Articles | Volume 27, issue 14
https://doi.org/10.5194/hess-27-2579-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Special issue:
https://doi.org/10.5194/hess-27-2579-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
17 Jul 2023
Research article |
| 17 Jul 2023
| 17 Jul 2023
Physics-informed machine learning for understanding rock moisture dynamics in a sandstone cave
Kai-Gao Ouyang
MWR Key Laboratory of Groundwater Conservation, China University of Geosciences, Beijing, China
MWR Key Laboratory of Groundwater Conservation, China University of Geosciences, Beijing, China
MNR Key Laboratory of Shallow Geothermal Energy, China University of Geosciences, Beijing, China
Gang Mei
School of Engineering and Technology, China University of Geosciences, Beijing, China
Hong-Bin Yan
Yungang Research Institute, Datong, Shanxi, China
Ran Niu
MWR Key Laboratory of Groundwater Conservation, China University of Geosciences, Beijing, China
Li Wan
MWR Key Laboratory of Groundwater Conservation, China University of Geosciences, Beijing, China
Yijian Zeng
Department of Water Resources, ITC Faculty of Geo-Information Science and Earth Observation, University of Twente, Enschede, the Netherlands
Related authors
No articles found.
Shijie Jiang, Georgios Blougouras, Tuya Wulan, Yijian Zeng, and Jialiang Zhou
EGUsphere, https://doi.org/10.5194/egusphere-2026-3294, https://doi.org/10.5194/egusphere-2026-3294, 2026
This preprint is open for discussion and under review for Geoscientific Model Development (GMD).
Short summary
Short summary
Land surface models simulate how plants and soils exchange water and carbon with the atmosphere, but they rely on many parameters that are hard to measure. We built a model that combines physics with machine learning so these parameters can be learned from observations, checked for reliability, and mapped from a few measurement sites across whole regions. Over Europe it improved predictions and produced coherent parameter maps, offering a foundation for land models that learn from data.
Matthias Cuntz, Stephan Thober, Anne Verhoef, Yijian Zeng, Aaron Boone, Agnès Ducharne, Sujan Koirala, David M. Lawrence, Philipp de Vrese, Carsten Montzka, Sonia I. Seneviratne, Harry Vereecken, Dani Or, Salma Tafasca, Naoki Mizukami, Rich Ellis, Patrick C. McGuire, and Lukas Gudmundsson
EGUsphere, https://doi.org/10.5194/egusphere-2026-4223, https://doi.org/10.5194/egusphere-2026-4223, 2026
Preprint archived
Short summary
Short summary
Land surface models (LSMs) are essential tools to analyze future climate change. Here we pushed state-of-the-art LSMs out of their comfort zone revealing hidden connections between processes encoded into the models. The LSMs reacted differently due to complex, sometimes counter-intuitive interactions of different water fluxes. We identified surface runoff and soil evaporation as the two main reasons behind model differences, which should be improved by using more mechanistic descriptions.
Pan Jiang, Zhengjing Ma, and Gang Mei
Nat. Hazards Earth Syst. Sci., 26, 487–529, https://doi.org/10.5194/nhess-26-487-2026, https://doi.org/10.5194/nhess-26-487-2026, 2026
Short summary
Short summary
In order to elucidate the potential for integrating deep learning with potential landslide identification, this paper focuses on four key dimensions: (1) Summarising data sources for potential landslide identification. (2) Compare the roles of commonly used deep learning models. (3) Analyse the practical applications of deep learning in early landslide detection. (4) Investigate key challenges and propose future priorities for potential landslide identification.
Qianqian Han, Yijian Zeng, Yunfei Wang, Fakhereh Alidoost, Francesco Nattino, Yang Liu, and Bob Su
Earth Syst. Sci. Data, 17, 7101–7117, https://doi.org/10.5194/essd-17-7101-2025, https://doi.org/10.5194/essd-17-7101-2025, 2025
Short summary
Short summary
Understanding how land interacts with the atmosphere is crucial for studying climate change, yet global high-resolution data on energy, water, and carbon exchanges remain limited. This study introduces a new dataset FluxHourly that estimates these exchanges hourly from 2000 to 2020 by combining physical process model, field measurements, and machine learning with satellite and meteorological data. Fluxhourly enables analysis of ecosystem responses to climate extremes.
Mostafa Gomaa Daoud, Fakhereh Alidoost, Yijian Zeng, Bart Schilperoort, Christiaan Van der Tol, Maciek W. Lubczynski, Mhd Suhyb Salama, Eric D. Morway, Christian D. Langevin, Prajwal Khanal, Zengjing Song, Lianyu Yu, Hong Zhao, Gualbert Oude Essink, Victor F. Bense, Michiel van der Molen, and Zhongbo Su
EGUsphere, https://doi.org/10.5194/egusphere-2025-4179, https://doi.org/10.5194/egusphere-2025-4179, 2025
Short summary
Short summary
This study investigates the groundwater role in soil-plant-atmosphere continuum. An integrated ecohydrological modelling approach was developed by coupling STEMMUS-SCOPE to MODFLOW 6 and applied at three sites over 8 years. The coupled model improved simulations of soil moisture and temperature, evapotranspiration, carbon fluxes and fluorescence. The findings highlight the groundwater critical role in ecosystem dynamics and its contribution to advancing water, energy and carbon cycle modelling.
Paolo Nasta, Günter Blöschl, Heye R. Bogena, Steffen Zacharias, Roland Baatz, Gabriëlle De Lannoy, Karsten H. Jensen, Salvatore Manfreda, Laurent Pfister, Ana M. Tarquis, Ilja van Meerveld, Marc Voltz, Yijian Zeng, William Kustas, Xin Li, Harry Vereecken, and Nunzio Romano
Hydrol. Earth Syst. Sci., 29, 465–483, https://doi.org/10.5194/hess-29-465-2025, https://doi.org/10.5194/hess-29-465-2025, 2025
Short summary
Short summary
The Unsolved Problems in Hydrology (UPH) initiative has emphasized the need to establish networks of multi-decadal hydrological observatories to tackle catchment-scale challenges on a global scale. This opinion paper provocatively discusses two endmembers of possible future hydrological observatory (HO) networks for a given hypothesized community budget: a comprehensive set of moderately instrumented observatories or, alternatively, a small number of highly instrumented supersites.
Gab Abramowitz, Anna Ukkola, Sanaa Hobeichi, Jon Cranko Page, Mathew Lipson, Martin G. De Kauwe, Samuel Green, Claire Brenner, Jonathan Frame, Grey Nearing, Martyn Clark, Martin Best, Peter Anthoni, Gabriele Arduini, Souhail Boussetta, Silvia Caldararu, Kyeungwoo Cho, Matthias Cuntz, David Fairbairn, Craig R. Ferguson, Hyungjun Kim, Yeonjoo Kim, Jürgen Knauer, David Lawrence, Xiangzhong Luo, Sergey Malyshev, Tomoko Nitta, Jerome Ogee, Keith Oleson, Catherine Ottlé, Phillipe Peylin, Patricia de Rosnay, Heather Rumbold, Bob Su, Nicolas Vuichard, Anthony P. Walker, Xiaoni Wang-Faivre, Yunfei Wang, and Yijian Zeng
Biogeosciences, 21, 5517–5538, https://doi.org/10.5194/bg-21-5517-2024, https://doi.org/10.5194/bg-21-5517-2024, 2024
Short summary
Short summary
This paper evaluates land models – computer-based models that simulate ecosystem dynamics; land carbon, water, and energy cycles; and the role of land in the climate system. It uses machine learning and AI approaches to show that, despite the complexity of land models, they do not perform nearly as well as they could given the amount of information they are provided with about the prediction problem.
Zengjing Song, Yijian Zeng, Yunfei Wang, Enting Tang, Danyang Yu, Fakhereh Alidoost, Mingguo Ma, Xujun Han, Xuguang Tang, Zhongjing Zhu, Yao Xiao, Debing Kong, and Zhongbo Su
EGUsphere, https://doi.org/10.5194/egusphere-2024-2940, https://doi.org/10.5194/egusphere-2024-2940, 2024
Preprint archived
Short summary
Short summary
The exchange of water and carbon between the plant and the atmosphere is affected under water stress conditions. In this study, a leaf-water-potential-based water stress factor is considered in the STEMMUS-SCOPE (hereafter STEMMUS-SCOPE-PHS), to replace the conventional soil-moisture-based water stress factor. The results show that leaf water potential reflects the plant water stress well, and the STEMMUS-SCOPE-PHS outperforms STEMMUS-SCOPE in the dynamics of the water, energy and carbon fluxes.
Tobias Karl David Weber, Lutz Weihermüller, Attila Nemes, Michel Bechtold, Aurore Degré, Efstathios Diamantopoulos, Simone Fatichi, Vilim Filipović, Surya Gupta, Tobias L. Hohenbrink, Daniel R. Hirmas, Conrad Jackisch, Quirijn de Jong van Lier, John Koestel, Peter Lehmann, Toby R. Marthews, Budiman Minasny, Holger Pagel, Martine van der Ploeg, Shahab Aldin Shojaeezadeh, Simon Fiil Svane, Brigitta Szabó, Harry Vereecken, Anne Verhoef, Michael Young, Yijian Zeng, Yonggen Zhang, and Sara Bonetti
Hydrol. Earth Syst. Sci., 28, 3391–3433, https://doi.org/10.5194/hess-28-3391-2024, https://doi.org/10.5194/hess-28-3391-2024, 2024
Short summary
Short summary
Pedotransfer functions (PTFs) are used to predict parameters of models describing the hydraulic properties of soils. The appropriateness of these predictions critically relies on the nature of the datasets for training the PTFs and the physical comprehensiveness of the models. This roadmap paper is addressed to PTF developers and users and critically reflects the utility and future of PTFs. To this end, we present a manifesto aiming at a paradigm shift in PTF research.
Yunfei Wang, Yijian Zeng, Zengjing Song, Danyang Yu, Qianqian Han, Enting Tang, Henk de Bruin, and Zhongbo Su
EGUsphere, https://doi.org/10.5194/egusphere-2024-1321, https://doi.org/10.5194/egusphere-2024-1321, 2024
Preprint archived
Short summary
Short summary
Various methods were proposed to estimate irrigation water requirements (IWR). However, the simulated IWR exhibits large differences. This study evaluates six potential evapotranspiration (PET) methods and proposes a practical approach to estimate IWR. The radiation-based methods show promise in approximating daily PET accurately, and the STEMMUS-SCOPE model can reliably estimate IWR. This research enhances our understanding of different PET methods and their implications for water management.
Enting Tang, Yijian Zeng, Yunfei Wang, Zengjing Song, Danyang Yu, Hongyue Wu, Chenglong Qiao, Christiaan van der Tol, Lingtong Du, and Zhongbo Su
Biogeosciences, 21, 893–909, https://doi.org/10.5194/bg-21-893-2024, https://doi.org/10.5194/bg-21-893-2024, 2024
Short summary
Short summary
Our study shows that planting shrubs in a semiarid grassland reduced the soil moisture and increased plant water uptake and transpiration. Notably, the water used by the ecosystem exceeded the rainfall received during the growing seasons, indicating an imbalance in the water cycle. The findings demonstrate the effectiveness of the STEMMUS–SCOPE model as a tool to represent ecohydrological processes and highlight the need to consider energy and water budgets for future revegetation projects.
Qianqian Han, Yijian Zeng, Lijie Zhang, Calimanut-Ionut Cira, Egor Prikaziuk, Ting Duan, Chao Wang, Brigitta Szabó, Salvatore Manfreda, Ruodan Zhuang, and Bob Su
Geosci. Model Dev., 16, 5825–5845, https://doi.org/10.5194/gmd-16-5825-2023, https://doi.org/10.5194/gmd-16-5825-2023, 2023
Short summary
Short summary
Using machine learning, we estimated global surface soil moisture (SSM) to aid in understanding water, energy, and carbon exchange. Ensemble models outperformed individual algorithms in predicting SSM under different climates. The best-performing ensemble included K-neighbours Regressor, Random Forest Regressor, and Extreme Gradient Boosting. This is important for hydrological and climatological applications such as water cycle monitoring, irrigation management, and crop yield prediction.
Lianyu Yu, Yijian Zeng, Huanjie Cai, Mengna Li, Yuanyuan Zha, Jicai Zeng, Hui Qian, and Zhongbo Su
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2022-221, https://doi.org/10.5194/gmd-2022-221, 2023
Revised manuscript not accepted
Short summary
Short summary
We developed a coupled soil water-groundwater (SW-GW) model, which is verified as physically accurate and applicable in large-scale groundwater problems. The role of vadose zone processes, coupling approach, and spatiotemporal heterogeneity of SW-GW interactions were highlighted as essential to represent the SW-GW system. Given the relevant dataset, the developed SW-GW modeling framework has the potential to portray the processes "from bedrock to atmosphere" in a physically consistent manner.
Pei Zhang, Donghai Zheng, Rogier van der Velde, Jun Wen, Yaoming Ma, Yijian Zeng, Xin Wang, Zuoliang Wang, Jiali Chen, and Zhongbo Su
Earth Syst. Sci. Data, 14, 5513–5542, https://doi.org/10.5194/essd-14-5513-2022, https://doi.org/10.5194/essd-14-5513-2022, 2022
Short summary
Short summary
Soil moisture and soil temperature (SMST) are important state variables for quantifying the heat–water exchange between land and atmosphere. Yet, long-term, regional-scale in situ SMST measurements at multiple depths are scarce on the Tibetan Plateau (TP). The presented dataset would be valuable for the evaluation and improvement of long-term satellite- and model-based SMST products on the TP, enhancing the understanding of TP hydrometeorological processes and their response to climate change.
Hong Zhao, Yijian Zeng, Jan G. Hofste, Ting Duan, Jun Wen, and Zhongbo Su
Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2022-333, https://doi.org/10.5194/hess-2022-333, 2022
Revised manuscript not accepted
Short summary
Short summary
This paper demonstrated the capability of our developed platform for simulating microwave emission and backscatter signals at multi-frequency. The results of associated investigations on impacts of vegetation water (VW) and temperature (T) imply the need to first disentangle the impact of T for the use of high-frequency signals as its variation is more due to dynamic T. Estimated vegetation optical depth is frequency-dependent, while its diurnal variation depends on that of VW despite frequency.
Shaoning Lv, Clemens Simmer, Yijian Zeng, Jun Wen, Yuanyuan Guo, and Zhongbo Su
The Cryosphere Discuss., https://doi.org/10.5194/tc-2021-369, https://doi.org/10.5194/tc-2021-369, 2022
Preprint withdrawn
Short summary
Short summary
The freeze-thaw of the ground is an interesting topic to climatology, hydrology, and other earth sciences. The global freeze-thaw distribution is available by passive microwave remote sensing technique. However, the remote sensing technique indirectly detects freeze-thaw states by measuring the brightness temperature difference between frozen and unfrozen soil. Thus, we present different interprets of the brightness signals to the FT-state by using its sub-daily character.
Lianyu Yu, Yijian Zeng, and Zhongbo Su
Geosci. Model Dev., 14, 7345–7376, https://doi.org/10.5194/gmd-14-7345-2021, https://doi.org/10.5194/gmd-14-7345-2021, 2021
Short summary
Short summary
We developed an integrated soil–snow–atmosphere model (STEMMUS-UEB) dedicated to the physical description of snow and soil processes with various complexities. With STEMMUS-UEB, we demonstrated that the snowpack affects not only the soil surface moisture conditions (in the liquid and ice phase) and energy-related states (albedo, LE) but also the subsurface soil water and vapor transfer, which contributes to a better understanding of the hydrothermal implications of the snowpack in cold regions.
Wouter Dorigo, Irene Himmelbauer, Daniel Aberer, Lukas Schremmer, Ivana Petrakovic, Luca Zappa, Wolfgang Preimesberger, Angelika Xaver, Frank Annor, Jonas Ardö, Dennis Baldocchi, Marco Bitelli, Günter Blöschl, Heye Bogena, Luca Brocca, Jean-Christophe Calvet, J. Julio Camarero, Giorgio Capello, Minha Choi, Michael C. Cosh, Nick van de Giesen, Istvan Hajdu, Jaakko Ikonen, Karsten H. Jensen, Kasturi Devi Kanniah, Ileen de Kat, Gottfried Kirchengast, Pankaj Kumar Rai, Jenni Kyrouac, Kristine Larson, Suxia Liu, Alexander Loew, Mahta Moghaddam, José Martínez Fernández, Cristian Mattar Bader, Renato Morbidelli, Jan P. Musial, Elise Osenga, Michael A. Palecki, Thierry Pellarin, George P. Petropoulos, Isabella Pfeil, Jarrett Powers, Alan Robock, Christoph Rüdiger, Udo Rummel, Michael Strobel, Zhongbo Su, Ryan Sullivan, Torbern Tagesson, Andrej Varlagin, Mariette Vreugdenhil, Jeffrey Walker, Jun Wen, Fred Wenger, Jean Pierre Wigneron, Mel Woods, Kun Yang, Yijian Zeng, Xiang Zhang, Marek Zreda, Stephan Dietrich, Alexander Gruber, Peter van Oevelen, Wolfgang Wagner, Klaus Scipal, Matthias Drusch, and Roberto Sabia
Hydrol. Earth Syst. Sci., 25, 5749–5804, https://doi.org/10.5194/hess-25-5749-2021, https://doi.org/10.5194/hess-25-5749-2021, 2021
Short summary
Short summary
The International Soil Moisture Network (ISMN) is a community-based open-access data portal for soil water measurements taken at the ground and is accessible at https://ismn.earth. Over 1000 scientific publications and thousands of users have made use of the ISMN. The scope of this paper is to inform readers about the data and functionality of the ISMN and to provide a review of the scientific progress facilitated through the ISMN with the scope to shape future research and operations.
Mengna Li, Yijian Zeng, Maciek W. Lubczynski, Jean Roy, Lianyu Yu, Hui Qian, Zhenyu Li, Jie Chen, Lei Han, Han Zheng, Tom Veldkamp, Jeroen M. Schoorl, Harrie-Jan Hendricks Franssen, Kai Hou, Qiying Zhang, Panpan Xu, Fan Li, Kai Lu, Yulin Li, and Zhongbo Su
Earth Syst. Sci. Data, 13, 4727–4757, https://doi.org/10.5194/essd-13-4727-2021, https://doi.org/10.5194/essd-13-4727-2021, 2021
Short summary
Short summary
The Tibetan Plateau is the source of most of Asia's major rivers and has been called the Asian Water Tower. Due to its remoteness and the harsh environment, there is a lack of field survey data to investigate its hydrogeology. Borehole core lithology analysis, an altitude survey, soil thickness measurement, hydrogeological surveys, and hydrogeophysical surveys were conducted in the Maqu catchment within the Yellow River source region to improve a full–picture understanding of the water cycle.
Cited articles
Arora, B., Dwivedi, D., Faybishenko, B., Jana, R. B., and Wainwright, H. M.:
Understanding and predicting vadose zone processes, Rev. Mineral. Geochem., 85, 303–328, https://doi.org/10.2138/rmg.2019.85.10, 2019.
Auler, A. S. and Smart, P. L.: Rates of condensation corrosion in speleothems of semiarid northeastern Brazil, Speleogenesis Evol. Karst Aquifers, 2, 2, https://speleogenesis.com/resources/journal/viewpoints-and-comments/638-rates-of-condensation-corrosion-in-speleothems-of-semi (last access: 15 July 2023), 2004.
Barzegar, R., Fijani, E., Moghaddam, A, A., and Tziritis E.:
Forecasting of groundwater level fluctuations using ensemble hybrid multi-wavelet neural network-based models, Sci. Total. Environ., 599–600, 20–31, https://doi.org/10.1016/j.scitotenv.2017.04.189, 2017.
Bengio, Y.: Practical recommendations for gradient-based training of deep architectures, in: Neural networks: Tricks of the trade, edited by: Montavon, G., Orr, G. B., and Müller, K.-R., Springer, 437–478, https://doi.org/10.1007/978-3-642-35289-8_26, 2012.
Bengio, Y., Simard, P., and Frasconi, P.: Learning long-term dependencies with gradient descent is difficult, IEEE T. Neur. Net. Lear., 5, 157–166, https://doi.org/10.1109/72.279181, 1994.
Broekhoff, J. C. P.: Sorption and Capillarity, Waltman, Delft, http://resolver.tudelft.nl/uuid:f3bfa125-8ab7-42c3-bc09-2f3929e32970 (last access: 15 July 2023), 1969.
Brubaker, K. L. and Entekhabi, D.:
Analysis of feedback mechanisms in land-atmosphere interaction, Water Resour. Res., 32, 1343–1357, https://doi.org/10.1029/96WR00005, 1996.
Camuffo, D.: Microclimate for cultural heritage, Elsevier, Amsterdam, ISBN 978-0-444-63296-8, 1998.
Cao, W. B., Wan, L., Zeng, Y. J., Gan, X. M., Jiang, X. W., Huang, J. Z., Yan, H. B., and Zhang, L.: Study on formation mechanism and prevention of condensation water in Yungang Grottoes, in: Collection of Papers of the 2005 Yungang International Academic Seminar (Protection Volume), Cultural Relics Publishing House, Beijing, 191–198, ISBN 978-7-501-01901-4, 2005.
Chen, J., Gao, X., Zheng, X., Miao, C., Zhang, Y., Du, Q., and Xu, Y.:
Simulation of Soil Freezing and Thawing for Different Groundwater Table Depths, Vadose Zone J., 18, 18057, https://doi.org/10.2136/vzj2018.08.0157, 2019.
de Freitas, C. R. and Schmekal, A.: Studies of condensation/evaporation processes in the Glowworm Cave, New Zealand, Int. J. Speleol., 35, 75–81, https://doi.org/10.5038/1827-806X.35.2.3, 2006.
Deprez M., De Kock, T., De Schutterc, G., and Cnudde, V.: A review on freeze–thaw action and weathering of rocks, Earth-Sci. Rev., 203, 103143, https://doi.org/10.1016/j.earscirev.2020.103143, 2020.
Dingman, S. L.: Physical hydrology, Prentice Hall, Upper Saddle River, ISBN 0-13-099695-5, 2002.
Fang, K., Pan, M., and Shen, C. P.: The value of SMAP for long-term soil moisture estimation with the help of deep learning, IEEE T. Geosci. Remote, 57, 2221–2233, https://doi.org/10.1109/TGRS.2018.2872131, 2019.
Fernández-Cortés, Á., Calaforra, J. M., and García-Guinea, J.:
The Pulpí gigantic geode (Almería, Spain): geology, metal pollution, microclimatology, and conservation, Environ. Geol., 50, 707–716, https://doi.org/10.1007/s00254-006-0243-z, 2006.
Fischer, T. and Krauss, C.:
Deep learning with long short-term memory networks for financial market predictions, Eur. J. Oper. Res., 270, 654–669, https://doi.org/10.1016/j.ejor.2017.11.054, 2018.
Gabrovšek, F., Dreybrodt, W., and Perne, M.:
Physics of condensation corrosion in caves, in: Advances in Research in Karst Media. Environmental Earth Sciences, edited by: Andreo, B., Carrasco, F., Durán, J. J., and LaMoreaux, J. W., Springer, Berlin, Heidelberg, 491–496, https://doi.org/10.1007/978-3-642-12486-0_75, 2010.
Gao, S., Huang Y. F., Zhang, S., Han, J. C., Wang, G. Q., Zhang, M. X., and Lin, Q. S.:
Short-term runoff prediction with GRU and LSTM networks without requiring time step optimization during sample generation, J. Hydrol., 589, 125188, https://doi.org/10.1016/j.jhydrol.2020.125188, 2020.
Guerrier, B., Doumenc, F., Roux, A., Mergui, S., and Jeannin, P. Y.:
Climatology in shallow caves with negligible ventilation: Heat and mass transfer, Int. J. Therm. Sci., 146, 106066, https://doi.org/10.1016/j.ijthermalsci.2019.106066, 2019.
Hochreiter, S. and Schmidhuber, J.:
Long Short-Term Memory, Neural Comput., 9, 1735–1780, https://doi.org/10.1162/neco.1997.9.8.1735, 1997.
Hopfield, J. J.:
Neural networks and physical systems with emergent collective computational abilities, P. Natl. Acad. Sci. USA, 79, 2554–2558, https://doi.org/10.1073/pnas.79.8.2554, 1982.
Irmak, S. and Irmak, A.:
Performance of frequency-domain reflectometer, capacitance, and psuedo-transit time-based soil water content probes in four coarse-textured soils, Appl. Eng. Agric., 21, 999–1008, https://doi.org/10.13031/2013.20035, 2005.
Jumin, E., Zaini, N., Ahmed A. N., Abdullah, S., Ismail, M., Sherif, M., Sefelnasr, A., and El-Shafie, A.:
Machine learning versus linear regression modelling approach for accurate ozone concentrations prediction, Eng. Appl. Comp. Fluid, 14, 713–725, https://doi.org/10.1080/19942060.2020.1758792, 2020.
Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S. F., and Yang, L.:
Physics-informed machine learning, Nat. Rev. Phys., 3, 422–440, https://doi.org/10.1038/s42254-021-00314-5, 2021.
Larson, J., Lidberg, W., Ågren, A. M., and Laudon, H.:
Predicting soil moisture conditions across a heterogeneous boreal catchment using terrain indices, Hydrol. Earth Syst. Sci., 26, 4837–4851, https://doi.org/10.5194/hess-26-4837-2022, 2022.
Lees, T., Buechel, M., Anderson, B., Slater, L., Reece, S., Coxon, G., and Dadson, S. J.:
Benchmarking data-driven rainfall–runoff models in Great Britain: a comparison of long short-term memory (LSTM)-based models with four lumped conceptual models, Hydrol. Earth Syst. Sci., 25, 5517–5534, https://doi.org/10.5194/hess-25-5517-2021, 2021.
Li, P. J., Zha, Y. Y., Tso, C. M., Shi, L. S., Yu, D. Y., Zhang, Y. G., and Zeng, W. Z.:
Data assimilation of uncalibrated soil moisture measurements from frequency-domain reflectometry, Geoderma, 374, 114432, https://doi.org/10.1016/j.geoderma.2020.114432, 2020.
Li, Y., Aemisegger, F., Riedl, A., Buchmann, N., and Eugster, W.:
The role of dew and radiation fog inputs in the local water cycling of a temperate grassland during dry spells in central Europe, Hydrol. Earth Syst. Sci., 25, 2617–2648, https://doi.org/10.5194/hess-25-2617-2021, 2021.
Liñán, C., Benavente, J., del Rosal, Y., Vadillo, I., Ojeda, L., and Carrasco, F.: Condensation water in heritage touristic caves: Isotopic and hydrochemical data and a new approach for its quantification through image analysis, Hydrol. Process., 35, e14083, https://doi.org/10.1002/hyp.14083, 2021.
Lipton, Z. C., Berkowitz, J., and Elkan, C.:
A critical review of recurrent neural networks for sequence learning, arXiv [preprint], arXiv:1506.00019, 2015.
Liu, G. J., Tang, Z. Y., Qin, H., Liu, S., Shen, Q., Qu, Y. H., and Zhou, J. Z.: Short-term runoff prediction using deep learning multi-dimensional ensemble method, J. Hydrol., 609, 127762, https://doi.org/10.1016/j.jhydrol.2022.127762, 2022.
Lu, N. and Likos, W. J.: Unsaturated soil mechanics, John Wiley, Hoboken, ISBN 978-0-471-44731-3, 2004.
Lundberg, S. M. and Lee, S.:
A unified approach to interpreting model predictions, in: Advances in neural information processing systems 30 (NIPS 2017), edited by: Guyon, I., Luxburg, U. V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., and Garnett, R., Neural information processing systems (NIPS), La Jolla, 2017.
Mao, D. Q., Wang, X. D., Meng, J., Ma, X. M., Jiang, X. W., Wan, L., Yan, H. B., and Fan, Y.:
Infiltration assessments on top of yungang grottoes by time–lapse electrical resistivity tomography, Hydrology, 9, 77, https://doi.org/10.3390/hydrology9050077, 2022.
Matsuoka, N. and Murton, J.:
Frost weathering: recent advances and future directions, Permafrost Periglac., 19, 195–210, https://doi.org/10.1002/ppp.620, 2008.
Meinzer, O. E.: The history and development of ground–water hydrology, J. Washington Academy of Sciences, 24, 6–32, http://www.jstor.org/stable/24532897 (last access: 15 July 2023), 1934.
Mollo, L. and Greco, R.:
Moisture measurements in masonry materials by time domain reflectometry, J. Mater. Civil Eng., 23, 441–444, 2011.
Nguyen, J. L., Schwartz, J., and Dockery, D. W.: The relationship between indoor and outdoor temperature, apparent temperature, relative humidity, and absolute humidity, Indoor Air, 24, 103–112, https://doi.org/10.1111/ina.12052, 2013.
Ouyang, K. G. and Jiang, X. W.: LSTM+SHAP_Rock water content, Zenodo [code], https://doi.org/10.5281/zenodo.7382827, 2022a.
Ouyang, K. G. and Jiang, X. W.: Rock moisture in the YunGang Grottoes, Zenodo [data set], https://doi.org/10.5281/zenodo.7382895, 2022b.
Rempe, D. M. and Dietrich, W. E.: Direct observations of rock moisture, a hidden component of the hydrologic cycle, P. Natl. Acad. Sci. USA, 115, 2664–2669, https://doi.org/10.1073/pnas.1800141115, 2018.
Sakaki, T. and Rajaram, H.: Performance of different types of time domain reflectometry probes for water content measurement in partially saturated rocks, Water Resour. Res., 42, W07404, https://doi.org/10.1029/2005WR004643, 2006.
Salve, R., Rempe, D. M., and Dietrich, W. E.:
Rain, rock moisture dynamics, and the rapid response of perched groundwater in weathered, fractured argillite underlying a steep hillslope, Water Resour. Res., 48, 1–25, https://doi.org/10.1029/2012WR012583, 2012.
Sass, O.:
Rock moisture measurements: techniques, results, and implications for weathering, Earth Surf. Proc. Land., 30, 359–374, https://doi.org/10.1002/esp.1214, 2005.
Schmidt, L. and Rempe, D.:
Quantifying dynamic water storage in unsaturated bedrock with borehole nuclear magnetic resonance, Geophys. Res. Lett., 47, e2020GL089600, https://doi.org/10.1029/2020GL089600, 2020.
Schoups, G., Hopmans, J. W., Young, C. A., Vrugt, J. A., and Wallender, W. W.:
Multi-criteria optimization of a regional spatially-distributed subsurface water flow model, J. Hydrol., 31, 20–48, https://doi.org/10.1016/j.jhydrol.2005.01.001, 2005.
Shapley, L. S.:
A value for n-person games, in: Contributions to the Theory of Games II, editd by: Kuhn, H. and Tucker, A., Princeton University Press, Princeton, 307–317, https://doi.org/10.1515/9781400881970-018, 1953.
Štrumbelj, E. and Kononenko, I.:
Explaining prediction models and individual predictions with feature contributions, Knowl. Inf. Syst., 41, 647–665, https://doi.org/10.1007/s10115-013-0679-x, 2014.
Sun, Z. H. and Scherer, G. W.:
Measurement and simulation of dendritic growth of ice in cement paste, Cement Concrete Res., 40, 1393–1402, https://doi.org/10.1016/j.cemconres.2010.03.008, 2010.
Tindall, J., Kunkel, J., and Anderson, D. E.: Unsaturated Zone Hydrology for Scientists and Engineers, Pearson Education Inc., New York, ISBN 978-0-136-60713-7, 1998.
Vereecken, H., Huisman, J. A., Pachepsky, Y., Montzka, C., van der Kruk, J., Bogena, H., Weihermuller, L., Herbst, M., Martinez, G., and Vanderborght, J.:
On the spatio-temporal dynamics of soil moisture at the field scale, J. Hydrol., 516, 76–96, https://doi.org/10.1016/j.jhydrol.2013.11.061, 2014.
Vinnikov, K. Y., Robock, A., Speranskaya, N. A., and Schlosser, C. A.:
Scales of temporal and spatial variability of midlatitude soil moisture, J. Geophys. Res.-Atmos., 101, 7163–7174, https://doi.org/10.1029/95JD02753, 1996.
Wang. X. S., Wan, L., Peng, T., Dong, P., and Qian, J.:
Infiltration water in Yungang Grottoes and its movement, Geotechnical Investigation & Surveying, 40, 12–16, 2012.
Xiang, Z. R., Yan, J., and Demir I.: A rainfall-runoff model with LSTM-based sequence-to- sequence learning, Water Resour. Res., 56, e2019WR025326, https://doi.org/10.1029/2019WR025326, 2020.
Xie, H.-Y., Jiang, X.-W., Tan, S.-C., Wan, L., Wang, X.-S., Liang, S.-H., and Zeng, Y.:
Interaction of soil water and groundwater during the freezing–thawing cycle: field observations and numerical modeling, Hydrol. Earth Syst. Sci., 25, 4243–4257, https://doi.org/10.5194/hess-25-4243-2021, 2021.
Xu, L., Lio, A., Hu, J., Ogletree, A. D. F., and Salmeron, M.: Wetting and capillary phenomena of water on mica, J. Phys. Chem. B, 102, 540–548, https://doi.org/10.1021/JP972289L, 1998.
Yu, L., Zeng, Y., Su, Z., Cai, H., and Zheng, Z.:
The effect of different evapotranspiration methods on portraying soil water dynamics and ET partitioning in a semi-arid environment in Northwest China, Hydrol. Earth Syst. Sci., 20, 975–990, https://doi.org/10.5194/hess-20-975-2016, 2016.
Yu, L., Zeng, Y., Wen, J., and Su, Z.:
Liquid-Vapor-Air Flow in the Frozen Soil, J. Geophys. Res.-Atmos., 123, 7393–7415, https://doi.org/10.1029/2018jd028502, 2018.
Zhang, Z. Y., Wang, W., Gong, C. C., Wang, Z. F., Duan, L., Yeh, T. J., and Yu, P. Y.:
Evaporation from seasonally frozen bare and vegetated ground at various groundwater table depths in the Ordos Basin, Northwest China, Hydrol. Process., 33, 1338–1348, https://doi.org/10.1002/hyp.13404, 2019.
Zhao, H. F., Li, J., Yuan, Q. Q., Lin, L. P., Yue, L. W., and Xu, H. Z.:
Downscaling of soil moisture products using deep learning: Comparison and analysis on Tibetan Plateau, J. Hydrol., 607, 127570, https://doi.org/10.1016/j.jhydrol.2022.127570, 2022.
Zhao, Z., Chen, W. H., Wu, X. M., Chen, P. C. Y., and Liu, J. M.:
LSTM Network: a deep learning approach for short-term traffic forecast, IET Intell. Transp. Sy., 11, 68–75, https://doi.org/10.1049/IET-ITS.2016.0208, 2017.
Short summary
Our knowledge on sources and dynamics of rock moisture is limited. By using frequency domain reflectometry (FDR), we monitored rock moisture in a cave. The results of an explainable deep learning model reveal that the direct source of rock moisture responsible for weathering in the studied cave is vapour, not infiltrating precipitation. A physics-informed deep learning model, which uses variables controlling vapor condensation as model inputs, leads to accurate rock water content predictions.
Our knowledge on sources and dynamics of rock moisture is limited. By using frequency domain...
Special issue