Here is another noise budget update, with measurements taken over the last several days.  The main difference between this and 45036 is the improvements in the MICH and SRCL feedforward (45110), and I have updated some of the ASC injections. 

The third attachment shows the noise budget sum compared to the DCPD cross corelation this morning:

This is based on only 5 minutes of data, because I was trying to avoid glitches and times when there are injections. At high frequencies, we don't have frequency noise included in the noise budget yet, which could explain part of the discrepancy between the cross corelation and the noise budget sum at high frequencies.  At low frequencies we don't know what this noise is, but there are some peaks on a smooth background.