Zhang et al. (2025) A multiple spatial scales water use simulation for capturing its spatial heterogeneity through cellular automata model

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-12-12
• Authors: Jiayu Zhang, Dedi Liu, Jiaoyang Wang, Yue Feng, Huaiyuan Liang, Zhenzhen Peng, Wei Guan
• DOI: 10.5194/hess-29-7149-2025

Research Groups

• State Key Laboratory of Water Resources Engineering and Management, Wuhan University, Wuhan, China
• Hubei Provincial Key Lab of Water System Science for Sponge City Construction, Wuhan University, Wuhan, China
• Department of Earth Science, University of the Western Cape, Republic of South Africa

Short Summary

This study proposes a multi-scale water use simulation framework, integrating a cellular automata model with Generalized Likelihood Uncertainty Estimation, to address spatial heterogeneity and uncertainty in water resource planning across China. The framework reveals that both model structure and spatial scale significantly impact water use heterogeneity and uncertainty, offering a flexible tool for adaptive water resource management.

Objective

• To develop a multi-scale water use simulation framework that specifically addresses the impact of spatial scale on the spatial heterogeneity of water use.
• To quantify the uncertainties of water use simulation to provide a robust foundation for water scarcity assessment and policy-making.
• To facilitate a deeper understanding of the influences of spatial scale on water use heterogeneity and improve the accuracy of water scarcity assessments for effective resource management.

Study Configuration

• Spatial Scale: 341 prefectures across China, with simulations conducted at three resolutions: 1 kilometer (km), an appropriate adaptive spatial scale (varying per prefecture), and the prefecture administrative scale.
• Temporal Scale: Annual simulations from 1998 to 2013, with a calibration period from 1998 to 2009 and a validation period from 2010 to 2013.

Methodology and Data

• Models used:
  • Cellular Automata (CA) model for multi-scale water use simulation, employing two update rules:
    • Probability rule: Captures stochastic transitions via distribution fitting (Normal, Lognormal, Exponential, Gamma, Uniform distributions selected using Akaike Information Criterion (AIC)).
    • Linear rule: Models neighborhood-weighted evolution using a linear combination of cell's own state and neighbors' states, with inverse distance weighting.
  • Generalized Likelihood Uncertainty Estimation (GLUE) for quantifying simulation uncertainty, identifying behavioral parameter sets based on Root Mean Squared Error (RMSE) and Relative Error (RE) thresholds.
  • Coefficient of Variation (CV) and Moran's I for analyzing spatial heterogeneity and autocorrelation of water use.
• Data sources:
  • Annual water use statistical survey data at the administrative (prefecture) scale from Zhou et al. (2020), compiled from China's National Water Resources Assessment Programs (1965–2000) and Provincial Water Resources Bulletins (2001–2013).
  • Sector-specific predictor variables for generating grid maps:
    • Irrigation: Potential evapotranspiration, Normalized Difference Vegetation Index (NDVI), rainfall, soil moisture.
    • Domestic: Population, rainfall, temperature, night-light.
    • Industrial: Gross Domestic Product (GDP), night-light, population, rainfall.
  • Datasets from China Meteorological Administration (CMA), Resource and Environment Science and Data Center (RESDC), and Global Drought and Flood Catalogue (GDFC).

Main Results

• Both the update rule and spatial scale significantly influence the spatial heterogeneity and uncertainty of water use simulations.
• The probability rule effectively captures broader variability and stochastic transitions but results in higher mean RMSE (0.36 billion cubic meters) and RE (±29.8%) during validation.
• The linear rule provides more stable and accurate simulations with lower mean RMSE (0.28 billion cubic meters) and RE (±22.4%) during validation, particularly in regions with smoother water use patterns.
• The 1 km scale captures the most detailed local variations but amplifies uncertainty due to sensitivity to local fluctuations.
• The prefecture scale oversimplifies spatial details, suppressing heterogeneity and masking sub-regional dynamics.
• The appropriate spatial scale offers the best trade-off between capturing spatial heterogeneity and maintaining model stability, yielding the most reliable simulations.
• Uncertainty, quantified by GLUE at a 95% confidence level, varies regionally, with larger ranges (0.75–2.0 × 10⁹ m³) in western/southwestern prefectures (e.g., Xinjiang, Qinghai) and narrower ranges (0–0.5 × 10⁹ m³) in eastern/northeastern regions (e.g., Beijing, Jiangsu).
• The probability rule consistently yields wider uncertainty intervals than the linear rule, especially in prefectures with unstable water use conditions (e.g., Kashgar, Bayannur).

Contributions

• Proposes a novel multi-scale water use simulation framework by integrating a Cellular Automata (CA) model with Generalized Likelihood Uncertainty Estimation (GLUE), specifically addressing the impact of spatial scale on water use heterogeneity.
• Develops and compares two CA update rules (probability and linear) to capture different aspects of water use dynamics (stochastic vs. deterministic).
• Provides a comprehensive analysis of how spatial scale (1 km, appropriate, prefecture) influences the representation of water use heterogeneity and simulation uncertainty across China.
• Offers a flexible and uncertainty-aware tool for multi-scale water use simulation, supporting adaptive water resource planning, integrated water management, infrastructure planning, and environmental policy under changing socio-economic and climatic conditions.

Funding

• National Key R&D Program of China (grant nos. 2024YFC3012402 and 2022YFC3202803)
• National Natural Science Foundation of China (grant nos. 52379022 and 51879194)

Citation

@article{Zhang2025multiple,
  author = {Zhang, Jiayu and Liu, Dedi and Wang, Jiaoyang and Feng, Yue and Liang, Huaiyuan and Peng, Zhenzhen and Guan, Wei},
  title = {A multiple spatial scales water use simulation for capturing its spatial heterogeneity through cellular automata model},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-7149-2025},
  url = {https://doi.org/10.5194/hess-29-7149-2025}
}