Partitioning of nutrient variation in alfalfa and corn silage by source on New York dairy farms.

Variation in feed components contributes to variation and uncertainty of diets delivered to dairy cows. Forages often have a high inclusion rate (50% to 70% of DM fed) and variable composition, and thus are an important contributor to nutrient variability of delivered diets. Our objective was to quantify the variation and identify the main sources of variability in corn silage and alfalfa-grass haylage composition at harvest (fresh forage) and feed-out (fermented forage) on New York dairy farms. Corn silage and alfalfa-grass haylage were sampled on 8 New York commercial dairy farms during harvest in the summer and fall of 2020 and during their subsequent feed-out in the winter and spring of 2021. At harvest, a composite sample of fresh chopped forage of every 8-ha section of individual fields was collected from piles delivered for silo filling. During a 16-wk feed-out period, 2 independent samples of each forage were collected 3 times per week. The fields of origin of each forage sample during feed-out were identified and recorded using silo maps created at filling. A mixed-model analysis quantified the variance of corn silage DM, NDF, and starch and haylage DM, NDF, and CP content. Fixed effects included soil type, weather conditions, and management practices during harvest and feed-out, and random effects were farm, silo unit, field, and day. At harvest, between-farm variability was the largest source of variation for both corn silage and haylage, but within-farm sources of variation exceeded farm-to-farm variation for haylage at feed-out. At feed-out, haylage DM and NDF content had higher within-farm variability than corn silage. In contrast, corn silage starch showed higher within-farm variation at feed-out than haylage CP content. For DM content at feed-out, day-to-day variation was the most relevant source of within-farm variation for both forages. However, for the nutrient components at feed-out (NDF and CP for haylage; NDF and starch for corn silage) silo-to-silo variation was the largest source of variability. Weather conditions systematically explained a proportion of the farm-to-farm variability for both forages at harvest and feed-out. We concluded that because of the high farm-to-farm variation, corn silage and haylage must be sampled on individual farms. We also concluded that due to the high silo-to-silo variability, and the still significant day-to-day and field-to-field variability within-farm, corn silage and haylage should be sampled within individual silos to better capture changes in forage components at feed-out.