In order to get a correct end result, the variables may not have to be seperated. But, you do have to actually have them in the model. Wether seperated or not may not be so important, but if they are missing altogether will likely have an impact. That's the bit that keeps going over your head.
For example, you do not include in your model any allowance for the fact that nitrate is primarily exported from the tank by means other than the skimmer. So, if you narrow it down so that in your model, ALL export is done by the skimmer, nothing else, your model does not reflect a real tank.
Can you see that now?
I'll agree with that, if by "the purposes here", you mean nothing other than the relationship between x amount of skimming, x amount of dirt, and x amount of volume. Your model will show that, and so will the one in the link, as that was all it was designed for.
However, you have stated that the reason for all this argument was to try to prove Cracker wrong when he made a comment about increased volume.
You need to realise that Cracker was talking about a real tank, not a model that only allows one form of export, a skimmer.
So as long as you continue to doggedly stick to your skimmer export only model, you will likely convince yourself of your argument, but you will not convince me, or Cracker, or anyone else, who realises that in an actual tank there are other forms of export, for example processing of nitrate by liverock, vacuuming of sand, etc. These other processes throw your skimmer only model, neat and tidy as it is, right out of whack.
So while, as you keep saying, I cannot see the forest for the trees, at least I know what the tree is. :lol: