“…Thus, re-dimensioning the input data to provide a robust calibration, and rescaling them to avoid losing the value of a certain variable are crucial prerequisites (Araghinejad 2014). Various techniques involving principal component analysis (PCA), wavelet analysis (WA), singular value decomposition (SVD), singular spectrum analysis (SSA), for example, have been applied by hydrologists for pre-possessing large volumes of data (Marques et al 2006, Hu et al 2007, Partal and Kisi 2007, Sivapragasam et al 2007, Wu et al 2009, Chen et al 2009, Aziz et al 2010, Martinez and Jones 2011, Anderson et al 2012, Córdoba-Machado et al 2015. Among the pre- A c c e p t e d M a n u s c r i p t 4 processing techniques, PCA is a common mathematical tool used to reduce the number of variables, whilst keeping their variation as far as possible .…”