Table 2: Statistical comparison of algorithms using nonlinear regression with PR observations (validation data sets).

Algorithm appliedNo. of data pointsCorrelation coefficientsRoot mean square error (RMSE) (mm/h)Bias (mm/h)

Nonlinear regression using SI only6842
0.722
3.186
2.171
Nonlinear regression using PCT only6842 0.741 2.763 1.962
Nonlinear regression using both SI and PCT6842 0.784 1.896 1.182