Figure 2.

Result of a singular value decomposition of the data matrix shown in figure 1. Panel (a) shows the seven first left-singular vectors U_i multiplied by their corresponding singular value, S_i,i, and offset for clarity and with U₁ lowest. Panel (b) shows the seven corresponding right-singular vectors, i.e. the seven first columns of V, where in particular the first column (lower-most trace) indicates significant time-dependence of the magnitude of U₁. Panel (c) shows, from top to bottom, the magnitude of the singular values as a function of column number i as well as the first-order autocorrelation functions of U_i and V_i. No clear cut-offs (see main text) where noise becomes a dominant part of the data are evident.