Evaluation of methodologies to quantify dry matter intake in grazing New Zealand dairy cattle : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Agricultural Science at Massey University, Manawatū, New Zealand

Abstract

Dry matter intake (DMI) is a key driver of enteric methane emissions in cattle, therefore the accurate measurement of DMI in grazing dairy cattle is necessary for research in methane emission reduction. A gap in methodologies currently available in New Zealand (NZ) to accurately predict individual grazing cow DMI for use in greenhouse gas (GHG) emission research has been identified. Therefore, the objective of this thesis is to quantify the variation between individual intake of “grazing” dairy cattle in NZ predicted from five methodologies; two indigestible markers techniques being n-alkanes and titanium dioxide paired with indigestible neutral detergent fibre (iNDF) and three standard energetics back calculation equations, all compared to actual intake measured by the Calan gate disappearance method. To test the marker techniques, an experiment was designed to provide data on actual individual cow DMI to compare intake predicted by the n-alkane, titanium dioxide and iNDF methodologies. The experiment occurred from 26 October to 4 November 2022, using two 5- day measurement periods, within the DairyNZ Calan gate facility, so that actual individual cow intakes were measured and alternative methodologies could be compared against each other. Using 40-multiparous early-lactation Holstein Friesian cattle, in two treatment groups - Control (pasture only diet) or Supplement (pasture and supplement diet), with treatment groups balanced for age, days in milk, liveweight and milk production. The cattle were previously adapted to diet, treatment and indigestible markers used, over a 3-week period prior to the measurement period beginning. All the cattle were dosed with n-alkane C32 (377.6 mg) and titanium dioxide (5 g) twice a day, following faecal collection. Supplement, pasture and faecal samples were bulked by measurement period for each individual cow, alongside samples of n alkane and titanium dioxide dosed, then analysed for n-alkanes C27 to C35, titanium dioxide and iNDF. The n-alkanes C29, C31, C33 and C35 were each paired with C32 to predict DMI, as was the pair titanium dioxide and iNDF. The n-alkane pair C32:C33 provided the most accurate prediction of pasture intake in relation to actual pasture intake, although there was an underestimation on average of 35 to 40%. Titanium dioxide and iNDF did not accurately predict total intake, with both over and underestimations of actual intake occurring. The back calculation methodologies included three energetic back calculation equations commonly used in NZ (NRC, MPI GHG Inventory and Nicol and Brookes). The ability of these methodologies to predict actual individual dairy cow DMI was tested across six datasets: five previous DairyNZ trials were used along with data from the experiment discussed above, all of which have DMI measured from the Calan gate system. Liveweight change is a variable included in the Nicol and Brookes equation and within these analyses added large variation to the energetic requirements and calculated daily DMI for individual cows. Based on this, the equation from Nicol and Brookes was not included in the full analysis, consequently as the other two equations that were analysed in full were not suitable for use in non-lactating cattle, a trial was excluded due to the use of non-lactating cattle. At an individual cow level, a single equation did not consistently provide predictions within the same range of accuracy across the trials. The data indicated that of the equations analysed, the NRC equation provided the best DMI predictions for individual lactating dairy cows in terms of accuracy and ease of use. At the herd level, DMI predictions were in a range of accuracy for all equations, with the NRC equation providing the most consistent results.