What is Singular Value Decomposition?
Singular value decomposition (SVD) is one of techniques that isolate
coupled modes of variability between
time series of two fields. Bretherton et al, 1992 (See Reference below ) reviewed the application of SVD methods to meteorological
data, and compared it with other methods in the context of studing coupled modes between
time series of two fields. Their results show that SVD is very simple to perform and interpret and requires no user-supplied parameters.
Bretherton, C.S., C. Smith, and J.M. Wallace, 1992: An Intercomparison of Methods for Finding Couples Patterns in Climate Data. J. Climate , 5, 541-560.
Wallace, J.M., C. Smith, and C.S. Bretherton, 1992: Singular Value Decomposition of Wintertime Sea Surface Temperature and 500-md Height Anomalies J. Climate ,