Advanced Machine Learning Techniques to Improve Hydrological Prediction: A Comparative Analysis of Streamflow Prediction Models
Open accessThe DOI matches its Crossref record; 59 of the 59 references it cites were found in a registry or an index.
Abstract
The management of water resources depends heavily on hydrological prediction, and advances in machine learning (ML) present prospects for improving predictive modelling capabilities. This study investigates the use of a variety of widely used machine learning algorithms, such as CatBoost, ElasticNet, k-Nearest Neighbors (KNN), Lasso, Light Gradient Boosting Machine Regressor (LGBM), Linear Regression (LR), Multilayer Perceptron (MLP), Random Forest (RF), Ridge, Stochastic Gradient Descent (SGD), and the Extreme Gradient Boosting Regression Model (XGBoost), to predict the river inflow of the Garudeshwar watershed, a key element in planning for flood control and water supply. The substantial engineering feature used in the study, which incorporates temporal lag and contextual data based on Indian seasons, leads it distinctiveness. The study concludes that the CatBoost method demonstrated remarkable performance across various metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared (R2) values, for both training and testing datasets. This was accomplished by an in-depth investigation and model comparison. In contrast to CatBoost, XGBoost and LGBM demonstrated a higher percentage of data points with prediction errors exceeding 35% for moderate inflow numbers above 10,000. CatBoost established itself as a reliable method for hydrological time-series modelling, easily managing both categorical and continuous variables, and thereby greatly enhancing prediction accuracy. The results of this study highlight the value and promise of widely used machine learning algorithms in hydrology and offer valuable insights for academics and industry professionals.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Biau, G., & Scornet, E. (2016) · Test · doi:10.1007/s11749-016-0481-7 · Cited by 4,149 · cited 1×
“Each tree predicts the mean value of the target variable [69].”
BackgroundGardner, M. W., & Dorling, S. R. (1998) · Atmospheric Environment · doi:10.1016/s1352-2310(97)00447-0 · Cited by 3,324
CatBoost: unbiased boosting with categorical features
Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.09516 · Cited by 1,110
What Role Does Hydrological Science Play in the Age of Machine Learning?
Nearing, G. S., Kratzert, F., Sampson, A. K., Pelissier, C. S., Klotz, D., Frame, J. M., Prieto, C., & Gupta, H. V. (2021) · Water Resources Research · doi:10.1029/2020wr028091 · Cited by 894 · cited 1×
“Physically based streamflow forecasting models are based on certain hydrological hypotheses and require a large quantity of hydrological data for calibration [9].”
BackgroundMcDonald, G. C. (2009) · WIREs Computational Statistics · doi:10.1002/wics.14 · Cited by 823 · cited 1×
“Regression analysis frequently uses Ridge regression, commonly referred to as Tikhonov regularization, to address the multicollinearity and overfitting issues [71].”
MethodTyralis, H., Papacharalampous, G., & Langousis, A. (2019) · Water · doi:10.3390/w11050910 · Cited by 771
Su, X., Yan, X., & Tsai, C. (2012) · WIREs Computational Statistics · doi:10.1002/wics.1198 · Cited by 652 · cited 1×
“Different types of functions or models can be employed for regression, wherein a linear function is the simplest one [67].”
BackgroundKrämer, O. (2013) · Springer eBooks · doi:10.1007/978-3-032-06637-4_2 · Cited by 639
Improving the representation of hydrologic processes in Earth System Models
Clark, M. P., Fan, Y., Lawrence, D. M., Adam, J. C., Bolster, D., Gochis, D. J., Hooper, R. P., Kumar, M., Leung, L. R., Mackay, D. S., et al. (2015) · Water Resources Research · doi:10.1002/2015wr017096 · Cited by 636 · cited 1×
“These models reveal information on the mechanics of runoff production and flow dynamics, making them helpful tools for understanding the behavior of watersheds [12].”
BackgroundArtificial intelligence based models for stream-flow forecasting: 2000–2015
Yaseen, Z. M., El-shafie, A., Jaafar, O., Afan, H. A., & Sayl, K. N. (2015) · Journal of Hydrology · doi:10.1016/j.jhydrol.2015.10.038 · Cited by 570
Extreme Gradient Boosting as a Method for Quantitative Structure–Activity Relationships
Sheridan, R. P., Wang, W. M., Liaw, A., Ma, J., & Gifford, E. M. (2016) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.6b00591 · Cited by 544 · cited 1×
“The XGBoost approach delivers a variety of performance-improving improvements while sharing a similar structure with other gradient-boosting regressors [74].”
BackgroundStochastic gradient descent with differentially private updates
Song, S., Chaudhuri, K., & Sarwate, A. D. (2013) · 2013 IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2013.6736861 · Cited by 501 · cited 1×
“It is a modification of the common Gradient Descent technique and is especially helpful in cases involving online and massively multi-user learning [73].”
MethodFan, J., Ma, X., Wu, L., Zhang, F., Yu, X., & Zeng, W. (2019) · Agricultural Water Management · doi:10.1016/j.agwat.2019.105758 · Cited by 455
Yang, S., Wan, M. P., Chen, W., Ng, B. F., & Dubey, S. (2020) · Applied Energy · doi:10.1016/j.apenergy.2020.115147 · Cited by 403
Noori, R., Karbassi, A. R., Moghaddamnia, A., Han, D., Zokaei-Ashtiani, M. H., Farokhnia, A., & Gousheh, M. G. (2011) · Journal of Hydrology · doi:10.1016/j.jhydrol.2011.02.021 · Cited by 371
Improving streamflow prediction in the WRF-Hydro model with LSTM networks
Cho, K., & Kim, Y. (2022) · Journal of Hydrology · doi:10.1016/j.jhydrol.2021.127297 · Cited by 332
Streamflow forecasting using extreme gradient boosting model coupled with Gaussian mixture model
Ni, L., Wang, D., Wu, J., Wang, Y., Tao, Y., Zhang, J., & Liu, J. (2020) · Journal of Hydrology · doi:10.1016/j.jhydrol.2020.124901 · Cited by 272
A robust method for non-stationary streamflow prediction based on improved EMD-SVM model
Meng, E., Huang, S., Huang, Q., Fang, W., Wu, L., & Wang, L. (2019) · Journal of Hydrology · doi:10.1016/j.jhydrol.2018.11.015 · Cited by 257
Comprehensive Overview of Flood Modeling Approaches: A Review of Recent Advances
Kumar, V., Sharma, K., Caloiero, T., Mehta, D., & Singh, K. (2023) · Hydrology · doi:10.3390/hydrology10070141 · Cited by 251 · cited 1×
“These methods provide a data-driven approach to hydrological modeling, allowing for the creation of prediction models that are more precise [43,44].”
MethodSelf-citationJiang, S., Zheng, Y., Wang, C., & Babovic, V. (2022) · Water Resources Research · doi:10.1029/2021wr030185 · Cited by 244 · cited 1×
“Artificial neural networks (ANNs), support vector machines (SVMs), Random Forests (RFs), gradient boosting machines (GBMs), deep learning (DL) [32], long short-term memory (LSTM) [33], Gaussian processes (GPs), and physics-informed ML [34,35] are a few ML techniques utilized in streamflow forecasting.”
BackgroundRyali, S., Chen, T., Supekar, K., & Menon, V. (2012) · NeuroImage · doi:10.1016/j.neuroimage.2011.11.054 · Cited by 220 · cited 1×
“ElasticNet is a linear regression model that combines the L 1 (Lasso) and L 2 (Ridge) regularization techniques [72].”
MethodHerath, H. M. V. V., Chadalawada, J., & Babovic, V. (2021) · Hydrology and Earth System Sciences · doi:10.5194/hess-25-4373-2021 · Cited by 218
Tyralis, H., Papacharalampous, G., & Langousis, A. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-05172-3 · Cited by 196
Kumar, V., Azamathulla, H. M., Sharma, K. V., Mehta, D. J., & Maharaj, K. T. (2023) · Sustainability · doi:10.3390/su151310543 · Cited by 190 · cited 1×
“For short-term forecasting applications in particular, LSTMs have proved effective in capturing temporal relationships in streamflow data and producing precise forecasts [55,56].”
BackgroundSelf-citationFlow forecast by SWAT model and ANN in Pracana basin, Portugal
Demirel, M. C., Venancio, A., & Kahya, E. (2009) · Advances in Engineering Software · doi:10.1016/j.advengsoft.2008.08.002 · Cited by 173 · cited 1×
“For predicting river inflow, physical based models like the Soil and Water Assessment Tool (SWAT) are frequently used in hydrology [40].”
MethodChadalawada, J., Herath, H. M. V. V., & Babovic, V. (2020) · Water Resources Research · doi:10.1029/2019wr026933 · Cited by 169 · cited 1×
“Artificial neural networks (ANNs), support vector machines (SVMs), Random Forests (RFs), gradient boosting machines (GBMs), deep learning (DL) [32], long short-term memory (LSTM) [33], Gaussian processes (GPs), and physics-informed ML [34,35] are a few ML techniques utilized in streamflow forecasting.”
BackgroundBhasme, P., Vagadiya, J., & Bhatia, U. (2022) · Journal of Hydrology · doi:10.1016/j.jhydrol.2022.128618 · Cited by 166
Lima, C., & Lall, U. (2010) · Journal of Hydrology · doi:10.1016/j.jhydrol.2009.12.045 · Cited by 163
Adapting California’s water management to climate change
Hanak, E., & Lund, J. R. (2012) · Climatic Change · doi:10.1007/s10584-011-0241-3 · Cited by 163 · cited 1×
“The use of accurate inflow predictions aids in mitigating risks, optimizing water storage, and facilitating efficient water resource management practices [6,7].”
BackgroundFeng, D., Lawson, K., & Shen, C. (2021) · Geophysical Research Letters · doi:10.1029/2021gl092999 · Cited by 162
Regression and classification using extreme learning machine based on L1-norm and L2-norm
Luo, X., Chang, X., & Ban, X. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.03.112 · Cited by 160
A review on the applications of machine learning for runoff modeling
Mohammadi, B. (2021) · Sustainable Water Resources Management · doi:10.1007/s40899-021-00584-y · Cited by 148 · cited 1×
“With the aid of these methods, it is possible to evaluate sizable amounts of historical data, spot trends, and build intricate connections between meteorological factors, hydrological parameters, and river inflows [29].”
MethodWang, Z., Yang, W., Liu, Q., Zhao, Y., Liu, P., Wu, D., Banu, M., & Chen, L. (2022) · Journal of Manufacturing Processes · doi:10.1016/j.jmapro.2022.02.053 · Cited by 139 · cited 1×
“Without using explicit physical equations, these models may discover patterns and connections directly from the available data [15].”
Singh, D., Vardhan, M., Sahu, R., Chatterjee, D., Chauhan, P., & Liu, S. (2023) · Hydrology and earth system sciences · doi:10.5194/hess-27-1047-2023 · Cited by 132
Ikram, R. M. A., Ewees, A. A., Parmar, K. S., Yaseen, Z. M., Shahid, S., & Kişi, Ö. (2022) · Applied Soft Computing · doi:10.1016/j.asoc.2022.109739 · Cited by 129
Venkataraman, K., Tummuri, S., Medina, A., & Perry, J. (2016) · Journal of Hydrology · doi:10.1016/j.jhydrol.2016.01.001 · Cited by 127
Akbarian, M., Saghafian, B., & Golian, S. (2023) · Journal of Hydrology · doi:10.1016/j.jhydrol.2023.129480 · Cited by 120
Regionalization of hydrological model parameters using gradient boosting machine
Song, Z., Xia, J., Wang, G., She, D., Hu, C., & Hong, S. (2022) · Hydrology and earth system sciences · doi:10.5194/hess-26-505-2022 · Cited by 117
Urban rail transit passenger flow forecast based on LSTM with enhanced long‐term features
Yang, D., Chen, K., Yang, M., & Zhao, X. (2019) · IET Intelligent Transport Systems · doi:10.1049/iet-its.2018.5511 · Cited by 115
Cai, H., Liu, S., Shi, H., Zhou, Z., Jiang, S., & Babovic, V. (2022) · Journal of Hydrology · doi:10.1016/j.jhydrol.2022.128495 · Cited by 111 · cited 1×
“Artificial neural networks (ANNs), support vector machines (SVMs), Random Forests (RFs), gradient boosting machines (GBMs), deep learning (DL) [32], long short-term memory (LSTM) [33], Gaussian processes (GPs), and physics-informed ML [34,35] are a few ML techniques utilized in streamflow forecasting.”
BackgroundSan, O., Rasheed, A., & Kvamsdal, T. (2021) · GAMM-Mitteilungen · doi:10.1002/gamm.202100007 · Cited by 91
Meresa, H. K. (2018) · Modeling Earth Systems and Environment · doi:10.1007/s40808-018-0532-z · Cited by 85
Kumar, V., & Yadav, S. M. (2022) · Water Science & Technology Water Supply · doi:10.2166/ws.2022.010 · Cited by 82 · cited 1×
“It helps in managing irrigation systems, operating reservoirs, and producing electricity [26].”
BackgroundSelf-citationApplied Machine Learning for Developing Next‐Generation Functional Materials
Dinic, F., Singh, K., Dong, T., Rezazadeh, M., Wang, Z., Khosrozadeh, A., Yuan, T., & Voznyy, O. (2021) · Advanced Functional Materials · doi:10.1002/adfm.202104195 · Cited by 82 · cited 1×
“The calibration and validation processes of these models are hampered by the absence of precise and comprehensive hydrological data, which reduces the forecasting accuracy [11].”
BackgroundReview and comparison of performance indices for automatic model induction
Chadalawada, J., & Babovic, V. (2017) · Journal of Hydroinformatics · doi:10.2166/hydro.2017.078 · Cited by 76 · cited 1×
“There are several assessment measures employed, including mean squared error (MSE), mean absolute error (MAE), root mean square error (RMSE), root mean square percent error (RMSPE), and R-squared (R 2 ) [75]. where the overall number of data points is n .”
MethodTurner, S. W. D., Marlow, D. R., Ekström, M., Rhodes, B., Kularathna, U., & Jeffrey, P. J. (2014) · Water Resources Research · doi:10.1002/2013wr015156 · Cited by 76 · cited 1×
“…while these conventional approaches have proved useful for understanding river inflow patterns and guiding water resource management decisions, they may have shortcomings in terms of capturing complicated non-linear interactions and managing huge datasets with a variety of influencing elements [37].”
BackgroundA review of data-driven modelling in drinking water treatment
Aliashrafi, A., Zhang, Y., Groenewegen, H., & Peleato, N. M. (2021) · Reviews in Environmental Science and Bio/Technology · doi:10.1007/s11157-021-09592-y · Cited by 71
Rivera-González, L., Bolonio, D., Mazadiego, L. F., & Valencia-Chapi, R. (2019) · Sustainability · doi:10.3390/su11195316 · Cited by 63
Ostadkalayeh, F. B., Moradi, S., Asadi, A., Nia, A. M., & Taheri, S. (2023) · Water Resources Management · doi:10.1007/s11269-023-03492-2 · Cited by 60
Comparative Study for Daily Streamflow Simulation with Different Machine Learning Methods
Hao, R., & Bai, Z. (2023) · Water · doi:10.3390/w15061179 · Cited by 58
Liang, J., Li, W., Bradford, S. A., & Šimůnek, J. (2019) · Water · doi:10.3390/w11020200 · Cited by 56 · cited 1×
“Physically based models require accurate hydrological data as inputs, such as rainfall volume, intensity, and dispersion [10].”
Dong, N., Guan, W., Cao, J., Zou, Y., Yang, M., Wei, J., Chen, L., & Wang, H. E. (2023) · Journal of Hydrology · doi:10.1016/j.jhydrol.2023.129246 · Cited by 55 · cited 1×
“These methods provide a data-driven approach to hydrological modeling, allowing for the creation of prediction models that are more precise [43,44].”
MethodStreamflow forecasting using a hybrid LSTM-PSO approach: the case of Seyhan Basin
Haznedar, B., Kılınç, H. Ç., Özkan, F., & Yurtsever, A. (2023) · Natural Hazards · doi:10.1007/s11069-023-05877-3 · Cited by 51
Mehta, D., Eslamian, S., & Prajapati, K. (2021) · Modeling Earth Systems and Environment · doi:10.1007/s40808-021-01259-5 · Cited by 42 · cited 1×
“Authorities can decide on evacuation, emergency response, and resource allocation in accordance with projections that are provided on an hourly or daily basis [23].”
BackgroundSelf-citationNeeds for Climate Information in Support of Decision-Making in the Water Sector
Stakhiv, E. Z., & Stewart, B. (2010) · Procedia Environmental Sciences · doi:10.1016/j.proenv.2010.09.008 · Cited by 42
Gangani, P., Mangukiya, N. K., Mehta, D., Muttil, N., & Rathnayake, U. (2023) · Climate · doi:10.3390/cli11050114 · Cited by 37
Self-citationMartinho, A. D., Saporetti, C. M., & Goliatt, L. (2022) · Hydrological Sciences Journal · doi:10.1080/02626667.2022.2141121 · Cited by 36
Improving Short-range Reservoir Inflow Forecasts with Machine Learning Model Combination
Rajesh, M., Anishka, S., Viksit, P. S., Arohi, S., & Rehana, S. (2022) · Water Resources Management · doi:10.1007/s11269-022-03356-1 · Cited by 31
Omukuti, J., Wanzala, M. A., Ngaina, J. N., & Ganola, P. (2023) · Climate Resilience and Sustainability · doi:10.1002/cli2.47 · Cited by 9 · cited 1×
“Long-term forecasting, however, covers the weekly, monthly, and yearly timescales [25].”
Background
Cited by
Babu, B. R., Yaswanth Sai Ram, C., Manoj, C., Venkatesh, G., Chandra, G. T., & Mohitha, C. (2025) · 2025 6th International Conference on IoT Based Control Networks and Intelligent Systems (ICICNIS) · doi:10.1109/icicnis66685.2025.11315730
Metadata
- Authors
- Kumar, V., Kedam, N., Sharma, K. V., Mehta, D. J., & Caloiero, T.
- Venue
- Water
- Publisher
- MDPI AG
- Published
- 2023-07-13
- Volume
- 15
- Issue
- 14
- Pages
- 2572
- ISSN
- 2073-4441 (ISSN-L)
- OpenAlex
- W4384341507
- Cited by
- 235
- References
- 59
- FWCI
- 19.57
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.