NavList:

My apologies if this sounds like a beginner's question. I need an example to illustrate it - I chose 16th January 2023 from the HMNAO Nautical Almanac and from my own 'skyalmanac'. I calculate the following data:

Note that I add a sign (declination moving Southerly is negative) and I calculate the EXACT difference of the Declination at h and h+1 and THEN round to one decimal place of minutes of arc. The official Nautical Almanac has the following data:

All d-value data in the HMNAO almanac is unsigned. Furthermore it is the precise difference OF THE ROUNDED Declinations in the previous column. As you see it begins with 13.6' decreasing to 13.4' then increasing to 13.5' then decreasing to 13.3' then increasing to 13.4' then decreasing to 13.2' then increasing to 13.3' and so on. The Moon doesn't have hourly booster rockets so I ask myself how this fake delta value can be more practical than my calculations above that have a smooth variation from hour to hour.

My next question relates to the sign of the d-value. My definiton above is not what David Burch advocates (Chapter 9 "Moon Sights",page 100, Celestial Navigation 2nd edition).:
"Select the sign of the d (±) the same as with the sun or planets by noting if the declination is increasing (+) or decreasing (-) with time. Don't make the mistake of using the d-value itself for this. The d-value can go up with time when the declination is going down, and vice-versa."

The last sentence supports the idea that the d-value is the difference of the already ROUNDED Declination figures. If I was to print the sign in the above data starting from 00h on 16th January 2023, I assume it would appear as follows: -13.6 -13.5 +13.4 -13.5 +13.3 -13.4 +13.2 -13.3  13.1  13.1 -13.1 +12.9 .......
which looks utterly confusing to me.
Is it feasibile to improve on the d-value representation in the official Nautical Almanac or is it best left alone to comply with the instructions regarding Sight Reduction of the Moon?

Kind Regards
Andrew