Articles | Volume 19, issue 21
https://doi.org/10.5194/gmd-19-10719-2026
© 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/gmd-19-10719-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Development and technical paper
| 
11 Nov 2026
Development and technical paper |
| 11 Nov 2026
| 11 Nov 2026
A hybrid physics–ML framework for integrating groundwater dynamics into land surface modeling: a ParFlow-based demonstration in headwater watersheds of the Pearl River Basin, China
Chen Yang
CORRESPONDING AUTHOR
School of Atmospheric Sciences, Sun Yat-sen University, Zhuhai, China
Hui Huang
School of Atmospheric Sciences, Sun Yat-sen University, Zhuhai, China
Zeyu Tang
School of Atmospheric Sciences, Sun Yat-sen University, Zhuhai, China
Yongjiu Dai
School of Atmospheric Sciences, Sun Yat-sen University, Zhuhai, China
Xiaobin Chang
School of Artificial Intelligence, Sun Yat-sen University, Zhuhai, China
Related authors
Aoqi Sun, Zeyu Tang, Chunmiao Zheng, Wei Shangguan, and Chen Yang
EGUsphere, https://doi.org/10.31223/X5HZ2G, https://doi.org/10.31223/X5HZ2G, 2026
This preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).
Short summary
Short summary
Groundwater can shape exchanges between land and the atmosphere, but representing it in large-scale models is costly. We tested whether a simple indicator could identify where groundwater matters by comparing it with detailed computer simulations. The indicator gave stable results across the whole region but often changed at individual locations and depended strongly on how easily water moves underground. It is useful for broad screening, but local results should be interpreted cautiously.
Aoqi Sun, Wenjie Xu, Enze Ma, Hua Yuan, and Chen Yang
Hydrol. Earth Syst. Sci., 30, 5215–5227, https://doi.org/10.5194/hess-30-5215-2026, https://doi.org/10.5194/hess-30-5215-2026, 2026
Short summary
Short summary
Groundwater is often viewed as a hidden reserve that supports evapotranspiration and streamflow during dry periods. We show that sustained warming and greening can weaken this buffering role. As groundwater levels decline, links between shallow and deeper stores reorganize, reducing older groundwater inputs to streams and evapotranspiration. Over time, water cycling shifts toward shallower, faster pathways, potentially lowering system resilience and predictability under long-term climate stress.
Chen Yang, Aoqi Sun, Shupeng Zhang, Yongjiu Dai, Stefan Kollet, and Reed Maxwell
Geosci. Model Dev., 19, 1849–1866, https://doi.org/10.5194/gmd-19-1849-2026, https://doi.org/10.5194/gmd-19-1849-2026, 2026
Short summary
Short summary
Groundwater plays a key role in land–atmosphere water and energy exchange, yet it is often simplified in large-scale Earth system models. We review 20 years of efforts to couple the groundwater model ParFlow with land surface and atmospheric models, showing how groundwater dynamics shape terrestrial fluxes. We also present an updated coupling framework that enhances model performance and flexibility, and outline a modular strategy to guide future development.
Peyman Abbaszadeh, Fadji Zaouna Maina, Chen Yang, Dan Rosen, Sujay Kumar, Matthew Rodell, and Reed Maxwell
Hydrol. Earth Syst. Sci., 29, 5429–5452, https://doi.org/10.5194/hess-29-5429-2025, https://doi.org/10.5194/hess-29-5429-2025, 2025
Short summary
Short summary
To manage Earth's water resources effectively amid climate change, it is crucial to understand both surface and groundwater processes. We developed a new modeling system that combines two advanced tools, ParFlow and LIS (Land Information System)/Noah-MP, to better simulate both land surface and groundwater interactions. By testing this integrated model in the Upper Colorado River Basin, we found it improves predictions of hydrologic processes, especially in complex terrains.
Chen Yang, Zitong Jia, Wenjie Xu, Zhongwang Wei, Xiaolang Zhang, Yiguang Zou, Jeffrey McDonnell, Laura Condon, Yongjiu Dai, and Reed Maxwell
Hydrol. Earth Syst. Sci., 29, 2201–2218, https://doi.org/10.5194/hess-29-2201-2025, https://doi.org/10.5194/hess-29-2201-2025, 2025
Short summary
Short summary
We developed the first high-resolution, integrated surface water–groundwater hydrologic model of the entirety of continental China using ParFlow. The model shows good performance in terms of streamflow and water table depth when compared to global data products and observations. It is essential for water resources management and decision-making in China within a consistent framework in the changing world. It also has significant implications for similar modeling in other places in the world.
Gaosong Shi, Anne Gobin, Zhongwang Wei, Hua Yuan, Qingliang Li, Wei Shangguan, and Yongjiu Dai
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2026-770, https://doi.org/10.5194/essd-2026-770, 2026
Preprint under review for ESSD
Short summary
Short summary
We developed a new global soil dataset that maps 20 important soil properties across six depth ranges at high spatial detail. It combines soil observations from around the world with environmental information and regional soil products. Tests showed improved average prediction accuracy, while land-surface experiments demonstrated the importance of soil data for simulations.
Tingting Wu, Shupeng Zhang, Xiaofan Yang, and Yongjiu Dai
Hydrol. Earth Syst. Sci., 30, 6867–6905, https://doi.org/10.5194/hess-30-6867-2026, https://doi.org/10.5194/hess-30-6867-2026, 2026
Short summary
Short summary
We coupled a land surface model (CoLM - Common Land Model) with an analysis tool (PSUADE – Problem Solving environment for Uncertainty Analysis and Design Exploration) to evaluate processes controlling water table depth (WTD) through sensitivity analysis and parameter optimization. A small set of sensitive parameters governing WTD was identified, showing subsurface runoff controls average groundwater levels, while soil properties govern variability. A stepwise calibration was then suggested to improve accuracy. This framework offers a transferable pathway to enhance groundwater simulation.
Feini Huang, Ziqian Li, Lijun Wang, Gustau Camps-Valls, Wei Su, Maria Piles, Ruqing Zhang, Wei Shangguan, and Yongjiu Dai
EGUsphere, https://doi.org/10.5194/egusphere-2026-5166, https://doi.org/10.5194/egusphere-2026-5166, 2026
This preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).
Short summary
Short summary
Observation-driven 1–14-day soil moisture forecasts remain reliable in some places but lose accuracy rapidly in others. We show that this contrast is governed by how soils filter rainfall. Under intense rainfall, low-silt soils provide little buffering, accelerate downward water movement, erode moisture memory, and constrain forecast skill. This finding shifts attention from better algorithms to physical forecast limits and guides the design of reliable models and monitoring networks.
Hongbin Liang, Shulei Zhang, Omarjan Obulkasim, Qiancheng Ye, Linyi Song, Zhongwang Wei, and Yongjiu Dai
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2026-694, https://doi.org/10.5194/essd-2026-694, 2026
Preprint under review for ESSD
Short summary
Short summary
China has built the world’s largest reservoir system, yet modeling of its climate and environmental impacts lacks attribute values. We developed a database of 5,142 reservoirs across China with over 20 key attributes, covering 87.5 % of total storage and 62 % of hydropower capacity. It enables more realistic identification of reservoir-served irrigation areas and improves streamflow simulations, offering a valuable resource for hydrology, agriculture, energy, climate, and environmental research.
Zhilong Fan, Lu Li, Zhongwang Wei, Shupeng Zhang, Sisi Chen, Xingjie Lu, Wei Shangguan, Nan Wei, Hua Yuan, Pan Zhang, Jiaxin Tian, Jianhong Zhou, Baoxiang Pan, Wei Han, and Yongjiu Dai
EGUsphere, https://doi.org/10.5194/egusphere-2026-5358, https://doi.org/10.5194/egusphere-2026-5358, 2026
Short summary
Short summary
We developed a new land data assimilation framework based on deep generative models. Compared with existing methods, it better handles non-Gaussian model errors and noisy observations. Tests in both an idealized system and a realistic land data assimilation experiment showed more accurate and consistent estimates, demonstrating its potential for improving land-state estimation with complex land surface models.
Xiaoying Shi, Daniel M. Ricciuto, Yaoping Wang, Joe R. Melton, Fortunat Joos, Yiqi Luo, Xiaofeng Xu, Dafeng Hui, Qianlai Zhuang, Chunjing Qiu, Hongxing He, Siya Shao, Xingjie Lu, Yongjiu Dai, Akihiko Ito, Qing Sun, Nitin Chaudhary, Ayesha Hussain, Mousong Wu, Jian Zhou, Jingwei Zhang, Ye Yuan, Rachael H. Turton, Eleanor J. Burke, Jiafu Mao, Yuefeng Hao, Natalie A. Griffiths, Verity Salmon, Paul J. Hanson, Melanie A. Mayes, and Peter E. Thornton
EGUsphere, https://doi.org/10.5194/egusphere-2026-5193, https://doi.org/10.5194/egusphere-2026-5193, 2026
Short summary
Short summary
Northern peatlands store vast amounts of carbon, but their response to climate change remains uncertain. We compared 15 ecosystem models with observations from a long-term warming and elevated carbon dioxide experiment in Minnesota. Models differed widely in their carbon responses, reflecting differences in ecosystem processes and model parameters. The results identify priorities for improving peatland models and projections of future carbon–climate feedbacks.
Aoqi Sun, Zeyu Tang, Chunmiao Zheng, Wei Shangguan, and Chen Yang
EGUsphere, https://doi.org/10.31223/X5HZ2G, https://doi.org/10.31223/X5HZ2G, 2026
This preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).
Short summary
Short summary
Groundwater can shape exchanges between land and the atmosphere, but representing it in large-scale models is costly. We tested whether a simple indicator could identify where groundwater matters by comparing it with detailed computer simulations. The indicator gave stable results across the whole region but often changed at individual locations and depended strongly on how easily water moves underground. It is useful for broad screening, but local results should be interpreted cautiously.
Aoqi Sun, Wenjie Xu, Enze Ma, Hua Yuan, and Chen Yang
Hydrol. Earth Syst. Sci., 30, 5215–5227, https://doi.org/10.5194/hess-30-5215-2026, https://doi.org/10.5194/hess-30-5215-2026, 2026
Short summary
Short summary
Groundwater is often viewed as a hidden reserve that supports evapotranspiration and streamflow during dry periods. We show that sustained warming and greening can weaken this buffering role. As groundwater levels decline, links between shallow and deeper stores reorganize, reducing older groundwater inputs to streams and evapotranspiration. Over time, water cycling shifts toward shallower, faster pathways, potentially lowering system resilience and predictability under long-term climate stress.
Xueqi Cao, Nan Wei, Hua Yuan, Shupeng Zhang, Zhongwang Wei, Xingjie Lu, Lu Li, and Yongjiu Dai
EGUsphere, https://doi.org/10.5194/egusphere-2026-4210, https://doi.org/10.5194/egusphere-2026-4210, 2026
Short summary
Short summary
Lakes influence weather by storing and releasing heat, but models often oversimplify how sunlight travels through water, snow, and ice. We improved how a model handles light entering a lake. Tests across nine diverse lakes show surface waters warm more, deep waters warm less, and under-ice heating is better captured. Benefits vary among lakes, so better light treatment must pair with better mixing and heat storage.
Zezhen Wu, Zhongwang Wei, Xingjie Lu, Nan Wei, Lu Li, Shupeng Zhang, Hua Yuan, Shaofeng Liu, and Yongjiu Dai
Hydrol. Earth Syst. Sci., 30, 2651–2666, https://doi.org/10.5194/hess-30-2651-2026, https://doi.org/10.5194/hess-30-2651-2026, 2026
Short summary
Short summary
Land surface models simulate exchanges at the Earth's surface. Traditional evaluation methods can be misleading because they may hide errors or be overly sensitive to outliers. We introduce the Model Fidelity Metric, which measures model performance in terms of accuracy, variability and distribution similarity. Tests with synthetic data and real streamflow observations show that this metric provides more stable and informative assessments of model performance.
Chen Yang, Aoqi Sun, Shupeng Zhang, Yongjiu Dai, Stefan Kollet, and Reed Maxwell
Geosci. Model Dev., 19, 1849–1866, https://doi.org/10.5194/gmd-19-1849-2026, https://doi.org/10.5194/gmd-19-1849-2026, 2026
Short summary
Short summary
Groundwater plays a key role in land–atmosphere water and energy exchange, yet it is often simplified in large-scale Earth system models. We review 20 years of efforts to couple the groundwater model ParFlow with land surface and atmospheric models, showing how groundwater dynamics shape terrestrial fluxes. We also present an updated coupling framework that enhances model performance and flexibility, and outline a modular strategy to guide future development.
Wanyi Lin, Hua Yuan, Wenzong Dong, Zhuo Liu, Jiayi Xiang, Xinran Yu, Shupeng Zhang, Zhongwang Wei, and Yongjiu Dai
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2026-21, https://doi.org/10.5194/essd-2026-21, 2026
Manuscript not accepted for further review
Short summary
Short summary
Land surface models require information on vegetation composition and seasonal leaf area dynamics, yet long-term high-resolution datasets remain limited. We present a global dataset of plant functional types and corresponding leaf area index for 1985–2020, derived from multiple satellite observations. The dataset reduces uncertainties in vegetation representation and improves the description of phenological dynamics, supporting more realistic land surface and climate modeling.
Peyman Abbaszadeh, Fadji Zaouna Maina, Chen Yang, Dan Rosen, Sujay Kumar, Matthew Rodell, and Reed Maxwell
Hydrol. Earth Syst. Sci., 29, 5429–5452, https://doi.org/10.5194/hess-29-5429-2025, https://doi.org/10.5194/hess-29-5429-2025, 2025
Short summary
Short summary
To manage Earth's water resources effectively amid climate change, it is crucial to understand both surface and groundwater processes. We developed a new modeling system that combines two advanced tools, ParFlow and LIS (Land Information System)/Noah-MP, to better simulate both land surface and groundwater interactions. By testing this integrated model in the Upper Colorado River Basin, we found it improves predictions of hydrologic processes, especially in complex terrains.
Zhongwang Wei, Qingchen Xu, Fan Bai, Xionghui Xu, Zixin Wei, Wenzong Dong, Hongbin Liang, Nan Wei, Xingjie Lu, Lu Li, Shupeng Zhang, Hua Yuan, Laibao Liu, and Yongjiu Dai
Geosci. Model Dev., 18, 6517–6540, https://doi.org/10.5194/gmd-18-6517-2025, https://doi.org/10.5194/gmd-18-6517-2025, 2025
Short summary
Short summary
Land surface models are used for simulating how Earth's surface interacts with the atmosphere. As models grow more complex and detailed, researchers need better tools to evaluate their performance. OpenBench, a new software system that makes the evaluation process more comprehensive and efficient, stands out by incorporating various factors and working with data at any scale, enabling scientists to incorporate new types of models and measurements as our understanding of Earth's systems evolves.
Shuyang Guo, Yongjiu Dai, Hua Yuan, and Hongbin Liang
The Cryosphere, 19, 3553–3570, https://doi.org/10.5194/tc-19-3553-2025, https://doi.org/10.5194/tc-19-3553-2025, 2025
Short summary
Short summary
The Snow, Ice, and Aerosol Radiation Model version 4 has only been used to evaluate bare-ice albedo in land surface models, with necessary ice property data lacking quality control. We integrated this model into our land surface model and improved bare-ice properties using quality-controlled satellite data. Our findings show regional warming and reduced snow cover in Greenland’s bare-ice region, driven by changes in bare-ice properties through bare-ice–snow albedo feedback.
Shulei Zhang, Hongbin Liang, Fang Li, Xingjie Lu, and Yongjiu Dai
Hydrol. Earth Syst. Sci., 29, 3119–3143, https://doi.org/10.5194/hess-29-3119-2025, https://doi.org/10.5194/hess-29-3119-2025, 2025
Short summary
Short summary
This study enhances irrigation modeling in the Common Land Model by capturing the full irrigation process, detailing water supplies from various sources, and enabling bidirectional coupling between water demand and supply. The proposed model accurately simulates irrigation water withdrawals, energy fluxes, river flow, and crop yields. It offers insights into irrigation-related climate impacts and water scarcity, contributing to sustainable water management and improved Earth system modeling.
Chen Yang, Zitong Jia, Wenjie Xu, Zhongwang Wei, Xiaolang Zhang, Yiguang Zou, Jeffrey McDonnell, Laura Condon, Yongjiu Dai, and Reed Maxwell
Hydrol. Earth Syst. Sci., 29, 2201–2218, https://doi.org/10.5194/hess-29-2201-2025, https://doi.org/10.5194/hess-29-2201-2025, 2025
Short summary
Short summary
We developed the first high-resolution, integrated surface water–groundwater hydrologic model of the entirety of continental China using ParFlow. The model shows good performance in terms of streamflow and water table depth when compared to global data products and observations. It is essential for water resources management and decision-making in China within a consistent framework in the changing world. It also has significant implications for similar modeling in other places in the world.
Gaosong Shi, Wenye Sun, Wei Shangguan, Zhongwang Wei, Hua Yuan, Lu Li, Xiaolin Sun, Ye Zhang, Hongbin Liang, Danxi Li, Feini Huang, Qingliang Li, and Yongjiu Dai
Earth Syst. Sci. Data, 17, 517–543, https://doi.org/10.5194/essd-17-517-2025, https://doi.org/10.5194/essd-17-517-2025, 2025
Short summary
Short summary
In this study, we developed the second version of China's high-resolution soil information grid using legacy soil samples and advanced machine learning. This version predicts over 20 soil properties at six depths, providing accurate soil variation maps across China. It outperforms previous versions and global products, offering valuable data for hydrological and ecological analyses and Earth system modelling, enhancing our understanding of soil roles in environmental processes.
Jiahao Shi, Hua Yuan, Wanyi Lin, Wenzong Dong, Hongbin Liang, Zhuo Liu, Jianxin Zeng, Haolin Zhang, Nan Wei, Zhongwang Wei, Shupeng Zhang, Shaofeng Liu, Xingjie Lu, and Yongjiu Dai
Earth Syst. Sci. Data, 17, 117–134, https://doi.org/10.5194/essd-17-117-2025, https://doi.org/10.5194/essd-17-117-2025, 2025
Short summary
Short summary
Flux tower data are widely recognized as benchmarking data for land surface models, but insufficient emphasis on and deficiency in site attribute data limits their true value. We collect site-observed vegetation, soil, and topography data from various sources. The final dataset encompasses 90 sites globally, with relatively complete site attribute data and high-quality flux validation data. This work has provided more reliable site attribute data, benefiting land surface model development.
Yangzi Che, Xuecao Li, Xiaoping Liu, Yuhao Wang, Weilin Liao, Xianwei Zheng, Xucai Zhang, Xiaocong Xu, Qian Shi, Jiajun Zhu, Honghui Zhang, Hua Yuan, and Yongjiu Dai
Earth Syst. Sci. Data, 16, 5357–5374, https://doi.org/10.5194/essd-16-5357-2024, https://doi.org/10.5194/essd-16-5357-2024, 2024
Short summary
Short summary
Most existing building height products are limited with respect to either spatial resolution or coverage, not to mention the spatial heterogeneity introduced by global building forms. Using Earth Observation (EO) datasets for 2020, we developed a global height dataset at the individual building scale. The dataset provides spatially explicit information on 3D building morphology, supporting both macro- and microanalysis of urban areas.
Qingliang Li, Gaosong Shi, Wei Shangguan, Vahid Nourani, Jianduo Li, Lu Li, Feini Huang, Ye Zhang, Chunyan Wang, Dagang Wang, Jianxiu Qiu, Xingjie Lu, and Yongjiu Dai
Earth Syst. Sci. Data, 14, 5267–5286, https://doi.org/10.5194/essd-14-5267-2022, https://doi.org/10.5194/essd-14-5267-2022, 2022
Short summary
Short summary
SMCI1.0 is a 1 km resolution dataset of daily soil moisture over China for 2000–2020 derived through machine learning trained with in situ measurements of 1789 stations, meteorological forcings, and land surface variables. It contains 10 soil layers with 10 cm intervals up to 100 cm deep. Evaluated by in situ data, the error (ubRMSE) ranges from 0.045 to 0.051, and the correlation (R) range is 0.866-0.893. Compared with ERA5-Land, SMAP-L4, and SoMo.ml, SIMI1.0 has higher accuracy and resolution.
Ziqi Lin, Yongjiu Dai, Umakant Mishra, Guocheng Wang, Wei Shangguan, Wen Zhang, and Zhangcai Qin
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2022-232, https://doi.org/10.5194/essd-2022-232, 2022
Manuscript not accepted for further review
Short summary
Short summary
Spatial soil organic carbon (SOC) data is critical for predictions in carbon climate feedbacks and future climate trends, but no conclusion has yet been reached on which dataset to be used for specific purposes. We evaluated the SOC estimates from five widely used global soil datasets and a regional permafrost dataset, and identify uncertainties of SOC estimates by region, biome, and data sources, hoping to help improve SOC/soil data in the future.
Cited articles
Bennett, A., Tran, H., De la Fuente, L., Triplett, A., Ma, Y., Melchior, P., Maxwell, R. M., and Condon, L. E.: Spatio-Temporal Machine Learning for Regional to Continental Scale Terrestrial Hydrology, J. Adv. Model. Earth Sy., 16, e2023MS004095, https://doi.org/10.1029/2023ms004095, 2024.
Chapman, W. E., Schreck, J. S., Sha, Y., Gagne Ii, D. J., Kimpara, D., Zanna, L., Mayer, K. J., and Berner, J.: CAMulator: Fast Emulation of the Community Atmosphere Model, arXiv [preprint], https://doi.org/10.48550/arXiv.2504.06007, 2025.
Chen, M., Qian, Z., Boers, N., Jakeman, A. J., Kettner, A. J., Brandt, M., Kwan, M.-P., Batty, M., Li, W., Zhu, R., Luo, W., Ames, D. P., Barton, C. M., Cuddy, S. M., Koirala, S., Zhang, F., Ratti, C., Liu, J., Zhong, T., Liu, J., Wen, Y., Yue, S., Zhu, Z., Zhang, Z., Sun, Z., Lin, J., Ma, Z., He, Y., Xu, K., Zhang, C., Lin, H., and Lü, G.: Iterative integration of deep learning in hybrid Earth surface system modelling, Nat. Rev. Earth Environ., 4, 568–581, https://doi.org/10.1038/s43017-023-00452-7, 2023.
Chen, S., Li, L., Wei, Z., Wei, N., Zhang, Y., Zhang, S., Yuan, H., Shangguan, W., Zhang, S., Li, Q., and Dai, Y.: Exploring Topography Downscaling Methods for Hyper-Resolution Land Surface Modeling, J. Geophys. Res.-Atmos., 129, e2024JD041338, https://doi.org/10.1029/2024jd041338, 2024.
Christopoulos, C., Lopez-Gomez, I., Beucler, T., Cohen, Y., Kawczynski, C., Dunbar, O. R. A., and Schneider, T.: Online Learning of Entrainment Closures in a Hybrid Machine Learning Parameterization, J. Adv. Model. Earth Sy., 16, e2024MS004485, https://doi.org/10.1029/2024ms004485, 2024.
Clark, S. K., Watt-Meyer, O., Kwa, A., McGibbon, J., Henn, B., Perkins, W. A., Wu, E., Bretherton, C. S., and Harris, L. M.: ACE2-SOM: Coupling to a slab ocean and learning the sensitivity of climate to changes in CO2, arXiv [preprint], https://doi.org/10.48550/arXiv.2412.04418, 2024.
Cremer, C. J. M., Mariegaard, J. S., and Andersson, H. J.: A hybrid data assimilation and machine learning approach for enhancing operational forecasting in 2D hydrodynamic models, J. Hydroinform., 27, 493–507, https://doi.org/10.2166/hydro.2025.289, 2025.
Cresswell-Clay, N., Liu, B., Durran, D. R., Liu, Z., Espinosa, Z. I., Moreno, R. A., and Karlbauer, M.: A Deep Learning Earth System Model for Efficient Simulation of the Observed Climate, AGU Adv., 6, e2025AV001706, https://doi.org/10.1029/2025av001706, 2025.
Dai, T., Maher, K., and Perzan, Z.: Machine learning surrogates for efficient hydrologic modeling: Insights from stochastic simulations of managed aquifer recharge, J. Hydrol., 652, 132606, https://doi.org/10.1016/j.jhydrol.2024.132606, 2025.
Duncan, J. P. C., Wu, E., Dheeshjith, S., Subel, A., Arcomano, T., Clark, S. K., Henn, B., Kwa, A., McGibbon, J., Perkins, W. A., Gregory, W., Fernandez-Granda, C., Busecke, J., Watt-Meyer, O., Hurlin, W. J., Adcroft, A., Zanna, L., and Bretherton, C.: SamudrACE: Fast and Accurate Coupled Climate Modeling with 3D Ocean and Atmosphere Emulators, arXiv [preprint], https://doi.org/10.48550/arXiv.2509.12490, 2025.
Eilander, D., van Verseveld, W., Yamazaki, D., Weerts, A., Winsemius, H. C., and Ward, P. J.: A hydrography upscaling method for scale-invariant parametrization of distributed hydrological models, Hydrol. Earth Syst. Sci., 25, 5287–5313, https://doi.org/10.5194/hess-25-5287-2021, 2021.
Fan, Y., Miguez-Macho, G., Weaver, C. P., Walko, R., and Robock, A.: Incorporating water table dynamics in climate modeling: 1. Water table observations and equilibrium water table simulations, J. Geophys. Res.-Atmos., 112, D10125, https://doi.org/10.1029/2006jd008111, 2007.
Friedl, M. and Sulla-Menashe, D.: MODIS/Terra+Aqua Land Cover Type Yearly L3 Global 500m SIN Grid V061 (6.1), NASA Land Processes Distributed Active Archive Center [data set], https://doi.org/10.5067/modis/mcd12q1.061, 2022.
Gleeson, T., Smith, L., Moosdorf, N., Hartmann, J., Dürr, H. H., Manning, A. H., van Beek, L. P. H., and Jellinek, A. M.: Mapping permeability over the surface of the Earth, Geophys. Res. Lett., 38, L02401, https://doi.org/10.1029/2010gl045565, 2011.
Huang, H., Sun, A., Tran, H., Dai, Y., and Yang, C.: A Transformer-Based Surrogate for Distributed Groundwater Dynamics with Zero-Shot Cross-Basin Transfer and Precipitation-Regime Data Augmentation, Earth ArXiv [preprint], https://doi.org/10.31223/X5QV3P, 2026.
Kollet, S. J. and Maxwell, R. M.: Capturing the influence of groundwater dynamics on land surface processes using an integrated, distributed watershed model, Water Resour. Res., 44, W02402, https://doi.org/10.1029/2007wr006004, 2008.
Lin, J., Yu, S., Peng, L., Beucler, T., Wong-Toi, E., Hu, Z., Gentine, P., Geleta, M., and Pritchard, M.: Sampling Hybrid Climate Simulation at Scale to Reliably Improve Machine Learning Parameterization, arXiv [preprint], https://doi.org/10.48550/arXiv.2309.16177, 2023.
Liu, H.-Y., Satoh, M., Gu, J.-F., Lei, L., Tang, J., Tan, Z.-M., Wang, Y., and Xu, J.: Predictability of the Most Long-Lived Tropical Cyclone Freddy (2023) During Its Westward Journey Through the Southern Tropical Indian Ocean, Geophys. Res. Lett., 50, e2023GL105729, https://doi.org/10.1029/2023gl105729, 2023.
Liu, H.-Y., Tan, Z.-M., Wang, Y., Tang, J., Satoh, M., Lei, L., Gu, J.-F., Zhang, Y., Nie, G.-Z., and Chen, Q.-Z.: A Hybrid Machine Learning/Physics-Based Modeling Framework for 2-Week Extended Prediction of Tropical Cyclones, J. Geophys. Res.-Mach. Learn., 1, e2024JH000207, https://doi.org/10.1029/2024jh000207, 2024.
Liu, Y., Huang, H., Wang, S.-C., Zhang, T., Xu, D., and Chen, Y.: ELM2.1-XGBfire1.0: improving wildfire prediction by integrating a machine learning fire model in a land surface model, Geosci. Model Dev., 18, 4103–4117, https://doi.org/10.5194/gmd-18-4103-2025, 2025.
Maxwell, R. M. and Miller, N. L.: Development of a Coupled Land Surface and Groundwater Model, J. Hydrometeorol., 6, 233–247, https://doi.org/10.1175/jhm422.1, 2005.
Maxwell, R. M., Condon, L. E., and Melchior, P.: A Physics-Informed, Machine Learning Emulator of a 2D Surface Water Model: What Temporal Networks and Simulation-Based Inference Can Help Us Learn about Hydrologic Processes, Water, 13, 3633, https://doi.org/10.3390/w13243633, 2021.
Muñoz-Sabater, J., Dutra, E., Agustí-Panareda, A., Albergel, C., Arduini, G., Balsamo, G., Boussetta, S., Choulga, M., Harrigan, S., Hersbach, H., Martens, B., Miralles, D. G., Piles, M., Rodríguez-Fernández, N. J., Zsoter, E., Buontempo, C., and Thépaut, J.-N.: ERA5-Land: a state-of-the-art global reanalysis dataset for land applications, Earth Syst. Sci. Data, 13, 4349–4383, https://doi.org/10.5194/essd-13-4349-2021, 2021.
Niu, Z., Wang, D., Mu, M., Huang, W., Fan, X., Yang, M., and Qin, B.: Machine-Learning (ML)-Physics Fusion Model Accelerates the Paradigm Shift in Typhoon Forecasting With a CNOP-Based Assimilation Framework, Geophys. Res. Lett., 52, e2025GL115926, https://doi.org/10.1029/2025gl115926, 2025.
O'Neill, M. M. F., Tijerina, D. T., Condon, L. E., and Maxwell, R. M.: Assessment of the ParFlow–CLM CONUS 1.0 integrated hydrologic model: evaluation of hyper-resolution water balance components across the contiguous United States, Geosci. Model Dev., 14, 7223–7254, https://doi.org/10.5194/gmd-14-7223-2021, 2021.
Shangguan, W., Dai, Y., Duan, Q., Liu, B., and Yuan, H.: A global soil data set for earth system modeling, J. Adv. Model. Earth Sy., 6, 249–263, https://doi.org/10.1002/2013ms000293, 2014.
Tran, H., Leonarduzzi, E., De la Fuente, L., Hull, R. B., Bansal, V., Chennault, C., Gentine, P., Melchior, P., Condon, L. E., and Maxwell, R. M.: Development of a Deep Learning Emulator for a Distributed Groundwater–Surface Water Model: ParFlow-ML, Water, 13, 3393, https://doi.org/10.3390/w13233393, 2021.
van Genuchten, M. T.: 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.
Wang, Y., Wu, H., Zhang, J., Gao, Z., Wang, J., Yu, P. S., and Long, M.: PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning, IEEE T. Pattern Anal., 45, 2208–2225, https://doi.org/10.1109/tpami.2022.3165153, 2023.
Watt-Meyer, O., Dresdner, G., McGibbon, J., Clark, S. K., Henn, B., Duncan, J., Brenowitz, N. D., Kashinath, K., Pritchard, M. S., Bonev, B., Peters, M. E., and Bretherton, C. S.: ACE: A fast, skillful learned global atmospheric model for climate prediction, arXiv [preprint], https://doi.org/10.48550/arXiv.2310.02074, 2023.
Yang, C.: ParFlow-CoLM, Zenodo [code], https://doi.org/10.5281/zenodo.16879407, 2025.
Yang, C.: Data for: A hybrid physics–ML framework for integrating groundwater dynamics into land surface modeling (2), Science Data Bank [data set], https://doi.org/10.57760/sciencedb.41621, 2026.
Yang, C., Li, H.-Y., Fang, Y., Cui, C., Wang, T., Zheng, C., Leung, L. R., Maxwell, R. M., Zhang, Y.-K., and Yang, X.: Effects of Groundwater Pumping on Ground Surface Temperature: A Regional Modeling Study in the North China Plain, J. Geophys. Res.-Atmos., 125, e2019JD031764, https://doi.org/10.1029/2019jd031764, 2020.
Yang, C., Jia, Z., Xu, W., Wei, Z., Zhang, X., Zou, Y., McDonnell, J., Condon, L., Dai, Y., and Maxwell, R.: CONCN: a high-resolution, integrated surface water–groundwater ParFlow modeling platform of continental China, Hydrol. Earth Syst. Sci., 29, 2201–2218, https://doi.org/10.5194/hess-29-2201-2025, 2025a.
Yang, C., Huang, H., Tang, Z., and Chang, X.: A Hybrid Physics–ML Framework for Integrating Groundwater Dynamics into Land Surface Modeling, Zenodo [code], https://doi.org/10.5281/zenodo.20791904, 2026a.
Yang, C., Sun, A., Zhang, S., Dai, Y., Kollet, S., and Maxwell, R.: 20 years of trials and insights: bridging legacy and next generation in ParFlow and Land Surface Model Coupling, Geosci. Model Dev., 19, 1849–1866, https://doi.org/10.5194/gmd-19-1849-2026, 2026b.
Yang, Y., Feng, D., Beck, H. E., Hu, W., Abbas, A., Sengupta, A., Delle Monache, L., Hartman, R., Lin, P., Shen, C., and Pan, M.: Global Daily Discharge Estimation Based on Grid Long Short-Term Memory (LSTM) Model and River Routing, Water Resour. Res., 61, e2024WR039764, https://doi.org/10.1029/2024wr039764, 2025b.
Yu, S., Hannah, W., Peng, L., Lin, J., Bhouri, M. A., Gupta, R., Lütjens, B., Will, J. C., Behrens, G., Busecke, J., Loose, N., Stern, C., Beucler, T., Harrop, B., Hillman, B., Jenney, A., Ferretti, S. L., Liu, N., Anandkumar, A., Brenowitz, N., Eyring, V., Geneva, N., Gentine, P., Mandt, S., Pathak, J., Subramaniam, A., Vondrick, C., Yu, R., Zanna, L., Zheng, T., Abernathey, R., Ahmed, F., Bader, D., Baldi, P., Barnes, E., Bretherton, C., Caldwell, P., Chuang, W., Han, Y., Huang, Y., Iglesias-Suarez, F., Jantre, S., Kashinath, K., Khairoutdinov, M., Kurth, T., Lutsko, N., Ma, P.-L., Mooers, G., Neelin, J. D., Randall, D., Shamekh, S., Taylor, M., Urban, N., Yuval, J., Zhang, G., and Pritchard, M.: ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation, Adv. Neur. In., 22070–22084, https://doi.org/10.52202/075280-0968, 2023.
Yu, S., Hu, Z., Subramaniam, A., Hannah, W., Peng, L., Lin, J., Bhouri, M. A., Gupta, R., Lütjens, B., Will, J. C., Behrens, G., Busecke, J. J. M., Loose, N., Stern, C. I., Beucler, T., Harrop, B., Heuer, H., Hillman, B. R., Jenney, A., Liu, N., White, A., Zheng, T., Kuang, Z., Ahmed, F., Barnes, E., Brenowitz, N. D., Bretherton, C., Eyring, V., Ferretti, S., Lutsko, N., Gentine, P., Mandt, S., Neelin, J. D., Yu, R., Zanna, L., Urban, N. M., Yuval, J., Abernathey, R., Baldi, P., Chuang, W., Huang, Y., Iglesias-Suarez, F., Jantre, S., Ma, P.-L., Shamekh, S., Zhang, G., and Pritchard, M.: ClimSim-Online: A Large Multi-Scale Dataset and Framework for Hybrid Physics-ML Climate Emulation, J. Mach. Learn. Res., 26, 1–85, 2025.
Yuan, H. and Dai, Y.: Description of the Common Land Model (CoLM 2024), Sun Yat-sen University Press, Guangzhou, China, ISBN 978-7-306-08654-9, 2025.
Short summary
Groundwater strongly influences exchanges of water and energy between land and atmosphere, but detailed simulation is computationally expensive. We developed a faster approach that preserves groundwater–land surface interactions. In a representative watershed within China's Pearl River Basin, it ran 18.24 times faster while closely matching the physics-based model, offering a practical way to represent groundwater in large-scale environmental modeling.
Groundwater strongly influences exchanges of water and energy between land and atmosphere, but...
Altmetrics
Final-revised paper
Preprint