A comparison of narrow lines in 133.5 hours of O2 DARM to those
seen in 107 hours of ER10 DARM shows very similar structures and
strengths. On the whole, the low-frequency lines are generally higher
in the O2 data than in ER10, but the upconversion around the quad
violin modes & harmonics is better in the O2 data, consistent with the
significant mitigation of low-frequency doublets with a splitting of
0.0467 Hz, suspected to be due (somehow) to beating of the 2nd quad violin
harmonics 1009.4414 and 1009.4881 Hz.

For reference, zipped tar files of sub band spectra up to 2000 Hz are attached,
where the orderings of the two overlain curves (ER10, O2's 1st 13 days) are
reversed, to make more apparent in which data sets individual lines are stronger.