Naeem et al. (2025) Simulating and predicting lake dynamics by fusing HBV modeling, machine learning approach and remote sensing data

Identification

• Journal: Journal of Hydrology
• Year: 2025
• Date: 2025-09-23
• Authors: Muhammad Naeem, Yongqiang Zhang, Ning Ma, Zixuan Tang, Ping Miao, Xiaoqiang Tian, Congcong Li, Qi Huang, Zhenwu Xu, Longhao Wang, Zhen Huang
• DOI: 10.1016/j.jhydrol.2025.134303

Research Groups

• Key Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China
• University of Chinese Academy of Sciences, Beijing, China
• River and Lake Protection Center, Ordos Water Conservancy Bureau, Inner Mongolia, Ordos, China
• Flood and Drought Disaster Prevention Technology Center, Ordos Water Conservancy Bureau, Inner Mongolia, Ordos, China

Short Summary

This study comprehensively analyzes historical (1990-2023) and projected (up to 2060) hydrological dynamics and land use changes in the Hongjiannao Lake Basin by integrating HBV modeling, Random Forest, CA-Markov, and remote sensing, revealing significant lake area fluctuations and predicting future increases despite ongoing pressures.

Objective

• To provide a comprehensive analysis of the hydrological dynamics and land use changes in the Hongjiannao Lake Basin from 1990 to 2023, with projections extending to 2060, to understand the interactions between climate change, anthropogenic activities, and ecosystem responses.

Study Configuration

• Spatial Scale: Hongjiannao Lake Basin
• Temporal Scale: Historical analysis: 1990–2023; Projections: up to 2060.

Methodology and Data

• Models used: Hydrologiska Byrans Vattenbalansavdelning (HBV) model, Random Forest (RF) machine learning algorithm, Cellular Automata (CA) Markov model.
• Data sources: Remote sensing data.

Main Results

• The Hongjiannao Lake experienced a 25.5 % reduction in area between 2000 and 2011, followed by a 26.2 % recovery from 2012 to 2023.
• Projections indicate a potential 29 % increase in the lake area by 2060 under various future climate scenarios, suggesting ecosystem resilience.
• The Random Forest model demonstrated strong predictive capabilities for lake area, with R² values of 0.92 during 1990–2013 calibration and 0.76 during 2014–2023 validation.
• Root mean square errors (RMSE) for the RF model were 0.12 square kilometers (km²) for calibration and 0.26 square kilometers (km²) for validation.
• The CA-Markov model predicted significant landscape changes, including vegetation growth and urbanization.

Contributions

• Provides a robust framework for understanding the complex interactions between climate change, anthropogenic activities, and ecosystem responses by integrating advanced hydrological modeling (HBV), machine learning (RF), cellular automata (CA-Markov), and remote sensing data.
• Offers a comprehensive historical analysis and future projections of lake dynamics and land use changes in a semi-arid region, highlighting both past fluctuations and future trends.
• Stresses the critical need for integrated water management strategies that consider climate, land use, and hydrological factors for sustainable conservation and restoration in semi-arid environments.

Funding

• Not explicitly stated in the provided text.

Citation

@article{Naeem2025Simulating,
  author = {Naeem, Muhammad and Zhang, Yongqiang and Ma, Ning and Tang, Zixuan and Miao, Ping and Tian, Xiaoqiang and Li, Congcong and Huang, Qi and Xu, Zhenwu and Wang, Longhao and Huang, Zhen},
  title = {Simulating and predicting lake dynamics by fusing HBV modeling, machine learning approach and remote sensing data},
  journal = {Journal of Hydrology},
  year = {2025},
  doi = {10.1016/j.jhydrol.2025.134303},
  url = {https://doi.org/10.1016/j.jhydrol.2025.134303}
}