Thank you for your comments – I have had a sleepless night redoing my calculations as a result of your post; and yes, there is an error in my calculations. However, that error does not change the key conclusion provided in my evidence that if built, the plant will require 555,508 cows to supply it, and this has a ‘people equivalent’ of 8,332,620 (which is where the comparison with the city of Jakarta comes from).

As noted by Jeanette, my figure of 555,508 cows includes replacements – so it needs to be halved to get a per drier figure – this is 277,754 cows per drier; which can be further decomposed to a ‘milking herd’ of 195,652 cows with 40,969 Rising 1 heifers and 40,133 Rising 2 heifers.

The 555,508 figure is then multiplied by 15 to get the 8,332,620 people figure.

For completeness, I assumed cows produce 23 litres of milk per day at the peak (so 4.5M litres / 23 litres per cow per day = 195,652 cows).

With respect to the 11.5 conversion between litres and kgMS (i.e. 100 litres of milk = 8.69 kgMS), I note in my evidence that milk composition changes across the season and have used 11.5 as a planning number – though accept other conversion figures can be used (indeed, when at MAF we used to use 12.61).

I also agree there is a 3.3% differential between litres of milk and kgs of liquid raw milk – but given I was doing an ‘order of magnitude’ comparison I do not consider this materially effects the conclusions.

The error I made relates to the ratio between converting kgMS to kg WMP. I assumed a ratio of 1:1 when it is more accurately 1:1.5-1:1.75. As a result, I under-estimate the plant production, finding that when processing 4.5M litres of milk per day it will be operating at about 56% capacity when it will actually be running at, or very near to, full capacity. I then posed the question of ‘if the plant is only running at 56% capacity how many cows will be required for it to operate at full capacity?’

This is where the 1 million cows number comes from and I acknowledge this number is incorrect.

This also implies that given the revision above, the scenario 2 and 3 numbers I state in my evidence need to be discounted.

For clarification, I never stated that if the plant was processing 4.5M litres per drier per day then it will require 1 million cows to supply it – I agree with you – the figure is about half of that.

I am deeply embarrassed and shocked by this error as I take tremendous pride – and considerable care – in producing credible figures. Indeed, I sought advice regarding the specific calculation as I could not understand why two plants would be built and then operated at a little above half capacity and I suspected I had made an error (and specifically asked whether I had). Unfortunately, not only was my error not picked up but the advice I got was that both scenarios were possible. I therefore proceeded as per my tendered evidence.

I take full responsibly for the error because it was issued under my name so it is my responsibility to ensure everything is correct – and the simple fact is it wasn’t. I therefore apologise unreservedly to all concerned for my error and for any distress or confusion it has caused.

I have also written to Mr Paul Rogers, Panel Chair, outlining the error, taking responsibility for it, and correcting it.

Again, I apologise unreservedly and thank you for bringing this to my attention.

Kind regards

Peter Fraser
Rōpere Consulting Limited

Just applying a very rough reconciliation from the total NZ dairy Cow numbers (~5 million) vs the total milk supplied in New Zealand (~21 billion litres) = 4.2 billion litres per million cows compared to the total milk expected to be processed by the proposed driers even if using 9 million litres per day at peak (1.8 billion litres per season) leads to an estimate of only 429,000 cows required… simple check that points out there is clearly an error…