Provides a more accurate assessment of anthropogenic climate change impacts in China by rigorously correcting for urbanization bias in the observational data used for detection, attribution, and projection.
• Offers a valuable methodology and insights for other regions globally experiencing similar urbanization-induced distortions in recorded warming trends.

Funding

• National Natural Science Foundation of China (42505171)

Citation

@article{Zhang2026Anthropogenic,
  author = {Zhang, Jintao and Ren, Guoyu and You, Qinglong and WEN, Kangmin},
  title = {Anthropogenic warming of China and constrained future projection: updated investigation based on urbanization-bias adjusted observations and CMIP6 models},
  journal = {Climatic Change},
  year = {2026},
  doi = {10.1007/s10584-026-04170-z},
  url = {https://doi.org/10.1007/s10584-026-04170-z}
}