Comparison of Data Driven Modelling Techniques For Rainfall Runoff Modelling of Kosi River
Abstract
Abstract
In the recent past, use of various machine learning techniques in predicting runoff from the catchment has become very popular. In this study, three empirical rainfall runoff models are employed to predict the discharge of the Kosi river for 11 years (2005-2015). The machine learning techniques such as support vector regression (SVR), multivariate adaptive regression splines (MARS) and random forest (RF) are employed for rainfall runoff modelling of Kosi watershed. The performances of all three prediction models have been successfully compared. Daily rainfall-runoff data for the period of 2005 to 2015 was collected for the Kosi river at Ramnagar barrage. It was seen that RF model outperformed over other two models. The gamma test was successfully applied in determination of the best input variables. The performance of the models is evaluated in terms of efficiency measures such as coefficient of determination (R2), root mean squared error (RMSE) and normalized squared error (NSE). The results revealed that random forest with R2 value 0.95 in testing phase performed superior than other two models. The performance of MARS model was satisfactory while SVR model resulted very poor values. Therefore, RF model can be considered as most accurate model for prediction of discharge.
Downloads
References
. Deen Dayal, Praveen K. Gupta and Ashish Pandey., (2021) Streamflow estimation using satellite-retrieved water fluxes and machine learning technique over monsoon-dominated catchments of India. Hydrological Sciences Journal, 66:4, 656-671.
. Maity Rajib, Bhagwat Parag and Bhatnagar Ashish (2010). Potential of support vector regression for prediction of monthly streamflow using endogenous property. Hydrological Processes, 24, 917–923.
. Zhenliang Yin, Qifeng, Xiaohu Wen, Ravinesh C. Deo, Linshan Yang, Jianhua Si, Zhibin He (2018). Design and evaluation of SVR, MARS and M5Tree models for 1, 2 and 3-day lead time forecasting of river flow data in a semiarid mountainous catchment. Stochastic Environmental Research and Risk Assessment.
. Dinh Tuan Vu, Xuan Linh Tran, Minh-Tu Cao, Thien Cuong Tran, Nhat-Duc Hoang (2020). Machine Learning Based Soil Erosion Susceptibility Prediction Using Social Spider Algorithm Optimized Multivariate Adaptive Regression Spline.
. Joshi Jignesh, Patel Vinod M., (2011) Rainfall-Runoff Modeling Using Artificial Neural Network. National Conference on Recent Trends in Engineering & Technology
. J.H. Friedman, C.B. Roosen., (1995) An Introduction to Multivariate Adaptive Regression Splines, Statistical Methods in Medical Research, 4 197-217
. J.H. Friedman, Multivariate Adaptive Regression Splines., (1991) The Annals of Statistics, 19 1-67
. Hsu K Gao X, Sorooshain S.,and Gupta H.V. (1995). Artificial Neural Network Modelling of the Rainfall-runoff Process Water Resources, 3, 2517-2530.
. Keith Beven., (2012) Rainfall-Runoff Modelling The Primer. 1-5
. Xiao M, Zhang Q, Singh VP, Chen X., (2017) Probabilistic forecasting of seasonal drought behaviors in the Huai river basin, China. Theor Appl Climatol 128:667–677. https://doi.org/10.1007/ s00704-016-1733-x
. Seth Richa, Singh Prashant, Manindra Mohan, Singh Rakesh, Gupta Vinod K., Uniyal D.P., Dobhal Rajendra and Gupta Sanjay (2013). Assessment of Water Quality of Kosi River, Almora, Uttarakhand (India) for Drinking and Irrigation Purposes, TACL 3 (4), 287 – 297
. Zhaoli Wang, Chengguang Lai, Xiaohong Chen, Bing Yang, Shiwei Zhao and Xiaoyan Bai. (2015). Flood hazard risk assessment model based on random forest. Journal of Hydrology 527, 1130-1141.
Copyright (c) 2021 Nikhil Jadhav, Pankaj Kumar, Abhinav Kumar Singh
This work is licensed under a Creative Commons Attribution 4.0 International License.
I/We agree with the provision of the Bye-Law 118 of The Institution of Engineers (India) which states that copyright of each paper published in Institution Journal or Annual Technical Volume in full or in Abstract at its centres shall lie with the Institution.
