You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Water
Download PDF
  • Review
  • Open Access

11 August 2025

A Systematic Literature Review of MODFLOW Combined with Artificial Neural Networks (ANNs) for Groundwater Flow Modelling

,
,
,
,
and
1
Energy Cluster, School of Advanced Engineering, UPES, Dehradun 248007, India
2
Department of Civil Engineering, College of Engineering, Kyung Hee University, 1732 Deogyeong-daero, Giheung-gu, Yongin-si 17104, Republic of Korea
3
Land and Housing Research Institute, Smart Climate Environment Research Center, 99, Expo-ro 539beon-gil, Yuseong-gu, Daejeon 34047, Republic of Korea
*
Author to whom correspondence should be addressed.
Water2025, 17(16), 2375;https://doi.org/10.3390/w17162375
This article belongs to the Special Issue Application of Hydrological Modelling to Water Resources Management
Version Notes
Order Reprints

Abstract

The sustainable management of global groundwater resources is increasingly challenged by climatic uncertainty and escalating anthropogenic stress. Thus, there is a need for simulation tools that are more robust and flexible. This systematic review addresses the integration of two dominant modeling paradigms: the physically grounded Modular Finite-Difference Flow (MODFLOW) model and the data-agile Artificial Neural Network (ANN). While the MODFLOW model provides deep process-based understanding, it is often limited by extensive data requirements and computational intensity. In contrast, an ANN offers remarkable predictive accuracy and computational efficiency, particularly in complex, non-linear systems, but traditionally lacks physical interpretability. This review synthesizes existing research to present a functional classification framework for MODFLOW–ANN integration, providing a systematic analysis of the literature within this structure. Our analysis of the literature, sourced from Scopus, Web of Science, and Google Scholar reveals a clear trend of the strategic integration of these models, representing a new trend in hydrogeological simulation. The literature reveals a classification framework that categorizes the primary integration strategies into three distinct approaches: (1) training an ANN on MODFLOW model outputs to create computationally efficient surrogate models; (2) using an ANN to estimate physical parameters for improved MODFLOW model calibration; and (3) applying ANNs as post-processors to correct systematic errors in MODFLOW model simulations. Our analysis reveals that these hybrid methods consistently outperform standalone approaches by leveraging ANNs for computational acceleration through surrogate modeling, for enhanced model calibration via intelligent parameter estimation, and for improved accuracy through systematic error correction.

1. Introduction

Groundwater modeling is essential for the management of present-day water resources. It provides an essential tool for protecting water quality, forecasting aquifer dynamics, and understanding complex subsurface hydrological systems. The field of groundwater modeling has evolved significantly over time, progressing from early empirical methods to the sophisticated computational tools available today. Early civilizations, such as the Romans and Persians, engineered advanced water management systems like aqueducts and qanats. The design and placement of these structures were based on empirical observations of local geology and hydrological conditions []. The scientific formalization of groundwater hydrology began in the 19th century with Henry Darcy’s foundational work on water movement through porous media []. This foundation was expanded upon by pioneers such as Jules Dupuit and Philip Forchheimer, who developed fundamental theories of well hydraulics and aquifer behavior [,]. A major advancement came in 1935 with the Theis equation, which provided the first analytical method for modeling transient (time-varying) groundwater flow [].
The advancement of digital computing in the latter half of the 20th century caused a fundamental shift in groundwater modeling, enabling the numerical solution of the complex partial differential equations that govern groundwater flow []. This computational shift led the development of advanced, physics-based simulation models []. Most important among them was the U.S. Geological Survey’s MODFLOW model, first released in 1984, which has since then emerged as a global standard for groundwater modeling [,]. The primary strengths of the MODFLOW model are its modular structure and its foundation in physical principles, enabling it to simulate a wide variety of aquifer systems and hydrological conditions [,]. Despite its physical basis, the model has notable limitations. It demands large datasets for calibration, is often computationally expensive, and requires a time-consuming iterative process to match simulation results to field observations [].
In parallel, data-driven methods, particularly Artificial Neural Networks (ANNs), have emerged as a powerful alternative to physics-based modeling. ANNs operate on a fundamentally different principle. Instead of being programmed with explicit physical laws, they learn complex, non-linear relationships directly from historical data [,]. This approach makes them highly effective for predictive tasks, especially in geologically complex settings, such as karstic or hard-rock aquifers, where the underlying physical processes are difficult to fully parameterize. However, the primary limitation of such data-driven models is their lack of direct physical interpretability, i.e., the black box problem. This is a critical drawback in decision-making scenarios that require not only an accurate prediction but also a clear understanding of its underlying physical basis [].
The most widely accepted research in this field today focuses on combining these two powerful but fundamentally different approaches. Hybrid models create a powerful synergy by leveraging the complementary strengths of numerical and machine learning methods []. The MODFLOW model grounds the system in physical reality, ensuring adherence to hydrological laws, while ANNs provide exceptional computational speed and the ability to identify complex non-linearities in the data. This integration of physical principles with data-driven pattern recognition unlocks a modeling capability that is more rapid, robust, and accurate than either paradigm used alone. While narrative reviews of MODFLOW models or ANNs exist, and some studies offer direct comparisons, a comprehensive systematic review that categorizes the functional classification of the integration strategies that constitute MODFLOW–ANN integration has been missing in the literature [,]. This review addresses this gap by presenting a functional classification framework for MODFLOW–ANN integration and synthesizes the literature within this structure. The primary objectives of this paper are: (1) To systematically identify and synthesize the literature on MODFLOW-ANN integration for groundwater flow modeling published between 2003 and 2005, using a documented and replicable search strategy. (2) To present a classification of integration strategies, such as surrogate modeling, parameter estimation, and error correction. (3) To analyze the applications, reported advantages, and persistent limitations within this classification framework, thereby identifying key trends and future research directions.

2. Systematic Review Methodology

We followed a clear plan to conduct this literature review. Our approach included setting specific research questions, using a well-documented search strategy, applying clear rules for selecting studies, and having a consistent process for gathering and combining data [,].

2.1. Research Questions

We used three specific research questions (RQs) to focus our review on our search, data collection, and analysis. These questions were created to meet the study’s main goals: to map out when and where MODFLOW models and ANNs have been used together, to group the different ways they are combined, and to assess the results of each method.
  • RQ1: What is the temporal and geographical distribution of research integrating MODFLOW models and ANNs for groundwater flow modeling?
  • RQ2: What are the primary strategies for integrating MODFLOW models and ANNs, and what is their relative prevalence in the literature?
  • RQ3: What are the reported advantages, limitations, and performance outcomes for each integration strategy?

2.2. Finding and Identifying Studies

We carried out a thorough and methodical search of published research to find all relevant studies. The search covered a 20-year period from 1 January 2003 to 3 March 2025. We chose this timeframe to track how these combined modeling techniques have grown from their early days to their recent increase in use [].
We searched three major online databases: Scopus, Web of Science, and Google Scholar. We selected these sources because they offer broad and overlapping coverage of research in water science, environmental studies, and computer engineering. This helped us make sure that we found as many relevant publications as possible.
Our search terms were developed carefully through a process of trial and error. This was to get a good balance between finding all relevant studies and filtering out irrelevant ones []. To make our search process clear and repeatable, the exact search terms we used are listed in Table 1. A consistent and well-defined search plan is vital for a good systematic review. It ensures that the initial set of articles is collected in a fair and repeatable way. Using inconsistent dates or search terms would have made our findings less reliable.
Table 1. Database Search Strategy.

2.3. Study Selection Criteria

To ensure the objectivity and consistency of the review, a formal set of inclusion and exclusion criteria was established prior to the screening process. A two-stage screening process was then conducted by two authors independently to minimize bias, with any disagreements resolved through consensus discussion. The first stage involved screening the titles and abstracts of all retrieved records. The second stage involved a full-text review of the articles that passed the initial screen []. The criteria are outlined in Table 2.
Table 2. Inclusion and exclusion criteria.

2.4. Data Extraction and Synthesis

For each study that met the final inclusion criteria, a standardized data extraction form was used to systematically collect relevant information. This process was designed to directly address the research questions and included capturing of the following details: author and year of publication, study location/country, primary scientific objective, the specific integration strategy employed, key findings and conclusions, and any reported performance metrics (e.g., Coefficient of Determination (R2), Root Mean Square Error (RMSE), Nash–Sutcliffe Efficiency (NSE)). The extracted data were then thematically synthesized to identify patterns, trends, and overarching conclusions, which form the analytical basis of the subsequent sections of this review.

3. The Physical-Based Model: MODFLOW

For four decades, the Modular Finite-Difference Flow (MODFLOW) model has been the cornerstone of quantitative groundwater investigation globally. Because it is dependable, flexible, and well-documented, this open source code distributed by the U.S. Geological Survey is now the go-to tool for modeling groundwater flow in both professional and academic settings []. MODFLOW is a widely used groundwater modeling software that numerically solves the three-dimensional, transient groundwater flow equation using a finite-difference approach []. The flow is governed by the Boussinesq Equation, which is a combination of Darcy’s Law and the principle of mass conservation [].
∂ ∂ x K x x ∂ h ∂ x + ∂ ∂ y K y y ∂ h ∂ y + ∂ ∂ z K z z ∂ h ∂ z − W = S s ∂ h ∂ T
where, K x x , K y y , and K z z are the values of hydraulic conductivity on the x, y, and z axes. W is volumetric flux per unit volume representing the source and sinks of water [t−1]. Ss is the specific storage of the permeable material [L−1], h the is potentiometric head [L], and T is time.
The widespread adoption and success of MODFLOW is due to its fundamental modular structure. The software is built around a series of independent modules, known as “Packages”. Each package is designed to simulate a specific hydrological process or feature, such as wells (WEL), recharge (RCH), or rivers (RIV) [,]. This design not only provides the flexibility to perform simulations to specific hydrogeological problems but also facilitates the integration of MODFLOW with other computational frameworks such as ANNs [].
MODFLOW is continuously evolving, with key versions like MODFLOW-2005 and MODFLOW 6 (2017) offering modern, object-oriented frameworks. As summarized in Table 3, this evolution is driven by the growing complexity of hydrogeological challenges and the corresponding demand for more integrated modeling solutions.
Table 3. Advancement in MODFLOW through time.

MODFLOW Applications in Groundwater Modelling

MODFLOW has proven to be a highly versatile tool, successfully used to address scientific problems across a wide spectrum of hydrogeological conditions. Globally, researchers have employed the model to simulate complex phenomena such as surface-subsurface interactions, solute transport, and groundwater flow dynamics. The diverse applications of the MODFLOW model demonstrate its utility. For instance, studies have used the model to assess the performance of multi-aquifer wells [], simulate the unique hydrology of peatlands [], and support critical infrastructure planning [].
Furthermore, the MODFLOW model’s capabilities are frequently expanded by coupling it with other specialized models to create more comprehensive simulations. A key example is its integration with the Soil and Water Assessment Tool (SWAT), which improves low-flow simulations in river basins by linking watershed-scale surface processes with subsurface flow []. In a similar manner, coupling the MODFLOW model with agent-based land-use models allows researchers to evaluate the long-term impacts of different zoning policies on groundwater resources, offering a powerful tool for sustainable planning [,].
The model has been particularly critical in nations like India, which confront acute water stress and significant hydrogeological diversity. MODFLOW has been instrumental in quantifying and managing groundwater depletion, a critical issue within India’s agricultural heartlands. For example, long-term simulations in central Punjab documented significant annual water table declines of 0.15–0.20 m over a 30-year period, providing clear, quantitative evidence of unsustainable groundwater extraction rates []. Two studies showed that the MODFLOW model was successful in simulating recharge-driven water table fluctuations in the Sonar sub-basin. One demonstrated the dominance of recharge over pumping in influencing water table changes, while the other created a two-layer aquifer model using observed and pumping well data to simulate dynamic responses under various stress conditions [,]. In southern India, such as in the Tirunelveli district of Tamil Nadu, GIS-based MODFLOW models have been used to assess over-extraction risks and delineate areas vulnerable to depletion, thereby informing more targeted and sustainable water management strategies []. MODFLOW and its associated transport models, like MT3DMS and SEAWAT, are vital for tackling a wide range of water quality. Along extensive coastlines, these models are critical for simulating and preventing seawater intrusion. Studies in the coastal aquifers of Balasore (Odisha) and Gujarat in India have used MODFLOW to evaluate different pumping scenarios and recommend sustainable extraction caps to mitigate salinity risk []. In industrial and agricultural hubs, such as Tamil Nadu’s Amaravathi River basin, MODFLOW simulations have been used to track the transport of contaminants like total dissolved solids (TDS) from industrial effluents, providing a scientific basis for pollution control measures [].
Recognizing that groundwater and surface water are interconnected, many studies have focused on integrated modeling. Research in the Mahesh River basin demonstrated the enhanced predictive capacity of coupled surface–groundwater models, which more accurately captured the aquifer’s response to hydrological stresses []. Similarly, work in the Mahanadi delta has successfully captured the dynamic river–aquifer interactions that govern the region’s hydrology [].
To provide a consolidated overview, Table 4 summarizes key studies utilizing MODFLOW worldwide, with an emphasis on Indian context, highlighting the diverse challenges addressed by this versatile model.
Table 4. Summary of the literature on groundwater modelling using MODFLOW.

4. Artificial Neural Networks (ANNs)

The field of machine learning (ML) has guided in a new era for groundwater flow modeling, significantly enhancing computational efficiency, predictive accuracy, and robustness []. Among the various ML techniques, Artificial Neural Networks (ANNs) have gained prominence for their ability to handle the complex, nonlinear relationships inherent in hydrological systems. ANNs excel at processing large datasets, identifying hidden patterns, and adapting to the non-stationary characteristics of groundwater dynamics, making them a powerful complement to traditional physics-based models [,].
An ANN is a computational system inspired by the structure and function of biological neural networks. It is composed of interconnected nodes, or “neurons,” arranged in layers: an input layer, one or more hidden layers, and an output layer []. The network learns by adjusting the numerical weights of the connections between neurons during a training process, effectively teaching itself to map a set of inputs (e.g., rainfall, pumping rates) to a desired output (e.g., groundwater level) without being explicitly programmed with governing physical equations [].
While conventional ANNs provide highly accurate approximations, a persistent criticism has been their potential lack of physical consistency i.e., the black box model []. To address this, a new class of models known as physics-informed neural networks (PINNs) have emerged. PINNs represent a significant advancement by incorporating the governing physical equations, such as the groundwater flow equation, directly into the ANN’s training process. This is achieved by adding a loss function that measures how much the network’s output conflicts with known physical laws. The model is then trained to minimize this conflict, resulting in a solution that agrees with both the training data and the fundamental principles of physics [,,].
Overall, these studies highlight the adaptability of ANNs in a wide range of hydrological contexts. ANNs have consistently demonstrated strong performance in predicting groundwater levels, quality, and flow. This effectiveness is maintained across various applications, whether they are used as standalone models, in hybrid systems, or integrated with GIS and optimization algorithms.
As illustrated in Figure 1, the annual publication trend on this topic shows a significant increase in research activity after 2020. This recent surge is likely driven by a convergence of technological advancements and pressing environmental challenges. The increased accessibility of high-performance computing and powerful open-source machine learning libraries has provided the necessary tools, while the urgent need to understand climate change’s impact on groundwater provides critical motivation.
Figure 1. Annual publication trend for research on MODFLOW–ANN integration (2003–2025), illustrating a consistent increase in output followed by a sharp rise after 2020.
  • We identified about 250 key terms by looking at the most frequently occurring words (at least ten times) in the titles and abstracts of research papers pertaining to groundwater. These terms were divided into four primary groups, each of which stood for a significant field of study in Figure 2.
    Figure 2. Word cloud representing frequently occurring terms in groundwater research publications.
  • Many terms were included in the first cluster (dark purple), which focused on subjects pertaining to ML applications, specifically the use of artificial neural networks for environmental prediction. Artificial, neural, network, model, machine, support vector, learning, prediction, ANN, and forecasting were among the most prominent terms in this cluster.
  • The second cluster (green–blue) concentrated on hydrological processes and the evaluation of groundwater quantity and quality. Groundwater, aquifer, level, fluctuation, depth, recharge, soil, quality, transport, and contamination were among the terms it included.
  • Terms related to time-series analysis and forecasting, frequently in combination with neural networks, made up the third cluster (yellow). Forecasting, wavelet, trends, time series, uncertainty, performance, and optimization were among the terms that appeared frequently.

Application of ANNs in Groundwater Flow Simulation

Standalone ANN models have been successfully applied to a wide range of groundwater problems across diverse hydrogeological and geographic contexts globally. They have been used to forecast the spatial distribution of nitrate contamination from agricultural sources [], predict groundwater level fluctuations in advance [], and forecast tidally influenced groundwater dynamics along coastal aquifers []. In geologically complex systems, such as karstic or fractured rock aquifers, where the assumptions of traditional numerical models may not hold, ANNs have proven to be particularly effective []. Hybrid models that combine ANNs with other data-driven techniques, such as wavelet decomposition for time-series pre-processing, have demonstrated even greater accuracy [,]. These studies collectively highlight the remarkable adaptability and predictive power of ANNs in a wide range of hydrological settings, consistently demonstrating strong performance in forecasting groundwater levels, quality, and flow dynamics.
In India, ANNs have become a widely adopted tool for modeling groundwater dynamics and forecasting future scenarios, particularly in regions with limited hydrogeological data. Early applications focused on solving complex inverse problems, such as identifying unknown sources of pollution by learning the relationship between observed contaminant concentrations and their potential fluxes []. More recent studies have demonstrated increasing sophistication. For instance, researchers have coupled ANNs with Genetic Algorithms to simulate flow and contaminant transport under conditions of uncertainty [], and have applied feedforward ANNs in the complex hard rock terrains of southern India to inform sustainable management strategies []. A significant trend has been the coupling of ANNs with MODFLOW models to optimize agricultural water use, a critical issue for India’s food security. In these applications, ANNs are used to develop optimal cropping patterns that maximize yield while minimizing groundwater depletion, with MODFLOW providing the underlying simulation of the aquifer’s response. These studies collectively highlight the remarkable adaptability and predictive power of ANNs in a wide range of hydrological settings. They consistently demonstrate strong performance in forecasting groundwater levels, quality, and flow dynamics.
Table 5 summarizes the use of ANNs in groundwater studies, illustrating the increasing importance and versatility of these models.
Table 5. Summary of the literature on groundwater modelling using ANNs.

5. A Classification Framework of MODFLOW–ANN Integration

The integration of MODFLOW models and ANNs is not a single approach but a collection of distinct strategies, each designed to leverage the complementary strengths of both models. Our systematic review of the literature revealed three primary integration strategies that form a classification framework for understanding the hybrid groundwater modeling, as shown in Figure 3.
Figure 3. Schematic representation of the hybrid MODFLOW–ANN framework for groundwater modeling.
This framework depicts the integration of MODFLOW models with ANNs through three interconnected pathways: parameter estimation, surrogate modeling, and error correction. In Pathway 1, field observations such as hydraulic head and geological data are used to train an ANN model for estimating hydrogeological parameters, which are then incorporated into MODFLOW. In Pathway 2, MODFLOW outputs are used to train ANN-based surrogate models that replicate their behavior for faster simulations; and in Pathway 3, residuals between MODFLOW predictions and observed data are used to train a secondary ANN that corrects the model outputs. To substantiate the utility of this functional framework, Table 6 provides a comparative analysis against other classification schemes found in the literature. This comparison highlights how this functional framework, synthesized from the literature, offers a practical perspective for hydrogeologists that is distinct from frameworks based on algorithm type or modeling scale.
Table 6. Comparative analysis of hydrogeological and machine learning classification frameworks.
Together, these pathways form a hybrid modeling approach that strengthens the predictive reliability and computational efficiency of groundwater flow simulations. Table 7 provides a concise overview of this analytical framework.
Table 7. A classification framework of MODFLOW–ANN integration strategies.

5.1. Strategy 1: An ANN as a Surrogate Model

In this strategy, a computationally expensive MODFLOW model is run numerous times to generate a large dataset of inputs and corresponding outputs. An ANN is then trained on this dataset to act as a fast and accurate emulator, or “surrogate”, of the physical model []. The primary motivation is computational efficiency, which is crucial for applications requiring thousands of model runs, such as uncertainty analysis via Monte Carlo simulations or simulation–optimization for resource management []. A recent example is the ‘Dual Path CNN-MLP’ model developed for the Qatar aquifer, which served as a surrogate for MODFLOW and achieved a 99% boost in computational efficiency, making large-scale uncertainty analysis practical [].

5.2. Strategy 2: ANNs for Parameter Estimation (Inverse Modeling)

This strategy uses an ANN to solve the challenging inverse problem of estimating key MODFLOW input parameters, such as the spatial distribution of hydraulic conductivity, from observed data []. This data-driven approach can replace or supplement traditional, time-consuming calibration methods []. The ANN offers a powerful method to map the complex, non-linear relationships between aquifer stresses, responses, and properties without a priori assumptions. A notable case study by [] used an ANN to derive regression equations for predicting potentiometric heads based on geological and mining parameters, complementing a detailed MODFLOW sensitivity analysis of a mine complex.

5.3. Strategy 3: An ANN as an Error-Correction Model

This advanced hybrid approach uses a calibrated MODFLOW model to produce a primary, physics-based prediction. The systematic errors (residuals), calculated as the difference between the MODFLOW predictions and observed field data, are then modeled using an ANN []. The final, improved prediction is the sum of the MODFLOW output and the ANN-predicted error. This strategy leverages the strengths of both paradigms: the physics-based model provides the primary prediction, while the data-driven model captures complex, unmodeled dynamics and local biases present in the residuals []. A case study from the Bist-Doab region in Punjab, India, successfully applied this residual modeling approach to integrate the MODFLOW model with machine learning models, significantly reducing model error and improving predictions in an agriculturally stressed region [].
Table 8 summarizes the key literature reviewed for these integrated hybrid approaches, providing an overview of the methodologies and findings in this emerging area.
Table 8. Summary of the literature on hybrid MODFLOW–ANN models.

6. Synthesis and Discussion

The integration of the physically based MODFLOW model and data-driven ANNs signals a new paradigm in how groundwater systems are simulated and managed. This systematic review has illuminated the powerful synergies that arise from their strategic combination. Hybrid approaches integrate the physical principles of numerical models with the computational efficiency of machine learning, thereby enabling more accurate, robust, and efficient hydrogeological analysis. The cumulative literature evidence from global and Indian case studies indicates that hybrid models consistently outperform their standalone counterparts. This superior performance obtained from a powerful symbiotic relationship: MODFLOW provides the physical framework, constraining the model to physically realistic solutions, while ANNs address key limitations by accelerating computations, enhancing parameterization, and correcting systematic biases.
This review has categorized the primary integration methods into three core strategies: surrogate modeling, intelligent parameterization, and error correction. The choice of strategy depends critically on the scientific question, data availability, and computational resources. This strategic diversity underscores the maturity of the field and highlights the need for modeling teams to possess expertise in both hydrogeology and data science.

7. Conclusions and Future Research Directions

This systematic review confirms that the integration of MODFLOW models and ANNs provides a more comprehensive and accurate depiction of aquifer systems than either method alone. By systematically categorizing the literature into a unique classification of integration strategies, we have provided a structured framework for understanding and advancing the field.
Although the progress in hybrid modeling has been considerable, it represents the early stages of a rapidly advancing field with vast potential. To realize the full potential of these integrated models, future research should focus on several key areas:
  • Explainable AI (XAI) and Physics-Informed Neural Networks (PINNs): The most significant barrier to broader adoption remains the black box problem. Future research must prioritize the development of techniques like PINNs, which bake physical laws directly into the network architecture, to make these models more transparent and trustworthy.
  • Scalability and High-Performance Computing (HPC): A major challenge is scaling these hybrid approaches to large regional or national scales, which will require leveraging HPC resources to handle the massive datasets and complex training procedures involved.
  • Real-Time Data Assimilation: Future work should focus on developing methods to dynamically assimilate real-time data streams from sources like remote-sensing satellites and IoT-enabled sensors into these hybrid models, enabling a shift from static to dynamic, adaptive forecasting systems.
By pursuing these research directions, the scientific community can continue to build upon the strong foundation of MODFLOW–ANN integration, creating the next generation of groundwater models that are not only more powerful but also more transparent, reliable, and integral to securing our global water future.

Author Contributions

Conceptualization, K.K. and A.A.; methodology, K.K. and A.A.; investigation, A.A. and P.K.S.; writing—original draft preparation, K.K.; writing—review and editing, A.A., P.K.S., Y.K.S., J.L., and F.G.; supervision, A.A.; project administration, A.A., J.L., and F.G.; and funding acquisition, A.A., J.L., and F.G. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean Government (MSIT) (No. NRF-RS-2023-00259995).

Data Availability Statement

This is a review article and does not contain any new data. All data discussed in this manuscript are sourced from the previously published literature, which has been cited appropriately in the text and listed in the reference section.

Acknowledgments

We extend our sincere gratitude to the University of Petroleum and Energy Studies in Dehradun, India, for generously providing financial and research resources.

Conflicts of Interest

One of the authors, Fatemeh Ghobadi, is currently serving as an Assistant Guest Editor for this journal. In accordance with the editorial policy of Water, to avoid any potential conflicts of interest, all decisions regarding manuscripts co-authored by Guest Editors are made independently by the journal’s editorial board. Guest Editors have no involvement or influence in the handling, peer review, or editorial decision-making process of manuscripts to which they contribute. Accordingly, Fatemeh Ghobadi was not involved in the review or decision process for this manuscript. The authors declare no other conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
AbbreviationFull Form
ANNArtificial Neural Network
ANFISAdaptive Neuro-Fuzzy Inference System
BCFBlock-Centered Flow package
DISVDiscretization by Vertices
DRNDrain package
FNNFeedforward Neural Network
GHBGeneral Head Boundary package
GISGeographic Information System
GMSGroundwater Modeling System
HPCHigh-Performance Computing
LSTMLong Short-Term Memory
MAW1Multi-Aquifer Well package (MODFLOW 6)
MLMachine Learning
MLPMultilayer Perceptron
MODFLOWModular Finite Difference Flow model
MODFLOW-NWTMODFLOW Newton Solver
MODFLOW-USGMODFLOW Unstructured Grid
MT3DMSModular Three-Dimensional Transport Model
NSENash-Sutcliffe Efficiency
PESTParameter ESTimation
PINNPhysics-Informed Neural Network
PSOParticle Swarm Optimization
R2Coefficient of Determination
RCHRecharge package
ReLURectified Linear Unit
RIVRiver package
RMSERoot Mean Square Error
SEAWATSEA Water Transport model
SWATSoil and Water Assessment Tool
SVMSupport Vector Machine
WULUMODWater-Use and Land-Use Change Model

References

  1. Angelakis, A.N.; Voudouris, K.S.; Mariolakos, I. Groundwater utilization through the centuries focusing οn the Hellenic civilizations. Hydrogeol. J. 2016, 24, 1311–1324. [Google Scholar] [CrossRef]
  2. Bobeck, P. Henry Darcy and the Public Fountains of the City of Dijon. In Henry P. G. Darcy and Other Pioneers in Hydraulics; American Society of Civil Engineers: Philadelphia, PA, USA, 2003; pp. 37–50. [Google Scholar] [CrossRef]
  3. Brown, G.O. Jules Dupuit’s Contributions in Water Resources. In Water Resources and Environmental History; American Society of Civil Engineers: Salt Lake City, UT, USA, 2004; pp. 104–110. [Google Scholar] [CrossRef]
  4. Okuyade, W.I.A.; Abbey, T.M.; Abbey, M.E. Application of the Dupuit–Forchheimer model to groundwater flow into a well. Model. Earth Syst. Environ. 2022, 8, 2359–2367. [Google Scholar] [CrossRef]
  5. Theis, C.V. The relation between the lowering of the Piezometric surface and the rate and duration of discharge of a well using ground-water storage. Eos Trans. Am. Geophys. Union 1935, 16, 519–524. [Google Scholar] [CrossRef]
  6. Igboekwe, M.U.; Achi, N.J. Finite Difference Method of Modelling Groundwater Flow. J. Water Resour. Prot. 2011, 3, 192–198. [Google Scholar] [CrossRef]
  7. Sarkar, B.C. Geostatistics in Groundwater Modelling. In Groundwater Development and Management; Sikdar, P.K., Ed.; Springer International Publishing: Cham, Switzerland, 2019; pp. 147–169. [Google Scholar] [CrossRef]
  8. Ma, Q.; Abily, M.; Du, M.; Gourbesville, P.; Fouché, O. Integrated Groundwater Resources Management: Spatially-Nested Modelling Approach for Water Cycle Simulation. Water Resour. Manag. 2020, 34, 1319–1333. [Google Scholar] [CrossRef]
  9. McDonald, M.G. A Modular Three-Dimensional Finite-Difference Ground-Water Flow Model; US Geological Survey: Reston, VA, USA, 1988. [Google Scholar] [CrossRef]
  10. Pathak, R.; Awasthi, M.K.; Sharma, S.K.; Hardaha, M.K.; Nema, R.K. Ground Water Flow Modelling Using MODFLOW—A Review. Int. J. Curr. Microbiol. Appl. Sci. 2018, 7, 83–88. [Google Scholar] [CrossRef]
  11. Harbaugh, A.W. MODFLOW-2005, U.S. Geological Survey Modular Ground-Water Model—The Groundwater Flow Process. U.S. Geological Survey Techniques and Methods 6-A16. 2005. Available online: https://pubs.usgs.gov/tm/2005/tm6A16/PDF/TM6A16.pdf (accessed on 5 August 2025).
  12. Doherty, J.E.; Hunt, R.J. Approaches to Highly Parameterized Inversion—A Guide to Using PEST for Groundwater-Model Calibration; U.S. Geological Survey Scientific Investigations Report 2010–5169; U.S. Geological Survey: Reston, VA, USA, 2010; 59p. [Google Scholar]
  13. Zeydalinejad, N. Artificial neural networks vis-à-vis MODFLOW in the simulation of groundwater: A review. Model. Earth Syst. Environ. 2022, 8, 2911–2932. [Google Scholar] [CrossRef]
  14. Kohzadi, N.; Boyd, M.S.; Kaastra, I.; Kermanshahi, B.S.; Scuse, D. Neural Networks for Forecasting: An Introduction. Can. J. Agric. Econ. Can. Agroecon. 1995, 43, 463–474. [Google Scholar] [CrossRef]
  15. Sharma, Y.K.; Kim, S.; Tayerani Charmchi, A.S.; Kang, D.; Batelaan, O. Strategic Imputation of Groundwater Data Using Machine Learning: Insights from Diverse Aquifers in the Chao-Phraya River Basin. Groundw. Sustain. Dev. 2024, 18, 101394. [Google Scholar] [CrossRef]
  16. Sun, A.Y. Predicting groundwater level changes using GRACE data: Predicting Groundwater Level Changes Using Grace Data. Water Resour. Res. 2013, 49, 5900–5912. [Google Scholar] [CrossRef]
  17. Chen, C.; He, W.; Zhou, H.; Xue, Y.; Zhu, M. A comparative study among machine learning and numerical models for simulating groundwater dynamics in the Heihe River Basin, northwestern China. Sci. Rep. 2020, 10, 3904. [Google Scholar] [CrossRef]
  18. Nikolos, I.K.; Stergiadi, M.; Papadopoulou, M.P.; Karatzas, G.P. Artificial neural networks as an alternative approach to groundwater numerical modelling and environmental design. Hydrol. Process. 2008, 22, 3337–3348. [Google Scholar] [CrossRef]
  19. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021, 372, n71. [Google Scholar] [CrossRef] [PubMed]
  20. Zizaan, A.; Idri, A. Machine learning based Breast Cancer screening: Trends, challenges, and opportunities. Comput. Methods Biomech. Biomed. Eng. Imaging Vis. 2023, 11, 976–996. [Google Scholar] [CrossRef]
  21. Silva, R.L.S.; Neiva, F.W. Systematic Literature Review in Computer Science—A Practical Guide; Federal University of Juiz de Fora: Juiz de Fora, Brazil, 2016. [Google Scholar] [CrossRef]
  22. Carrera-Rivera, A.; Ochoa, W.; Larrinaga, F.; Lasa, G. How-to conduct a systematic literature review: A quick guide for computer science research. MethodsX 2022, 9, 101895. [Google Scholar] [CrossRef] [PubMed]
  23. Xu, X.; Huang, G.; Zhan, H.; Qu, Z.; Huang, Q. Integration of SWAP and MODFLOW-2000 for modeling groundwater dynamics in shallow water table areas. J. Hydrol. 2012, 412–413, 170–181. [Google Scholar] [CrossRef]
  24. Zaresefat, M.; Derakhshani, R. Revolutionizing Groundwater Management with Hybrid AI Models: A Practical Review. Water 2023, 15, 1750. [Google Scholar] [CrossRef]
  25. Langevin, C.D.; Hughes, J.D.; Banta, E.R.; Provost, A.; Niswonger, R.; Panday, S. MODFLOW 6, The U.S. Geological Survey Modular Hydrologic Model, U.S. Geological Survey: Reston, VA, USA, 2017. [CrossRef]
  26. Grodzka-Łukaszewska, M.; Sinicyn, G.; Grygoruk, M.; Mirosław-Świątek, D.; Kardel, I.; Okruszko, T. The role of the river in the functioning of marginal fen: A case study from the Biebrza Wetlands. PeerJ 2022, 10, e13418. [Google Scholar] [CrossRef]
  27. Langevin, C.D.; Hughes, J.D.; Provost, A.M.; Russcher, M.J.; Morway, E.D.; Reno, M.J.; Bonelli, W.P.; Panday; Sorab; Merrick; et al. MODFLOW Version 6.5.0, U.S. Geological Survey: Reston, VA, USA, 2024. [CrossRef]
  28. Panday, S.; Langevin, C.D.; Niswonger, R.G.; Ibaraki, M.; Hughes, J.D. MODFLOW–USG Version 1: An Unstructured Grid Version of MODFLOW for Simulating Groundwater Flow and Tightly Coupled Processes Using a Control Volume Finite-Difference Formulation; U.S. Geological Survey Techniques and Methods; U.S. Geological Survey: Reston, VA, USA, 2013. [Google Scholar]
  29. Arnold, J.G.; Moriasi, D.N.; Gassman, P.W.; Abbaspour, K.C.; White, M.J.; Srinivasan, R.; Santhi, C.; Harmel, R.D.; Van Griensven, A.; Van Liew, M.W.; et al. SWAT: Model Use, Calibration, and Validation. Trans. ASABE 2012, 55, 1491–1508. [Google Scholar] [CrossRef]
  30. Sharma, Y.K.; Mohanasundaram, S.; Kim, S.; Shrestha, S.; Babel, M.S.; Loc, H.H. Enhancing Aquifer Reliability and Resilience Assessment in Data-Scarce Regions Using Satellite Data: Application to the Chao Phraya River Basin. Remote Sens. 2025, 17, 1731. [Google Scholar] [CrossRef]
  31. Neville, C.J. Modeling Multiaquifer Wells with MODFLOW. Groundwater 2004, 42, 910–919. [Google Scholar] [CrossRef]
  32. Gao, H. Groundwater Modeling for Flow Systems with Complex Geological and Hydrogeological Conditions. Procedia Earth Planet. Sci. 2011, 3, 23–28. [Google Scholar] [CrossRef]
  33. Brunner, P.; Simmons, C.T.; Cook, P.G.; Therrien, R. Modeling Surface Water-Groundwater Interaction with MODFLOW: Some Considerations. Groundwater 2010, 48, 174–180. [Google Scholar] [CrossRef]
  34. Wróbel, M.; Brandyk, A.; Tereba, A. Modflow model in the assessment of water conditions in forest areas. Int. J. Hydrol. Sci. Technol. 2023, 15, 207. [Google Scholar] [CrossRef]
  35. Sharma, Y.; Aggarwal, A.; Singh, J. Development of Flow Duration Curves and Eco-Flow Metrics for the Tawi River Basin—(Jammu, India). Int. J. Adv. Remote Sens. GIS 2019, 8, 3114–3125. [Google Scholar] [CrossRef]
  36. Rejani, R.; Jha, M.K.; Panda, S.N.; Mull, R. Simulation Modeling for Efficient Groundwater Management in Balasore Coastal Basin, India. Water Resour. Manag. 2008, 22, 23–50. [Google Scholar] [CrossRef]
  37. Khadri, S.F.R.; Pande, C. Ground water flow modeling for calibrating steady state using MODFLOW software: A case study of Mahesh River basin, India. Model. Earth Syst. Environ. 2016, 2, 39. [Google Scholar] [CrossRef]
  38. Varalakshmi, V.; Venkateswara Rao, B.; SuriNaidu, L.; Tejaswini, M. Groundwater Flow Modeling of a Hard Rock Aquifer: Case Study. J. Hydrol. Eng. 2014, 19, 877–886. [Google Scholar] [CrossRef]
  39. Kirubakaran, M.; Colins Johnny, J.; Samson, S. MODFLOW Based Groundwater Budgeting Using GIS: A Case Study from Tirunelveli Taluk, Tirunelveli District, Tamil Nadu, India. J. Indian Soc. Remote Sens. 2018, 46, 783–792. [Google Scholar] [CrossRef]
  40. Momejian, N.; Abou Najm, M.; Alameddine, I.; El-Fadel, M. Groundwater Vulnerability Modeling to Assess Seawater Intrusion: A Methodological Comparison with Geospatial Interpolation. Water Resour. Manag. 2019, 33, 1039–1052. [Google Scholar] [CrossRef]
  41. Rajamanickam, R.; Nagan, S. Groundwater Quality Modeling of Amaravathi River Basin of Karur District, Tamil Nadu, Using Visual Modflow. Int. J. Environ. Sci. 2010, 1, 91–108. [Google Scholar]
  42. Siva Prasad, Y.; Venkateswara Rao, B.; Surinaidu, L. Groundwater flow modeling and prognostics of Kandivalasa river sub-basin, Andhra Pradesh, India. Environ. Dev. Sustain. 2021, 23, 1823–1843. [Google Scholar] [CrossRef]
  43. Thomas, B.; Behrangi, A.; Famiglietti, J. Precipitation Intensity Effects on Groundwater Recharge in the Southwestern United States. Water 2016, 8, 90. [Google Scholar] [CrossRef]
  44. Martinez, M.B.; Widdowson, M.A. Evaluating Flow Distribution in a Multiaquifer Recharge Well Using an In Situ Flowmeter. Groundwater 2024, 62, 605–616. [Google Scholar] [CrossRef] [PubMed]
  45. Langevin, C.D.; Guo, W. MODFLOW/MT3DMS–Based Simulation of Variable-Density Ground Water Flow and Transport. Groundwater 2006, 44, 339–351. [Google Scholar] [CrossRef]
  46. Kim, N.W.; Chung, I.M.; Won, Y.S.; Arnold, J.G. Development and application of the integrated SWAT–MODFLOW model. J. Hydrol. 2008, 356, 1–16. [Google Scholar] [CrossRef]
  47. Merchán-Rivera, P.; Wohlmuth, B.; Chiogna, G. Identifying stagnation zones and reverse flow caused by river-aquifer interaction: An approach based on polynomial chaos expansions. Water Resour. Res. 2021, 57, e2021WR029824. [Google Scholar] [CrossRef]
  48. Reeves, H.W.; Zellner, M.L. Linking MODFLOW with an Agent-Based Land-Use Model to Support Decision Making. Groundwater 2010, 48, 649–660. [Google Scholar] [CrossRef]
  49. Batelaan, O.; De Smedt, F. SEEPAGE, a New MODFLOW DRAIN Package. Groundwater 2004, 42, 576–588. [Google Scholar] [CrossRef]
  50. Bedekar, V.; Niswonger, R.G.; Kipp, K.; Panday, S.; Tonkin, M. Approaches to the Simulation of Unconfined Flow and Perched Groundwater Flow in MODFLOW. Groundwater 2012, 50, 187–198. [Google Scholar] [CrossRef]
  51. Rahnama, M.B.; Zamzam, A. Quantitative and qualitative simulation of groundwater by mathematical models in Rafsanjan aquifer using MODFLOW and MT3DMS. Arab. J. Geosci. 2013, 6, 901–912. [Google Scholar] [CrossRef]
  52. Wang, Q.; Zhan, H.; Tang, Z. A New Package in MODFLOW to Simulate Unconfined Groundwater Flow in Sloping Aquifers. Groundwater 2014, 52, 924–935. [Google Scholar] [CrossRef] [PubMed]
  53. Antoniou, M.; Theodossiou, N.; Karakatsanis, D. Coupling groundwater simulation and optimization models using MODFLOW and Harmony Search Algorithm. Desalination Water Treat. 2017, 86, 297–304. [Google Scholar] [CrossRef]
  54. AL-Hashmi, S. A numerical groundwater flow model of Wadi Samail Catchment using MODFLOW software. Int. J. GEOMATE 2020, 18, 30–36. [Google Scholar] [CrossRef]
  55. Bailey, R.T.; Park, S.; Bieger, K.; Arnold, J.G.; Allen, P.M. Enhancing SWAT+ simulation of groundwater flow and groundwater-surface water interactions using MODFLOW routines. Environ. Model. Softw. 2020, 126, 104660. [Google Scholar] [CrossRef]
  56. Boczoń, A.; Wróbel, M.; Kowalska, A. Long-Term Changes in Groundwater Levels in the Białowieża Forest, Poland, Under Climate Change. Water 2025, 17, 2027. [Google Scholar] [CrossRef]
  57. Senthilkumar, M.; Elango, L. Three-dimensional mathematical model to simulate groundwater flow in the lower Palar River basin, southern India. Hydrogeol. J. 2004, 12, 197–208. [Google Scholar] [CrossRef]
  58. Chatterjee, C.; Kumar, R.; Chakravorty, B.; Lohani, A.K.; Kumar, S. Integrating Remote Sensing and GIS Techniques with Groundwater Flow Modeling for Assessment of Waterlogged Areas. Water Resour. Manag. 2005, 19, 539–554. [Google Scholar] [CrossRef]
  59. Majumdar, P.K.; Kumar, S.; Singh, V.; Jose, M.K. Characterization of groundwater flow in the depleting water table areas in central Punjab. In Proceedings of the XII World Water Congress of IWRA, New Delhi, India, 22–25 November 2005. [Google Scholar]
  60. Kant, S.; Singh, S.; Nema, A.K.; Chandola, V.K. Modeling of Groundwater Levels in the Sonar sub-basin, Madhya Pradesh Using Visual MODFLOW. Indian J. Dryland Agric. Res. Dev. 2013, 28, 33–37. [Google Scholar]
  61. Surinaidu, L. Quantifying Stream Flows and Groundwater Response under the Climate and Land Use Change through Integrated Hydrological Modelling in a South Indian River Basin. Water Security 2022, 17, 100129. [Google Scholar] [CrossRef]
  62. Surinaidu, L. Application of MODFLOW for groundwater Seepage Problems in the Subsurface Tunnels. J. Indian Geophys. Union 2015, 19, 422–443. [Google Scholar]
  63. Sajeena, S.; Kurien, E.K. Studies on groundwater resources using visual MODFLOW—A case study of Kadalundi river basin, Malappuram, Kerala. Indian J. Soil Conserv. 2019, 47, 21–29. [Google Scholar]
  64. Behera, A.K.; Pradhan, R.M.; Kumar, S.; Chakrapani, G.J.; Kumar, P. Assessment of Groundwater Flow Dynamics Using MODFLOW in Shallow Aquifer System of Mahanadi Delta (East Coast), India. Water 2022, 14, 611. [Google Scholar] [CrossRef]
  65. Timaniya, A.; Soni, N. Modeling of Saline Water Intrusion using MODFLOW in Una Coastal Aquifer of Gujarat, India. Int. J. Res. Appl. Sci. Eng. Technol. 2022, 10, 634–640. [Google Scholar] [CrossRef]
  66. Siddiqi, F.U.R.; Ahmad, S.; Akram, T.; Ali, M.U.; Zafar, A.; Lee, S.W. Artificial Neural Network-Based Data-Driven Parameter Estimation Approach: Applications in PMDC Motors. Mathematics 2024, 12, 3407. [Google Scholar] [CrossRef]
  67. Malakar, P.; Sarkar, S.; Mukherjee, A.; Bhanja, S.; Sun, A.Y. Use of machine learning and deep learning methods in groundwater. In Global Groundwater; Elsevier: Amsterdam, The Netherlands, 2021; pp. 545–557. [Google Scholar] [CrossRef]
  68. Sharma, Y.; Srivastava, M.; Sharma, P.; Kumar, D. Deterministic Seismic Hazard Assessment by Revisiting 1991 Uttarkashi and 1999 Chamoli Earthquake for Uttarakhand, India. Available online: https://meetingorganizer.copernicus.org/EGU23/EGU23-11563.html (accessed on 19 March 2025).
  69. Tao, H.; Hameed, M.M.; Marhoon, H.A.; Zounemat-Kermani, M.; Heddam, S.; Kim, S.; Sulaiman, S.O.; Tan, M.L.; Sa’adi, Z.; Mehr, A.D.; et al. Groundwater level prediction using machine learning models: A comprehensive review. Neurocomputing 2022, 489, 271–308. [Google Scholar] [CrossRef]
  70. D’Oria, M.; Fienen, M.N. MODFLOW-Style Parameters in Underdetermined Parameter Estimation. Groundwater 2012, 50, 149–153. [Google Scholar] [CrossRef]
  71. Fahimi, F.; Yaseen, Z.M.; El-shafie, A. Application of soft computing based hybrid models in hydrological variables modeling: A comprehensive review. Theor. Appl. Climatol. 2017, 128, 875–903. [Google Scholar] [CrossRef]
  72. Chu, H.; Bian, J.; Lang, Q.; Sun, X.; Wang, Z. Daily Groundwater Level Prediction and Uncertainty Using LSTM Coupled with PMI and Bootstrap Incorporating Teleconnection Patterns Information. Sustainability 2022, 14, 11598. [Google Scholar] [CrossRef]
  73. Solgi, R.; Loáiciga, H.A.; Kram, M. Long short-term memory neural network (LSTM-NN) for aquifer level time series forecasting using in-situ piezometric observations. J. Hydrol. 2021, 601, 126800. [Google Scholar] [CrossRef]
  74. Liang, Z.; Liu, Y.; Hu, H.; Li, H.; Ma, Y.; Khan, M.Y.A. Combined Wavelet Transform With Long Short-Term Memory Neural Network for Water Table Depth Prediction in Baoding City, North China Plain. Front. Environ. Sci. 2021, 9, 780434. [Google Scholar] [CrossRef]
  75. Loudyi, D.; Falconer, R.; Lin, B. MODFLOW: An Insight into Thirty Years Development of a Standard Numerical Code for Groundwater Simulations; CUNY Academic Works; CUNY: New York, NY, USA, 2014. [Google Scholar]
  76. Noureldeen, A.S.; Ghaly, S.; Ali, K.; Abozaid, G. ANN-Based Approach to Predict Changes in Ground Water Levels as a Result of Constructing New Naga-Hammadi Barrage, Egypt. In Recent Advances in Environmental Science from the Euro-Mediterranean and Surrounding Regions; Kallel, A., Ksibi, M., Ben Dhia, H., Khélifi, N., Eds.; Advances in Science, Technology & Innovation; Springer International Publishing: Cham, Switzerland, 2018; pp. 853–856. ISBN 978-3-319-70547-7. [Google Scholar]
  77. Motawej, H. Integrating MODFLOW and LSTM models for enhanced groundwater management in the coastal plains of Lattakia Governorate. Water Pract. Technol. 2025, 20, 413–423. [Google Scholar] [CrossRef]
  78. Trichakis, I.C.; Nikolos, I.K.; Karatzas, G.P. Artificial Neural Network (ANN) Based Modeling for Karstic Groundwater Level Simulation. Water Resour. Manag. 2011, 25, 1143–1152. [Google Scholar] [CrossRef]
  79. Chidepudi, S.K.R.; Massei, N.; Jardani, A.; Henriot, A.; Allier, D.; Baulon, L. A wavelet-assisted deep learning approach for simulating groundwater levels affected by low-frequency variability. Sci. Total Environ. 2023, 865, 161035. [Google Scholar] [CrossRef]
  80. Sharma, Y.; Tyagi, A.; Sharma, M.L.; Sharma, P.; Aggarwal, A. Building Vulnerability Assessment Using Artificial Intelligence for Landslide Susceptibility Zone in Champawat District, India. Available online: https://meetingorganizer.copernicus.org/EGU23/EGU23-1957.html (accessed on 30 March 2025).
  81. Chowdhury, T.N.; Battamo, A.; Nag, R.; Zekker, I.; Salauddin, M. Impacts of climate change on groundwater quality: A systematic literature review of analytical models and machine learning techniques. Environ. Res. Lett. 2025, 20, 033003. [Google Scholar] [CrossRef]
  82. Khosravi, K.; Farooque, A.A.; Naghibi, A.; Heddam, S.; Sharafati, A.; Hatamiafkoueieh, J.; Abolfathi, S. Enhancing Pan evaporation predictions: Accuracy and uncertainty in hybrid machine learning models. Ecol. Inform. 2025, 85, 102933. [Google Scholar] [CrossRef]
  83. Hemmat Esfe, M.; Toghraie, D. An optimal feed-forward artificial neural network model and a new empirical correlation for prediction of the relative viscosity of Al2O3-engine oil nanofluid. Sci. Rep. 2021, 11, 17072. [Google Scholar] [CrossRef]
  84. Li, J.; Yoder, R.E.; Odhiambo, L.O.; Zhang, J. Simulation of nitrate distribution under drip irrigation using artificial neural networks. Irrig. Sci. 2004, 23, 29–37. [Google Scholar] [CrossRef]
  85. Daliakopoulos, I.N.; Coulibaly, P.; Tsanis, I.K. Groundwater level forecasting using artificial neural networks. J. Hydrol. 2005, 309, 229–240. [Google Scholar] [CrossRef]
  86. Giustolisi, O.; Simeone, V. Optimal design of artificial neural networks by a multi-objective strategy: Groundwater level predictions. Hydrol. Sci. J. 2006, 51, 502–523. [Google Scholar] [CrossRef]
  87. Joorabchi, A.; Zhang, H.; Blumenstein, M. Application of Artificial Neural Networks to Groundwater Dynamics in Coastal Aquifers. J. Coast. Res. 2007, II, 966–970. [Google Scholar]
  88. Szidarovszky, F.; Coppola, E.A.; Long, J.; Hall, A.D.; Poulton, M.M. A Hybrid Artificial Neural Network-Numerical Model for Ground Water Problems. Groundwater 2007, 45, 590–600. [Google Scholar] [CrossRef] [PubMed]
  89. Nourani, V.; Mogaddam, A.A.; Nadiri, A.O. An ANN-based model for spatiotemporal groundwater level forecasting. Hydrol. Process. 2008, 22, 5054–5066. [Google Scholar] [CrossRef]
  90. Kim, K.-D.; Lee, S.; Oh, H.-J. Prediction of ground subsidence in Samcheok City, Korea using artificial neural networks and GIS. Environ. Geol. 2009, 58, 61–70. [Google Scholar] [CrossRef]
  91. Jalalkamali, A.; Sedghi, H.; Manshouri, M. Monthly groundwater level prediction using ANN and neuro-fuzzy models: A case study on Kerman plain, Iran. J. Hydroinformatics 2011, 13, 867–876. [Google Scholar] [CrossRef]
  92. Nakhaei, M.; Nasr, A.S. A combined Wavelet-Artificial Neural Network model and its application to the prediction of groundwater level fluctuations. Geopersia 2012, 2, 77–91. [Google Scholar] [CrossRef]
  93. Fallah-Mehdipour, E.; Bozorg Haddad, O.; Mariño, M.A. Prediction and simulation of monthly groundwater levels by genetic programming. J. Hydro-Environ. Res. 2013, 7, 253–260. [Google Scholar] [CrossRef]
  94. Demirci, M.; Üneş, F.; Körlü, S. Modeling of groundwater level using artificial intelligence techniques: A case study of Reyhanli region in Turkey. Appl. Ecol. Environ. Res. 2019, 17, 2651–2663. [Google Scholar] [CrossRef]
  95. Hanifian, S.; Khaleghi, M.R.; Najarchi, M.; Jafarnia, R.; Varvani, J. A comparative study of artificial neural networks and multivariate regression for predicting groundwater depths in the Arak aquifer. Acta Geophys. 2023, 72, 419–432. [Google Scholar] [CrossRef]
  96. Shahbazi, M.; Zarei, H.; Solgi, A. A new approach in using the GRACE satellite data and artificial intelligence models for modeling and predicting the groundwater level (case study: Aspas aquifer in Southern Iran). Environ. Earth Sci. 2024, 83, 240. [Google Scholar] [CrossRef]
  97. Singh, R.M.; Datta, B.; Jain, A. Identification of Unknown Groundwater Pollution Sources Using Artificial Neural Networks. J. Water Resour. Plan. Manag. 2004, 130, 506–514. [Google Scholar] [CrossRef]
  98. Agarwal, A.; Singh, R.; Mishra, S.; Bhunya, P. ANN-based sediment yield models for Vamsadhara river basin (India). Water SA 2005, 31, 85–100. [Google Scholar] [CrossRef]
  99. Nayak, P.C.; Rao, Y.R.S.; Sudheer, K.P. Groundwater Level Forecasting in a Shallow Aquifer Using Artificial Neural Network Approach. Water Resour. Manag. 2006, 20, 77–90. [Google Scholar] [CrossRef]
  100. Prasad, R.K.; Mathur, S. Groundwater Flow and Contaminant Transport Simulation with Imprecise Parameters. J. Irrig. Drain. Eng. 2007, 133, 61–70. [Google Scholar] [CrossRef]
  101. Sreekanth, P.D.; Geethanjali, N.; Sreedevi, P.D.; Ahmed, S.; Kumar, N.R.; Jayanthi, P.D.K. Forecasting groundwater level using artificial neural networks. Curr. Sci. 2009, 96, 933–939. [Google Scholar]
  102. Mohanty, S.; Jha, M.K.; Kumar, A.; Sudheer, K.P. Artificial Neural Network Modeling for Groundwater Level Forecasting in a River Island of Eastern India. Water Resour. Manag. 2010, 24, 1845–1865. [Google Scholar] [CrossRef]
  103. Banerjee, P.; Singh, V.S.; Chatttopadhyay, K.; Chandra, P.C.; Singh, B. Artificial neural network model as a potential alternative for groundwater salinity forecasting. J. Hydrol. 2011, 398, 212–220. [Google Scholar] [CrossRef]
  104. Sreekanth, P.D.; Sreedevi, P.D.; Ahmed, S.; Geethanjali, N. Comparison of FFNN and ANFIS models for estimating groundwater level. Environ. Earth Sci. 2011, 62, 1301–1310. [Google Scholar] [CrossRef]
  105. Charulatha, G.; Srinivasalu, S.; Uma Maheswari, O.; Venugopal, T.; Giridharan, L. Evaluation of ground water quality contaminants using linear regression and artificial neural network models. Arab. J. Geosci. 2017, 10, 128. [Google Scholar] [CrossRef]
  106. Balavalikar, S.; Nayak, P.; Shenoy, N.; Nayak, K. Particle swarm optimization based artificial neural network model for forecasting groundwater level in Udupi district. In Proceedings of the International Conference on Electrical, Electronics, Materials and Applied Science, Secunderabad, India, 22–23 December 2017; p. 020021. [Google Scholar] [CrossRef]
  107. Saran, S.; Chaudhary, P.; Uttam, A.; Gupta, S. Analysis and Optimization of Groundwater Distribution Using SVM and Neural Networks. SSRN Electron. J. 2021, 17, 17–36. [Google Scholar] [CrossRef]
  108. Malik, A.; Bhagwat, A. Modelling groundwater level fluctuations in urban areas using artificial neural network. Groundw. Sustain. Dev. 2021, 12, 100484. [Google Scholar] [CrossRef]
  109. Kochhar, A.; Singh, H.; Sahoo, S.; Litoria, P.K.; Pateriya, B. Prediction and forecast of pre-monsoon and post-monsoon groundwater level: Using deep learning and statistical modelling. Model. Earth Syst. Environ. 2022, 8, 2317–2329. [Google Scholar] [CrossRef]
  110. Navale, V.; Mhaske, S. Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) model for Forecasting Groundwater Level in the Pravara River Basin, India. Model. Earth Syst. Environ. 2023, 9, 2663–2676. [Google Scholar] [CrossRef]
  111. Asher, M.J.; Croke, B.F.W.; Jakeman, A.J.; Peeters, L.J.M. A review of surrogate models and their application to groundwater modeling. Water Resour. Res. 2015, 51, 5957–5973. [Google Scholar] [CrossRef]
  112. Khan, J.; Lee, E.; Balobaid, A.S.; Kim, K. A Comprehensive Review of Conventional, Machine Leaning, and Deep Learning Models for Groundwater Level (GWL) Forecasting. Appl. Sci. 2023, 13, 2743. [Google Scholar] [CrossRef]
  113. Haggerty, R.; Sun, J.; Yu, H.; Li, Y. Application of machine learning in groundwater quality modeling—A comprehensive review. Water Res. 2023, 233, 119745. [Google Scholar] [CrossRef]
  114. Scheibe, T.D.; Murphy, E.M.; Chen, X.; Rice, A.K.; Carroll, K.C.; Palmer, B.J.; Tartakovsky, A.M.; Battiato, I.; Wood, B.D. An Analysis Platform for Multiscale Hydrogeologic Modeling with Emphasis on Hybrid Multiscale Methods. Groundwater 2015, 53, 38–56. [Google Scholar] [CrossRef]
  115. Deb, K.; Roy, P.; Hussein, R. Surrogate Modeling Approaches for Multiobjective Optimization: Methods, Taxonomy, and Results. Math. Comput. Appl. 2020, 26, 5. [Google Scholar] [CrossRef]
  116. Al-Maktoumi, A.; Rajabi, M.M.; Zekri, S.; Govindan, R.; Panjehfouladgaran, A.; Hajibagheri, Z. Accelerating regional-scale groundwater flow simulations with a hybrid deep neural network model incorporating mixed input types: A case study of the northeast Qatar aquifer. J. Hydroinformatics 2024, 26, 1333–1350. [Google Scholar] [CrossRef]
  117. Mohammadi, K. Groundwater Table Estimation Using MODFLOW and Artificial Neural Networks. In Practical Hydroinformatics; Abrahart, R.J., See, L.M., Solomatine, D.P., Eds.; Water Science and Technology Library; Springer: Berlin/Heidelberg, Germany, 2008; Volume 68, pp. 127–138. [Google Scholar] [CrossRef]
  118. Twumasi, F. Applying MODFLOW and Artificial Neural Networks to Model the Formation of MinePools in Underground Coal Mines. Master’s Thesis, Ohio University, Athens, OH, USA, 2018. [Google Scholar]
  119. Demissie, Y.; Valocchi, A.; Minsker, B.; Bailey, B. Bias-corrected groundwater model prediction uncertainty analysis. In Proceedings of the International Conference on Calibration and Reliability in Groundwater Modelling: Credibility of Modelling, Copenhagen, Denmark, 9–13 September 2007. [Google Scholar]
  120. Kantode, A.; Prashanth, T.; Ganguly, S. Development of a Precise Regional-Scale Groundwater Model by Coupling MODFLOW & Machine Learning Algorithms: A Case Study in Bist-Doab Region, Punjab, India. Available online: https://meetingorganizer.copernicus.org/EGU25/EGU25-1365.html (accessed on 30 July 2025).
  121. Coulibaly, P.; Anctil, F.; Aravena, R.; Bobée, B. Artificial neural network modeling of water table depth fluctuations. Water Resour. Res. 2001, 37, 885–896. [Google Scholar] [CrossRef]
  122. Verma, S.; Parthiban, P.; Ravikumar, K.; Das, I.C.; Das, A. Steady-state Assessment of Hydraulic Potential at Water Scarce regions of Agniyar River Basin, India using GMS-MODFLOW. Disaster Adv. 2023, 16, 38–43. [Google Scholar] [CrossRef]
  123. Demichele, F.; Micallef, F.; Portoghese, I.; Mamo, J.A.; Sapiano, M.; Schembri, M.; Schüth, C. Determining Aquifer Hydrogeological Parameters in Coastal Aquifers from Tidal Attenuation Analysis, Case Study: The Malta Mean Sea Level Aquifer System. Water 2023, 15, 177. [Google Scholar] [CrossRef]
  124. Ghaseminejad, A.; Uddameri, V. Physics-inspired integrated space-time Artificial Neural Networks for regional groundwater flow modeling. Hydrol. Earth Syst. Sci. 2020, 24, 5759–5779. [Google Scholar] [CrossRef]
  125. Payne, K.; Chami, P.; Odle, I.; Yawson, D.O.; Paul, J.; Maharaj-Jagdip, A.; Cashman, A. Machine Learning for Surrogate Groundwater Modelling of a Small Carbonate Island. Hydrology 2022, 10, 2. [Google Scholar] [CrossRef]
  126. Ruidas, D.; Pal, S.C.; Towfiqul Islam, A.R.M.; Saha, A. Hydrogeochemical Evaluation of Groundwater Aquifers and Associated Health Hazard Risk Mapping Using Ensemble Data Driven Model in a Water Scares Plateau Region of Eastern India. Expo Health 2023, 15, 113–131. [Google Scholar] [CrossRef]
  127. Alghafli, K.; Shi, X.; Sloan, W.; Ali, A.M. Investigating the Role of ENSO in Groundwater Temporal Variability across Abu Dhabi Emirate, United Arab Emirates Using Machine Learning Algorithms. Groundw. Sustain. Dev. 2025, 28, 101389. [Google Scholar] [CrossRef]
  128. Mohanty, S.; Jha, M.K.; Kumar, A.; Panda, D.K. Comparative evaluation of numerical model and artificial neural network for simulating groundwater flow in Kathajodi–Surua Inter-basin of Odisha, India. J. Hydrol. 2013, 495, 38–51. [Google Scholar] [CrossRef]
  129. Devarajan, K. Application of numerical and empirical models for groundwater level forecasting. Int. J. Res. Eng. Technol. 2015, 4, 127–133. [Google Scholar] [CrossRef]
  130. Moghaddam, H.K.; Moghaddam, H.K.; Kivi, Z.; Bahreinimotlagh, M.; Alizadeh, M.J. Developing comparative mathematic models, BN and ANN for forecasting of groundwater levels. Groundw. Sustain. Dev. 2019, 9, 100237. [Google Scholar] [CrossRef]
  131. Sanginabadi, H.; Saghafian, B.; Delavar, M. Coupled Groundwater Drought and Water Scarcity Index for Intensively Overdrafted Aquifers. J. Hydrol. Eng. 2019, 24, 04019003. [Google Scholar] [CrossRef]
  132. Malekzadeh, M.; Kardar, S.; Shabanlou, S. Simulation of groundwater level using MODFLOW, extreme learning machine and Wavelet-Extreme Learning Machine models. Groundw. Sustain. Dev. 2019, 9, 100279. [Google Scholar] [CrossRef]
  133. Nassery, H.R.; Zeydalinejad, N.; Alijani, F.; Shakiba, A. A proposed modelling towards the potential impacts of climate change on a semi-arid, small-scaled aquifer: A case study of Iran. Environ. Monit. Assess. 2021, 193, 182. [Google Scholar] [CrossRef]
  134. Mohammed, K.S.; Shabanlou, S.; Rajabi, A.; Yosefvand, F.; Izadbakhsh, M.A. Prediction of groundwater level fluctuations using artificial intelligence-based models and GMS. Appl. Water Sci. 2023, 13, 54. [Google Scholar] [CrossRef]
  135. Akbarifard, S.; Madadi, M.R.; Zounemat-Kermani, M. An artificial intelligence-based model for optimal conjunctive operation of surface and groundwater resources. Nat. Commun. 2024, 15, 553. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Deformation and Stress of a Runner in Large Francis Turbines Under Wide-Load Operating Conditions
Previous
Establishing a Sea Level Rise-Adjusted Design Flood Elevation for Buildings: A Comparative Study of Methods
Next