Next Article in Journal
Geographical Information Systems-Based Assessment of Evacuation Accessibility to Special Needs Shelters Comparing Storm Surge Impacts of Hurricane Irma (2017) and Ian (2022)
Previous Article in Journal
Application of Aerial Photographs and Coastal Field Data to Understand Sea Turtle Landing and Spawning Behavior at Kili-Kili Beach, Indonesia
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Enhanced Forecasting of Groundwater Level Incorporating an Exogenous Variable: Evaluating Conventional Multivariate Time Series and Artificial Neural Network Models

by
Md Abrarul Hoque
1,*,
Asib Ahmmed Apon
2,
Md Arafat Hassan
3,
Sajal Kumar Adhikary
4 and
Md Ariful Islam
5
1
Department of Civil Engineering, European University of Bangladesh, 2/4 Gabtoli, Mirpur, Dhaka 1216, Bangladesh
2
Department of Civil Engineering, Bangladesh University of Engineering and Technology, Dhaka 1000, Bangladesh
3
Department of Geography, Rutgers University, New Brunswick, NJ 08901, USA
4
Department of Civil Engineering, Khulna University of Engineering & Technology, Khulna 9203, Bangladesh
5
Department of Earth and Atmospheric Sciences, University of Nebraska–Lincoln, Lincoln, NE 68588, USA
*
Author to whom correspondence should be addressed.
Geographies 2025, 5(1), 1; https://doi.org/10.3390/geographies5010001
Submission received: 5 December 2024 / Revised: 27 December 2024 / Accepted: 29 December 2024 / Published: 31 December 2024

Abstract

:
Continuous and uncontrolled extraction of groundwater often creates tremendous pressure on groundwater levels (GWLs). As a part of sustainable planning and effective management of water resources, it is crucial to assess the existing and forecasted GWL conditions. In this study, an attempt was made to model and forecast GWL using artificial neural networks (ANNs) and multivariate time series models. Autoregressive integrated moving average (ARIMA) and ARIMA models incorporating exogenous variables (ARIMAX) were adopted as the time series models. GWL data from five monitoring wells from the study area of the Kushtia District in Bangladesh were used to demonstrate the modeling exercise. Rainfall (RF) was taken as the exogenous variable to explore whether its inclusion enhanced the performance of GWL forecasting using the developed models. It was evident from the results that the multivariate ARIMAX model (with the sum of squared errors, SSE, of 15.143) performed better than the univariate ARIMA model with an SSE of 16.585 for GWL forecasting. This demonstrates the fact that the multivariate time series models generated enhanced forecasting of GWL compared to the univariate time series models. When comparing the models, it was found that the ANN-based model outperformed the time series models with enhanced forecasting accuracy (SSE of 9.894). The results also exhibit a significant correlation coefficient (R) of 0.995 (model ANN 6-8-1) for the existing and predicted data. The current study conclusively proves the superiority of ANN over the time series models for the enhanced forecasting of GWL in the study area.

1. Introduction

Bangladesh is one of the most densely populated countries in the world. There is population growth and increasing urbanization in addition to climate change. For this reason, the stress on groundwater is growing rapidly. This earthly resource is essential, particularly for developing countries that rely on it for agricultural needs. Additionally, it plays a crucial role in the earth’s water cycle. Like many districts in Bangladesh, Kushtia relies heavily on agriculture, as a large irrigation project named the Ganges–Kobadak Project covers an area of 197,500 hectares and serves the region. Approximately 68% of the area is covered by irrigation [1]. According to the Statistical Yearbook of Bangladesh 2022 (by the Bangladesh Bureau of Statistics), irrigation in Kushtia District is carried out using various types of tube wells, including deep, shallow, and hand tube wells, as well as power pumps. Some authors [2] analyzed the importance of safe water sources and assessed conventional water sources, including groundwater. So, accurate and reliable forecasting of GWL is necessary. Furthermore, it is also essential for the sustainable management of water resources and to facilitate the conjunctive use of groundwater and surface water resources.
Degradation of GWL is common, and research [3] indicates that during the previous 25 years, the total volume of groundwater storage exhibited a declining trend throughout the Southeast Asian region. The increase in impervious surface caused by urbanization reduces infiltration, which has an impact on groundwater storage [4]. Groundwater is sometimes the only dependable freshwater supply in many developing nations since it is widely accessible, relatively inexpensive to collect, and usually of higher quality than surface water [5,6]. The dependence on groundwater is especially noticeable in areas with little or highly contaminated surface water [7]. The nation’s dense population, expanding agricultural needs, and quick industrialization are the main causes of this widespread reliance [8]. Lowering GWLs in agricultural areas might result in reduced crop yields because of limited water supplies, exacerbating food security problems [9,10]. Apart from the physical loss of groundwater, another major concern is its quality, especially in areas where contamination is common [11].
Time series analysis has recently become substantially more appealing as a statistical tool for developing forecast models, and its applicability has increased globally [12]. The authors’ studies found that RF is an essential component of groundwater recharge potential [13]. Furthermore, RF significantly influences groundwater recharge potential [14]. GWLs can be significantly impacted by the fluctuation of RF, which is a key source of groundwater recharge [15]. It is generally acknowledged that one of the most important ways to increase the precision of GWL predictions is to incorporate RF data into both ANN and ARIMA models [16]. Because of Bangladesh’s monsoon climate, the relationship between RF and GWL is significant [17]. For these reasons, RF is considered an endogenous variable in this study.
The goals of this study are to model and predict GWL fluctuations using ANN models and statistical time series ARIMA-based models, incorporating an exogenous variable to analyze its effects. It is widely accepted that ANN and ARIMA-based time series models are powerful tools for predicting various data. The performances of ARIMA-based ANN models were compared to determine which modeling strategy is better for predicting GWL fluctuations. In addition, assessing the ability of models to predict using various model evaluation criteria is also a primary focus of this study. One important issue in groundwater management is predicting GWL changes, which provides essential information about future groundwater availability and can help in developing policies for sustainable usage [18,19]. Furthermore, these models can be used to conduct additional research on groundwater issues.

2. The Study Area

The Kushtia District in the Khulna division of Bangladesh was selected for this study. The area covered by the district is 1608.8 km2. It is positioned in southwest Bangladesh, encompassing the coordinates of 23°42′ to 24°12′ north and 88°42′ to 89°22′ east. Figure 1 shows the location of the study area. The area is characterized by several rivers, such as the Ganges, Mathabhanga, Kaligonga, and Kumar, which flow across the district. Additionally, the district has a flat, alluvial landscape, which is typical for the delta region. It has fertile soil that makes it favorable to agriculture and is highly dependent on irrigation. Groundwater is an almost worldwide source of superior freshwater [20]. The district also largely depends on it.

2.1. Aquifer Characteristics

The aquifers in the area consist of alluvial deposits, which include sand, silt, and clay. These formations typically exhibit varying permeability that influences groundwater movement. Shallow aquifers are usually found at lesser depths and are directly influenced by surface water and recharge primarily through RF and seasonal flooding. The authors found that the diffusivity (degrees of flow) varied from 181,143 m2/day to 256,788 m2/day, and the overall estimated parameter results of the aquifer system show that the area is hydrogeologically favorable for groundwater development provided the other conditions are fulfilled [21].

2.2. Data Description

For GWL modeling and prediction, in the current study, groundwater monitoring stations from five upazilas (i.e., Bheramara, Daulatpur, Kushtia Sadar, Kumarkhali, and Mirpur), along with an RF station of the district, were selected. The GWL data were collected on a weekly basis. Accordingly, 414 weekly GWL data points from 1999 to 2006 were sourced from the Bangladesh Water Development Board (BWDB). The RF data obtained here were exogenous input and collected for the same periods. The available GWL data were measured in m.PWD (meters above the public works datum (PWD)) used by BWDB. This datum is located 0.46 m below the mean sea level. Information on the selected GWL monitoring stations and RF stations, along with their locations, are given in Table 1. The data are visualized for each currently available station in Figure 2 for the same period. The weekly data are provided in Appendix A.

3. Methodology

The primary objective of this study is to model and forecast GWL changes using ANN and statistical ARIMA-based time series models. Univariate and multivariate models were constructed, and the top-performing model was identified. In both cases, model development and evaluations were carried out. Afterward, GWL predictions were carried out based on the existing data and variables. Moreover, the impact of incorporating climate data was also observed.

3.1. Advantages of ANN Models

Time series modeling is considered one of the most robust statistical methods for studying the connection between climate variation, GWL, and forecasting [18]. The main advantage of using ANN models over conventional techniques is that they do not require underlying processes, which are complex in nature, to be explicitly defined in mathematical form [22]. In [23], researchers employed ANN models to forecast GWL changes for future periods. Results from [24] showed that forecasting time series is more accurate in ANN models than in ARIMA-based models. Accepting all complex parameters as input, ANN models generate patterns during model training, and then they use the same patterns to generate forecasts or predictions [25]. The strength of ANN models in prediction lies in their data-driven nature, their ability to detect previously unseen patterns, and their efficiency with large datasets [26].

3.2. Prediction Approaches of the Models

Several researchers have identified two main approaches in GWL prediction: data-driven models and numerical models. Researchers have also found that data-driven models are useful in assessing different aspects such as uncertainty, variability, and complexities in water resources and environmental problems [27]. In recent decades, artificial intelligence has been widely applied in studies related to water resources [27]. For example, ANN is used for daily weather forecasting [25], RF forecasting [26], and GWL prediction purposes [27,28]. However, this method has not been widely used until recently [28]. Due to the limited understanding of aquifer properties in Bangladesh, the applicability of conventional GWL prediction models is limited; for this reason, soft computing tools are good alternatives that provide higher efficacy [29]. The ANN model can discern connections in historical data that cannot readily be seen. In this way, it aids in prediction and forecasting. The authors of [30] worked on water quality forecasting. ARIMAX models have also been applied in forecasting electricity usage [31], grain production [32], domestic water consumption [33], and droughts [34]. “ARIMA models are well-known for their notable forecasting accuracy and flexibility in representing several different types of time series” [35].

3.3. Development Strategy of ANN Models

There are numerous techniques used to create and simulate a neural network [36]. In the current study, several model architectures were analyzed to find the most accurate prediction model via the MATLAB platform using the neural network toolbox. This toolbox offers proficiency in designing various neural network configurations with many applications [36]. Creating an ANN model requires defining its type, structure, variable processing, training algorithm, and stopping criteria [37].

3.3.1. Dataset Preparation

First, the datasets were collected and processed as inputs for the models. For univariate models, only GWL data were used. However, GWL and RF data were taken as inputs for multivariate models. The data were allocated as follows: 70% (i.e., 290 data points) for training, 15% (i.e., 62 data points) for validation, and 15% (i.e., 62 data points) for testing, which is typical. Before inputting the data into the ANN model, the data should be processed. In this study, weekly GWL and RF data were prepared on a one-week lag basis. This means that the data were sorted as follows: GWLt, GWLt−1, RFt, and so on, depending on the number of inputs.

3.3.2. ANN Model Architecture and Model Training

The ANN model is inspired by the human brain and neurons are the basic units of its structure. It is similar to brain cells. Each neuron takes inputs, processes them, and produces an output. ANNs are organized into three layers. The input layer is where the network receives data. It contains neurons that represent input features. If there are n input features, this layer will have n neurons. Hidden layers are the layers between the input and output. Each hidden layer can have hi neurons, where i indexes the hidden layer (e.g., h1, h2,…, hk). Each neuron in a hidden layer transforms the input using weights and activation functions. Many researchers, e.g., in Ref. [27], have selected the hidden layer node count based on a trial-and-error procedure. Therefore, in the current study, the hidden layer size was specified by following the trial-and-error approach, based on the model’s performance. The output layer produced the final result or prediction of the network. The multilayer perceptron feedforward ANN model was applied. It is widely used in hydrological modeling [37]. In particular, a configuration with a single hidden layer is one of the most widely used ANNs for time series modeling and prediction [38]. The ANN 6-3-1 structure denotes that this model has six neurons in the input layer, three in the hidden layer, and one in the output layer.
Connections between neurons have weights, and each neuron has a bias. To compute the output, σj, of a neuron, the weighted inputs are summed up, a bias is added, and then an activation function is applied. The output can be expressed mathematically as Equation (1).
σ j = f ( w ij × x i + b j )
where wij is the weight that connects input i to neuron j, xi is the input value from the previous layer, and bj is the bias term for neuron j.
The sigmoid activation function transforms the input (the weighted sum w ij × x i + b j passes through the activation function) into an output f(x). The function transforms the inputs into a range of (0, 1) so that the input value of the next layer is within a fixed range. In summary, it helps determine the output, σj, of a neuron. The function can be expressed mathematically using Equation (2):
f ( x ) = 1 1 + e - x
where x is the given input.
Model training involves adjusting the network’s weights and biases to minimize the difference between the predicted and actual outputs. This process allows the ANN to learn from the data. This study utilizes the commonly utilized Levenberg–Marquardt backpropagation technique for model training. The training was stopped after 1000 epochs of backpropagation, which is considered acceptable in this study. On the MATLAB platform (version R2018a), along with the prepared weekly GWL and RF data, the hidden layer size was also provided as input. The software performed model training and prediction.

3.4. ARIMA-Based Model Development

When predicting various extreme weather events such as heavy RF and droughts, ARIMAX models proved effective [39]. Apart from ANN models, this study adopted time series models for predicting GWL fluctuations: ARIMAX and ARIMA. One of the most frequently used time series models for analyzing and forecasting hydrologic data is ARIMA, which combines autoregressive (AR) and moving average (MA) components. For ARIMA-based models, the procedure involves specifying the model structure, estimating parameters, conducting residual diagnostics, and forecasting the data series. This study was carried out using the econometric modeler toolbox on the MATLAB platform. In the toolbox, the model parameters (p, d, and q) were provided along with the GWL data and RF as inputs. The toolbox performed the prediction using the model’s formula and provided the detailed results of different plots to evaluate performances. Explanations of the model’s parameters and formulas are provided in the following sections.

3.4.1. Mathematical Expression of the ARIMA Model

The ARIMA-based time series model contains three parts (p, d, and q), where p = the order of autoregression, d = the order of integration (differencing), and q = the order of moving average. The general mathematical form, or multiplicative equation, of an ARIMAX (p, d, q) model with one exogenous variable is given by Equation (3):
ϕ ( L ) ( 1 L ) d Y t = c + β x t + θ ( L ) ϵ t
where c is the constant, ϵ t is the error term, L = lag operator, ϕ(L) = (1 − ϕ1L − ϕ2L2 −… − ϕpLp) is the AR polynomial, θ(L) = (1 + θ1L + θ2L2 + … + θqLq) is the MA polynomial, and β denotes the coefficient for the exogenous variable.
For example, this study considers a time series with AR(2), d(1), MA(1), and one exogenous variable. The model could be specified as Equation (4):
Y t = c + ϕ 1 Y t - 1 + ϕ 2 Y t - 2 + β 1 x t - 1 + θ 1 ϵ t - 1 + ϵ t
where Yt represents the prediction output, c is the constant term, ϕ1 and ϕ2 are the autoregressive coefficients for the lagged values of Y, specifically Yt−1 and Yt−2; β1 is the coefficient for the lagged exogenous variable, Xt−1, θ1 represents the coefficient associated with the lagged error term (ϵt−1) in the moving average component of the model, and ϵ t represents the error term at time t.

3.4.2. Model Identification

For the ARIMA-based models, the time series (GWL) was transformed into a stationary series by differencing (d = 1, 2, 3,…). From this step, the order of integration d was obtained and taken as 3 (as the data become stationary in this order). The values for p and q were selected based on ACF (autocorrelation function) and PACF (partial autocorrelation function) plots. With no significant correlation beyond lag 3, p and q were set to 3. The blue lines in the plots of the software outputs indicate the confidence intervals to assess whether the autocorrelations are statistically significant or not. If the autocorrelation value falls outside the significance lines, then the autocorrelation at that lag is statistically significant. After that, each possible model combination was analyzed and evaluated for accuracy to pick the best model. Once the model was identified, its parameters were estimated. The ACF and PACF plots are shown in Figure 3a,b.

3.5. Model Evaluation Criteria

In forecasting, accuracy is the primary concern, not the processing time. It has been observed that there are no models that can forecast with complete accuracy; however, errors can be reduced by using various techniques [25]. In this study, the mean squared error (MSE), root mean square error (RMSE), Nash–Sutcliffe efficiency (NSE), and the sum of squared errors (SSE) are used for ANNs performance evaluation. ANNs have proven to be an essential tool for accurate GWL modeling, along with various other AI methods [40]. Many authors used RMSE values to evaluate the performance of the ANN model [41]. Mathematically, NSE can be expressed using Equation (5). In this study, SSE is considered a crucial component of the model’s performance evaluation. Additional metrics are used, providing a more comprehensive assessment of the model’s performance. In order to analyze the accuracy of ARIMA-based models, the Akaike information criterion (AIC) and Bayesian information criterion (BIC) were also applied to evaluate the performance of the models.
NSE = 1 q = 1 n [ Y obs ( q ) Y est ( q ) ] 2 q = 1 n [ Y obs ( q ) Y ¯ est ( q ) ] 2
where Y obs denotes the observed data, Y est denotes the estimated data, Y - est denotes the mean values of estimated data, and n denotes the number of observations.

4. Results and Discussion

In the current study, we analyzed ANN-based and ARIMA-based models to predict GWLs in the Kushtia District in Bangladesh, after developing and evaluating these models. In addition, an attempt was also made to test the prediction capability of the models. The results indicated that incorporating exogenous variables provided enhanced results. As further evidence, studies also indicate that the relationship between GWL changes and meteorological parameters, such as precipitation, is significant [42]. The study showed that ANN models yield more accurate predictions than ARIMA-based models.

4.1. Performance of ANN Models

The ANN 6-8-1 multivariate model with Station ID: KG-1 produced the lowest error (SSE 9.894) among the five stations. On the other hand, for univariate models, the best-performing model was found for the same station with an SSE value of 10.809; the model architecture was ANN 3-7-1. It is evident that the multivariate model performed more efficiently than the univariate model. The results of the best-performing models for each station are shown in Table 2 and Table 3.

4.2. Performance Overview of Different ANN Models in the Building Stages

Table 4 presents thorough results for the KG-5 station for the training, validation, and testing stages of the models. The table illustrates the performances of these models at different stages of building ANN models. The performances were evaluated using MSE and NSE for all stages. Since ANN modeling has three phases, the performance of each stage is compared for greater insight and comprehensiveness. Here, two model evaluation criteria (MSE and NSE) were sufficiently used to provide an overall evaluation of the model’s performance.

4.3. Graphical Evaluation of the Selected Performances of the ANN Model

The scatterplot shown in Figure 4a depicts the ANN 8-9-1 model’s actual versus predicted data for station KG-5. This model’s output is illustrated to provide an overview of the iterative procedure involved in finding the best predictive model and its comparative performance. The correlation coefficient value (R = 0.993) indicates a highly significant positive linear relationship between the observed and predicted values. A graphical representation of the actual and predicted data plots for that station is shown in Figure 4b.

4.4. Performances of ARIMA-Based Models

The best performance of the ARIMAX models (SSE = 15.143) was observed for station KG-5 with the model architecture ARIMAX (3,0,2). However, the best performance of the univariate models was observed at the same station, with the model architecture ARIMA (2,0,1) and an SSE of 16.585. It is noticeable that multivariate models performed more efficiently than univariate models. Detailed results are provided in Table 5 and Table 6.

4.5. Performance Overview of Different ARIMA Models

Several ARIMA-based models were built after the model’s identification; detailed results of the selected ARIMA (p, d, q) models for station KG-5 are presented in Table 7, along with AIC and BIC values. This table provides a comprehensive overview of the performances of different models and insights into how the procedure is incorporated to find an appropriate prediction model. It also illustrates model evaluation criteria with AIC and BIC values.

4.6. Evaluation of the Parameters of the ARIMA Models

Additionally, the model parameters and their statistics for ARIMA (2,0,1) at station KG-5 are provided in Table 8. The parameter values are the estimated coefficients for each parameter (e.g., AR coefficients); these parameters have a significant role in the predictive equations, and detailed values confirm statistical significance for the model.

4.7. Graphical Evaluation of the Selected ARIMA Model Performances

The ARIMA (2,0,1) model’s performance for Station ID: KG-5 is graphically presented in Figure 5a–e. It should be noted that 48 unique ARIMA models for each of the five GWL stations were tested (totaling 240 ARIMA models) to evaluate their fits. The plots illustrate a segment of an extensive iterative process in which the model is refined through the analysis of observed results. The blue lines in the ACF and PACF plots indicate the confidence intervals to assess whether the autocorrelations are statistically significant or not as described earlier. A linear alignment of points in the Q-Q plot along the reference line suggests that the residuals adhere to normality, and it indicates that the model adequately captures the underlying data patterns. The plot shows that there is little linear alignment of the points and the model building process should be repeated.

4.8. Forecasting of GWL

Finally, using the best-performing models (ANN 6-8-1 and ARIMAX 3,0,2), GWL data were predicted using existing data. It was observed that ANNs predicted GWL values, ranging from a maximum of 10.797 m to a minimum of 5.875 m. However, the highest and lowest values for the actual data were 12.610 m and 5.730 m, respectively. The main findings are summarized in Table 9, along with the corresponding stations. The performances of ANN models can be attributed to the model’s architecture or training parameters. The ARIMAX results could be attributed to its incorporation of exogenous regressors, which might better account for factors influencing water levels. Future work should also consider incorporating a wider range of models and hybrid approaches to balance the strengths of each predictive technique. Moreover, additional validation against independent datasets will be crucial to confirm the robustness of the models and ensure their generalizability. In addition, a graphical representation of predicted GWL data is shown in Figure 6a for the ARIMAX (3,0,2) model and in Figure 6b for the ANN 6-8-1 model.

4.9. Relative Improvement in the Performances of Models and Comparison of the Best Models

A detailed analysis was conducted to validate the performances of the models and to provide an overall comparison between the ANN and ARIMA-based models. The evaluation was based on three primary performance metrics: performance enhancement, performance degradation, and reference model, against which other models are assessed.

4.9.1. Assessing the Efficiency of Models Relative to ANN

Compared to the ANN multivariate model with a 6-8-1 architecture, the results of other models indicate a notable drop; specifically, this model outperforms the best-performing ANN univariate, ARIMAX, and ARIMA models, with reductions of −8.459%, −34.658%, and −40.340%, respectively. In contrast to the ANN univariate model with a 3-7-1 architecture, the best-performing ANN multivariate model showed an enhancement of the performance with an increase of 9.241%, while ARIMAX and ARIMA experienced performance declines of −28.620% and −34.826%, respectively. The comparison indicates that, although the ANN univariate delivers enhanced predictive accuracy, it also exhibits substantial prediction deviations.

4.9.2. Assessing the Efficiency of Models Relative to the ARIMA-Based Model

Relative to the ARIMAX (3,0,2) configuration, both ANN multivariate and ANN univariate models showed significant performance improvements, with increases of 53.043% and 40.096%, respectively. However, the ARIMA model underperformed, demonstrating a performance decline of −8.694%. With the ARIMA (2,0,1) architecture, all models demonstrated a marked performance enhancement: the ANN multivariate model achieved a 67.617% improvement, the ANN univariate model improved by 53.436%, and ARIMAX showed a 9.522% enhancement. Comparative performance metrics of different forecasting models are presented in Table 10.

4.9.3. Model Accuracy via Regression Plot

In order to provide a more comprehensive overview of the performance of the best models, ARIMAX (3,0,2) and ANN 6-8-1, the scatterplots are presented in Figure 7a,b. It is clear from the figures that the models achieved significant accuracy. Their corresponding correlation coefficient (R) values are also shown.

5. Conclusions

The purpose of this study was to assess and evaluate the efficacy of ANN-based and ARIMA-based models for predicting GWL in Kushtia, Bangladesh. Such modeling approaches, particularly those that incorporate exogenous factors, are not extensively used in this region. In order to achieve the best model fit, the ANN model was extensively tuned by testing a large number of model configurations, including hidden layer sizes and network architectures, on the MATLAB platform. To ensure that the ARIMA model accurately captured the underlying time series structure, ACF and PACF plots were used to determine the model orders.
The key findings of this study are as follows:
  • The ANN model (6-8-1) outperformed the ARIMA-based best model (SSE 15.143) by 53.043% and achieved the highest predictive accuracy with an SSE 9.894.
  • The multivariate ANN model showed 9.241% higher accuracy than the best univariate ANN model.
  • The ARIMAX model improved prediction accuracy by 9.522% over the best ARIMA model.
  • It is evident that models that included exogenous variables provided more reliable GWL predictions than univariate models.
Even though this work provides useful insights into GWL prediction, it is necessary to acknowledge several limitations. A primary constraint is the limited availability of GWL data. By including more monitoring wells, the extended dataset would significantly enhance the reliability of the model.
The current study is expected to be a useful tool for water resource managers and policymakers. Despite its limitations, this study provides significant insights into improving GWL forecasting, with practical implications for groundwater management, such as studying future GWL changes in regions with similar aquifer characteristics or hydrogeological conditions. Moreover, it lays the groundwork for future research. Overall, the findings of the current study could be helpful to improve groundwater monitoring, decision-making, and aquifer management strategies.

Author Contributions

Conceptualization, M.A.H. (Md Abrarul Hoque), A.A.A., M.A.H. (Md Arafat Hassan), S.K.A. and M.A.I.; methodology, M.A.H. (Md Abrarul Hoque), A.A.A. and S.K.A.; software (MATLAB: version R2018a), M.A.H. (Md Abrarul Hoque) and A.A.A.; validation, S.K.A.; formal analysis, M.A.H. (Md Abrarul Hoque) and A.A.A.; investigation, S.K.A.; resources, S.K.A. and M.A.I.; data curation, S.K.A.; writing—original draft preparation, M.A.H. (Md Abrarul Hoque), A.A.A. and M.A.H. (Md Arafat Hassan); writing—review and editing, M.A.H. (Md Abrarul Hoque), A.A.A., M.A.H. (Md Arafat Hassan) and S.K.A.; visualization, M.A.H. (Md Abrarul Hoque); supervision, M.A.H. (Md Arafat Hassan) and S.K.A.; project administration, M.A.H. (Md Arafat Hassan) and S.K.A.; funding acquisition, M.A.I. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Bangladesh Ministry of Science and Technology. The funding was provided for the years 2021–2022.

Data Availability Statement

The data and codes will be available from the corresponding author upon reasonable request.

Acknowledgments

The authors acknowledge the necessary data support provided by the Bangladesh Water Development Board (BWDB) to carry out this study.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A. Data Details

Table A1 and Table A2 show the weekly RF data and GWL data that were used in this study. The weekly RF and GWL data exhibit clear temporal variations.
Table A1. RF data for Station ID: KR-1 (Location: Mirpur–Kushtia).
Table A1. RF data for Station ID: KR-1 (Location: Mirpur–Kushtia).
WeekRFt
(mm)
WeekRFt
(mm)
WeekRFt
(mm)
WeekRFt
(mm)
3 March 20030.0016 February 20040.0031 January 20050.0016 January 20060.00
10 March 20030.0023 February 20040.007 February 20050.0023 January 20060.00
17 March 200333.001 March 20040.0014 February 20050.0030 January 20060.00
24 March 200332.508 March 20040.0021 February 20050.006 February 20060.00
31 March 200377.0015 March 20040.0028 February 20050.0013 February 20060.00
7 April 20030.0022 March 20040.007 March 20050.0020 February 20060.00
14 April 20030.0029 March 20040.0014 March 200514.0027 February 20060.00
21 April 200325.505 April 20040.0021 March 20050.006 March 20060.00
28 April 200314.5012 April 20046.7028 March 200521.0013 March 20060.00
5 May 20033.5019 April 20040.004 April 200545.0020 March 20060.00
12 May 200369.7026 April 200424.0011 April 20050.0027 March 20060.00
19 May 20030.003 May 200426.0018 April 20050.003 April 200610.00
26 May 200315.0010 May 20040.0025 April 20050.0010 April 200614.00
2 June 20039.0017 May 20040.002 May 200566.5017 April 20060.00
9 June 200335.0024 May 200482.009 May 200590.0024 April 200618.00
16 June 200346.0031 May 200459.5016 May 200535.001 May 20060.00
23 June 200364.007 June 200460.0023 May 200514.008 May 200661.40
30 June 2003121.5014 June 200432.0030 May 20050.0015 May 200611.00
7 July 20030.0021 June 2004116.306 June 20050.0022 May 20060.00
14 July 2003164.0028 June 200462.5013 June 20050.0029 May 2006146.00
21 July 20032.205 July 200426.0020 June 20056.005 June 200677.00
28 July 200329.0012 July 200428.0027 June 200529.0012 June 200651.00
4 August 2003111.5019 July 200470.004 July 200544.5019 June 200650.00
11 August 200313.2026 July 2004161.8011 July 200545.0026 June 200639.00
18 August 200384.002 August 200444.5018 July 2005247.503 July 200612.00
25 August 200358.009 August 200419.0025 July 200573.0010 July 200699.00
1 September 200339.5016 August 200447.001 August 200514.0017 July 200629.00
8 September 200384.0023 August 200412.008 August 200518.0024 July 200646.00
15 September 200333.5030 August 200428.5015 August 200547.0031 July 200663.50
22 September 20030.006 September 200475.0022 August 20050.007 August 20067.50
29 September 200341.0013 September 2004188.0029 August 200520.5014 August 200662.00
6 October 200368.5020 September 2004340.005 September 200595.0021 August 200623.00
13 October 2003102.0027 September 200420.0012 September 200593.0028 August 200634.00
20 October 20038.004 October 200495.0019 September 2005435.004 September 200659.00
27 October 200311.0011 October 2004117.5026 September 200520.0011 September 200627.00
3 November 20037.0018 October 200424.003 October 2005333.5018 September 200695.00
10 November 20030.0025 October 20040.0010 October 2005118.0025 September 2006339.00
17 November 20030.001 November 20040.0017 October 20050.002 October 200622.00
24 November 20030.008 November 20040.0024 October 2005349.509 October 20060.00
1 December 20030.0015 November 20040.0031 October 20050.0016 October 20060.00
8 December 20030.0022 November 20040.007 November 20050.0023 October 20060.00
15 December 20030.0029 November 20040.0014 November 20050.0030 October 20060.00
22 December 20030.006 December 20040.0021 November 20050.006 November 20060.00
29 December 20035.5013 December 20040.0028 November 20050.0013 November 200610.00
5 January 20040.0020 December 20040.005 December 20050.0020 November 20060.00
12 January 20040.0027 December 20040.0012 December 20050.0027 November 20060.00
19 January 20040.003 January 20050.0019 December 20050.004 December 20060.00
26 January 20042.0010 January 20050.0026 December 20050.0011 December 20060.00
2 February 20040.0017 January 20050.002 January 20060.0018 December 20060.00
9 February 20040.0024 January 200519.009 January 20060.0025 December 20060.00
Table A2. GWL data for Station ID: KG-1 (Location: Bheramara–Kushtia).
Table A2. GWL data for Station ID: KG-1 (Location: Bheramara–Kushtia).
WeekGWLt
(m.PWD)
WeekGWLt
(m.PWD)
WeekGWLt
(m.PWD)
WeekGWLt
(m.PWD)
3 March 20036.6516 February 20047.3631 January 20057.2516 January 20067.65
10 March 20036.5623 February 20047.247 February 20057.1223 January 20067.42
17 March 20036.491 March 20047.1014 February 20056.9930 January 20067.27
24 March 20036.438 March 20046.9621 February 20056.886 February 20067.11
31 March 20036.3615 March 20046.8228 February 20056.7813 February 20066.95
7 April 20036.2922 March 20046.687 March 20056.6320 February 20066.84
14 April 20036.2129 March 20046.5614 March 20056.5727 February 20066.71
21 April 20036.145 April 20046.4421 March 20056.456 March 20066.58
28 April 20036.2012 April 20046.9628 March 20056.3713 March 20066.46
5 May 20036.0119 April 20046.264 April 20056.3020 March 20066.35
12 May 20036.0926 April 20046.2211 April 20056.2227 March 20066.24
19 May 20036.053 May 20046.1818 April 20056.123 April 20066.12
26 May 20036.0210 May 20046.1225 April 20056.0110 April 20066.06
2 June 20036.0017 May 20046.062 May 20055.9817 April 20065.95
9 June 20036.0224 May 20046.159 May 20055.9824 April 20065.86
16 June 20036.0631 May 20046.2716 May 20055.961 May 20065.81
23 June 20036.247 June 20046.3223 May 20055.938 May 20065.77
30 June 20036.5314 June 20046.3630 May 20055.8915 May 20065.73
7 July 20036.9021 June 20046.426 June 20055.8322 May 20065.73
14 July 20037.4328 June 20046.6313 June 20055.8329 May 20065.73
21 July 20038.055 July 20046.9220 June 20055.795 June 20065.78
28 July 20038.5612 July 20047.2227 June 20055.8912 June 20065.99
4 August 20039.0119 July 20047.804 July 20056.0719 June 20066.23
11 August 20039.3426 July 20048.5511 July 20056.3326 June 20066.39
18 August 20039.702 August 20049.0718 July 20056.853 July 20066.57
25 August 200310.079 August 20049.3725 July 20057.4410 July 20066.74
1 September 200310.4416 August 20049.561 August 20057.4517 July 20067.10
8 September 200310.6623 August 20049.768 August 20058.3024 July 20067.49
15 September 200311.1330 August 200410.0615 August 20058.9031 July 20067.95
22 September 200311.266 September 200410.4022 August 20059.137 August 20068.34
29 September 200311.6013 September 200410.5529 August 20059.5914 August 20068.68
6 October 200311.6120 September 200410.665 September 20059.9221 August 20068.82
13 October 200311.6027 September 200411.1412 September 20059.8528 August 20069.11
20 October 200311.104 October 200411.1319 September 20059.674 September 20069.50
27 October 200310.6711 October 200411.5726 September 20059.6511 September 20069.75
3 November 200310.4618 October 200411.193 October 200510.3618 September 20069.92
10 November 200310.0825 October 200410.6610 October 200510.5625 September 200610.57
17 November 20039.841 November 200410.1917 October 200510.912 October 200610.65
24 November 20039.548 November 20049.8724 October 200511.429 October 200610.34
1 December 20039.3015 November 20049.6031 October 200510.8816 October 20069.96
8 December 20039.0622 November 20049.297 November 200510.3723 October 20069.56
15 December 20038.8229 November 20049.0114 November 20059.8930 October 20069.22
22 December 20038.656 December 20048.7421 November 20059.526 November 20068.93
29 December 20038.4613 December 20048.5028 November 20059.2113 November 20068.67
5 January 20048.2820 December 20048.245 December 20058.9420 November 20068.43
12 January 20048.1127 December 20048.0412 December 20058.6727 November 20068.26
19 January 20047.913 January 20057.8519 December 20058.444 December 20068.05
26 January 20047.7610 January 20057.6626 December 20058.2111 December 20067.85
2 February 20047.6117 January 20057.482 January 20067.9918 December 20067.67
9 February 20047.4724 January 20057.379 January 20067.8125 December 20067.51

References

  1. Hossain, M.B.; Roy, D.; Mahmud, M.N.H.; Paul, P.L.C.; Yesmin, M.S.; Kundu, P.K. Early transplanting of rainfed rice minimizes irrigation demand by utilizing rainfall. Environ. Syst. Res. 2021, 10, 34. [Google Scholar] [CrossRef]
  2. Adhikary, S.K.; Das, S.K.; Chaki, T.; Rahman, M. Identifying safe drinking water source for establishing sustainable urban water supply scheme in Rangunia municipality, Bangladesh. In Proceedings of the 20th International Congress on Modelling and Simulation, Adelaide, Australia, 1–6 December 2013; pp. 3134–3140. [Google Scholar]
  3. Lu, Y.; Dai, L.; Yan, G.; Huo, Z.; Chen, W.; Lan, J.; Zhang, C.; Xu, Q.; Deng, S.; Chen, J. Effects of various land utilization types on groundwater at different temporal scales: A case study of Huocheng plain, Xinjiang, China. Front. Environ. Sci. 2023, 11, 1225916. [Google Scholar] [CrossRef]
  4. Mishra, N.; Khare, D.; Gupta, K.K.; Shukla, R. Impact of land use change on groundwater—A review. Adv. Water Resour. Prot. 2014, 2, 28–41. [Google Scholar]
  5. Wada, Y.; Van Beek, L.P.; Van Kempen, C.M.; Reckman, J.W.; Vasak, S.; Bierkens, M.F. Global depletion of groundwater resources. Geophys. Res. Lett. 2010, 37, L20402. [Google Scholar] [CrossRef]
  6. Foster, S.; Pulido-Bosch, A.; Vallejos, Á.; Molina, L.; Llop, A.; MacDonald, A.M. Impact of irrigated agriculture on groundwater-recharge salinity: A major sustainability concern in semi-arid regions. Hydrogeol. J. 2018, 26, 2781–2791. [Google Scholar] [CrossRef]
  7. Schmoll, O. (Ed.) Protecting Groundwater for Health: Managing the Quality of Drinking-Water Sources; World Health Organization: Geneva, Switzerland, 2006. [Google Scholar]
  8. Shahid, S.; Wang, X.J.; Moshiur Rahman, M.; Hasan, R.; Harun, S.B.; Shamsudin, S. Spatial assessment of groundwater over-exploitation in northwestern districts of Bangladesh. J. Geol. Soc. India 2015, 85, 463–470. [Google Scholar] [CrossRef]
  9. Dangar, S.; Asoka, A.; Mishra, V. Causes and implications of groundwater depletion in India: A review. J. Hydrol. 2021, 596, 126103. [Google Scholar] [CrossRef]
  10. Jain, M.; Fishman, R.; Mondal, P.; Galford, G.L.; Bhattarai, N.; Naeem, S.; Lall, U.; Balwinder-Singh; DeFries, R.S. Groundwater depletion will reduce cropping intensity in India. Sci. Adv. 2021, 7, eabd2849. [Google Scholar] [CrossRef]
  11. Jia, X.; O’Connor, D.; Hou, D.; Jin, Y.; Li, G.; Zheng, C.; Ok, Y.S.; Tsang, D.C.; Luo, J. Groundwater depletion and contamination: Spatial distribution of groundwater resources sustainability in China. Sci. Total Environ. 2019, 672, 551–562. [Google Scholar] [CrossRef] [PubMed]
  12. Islam, F.; Imteaz, M.A. The effectiveness of ARIMAX model for prediction of summer rainfall in northwest Western Australia. IOP Conf. Ser. Mater. Sci. Eng. 2021, 1067, 012037. [Google Scholar] [CrossRef]
  13. Hossain, M.Z.; Adhikary, S.K. Identifying groundwater recharge potential zones in Barind Tract of Bangladesh using geospatial technique. AIP Conf. Proc. 2023, 2713, 050001. [Google Scholar] [CrossRef]
  14. Hossain, M.Z.; Adhikary, S.K.; Nath, H.; Kafy, A.A.; Altuwaijri, H.A.; Rahman, M.T. Integrated Geospatial and Analytical Hierarchy Process Approach for Assessing Sustainable Management of Groundwater Recharge Potential in Barind Tract. Water 2024, 16, 2918. [Google Scholar] [CrossRef]
  15. Mancini, S.; Egidio, E.; De Luca, D.A.; Lasagna, M. Application and comparison of different statistical methods for the analysis of groundwater levels over time: Response to rainfall and resource evolution in the Piedmont Plain (NW Italy). Sci. Total Environ. 2022, 846, 157479. [Google Scholar] [CrossRef]
  16. Sekhar, P.H.; Kesavulu Poola, K.; Bhupathi, M. Modelling and prediction of coastal Andhra rainfall using ARIMA and ANN models. Int. J. Stat. Appl. Math. 2020, 5, 104–110. [Google Scholar]
  17. Elbeltagi, A.; Salam, R.; Pal, S.C.; Zerouali, B.; Shahid, S.; Mallick, J.; Islam, M.S.; Islam, A.R.M.T. Groundwater level estimation in northern region of Bangladesh using hybrid locally weighted linear regression and Gaussian process regression modeling. Theor. Appl. Climatol. 2022, 149, 131–151. [Google Scholar] [CrossRef]
  18. Sun, J.; Hu, L.; Li, D.; Sun, K.; Yang, Z. Data-driven models for accurate groundwater level prediction and their practical significance in groundwater management. J. Hydrol. Reg. Stud. 2022, 608, 127630. [Google Scholar] [CrossRef]
  19. Iqbal, N.; Khan, A.N.; Rizwan, A.; Ahmad, R.; Kim, B.W.; Kim, K.; Kim, D.H. Groundwater level prediction model using correlation and difference mechanisms based on boreholes data for sustainable hydraulic resource management. IEEE Access 2021, 9, 96092–96113. [Google Scholar] [CrossRef]
  20. Qadir, B.H.; Mohammed, M.A. Comparison between SARIMA and SARIMAX time series Models with application on Groundwater in Sulaymaniyah. Sci. J. Cihan Univ. Sulaimaniya 2021, 5, 30–48. [Google Scholar]
  21. Shahinuzzaman, M.; Haque, M.N.; Uddin, K.M.N.; Alibuddin, M. Identification of Aquifer Properties in the Eastern Part of Kushtia District, Bangladesh. J. Geosci. Environ. Prot. 2020, 8, 222. [Google Scholar] [CrossRef]
  22. Porte, P.; Isaac, R.K.; Mahilang, K.K.S.; Sonboier, K.; Minj, P. Groundwater level prediction using artificial neural network model. Int. J. Curr. Microbiol. Appl. Sci. 2018, 72, 2947–2954. [Google Scholar] [CrossRef]
  23. Ghazi, B.; Jeihouni, E.; Kalantari, Z. Predicting groundwater level fluctuations under climate change scenarios for Tasuj plain, Iran. Arab. J. Geosci. 2021, 14, 115. [Google Scholar] [CrossRef]
  24. Kassem, A.A.; Raheem, A.M.; Khidir, K.M. Daily streamflow prediction for khazir river basin using ARIMA and ANN models. ZANCO J. Pure Appl. Sci. 2020, 32, 30–39. [Google Scholar]
  25. Narvekar, M.; Fargose, P. Daily weather forecasting using artificial neural network. Int. J. Comput. Appl. 2015, 121, 0975–8887. [Google Scholar] [CrossRef]
  26. Liu, Q.; Zou, Y.; Liu, X.; Linge, N. A survey on rainfall forecasting using artificial neural network. Int. J. Embed. Syst. 2019, 11, 240–249. [Google Scholar] [CrossRef]
  27. Khedri, A.; Kalantari, N.; Vadiati, M. Comparison study of artificial intelligence method for short term groundwater level prediction in the northeast Gachsaran unconfined aquifer. Water Supply 2020, 20, 909–921. [Google Scholar] [CrossRef]
  28. Husna, N.E.A.; Bari, S.H.; Hussain, M.M.; Ur-rahman, M.T.; Rahman, M. Ground water level prediction using artificial neural network. Int. J. Hydrol. Sci. Technol. 2016, 6, 371–381. [Google Scholar] [CrossRef]
  29. Pham, Q.B.; Kumar, M.; Di Nunno, F.; Elbeltagi, A.; Granata, F.; Islam, A.R.M.T.; Talukdar, S.; Nguyen, X.C.; Ahmed, A.N.; Anh, D.T. Groundwater level prediction using machine learning algorithms in a drought-prone area. Neural Comput. Appl. 2022, 34, 10751–10773. [Google Scholar] [CrossRef]
  30. Palani, S.; Liong, S.Y.; Tkalich, P. An ANN application for water quality forecasting. Mar. Pollut. 2008, 56, 1586–1597. [Google Scholar] [CrossRef] [PubMed]
  31. Rabbi, F.; Tareq, S.U.; Islam, M.M.; Chowdhury, M.A.; Kashem, M.A. A multivariate time series approach for forecasting of electricity demand in bangladesh using arimax model. In Proceedings of the 2020 2nd International Conference on Sustainable Technologies for Industry 4.0, Dhaka, Bangladesh, 19–20 December 2020; pp. 1–5. [Google Scholar]
  32. Lemos, J.D.J.S.; Bezerra, F.N.R. ARIMAX Model to Forecast Grain Production under Rainfall Instabilities in Brazilian Semi-Arid Region. Glob. J. Hum. Soc. Sci. 2024, 24, 1–9. [Google Scholar]
  33. Khairi, S.M.; Aziz, I.A. Domestic water consumption forecasting with sociodemographic features using ARIMA and ARIMAX: A case study in Malaysia. Platf. A J. Sci. Technol. 2022, 5, 16–30. [Google Scholar] [CrossRef]
  34. Jalalkamali, A.; Moradi, M.; Moradi, N. Application of several artificial intelligence models and ARIMAX model for forecasting drought using the Standardized Precipitation Index. Int. J. Environ. Sci. Technol. 2015, 12, 1201–1210. [Google Scholar] [CrossRef]
  35. Khandelwal, I.; Adhikari, R.; Verma, G. Time series forecasting using hybrid ARIMA and ANN models based on DWT decomposition. Procedia Comput. Sci. 2015, 48, 173–179. [Google Scholar] [CrossRef]
  36. Chitsazan, M.; Rahmani, G.; Neyamadpour, A. Forecasting groundwater level by artificial neural networks as an alternative approach to groundwater modeling. J. Geol. Soc. India 2015, 85, 98–106. [Google Scholar] [CrossRef]
  37. Wang, W.; Van Gelder, P.H.; Vrijling, J.K.; Ma, J. Forecasting daily streamflow using hybrid ANN models. J. Hydrol. Reg. Stud. 2006, 324, 383–399. [Google Scholar] [CrossRef]
  38. Azad, A.S.; Sokkalingam, R.; Daud, H.; Adhikary, S.K.; Khurshid, H.; Mazlan, S.N.A.; Rabbani, M.B.A. Water level prediction through hybrid SARIMA and ANN models based on time series analysis: Red hills reservoir case study. Sustainability 2022, 14, 1843. [Google Scholar] [CrossRef]
  39. Islam, F.; Imteaz, M.A. Use of teleconnections to predict Western Australian seasonal rainfall using ARIMAX model. Hydrology 2020, 7, 52. [Google Scholar] [CrossRef]
  40. Pourmorad, S.; Kabolizade, M.; Dimuccio, L.A. Artificial Intelligence Advancements for Accurate Groundwater Level Modelling: An Updated Synthesis and Review. Appl. Sci. 2024, 14, 7358. [Google Scholar] [CrossRef]
  41. Li, W.; Finsa, M.M.; Laskey, K.B.; Houser, P.; Douglas-Bate, R. Groundwater level prediction with machine learning to support sustainable irrigation in water scarcity regions. Water 2023, 15, 3473. [Google Scholar] [CrossRef]
  42. Haji-Aghajany, S.; Amerian, Y.; Amiri-Simkooei, A. Impact of climate change parameters on groundwater level: Implications for two subsidence regions in Iran using geodetic observations and artificial neural networks (ANN). Remote Sens. 2023, 15, 1555. [Google Scholar] [CrossRef]
Figure 1. Location of the study area (Kushtia District) in Bangladesh showing the GWL and RF stations.
Figure 1. Location of the study area (Kushtia District) in Bangladesh showing the GWL and RF stations.
Geographies 05 00001 g001
Figure 2. GWL data for the stations (a) Bheramara, (b) Daulatpur, (c) Kushtia Sadar, (d) Kumarkhali (e) Mirpur, and RF data for (f) station Mirpur.
Figure 2. GWL data for the stations (a) Bheramara, (b) Daulatpur, (c) Kushtia Sadar, (d) Kumarkhali (e) Mirpur, and RF data for (f) station Mirpur.
Geographies 05 00001 g002
Figure 3. (a) ACF and (b) PACF plots for GWL fluctuations; Station ID: KG-1; location: Bheramara.
Figure 3. (a) ACF and (b) PACF plots for GWL fluctuations; Station ID: KG-1; location: Bheramara.
Geographies 05 00001 g003
Figure 4. (a) Scatterplot and (b) actual and predicted GWL plot based on the ANN 8-9-1 model for Station ID: KG-5; location: Mirpur.
Figure 4. (a) Scatterplot and (b) actual and predicted GWL plot based on the ANN 8-9-1 model for Station ID: KG-5; location: Mirpur.
Geographies 05 00001 g004
Figure 5. (a) ACF, (b) PACF, (c) QQ plot, (d) scatterplot, and (e) actual vs. model prediction and residual plots based on ARIMA (2,0,1), model Station ID: KG-5; location: Mirpur.
Figure 5. (a) ACF, (b) PACF, (c) QQ plot, (d) scatterplot, and (e) actual vs. model prediction and residual plots based on ARIMA (2,0,1), model Station ID: KG-5; location: Mirpur.
Geographies 05 00001 g005
Figure 6. Existing and predicted GWL data based on (a) ARIMAX (3,0,2), Station ID: KG-5, location: Mirpur, and (b) ANN 6-8-1, Station ID: KG-1, location: Bheramara.
Figure 6. Existing and predicted GWL data based on (a) ARIMAX (3,0,2), Station ID: KG-5, location: Mirpur, and (b) ANN 6-8-1, Station ID: KG-1, location: Bheramara.
Geographies 05 00001 g006
Figure 7. (a) Scatterplot for ARIMAX (3,0,2); location: Mirpur. (b) Scatterplot for ANN 6-8-1; location: Bheramara.
Figure 7. (a) Scatterplot for ARIMAX (3,0,2); location: Mirpur. (b) Scatterplot for ANN 6-8-1; location: Bheramara.
Geographies 05 00001 g007
Table 1. GWL and RF station details.
Table 1. GWL and RF station details.
SL
No
Station
ID
Station
Type
Location of Station
(Sub-District Name)
Latitude
(Degree)
Longitude
(Degree)
1KG-1GWLBheramara24.0988.96
2KG-2GWLDaulatpur23.9888.83
3KG-3GWLKushtia Sadar23.8389.10
4KG-4GWLKumarkhali23.8489.20
5KG-5GWLMirpur23.9389.02
6KR-1RFMirpur24.0588.99
Table 2. Performances of the best-selected ANN (multivariate) models.
Table 2. Performances of the best-selected ANN (multivariate) models.
Station IDLocationModel
Architecture
Model Performance
RMSENSESSE
KG-1BheramaraANN 6-8-10.1540.9889.894
KG-2DaulatpurANN 7-8-10.1680.97911.799
KG-3Kushtia SadarANN 10-4-10.2310.96526.910
KG-4KumarkhaliANN 6-7-10.2550.98622.273
KG-5MirpurANN 8-9-10.1800.98413.434
Table 3. Performances of the best-selected ANN (univariate) models.
Table 3. Performances of the best-selected ANN (univariate) models.
Station IDLocationModel
Architecture
Model Performance
RMSENSESSE
KG-1BheramaraANN 3-7-10.1610.98710.809
KG-2DaulatpurANN 2-3-10.1710.97912.171
KG-3Kushtia SadarANN 4-9-10.2580.95726.802
KG-4KumarkhaliANN 2-4-10.2540.98627.725
KG-5MirpurANN 5-10-10.1810.98413.595
Table 4. Performance overview of the selected ANN (multivariate) models for Station ID: KG-5.
Table 4. Performance overview of the selected ANN (multivariate) models for Station ID: KG-5.
ModelTrainingValidationTest
MSENSEMSENSEMSENSE
ANN 8-2-10.0300.9840.0660.9640.0460.978
ANN 8-3-10.0320.9820.0610.9670.0450.978
ANN 8-4-10.0460.9750.0690.9630.0760.963
ANN 8-5-10.0360.9800.0740.9600.0370.982
ANN 8-6-10.0310.9830.0780.9580.0420.980
ANN 8-7-10.0300.9830.2010.8920.0400.981
ANN 8-8-10.0350.9810.0790.9580.0420.980
ANN 8-9-10.0320.9830.0830.9550.0320.984
ANN 8-10-10.0270.9850.0710.9620.0390.981
Table 5. Best ARIMAX model results.
Table 5. Best ARIMAX model results.
Station IDLocationModel ArchitectureModel Performance (SSE)
KG-1BheramaraARIMAX (3,0,3)15.361
KG-2DaulatpurARIMAX (3,0,2)18.721
KG-3Kushtia SadarARIMAX (1,0,3)25.449
KG-4KumarkhaliARIMAX (2,0,0)63.680
KG-5MirpurARIMAX (3,0,2)15.143
Table 6. Best ARIMA model results.
Table 6. Best ARIMA model results.
Station IDLocationModel ArchitectureModel Performance (SSE)
KG-1BheramaraARIMA (2,0,1)17.217
KG-2DaulatpurARIMA (2,0,1)26.880
KG-3Kushtia SadarARIMA (2,0,3)28.207
KG-4KumarkhaliARIMA (3,0,1)64.582
KG-5MirpurARIMA (2,0,1)16.585
Table 7. Performances of the selected ARIMA (p, d, q) models for Station ID: KG-5.
Table 7. Performances of the selected ARIMA (p, d, q) models for Station ID: KG-5.
ModelSSEMSERMSEAICBIC
ARIMA (0,2,1)20.6880.0500.224−59.599−47.521
ARIMA (1,2,2)20.6880.0500.224−55.600−35.471
ARIMA (1,2,3)20.6800.0500.223−53.750−29.594
ARIMA (2,0,0)21.2200.0510.226−47.077−30.974
ARIMA (2,0,1)16.5850.0400.200−147.100−126.971
ARIMA (2,0,2)16.5190.0400.200−146.764−122.609
ARIMA (3,0,1)16.5370.0400.200−146.317−122.162
ARIMA (3,2,1)20.6270.0500.223−54.820−30.665
ARIMA (3,2,2)20.5630.0500.223−54.092−25.911
ARIMA (3,2,3)20.0210.0480.220−63.151−30.944
Table 8. Model parameters for ARIMA (2,0,1) (Station ID: KG-5).
Table 8. Model parameters for ARIMA (2,0,1) (Station ID: KG-5).
ParametersValueStandard ErrorT Statisticp Value
Constant0.1310.00914.3521.02 × 10−46
AR{1}1.9690.008244.5400
AR{2}−0.9840.007−123.0570
MA{1}−0.9310.020−45.9210
Variance0.0400.00219.1471.02 × 10−81
Table 9. Predicted highest, lowest, and average GWL values.
Table 9. Predicted highest, lowest, and average GWL values.
Station IDLocationModel/DataHighest
(m.PWD)
Lowest
(m.PWD)
Average
(m.PWD)
KG-1BheramaraExisting
(actual data)
12.6105.7308.148
KG-1BheramaraANN 6-8-1
(multivariate)
10.7975.8757.742
KG-5MirpurARIMAX
(3,0,2)
11.6946.6228.951
Table 10. Evaluation metrics and comparative performance improvements of the predictive models.
Table 10. Evaluation metrics and comparative performance improvements of the predictive models.
Performance Comparison of the ModelsRemarks
Station IDResults
(SSE)
Model
Architecture
ANN
(Multivariate)
(%)
ANN
(Univariate)
(%)
ARIMAX
(%)
ARIMA
(%)
KG-19.894ANN 6-8-1
(multivariate)
0−8.459−34.658−40.340Reference model ANN (multivariate):
Performance Enhancement: -;
Performance Degradation: ANN (univariate), ARIMAX, ARIMA
KG-110.809ANN 3-7-1
(univariate)
9.2410−28.620−34.826Reference model ANN (univariate):
Performance Enhancement: ANN (multivariate);
Performance Degradation: ARIMAX, ARIMA
KG-515.143ARIMAX
(3,0,2)
53.04340.0960−8.694Reference model ARIMAX:
Performance Enhancement: ANN (multivariate), ANN (univariate);
Performance Degradation: ARIMA
KG-116.585ARIMA
(2,0,1)
67.61653.4369.5220Reference model ARIMA:
Performance Enhancement: ANN (multivariate), ANN (univariate), ARIMAX;
Performance Degradation: -
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.

Share and Cite

MDPI and ACS Style

Hoque, M.A.; Apon, A.A.; Hassan, M.A.; Adhikary, S.K.; Islam, M.A. Enhanced Forecasting of Groundwater Level Incorporating an Exogenous Variable: Evaluating Conventional Multivariate Time Series and Artificial Neural Network Models. Geographies 2025, 5, 1. https://doi.org/10.3390/geographies5010001

AMA Style

Hoque MA, Apon AA, Hassan MA, Adhikary SK, Islam MA. Enhanced Forecasting of Groundwater Level Incorporating an Exogenous Variable: Evaluating Conventional Multivariate Time Series and Artificial Neural Network Models. Geographies. 2025; 5(1):1. https://doi.org/10.3390/geographies5010001

Chicago/Turabian Style

Hoque, Md Abrarul, Asib Ahmmed Apon, Md Arafat Hassan, Sajal Kumar Adhikary, and Md Ariful Islam. 2025. "Enhanced Forecasting of Groundwater Level Incorporating an Exogenous Variable: Evaluating Conventional Multivariate Time Series and Artificial Neural Network Models" Geographies 5, no. 1: 1. https://doi.org/10.3390/geographies5010001

APA Style

Hoque, M. A., Apon, A. A., Hassan, M. A., Adhikary, S. K., & Islam, M. A. (2025). Enhanced Forecasting of Groundwater Level Incorporating an Exogenous Variable: Evaluating Conventional Multivariate Time Series and Artificial Neural Network Models. Geographies, 5(1), 1. https://doi.org/10.3390/geographies5010001

Article Metrics

Back to TopTop