Machine Learning-Based Groundwater Level Simulation for Industrial Water Management: A Comparative Study of ARIMA, BP-ANN and Random Forest
DOI:
https://doi.org/10.4108/eetsis.14473Keywords:
Data-driven model, Groundwater level dynamics, Random Forest (RF), Backpropagation Artificial Neural Network (BP-ANN), Autoregressive Integrated Moving Average (ARIMA), Variable Projection Importance (VPI), Industrial water consumption, Groundwater simulation, Dingxi CityAbstract
Timely and accurate simulation of regional groundwater level dynamics is critical to guarantee stable water supply for agricultural production and manufacturing sectors in arid semi-arid loess regions. This study adopts Variable Projection Importance (VPI) to screen feature variables controlling groundwater fluctuation, and constructs three data-driven models: Autoregressive Integrated Moving Average (ARIMA), Backpropagation Artificial Neural Network (BP-ANN) and Random Forest (RF). The three models are applied to simulate monthly groundwater dynamics across seven county-level administrative divisions of water-deficient Dingxi City with scattered agricultural processing and light manufacturing enterprises. Training and independent validation results show the overall simulation accuracy ranking follows RF > BP-ANN > ARIMA. For one-year groundwater level prediction, RF provides higher fitting precision, reliability and better regional adaptability, which can be prioritized for groundwater dynamic simulation to support industrial and agricultural water allocation in Dingxi. In terms of feature contribution, lagged historical groundwater level (L1) contributes over 80% to model outputs for most districts, followed by air humidity and precipitation; human interference factors including total groundwater exploitation and manufacturing water intake occupy low weight. Taking county-level administrative units (basic industrial water management units) as independent modeling objects, this paper compares the performance of three mainstream data-driven models, quantifies the contribution of manufacturing groundwater extraction to aquifer fluctuation, and provides a complete, operable modeling process for groundwater simulation in similar arid areas with small-scale manufacturing clusters. The research results can offer quantitative support for balancing groundwater exploitation limits among manufacturing, farming and ecological protection in Dingxi.
References
[1] Feng S, Huo Z, Kang S, et al. Groundwater simulation using a numerical model under different water resources management scenarios in an arid region of china[J]. Environmental Earth Sciences, 2011, 62(5): 961-971.
[2] Hu L, Jiao J J. Calibration of a large-scale groundwater flow model using GRACE data: a case study in the Qaidam Basin, China[J]. Hydrogeology Journal, 2015, 23(7): 1305-1317.
[3] Maxwell R M, Condon L E, Kollet S J. A high-resolution simulation of groundwater and surface water over most of the continental US with the integrated hydrologic model ParFlow v3[J]. Geoscientific Model Development, 2015, 8(3):923-937.
[4] Rahman A T M S, Hosono T, Quilty J M,et al. Multiscale groundwater level forecasting: Coupling new machine learning approaches with wavelet transforms[J]. Advances in Water Resources, 2020, 141:103595.
[5] Sahu R K, Müller J, Park J, et al. Impact of input feature selection on groundwater level prediction from a multi-layer perceptron neural network[J]. Frontiers in Water, 2020, 2: 573034.
[6] Sahoo S, Jha M K. Groundwater-level prediction using multiple linear regression and artificial neural network techniques: a comparative assessment[J]. Hydrogeology Journal, 2013, 21(8): 1865-1887. (Supplemented in-text citation in Introduction data-driven model paragraph)
[7] Adhikary S K, Rahman M M, Gupta A D, et al. A stochastic modelling technique for predicting groundwater table fluctuations with time series analysis[J]. International Journal of Applied Science and Engineering Research, 2012, 1(2): 238-249.
[8] Mirzavand M, Ghazavi R. A stochastic modelling technique for groundwater level forecasting in an arid environment using time series methods[J]. Water Resources Management, 2014, 29(4): 1315-1328.
[9] Luo J W, Deng Y E. Prediction of dynamic changes of groundwater level in a plain area using ARIMA model[J]. Groundwater, 2020, 42(2): 43-45.
[10] Wang Y, Lu W X, Bian J M, et al. Application and comparison of three groundwater level dynamic prediction models in western Jilin[J]. Journal of Jilin University (Earth Science Edition), 2015, 45(3): 886-891.
[11] Li W C, Hu T R, Wei Y X. Application of BP-ANN model in groundwater dynamic prediction[J]. Journal of Northeast Agricultural University, 2009, 40(9): 44-47. (Supplemented in-text citation in BP-ANN model subsection)
[12] Fan Y, Hu L, Wang H, et al. Machine learning methods for improved understanding of a pumping test in heterogeneous aquifers[J]. Water, 2020, 12(5): 1342. (Supplemented in-text citation in BP-ANN activation function description paragraph)
[13] Jones A L, Smart P L. Spatial and temporal changes in the structure of groundwater nitrate concentration time series (1935-1999) as demonstrated by autoregressive modelling[J]. Journal of Hydrology, 2005, 310(1-4):201-215.
[14] Zhang H, Liu Z. Random forest-based groundwater level prediction considering multi-source meteorological and human factors[J]. Journal of Hydrologic Engineering, 2021, 26(8): 04021046.
[15] Gao Y, Yang J. Long short-term memory neural network for groundwater level forecasting in arid inland basins[J]. Journal of Hydrology: Regional Studies, 2022, 42: 101123.
[16] Chen L, Wang Q. Comparison of LSTM and RF models for groundwater dynamic simulation in semi-arid loess regions[J]. Hydrological Processes, 2023, 37(11): e14689.
[17] Zhao X. Feature screening and machine learning simulation of regional groundwater level based on VPI[J]. Journal of Arid Land Resources and Environment, 2024, 38(2): 112-119.
[18] Liu Y, Zhang Y. Comparison of BP neural network and random forest for groundwater level prediction in loess hilly area[J]. Water Saving Irrigation, 2020, (6): 89-95.
[19] Wu S, Li J. Variable projection importance-based dominant factor identification of groundwater dynamics in semi-arid areas[J]. Hydrogeology & Engineering Geology, 2022, 49(3): 78-86.
[20] Yang C, Zhou M. Influencing factor contribution analysis of groundwater level variation based on random forest algorithm[J]. Journal of Groundwater Science and Engineering, 2023, 11(4): 512-520.
[21] Zhang L, Wang H. Performance comparison of ARIMA, BP-ANN and RF models for monthly groundwater level prediction[J]. Water Resources and Power, 2021, 39(7): 16-20.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Jia Cuixia, Lu Bin, Pengjun Bai
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
