Articles | Volume 26, issue 9
https://doi.org/10.5194/hess-26-2405-2022
© Author(s) 2022. 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-26-2405-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
09 May 2022
Research article |
| 09 May 2022
| 09 May 2022
Karst spring discharge modeling based on deep learning using spatially distributed input data
Karlsruhe Institute of Technology (KIT), Institute of Applied Geosciences, Hydrogeology, Kaiserstr. 12, 76131 Karlsruhe, Germany
Tanja Liesch
Karlsruhe Institute of Technology (KIT), Institute of Applied Geosciences, Hydrogeology, Kaiserstr. 12, 76131 Karlsruhe, Germany
Guillaume Cinkus
HydroSciences Montpellier (HSM), Université de Montpellier, CNRS, IRD, 34090 Montpellier, France
Nataša Ravbar
ZRC SAZU, Karst Research Institute, Titov trg 2, 6230 Postojna, Slovenia
Zhao Chen
Institute of Groundwater Management, Technical University of Dresden, 01062 Dresden
Naomi Mazzilli
UMR 1114 EMMAH (AU-INRAE), Université d'Avignon, 84000 Avignon, France
Hervé Jourde
HydroSciences Montpellier (HSM), Université de Montpellier, CNRS, IRD, 34090 Montpellier, France
Nico Goldscheider
Karlsruhe Institute of Technology (KIT), Institute of Applied Geosciences, Hydrogeology, Kaiserstr. 12, 76131 Karlsruhe, Germany
Related authors
Raoul A. Collenteur, Ezra Haaf, Mark Bakker, Tanja Liesch, Andreas Wunsch, Jenny Soonthornrangsan, Jeremy White, Nick Martin, Rui Hugman, Ed de Sousa, Didier Vanden Berghe, Xinyang Fan, Tim J. Peterson, Jānis Bikše, Antoine Di Ciacca, Xinyue Wang, Yang Zheng, Maximilian Nölscher, Julian Koch, Raphael Schneider, Nikolas Benavides Höglund, Sivarama Krishna Reddy Chidepudi, Abel Henriot, Nicolas Massei, Abderrahim Jardani, Max Gustav Rudolph, Amir Rouhani, J. Jaime Gómez-Hernández, Seifeddine Jomaa, Anna Pölz, Tim Franken, Morteza Behbooei, Jimmy Lin, and Rojin Meysami
Hydrol. Earth Syst. Sci., 28, 5193–5208, https://doi.org/10.5194/hess-28-5193-2024, https://doi.org/10.5194/hess-28-5193-2024, 2024
Short summary
Short summary
We show the results of the 2022 Groundwater Time Series Modelling Challenge; 15 teams applied data-driven models to simulate hydraulic heads, and three model groups were identified: lumped, machine learning, and deep learning. For all wells, reasonable performance was obtained by at least one team from each group. There was not one team that performed best for all wells. In conclusion, the challenge was a successful initiative to compare different models and learn from each other.
Andreas Wunsch, Tanja Liesch, and Nico Goldscheider
Hydrol. Earth Syst. Sci., 28, 2167–2178, https://doi.org/10.5194/hess-28-2167-2024, https://doi.org/10.5194/hess-28-2167-2024, 2024
Short summary
Short summary
Seasons have a strong influence on groundwater levels, but relationships are complex and partly unknown. Using data from wells in Germany and an explainable machine learning approach, we showed that summer precipitation is the key factor that controls the severeness of a low-water period in fall; high summer temperatures do not per se cause stronger decreases. Preceding winters have only a minor influence on such low-water periods in general.
Guillaume Cinkus, Naomi Mazzilli, Hervé Jourde, Andreas Wunsch, Tanja Liesch, Nataša Ravbar, Zhao Chen, and Nico Goldscheider
Hydrol. Earth Syst. Sci., 27, 2397–2411, https://doi.org/10.5194/hess-27-2397-2023, https://doi.org/10.5194/hess-27-2397-2023, 2023
Short summary
Short summary
The Kling–Gupta Efficiency (KGE) is a performance criterion extensively used to evaluate hydrological models. We conduct a critical study on the KGE and its variant to examine counterbalancing errors. Results show that, when assessing a simulation, concurrent over- and underestimation of discharge can lead to an overall higher criterion score without an associated increase in model relevance. We suggest that one carefully choose performance criteria and use scaling factors.
Guillaume Cinkus, Andreas Wunsch, Naomi Mazzilli, Tanja Liesch, Zhao Chen, Nataša Ravbar, Joanna Doummar, Jaime Fernández-Ortega, Juan Antonio Barberá, Bartolomé Andreo, Nico Goldscheider, and Hervé Jourde
Hydrol. Earth Syst. Sci., 27, 1961–1985, https://doi.org/10.5194/hess-27-1961-2023, https://doi.org/10.5194/hess-27-1961-2023, 2023
Short summary
Short summary
Numerous modelling approaches can be used for studying karst water resources, which can make it difficult for a stakeholder or researcher to choose the appropriate method. We conduct a comparison of two widely used karst modelling approaches: artificial neural networks (ANNs) and reservoir models. Results show that ANN models are very flexible and seem great for reproducing high flows. Reservoir models can work with relatively short time series and seem to accurately reproduce low flows.
Marc Ohmer, Tanja Liesch, and Andreas Wunsch
Hydrol. Earth Syst. Sci., 26, 4033–4053, https://doi.org/10.5194/hess-26-4033-2022, https://doi.org/10.5194/hess-26-4033-2022, 2022
Short summary
Short summary
We present a data-driven approach to select optimal locations for groundwater monitoring wells. The applied approach can optimize the number of wells and their location for a network reduction (by ranking wells in order of their information content and reducing redundant) and extension (finding sites with great information gain) or both. It allows us to include a cost function to account for more/less suitable areas for new wells and can help to obtain maximum information content for a budget.
Andreas Wunsch, Tanja Liesch, and Stefan Broda
Hydrol. Earth Syst. Sci., 25, 1671–1687, https://doi.org/10.5194/hess-25-1671-2021, https://doi.org/10.5194/hess-25-1671-2021, 2021
Fabienne Doll, Tanja Liesch, Maria Wetzel, Stefan Kunz, and Stefan Broda
EGUsphere, https://doi.org/10.5194/egusphere-2025-3539, https://doi.org/10.5194/egusphere-2025-3539, 2025
This preprint is open for discussion and under review for Geoscientific Model Development (GMD).
Short summary
Short summary
With the growing use of machine learning for groundwater level (GWL) prediction, proper performance estimation is crucial. This study compares three validation strategies—blocked cross-validation (bl-CV), repeated out-of-sample (repOOS), and out-of-sample (OOS)—for 1D-CNN models using meteorological inputs. Results show that bl-CV offers the most reliable performance estimates, while OOS is the most uncertain, highlighting the need for careful method selection.
Marc Ohmer, Tanja Liesch, Bastian Habbel, Benedikt Heudorfer, Mariana Gomez, Patrick Clos, Maximilian Nölscher, and Stefan Broda
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2025-321, https://doi.org/10.5194/essd-2025-321, 2025
Preprint under review for ESSD
Short summary
Short summary
We present a public dataset of weekly groundwater levels from more than 3,000 wells across Germany, spanning 32 years. It combines weather data and site-specific environmental information to support forecasting groundwater changes. Three benchmark models of varying complexity show how data and modeling approaches influence predictions. This resource promotes open, reproducible research and helps guide future water management decisions.
Vianney Sivelle, Guillaume Cinkus, Naomi Mazzilli, David Labat, Bruno Arfib, Nicolas Massei, Yohann Cousquer, Dominique Bertin, and Hervé Jourde
Hydrol. Earth Syst. Sci., 29, 1259–1276, https://doi.org/10.5194/hess-29-1259-2025, https://doi.org/10.5194/hess-29-1259-2025, 2025
Short summary
Short summary
KarstMod provides a platform for global modelling of the rain level–flow relationship in karstic basins. The platform provides a set of tools to assess the dynamics of the compartments considered in the model and to detect possible flaws in structure and parameterization. This platform is developed as part of the French observatory network on karst hydrology (SNO KARST), which aims to strengthen the sharing of knowledge and promote interdisciplinary research on karst systems at a national level.
Raoul A. Collenteur, Ezra Haaf, Mark Bakker, Tanja Liesch, Andreas Wunsch, Jenny Soonthornrangsan, Jeremy White, Nick Martin, Rui Hugman, Ed de Sousa, Didier Vanden Berghe, Xinyang Fan, Tim J. Peterson, Jānis Bikše, Antoine Di Ciacca, Xinyue Wang, Yang Zheng, Maximilian Nölscher, Julian Koch, Raphael Schneider, Nikolas Benavides Höglund, Sivarama Krishna Reddy Chidepudi, Abel Henriot, Nicolas Massei, Abderrahim Jardani, Max Gustav Rudolph, Amir Rouhani, J. Jaime Gómez-Hernández, Seifeddine Jomaa, Anna Pölz, Tim Franken, Morteza Behbooei, Jimmy Lin, and Rojin Meysami
Hydrol. Earth Syst. Sci., 28, 5193–5208, https://doi.org/10.5194/hess-28-5193-2024, https://doi.org/10.5194/hess-28-5193-2024, 2024
Short summary
Short summary
We show the results of the 2022 Groundwater Time Series Modelling Challenge; 15 teams applied data-driven models to simulate hydraulic heads, and three model groups were identified: lumped, machine learning, and deep learning. For all wells, reasonable performance was obtained by at least one team from each group. There was not one team that performed best for all wells. In conclusion, the challenge was a successful initiative to compare different models and learn from each other.
Andreas Wunsch, Tanja Liesch, and Nico Goldscheider
Hydrol. Earth Syst. Sci., 28, 2167–2178, https://doi.org/10.5194/hess-28-2167-2024, https://doi.org/10.5194/hess-28-2167-2024, 2024
Short summary
Short summary
Seasons have a strong influence on groundwater levels, but relationships are complex and partly unknown. Using data from wells in Germany and an explainable machine learning approach, we showed that summer precipitation is the key factor that controls the severeness of a low-water period in fall; high summer temperatures do not per se cause stronger decreases. Preceding winters have only a minor influence on such low-water periods in general.
Benedikt Heudorfer, Tanja Liesch, and Stefan Broda
Hydrol. Earth Syst. Sci., 28, 525–543, https://doi.org/10.5194/hess-28-525-2024, https://doi.org/10.5194/hess-28-525-2024, 2024
Short summary
Short summary
We build a neural network to predict groundwater levels from monitoring wells. We predict all wells at the same time, by learning the differences between wells with static features, making it an entity-aware global model. This works, but we also test different static features and find that the model does not use them to learn exactly how the wells are different, but only to uniquely identify them. As this model class is not actually entity aware, we suggest further steps to make it so.
Chloé Fandel, Ty Ferré, François Miville, Philippe Renard, and Nico Goldscheider
Hydrol. Earth Syst. Sci., 27, 4205–4215, https://doi.org/10.5194/hess-27-4205-2023, https://doi.org/10.5194/hess-27-4205-2023, 2023
Short summary
Short summary
From the surface, it is hard to tell where underground cave systems are located. We developed a computer model to create maps of the probable cave network in an area, based on the geologic setting. We then applied our approach in reverse: in a region where an old cave network was mapped, we used modeling to test what the geologic setting might have been like when the caves formed. This is useful because understanding past cave formation can help us predict where unmapped caves are located today.
Guillaume Cinkus, Naomi Mazzilli, Hervé Jourde, Andreas Wunsch, Tanja Liesch, Nataša Ravbar, Zhao Chen, and Nico Goldscheider
Hydrol. Earth Syst. Sci., 27, 2397–2411, https://doi.org/10.5194/hess-27-2397-2023, https://doi.org/10.5194/hess-27-2397-2023, 2023
Short summary
Short summary
The Kling–Gupta Efficiency (KGE) is a performance criterion extensively used to evaluate hydrological models. We conduct a critical study on the KGE and its variant to examine counterbalancing errors. Results show that, when assessing a simulation, concurrent over- and underestimation of discharge can lead to an overall higher criterion score without an associated increase in model relevance. We suggest that one carefully choose performance criteria and use scaling factors.
Guillaume Cinkus, Andreas Wunsch, Naomi Mazzilli, Tanja Liesch, Zhao Chen, Nataša Ravbar, Joanna Doummar, Jaime Fernández-Ortega, Juan Antonio Barberá, Bartolomé Andreo, Nico Goldscheider, and Hervé Jourde
Hydrol. Earth Syst. Sci., 27, 1961–1985, https://doi.org/10.5194/hess-27-1961-2023, https://doi.org/10.5194/hess-27-1961-2023, 2023
Short summary
Short summary
Numerous modelling approaches can be used for studying karst water resources, which can make it difficult for a stakeholder or researcher to choose the appropriate method. We conduct a comparison of two widely used karst modelling approaches: artificial neural networks (ANNs) and reservoir models. Results show that ANN models are very flexible and seem great for reproducing high flows. Reservoir models can work with relatively short time series and seem to accurately reproduce low flows.
Leïla Serène, Christelle Batiot-Guilhe, Naomi Mazzilli, Christophe Emblanch, Milanka Babic, Julien Dupont, Roland Simler, Matthieu Blanc, and Gérard Massonnat
Hydrol. Earth Syst. Sci., 26, 5035–5049, https://doi.org/10.5194/hess-26-5035-2022, https://doi.org/10.5194/hess-26-5035-2022, 2022
Short summary
Short summary
This work aims to develop the Transit Time index (TTi) as a natural tracer of karst groundwater transit time, usable in the 0–6-month range. Based on the fluorescence of organic matter, TTi shows its relevance to detect a small proportion of fast infiltration water within a mix, while other natural transit time tracers provide no or less sensitive information. Comparison of the average TTi of different karst springs also provides consistent results with the expected relative transit times.
Marc Ohmer, Tanja Liesch, and Andreas Wunsch
Hydrol. Earth Syst. Sci., 26, 4033–4053, https://doi.org/10.5194/hess-26-4033-2022, https://doi.org/10.5194/hess-26-4033-2022, 2022
Short summary
Short summary
We present a data-driven approach to select optimal locations for groundwater monitoring wells. The applied approach can optimize the number of wells and their location for a network reduction (by ranking wells in order of their information content and reducing redundant) and extension (finding sites with great information gain) or both. It allows us to include a cost function to account for more/less suitable areas for new wells and can help to obtain maximum information content for a budget.
Markus Merk, Nadine Goeppert, and Nico Goldscheider
Hydrol. Earth Syst. Sci., 25, 3519–3538, https://doi.org/10.5194/hess-25-3519-2021, https://doi.org/10.5194/hess-25-3519-2021, 2021
Short summary
Short summary
Soil moisture levels have decreased significantly over the past 2 decades. This decrease is not uniformly distributed over the observation period. The largest changes occur at tipping points during years of extreme drought, after which soil moisture levels reach significantly different alternate stable states. Not only the overall trend in soil moisture is affected, but also the seasonal dynamics.
Andreas Wunsch, Tanja Liesch, and Stefan Broda
Hydrol. Earth Syst. Sci., 25, 1671–1687, https://doi.org/10.5194/hess-25-1671-2021, https://doi.org/10.5194/hess-25-1671-2021, 2021
Cited articles
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado,
G. S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M.,
Levenberg, J., Mane, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Viegas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., and Zheng, X.: TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems, p. 19, https://www.tensorflow.org/ (last access: 6 May 2022), 2015. a
Afzaal, H., Farooque, A. A., Abbas, F., Acharya, B., and Esau, T.: Groundwater Estimation from Major Physical Hydrology Components Using Artificial Neural Networks and Deep Learning, Water, 12, 5, https://doi.org/10.3390/w12010005, 2020. a
Bandhauer, M., Isotta, F., Lakatos, M., Lussana, C., Båserud, L.,
Izsák, B., Szentes, O., Tveito, O. E., and Frei, C.: Evaluation of Daily
Precipitation Analyses in E-OBS (V19.0e) and ERA5 by Comparison to Regional High-Resolution Datasets in European Regions, Int. J. Climatol., 42, 727–747, https://doi.org/10.1002/joc.7269, 2021. a
Bergström, S.: The Development of a Snow Routine for the HBV-2 Model,
Hydrol. Res., 6, 73–92, https://doi.org/10.2166/nh.1975.0006, 1975. a
Bergström, S.: The HBV Model, in: Computer Models of Watershed Hydrology, edited by: Singh, V. P., Water Resources Publications, Colorado, USA, 443–476, ISBN 0-918334-91-8, 1995. a
Bicalho, C. C., Batiot-Guilhe, C., Seidel, J. L., Van Exter, S., and Jourde,
H.: Hydrodynamical Changes and Their Consequences on Groundwater Hydrochemistry Induced by Three Decades of Intense Exploitation in a
Mediterranean Karst System, Environ. Earth Sci., 65, 2311–2319,
https://doi.org/10.1007/s12665-011-1384-2, 2012. a
Breiman, L.: Random Forests, Mach. Learn., 45, 5–32, https://doi.org/10.1023/A:1010933404324, 2001. a
Cai, Z., Fan, Q., Feris, R. S., and Vasconcelos, N.: A Unified Multi-Scale Deep Convolutional Neural Network for Fast Object Detection, in: Computer Vision – ECCV 2016, edited by: Leibe, B., Matas, J., Sebe, N., and Welling, M., Springer International Publishing, Cham, 354–370, ISBN 978-3-319-46493-0, 2016. a
Chen, Z. and Goldscheider, N.: Modeling Spatially and Temporally Varied
Hydraulic Behavior of a Folded Karst System with Dominant Conduit Drainage at
Catchment Scale, Hochifen–Gottesacker, Alps, J. Hydrol., 514, 41–52, https://doi.org/10.1016/j.jhydrol.2014.04.005, 2014. a, b, c, d
Chen, Z., Auler, A. S., Bakalowicz, M., Drew, D., Griger, F., Hartmann, J.,
Jiang, G., Moosdorf, N., Richts, A., Stevanovic, Z., Veni, G., and Goldscheider, N.: The World Karst Aquifer Mapping Project: Concept, Mapping Procedure and Map of Europe, Hydrogeol. J., 25, 771–785, https://doi.org/10.1007/s10040-016-1519-3, 2017a. a
Chen, Z., Hartmann, A., and Goldscheider, N.: A New Approach to Evaluate
Spatiotemporal Dynamics of Controlling Parameters in Distributed Environmental Models, Environ. Model. Softwa., 87, 1–16, https://doi.org/10.1016/j.envsoft.2016.10.005, 2017b. a, b, c, d
Chen, Z., Hartmann, A., Wagener, T., and Goldscheider, N.: Dynamics of water fluxes and storages in an Alpine karst catchment under current and potential future climate conditions, Hydrol. Earth Syst. Sci., 22, 3807–3823, https://doi.org/10.5194/hess-22-3807-2018, 2018. a, b, c
Chollet, F.: Keras, https://github.com/keras-team/keras (last access: 22 May 2020), 2015. a
Copernicus Climate Change Service: E-OBS daily gridded meteorological data for Europe from 1950 to present derived from in-situ observations, Copernicus Climate Change Service [data set], https://doi.org/10.24381/CDS.151D3EC6, 2020. a
Cornes, R. C., van der Schrier, G., van den Besselaar, E. J. M., and Jones, P. D.: An Ensemble Version of the E-OBS Temperature and Precipitation Data Sets, J. Geophys. Res.-Atmos., 123, 9391–9409, https://doi.org/10.1029/2017jd028200, 2018. a, b, c
Darras, T., Borrell Estupina, V., Kong-A-Siou, L., Vayssade, B., Johannet, A., and Pistre, S.: Identification of spatial and temporal contributions of rainfalls to flash floods using neural network modelling: case study on the Lez basin (southern France), Hydrol. Earth Syst. Sci., 19, 4397–4410, https://doi.org/10.5194/hess-19-4397-2015, 2015. a
Darras, T., Kong-A-Siou, L., Vayssade, B., Johannet, A., and Pistre, S.: Karst Flash Flood Forecasting Using Recurrent and Nonrecurrent Artificial Neural Network Models: The Case of the Lez Basin (Southern France), in: EuroKarst 2016, Neuchâtel, Advances in Karst Science, Springer International Publishing, Cham, https://doi.org/10.1007/978-3-319-45465-8, 2017. a
DWD: DWD Opendata, https://opendata.dwd.de/, last access: 6 May 2022. a
Fleury, P., Ladouche, B., Conroux, Y., Jourde, H., and Dörfliger, N.:
Modelling the Hydrologic Functions of a Karst Aquifer under Active Water
Management – The Lez Spring, J. Hydrol., 365, 235–243, https://doi.org/10.1016/j.jhydrol.2008.11.037, 2009. a, b
Fresnay, S., Hally, A., Garnaud, C., Richard, E., and Lambert, D.: Heavy
Precipitation Events in the Mediterranean: Sensitivity to Cloud Physics
Parameterisation Uncertainties, Nat. Hazards Earth Syst. Sci., 12,
2671–2688, https://doi.org/10.5194/nhess-12-2671-2012, 2012. a
Friedman, J. H.: Greedy Function Approximation: A Gradient Boosting Machine, Ann. Stat., 29, 1189–1232, https://doi.org/10.1214/aos/1013203451, 2001. a
Goldscheider, N.: Fold Structure and Underground Drainage Pattern in the Alpine Karst System Hochifen-Gottesacker, Eclogae Geol. Helv., 98, 1–17,
https://doi.org/10.1007/s00015-005-1143-z, 2005. a, b, c
Goodfellow, I., Bengio, Y., and Courville, A.: Deep Learning, Adaptive
Computation and Machine Learning, The MIT Press, Cambridge, Massachusetts, ISBN 978-0-262-03561-3, 2016. a
Gupta, H. V., Kling, H., Yilmaz, K. K., and Martinez, G. F.: Decomposition of
the Mean Squared Error and NSE Performance Criteria: Implications for Improving Hydrological Modelling, J. Hydrol., 377, 80–91,
https://doi.org/10.1016/j.jhydrol.2009.08.003, 2009. a
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. a
Hock, R.: A Distributed Temperature-Index Ice- and Snowmelt Model Including
Potential Direct Solar Radiation, J. Glaciol., 45, 101–111, https://doi.org/10.3189/s0022143000003087, 1999. a
Hunter, J. D.: Matplotlib: A 2D Graphics Environment, Comput. Sci. Eng., 9, 90–95, https://doi.org/10.1109/mcse.2007.55, 2007. a
Hussain, D., Hussain, T., Khan, A. A., Naqvi, S. A. A., and Jamil, A.: A Deep
Learning Approach for Hydrological Time-Series Prediction: A Case Study of Gilgit River Basin, Earth Sci. Inform., 13, 915–927, https://doi.org/10.1007/s12145-020-00477-2, 2020. a
Innamorati, C., Ritschel, T., Weyrich, T., and Mitra, N. J.: Learning on the
Edge: Investigating Boundary Filters in CNNs, Int. J. Comput. Vis., 128, 773–782, https://doi.org/10.1007/s11263-019-01223-y, 2020. a
Ioffe, S. and Szegedy, C.: Batch Normalization: Accelerating Deep Network
Training by Reducing Internal Covariate Shift, arXiv preprint: 1502.03167 [cs], http://arxiv.org/abs/1502.03167 (last access: 14 November 2021), 2015. a
Jeannin, P.-Y., Artigue, G., Butscher, C., Chang, Y., Charlier, J.-B., Duran,
L., Gill, L., Hartmann, A., Johannet, A., Jourde, H., Kavousi, A., Liesch,
T., Liu, Y., Lüthi, M., Malard, A., Mazzilli, N., Pardo-Igúzquiza, E., Thiéry, D., Reimann, T., Schuler, P., Wöhling, T., and Wunsch, A.: Karst Modelling Challenge 1: Results of Hydrological Modelling, J. Hydrol., 600, 126508, https://doi.org/10.1016/j.jhydrol.2021.126508, 2021. a, b, c
Johannet, A., Mangin, A., and D'Hulst, D.: Subterranean Water Infiltration
Modelling by Neural Networks: Use of Water Source Flow, in: Volume 1,
Parts 1 and 2, ICANN '94: Proceedings of the International Conference on Artificial Neural Networks, 26–29 May 1994, Sorrento, Italy, Springe, Berlin, Heidelberg, Sorrento, Italy, 1033–1036, ISBN 978-3-540-19887-1, 1994. a
Jourde, H., Lafare, A., Mazzilli, N., Belaud, G., Neppel, L., Dörfliger,
N., and Cernesson, F.: Flash Flood Mitigation as a Positive Consequence of
Anthropogenic Forcing on the Groundwater Resource in a Karst Catchment,
Environ. Earth Sci., 71, 573–583, https://doi.org/10.1007/s12665-013-2678-3, 2014. a, b
Jourde, H., Massei, N., Mazzilli, N., Binet, S., Batiot-Guilhe, C., Labat,
D., Steinmann, M., Bailly-Comte, V., Seidel, J. L., Arfib, B., Charlier, J. B., Guinot, V., Jardani, A., Fournier, M., Aliouache, M., Babic, M., Bertrand, C., Brunet, P., Boyer, J. F., Bricquet, J. P., Camboulive, T.,
Carrière, S. D., Celle-Jeanton, H., Chalikakis, K., Chen, N., Cholet, C., Clauzon, V., Soglio, L. D., Danquigny, C., Défargue, C., Denimal, S.,
Emblanch, C., Hernandez, F., Gillon, M., Gutierrez, A., Sanchez, L. H., Hery,
M., Houillon, N., Johannet, A., Jouves, J., Jozja, N., Ladouche, B., Leonardi, V., Lorette, G., Loup, C., Marchand, P., de Montety, V., Muller,
R., Ollivier, C., Sivelle, V., Lastennet, R., Lecoq, N., Maréchal, J. C.,
Perotin, L., Perrin, J., Petre, M. A., Peyraube, N., Pistre, S., Plagnes, V.,
Probst, A., Probst, J. L., Simler, R., Stefani, V., Valdes-Lao, D., Viseur,
S., and Wang, X.: SNO KARST: A French Network of Observatories for the Multidisciplinary Study of Critical Zone Processes in Karst Watersheds and Aquifers, Vadose Zone J., 17, 180094, https://doi.org/10.2136/vzj2018.04.0094, 2018. a
Kaufmann, G., Mayaud, C., Kogovšek, B., and Gabrovšek, F.:
Understanding the Temporal Variation of Flow Direction in a Complex Karst
System (Planinska Jama, Slovenia), Acta Carsolog., 49, 213–228, https://doi.org/10.3986/ac.v49i2-3.7373, 2020. a
Kaufmann, G., Gabrovšek, F., and Turk, J.: Modelling Flow of Subterranean
Pivka River in Postojnska Jama, Slovenia, Acta Carsolog., 45, 57–70, https://doi.org/10.3986/ac.v45i1.3059, 2016. a
King, D. A., Bachelet, D. M., Symstad, A. J., Ferschweiler, K., and Hobbins,
M.: Estimation of Potential Evapotranspiration from Extraterrestrial
Radiation, Air Temperature and Humidity to Assess Future Climate Change
Effects on the Vegetation of the Northern Great Plains, USA, Ecol. Model., 297, 86–97, https://doi.org/10.1016/j.ecolmodel.2014.10.037, 2015. a
Kiranyaz, S., Ince, T., Abdeljaber, O., Avci, O., and Gabbouj, M.: 1-D
Convolutional Neural Networks for Signal Processing Applications, in:
ICASSP 2019–2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 12–17 May 2019, Brighton, UK, 8360–8364, https://doi.org/10.1109/ICASSP.2019.8682194, 2019. a
Kohavi, R. and John, G. H.: Wrappers for Feature Subset Selection, Artific.
Intel., 97, 273–324, https://doi.org/10.1016/S0004-3702(97)00043-X, 1997. a
Kollat, J. B., Reed, P. M., and Wagener, T.: When Are Multiobjective
Calibration Trade-Offs in Hydrologic Models Meaningful?, Water Resour. Res., 48, W03520, https://doi.org/10.1029/2011wr011534, 2012. a
Kong A Siou, L., Johannet, A., Valérie, B. E., and Pistre, S.: Optimization of the Generalization Capability for Rainfall–Runoff Modeling by Neural Networks: The Case of the Lez Aquifer (Southern France), Environ. Earth Sci., 65, 2365–2375, https://doi.org/10.1007/s12665-011-1450-9, 2012. a, b, c
Kong-A-Siou, L., Cros, K., Johannet, A., Borrell-Estupina, V., and Pistre,
S.: KnoX Method, or Knowledge eXtraction from Neural Network Model. Case Study on the Lez Karst Aquifer (Southern France), J. Hydrol., 507, 19–32, https://doi.org/10.1016/j.jhydrol.2013.10.011, 2013. a, b, c
Kong-A-Siou, L., Fleury, P., Johannet, A., Borrell Estupina, V., Pistre, S.,
and Dörfliger, N.: Performance and Complementarity of Two Systemic Models
(Reservoir and Neural Networks) Used to Simulate Spring Discharge and
Piezometry for a Karst Aquifer, J. Hydrol., 519, 3178–3192,
https://doi.org/10.1016/j.jhydrol.2014.10.041, 2014. a, b, c, d
Kong-A-Siou, L., Johannet, A., Borrell Estupina, V., and Pistre, S.: Neural
Networks for Karst Groundwater Management: Case of the Lez Spring (Southern France), Environ. Earth Sci., 74, 7617–7632, https://doi.org/10.1007/s12665-015-4708-9, 2015. a
Kovačič, G., Petrič, M., and Ravbar, N.: Evaluation and
Quantification of the Effects of Climate and Vegetation Cover Change on Karst Water Sources: Case Studies of Two Springs in South-Western Slovenia, Water, 12, 3087, https://doi.org/10.3390/w12113087, 2020. a, b
Kratzert, F., Klotz, D., Brenner, C., Schulz, K., and Herrnegger, M.:
Rainfall–Runoff Modelling Using Long Short-Term Memory (LSTM) Networks, Hydrol. Earth Syst. Sci., 22, 6005–6022, https://doi.org/10.5194/hess-22-6005-2018, 2018. a
Lähivaara, T., Malehmir, A., Pasanen, A., Kärkkäinen, L., Huttunen, J. M. J., and Hesthaven, J. S.: Estimation of Groundwater Storage from Seismic Data Using Deep Learning, Geophys. Prospect., 67, 2115–2126,
https://doi.org/10.1111/1365-2478.12831, 2019. a
Lebigot, E. O.: Uncertainties: A Python Package for Calculations with Uncertainties, https://pythonhosted.org/uncertainties/numpy_guide.html (last access: 18 February 2021), 2010. a
LeCun, Y., Bengio, Y., and Hinton, G.: Deep Learning, Nature, 521, 436–444,
https://doi.org/10.1038/nature14539, 2015. a, b
Li, Q., Cai, W., Wang, X., Zhou, Y., Feng, D. D., and Chen, M.: Medical Image
Classification with Convolutional Neural Network, in: 2014 13th International Conference on Control Automation Robotics Vision (ICARCV), 10–12 December 2014, Singapore, 844–848, https://doi.org/10.1109/ICARCV.2014.7064414, 2014. a
Longenecker, J., Bechtel, T., Chen, Z., Goldscheider, N., Liesch, T., and
Walter, R.: Correlating Global Precipitation Measurement Satellite Data with Karst Spring Hydrographs for Rapid Catchment Delineation, Geophys. Res. Lett., 44, 4926–4932, https://doi.org/10.1002/2017GL073790, 2017. a, b
Louppe, G., Wehenkel, L., Sutera, A., and Geurts, P.: Understanding Variable
Importances in Forests of Randomized Trees, in: Advances in Neural Information Processing Systems, vol. 26, Curran Associates, Inc.,
https://papers.nips.cc/paper/2013/hash/e3796ae838835da0b6f6ea37bcf8bcb7-Abstract.html
(last access: 30 January 2022), 2013. a
Maier, H. R. and Dandy, G. C.: Neural Networks for the Prediction and
Forecasting of Water Resources Variables: A Review of Modelling Issues and
Applications, Environ. Model. Softw., 15, 101–124, https://doi.org/10.1016/s1364-8152(99)00007-9, 2000. a
Maier, H. R., Jain, A., Dandy, G. C., and Sudheer, K.: Methods Used for the
Development of Neural Networks for the Prediction of Water Resource Variables
in River Systems: Current Status and Future Directions, Environ. Model. Softw., 25, 891–909, https://doi.org/10.1016/j.envsoft.2010.02.003, 2010. a
Malard, A., Jeannin, P.-Y., Vouillamoz, J., and Weber, E.: An Integrated
Approach for Catchment Delineation and Conduit-Network Modeling in Karst
Aquifers: Application to a Site in the Swiss Tabular Jura, Hydrogeol. J., 23, 1341–1357, https://doi.org/10.1007/s10040-015-1287-5, 2015. a
Mayaud, C., Gabrovšek, F., Blatnik, M., Kogovšek, B., Petrič,
M., and Ravbar, N.: Understanding Flooding in Poljes: A Modelling Perspective, J. Hydrol., 575, 874–889, https://doi.org/10.1016/j.jhydrol.2019.04.092, 2019. a, b
Mazzilli, N., Jourde, H., Guinot, V., Bailly-Comte, V., and Fleury, P.:
Hydrological Modelling of a Karst Aquifer under Active Groundwater Management
Using a Parsimonious Conceptual Model, in: H2Karst, Besançon,
France, https://hal.archives-ouvertes.fr/hal-01844603 (last access: 28 July 2021), 2011. a
McGovern, A., Lagerquist, R., John Gagne, D., Jergensen, G. E., Elmore, K. L., Homeyer, C. R., and Smith, T.: Making the Black Box More Transparent:
Understanding the Physical Implications of Machine Learning, B. Am. Meteorol. Soc., 100, 2175–2199, https://doi.org/10.1175/BAMS-D-18-0195.1, 2019. a, b
McKinney, W.: Data Structures for Statistical Computing in Python, in: Python in Science Conference, Austin, Texas, 56–61,
https://doi.org/10.25080/majora-92bf1922-00a, 2010. a
Müller, J., Park, J., Sahu, R., Varadharajan, C., Arora, B., Faybishenko,
B., and Agarwal, D.: Surrogate Optimization of Deep Neural Networks for
Groundwater Predictions, J. Global Optim., 81, 203–231, https://doi.org/10.1007/s10898-020-00912-0, 2020. a, b
NASA: GPM – Global Precipitation Measurement,
http://www.nasa.gov/mission_pages/GPM/main/index.html (last access: 8 June 2021), 2016. a
Nash, J. E. and Sutcliffe, J. V.: River Flow Forecasting through Conceptual
Models Part I – A Discussion of Principles, J. Hydrol., 10, 282–290, https://doi.org/10.1016/0022-1694(70)90255-6, 1970. a
Nogueira, F.: Bayesian Optimization: Open Source Constrained Global Optimization Tool for Python, GitHub,
https://github.com/fmfn/BayesianOptimization (last access: 15 April 2020) 2014. a
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel,
O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J.,
Passos, A., and Cournapeau, D.: Scikit-Learn: Machine Learning in Python, J. Mach. Learn. Res., 12, 2825–2830, 2011. a
Petrič, M., Kogovšek, J., and Ravbar, N.: Effects of the Vadose Zone
on Groundwater Flow and Solute Transport Characteristics in Mountainous Karst
Aquifers – the Case of the Javorniki–Snežnik Massif (SW Slovenia), Acta Carsolog., 47, 35–51, https://doi.org/10.3986/ac.v47i1.5144, 2018. a
Rajaee, T., Ebrahimi, H., and Nourani, V.: A Review of the Artificial
Intelligence Methods in Groundwater Level Modeling, J. Hydrol., 572, 336–351, https://doi.org/10.1016/j.jhydrol.2018.12.037, 2019. a
Seibert, J.: Multi-criteria calibration of a conceptual runoff model using a genetic algorithm, Hydrol. Earth Syst. Sci., 4, 215–224, https://doi.org/10.5194/hess-4-215-2000, 2000. a
Sezen, C., Bezak, N., Bai, Y., and Šraj, M.: Hydrological Modelling of
Karst Catchment Using Lumped Conceptual and Data Mining Models, J. Hydrol., 576, 98–110, https://doi.org/10.1016/j.jhydrol.2019.06.036, 2019. a
Sit, M., Demiray, B. Z., Xiang, Z., Ewing, G. J., Sermet, Y., and Demir, I.: A Comprehensive Review of Deep Learning Applications in Hydrology and Water
Resources, Water Sci. Technol., 82, 2635–2670, https://doi.org/10.2166/wst.2020.369, 2020. a
SNO KARST: Time Series of Type Hydrology-Hydrogeology in Le Lez
(Méditerranée) Basin – MEDYCYSS Observatory – KARST Observatory Network – OZCAR Critical Zone Network Research Infrastructure, SNO KARST [data set], https://doi.org/10.15148/CFD01A5B-B7FD-41AA-8884-84DBDDAC767E, 2021. a, b
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov,
R.: Dropout: A Simple Way to Prevent Neural Networks from Overfitting, J. Mach. Learn. Res., 15, 1929–1958, 2014. a
Stevanović, Z.: Karst Waters in Potable Water Supply: A Global Scale
Overview, Environ. Earth Sci., 78, 662, https://doi.org/10.1007/s12665-019-8670-9, 2019. a
The pandas development team: Pandas-Dev/Pandas: Pandas 1.0.3, Zenodo [code],
https://doi.org/10.5281/zenodo.3509134, 2020.
a
Thiéry, D. and Bérard, P.: Alimentation en eau de la ville de
Montpellier – captage de la source du Lez – études des relations entre la source et son réservoir aquifère, Tech. rep., BRGM No. 83, SNG 167 LRO, BRGM, http://infoterre.brgm.fr/rapports/83-SGN-167-LRO.pdf (last access: 30 June 2021), 1983. a
Van, S. P., Le, H. M., Thanh, D. V., Dang, T. D., Loc, H. H., and Anh, D. T.:
Deep Learning Convolutional Neural Network in Rainfall–Runoff Modelling, J. Hydroinform., 22, 541–561, https://doi.org/10.2166/hydro.2020.095, 2020. a, b
van der Walt, S., Colbert, S. C., and Varoquaux, G.: The NumPy Array: A
Structure for Efficient Numerical Computation, Comput. Sci. Eng., 13, 22–30, https://doi.org/10.1109/mcse.2011.37, 2011. a
van Rossum, G.: Python Tutorial, https://ir.cwi.nl/pub/5008/05008D.pdf
(last access: 4 May 2022), 1995. a
Wunsch, A.: AndreasWunsch/CNN_KarstSpringModeling: v0.1 (v0.1), Zenodo [code], https://doi.org/10.5281/zenodo.5184692, 2021. a
Wunsch, A., Liesch, T., and Broda, S.: Groundwater Level Forecasting with
Artificial Neural Networks: A Comparison of Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNNs), and Non-Linear Autoregressive Networks with Exogenous Input (NARX), Hydrol. Earth Syst. Sci., 25, 1671–1687, https://doi.org/10.5194/hess-25-1671-2021, 2021. a, b, c, d, e
Yin, W., Kann, K., Yu, M., and Schütze, H.: Comparative Study of CNN and RNN for Natural Language Processing, arXiv preprint: 1702.01923 [cs], http://arxiv.org/abs/1702.01923 (last access: 29 January 2022), 2017. a
Short summary
Modeling complex karst water resources is difficult enough, but often there are no or too few climate stations available within or close to the catchment to deliver input data for modeling purposes. We apply image recognition algorithms to time-distributed, spatially gridded meteorological data to simulate karst spring discharge. Our models can also learn the approximate catchment location of a spring independently.
Modeling complex karst water resources is difficult enough, but often there are no or too few...
