Sauvignon blanc grape marc feeding for sheep and dairy cows in New Zealand : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

The objective of this research was to evaluate the nutritional value of ensiled sauvignon blanc grape marc (GM) as a supplement for sheep and dairy cows in New Zealand pasture based systems, establish the limitations of use in these systems, and to evaluate any advantage in feed value gained by physically transforming this GM. This was achieved with three separate experiments. The first experiment compared this GM without physical transformation with three common supplements using lactating cows grazing high quality pasture. The second experiment evaluated physically transformed and non-transformed GM as a supplement to lucerne in a digestibilty study with sheep, and the third experiment used rumen fistulated, non-lactating cows with an in sacco assessment of physically transformed and non-transformed GM. The first experiment evaluated the effect of non-transformed GM fed as a supplement at 25 % of the diet on milk production, milk fatty acid profile, rumen function (pH, volatile fatty acid and ammonia concentration), and urinary and faecal nitrogen concentrations. Supplement treatments were GM, grass silage (GS), canola meal (CM) and palm kernel expeller (PKE). Milk yield and milk fat percentage were greater in cows fed CM (P=0.01) but not different between other treatment groups. Milk total FA concertation was not affected by dietary treatments (P=0.13). Total C18`s were significantly greater in CM (P=0.002), but there was no difference between other treatment groups. Rumen pH, ammonia concentration and volatile fatty acids concentration were not different among treatments (P>0.05). Urinary nitrogen, NH3 and urea also were not affected by dietary treatments (P>0.05). The faecal N was significantly lowest in cows fed GM (2.68 %; P=0.01), compared to cows fed PKE (2.97 %), GS (2.85 %) and CM (2.86 %). It was concluded that ensiled sauvignon blanc GM use as a supplement at 25 % of the diet for periods of less than 30 d in mid-lactation dairy cows in a pasture-based system had no adverse impact on milk production, milk composition or rumen function compared with common supplements. The second experiment evaluated ensiled sauvignon blanc GM component digestibility and the effect of physically transforming the GM by roller milling to disrupt seed coats and stem material. Thirty-two sheep were used in a nested design with four groups of eight sheep fed a basal diet (lucerne) with 0, 5, 10 and 30 % inclusion of GM. Groups fed GM were divided into two subgroups of four sheep fed either physically transformed or non-transformed GM. The experiment had three runs, with Run 1 as a control, and Runs 2 and 3 with sheep groups fed as per diet treatments. Sheep were kept 14 d in metabolism crates with the first 7 d as an adaption period and the next 7 d for collecting and sampling. Whole tract apparent dry matter digestibility (DMD %) and neutral detergent fibre digestibility (NDFD %) were significantly (P<0.001) reduced with increasing inclusion of GM in the diet, but no difference between physically transformed or non-transformed GM treatments was observed. The nitrogen (N) efficiency decreased with increasing inclusion of GM in the diet, with 30 % GM treatments significantly lower than control (P<0.05), with no difference between physically transformed and non-transformed subgroups. The whole tract apparent ether extract digestibility (EED %) increased with increasing GM inclusion in Run 1 (P=0.03) and Run 2 (P=0.04), but no difference between physically transformed and non-transformed GM treatments was observed. The urinary and faecal N output (g/d) were not affected by inclusion of GM (P>0.05), and the faecal N : urine N ratio was not different (P>0.05) but tended to be greater in physically transformed treatments at the 30 % inclusion. It was concluded this GM component digestibility decreased with increasing GM inclusion to 30 % of the diet, which may be due to the high lignin and NDF content or the effect of GM tannins and seed oil on NDF degradation. However, there was no observed feed value advantage with physically transformed GM treatments, so no practical use in transforming this GM when feeding sheep. This may be due to a more efficient grape seed coat disruption by chewing in sheep compared with cattle. The third experiment evaluated the rumen degradation of GM components of physically transformed and non-transformed GM for cattle, and the effect of greater rumen availability of seed tannin and oil content on this degradation. The effect of physically transformed GM compared to non-transformed GM was quantified in in sacco trials in a cross-over design to compare the DM, NDF, N and EE degradation rate. Six multiparous, non-lactating, ruminally fistulated cows were used in sequential runs (1 and 2) of 17 d each with a 30 d washout, where the cows in two fixed, random groups of three were fed a base forage at 8 kg DM plus either physically transformed or non-transformed GM (feed treatment) at 3 kg DM daily, then reversed in Run 2. The DM, NDF, N and EE disappearance was significantly (P=0.001) higher in physically transformed GM than non-transformed GM bag treatments. However, the feed treatments did not significantly affect DM, NDF, N and EE rumen degradation (P<0.05), suggesting there was no effect of any tannin or seed oil made rumen available with physical transformation on these component degradations. It was concluded physical transformation could improve feed value in cattle, where typically grape seed rupture by chewing may be poor, by increasing available oil and rumen degradation of NDF, and that the typical tannin and LCFA profile of this GM had no effect on rumen NDF degradation. This research has shown that New Zealand ensiled sauvignon blanc GM can be safely fed to lactating cows and to sheep as a supplement to a forage diet to 25 – 30 %, respectively, but even when physically transformed to disrupt the seed coat the feed value was restricted by a low NDF digestibility and N efficiency. These low effective digestibilities did not appear to be the result of this GM tannin or oil content, and may be due simply to a highly lignified seed and stem content. Future research requirements identified are: a primary investigation of NZ sauvignon blanc tannin content and composition across different operations and regions, before and after ensiling; the specific effects of these tannins in vitro on NDF and protein degradation; the assessment of the effects of physical transformation of this GM in lactating cows on milk fat LCFA PUFA content, and N partitioning; and the advantage in feed value of alternative processing methods to increase NDF degradation using heat, pressure and some hydroxide treatments with ensiled, sauvignon blanc GM of New Zealand.