Application of two-Dimensional (2-D) correlation on time-resolved
spectra of dispersive instruments with 10 percent of noise or more produced poor contrast results in both the synchronous and disrelational spectra. To overcome this problem, singular value decomposition (SVD) may be used to determine the principal components in the original data set. All temporal components that show random fluctuations around zero, with variances below the noise level can be ignored. 2-D correlation analysis on the data after removing noise produced consistent results.