This dissertation is comprised of six chapters which include a literature review (chapter 1), four experimental chapters (chapters 2, 3, 4 and 5) and a future research chapter (chapter 6). In chapter 2, measures of efficiency of Holstein and Jersey cows were assessed when fed alfalfa silage or corn silage in low or high forage fiber diets. Breed did not affect digestive and metabolic efficiencies. In contrast, methane and urinary energy (% gross energy intake) were lower for corn silage than alfalfa silage-fed cows; and compared to high, low forage fiber diets reduced loss of urinary N (g/d and % N intake). Neither breed nor dietary treatments affected methane intensity (g/kg fat-protein corrected milk). In chapter 3, the carry-over effects of same three treatment factors (cow breed, dietary forage source and forage level) on manure greenhouse gas (GHG) emissions (methane and nitrous oxide) during 50-d storage and followed by a 50-d field application were evaluated. Compared to high, low forage-fed cows tended to emit 51 to 72% (depending on mode of expressions) greater combined (storage plus field) GHG emissions which were not affected by cow breed and forage source. In chapter 4, we evaluated the carry-over effects of the same treatment factors on manure ammonia emissions. Compared to high, low forage fed cows emitted less ammonia expressed as per cow, per kg manure or percentage of manure N. Although, forage source did not affect ammonia emissions, cow breed did impact ammonia emissions expressed per cow being 17% greater for Holstein than Jersey. In chapter 5, we performed a cradle-to-gate life cycle assessment to determine the carbon footprint (CF) of milk for the same treatment factors using emission factors measured in our studies. Low forage-fed cows had 11% greater CF than high forage-fed cows whereas both forage sources and cow breeds (Holstein and Jersey) had similar CF. We concluded that GHG mitigation strategies (choice of cow breed or diet) need to be evaluated holistically using measurements specific to the production system under consideration since evaluations at the whole-farm scale led to different results than when completed at the animal scale.