# Successive-Station Streamflow Prediction and Precipitation Uncertainty Analysis in the Zarrineh River Basin Using a Machine Learning Technique

^{1}

^{2}

^{3}

^{4}

^{5}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Study Area

^{2}, is the largest lake in Iran and accounts for 7% of the country’s surface water [4]. Zarrineh River (ZR) basin is the largest and most crucial sub-basin of the LU basin, providing more than 41% (i.e., 1271 MCM) of the environmental flow into LU [3]. ZR is situated to the southeast of Lake Urmia and covers an area of about 12,025 km

^{2}with a length of around 300 km, as depicted in Figure 1 [41]. However, the lake surface area has drastically decreased to one-tenth, to 500 km, with the volume of half a billion cubic meters due to the unconventional use of available water and climate changes [2,3]. The Boukan Dam is the largest and most significant dam operating in the ZR basin with a live storage capacity of 650 MCM, storing water for drinking, agricultural, and industrial uses [42]. The average annual precipitation over the basin for the last four decades was 352 mm, which classifies the region as semi-arid with a Mediterranean climate.

## 3. Data Collection

## 4. Model Description

#### 4.1. Gated Recurrent Unit (GRU) Cell Structure

_{t−1}(h

^{′}

_{t−1}) is transferred to the up-to-date time step from the earlier one. More state data from the prior time step is produced by the greater number of update gates. The reset gate is applied to determine the degree to which the information from the previous state is forgotten. The lesser the reset gate, the more state information is forgotten. The update equations in the GRU cell structure are computed as per Equations (1)–(4):

#### 4.2. GLUE Theory

## 5. Methodology

#### 5.1. GRU Model Development

_{1}and S

_{2}consider all the features, except the current station’s streamflow with zero- and one-month lag times, respectively. However, S

_{3}, S

_{4}, and S

_{5}structures contain all the input variables with two- to four-month lag times. A total of five model structures were used to simulate all hydrometric stations, except for the first station (Safakhaneh station), which does not include the streamflow from the upstream station.

_{o}and Q

_{s}are the observed and estimated streamflow at time t, correspondingly.

#### 5.2. Data Normalization

_{i}and X

_{norm}denote the raw and normalized data, correspondingly. X

_{max}and X

_{min}represent the maximum and minimum of raw dataset, respectively.

#### 5.3. Model Evaluation Criteria

^{2}) (Equation (8)), which has a range of [0, 1], represents the linear relation between the observed and predicted data. The prediction model shows more reliable results if the value of R

^{2}is closer to 1. The root mean square error (RMSE) (Equation (9)) evaluates the magnitude of the difference between the observed and predicted values. The closer the value of RMSE to 0, the higher the accuracy of the prediction.

#### 5.4. Bias Correction Method

_{BC}is the bias-corrected GCM output, T

_{RAW}is the raw GCM output for the historical or future period, T

_{REF}is the GCM output from the historical reference period, and ${\sigma}_{T,REF}$ and ${\sigma}_{o,REF}$ are the standard deviation of GCM output and the standard deviation of reference observations from the reference period, respectively.

#### 5.5. Quantification of Input Data Uncertainty Using GLUE

## 6. Results

#### 6.1. Evaluation of GRU Networks

_{3}model structure. The GRU network might reach sub-optimal solutions using a random start point. Therefore, ten identical runs were performed for each structure, and the final model was selected based on the replication with the best performance in the testing period.

^{2}, and RMSE in the validation and testing phases to obtain high and comparable performance and avoid model overfitting. The S

_{1}structure with no lag time shows the poorest performance among the other models in that station. However, introducing antecedent streamflow of the station and a one-month lag time of other input parameters in the S

_{2}model increases the model performance significantly compared to the S

_{1}structure. The S

_{2}, S

_{3}, S

_{4}, and S

_{5}model structures have the same input variables with a one- to four-month lag time. All the available climate data with various lag times were considered in the model structures to obtain the best combination of these inputs and their period. In addition, lag times were chosen in order to analyze how temporal variations in inputs affect the results.

_{5}model, which showed lower results than the S4 model. This indicates that the model’s performance declines when complicating the model with excessive inputs. Overall, the S4 structure shows the best performance among the other models, with NSE, R

^{2}, and RMSE of 0.75, 0.78, and 5.7, respectively, in the testing phase. While the streamflow of an upstream station is not considered in the model structures of this station, the downstream stations of the Safakhaneh benefit from the upstream streamflow. The monthly inflow to the Boukan dam was predicted using five structures, in which S1 presents inferior performance compared to the other models. However, applying the station’s streamflow and various lag times substantially improves the statistical criteria of the model. The S

_{5}structure with all the input variables and a four-month lag time shows the best output results, with NSE, R

^{2}, and RMSE of 0.85, 0.86, and 20.7, respectively.

_{4}model composed of all the input variables and a three-month lag time. The evaluation criteria for the S

_{4}model are 0.98, 0.99, and 8.2 for NSE, R

^{2}, and RMSE, respectively, demonstrating the most accurate model. The most critical station in the Zarrineh River basin is the outlet station, i.e., Nezamabad, which yields the outflow to Lake Urmia. The model illustrates satisfactory output results in all the structures with the highest model performance in the S

_{3}model, with NSE, R

^{2}, and RMSE of 0.87, 0.88, and 18.3, respectively. Thus, the GRU network shows a significant capability to predict the successive-station monthly streamflow of the basin, particularly at the outlet station contributing to the Lake Urmia inflow.

^{2}in each station. Although the model shows some inconsistencies at high flows at the Safakhaneh station and Boukan dam, they performed reasonably for the low- and medium-range flows. The GRU model generally performed significantly for all the flows at the Qezkorpi and Nezamabad stations. Various climate data, the land use, and the location of stations are responsible for the inconsistency in the results for the same model structure in different stations. The results demonstrate that the model performed better for downstream stations compared to the upstream stations considering that the calibrated river flow reaches the downstream stations.

#### 6.2. Uncertainty

## 7. Conclusions

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Nakhaei, M.; Akrami, M.; Gheibi, M.; Coronado PD U, 2.0.0.4.; Hajiaghaei-Keshteli, M.; Mahlknecht, J. A novel framework for technical performance evaluation of water distribution networks based on the water-energy nexus concept. Energy Convers. Manag.
**2022**, 273, 116422. [Google Scholar] [CrossRef] - Jalili, S.; Hamidi, S.A.; Namdar Ghanbari, R. Climate variability and anthropogenic effects on Lake Urmia water level fluctuations, northwestern Iran. Hydrol. Sci. J.
**2016**, 61, 1759–1769. [Google Scholar] [CrossRef] [Green Version] - Yazdandoost, F.; Moradian, S.; Izadi, A. Evaluation of Water Sustainability under a Changing Climate in Zarrineh River Basin, Iran. Water Resour. Manag.
**2020**, 34, 4831–4846. [Google Scholar] [CrossRef] - Farajzadeh, J.; Fard, A.F.; Lotfi, S. Modeling of monthly rainfall and runoff of Urmia lake basin using “feed-forward neural network” and “time series analysis” model. Water Resour. Ind.
**2014**, 7, 38–48. [Google Scholar] [CrossRef] [Green Version] - Zhu, S.; Zhou, J.; Ye, L.; Meng, C. Streamflow estimation by support vector machine coupled with different methods of time series decomposition in the upper reaches of Yangtze River, China. Environ. Earth Sci.
**2016**, 75, 531. [Google Scholar] [CrossRef] - Zhang, J.; Chen, X.; Khan, A.; Zhang, Y.K.; Kuang, X.; Liang, X.; Nuttall, J. Daily runoff forecasting by deep recursive neural network. J. Hydrol.
**2021**, 596, 126067. [Google Scholar] [CrossRef] - Yuan, X.; Chen, C.; Lei, X.; Yuan, Y.; Adnan, R.M. Monthly runoff forecasting based on LSTM–ALO model. Stoch. Environ. Res. Risk Assess.
**2018**, 32, 2199–2212. [Google Scholar] [CrossRef] - Ghadami, N.; Gheibi, M.; Kian, Z.; Faramarz, M.G.; Naghedi, R.; Eftekhari, M.; Fathollahi-Fard, A.M.; Dulebenets, M.A.; Tian, G. Implementation of solar energy in smart cities using an integration of artificial neural network, photovoltaic system and classical Delphi methods. Sustain. Cities Soc.
**2021**, 74, 103149. [Google Scholar] [CrossRef] - Morovati, K.; Tian, F.; Kummu, M.; Shi, L.; Tudaji, M.; Nakhaei, P.; Olivares, M.A. Contributions from climate variation and human activities to flow regime change of Tonle Sap Lake from 2001 to 2020. J. Hydrol.
**2023**, 616, 128800. [Google Scholar] [CrossRef] - Duan, Q.; Sorooshian, S.; Gupta, V. Effective and efficient global optimization for conceptual rainfall-runoff models. Water Resour. Res.
**1992**, 28, 1015–1031. [Google Scholar] [CrossRef] - Wu, H.; Chen, B. Evaluating uncertainty estimates in distributed hydrological modeling for the Wenjing River watershed in China by GLUE, SUFI-2, and ParaSol methods. Ecol. Eng.
**2015**, 76, 110–121. [Google Scholar] [CrossRef] - Akbarian, H.; Jalali, F.M.; Gheibi, M.; Hajiaghaei-Keshteli, M.; Akrami, M.; Sarmah, A.K. A sustainable Decision Support System for soil bioremediation of toluene incorporating UN sustainable development goals. Environ. Pollut.
**2022**, 307, 119587. [Google Scholar] [CrossRef] [PubMed] - Mousavi, S.J.; Nakhaei, P.; Sadollah, A.; Kim, J.H. Optimization of Hydropower Storage Projects Using Harmony Search Algorithm. In Harmony Search Algorithm, Proceedings of the 3rd International Conference on Harmony Search Algorithm (ICHSA 2017), Bilbao, Spain, 22–24 Februar 2017; Del Ser, J., Ed.; Springer: Singapore, 2017; Volume 514. [Google Scholar] [CrossRef]
- Talebidaloueia, M.; Mirbagheria, S.A.; Nakhaeib, P. Treatment prediction of sugar industry wastewater in moving-bed biofilm reactor using multi expression programming. Desalination Water Treat.
**2020**, 191, 82–92. [Google Scholar] [CrossRef] - Gheibi, M.; Eftekhari, M.; Akrami, M.; Emrani, N.; Hajiaghaei-Keshteli, M.; Fathollahi-Fard, A.M.; Yazdani, M. A sustainable decision support system for drinking water systems: Resiliency improvement against cyanide contamination. Infrastructures
**2022**, 7, 88. [Google Scholar] [CrossRef] - Bárdossy, A. Neural network-based modelling of the river Danube. J. Hydrol.
**2008**, 349, 88–100. [Google Scholar] - Jang, D.; Kim, T.W.; Park, H. A Comparative Study of Machine Learning Methods for Streamflow Prediction of Unimpaired Rivers in the United States. Water
**2016**, 8, 438. [Google Scholar] - Liu, J.; Yang, W.; Xiong, L.; He, X. Prediction of Daily Streamflow Using Random Forest Model. J. Hydrol. Eng.
**2018**, 23, 04018041. [Google Scholar] - Xu, Z.; Xu, C.Y.; Song, J. A hybrid model based on Gaussian process regression and an improved differential evolution algorithm for river flow forecasting. J. Hydrol.
**2016**, 533, 143–153. [Google Scholar] - Zhang, H.; Sun, Z.; Liu, C.; Zhang, J.; Zhang, B. LSTM-based streamflow forecasting for a river with multiple dams considering hydrologic similarity. J. Hydrol.
**2019**, 574, 697–710. [Google Scholar] - Amiri, E. Forecasting daily river flows using nonlinear time series models. J. Hydrol.
**2015**, 527, 1054–1072. [Google Scholar] [CrossRef] - Arab, M.; Akbarian, H.; Gheibi, M.; Akrami, M.; Fathollahi-Fard, A.M.; Hajiaghaei-Keshteli, M.; Tian, G. A soft-sensor for sustainable operation of coagulation and flocculation units. Eng. Appl. Artif. Intell.
**2022**, 115, 105315. [Google Scholar] [CrossRef] - Hu, R.; Fang, F.; Pain, C.C.; Navon, I.M. Rapid spatio-temporal flood prediction and uncertainty quantification using a deep learning method. J. Hydrol.
**2019**, 575, 911–920. [Google Scholar] [CrossRef] - Morovati, K.; Nakhaei, P.; Tian, F.; Tudaji, M.; Hou, S. A Machine Learning Framework to Predict Reverse Flow and Water Level: A Case Study of Tonle Sap Lake. J. Hydrol.
**2021**, 603, 127168. [Google Scholar] [CrossRef] - Shahsavar, M.M.; Akrami, M.; Gheibi, M.; Kavianpour, B.; Fathollahi-Fard, A.M.; Behzadian, K. Constructing a smart framework for supplying the biogas energy in green buildings using an integration of response surface methodology, artificial intelligence and petri net modelling. Energy Convers. Manag.
**2021**, 248, 114794. [Google Scholar] [CrossRef] - Raghu, M.; Poole, B.; Kleinberg, J.; Ganguli, S.; Sohl-Dickstein, J. On the expressive power of deep neural networks. In Proceedings of the International Conference on Machine Learning 2017, Sydney, Australia, 6–11 August 2017; pp. 2847–2854. [Google Scholar]
- Cho, K.; Van Merriënboer, B.; Gulcehre, C.; Bahdanau, D.; Bougares, F.; Schwenk, H.; Bengio, Y. Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv
**2014**, arXiv:1406.1078. [Google Scholar] [CrossRef] - Fu, R.; Zhang, Z.; Li, L. Using LSTM and GRU neural network methods for traffic flow prediction. In Proceedings of the 2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC), Wuhan, China, 11–13 November 2016; pp. 324–328. [Google Scholar] [CrossRef]
- Apaydin, H.; Feizi, H.; Sattari, M.T.; Colak, M.S.; Shamshirband, S.; Chau, K.W. Comparative analysis of recurrent neural network architectures for reservoir inflow forecasting. Water
**2020**, 12, 1500. [Google Scholar] [CrossRef] - Zhu, S.; Heddam, S.; Nyarko, E.K.; Hadzima-Nyarko, M.; Piccolroaz, S.; Wu, S. Modeling daily water temperature for rivers: Comparison between adaptive neuro-fuzzy inference systems and artificial neural networks models. Environ. Sci. Pollut. Res.
**2019**, 26, 402–420. [Google Scholar] [CrossRef] - Bae, D.H.; Trinh, H.L.; Nguyen, H.M. Uncertainty estimation of the SURR model parameters and input data for the Imjin River basin using the GLUE method. J. Hydro-Environ. Res.
**2018**, 20, 52–62. [Google Scholar] [CrossRef] - Lee, H.; Balin, D.; Shrestha, R.R.; Rode, M. Streamflow prediction with uncertainty analysis, Weida catchment, Germany. KSCE J. Civ. Eng.
**2010**, 14, 413–420. [Google Scholar] [CrossRef] - Tang, X.; Zhang, J.; Wang, G.; Jin, J.; Liu, C.; Liu, Y.; Bao, Z. Uncertainty Analysis of SWAT Modeling in the Lancang River Basin Using Four Different Algorithms. Water
**2021**, 13, 341. [Google Scholar] [CrossRef] - Zhang, C.; Yan, H.; Takase, K.; Oue, H. Comparison of the soil physical properties and hydrological processes in two different forest type catchments. Water Resour.
**2016**, 43, 225–237. [Google Scholar] [CrossRef] - Her, Y.; Yoo, S.H.; Cho, J.; Hwang, S.; Jeong, J.; Seong, C. Uncertainty in hydrological analysis of climate change: Multi-parameter vs. multi-GCM ensemble predictions. Sci. Rep.
**2019**, 9, 4974. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Fabry, F. Obstacles to the greater use of weather radar information. In Proceedings of the 6th International Symposium on Hydrological Applications of Weather Radar, Melbourne, Australia, 2–4 February 2004. [Google Scholar]
- Beven, K.; Binley, A. The future of distributed models: Model calibration and uncertainty prediction. Hydrol. Process.
**1992**, 6, 279–298. [Google Scholar] [CrossRef] - Vrugt, J.A.; Diks, C.G.; Gupta, H.V.; Bouten, W.; Verstraten, J.M. Improved treatment of uncertainty in hydrologic modeling: Combining the strengths of global optimization and data assimilation. Water Resour. Res.
**2005**, 41. [Google Scholar] [CrossRef] - Vrugt, J.A.; Gupta, H.V.; Bouten, W.; Sorooshian, S. A Shuffled Complex Evolution Metropolis algorithm for optimization and uncertainty assessment of hydrologic model parameters. Water Resour. Res.
**2003**, 39, 1201. [Google Scholar] [CrossRef] [Green Version] - Liang, Y.; Cai, Y.; Sun, L.; Wang, X.; Li, C.; Liu, Q. Sensitivity and uncertainty analysis for streamflow prediction based on multiple optimization algorithms in Yalong River Basin of southwestern China. J. Hydrol.
**2021**, 601, 126598. [Google Scholar] [CrossRef] - Amini, A.; Ghazvinei, P.T.; Javan, M.; Saghafian, B. Evaluating the impacts of watershed management on runoff storage and peak flow in Gav-Darreh watershed, Kurdistan, Iran. Arab. J. Geosci.
**2014**, 7, 3271–3279. [Google Scholar] [CrossRef] - Emami, F.; Koch, M. Agricultural water productivity-based hydro-economic modeling for optimal crop pattern and water resources planning in the Zarrine River Basin, Iran, in the wake of climate change. Sustainability
**2018**, 10, 3953. [Google Scholar] [CrossRef] [Green Version] - Jubb, I.; Canadell, P.; Dix, M. Representative Concentration Pathways (RCPs); Australian Climate Change Science Program; Australian Government, Department of the Environment, Canberra: Canberra, Australia, 2013; pp. 5–7.
- Ramirez-Villegas, J.; Challinor, A.J.; Thornton, P.K.; Jarvis, A. Implications of regional improvement in global climate models for agricultural impact research. Environ. Res. Lett.
**2013**, 8, 024018. [Google Scholar] [CrossRef] - Mengistu, A.G.; Woldesenbet, T.A.; Dile, Y.T. Evaluation of the performance of bias-corrected CORDEX regional climate models in reproducing Baro–Akobo basin climate. Theor. Appl. Climatol.
**2021**, 144, 751–767. [Google Scholar] [CrossRef] - Kumar, P.S.; Praveen, T.V.; Prasad, M.A. Artificial Neural Network Model for Rainfall-Runoff—A Case Study. Int. J. Hybrid Inf. Technol.
**2016**, 9, 263–272. [Google Scholar] [CrossRef] - Muhammad, A.U.; Li, X.; Feng, J. Using LSTM GRU and Hybrid Models for Streamflow Forecasting. In Machine Learning and Intelligent Communications, Proceedings of the 4th International Conference—MLICOM 2019, Nanjing, China, 24–25 August 2019; Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering; Zhai, X., Chen, B., Zhu, K., Eds.; Springer: Cham, Switzerland, 2019; Volume 294. [Google Scholar] [CrossRef]
- Hutter, F.; Hoos, H.; Leyton-Brown, K. An efficient approach for assessing hyperparameter importance. In Proceedings of the 31st International Conference on Machine Learning, Beijing, China, 21–26 June 2014; pp. 754–762. [Google Scholar]
- Liang, C.; Li, H.; Lei, M.; Du, Q. Dongting lake water level forecast and its relationship with the three gorges dam based on a long short-term memory network. Water
**2018**, 10, 1389. [Google Scholar] [CrossRef] [Green Version] - Piani, C.; Weedon, G.P.; Best, M.; Gomes, S.M.; Viterbo, P.; Hagemann, S.; Haerter, J.O. Statistical bias correction of global simulated daily precipitation and temperature for the application of hydrological models. J. Hydrol.
**2010**, 395, 199–215. [Google Scholar] [CrossRef] - Tieleman, T.; Hinton, G. Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude. COURSERA: Neural Netw. Mach. Learn.
**2012**, 4, 26–31. [Google Scholar] - Kochenderfer, M.J.; Wheeler, T.A. Algorithms for Optimization; MIT Press: Cambridge, MA, USA, 2019; ISBN 978-0262039420. [Google Scholar]
- Mehan, S.; Neupane, R.P.; Kumar, S. Coupling of SUFI 2 and SWAT for Improving the Simulation of Streamflow in an Agricultural Watershed of South Dakota. Hydrol. Curr. Res.
**2017**, 8, 280. [Google Scholar] [CrossRef] - Kiyan, A.; Gheibi, M.; Akrami, M.; Moezzi, R.; Behzadian, K. A Comprehensive Platform for Air Pollution Control System Operation in Smart Cities of Developing Countries: A Case Study of Tehran. Environ. Ind. Lett.
**2023**, 1, 10–27. [Google Scholar] [CrossRef] - Kiyan, A.; Gheibi, M.; Akrami, M.; Moezzi, R.; Behzadian, K.; Taghavian, H. The Operation of Urban Water Treatment Plants: A Review of Smart Dashboard Frameworks. Environ. Ind. Lett.
**2023**, 1, 28–45. [Google Scholar] [CrossRef] - Gheibi, M.; Chahkandi, B.; Behzadian, K.; Akrami, M.; Moezzi, R. Evaluation of Ceramic Water Filters’ Performance and Analysis of Managerial Insights by SWOT Matrix. Environ. Ind. Lett.
**2023**, 1, 1–9. [Google Scholar] [CrossRef]

**Figure 1.**Zarrineh River basin: Part (

**A**) shows the location map of the Zarrineh River basin in the northwest of Iran and part (

**B**) shows the Zarrineh River basin and its rivers network along with hydrometric and meteorological stations 1–4.

**Figure 6.**Comparison of observed and predicted streamflow for (

**a**) Safakhaneh, (

**b**) Boukan dam, (

**c**) Qezkorpi, (

**d**) Nezamabad stations.

**Figure 7.**Uncertainty interval (95 PPU) of input data for (

**a**) Safakhaneh, (

**b**) Boukan dam, (

**c**) Qezkorpi, (

**d**) Nezamabad; the gray area is the uncertainty interval and the dots are observations.

Name | Model Structure |
---|---|

S_{1} | ${Q}^{t}=f({Q}_{us}^{t},{P}^{t},{T}_{max}^{t},{T}_{min}^{t},{T}_{avg}^{t})$ |

S_{2} | ${Q}^{t}=f\left({Q}_{us}^{t-1},{Q}_{us}^{t},{Q}^{t-1},{P}^{t-1},{P}^{t},{T}_{max}^{t-1},{T}_{max}^{t},{T}_{min}^{t-1},{T}_{min}^{t},{T}_{avg}^{t-1},{T}_{avg}^{t}\right)$ |

S_{3} | ${Q}^{t}=f({Q}_{us}^{t-2},{Q}_{us}^{t-1},{Q}_{us}^{t},{Q}^{t-2},{Q}^{t-1},{P}^{t-2},{P}^{t-1},{P}^{t},{T}_{max}^{t-2},{T}_{max}^{t-1},{T}_{max}^{t}$ ${T}_{min}^{t-2},{T}_{min}^{t-1},{T}_{min}^{t},{T}_{avg}^{t-2},{T}_{avg}^{t-1},{T}_{avg}^{t})$ |

S_{4} | ${Q}^{t}=f({Q}_{us}^{t-3},{Q}_{us}^{t-2},{Q}_{us}^{t-1},{Q}_{us}^{t},{Q}^{t-3},{Q}^{t-2},{Q}^{t-1},{P}^{t-3},{P}^{t-2},{P}^{t-1},{P}^{t},{T}_{max}^{t-3}$ ${T}_{max}^{t-2},{T}_{max}^{t-1},{T}_{max}^{t},{T}_{min}^{t-3},{T}_{min}^{t-2},{T}_{min}^{t-1},{T}_{min}^{t},{T}_{avg}^{t-3},{T}_{avg}^{t-2},{T}_{avg}^{t-1},{T}_{avg}^{t})$ |

S_{5} | ${Q}^{t}=f({Q}_{us}^{t-4},{Q}_{us}^{t-3},{Q}_{us}^{t-2},{Q}_{us}^{t-1},{Q}_{us}^{t},{Q}^{t-4},{Q}^{t-3},{Q}^{t-2},{Q}^{t-1},{P}^{t-4},{P}^{t-3},$ ${P}^{t-2},{P}^{t-1},{P}^{t},{T}_{max}^{t-4},{T}_{max}^{t-3},{T}_{max}^{t-2},{T}_{max}^{t-1},{T}_{max}^{t},{T}_{min}^{t-4},{T}_{min}^{t-3},{T}_{min}^{t-2},$ ${T}_{min}^{t-1},{T}_{min}^{t},{T}_{avg}^{t-4},{T}_{avg}^{t-3},{T}_{avg}^{t-2},{T}_{avg}^{t-1},{T}_{avg}^{t})$ |

**Table 2.**Performance of GRU-based streamflow forecasting models for five station structures in the Zarrineh River basin with varying monthly lag time.

Training Phase | Validation Phase | Testing Phase | ||||||||
---|---|---|---|---|---|---|---|---|---|---|

Station | Structure | NSE | R^{2} | RMSE | NSE | R^{2} | RMSE | NSE | R^{2} | RMSE |

Safakhaneh (#1) | S_{1} | 0.34 | 0.35 | 13.9 | 0.49 | 0.34 | 15.1 | 0.46 | 0.29 | 12.3 |

S_{2} | 0.53 | 0.66 | 10.2 | 0.52 | 0.63 | 11.1 | 0.54 | 0.57 | 8.1 | |

S_{3} | 0.74 | 0.8 | 7.5 | 0.73 | 0.69 | 5.8 | 0.69 | 0.71 | 6.6 | |

S_{4} | 0.75 | 0.86 | 5.8 | 0.75 | 0.78 | 6.7 | 0.8 | 0.8 | 5.3 | |

S_{5} | 0.73 | 0.81 | 7.7 | 0.79 | 0.75 | 6.6 | 0.75 | 0.78 | 5.7 | |

Boukan dam (#2) | S_{1} | −7.9 | 0.79 | 23.4 | −10.8 | 0.82 | 27.4 | −12.6 | 0.85 | 20.7 |

S_{2} | 0.79 | 0.89 | 35.1 | 0.83 | 0.88 | 31.5 | 0.75 | 0.92 | 28.1 | |

S_{3} | 0.8 | 0.81 | 35.5 | 0.73 | 0.76 | 27.7 | 0.78 | 0.81 | 25.9 | |

S_{4} | 0.84 | 0.84 | 31.3 | 0.78 | 0.88 | 28.5 | 0.81 | 0.83 | 24.2 | |

S_{5} | 0.88 | 0.89 | 26.8 | 0.88 | 0.89 | 23.5 | 0.85 | 0.86 | 20.7 | |

Qezkorpi (#3) | S_{1} | 0.94 | 0.96 | 15.2 | 0.86 | 0.84 | 18.6 | 0.95 | 0.99 | 12.7 |

S_{2} | 0.93 | 0.96 | 19.2 | 0.93 | 0.95 | 16.5 | 0.94 | 0.99 | 13.4 | |

S_{3} | 0.95 | 0.96 | 15.9 | 0.92 | 0.94 | 11.4 | 0.96 | 0.99 | 10.2 | |

S_{4} | 0.96 | 0.96 | 15.1 | 0.97 | 0.98 | 7.6 | 0.98 | 0.99 | 8.2 | |

S_{5} | 0.94 | 0.95 | 18.7 | 0.91 | 0.94 | 14.1 | 0.94 | 0.99 | 12.6 | |

Nezamabad (#4) | S_{1} | 0.72 | 0.72 | 42.3 | 0.66 | 0.72 | 34.7 | 0.71 | 0.77 | 27.7 |

S_{2} | 0.81 | 0.85 | 34.8 | 0.84 | 0.87 | 26.3 | 0.79 | 0.82 | 23.7 | |

S_{3} | 0.95 | 0.95 | 18.1 | 0.85 | 0.89 | 17.6 | 0.87 | 0.88 | 18.3 | |

S_{4} | 0.94 | 0.94 | 18.7 | 0.89 | 0.93 | 22.4 | 0.85 | 0.88 | 19.8 | |

S_{5} | 0.84 | 0.87 | 31.7 | 0.84 | 0.87 | 24.4 | 0.82 | 0.83 | 21.3 |

Station Names | $\mathit{L}\left(\mathit{P}|\mathit{Q}\right)$ | p-Factor (%) | r-Factor |
---|---|---|---|

Safakhaneh | 86 | 78.5 | 0.53 |

Boukan dam | 89 | 89.3 | 0.57 |

Qezkorpi | 91 | 86.6 | 0.52 |

Nezam Abad | 85 | 61.6 | 0.47 |

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. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Nakhaei, M.; Ghazban, F.; Nakhaei, P.; Gheibi, M.; Wacławek, S.; Ahmadi, M.
Successive-Station Streamflow Prediction and Precipitation Uncertainty Analysis in the Zarrineh River Basin Using a Machine Learning Technique. *Water* **2023**, *15*, 999.
https://doi.org/10.3390/w15050999

**AMA Style**

Nakhaei M, Ghazban F, Nakhaei P, Gheibi M, Wacławek S, Ahmadi M.
Successive-Station Streamflow Prediction and Precipitation Uncertainty Analysis in the Zarrineh River Basin Using a Machine Learning Technique. *Water*. 2023; 15(5):999.
https://doi.org/10.3390/w15050999

**Chicago/Turabian Style**

Nakhaei, Mahdi, Fereydoun Ghazban, Pouria Nakhaei, Mohammad Gheibi, Stanisław Wacławek, and Mehdi Ahmadi.
2023. "Successive-Station Streamflow Prediction and Precipitation Uncertainty Analysis in the Zarrineh River Basin Using a Machine Learning Technique" *Water* 15, no. 5: 999.
https://doi.org/10.3390/w15050999