IWA Publishing

Finally, the accuracy of SWAT-SVR models for estimating streamflow was evaluated and compared according to Moriasi et al. (2007) criteria. The statistical performance indices R², NSE, PBIAS, and RSR for GPM, PERSIANN, CHIRPS, and CMORPH SWAT-ANN hybrid models for training and testing varied from 0.76 to 0.87 and 0.75 to 0.85, 0.75 to 0.87 and 0.75 to 0.85, −7.6 to −2.1% and −7 to −1.8%, and 0.36 to 0.5 and 0.38 to 0.50, respectively, as presented in Table 9. Furthermore, all the SVR models' performance was good compared with the calibrated SWAT-CUP model in predicting streamflow. Overall, it was observed from the results (Tables 6,⁷⁸–9) that the rank of the overall performance of the SWAT-based models in predicting streamflow was SWAT-RF > SWAT-ANN > SWAT-SVR > SWAT-CUP.

Table 9

SWAT-SVR model performance using different SPDs

SPDs model	R²		NSE		PBIAS (%)		RMSE (m³/s)		RSR
SPDs model	Training	Testing	Training	Testing	Training	Testing	Training	Testing	Training	Testing
GPM	0.87	0.85	0.87	0.85	−2.10	−1.80	216.68	224.98	0.36	0.38
PERSIANN	0.83	0.82	0.83	0.82	−3.30	−2.10	247.86	249.25	0.41	0.43
CHIRPS	0.83	0.82	0.82	0.82	−2.90	−2.40	252.93	248.18	0.42	0.42
CMORPH	0.76	0.75	0.75	0.75	−7.60	−7.00	301.62	300.00	0.50	0.50

SPDs model	R²		NSE		PBIAS (%)		RMSE (m³/s)		RSR
SPDs model	Training	Testing	Training	Testing	Training	Testing	Training	Testing	Training	Testing
GPM	0.87	0.85	0.87	0.85	−2.10	−1.80	216.68	224.98	0.36	0.38
PERSIANN	0.83	0.82	0.83	0.82	−3.30	−2.10	247.86	249.25	0.41	0.43
CHIRPS	0.83	0.82	0.82	0.82	−2.90	−2.40	252.93	248.18	0.42	0.42
CMORPH	0.76	0.75	0.75	0.75	−7.60	−7.00	301.62	300.00	0.50	0.50

View Large

This Feature Is Available To Subscribers Only