Articles | Volume 30, issue 20
https://doi.org/10.5194/hess-30-6459-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/hess-30-6459-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
A data-driven method coupling multiple physical constraints for correcting structural errors in groundwater contaminant transport models
Jinglong Tian
School of Earth Sciences and Engineering, Key Laboratory of Surficial Geochemistry, Ministry of Education, Nanjing University, Nanjing, China
Xiankui Zeng
CORRESPONDING AUTHOR
School of Earth Sciences and Engineering, Key Laboratory of Surficial Geochemistry, Ministry of Education, Nanjing University, Nanjing, China
Yue Pan
College of Transport Science and Engineering, Nanjing Tech University, Nanjing, China
Dong Wang
School of Earth Sciences and Engineering, Key Laboratory of Surficial Geochemistry, Ministry of Education, Nanjing University, Nanjing, China
Jichun Wu
School of Earth Sciences and Engineering, Key Laboratory of Surficial Geochemistry, Ministry of Education, Nanjing University, Nanjing, China
Related authors
No articles found.
Luyao Ren, Xiankui Zeng, Dong Wang, and Jichun Wu
EGUsphere, https://doi.org/10.5194/egusphere-2026-5732, https://doi.org/10.5194/egusphere-2026-5732, 2026
This preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).
Short summary
Short summary
Groundwater can cross surface-watershed boundaries, transferring water between watersheds, yet this process is often overlooked at large scales. We reconstructed inter-basin groundwater flow (IGF) in 36 major basins worldwide. In nearly half, the magnitude of long-term mean IGF exceeded 20 % of precipitation. Precipitation could rapidly drive IGF variations, and IGF could support evapotranspiration as an external water source. Ignoring IGF may limit understanding of the hydrological cycle.
Chenlu Yu, Dong Wang, Vijay P. Singh, Pengcheng Xu, Chenjian Yan, Xiankui Zeng, Jianguo Jiang, Mei Li, Qun Li, Shengtian Zhang, and Jichun Wu
EGUsphere, https://doi.org/10.5194/egusphere-2026-4803, https://doi.org/10.5194/egusphere-2026-4803, 2026
This preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).
Short summary
Short summary
Soil moisture affects water availability and ecosystems, but predicting its changes is difficult under climate change. We developed a framework that separates seasonal, long-term, and random variations using statistics and artificial intelligence. Long-term data across China showed seasonal signals are strongest predictors, while long-term changes have effects. The results reveal shifting controls from water loss to rainfall along dry-to-wet regions, supporting drought monitoring and management.
Danhong Chen, Xiankui Zeng, Dongwei Gui, Dong Wang, and Jichun Wu
Hydrol. Earth Syst. Sci., 30, 3741–3762, https://doi.org/10.5194/hess-30-3741-2026, https://doi.org/10.5194/hess-30-3741-2026, 2026
Short summary
Short summary
Our study in the Tarim River Basin tackles the tough balance between farming and nature in dry areas. Using a model, we found that boosting farm income often means less water for ecosystems and more pollution. The solution requires different water use strategies across the region. Importantly, farmland area and crop choices must change with the weather, shrinking in dry years. This approach provides a practical plan to manage water, ecology, and agriculture together.
Jitian Wu, Xiankui Zeng, Qihui Wu, Dong Wang, and Jichun Wu
Hydrol. Earth Syst. Sci., 30, 1563–1583, https://doi.org/10.5194/hess-30-1563-2026, https://doi.org/10.5194/hess-30-1563-2026, 2026
Short summary
Short summary
Basin-scale river channel movement shapes water distribution and ecosystems. Accurately measuring this process is vital for basin management and climate change adaptation. This study introduces a fully coupled landscape evolution model with channel extraction to simulate such large-scale channel migration. Additionally, a modified Bayesian uncertainty, combined with a surrogate model, is used for parameter calibration. This approach helps reconstruct and predict river channel migration patterns.
Ziyue Yin, Jian Song, Dianguang Liu, Jianfeng Wu, Yun Yang, Yuanyuan Sun, and Jichun Wu
Hydrol. Earth Syst. Sci., 29, 3957–3973, https://doi.org/10.5194/hess-29-3957-2025, https://doi.org/10.5194/hess-29-3957-2025, 2025
Short summary
Short summary
Mining activities threaten aquatic ecosystems, soil ecosystems, and human health worldwide. This study establishes a high-quality database and a national 0.5° gridded dataset to reveal the status and spatial pattern of mining-affected water pollution, human health risks, and potential multifaceted challenges in China. It provides in-depth insights to guide policymakers in designing differentiated management strategies for the sustainable development of mines.
Cited articles
Abu Doush, I., El-Abd, M., Hammouri, A., and Bataineh, M.: The effect of different stopping criteria on multi-objective optimization algorithms, Neural Comput. Appl., 35, 1125–1155, https://doi.org/10.1007/s00521-021-05805-1, 2023.
Ayensa-Jiménez, J., Doweidar, M., Sanz-Herrera, J., and Doblaré, M.: A new reliability-based data-driven approach for noisy experimental data with physical constraints, Comput. Method. Appl. M., 328, 752–774, https://doi.org/10.1016/j.cma.2017.08.027, 2018.
Cao, M. and Weng, Q.: Embedded physical constraints in machine learning to enhance vegetation phenology prediction, GISci. Remote Sens., 61, 2426598, https://doi.org/10.1080/15481603.2024.2426598, 2024.
Clement, T. P.: A Modular Computer Code for Simulating Reactive Multi-Species Transport in 3-Dimensional Groundwater Systems, PNNL-11720, Pacific Northwest National Laboratory, Richland, Washington, USA, https://doi.org/10.2172/8022, 1999.
Cooper, C., Zhang, J., and Gao, R.: Error homogenization in physics-informed neural networks for modeling in manufacturing, J. Manuf. Syst., 71, 298–308, https://doi.org/10.1016/j.jmsy.2023.09.013, 2023.
Engdahl, N.: Impacts of uncertain permeability fields on the transient hydrologic response in coupled surface-subsurface simulations of a headwaters catchment, Water Resour. Res., 61, e2025WR040668, https://doi.org/10.1029/2025WR040668, 2025.
Guger, M., Enzinger, C., Leutmezer, F., Kraus, J., Kalcher, S., Kvas, E., and Berger, T. on behalf of the Austrian MS Treatment Registry (AMSTR): Oral therapies for treatment of relapsing-remitting multiple sclerosis in Austria: a 2-year comparison using an inverse probability weighting method, J. Neurol., 267, 2090–2100, https://doi.org/10.1007/s00415-020-09811-6, 2020.
Guo, S., Agarwal, M., Cooper, C., Tian, Q., Gao, R., Grace, W., and Guo, Y.: Machine learning for metal additive manufacturing: Towards a physics-informed data-driven paradigm, J. Manuf. Syst., 62, 145–163, https://doi.org/10.1016/j.jmsy.2021.11.003, 2022.
Gupta, H., Clark, M., Vrugt, J., Abramowitz, G., and Ye, M.: Towards a comprehensive assessment of model structural adequacy, Water Resour. Res., 48, W08301, https://doi.org/10.1029/2011WR011044, 2012.
Harbaugh, A. W., Langevin, C. D., Hughes, J. D., Niswonger, R. N., and Konikow, L. F.: MODFLOW-2005 version 1.12.00, the U.S. Geological Survey modular groundwater model, U.S. Geological Survey [code], https://doi.org/10.5066/F7RF5S7G, 2017.
He, Y. and Xiu, D.: Numerical strategy for model correction using physical constraints, J. Comput. Phys., 313, 617–634, https://doi.org/10.1016/j.jcp.2016.02.054, 2016.
Hendrikx, T., Bierkens, M., de Louw, P., and Essink, G.: Optimizing scavenger well strategies under parameter uncertainty to maximize allowable freshwater pumping rates in a coastal aquifer, The Netherlands, J. Hydrol., 664, 134554, https://doi.org/10.1016/j.jhydrol.2025.134554, 2026.
Hu, X., Shi, L., and Lin, G.: The data-driven solution of energy imbalance-induced structural error in evapotranspiration models, J. Hydrol., 597, 126205, https://doi.org/10.1016/j.jhydrol.2021.126205, 2021.
Kadhar, K. and Baskar, S.: A stopping criterion for decomposition-based multi-objective evolutionary algorithms, Soft Comput., 22, 253–272, https://doi.org/10.1007/s00500-016-2331-7, 2018.
Kasiviswanathan, K., He, J., Tay, J., and Sudheer, K.: Enhancement of Model Reliability by Integrating Prediction Interval Optimization into Hydrogeological Modeling, Water Resour. Manag., 33, 229–243, https://doi.org/10.1007/s11269-018-2099-x, 2019.
Kennedy, M. and O'Hagan, A.: Bayesian calibration of computer models, J. Roy. Stat. Soc. B, 63, 425–464, https://doi.org/10.1111/1467-9868.00294, 2001.
Kim, T., Han, W., Shin, J., Yoon, S., Kang, P., Yang, W., and Lee, S.: Quantitative evaluation of hydrogeologic properties controlling field-scale DNAPL contamination across porous and fractured media, J. Hydrol., 666, 134819, https://doi.org/10.1016/j.jhydrol.2025.134819, 2026.
Li, F., Morgan, K., and Zaslavsky, A.: Balancing Covariates via Propensity Score Weighting, J. Am. Stat. Assoc., 113, 390–400, https://doi.org/10.1080/01621459.2016.1260466, 2018.
Lu, X., Liao, W., Zhang, Y., and Huang, Y.: Intelligent structural design of shear wall residence using physics-enhanced generative adversarial networks, Earthq. Eng. Struct. D., 51, 1657–1676, https://doi.org/10.1002/eqe.3632, 2022.
Melching, D., Schultheis, E., and Breitbarth, E.: Generating artificial displacement data of cracked specimen using physics-guided adversarial networks, Mach. Learn., 4, 045063, https://doi.org/10.1088/2632-2153/ad15b2, 2023.
Mendenhall, C., Hardan, J., Chiang, T., Blumenschein, L., and Tepole, A.: Physics-Informed Neural Network for Scalable Soft Multi-Actuator Systems, in: Proceedings of the 2024 IEEE 7th International Conference on Soft Robotics (RoboSoft), San Diego, CA, USA, 14–17 April 2024, 716–721, https://doi.org/10.1109/robosoft60065.2024.10522053, 2024.
Neuman, S. P. and Wierenga, P. J.: A comprehensive strategy of hydrogeologic modeling and uncertainty analysis for nuclear facilities and sites, NUREG/CR-6805, U.S. Nuclear Regulatory Commission, Washington, DC, https://www.nrc.gov/regulations-legislation/nureg-series-publications/publications-prepared-by-nrc-contractors/cr6805 (last access: 8 October 2026), 2003.
Nkedikizza, P., Biggar, J., Selim, H., Vangenuchten, M., Wierenga, P., Davidson, J., and Nielsen, D.: On The Equivalence Of 2 Conceptual Models For Describing Ion-Exchange During Transport Through An Aggregated Oxisol, Water Resour. Res., 20, 1123–1130, https://doi.org/10.1029/WR020i008p01123, 1984.
Pan, Y., Zeng, X., Xu, H., Sun, Y., Wang, D., and Wu, J.: Assessing human health risk of groundwater DNAPL contamination by quantifying the model structure uncertainty, J. Hydrol., 584, 124690, https://doi.org/10.1016/j.jhydrol.2020.124690, 2020.
Pan, Y., Zeng, X., Xu, H., Sun, Y., Wang, D., and Wu, J.: Use of stacked Gaussian processes regression method to improve prediction of groundwater solute transport model, J. Hydrol., 620, 129530, https://doi.org/10.1016/j.jhydrol.2023.129530, 2023.
Pathiraja, S., Moradkhani, H., Marshall, L., Sharma, A., and Geenens, G.: Data-Driven Model Uncertainty Estimation in Hydrologic Data Assimilation, Water Resour. Res., 54, 1252–1280, https://doi.org/10.1002/2018WR022627, 2018.
Rasmussen, C. E. and Williams, C. K. I.: Covariance functions, in: Gaussian Processes for Machine Learning, The MIT Press, 79–104, https://doi.org/10.7551/mitpress/3206.003.0007, 2005.
Refsgaard, J. C., Van Der Sluijs, J. P., Brown, J., and Van Der Keur, P.: A framework for dealing with uncertainty due to model structure error, Adv. Water Resour., 29, 1586–1597, https://doi.org/10.1016/j.advwatres.2005.11.013, 2006.
Saha, P., Dash, S., and Mukhopadhyay, S.: Physics-incorporated convolutional recurrent neural networks for source identification and forecasting of dynamical systems, Neural Networks, 144, 359–371, https://doi.org/10.1016/j.neunet.2021.08.033, 2021.
Šimůnek, J., Šejna M., Brunetti G., and van Genuchten, M. Th.: The HYDRUS Software Package for Simulating the One-, Two-, and Three-Dimensional Movement of Water, Heat, and Multiple Solutes in Variably Saturated Media, Technical Manual I, HYDRUS-1D, Version 5.0, PC Progress, Prague, Czech Republic, 334 pp., https://www.pc-progress.com/en/Default.aspx?hydrus (last access: 8 October 2026), 2022.
Sun, X., Zeng, X., Wu, J., and Wang, D.: A Two-Stage Bayesian Data-Driven Method to Improve Model Prediction, Water Resour. Res., 57, e2021WR030436, https://doi.org/10.1029/2021WR030436, 2021.
Tian, J.: Data set for “A data-driven method coupling multiple physical constraints for correcting structural errors in groundwater contaminant transport models”, HydroShare [data set], https://www.hydroshare.org/resource/60f2db9c87914233b7fac0dcbe1f4f7b/ (last access: 8 October 2026), 2026.
Tian, J., Zeng, X., Wang, D., and Wu, J.: A data-driven approach coupled with physical constraints to improve groundwater models with structural error, Water Resour. Res., 62, e2025WR040247, https://doi.org/10.1029/2025WR040247, 2026.
Vangenuchten, M.: A Closed-form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils, Soil Sci. Soc. Am. J., 44, 892–898, https://doi.org/10.2136/sssaj1980.03615995004400050002x, 1980.
Watson, T., Doherty, J., and Christensen, S.: Parameter and predictive outcomes of model simplification, Water Resour. Res., 49, 3952–3977, https://doi.org/10.1002/wrcr.20145, 2013.
Wen, Z., Zhu, Q., Krause, S., and Chen, W.: Advancing Multi-Phase LNAPL Extraction Models by Incorporating a Wellbore Module, Water Resour. Res., 61, e2025WR041014, https://doi.org/10.1029/2025WR041014, 2025.
Wu, J., Wang, J.-X., and Shadden, S. C.: Adding Constraints to Bayesian Inverse Problems, Proc. AAAI Conf. Artif. Intell., 33, 1666–1673, https://doi.org/10.1609/aaai.v33i01.33011666, 2019.
Wu, J.-L, Levine, M., Schneider, T., and Stuart, A.: Learning about structural errors in models of complex dynamical systems, J. Comput. Phys., 513, 113157, https://doi.org/10.1016/j.jcp.2024.113157, 2024.
Wu, X., Gao, B., Lyu, X., Zeng, X., Wu, J., and Sun, Y.: Insight into the mechanism of phosphate and cadmium co-transport in natural soils, J. Hazard. Mater., 435, 129095, https://doi.org/10.1016/j.jhazmat.2022.129095, 2022.
Xu, T. and Valocchi, A. J.: A Bayesian approach to improved calibration and prediction of groundwater models with structural error, Water Resour. Res., 51, 9290–9311, https://doi.org/10.1002/2015WR017912, 2015.
Xu, T., Valocchi, A., Ye, M., and Liang, F.: Quantifying model structural error: Efficient Bayesian calibration of a regional groundwater flow model using surrogates and a data-driven error model, Water Resour. Res., 53, 4084–4105, https://doi.org/10.1002/2016WR019831, 2017.
Ye, M., Neuman, S., and Meyer, P.: Maximum likelihood Bayesian averaging of spatial variability models in unsaturated fractured tuff, Water Resour. Res., 40, W05113, https://doi.org/10.1029/2003WR002557, 2004.
Yu, H., Wang, X., Ren, B., Zeng, T., Lv, M., and Wang, C.: An efficient Bayesian inversion method for seepage parameters using a data-driven error model and an ensemble of surrogates considering the interactions between prediction performance indicators, J. Hydrol., 604, 127235, https://doi.org/10.1016/j.jhydrol.2021.127235, 2022.
Zeeshan, M. and Ruhl, A.: Fates of potentially persistent and mobile organic substances in embedded outdoor columns for artificial groundwater recharge simulation, Water Res., 245, 120615, https://doi.org/10.1016/j.watres.2023.120615, 2023.
Zhang, J., Liu, C., and Gao, R.: Physics-guided Gaussian process for HVAC system performance prognosis, Mech. Syst. Signal Pr., 179, 109336, https://doi.org/10.1016/j.ymssp.2022.109336, 2022.
Zhang, J., Zhao, Y., Shone, F., Li, Z., Frangi, A., Xie, S., and Zhang, Z.: Physics-Informed Deep Learning for Musculoskeletal Modeling: Predicting Muscle Forces and Joint Kinematics From Surface EMG, IEEE Trans. Neural Syst. Rehabil. Eng., 31, 484–493, https://doi.org/10.1109/TNSRE.2022.3226860, 2023.
Zhang, Q., Shi, L., Holzman, M., Ye, M., Wang, Y., Carmona, F., and Zha, Y.: A dynamic data-driven method for dealing with model structural error in soil moisture data assimilation, Adv. Water Resour., 132, 103407, https://doi.org/10.1016/j.advwatres.2019.103407, 2019.
Zhang, X., Gong, J., and Xuan, F.: A physics-informed neural network for creep-fatigue life prediction of components at elevated temperatures, Eng. Fract. Mech., 258, 108130, https://doi.org/10.1016/j.engfracmech.2021.108130, 2021.
Short summary
Computer models that predict how pollutants spread through groundwater often contain errors, making forecasts unreliable. To fix this, we developed a new method that combines data-driven learning with physical laws. Our approach uses multiple physical constraints simultaneously, and dynamically adjusts their importance during calculations. This works because the extra constraints act like having additional measurements, helping the model better identify key parameters and reduce uncertainty.
Computer models that predict how pollutants spread through groundwater often contain errors,...