{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T01:20:57Z","timestamp":1778808057112,"version":"3.51.4"},"reference-count":50,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2025,12,21]],"date-time":"2025-12-21T00:00:00Z","timestamp":1766275200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Xinjiang Uygur Autonomous Region Major Science and Technology Special Project","award":["2024A03007-3"],"award-info":[{"award-number":["2024A03007-3"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJGI"],"abstract":"<jats:p>Accurate forecasting of groundwater level dynamics poses a critical challenge for sustainable water management in arid regions. However, the strong spatiotemporal heterogeneity inherent in groundwater systems and their complex interactions between natural processes and human activities often limit the effectiveness of conventional prediction methods. To address this, a hybrid CNN-LSTM deep learning model is constructed. This model is designed to extract multivariate coupled features and capture temporal dependencies from multi-variable time series data, while simultaneously simulating the nonlinear and delayed responses of aquifers to groundwater abstraction. Specifically, the convolutional neural network (CNN) component extracts the multivariate coupled features of hydro-meteorological driving factors, and the long short-term memory (LSTM) network component models the temporal dependencies in groundwater level fluctuations. This integrated architecture comprehensively represents the combined effects of natural recharge\u2013discharge processes and anthropogenic pumping on the groundwater system. Utilizing monitoring data from 2021 to 2024, the model was trained and tested using a rolling time-series validation strategy. Its performance was benchmarked against traditional models, including the autoregressive integrated moving average (ARIMA) model, recurrent neural network (RNN), and standalone LSTM. The results show that the CNN-LSTM model delivers superior performance across diverse hydrogeological conditions: at the upstream well AJC-7, which is dominated by natural recharge and discharge, the Nash\u2013Sutcliffe efficiency (NSE) coefficient reached 0.922; at the downstream well AJC-21, which is subject to intensive pumping, the model maintained a robust NSE of 0.787, significantly outperforming the benchmark models. Further sensitivity analysis reveals an asymmetric response of the model\u2019s predictions to uncertainties in pumping data, highlighting the role of key hydrogeological processes such as delayed drainage from the vadose zone. This study not only confirms the strong applicability of the hybrid deep learning model for groundwater level prediction in data-scarce arid regions but also provides a novel analytical pathway and mechanistic insight into the nonlinear behavior of aquifer systems under significant human influence.<\/jats:p>","DOI":"10.3390\/ijgi15010006","type":"journal-article","created":{"date-parts":[[2025,12,23]],"date-time":"2025-12-23T14:27:47Z","timestamp":1766500067000},"page":"6","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Application of a Hybrid CNN-LSTM Model for Groundwater Level Forecasting in Arid Regions: A Case Study from the Tailan River Basin"],"prefix":"10.3390","volume":"15","author":[{"given":"Shuting","family":"Hu","sequence":"first","affiliation":[{"name":"College of Hydraulic and Civil Engineering, Xinjiang Agricultural University, Urumqi 830052, China"},{"name":"Xinjiang Key Laboratory of Hydraulic Engineering Security and Water Disasters Prevention, Xinjiang Agricultural University, Urumqi 830052, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-6892-1286","authenticated-orcid":false,"given":"Mingliang","family":"Du","sequence":"additional","affiliation":[{"name":"College of Hydraulic and Civil Engineering, Xinjiang Agricultural University, Urumqi 830052, China"},{"name":"Xinjiang Key Laboratory of Hydraulic Engineering Security and Water Disasters Prevention, Xinjiang Agricultural University, Urumqi 830052, China"},{"name":"Wensu Future Irrigation District Field Station, Wensu, Aksu 843100, China"}]},{"given":"Jiayun","family":"Yang","sequence":"additional","affiliation":[{"name":"Xinjiang Key Laboratory of Hydraulic Engineering Security and Water Disasters Prevention, Xinjiang Agricultural University, Urumqi 830052, China"}]},{"given":"Yankun","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Hydraulic and Civil Engineering, Xinjiang Agricultural University, Urumqi 830052, China"},{"name":"Xinjiang Key Laboratory of Hydraulic Engineering Security and Water Disasters Prevention, Xinjiang Agricultural University, Urumqi 830052, China"}]},{"given":"Ziyun","family":"Tuo","sequence":"additional","affiliation":[{"name":"College of Hydraulic and Civil Engineering, Xinjiang Agricultural University, Urumqi 830052, China"},{"name":"Xinjiang Key Laboratory of Hydraulic Engineering Security and Water Disasters Prevention, Xinjiang Agricultural University, Urumqi 830052, China"}]},{"given":"Xiaofei","family":"Ma","sequence":"additional","affiliation":[{"name":"College of Hydraulic and Civil Engineering, Xinjiang Agricultural University, Urumqi 830052, China"},{"name":"Xinjiang Key Laboratory of Hydraulic Engineering Security and Water Disasters Prevention, Xinjiang Agricultural University, Urumqi 830052, China"}]}],"member":"1968","published-online":{"date-parts":[[2025,12,21]]},"reference":[{"key":"ref_1","first-page":"1055","article-title":"Pulmonary Edema and Pleural Effusion Detection Using EfficientNet-V1-B4 Architecture and AdamW Optimizer from Chest X-Rays Images","volume":"80","author":"AbuKaraki","year":"2024","journal-title":"CMC-Comput. Mater. Contin."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Afzaal, H., Farooque, A.A., Abbas, F., Acharya, B., and Esau, T. (2019). Groundwater estimation from major physical hydrology components using artificial neural networks and deep learning. Water, 12.","DOI":"10.3390\/w12010005"},{"key":"ref_3","first-page":"1","article-title":"CNN-Bi LSTM neural network for simulating groundwater level","volume":"8","author":"Ali","year":"2022","journal-title":"Environ. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"112896","DOI":"10.1016\/j.eswa.2019.112896","article-title":"Forecasting across time series databases using recurrent neural networks on groups of similar series: A clustering approach","volume":"140","author":"Bandara","year":"2020","journal-title":"Expert Syst. Appl."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"89","DOI":"10.5194\/hess-25-89-2021","article-title":"Using multiple methods to investigate the effects of land-use changes on groundwater recharge in a semi-arid area","volume":"25","author":"Barua","year":"2021","journal-title":"Hydrol. Earth Syst. Sci."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Bowes, B.D., Sadler, J.M., Morsy, M.M., Behl, M., and Goodall, J.L. (2019). Forecasting groundwater table in a flood prone coastal city with long short-term memory and recurrent neural networks. Water, 11.","DOI":"10.1002\/essoar.10500507.1"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1111\/gwat.13336","article-title":"Estimating Groundwater Pumping for Irrigation: A Method Comparison","volume":"62","author":"Brookfield","year":"2024","journal-title":"Groundwater"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"e35773","DOI":"10.1016\/j.heliyon.2024.e35773","article-title":"The influence of withdrawal-recharging pattern on the deformation characteristics of sand in confined aquifer","volume":"10","author":"Chang","year":"2024","journal-title":"Heliyon"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e623","DOI":"10.7717\/peerj-cs.623","article-title":"The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation","volume":"7","author":"Chicco","year":"2021","journal-title":"Peerj Comput. Sci."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Chu, H., Bian, J., Lang, Q., Sun, X., and Wang, Z. (2022). Daily Groundwater Level Prediction and Uncertainty Using LSTM Coupled with PMI and Bootstrap Incorporating Teleconnection Patterns Information. Sustainability, 14.","DOI":"10.3390\/su141811598"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1038\/s41586-019-1594-4","article-title":"Environmental flow limits to global groundwater pumping","volume":"574","author":"Gleeson","year":"2019","journal-title":"Nature"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Di Salvo, C. (2022). Improving Results of Existing Groundwater Numerical Models Using Machine Learning Techniques: A Review. Water, 14.","DOI":"10.3390\/w14152307"},{"key":"ref_13","first-page":"217","article-title":"A New Sequential Image Prediction Method Based on LSTM and DCGAN","volume":"64","author":"Fang","year":"2020","journal-title":"Comput. Mater. Contin."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2451","DOI":"10.1162\/089976600300015015","article-title":"Learning to forget: Continual prediction with LSTM","volume":"12","author":"Gers","year":"2000","journal-title":"Neural Comput."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Graves, A., and Graves, A. (2012). Long short-term memory. Supervised Sequence Labelling with Recurrent Neural Networks, Springer.","DOI":"10.1007\/978-3-642-24797-2"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Huang, J., Ge, Y., and Li, S. (2022). Mixed-Unit-Model-Based and Quantitative Studies on Groundwater Recharging and Discharging between Aquifers of Aksu River. Sustainability, 14.","DOI":"10.3390\/su14116936"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"093111","DOI":"10.1063\/5.0055371","article-title":"Initializing LSTM internal states via manifold learning","volume":"31","author":"Kemeth","year":"2021","journal-title":"Chaos"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1691","DOI":"10.1080\/02626667.2024.2393421","article-title":"Prediction of groundwater level using artificial neural network as an alternative approach: A comparison assessment with numerical groundwater flow model","volume":"69","author":"Kerebih","year":"2024","journal-title":"Hydrol. Sci. J."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Khan, J., Lee, E., Balobaid, A.S., and Kim, K. (2023). A comprehensive review of conventional, machine leaning, and deep learning models for groundwater level (GWL) forecasting. Appl. Sci., 13.","DOI":"10.3390\/app13042743"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"119789","DOI":"10.1016\/j.jenvman.2023.119789","article-title":"Watershed groundwater level multistep ahead forecasts by fusing convolutional-based autoencoder and LSTM models","volume":"351","author":"Kow","year":"2024","journal-title":"J. Environ. Manag."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/j.patrec.2014.01.008","article-title":"A review of unsupervised feature learning and deep learning for time-series modeling","volume":"42","author":"Karlsson","year":"2014","journal-title":"Pattern Recognit. Lett."},{"key":"ref_23","unstructured":"Le, Q.V., Jaitly, N., and Hinton, G.E. (2015). A simple way to initialize recurrent networks of rectified linear units. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1134\/S0097807823601346","article-title":"Short-Term Prediction of Groundwater Level Based on Spatiotemporal Correlation","volume":"51","author":"Liu","year":"2024","journal-title":"Water Resour."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"688","DOI":"10.1111\/gwat.12557","article-title":"Prospective interest of deep learning for hydrological inference","volume":"55","year":"2017","journal-title":"Groundwater"},{"key":"ref_26","first-page":"e2024JH000322","article-title":"Modeling groundwater levels in California\u2019s Central Valley by hierarchical Gaussian process and neural network regression","volume":"1","author":"Pradhan","year":"2024","journal-title":"J. Geophys. Res. Mach. Learn. Comput."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"757","DOI":"10.1016\/j.jksuci.2023.01.014","article-title":"A comprehensive review on ensemble deep learning: Opportunities and challenges","volume":"35","author":"Mohammed","year":"2023","journal-title":"J. King Saud Univ.-Comput. Inf. Sci."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"773","DOI":"10.1080\/02626667.2022.2046755","article-title":"Uncertainty assessment of LSTM based groundwater level predictions","volume":"67","author":"Nourani","year":"2022","journal-title":"Hydrol. Sci. J."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"60090","DOI":"10.1109\/ACCESS.2020.2982433","article-title":"Water level prediction model based on GRU and CNN","volume":"8","author":"Pan","year":"2020","journal-title":"IEEE Access"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"125033","DOI":"10.1016\/j.jhydrol.2020.125033","article-title":"Spatial prediction of groundwater potential mapping based on convolutional neural network (CNN) and support vector regression (SVR)","volume":"588","author":"Panahi","year":"2020","journal-title":"J. Hydrol."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"127294","DOI":"10.1016\/j.jhydrol.2021.127294","article-title":"Rescue of groundwater level time series: How to visually identify and treat errors","volume":"605","author":"Retike","year":"2022","journal-title":"J. Hydrol."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"356","DOI":"10.1016\/j.jhydrol.2017.08.006","article-title":"Wavelet coupled MARS and M5 Model Tree approaches for groundwater level forecasting","volume":"553","author":"Naganna","year":"2017","journal-title":"J. Hydrol."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Roy, D.K., Biswas, S.K., Mattar, M.A., El-Shafei, A.A., Murad, K.F.I., Saha, K.K., Datta, B., and Dewidar, A.Z. (2021). Groundwater level prediction using a multiple objective genetic algorithm-grey relational analysis based weighted ensemble of ANFIS models. Water, 13.","DOI":"10.3390\/w13213130"},{"key":"ref_34","first-page":"102129","article-title":"Permanent aquifer storage loss from long-term groundwater withdrawal: A case study of subsidence in Bandung (Indonesia)","volume":"57","author":"Rygus","year":"2025","journal-title":"J. Hydrol.-Reg. Stud."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"119231","DOI":"10.1016\/j.enconman.2024.119231","article-title":"Robust parameter estimation of proton exchange membrane fuel cell using Huber loss statistical function","volume":"323","author":"Saad","year":"2025","journal-title":"Energy Convers. Manag."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Salehi, A.W., Khan, S., Gupta, G., Alabduallah, B.I., Almjally, A., Alsolai, H., Siddiqui, T., and Mellit, A. (2023). A study of CNN and transfer learning in medical imaging: Advantages, challenges, future scope. Sustainability, 15.","DOI":"10.3390\/su15075930"},{"key":"ref_37","first-page":"101577","article-title":"Enhancing groundwater level prediction accuracy at a daily scale through combined machine learning and physics-based modeling","volume":"50","author":"Sun","year":"2023","journal-title":"J. Hydrol.-Reg. Stud."},{"key":"ref_38","first-page":"38204","article-title":"Sequencer: Deep lstm for image classification","volume":"35","author":"Tatsunami","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"130145","DOI":"10.1016\/j.jhydrol.2023.130145","article-title":"Groundwater level monitoring network design with machine learning methods","volume":"625","author":"Teimoori","year":"2023","journal-title":"J. Hydrol."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"11397","DOI":"10.1007\/s13762-022-04356-9","article-title":"Groundwater quality prediction based on LSTM RNN: An Iranian experience","volume":"19","author":"Valadkhan","year":"2022","journal-title":"Int. J. Environ. Sci. Technol."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"125776","DOI":"10.1016\/j.jhydrol.2020.125776","article-title":"Reconstruction of missing groundwater level data by using Long Short-Term Memory (LSTM) deep neural network","volume":"597","author":"Vu","year":"2021","journal-title":"J. Hydrol."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"1671","DOI":"10.5194\/hess-25-1671-2021","article-title":"Groundwater level forecasting with artificial neural networks: A comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX)","volume":"25","author":"Wunsch","year":"2021","journal-title":"Hydrol. Earth Syst. Sci."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"131775","DOI":"10.1016\/j.jhydrol.2024.131775","article-title":"Global sensitivity analysis of water level response to harmonic aquifer disturbances through a Monte-Carlo based surrogate model with random forest algorithm","volume":"641","author":"Xing","year":"2024","journal-title":"J. Hydrol."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"127678","DOI":"10.1016\/j.jhydrol.2022.127678","article-title":"A general numerical model for water level response to harmonic disturbances in aquifers considering wellbore effects","volume":"609","author":"Xing","year":"2022","journal-title":"J. Hydrol."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"132188","DOI":"10.1016\/j.jhydrol.2024.132188","article-title":"Application of seismically derived tilt signals to characterize groundwater flow regimes: An example from a constant-rate pumping test in Taiwan","volume":"645","author":"Yang","year":"2024","journal-title":"J. Hydrol."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Yang, J., Rajanayaka, C., Daughney, C.J., Booker, D., Morris, R., and Thompson, M. (2023). Metamodelling of Naturalised Groundwater Levels at a Regional Level in New Zealand. Sustainability, 15.","DOI":"10.3390\/su151813393"},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Yang, X., and Zhang, Z. (2022). A CNN-LSTM model based on a meta-learning algorithm to predict groundwater level in the middle and lower reaches of the Heihe River, China. Water, 14.","DOI":"10.3390\/w14152377"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"918","DOI":"10.1016\/j.jhydrol.2018.04.065","article-title":"Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas","volume":"561","author":"Zhang","year":"2018","journal-title":"J. Hydrol."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"128","DOI":"10.1016\/j.jes.2024.03.052","article-title":"Spatio-temporal prediction of groundwater vulnerability based on CNN-LSTM model with self-attention mechanism: A case study in Hetao Plain, northern China","volume":"153","author":"Zhao","year":"2025","journal-title":"J. Environ. Sci."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"6486","DOI":"10.1109\/TPAMI.2024.3382294","article-title":"Towards Understanding Convergence and Generalization of AdamW","volume":"46","author":"Zhou","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."}],"container-title":["ISPRS International Journal of Geo-Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2220-9964\/15\/1\/6\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,23]],"date-time":"2025-12-23T14:49:11Z","timestamp":1766501351000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2220-9964\/15\/1\/6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,21]]},"references-count":50,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["ijgi15010006"],"URL":"https:\/\/doi.org\/10.3390\/ijgi15010006","relation":{},"ISSN":["2220-9964"],"issn-type":[{"value":"2220-9964","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,21]]}}}