Figure 2

Output from certain GFDL 20thc models showed promise. While the two M-SSA-identified leading modes showed potential for an oscillatory pair in GFDL_2_0_20thc Run 1, (a) as deduced from the two leading modes (at p < 5% significance) being mostly outside the red-noise-established envelope of uncertainty in the M-SSA spectrum; (b) and the RCs of leading modes one and two showing similar periodicity; (c) their failure to produce a propagating signal is due to the normalized RCs of the combined modes being in exact phasing, not offset (see text). Thus, this cannot be considered the “stadium wave”.