Constants change for converstion factors for fat and protein to energy

[hut104]

working on #11246

[sno036]

Hi @hut104 - these are pretty big changes. I assume (while you are there it would be good to add some units/description!) that these are DM that is fat/protein and the conversion of DM to energy in MJ? That value seems rather high to me for fat and that is nearly double the value for protein (and I don't have sensible values in my head for protein). Are you incorporating some sort of efficiency in these that was not in the original?

[DocJamesD]

@sno036 The units are MJ/kg DM. The values for fat and protein been changed to that used by SCA (2007) for the energetic content of fat and protein gain/loss in animals. These are also derived from the average enthalpies of fat and protein, converted from Mcal to MJ. We decided that are also suitable for estimating the energy density (MJ/kg DM) of protein and fat in feeds - this is the method Hutton Oddy and I were using in the with our modelling efforts (albeit with slightly higher numbers from Jim Oltjen's maths, 23.8 for protein and 39.6 for fat) but I decided to use 23.6 and 39.3 for consistency with the rest of the model.

The SCA (1990) book doesn't use fermentable energy, it calculates things from OM apparently digested in the rumen (OMADR) or digestible OM intake, or from MD (MJ ME/kg DM) but without adjustment for fat or RUP, so this was a change in the logic/formulation between the 1990 and 2007 version. Unfortunately the 2007 SCA doesn't define FME via an equation, it just says MEI minus the energy contained in fat, UDP, and silage acids, but doesn't provide coefficients for estimating those values. The 2007 SCA book does provide a regression-based approach for estimating MD from DMD and fat (0.134 DMD + 0.235 EE + 1.23) which is similar to Eqn 32 in the Freer 2012 report but I can't find a value for RUP there either. U

I did reach out to Hutton to ask if he knew where the 14 value came from for protein and his theory was that it was possibly the 23.6 value with an estimate of 60% digestibility in the lower gut, but agreed that that wouldn't really affect fermentable energy in the rumen per se and said to use the higher values. He said that unfortunately it's hard to tell where Freer got the values from but in his opinion the protein value in particular was an underestimation. However, if there's more information as to why Freer took that approach that would be great - I have been asking around and trying to do some sleuthing as to where the numbers came from but it's not clear.

I am also working on a broader document unpicking the Freer intake model and exploring the coefficients/structure and trying to figure out where the different coefficients came from. It's a work in progress but I'm making good headway; I've had to set it aside for a few weeks due to an MLA deadline at the end of the month but have been working through the equations and doing some sensitivity analyses that I think folks will find useful.