Reservoir water level prediction at Hatia Dam, India using LSTM and RNN models
Sneh Kumar1, Vivekanand Singh1, Thendiyath Roshni1 and Jayprakash Kumar1
Corresponding Author : Sneh Kumar
Recieved : 17-January-2025; Revised : 16-April-2026; Accepted : 23-April-2026
Abstract
The water supply of Ranchi has been significantly affected by the declining capacity of a reservoir over time, primarily due to various infrastructural developments that have encroached upon its drainage area and reduced water inflow. This study aims to predict the reservoir water level (RWL) at Hatia Dam, a crucial water source for Ranchi, Jharkhand, India. For RWL prediction, machine learning (ML)-based models, including recurrent neural network (RNN) and long short-term memory (LSTM) models, were employed with lead times ranging from 10 to 90 days. The performance of these models was evaluated using several statistical metrics. The results indicate that the RNN model outperformed the LSTM model for a 10-day lead time, achieving a correlation coefficient (R) of 0.982, Nash–Sutcliffe efficiency (NSE) of 0.964, root mean square error (RMSE) of 0.414, and mean absolute error (MAE) of 0.227. Overall, both RNN and LSTM models demonstrated very good performance for lead times up to 60 days and satisfactory performance for a 90-day lead time. These findings provide valuable insights for strategic water resource planning and future RWL prediction.
Keywords
Reservoir water level prediction, Recurrent neural network (RNN), Long short-term memory (LSTM), Machine learning, Hatia dam, Water resource management.
Cite this article
Kumar S, Singh V, Roshni T, Kumar J. Reservoir water level prediction at Hatia Dam, India using LSTM and RNN models. International Journal of Advanced Technology and Engineering Exploration. 2026;13(137):474-489. DOI : 10.19101/IJATEE.2025.121220078
[1] Li H, Zhang L, Yao Y, Zhang Y. Prediction of reservoir water levels via an improved attention mechanism based on CNN-LSTM. Applied Intelligence. 2025; 55(7):1-20.
[2] Wang S, Xu O. Confidence interval forecasting model of small watershed flood based on compound recurrent neural networks and bayesian. Plos One. 2025; 20(4):1-20.
[3] Tandon K, Sen S. A probabilistic integration of LSTM and gaussian process regression for uncertainty-aware reservoir water level predictions. Hydrological Sciences Journal. 2025; 70(1):144-61.
[4] Han J, Bae JH. Developing an hourly water level prediction model for small-and medium-sized agricultural reservoirs using AutoML: case study of Baekhak reservoir, South Korea. Agriculture. 2024; 15(1):1-21.
[5] Raza MA, Karim A, Alqarni M, Al-khasawneh MA, Jumani TA, Aman M, et al. An intelligent long short-term memory-based machine learning model for the potential assessment of global hydropower capacity in sustainable energy transition and security. Energies. 2025; 18(13):1-27.
[6] Lachure SS, Tiwari A. Advanced deep learning approaches for superior temporal analysis and forecasting of water level discharge for the Bamni River. International Journal of River Basin Management. 2025: 1-19.
[7] Sharma RK, Kumar S, Padmalal D, Roy A. Streamflow prediction using machine learning models in selected rivers of Southern India. International Journal of River Basin Management. 2024; 22(4):529-55.
[8] VVR MR, Silpa N, Gadiraju M, Reddy SS, Bonthu S, Kurada RR. A plausible RNN-LSTM based profession recommendation system by predicting human personality types on social media forums. In 7th international conference on computing methodologies and communication (ICCMC) 2023 (pp. 850-5). IEEE.
[9] Kim CS, Kok KH, Kim CR. Quasi-Optimized LSTM approach for river water level forecasting. Water. 2025; 17(14):1-27.
[10] Danesh M, Gharehbaghi A, Mehdizadeh S, Danesh A. A comparative assessment of machine learning and deep learning models for the daily river streamflow forecasting. Water Resources Management. 2025; 39(4):1911-30.
[11] Hastie T, Tibshirani R, Friedman JH, Friedman JH. The elements of statistical learning: data mining, inference, and prediction. New York: springer; 2009.
[12] Deulkar AM, Dixit PR, Londhe SN, Jain RK. Comparative assessment of artificial neural networks (ANNs), long short term memory network (LSTM) and hydrologic engineering centre-hydrologic modelling system (HEC-HMS) for runoff modelling. Water Resources Management. 2025; 39(5):2049-68.
[13] Bassah R, Corzo G, Bhattacharya B, Haider SM, Swain ED, Aumen N. Forecasting water levels using the ConvLSTM algorithm in the Everglades, USA. Journal of Hydrology. 2025; 652:1-17.
[14] Sarkar A, Shahriyar MF, Chy AM, Tanzin MA, Mashrafi MJ, Fahim A. Hybrid convolutional neural network and random forest model for predicting water level fluctuations in Kaptai reservoir to enhance water resource management. In international conference on electrical, computer and communication engineering (ECCE) 2025 (pp. 1-6). IEEE.
[15] Herath M, Jayathilaka T, Hoshino Y, Rathnayake U. Deep machine learning-based water level prediction model for Colombo flood detention area. Applied Sciences. 2023; 13(4):1-17.
[16] Hafezifar E, Shourian M. Groundwater level prediction using deep learning-based recurrent neural network and numerical modeling: a comparative study. Earth Science Informatics. 2025; 18(2):1-18.
[17] Li Y, Qu J, Zhang H, Long Y, Li S. Water level prediction of Liuxihe reservoir based on improved long short-term memory neural network. Water Supply. 2023; 23(11):4563-82.
[18] Ruma JF, Adnan MS, Dewan A, Rahman RM. Particle swarm optimization based LSTM networks for water level forecasting: a case study on Bangladesh river network. Results in Engineering. 2023; 17:1-14.
[19] Li H, Zhang L, Zhang Y, Yao Y, Wang R, Dai Y. Water-level prediction analysis for the three gorges reservoir area based on a hybrid model of LSTM and its variants. Water. 2024; 16(9):1-19.
[20] Azad AS, Islam N, Nabi MN, Khurshid H, Siddique MA. Developments and trends in water level forecasting using machine learning models–a review. IEEE Access. 2025.
[21] Li H, Zhang L, Yao Y, Zhang Y. Prediction of water levels in large reservoirs base on optimization of deep learning algorithms. Earth Science Informatics. 2025; 18(1):1-15.
[22] Singh AK, Singh V, Kumar A, Roshni T. From bias to forecast: advancing satellite rainfall accuracy and flood prediction with transformer modeling in the Kosi basin (India). Stochastic Environmental Research and Risk Assessment. 2025; 39(12):5933-57.
[23] Kovvuri AR, Uppalapati PJ, Bonthu S, Kandula NR. Water level forecasting in reservoirs using time series analysis–auto ARIMA model. In international conference on cognitive computing and cyber physical systems 2022 (pp. 192-200). Cham: Springer Nature Switzerland.
[24] Luo J, Zhu D, Li D. Classification-enhanced LSTM model for predicting river water levels. Journal of Hydrology. 2025; 650:132535.
[25] Wee WJ, Zaini NA, Ahmed AN, El-shafie A. A review of models for water level forecasting based on machine learning. Earth Science Informatics. 2021; 14(4):1707-28.
[26] Jayathilake T, Gunathilake MB, Wimalasiri EM, Rathnayake U. Wetland water-level prediction in the context of machine-learning techniques: where do we stand?. Environments. 2023; 10(5):1-16.
[27] Truong VA, Hoang TN, Truong DV. Deep learning for downstream water level prediction in complex hydrology systems: an LSTM approach. Open Journal of Modern Hydrology. 2025; 15:218-32.
[28] Malekpour MM, Malekpoor H. Reservoir water level forecasting using wavelet support vector regression (WSVR) based on teaching learning-based optimization algorithm (TLBO). Soft Computing. 2022; 26(17):8897-909.
[29] Renteria-mena JB, Plaza D, Giraldo E. Multivariate hydrological modeling based on long short-term memory networks for water level forecasting. Information. 2024; 15(6):1-18.
[30] Tran TT, Nguyen LD, Hoai PN, Pham QB, Huyen PT, Dong NP, et al. Long short-term memory (LSTM) neural networks for short-term water level prediction in Mekong river estuaries. Songklanakarin Journal of Science & Technology. 2022; 44(4):1057–66.
[31] Vizi Z, Batki B, Rátki L, Szalánczi S, Fehérváry I, Kozák P, et al. Water level prediction using long short-term memory neural network model for a lowland river: a case study on the Tisza River, Central Europe. Environmental Sciences Europe. 2023; 35(1):1-18.
[32] Pandit CK. Research article of temporal rainfall and its correlation W temperature in Ranchi, Jharkhand. Journal of Current Research. 2023; 15(8):25519-22.
[33] Kumar S, Singh V, Kumar VP, Roshni T. Studying the impact of rainfall and LULC changes in the reservoir water level of the Hatia dam, Ranchi, India. Water Supply. 2025; 25(3):522-44.
[34] Singh AK, Singh V. Advancing real time flood prediction in the Kosi river basin (India): a machine learning framework leveraging satellite precipitation products. Theoretical and Applied Climatology. 2025; 156(9):1-23.
[35] Shuofeng L, Puwen L, Koyamada K. LSTM based hybrid method for basin water level prediction by using precipitation data. Journal of Advanced Simulation in Science and Engineering. 2021; 8(1):40-52.
[36] Widiasari IR, Efendi R. Utilizing LSTM-GRU for IoT-based water level prediction using multi-variable rainfall time series data. Informatics. 2024; 11:1-19.
[37] Nguyen XH. Combining statistical machine learning models with ARIMA for water level forecasting: the case of the red river. Advances in Water Resources. 2020; 142:103656.
[38] Kinga D, Adam JB. A method for stochastic optimization. In international conference on learning representations (ICLR) 2015 (pp.1-15).
[39] Gal Y, Ghahramani Z. A theoretically grounded application of dropout in recurrent neural networks. Advances in Neural Information Processing Systems. 2016; 29.
[40] Kumar A, Singh V. Real-time flood forecasting using satellite precipitation product and machine learning approach in Bagmati river basin, India. Acta Geophysica. 2024; 72(6):4431-51.
[41] Sapitang M, M RW, Faizal KK, Najah AA, El-shafie A. Machine learning application in reservoir water level forecasting for sustainable hydropower generation strategy. Sustainability. 2020; 12(15):1-19.
[42] Lee EH. Groundwater level prediction using modified recurrent neural network combined with meta-heuristic optimization algorithm. Groundwater for Sustainable Development. 2025; 28:101398.
[43] Abdi E, Taghi SM, Milewski A, Ibrahim OR. Advancements in hydrological modeling: the role of bRNN-CNN-GRU in predicting dam reservoir inflow patterns. Water. 2025; 17(11):1-21.
[44] Wang S, Xu O. Exploring a long short-term memory for mountain flood forecasting based on watershed-internal knowledge graph and large language model. Plos One. 2025; 20(3):e0318644.