Improved utilisation of grass Silage in milk production

The aim of the studies reported in this thesis was to optimise grass silage utilisation in milk production.  In the first paper of this thesis data from digestibility trials with sheep fed at maintenance level were used to investigate the effect of forage type and harvesting time of silage on energy and N metabolism. The assumption of constant loss of energy of digestible organic matter from energy losses in urine and CH4 in evaluation of silage metabolisable energy (ME) did not seem valid for highly-digestible primary growth grass silages. Urinary excretion of high-energy phenolic acids related to solubilisation of lignin could not be related to increased urinary energy excretion from sheep fed highly digestible grass silage. In the second paper a concentrate made from agro-industrial by-products was used in comparison with a conventional grain-based concentrate to dairy cows. The concentrates were fed with early or late harvested grass silage. A concentrate made of by-products could replace a conventional grain-based concentrate without impairing milk production, diet digestibility and CH4 emissions. Feeding agro-industrial by-product with highly digestible grass silage increased edible feed conversion ratios. Silage digestibility had a stronger effect on production performance by dairy cows than the source of concentrate. In the third paper the effect of strategy for harvesting regrowth (RG) grass silage on performance in dairy cows was studied. Increasing maturity stage in both primary growth (PG) and RG silage decreased intake potential, dairy performance and energy utilisation of lactating dairy cows, but improved N utilisation efficiency and herbage dry matter (DM) yield. Feeding more digestible silages from three-cut system promoted better dairy performance and energy utilisation compared with silages from two-cut systems. In the fourth paper data of intake and milk yield responses with different grass ley harvesting strategies of dairy cows were combined in a meta-analysis. The average increases in milk yields was 0.303 kg d-1 per 10 g kg-1 DM increase in silage digestible organic matter concentration (D-value). But each 10-unit increase in D-value reduced milk yield 0.092 kg when dietary ME intake was the same, suggesting the ME concentration of high D-value silages was overestimated. Cows fed RG compared to PG silage produced 0.55 kg d-1 more energy corrected milk when dietary ME intake was the same, suggesting more efficient digestion of RG silage. Overall, results from this thesis can be used to improve ration formulation and grassland management, or in economic models optimising milk production.