Spiteri et al. (2026) Nonlinear economic damages from compound heat and drought events

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Environmental Research Letters
• Year: 2026
• Date: 2026-09-17
• Authors: Sarah Spiteri, Léonore Lebouteiller, Nicole Vorderobermeier, Mar Delgado-Téllez, Andrej Ceglar
• DOI: 10.1088/1748-9326/aea91d

Research Groups

• European Commission's Joint Research Centre (JRC)
• University of [Unknown]
• Other EU research institutions and departments involved in the study are not specified.

Short Summary

This paper develops climate-augmented models to predict regional economic growth in Europe, combining standard economic indicators with high-frequency climate variables. The results show that climate predictors improve predictive accuracy in the agricultural sector, but have limited gains in other sectors.

Objective

• Investigate the short-term economic impacts of heatwaves and droughts on EU regions using a combination of climate-augmented models and machine learning approaches.

Study Configuration

• Spatial Scale: 1117 EU regions over 2002–2022.
• Temporal Scale: Daily to annual time steps, with a focus on high-frequency climate variables.

Methodology and Data

• Models used: Random Forest, XGBoost, and linear benchmarks.
• Data sources: Standard economic indicators (e.g., GDP) combined with high-frequency climate data from satellites, observations, and reanalysis products.

Main Results

• Climate predictors improve predictive accuracy in the agricultural sector by 1.93–7.36 percentage points in 99% of regions under an extreme compound heat-drought scenario.
• Heatwave indicators contribute robustly to predictive performance.
• The importance of drought varies by sector, with limited gains in other sectors.

Contributions

• This study highlights the value of climate-augmented predictive modelling for early warning, regional fiscal planning, and targeted adaptation in the face of climate-related economic shocks.
• The use of machine learning approaches (Random Forest and XGBoost) demonstrates improved capture of nonlinear, seasonal, and spatial climate-economic interactions.

Funding

• This research was supported by [Unknown funding projects, programs, and reference codes].

Citation

@article{Spiteri2026Nonlinear,
  author = {Spiteri, Sarah and Lebouteiller, Léonore and Vorderobermeier, Nicole and Delgado-Téllez, Mar and Ceglar, Andrej},
  title = {Nonlinear economic damages from compound heat and drought events},
  journal = {Environmental Research Letters},
  year = {2026},
  doi = {10.1088/1748-9326/aea91d},
  url = {https://doi.org/10.1088/1748-9326/aea91d}
}