Kirchengast et al. (2026) A new class of climate hazard metrics and its demonstration: revealing a ten-fold increase of extreme heat over Europe

Identification

• Journal: Weather and Climate Extremes
• Year: 2026
• Date: 2026-01-13
• Authors: Gottfried Kirchengast, Stephanie Haas, Jürgen Fuchsberger
• DOI: 10.1016/j.wace.2026.100855

Research Groups

• Wegener Center for Climate and Global Change, University of Graz, Graz, Austria
• Institute of Physics, University of Graz, Graz, Austria

Short Summary

This study introduces a new class of threshold-exceedance-amount (TEA) metrics to holistically quantify climate hazard extremity, demonstrating its utility by revealing an approximate ten-fold increase in the total events extremity (TEX) of extreme heat over Europe from 1961–1990 to 2010–2024, a change far exceeding natural variability.

Objective

• To introduce and demonstrate a new class of threshold-exceedance-amount (TEA) metrics that rigorously and consistently track changes in multiple characteristics of extreme events (frequency, duration, magnitude, area, and timing aspects like daily exposure and seasonal shift) as separate, partially compound, and fully compound metrics.
• To apply these new metrics to extreme heat events across Europe to quantify their amplification under recent climate change and assess the anthropogenic contribution beyond natural variability.

Study Configuration

• Spatial Scale: Local-to-country scale (e.g., Austria, ~100 km² to 83,883 km²) and continental-scale across Europe (e.g., total European land area 8,173,095 km²). Input data grid resolutions range from 1 km x 1 km to 30 km x 30 km. Data grid cells above 1500 meters altitude were excluded.
• Temporal Scale: Daily temperature and precipitation datasets covering 1961 to 2024. Analysis compares a reference period (1961–1990) with a recent "current climate" period (2010–2024). Decadal-mean annual time series (1966–2020) and long-term historic data (late 19th century, early 15th century) were used for natural variability estimation.

Methodology and Data

• Models used:
  • Threshold-Exceedance-Amount (TEA) metrics framework: A self-developed Python 3 software package implementing a customized space-time filter for cascaded and multi-scale extraction of extreme event statistics.
  • Computation workflow: Involves preprocessing, computation of daily basis variables, annual indicator variables, decadal-mean indicator variables, amplification factors, and aggregate georegion variables.
• Data sources:
  • SPARTACUS v2.1 daily gridded data (1 km x 1 km, Austria, 1961–2024) for daily maximum temperature (Tmax) and daily precipitation amount (P24H).
  • ERA5-Land reanalysis hourly gridded data (0.1° x 0.1°, Europe, 1950-present) for Tmax, hourly temperatures (Tm), and P24H.
  • ERA5 reanalysis hourly gridded data (0.25° x 0.25°, Europe, 1940-present) for Tmax, Tm, and P24H.
  • E-OBS v29.0e daily gridded data (0.25° x 0.25°, Europe, 1950-present) for cross-checking ERA5 results.
  • GeoSphere Austria Measurement stations Daily data v2 (7 representative stations, late 19th century-1990) and HISTALP station data for natural variability estimation.
  • ModE-RA monthly reanalysis data (1422-2008) for further natural variability assessment.
  • Reference threshold maps: 99th percentile of daily maximum temperature (Tmax) and 95th percentile of daily precipitation amount (P24H) over 1961–1990.

Main Results

• The new TEA metrics reveal an approximate ten-fold amplification (factors of around 10, with a range of 5 to 25) in the Total Events Extremity (TEX) of extreme heat over Austrian and most central and southern European regions when comparing the 2010–2024 period to the 1961–1990 reference period.
• For Austria, the decadal-mean TEX of extreme heat amplified by a factor of ~7.8. This was driven by frequency amplification (~1.7), duration amplification (~1.5–1.7), magnitude amplification (~1.5–1.7), and exceedance area expansion (~1.8).
• This TEX amplification in Austria emerged strongly from the mid-1990s and reached beyond an estimated ten-standard-deviations level in 2010–2024, providing unequivocal evidence of anthropogenic climate change.
• Local-scale analysis within Austria showed significant spatial variability in amplification, with smaller southeastern Alpine foreland regions experiencing stronger heat amplification (e.g., TEX amplification factors of ~16–20).
• Including the daily exposure time metric, the estimated TEX amplification for Austrian regions increased to ~10–31, indicating a striking ten-to thirty-fold increase in extreme heat exposure.
• The annual exposure period for extreme heat expanded by ~1 to 1.5 months since 1961–1990, corresponding to amplification factors of ~1.3–1.8 for Austria and ~2.8–3.7 for smaller sub-regions.
• Across Europe, the strongest TEX amplifications (around 11) were found in central Europe (with hotspots exceeding 20 in Eastern Europe), around 7 in southern Europe, and smallest (around 3) in northern Europe.
• The about ten-fold TEX amplification over continental Europe is approximately five times stronger than the amplification found for the more limited warm-days metric (TX90p).
• The atmospheric-boundary-layer exceedance heat content (AEHC) amplification over Central Europe (2001–2020|2010–2024 vs. 1961–1990) was about eight-to eleven-fold (~1020|1350 PJ/yr vs. ~121 PJ/yr), which is about twice as much as the northern mid-latitude atmospheric heat content gain (AHCg) amplification.
• For daily precipitation extremes in Austria, amplifications were much smaller (about 1.0–1.2), with only event frequency marginally emerging from natural variability, highlighting the challenges of tracking local-scale precipitation extremes with current datasets.

Contributions

• Introduces a novel, holistic class of Threshold-Exceedance-Amount (TEA) metrics that provides a rigorous and consistent framework for quantifying the multi-faceted characteristics of extreme events (frequency, duration, magnitude, area, and timing) in both fine-grained and compound forms (e.g., Total Events Extremity, TEX).
• Offers a unified methodology to track changes in extreme event characteristics, providing a deeper and more comprehensive understanding of extremity amplification under climate change than previously available dispersed indices.
• Demonstrates the significant added value of compound metrics like TEX by revealing a substantially higher (approx. ten-fold) amplification of extreme heat over Europe compared to conventional metrics, unequivocally attributing this to anthropogenic climate change.
• Provides a robust tool for quantifying anthropogenic amplification against natural variability, which is crucial for climate change impact analyses, extreme event attribution, and potential climate litigation.
• The generic nature of the underlying space-time filter method allows for broad applicability beyond weather and climate extremes, extending to other physical, socio-economic, or biological systems where threshold exceedances are relevant.

Funding

• WegenerNet funding (provided by the University of Graz based on field of excellence and core research infrastructure funds from the Austrian Federal Ministry for Women, Science and Research, the State of Styria, and the City of Graz).
• Research is part of the Field of Excellence Climate Change Graz.
• Publication costs funded by the University of Graz.

Citation

@article{Kirchengast2026new,
  author = {Kirchengast, Gottfried and Haas, Stephanie and Fuchsberger, Jürgen},
  title = {A new class of climate hazard metrics and its demonstration: revealing a ten-fold increase of extreme heat over Europe},
  journal = {Weather and Climate Extremes},
  year = {2026},
  doi = {10.1016/j.wace.2026.100855},
  url = {https://doi.org/10.1016/j.wace.2026.100855}
}