Meng et al. (2026) Similarity-based classification of groundwater hydrographs to reveal regional groundwater dynamics patterns in a semi-arid agricultural area

Identification

• Journal: Hydrogeology Journal
• Year: 2026
• Date: 2026-02-10
• Authors: Xiangbo Meng, Xi Chen, Yonggen Zhang, Man GAO, Linlin Lu
• DOI: 10.1007/s10040-026-03023-4

Research Groups

• Institute of Surface-Earth System Science, Tianjin University, Tianjin, China
• Tianjin Key Laboratory of Earth Critical Zone Science and Sustainable Development in Bohai Rim, Tianjin, China
• International Research Center of Big Data for Sustainable Development Goals, Beijing, China

Short Summary

This study develops a similarity-based classification of groundwater hydrographs in the North China Plain to reveal regional groundwater dynamics patterns, identifying six distinct clusters primarily driven by crop-related irrigation demands and the long-term cumulative water balance.

Objective

• To develop and apply a similarity-based classification method for groundwater hydrographs to identify regional groundwater dynamics patterns and elucidate their driving climatic and anthropogenic factors in a semi-arid agricultural area.

Study Configuration

• Spatial Scale: Handan City, North China Plain, covering an area of 7587 square kilometers, utilizing data from 79 monitoring wells in an unconfined aquifer.
• Temporal Scale: Water table measurements from 2018 to 2021, with a monitoring interval of 10 days. Climate data (precipitation, evapotranspiration) also covers 2018–2021.

Methodology and Data

• Models used: Discrete Wavelet Transformation (DWT), Hierarchical Clustering (HC) with "ward" linkage and Euclidean distance, Silhouette Score (SC) for cluster quality analysis, Mann–Kendall (MK) trend test, Random Forest (RF) for feature importance ranking, Cross-correlation functions for lag analysis.
• Data sources:
  • Water table (WT) measurements: 79 monitoring wells (2018–2021, 10-day interval) provided by the Handan Water Resources Bureau and Haihe Conservancy Commission.
  • Climate data (precipitation, evapotranspiration): Multi-Source Weight-Ensemble Precipitation (MSWEP; 0.1° × 0.1° spatial, 3-hour temporal) and ERA5 reanalysis dataset (0.25° × 0.25° spatial, 1-hour temporal) for 2018–2021.
  • Crop type: Land use and land cover data.
  • Groundwater abstraction data: Handan Water Resources Bulletins (county-level, 2018–2021).

Main Results

• Six distinct clusters of groundwater hydrographs (CL1-CL6) were identified, representing unique patterns of water table dynamics.
• These clusters exhibit unique temporal periodicities and trends, spatially corresponding to specific groundwater depth ranges and crop patterns.
• Water tables show rising trends in areas with low-irrigation cotton cultivation (CL1, CL2) and declining trends in regions with high-water-demand crops like winter wheat and summer maize (CL4, CL5).
• The cumulative net balance of precipitation minus evapotranspiration (P-ET) was identified as the primary driver of water table variability across most sites, reflecting the combined effects of groundwater abstraction for irrigation and reductions in recharge.
• In groundwater depression cones, water table response to precipitation-driven recharge is severely weakened, and increases in evapotranspiration driven by groundwater irrigation become the dominant factor controlling continuous water table decline.
• Cross-correlation analysis revealed infiltration recharge lag times of approximately 1 to 4 months, with shallower groundwater depths corresponding to shorter lag responses.

Contributions

• Develops and applies a novel similarity-based classification method (DWT-HC combined with random forest and cross-correlation) to effectively analyze complex groundwater dynamics in semi-arid agricultural areas, addressing limitations of traditional methods in data-scarce or heterogeneous settings.
• Provides a simpler and more scalable tool for groundwater dynamics analysis, enhancing the efficient utilization of monitoring data.
• Quantifies the relative contributions of climatic variables and anthropogenic activities (specifically crop patterns and irrigation demands) to distinct groundwater dynamics patterns.
• Offers critical insights into the formation and evolution of groundwater depression cones and the lag times in infiltration recharge within the North China Plain, which were previously underexplored.
• The findings provide actionable information for guiding groundwater resource management policies, such as implementing adaptation strategies like reducing irrigation or enhancing recharge in vulnerable areas.

Funding

• National Natural Science Foundation of China (U21A2004)
• Major Science and Technology Projects of the Ministry of Water Resources of China (SKS-2022041)

Citation

@article{Meng2026Similaritybased,
  author = {Meng, Xiangbo and Chen, Xi and Zhang, Yonggen and GAO, Man and Lu, Linlin},
  title = {Similarity-based classification of groundwater hydrographs to reveal regional groundwater dynamics patterns in a semi-arid agricultural area},
  journal = {Hydrogeology Journal},
  year = {2026},
  doi = {10.1007/s10040-026-03023-4},
  url = {https://doi.org/10.1007/s10040-026-03023-4}
}