NavList:

I used the formulas from the Nautical Almanac for refraction and parallax in altitude and Frank's formula for calculating the LDs:

Refraction = -0.0167° / tan(H + 7.32/(H + 4.32))
PA = HP · cos(H)
Calculated LD = acos(sin(Dec.1) · sin(Dec.2) + cos(Dec.1) · cos(Dec.2) · cos(GHA2-GHA1))

Refraction for the Moon was about -1.2', and for Procyon about -5.2'. Since Procyon was "lowered" about 5 minutes but the Moon was also "lowered" about 1 minute, I reasoned that the distance should increase by the difference of 4 minutes. The Moon's parallax in altitude was about +41.4' ... which should increase the geocentric distance by as much. And the semi-diameter of the Moon was 14.7'. Again, increasing the distance by as much. So, the cleared lunar would then be about 31°09.7' (I used Excel to calculate everything, so there is a difference of 0.1' due to rounding).

The calculated LD at 0300 UT1 was 31°23.6' and at 0400 UT1 it was 31°02.0'. Interpolation yields a time of 03:38:37 UT1. I used Jean Meeus' formulas (i.e., my own spreadsheet) for the GHAs and declinations of the Moon and Procyon. I used the 2022 Nautical Almanac values for the semi-diameter and horizontal parallax of the Moon (14.7' & 54.1' ... all day).

I notice that these figures are almost identical to the those which Lars Bergman arrived at ... except curiously, I forgot to include the augmentation of the Moon's semi-diameter and I did not take oblateness into account.

Cheers!
Sean C.