Articles | Volume 26, issue 16
https://doi.org/10.5194/hess-26-4469-2022
© 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-26-4469-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Forward and inverse modeling of water flow in unsaturated soils with discontinuous hydraulic conductivities using physics-informed neural networks with domain decomposition
Life and Environmental Science Department, University of California, Merced, CA, USA
Teamrat A. Ghezzehei
Life and Environmental Science Department, University of California, Merced, CA, USA
Related authors
No articles found.
Samuel N. Araya, Jeffrey P. Mitchell, Jan W. Hopmans, and Teamrat A. Ghezzehei
SOIL, 8, 177–198, https://doi.org/10.5194/soil-8-177-2022, https://doi.org/10.5194/soil-8-177-2022, 2022
Short summary
Short summary
We studied the long-term effects of no-till (NT) and winter cover cropping (CC) practices on soil hydraulic properties. We measured soil water retention and conductivity and also conducted numerical simulations to compare soil water storage abilities under the different systems. Soils under NT and CC practices had improved soil structure. Conservation agriculture practices showed marginal improvement with respect to infiltration rates and water storage.
Jing Yan and Teamrat Ghezzehei
Biogeosciences Discuss., https://doi.org/10.5194/bg-2022-52, https://doi.org/10.5194/bg-2022-52, 2022
Publication in BG not foreseen
Short summary
Short summary
Although hydraulic redistribution (HR) is a well-documented phenomenon, whether it is a passive happy accident or actively controlled by roots is not well understood. Our modeling study suggests HR is long-range feedback between roots that inhabit heterogeneously resourced soil regions. When nutrients and organic matter are concentrated in shallow layers that experience frequent drying, root-exudation facilitated HR allows plants to mineralize and extract the otherwise inaccessible nutrients.
Daniel Rath, Nathaniel Bogie, Leonardo Deiss, Sanjai J. Parikh, Daoyuan Wang, Samantha Ying, Nicole Tautges, Asmeret Asefaw Berhe, Teamrat A. Ghezzehei, and Kate M. Scow
SOIL, 8, 59–83, https://doi.org/10.5194/soil-8-59-2022, https://doi.org/10.5194/soil-8-59-2022, 2022
Short summary
Short summary
Storing C in subsoils can help mitigate climate change, but this requires a better understanding of subsoil C dynamics. We investigated changes in subsoil C storage under a combination of compost, cover crops (WCC), and mineral fertilizer and found that systems with compost + WCC had ~19 Mg/ha more C after 25 years. This increase was attributed to increased transport of soluble C and nutrients via WCC root pores and demonstrates the potential for subsoil C storage in tilled agricultural systems.
Samuel N. Araya, Anna Fryjoff-Hung, Andreas Anderson, Joshua H. Viers, and Teamrat A. Ghezzehei
Hydrol. Earth Syst. Sci., 25, 2739–2758, https://doi.org/10.5194/hess-25-2739-2021, https://doi.org/10.5194/hess-25-2739-2021, 2021
Short summary
Short summary
We took aerial photos of a grassland area using an unoccupied aerial vehicle and used the images to estimate soil moisture via machine learning. We were able to estimate soil moisture with high accuracy. Furthermore, by analyzing the machine learning models we developed, we learned how different factors drive the distribution of moisture across the landscape. Among the factors, rainfall, evapotranspiration, and topography were most important in controlling surface soil moisture distribution.
Jing Yan, Nathaniel A. Bogie, and Teamrat A. Ghezzehei
Biogeosciences, 17, 6377–6392, https://doi.org/10.5194/bg-17-6377-2020, https://doi.org/10.5194/bg-17-6377-2020, 2020
Short summary
Short summary
An uneven supply of water and nutrients in soils often drives how plants behave. We observed that plants extract all their required nutrients from dry soil patches in sufficient quantity, provided adequate water is available elsewhere in the root zone. Roots in nutrient-rich dry patches facilitate the nutrient acquisition by extensive growth, water release, and modifying water retention in their immediate environment. The findings are valuable in managing nutrient losses in agricultural systems.
Teamrat A. Ghezzehei, Benjamin Sulman, Chelsea L. Arnold, Nathaniel A. Bogie, and Asmeret Asefaw Berhe
Biogeosciences, 16, 1187–1209, https://doi.org/10.5194/bg-16-1187-2019, https://doi.org/10.5194/bg-16-1187-2019, 2019
Short summary
Short summary
Soil water is a medium from which microbes acquire resources and within which they are able to move. Occupancy and availability of water and oxygen gas in soils are mutually exclusive. In addition, as soil dries the remaining water is held with an increasing degree of adhesive energy, which restricts microbes' ability to extract resources from water. We introduce a mathematical model that describes these interacting effects and organic matter decomposition.
Mehdi Rahmati, Lutz Weihermüller, Jan Vanderborght, Yakov A. Pachepsky, Lili Mao, Seyed Hamidreza Sadeghi, Niloofar Moosavi, Hossein Kheirfam, Carsten Montzka, Kris Van Looy, Brigitta Toth, Zeinab Hazbavi, Wafa Al Yamani, Ammar A. Albalasmeh, Ma'in Z. Alghzawi, Rafael Angulo-Jaramillo, Antônio Celso Dantas Antonino, George Arampatzis, Robson André Armindo, Hossein Asadi, Yazidhi Bamutaze, Jordi Batlle-Aguilar, Béatrice Béchet, Fabian Becker, Günter Blöschl, Klaus Bohne, Isabelle Braud, Clara Castellano, Artemi Cerdà, Maha Chalhoub, Rogerio Cichota, Milena Císlerová, Brent Clothier, Yves Coquet, Wim Cornelis, Corrado Corradini, Artur Paiva Coutinho, Muriel Bastista de Oliveira, José Ronaldo de Macedo, Matheus Fonseca Durães, Hojat Emami, Iraj Eskandari, Asghar Farajnia, Alessia Flammini, Nándor Fodor, Mamoun Gharaibeh, Mohamad Hossein Ghavimipanah, Teamrat A. Ghezzehei, Simone Giertz, Evangelos G. Hatzigiannakis, Rainer Horn, Juan José Jiménez, Diederik Jacques, Saskia Deborah Keesstra, Hamid Kelishadi, Mahboobeh Kiani-Harchegani, Mehdi Kouselou, Madan Kumar Jha, Laurent Lassabatere, Xiaoyan Li, Mark A. Liebig, Lubomír Lichner, María Victoria López, Deepesh Machiwal, Dirk Mallants, Micael Stolben Mallmann, Jean Dalmo de Oliveira Marques, Miles R. Marshall, Jan Mertens, Félicien Meunier, Mohammad Hossein Mohammadi, Binayak P. Mohanty, Mansonia Pulido-Moncada, Suzana Montenegro, Renato Morbidelli, David Moret-Fernández, Ali Akbar Moosavi, Mohammad Reza Mosaddeghi, Seyed Bahman Mousavi, Hasan Mozaffari, Kamal Nabiollahi, Mohammad Reza Neyshabouri, Marta Vasconcelos Ottoni, Theophilo Benedicto Ottoni Filho, Mohammad Reza Pahlavan-Rad, Andreas Panagopoulos, Stephan Peth, Pierre-Emmanuel Peyneau, Tommaso Picciafuoco, Jean Poesen, Manuel Pulido, Dalvan José Reinert, Sabine Reinsch, Meisam Rezaei, Francis Parry Roberts, David Robinson, Jesús Rodrigo-Comino, Otto Corrêa Rotunno Filho, Tadaomi Saito, Hideki Suganuma, Carla Saltalippi, Renáta Sándor, Brigitta Schütt, Manuel Seeger, Nasrollah Sepehrnia, Ehsan Sharifi Moghaddam, Manoj Shukla, Shiraki Shutaro, Ricardo Sorando, Ajayi Asishana Stanley, Peter Strauss, Zhongbo Su, Ruhollah Taghizadeh-Mehrjardi, Encarnación Taguas, Wenceslau Geraldes Teixeira, Ali Reza Vaezi, Mehdi Vafakhah, Tomas Vogel, Iris Vogeler, Jana Votrubova, Steffen Werner, Thierry Winarski, Deniz Yilmaz, Michael H. Young, Steffen Zacharias, Yijian Zeng, Ying Zhao, Hong Zhao, and Harry Vereecken
Earth Syst. Sci. Data, 10, 1237–1263, https://doi.org/10.5194/essd-10-1237-2018, https://doi.org/10.5194/essd-10-1237-2018, 2018
Short summary
Short summary
This paper presents and analyzes a global database of soil infiltration data, the SWIG database, for the first time. In total, 5023 infiltration curves were collected across all continents in the SWIG database. These data were either provided and quality checked by the scientists or they were digitized from published articles. We are convinced that the SWIG database will allow for a better parameterization of the infiltration process in land surface models and for testing infiltration models.
Roland Baatz, Pamela L. Sullivan, Li Li, Samantha R. Weintraub, Henry W. Loescher, Michael Mirtl, Peter M. Groffman, Diana H. Wall, Michael Young, Tim White, Hang Wen, Steffen Zacharias, Ingolf Kühn, Jianwu Tang, Jérôme Gaillardet, Isabelle Braud, Alejandro N. Flores, Praveen Kumar, Henry Lin, Teamrat Ghezzehei, Julia Jones, Henry L. Gholz, Harry Vereecken, and Kris Van Looy
Earth Syst. Dynam., 9, 593–609, https://doi.org/10.5194/esd-9-593-2018, https://doi.org/10.5194/esd-9-593-2018, 2018
Short summary
Short summary
Focusing on the usage of integrated models and in situ Earth observatory networks, three challenges are identified to advance understanding of ESD, in particular to strengthen links between biotic and abiotic, and above- and below-ground processes. We propose developing a model platform for interdisciplinary usage, to formalize current network infrastructure based on complementarities and operational synergies, and to extend the reanalysis concept to the ecosystem and critical zone.
T. A. Ghezzehei, D. V. Sarkhot, and A. A. Berhe
Solid Earth, 5, 953–962, https://doi.org/10.5194/se-5-953-2014, https://doi.org/10.5194/se-5-953-2014, 2014
Related subject area
Subject: Vadose Zone Hydrology | Techniques and Approaches: Mathematical applications
Differentiating between crop and soil effects on soil moisture dynamics
Parametric soil water retention models: a critical evaluation of expressions for the full moisture range
Hydraulic and transport parameter assessment using column infiltration experiments
On the consistency of scale among experiments, theory, and simulation
Solar-forced diurnal regulation of cave drip rates via phreatophyte evapotranspiration
Multi-scale analysis of bias correction of soil moisture
Generalized analytical solution for advection-dispersion equation in finite spatial domain with arbitrary time-dependent inlet boundary condition
Helen Scholz, Gunnar Lischeid, Lars Ribbe, Ixchel Hernandez Ochoa, and Kathrin Grahmann
Hydrol. Earth Syst. Sci., 28, 2401–2419, https://doi.org/10.5194/hess-28-2401-2024, https://doi.org/10.5194/hess-28-2401-2024, 2024
Short summary
Short summary
Sustainable management schemes in agriculture require knowledge of site-specific soil hydrological processes, especially the interplay between soil heterogeneities and crops. We disentangled such effects on soil moisture in a diversified arable field with different crops and management schemes by applying a principal component analysis. The main effects on soil moisture variability were quantified. Meteorological drivers, followed by different seasonal behaviour of crops, had the largest impact.
Raneem Madi, Gerrit Huibert de Rooij, Henrike Mielenz, and Juliane Mai
Hydrol. Earth Syst. Sci., 22, 1193–1219, https://doi.org/10.5194/hess-22-1193-2018, https://doi.org/10.5194/hess-22-1193-2018, 2018
Short summary
Short summary
Water flows through soils with more difficulty when the soil is dried out. Scant rainfall in deserts may therefore result in a seemingly wet soil, but the water will often not penetrate deeply enough to replenish the groundwater. We compared the mathematical functions that describe how well different soils hold their water and found that only a few of them are realistic. The function one chooses to model the soil can have a large impact on the estimate of groundwater recharge.
Anis Younes, Thierry Mara, Marwan Fahs, Olivier Grunberger, and Philippe Ackerer
Hydrol. Earth Syst. Sci., 21, 2263–2275, https://doi.org/10.5194/hess-21-2263-2017, https://doi.org/10.5194/hess-21-2263-2017, 2017
Short summary
Short summary
The estimation of flow and solute transport in unsaturated soil is essential for quantifying groundwater resources or pollution. Usual column laboratory experiments and a new method are analyzed using a global sensitivity analysis. The data sets are composed of water pressure and water content measured inside the column and water flow rate and solute BTC measured at the outflow. Non-invasive methods (using flow rate and BTC only) provide comparable results than usual invasive methods.
James E. McClure, Amanda L. Dye, Cass T. Miller, and William G. Gray
Hydrol. Earth Syst. Sci., 21, 1063–1076, https://doi.org/10.5194/hess-21-1063-2017, https://doi.org/10.5194/hess-21-1063-2017, 2017
Short summary
Short summary
A complicating factor in describing the flow of two immiscible fluids in a porous medium is ensuring that experiments, theory, and simulation are all formulated at the same length scale. We have quantitatively analyzed the internal structure of a two-fluid system including the distribution of phases and the location of interfaces between phases. The data we have obtained allow for a clearer definition of capillary pressure at the averaged scale as a state function that describes the system.
Katie Coleborn, Gabriel C. Rau, Mark O. Cuthbert, Andy Baker, and Owen Navarre
Hydrol. Earth Syst. Sci., 20, 4439–4455, https://doi.org/10.5194/hess-20-4439-2016, https://doi.org/10.5194/hess-20-4439-2016, 2016
Short summary
Short summary
This is the first observation of tree water use in cave drip water. Our novel time series analysis using the synchrosqueeze transform identified daily and sub-daily oscillations in drip rate. The only hypothesis consistent with hydrologic theory and the data was that the oscillations were caused by solar driven pumping by trees above the cave. We propose a new protocol for inferring karst architecture and our findings support research on the impact trees on speleothem paleoclimate proxies.
C.-H. Su and D. Ryu
Hydrol. Earth Syst. Sci., 19, 17–31, https://doi.org/10.5194/hess-19-17-2015, https://doi.org/10.5194/hess-19-17-2015, 2015
Short summary
Short summary
Global environmental monitoring requires geophysical measurements from a variety of sources and sensors to close the information gap. This paper proposes a novel approach for analysing temporal scale-by-scale differences (biases and errors) between geophysical estimates from disparate sources. This allows assessment of different bias correction schemes, and forms the basis for a multi-scale bias correction scheme and data-adaptive, non-linear de-noising.
J.-S. Chen and C.-W. Liu
Hydrol. Earth Syst. Sci., 15, 2471–2479, https://doi.org/10.5194/hess-15-2471-2011, https://doi.org/10.5194/hess-15-2471-2011, 2011
Cited articles
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado,
G. S., Davis, A., J., D., Devin, M., Ghemawat, S., I., G., Harp, A., Irving,
G., Isard, M., Jozefowicz, R., Jia, Y., Kaiser, L., Kudlur, M., Levenberg,
J., Mané, D., Schuster, M., Monga, R., Moore, S., Murray, D., Olah, C.,
Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke,
V., Vasudevan, V., Viégas, F., Vinyals, O., Warden, P., Wattenberg, M.,
Wicke, M., Yu, Y., and Zheng, X.: TensorFlow: Large-scale machine learning
on heterogeneous distributed systems, https://www.tensorflow.org/ (last access: 28 August 2022), 2015. a, b, c
Assouline, S. and Or, D.: Conceptual and parametric representation of soil
hydraulic properties: A review, Vadose Zone J., 12, 1–20,
https://doi.org/10.2136/vzj2013.07.0121, 2013. a
Babaeian, E., Sadeghi, M., Jones, S. B., Montzka, C., Vereecken, H., and
Tuller, M.: Ground, proximal, and satellite remote sensing of soil moisture, Rev. Geophys., 57, 530–616, https://doi.org/10.1029/2018RG000618, 2019. a, b
Bandai, T. and Ghezzehei, T. A.: Physics‐informed neural networks with
monotonicity constraints for Richardson–Richards equation: Estimation of
constitutive relationships and soil water flux density from volumetric water
content measurements, Water Resour. Res., 57, e2020WR027642,
https://doi.org/10.1029/2020WR027642, 2021. a, b, c
Bandai, T. and Ghezzehei, T. A.: Author comment on “Forward and inverse
modeling of water flow in unsaturated soils with discontinuous hydraulic
conductivities using physics-informed neural networks with domain decomposition” by Toshiyuki Bandai and Teamrat A. Ghezzehei,
https://doi.org/10.5194/hess-2022-73-AC2, 2022a. a
Bandai, T. and Ghezzehei, T. A.: DD-PINNS-RRE, Zenodo [code],
https://doi.org/10.5281/zenodo.6030635, 2022b. a, b, c
Baydin, A. G., Pearlmutter, B. A., Radul, A. A., and Siskind, J. M.: Automatic differentiation in machine learning: a survey, J. Mach. Learn. Res., 18, 1–43, https://doi.org/10.1016/j.advwatres.2018.01.009, 2018. a, b
Bengio, Y.: Practical recommendations for gradient-based training of deep
architectures, in: Neural Networks: Tricks of the Trade, 2nd Edn., edited by: Montavon, G., Orr, G. B., and Müller, K.-R., Springer, 437–478,
https://doi.org/10.1007/978-3-642-35289-8, 2012. a
Brocca, L., Moramarco, T., Melone, F., and Wagner, W.: A new method for
rainfall estimation through soil moisture observations, Geophys. Res. Lett.,
40, 853–858, https://doi.org/10.1002/grl.50173, 2013. a
Brooks, R. H. and Corey, A. T.: Hydraulic properties of porous media, Hydrology Papers 3, Colorado State University, 1964. a
Brunone, B., Ferrante, M., Romano, N., and Santini, A.: Numerical simulations
of one-dimensional infiltration into layered soils with the Richards equation
using different estimates of the interlayer conductivity, Vadose Zone J., 2,
193–200, https://doi.org/10.2113/2.2.193, 2003. a
Buckingham, E.: Studies on the movement of soil moisture, Bull. 38, USDA,
Bureau of Soils, 1907. a
Byrd, R. H., Lu, P., Nocedal, J., and Zhu, C.: A limited memory algorithm for
bound constrained optimization, J. Sci. Comput., 16, 1190–1208,
https://doi.org/10.1137/0916069, 1995. a
Celia, M. A., Bouloutas, E. T., and Zarba, R. L.: A general mass-conservative numerical solution for the unsaturated flow equation, Water Resour. Res., 26, 1483–1496, https://doi.org/10.1029/WR026i007p01483, 1990. a
Clément, J.-B., Golay, F., Ersoy, M., and Sous, D.: An adaptive strategy
for discontinuous Galerkin simulations of Richards' equation: Application to
multi-materials dam wetting, Adv. Water Resour., 151, 103897,
https://doi.org/10.1016/j.advwatres.2021.103897, 2021. a
Cybenko, G.: Approximation by superpositions of a sigmoidal function, Math.
Control Sig. Syst., 2, 303–314, https://doi.org/10.1007/BF02551274, 1989. a
Depina, I., Jain, S., Valsson, S. M., and Gotovac, H.: Application of
physics-informed neural networks to inverse problems in unsaturated
groundwater flow, Georisk, 16, 21–36, https://doi.org/10.1080/17499518.2021.1971251, 2021. a
Dijkema, J., Koonce, J. E., Shillito, R. M., Ghezzehei, T. A., Berli, M.,
van der Ploeg, M. J., and van Genuchten, M. T.: Water distribution in an
arid zone soil: Numerical analysis of data from a large weighing lysimeter,
Vadose Zone J., 1–17, https://doi.org/10.2136/vzj2017.01.0035, 2017. a
Farthing, M. W. and Ogden, F. L.: Numerical solution of Richards' equation: A
Review of advances and challenges, Soil Sci. Soc. Am. J., 81, 1257–1269,
https://doi.org/10.2136/sssaj2017.02.0058, 2017. a
Fuks, O. and Tchelepi, H. A.: Limitations of physics informed machine learning for nonlinear two-phase transport in porous media, J. Mach. Learn. Model. Comput., 1, 19–37, https://doi.org/10.1615/.2020033905, 2020. a
Gardner, W. R.: Some steady-state solutions of the unsaturated moisture flow
equation with application to evaporation from a water table, Soil Sci., 85,
228–232, 1958. a
Ghezzehei, T. A., Kneafsey, T. J., and Su, G. W.: Correspondence of the
Gardner and van Genuchten-Mualem relative permeability function parameters,
Water Resour. Res., 43, 1–7, https://doi.org/10.1029/2006WR005339, 2007. a
Glorot, X. and Bengio, Y.: Understanding the difficulty of training deep
feedforward neural networks, in: Proceedings of the thirteenth international
conference on artificial intelligence and statistics, 13–15 May 2010, Sardinia, Italy, 249–256, 2010. a
He, Q., Brajas-solano, D., Tartakovsky, G., and Tartakovsky, A. M.:
Physics-informed neural networks for multiphysics data assimilation with
application to subsurface transport, Adv. Water Resour., 141, 103610,
https://doi.org/10.1016/j.advwatres.2020.103610, 2020. a, b
Hornik, K.: Approximation capabilities of multilayer neural network, Neural
Networks, 4, 251–257, 1991. a
Jagtap, A. D. and Karniadakis, G. E.: Extended physics-informed neural
networks (XPINNs): A generalized space-time domain decomposition based deep
learning framework for nonlinear partial differential equations, Commun.
Comput. Phys., 28, 2002–2041, https://doi.org/10.4208/cicp.oa-2020-0164, 2020. a, b, c, d, e, f, g
Jin, X., Cai, S., Li, H., and Karniadakis, G. E.: NSFnets (Navier-Stokes flow
nets): Physics-informed neural networks for the incompressible Navier-Stokes
equations, J. Comput. Phys., 426, 109951, https://doi.org/10.1016/j.jcp.2020.109951, 2021. a
Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S., and
Yang, L.: Physics-informed machine learning, Nat. Rev. Phys., 3, 422–440,
https://doi.org/10.1038/s42254-021-00314-5, 2021. a, b, c, d
Kingma, D. P. and Ba, J. B.: Adam: A method for stochastic optimization, in:
Proceedings of the 3rd international conference on learning representations,
7–9 May 2014, San Diego, CA, USA, https://doi.org/10.48550/arXiv.1412.6980, 2014. a
Kosugi, K.: Lognormal distribution model for unsaturated soil hydraulic
properties, Water Resour. Res., 32, 2697–2703, https://doi.org/10.1029/96WR01776,
1996. a
Labolle, E. M. and Clausnitzer, V.: Comment on Russo [1991], Serrano [1990,
1998], and other applications of the water-content-based form of Richards'
equation to heterogeneous soils, Water Resour. Res., 35, 605–607, 1999. a
Lagaris, I. E., Likas, A., and Fotiadis, D. I.: Artificial neural networks for solving ordinary and partial differential equations, IEEE T. Neural Netw. Learn. Syst., 9, 987–1000, https://doi.org/10.1109/72.712178, 1998. a, b
LeCun, Y., Bottou, L., Orr, G. B., and Müller, K.-R.: Efficient
backprop, in: Neural Networks: Tricks of the Trade, 2nd Edn., edited by: Montavon, G., Orr, G. B., and Müller, K.-R., Springer, 9–48, https://doi.org/10.1007/978-3-642-35289-8_3, 2012. a, b
Li, N., Yue, X. Y., and Wang, W. K.: Inverse estimation of spatiotemporal flux boundary conditions in unsaturated water flow modeling, Water Resour. Res., 57, e2020WR028030, https://doi.org/10.1029/2020WR028030, 2021. a
Li, Z., Kovachki, N., Azizzadenesheli, K., Liu, B., Bhattacharya, K., Stuart,
A., and Anandkumar, A.: Fourier neural operator for parametric partial differential equations, in: International conference on learning and
representations, 4 May 2021, Vienna, Austria, https://resolver.caltech.edu/CaltechAUTHORS:20201106-120140981 (last access: 30 August 2022), 2021. a
Lu, L., Jin, P., Pang, G., Zhang, Z., and Karniadakis, G. E.: Learning
nonlinear operators via DeepONet based on the universal approximation theorem
of operators, Nat. Mach. Intel., 3, 218–229, https://doi.org/10.1038/s42256-021-00302-5, 2021a. a
Lu, L., Meng, X., Mao, Z., and Karniadakis, G. E.: DeepXDE: A deep learning
library for solving differential equations, SIAM Rev., 63, 208–228,
https://doi.org/10.1137/19M1274067, 2021b. a, b
Mishra, S. and Molinaro, R.: Estimates on the generalization error of physics-informed neural networks for approximating PDEs, IMA J. Numer. Anal., drab093, https://doi.org/10.1093/imanum/drab093, 2022. a, b, c, d
Mitra, K. and Vohralík, M.: A posteriori error estimates for the
Richards equation, arXiv [preprint], https://doi.org/10.48550/arXiv.2108.12507, 2021. a
Mualem, Y.: A new model for predicting the hydraulic conductivity of
unsaturated porous media, Water Resour. Res., 12, 513–522, https://doi.org/10.1029/WR012i003p00513, 1976. a
Nearing, G. S., Kratzert, F., Sampson, A. K., Pelissier, C. S., Klotz, D.,
Frame, J. M., Prieto, C., and Gupta, H. V.: What role does hydrological
science play in the age of machine learning?, Water Resour. Res., 57,
e2020WR028091, https://doi.org/10.1029/2020WR028091, 2021. a
Paniconi, C. and Putti, M.: Physically based modeling in catchment hydrology
at 50: Survey and outlook, Water Resour. Res., 51, 7090–7129,
https://doi.org/10.1002/2015WR017780, 2015. a
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Köpf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., and Chintala, S.: PyTorch: An imperative style,
high-performance deep learning library, in: 33rd conference on neural
information processing systems, 8–14 December 2019, Vancouver, Canada, https://doi.org/10.48550/arXiv.1912.01703, 2019. a
Petra, N. and Stadler, G.: Model variational inverse problems governed by
partial differential equations, Tech. rep., The Institute for Computational
Engineering and Sciences, The University of Texas at Austin, 2011. a
Philip, J. R.: Theory of infiltration, Adv. Hydrosci., 5, 215–296,
https://doi.org/10.1016/B978-1-4831-9936-8.50010-6, 1969. a
Pries, C. E. H., Castanha, C., Porras, R. C., and Torn, M. S.: The whole-soil
carbon flux in response to warming, Science, 359, 1420–1423,
https://doi.org/10.1126/science.aao0457, 2017. a
Psaros, A. F., Meng, X., Zou, Z., Guo, L., and Karniadakis, G. E.: Uncertainty quantification in scientific machine learning: Methods, metrics, and comparisons, arXiv [preprint], arXiv:2201.07766v1, https://doi.org/10.48550/arXiv.2201.07766, 2022. a
Radu, F. A., Pop, I. S., and Knabner, P.: Error estimates for a mixed finite
element discretization of some degenerate parabolic equations, Numer. Math.,
109, 285–311, https://doi.org/10.1007/s00211-008-0139-9, 2008. a
Raghu, M., Poole, B., Kleinberg, J., Ganguli, S., and Dickstein, J. S.: On the expressive power of deep neural networks, in: 34th International Conference on Machine Learning, 6–11 August 2017, Sydney, Australia, https://doi.org/10.48550/arXiv.1606.05336, 2017. a
Raissi, M., Perdikaris, P., and Karniadakis, G. E.: Physics-informed neural
networks: A deep learning framework for solving forward and inverse problems
involving nonlinear partial differential equations, J. Comput. Phys., 378, 686–707, https://doi.org/10.1016/j.jcp.2018.10.045, 2019. a, b
Raissi, M., Yazdani, A., and Karniadakis, G. E.: Hidden fluid mechanics:
Learning velocity and pressure fields from flow visualizations, Science, 367, 1026–1030, https://doi.org/10.1126/science.aaw4741, 2020. a
Richards, L. A.: Capillary conduction of liquids through porous mediums,
Physics, 1, 318–333, https://doi.org/10.1063/1.1745010, 1931. a, b
Richardson, L. F.: Weather prediction by numerical process, Cambridge
University Press, https://doi.org/10.1017/CBO9780511618291, 1922. a, b
Robinson, D. A., Campbell, C. S., Hopmans, J. W., Hornbuckle, B. K., Jones, S. B., Knight, R., Ogden, F., Selker, J., and Wendroth, O.: Soil moisture
measurement for ecological and hydrological watershed-scale observatories: A
review, Vadose Zone J., 7, 358–389, https://doi.org/10.2136/vzj2007.0143, 2008. a
Sadeghi, M., Tuller, M., Warrick, A., Babaeian, E., Parajuli, K., Gohardoust,
M. R., and Jones, S. B.: An analytical model for estimation of land surface
net water flux from near-surface soil moisture observations, J. Hydrol., 570, 26–37, https://doi.org/10.1016/j.jhydrol.2018.12.038, 2019. a
Scudeler, C., Putti, M., and Paniconi, C.: Mass-conservative reconstruction of Galerkin velocity fields for transport simulations, Adv. Water Resour., 94, 470–485, https://doi.org/10.1016/j.advwatres.2016.06.011, 2016. a
Sheng, W., Zhou, R., Sadeghi, M., Babaeian, E., Robinson, D. A., Tuller, M.,
and Jones, S. B.: A TDR array probe for monitoring near-surface soil moisture distribution, Vadose Zone J., 16, 1–8, https://doi.org/10.2136/vzj2016.11.0112, 2017. a
Shin, Y., Darbon, J., and Karniadakis, G. E.: On the convergence of physics
informed neural networks for linear second-order elliptic and parabolic type
PDEs, Commun. Comput. Phys., 28, 2042–2074, https://doi.org/10.4208/cicp.oa-2020-0193, 2020. a, b
Šimůnek, J., Šejna, M., Saito, H., Sakai, M., and van Genuchten, M. T.: The HYDRUS-1D software package for simulating the one-dimensional
movement of water, heat, and multiple solutes in variably saturated media,
Version 4.17, Department of Environmental Sciences, University of California
Riverside, Riverside, 2013. a, b, c, d, e
Sun, L., Gao, H., Pan, S., and Wang, J.: Surrogate modeling for fluid flows
based on physics-constrained deep learning without simulation data, Comput.
Meth. Appl. Mech. Eng., 361, 112732, https://doi.org/10.1016/j.cma.2019.112732, 2020. a, b
Tartakovsky, A. M., Marrero, C. O., Perdikaris, P., Tartakovsky, G. D., and
Barajas-Solano, D.: Physics-informed deep neural networks for learning
parameters and constitutive relationships in subsurface flow problems, Water
Resour. Res., 56, e2019WR026731, https://doi.org/10.1029/2019wr026731, 2020.
a
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/10.1016/j.pan.2017.07.214, 1980. a, b, c
Vereecken, H., Huisman, J. A., Bogena, H., Vanderborght, J., Vrugt, J. A., and Hopmans, J. W.: On the value of soil moisture measurements in vadose zone
hydrology: A review, Water Resour. Res., 44, W00D06, https://doi.org/10.1029/2008WR006829, 2008. a
Virtanen, P., Gommers, R., Oliphant, T. E., Haberland, M., Reddy, T.,
Cournapeau, D., Burovski, E., Peterson, P., Weckesser, W., Bright, J.,
van der Walt, S. J., Brett, M., Wilson, J., Millman, K. J., Mayorov, N.,
Nelson, A. R. J., Jones, E., Kern, R., Larson, E., Carey, C. J., Polat, I.,
Feng, Y., Moore, E. W., VanderPlas, J., Laxalde, D., Perktold, J., Cimrman,
R., Henriksen, I., Quintero, E. A., Harris, C. R., Archibald, A. M., Ribeiro,
A. H., Pedregosa, F., van Mulbregt, P., and SciPy 1.0 Contributors: fundamental algorithms for scientific computing in Python, Nat. Methods, 17, 261–272, https://doi.org/10.1038/s41592-019-0686-2, 2020. a
Short summary
Scientists use a physics-based equation to simulate water dynamics that influence hydrological and ecological phenomena. We present hybrid physics-informed neural networks (PINNs) to leverage the growing availability of soil moisture data and advances in machine learning. We showed that PINNs perform comparably to traditional methods and enable the estimation of rainfall rates from soil moisture. However, PINNs are challenging to train and significantly slower than traditional methods.
Scientists use a physics-based equation to simulate water dynamics that influence hydrological...