Niu et al. (2026) Cascade structure and event-level attribution of three-dimensional compound drought–heat events in China: contrasting meteorological and agricultural systems

Identification

• Journal: Journal of Hydrology
• Year: 2026
• Date: 2026-07-17
• Authors: Ben Niu, Yi Li, Shasha Han, Jinxuan ZHAO, Darrell W.S. Tang, David M. Hannah
• DOI: 10.1016/j.jhydrol.2026.136057

Research Groups

• College of Water Resources and Architectural Engineering at Northwest Agriculture and Forestry University/Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas at Ministry of Education, Northwest A&F University, Yangling, Shaanxi, PR China
• Northwest A&F University Shenzhen Research Institute, Shenzhen, PR China
• School of Geography, Earth and Environmental Sciences & Birmingham Institute for Sustainability and Climate Action, University of Birmingham, Edgbaston, Birmingham, UK
• Water, Energy, and Environmental Engineering, University of Oulu, Finland

Short Summary

This study developed a three-dimensional framework to characterize the evolution and cascade structure of compound drought-heat events in China (1961-2023), revealing their complex spatiotemporal features and distinct drivers for meteorological versus agricultural systems. The findings emphasize treating these events as evolving three-dimensional systems rather than simple temporal overlaps.

Objective

• To investigate the three-dimensional cascade structure and event-level attribution of compound drought-heat events in China, contrasting their characteristics and drivers in meteorological and agricultural systems.

Study Configuration

• Spatial Scale: China, using 0.25° × 0.25° gridded data.
• Temporal Scale: Monthly data from 1961 to 2023.

Methodology and Data

• Models used:
  • GAMLSS (Generalized Additive Models for Location, Scale and Shape) for developing non-stationary drought and heat indices.
  • Copula models for deriving meteorological and agricultural compound drought-heat indices.
  • Spatial-overlap-constrained three-dimensional event-identification framework for characterizing event evolution and cascade structure.
  • Explainable machine-learning models for diagnosing dominant event-level predictors.
  • Multi-level event-cascade (Grade-cascade) analysis.
• Data sources:
  • CN05.1 gridded dataset.
  • GLDAS (Global Land Data Assimilation System) dataset.

Main Results

• Compound drought–heat events were more concentrated in the upper tails of area–severity distributions compared to single drought or heat events, particularly in the meteorological system.
• Since the 1990s, both meteorological and agricultural compound drought-heat systems intensified; meteorological events exhibited higher intensity, peak, and migration, while agricultural events showed longer persistence once established.
• Multi-level event-cascade analysis revealed that many apparently mild long-duration events contained embedded higher-grade subevents.
• Agricultural systems demonstrated a substantially greater proportion of many-to-many cascades, indicating stronger internal fragmentation and heterogeneity.
• Meteorological compound events were primarily associated with atmospheric moisture-demand and thermal predictors.
• Agricultural compound events were linked to both climatic forcing and land-surface energy–water predictors.

Contributions

• Developed and applied non-stationary drought and heat indices using GAMLSS to account for time-varying marginal distributions.
• Introduced a novel spatial-overlap-constrained three-dimensional event-identification framework to characterize the evolution and cascade structure of compound events.
• Utilized explainable machine-learning models to attribute event-level predictors, distinguishing drivers for meteorological and agricultural systems.
• Provided a comprehensive understanding of compound drought-heat events as evolving three-dimensional systems, moving beyond simple temporal overlap analyses.
• Offered contrasting insights into the characteristics, evolution, and drivers of meteorological versus agricultural compound drought-heat events in China.

Funding

• [No funding information was provided in the paper text.]

Citation

@article{Niu2026Cascade,
  author = {Niu, Ben and Li, Yi and Han, Shasha and ZHAO, Jinxuan and Tang, Darrell W.S. and Hannah, David M.},
  title = {Cascade structure and event-level attribution of three-dimensional compound drought–heat events in China: contrasting meteorological and agricultural systems},
  journal = {Journal of Hydrology},
  year = {2026},
  doi = {10.1016/j.jhydrol.2026.136057},
  url = {https://doi.org/10.1016/j.jhydrol.2026.136057}
}