Does Replacing Maize with Barley Affect the Animal Performance and Rumen Fermentation, including Methane Production, of Beef Cattle Fed High-Concentrate Diets On-Farm?

Ruminants fed high-concentrate diets produce less enteric methane than those fed high-forage diets, but not all grains are equally effective in reducing methane production. This study aimed to examine, in farm conditions, the effects of a partial substitution of maize with barley on animal performance and rumen fermentation, including methane production, of intensively reared beef calves (ca. 0.9:0.1 concentrate to forage ratio). Ninety-six beef calves were fed a concentrate with 45.5% maize and 15% barley (n = 48; M) or a concentrate with 15.5% maize and 45% barley (n = 48; B). Both the concentrate and barley straw were offered ad libitum. The type of concentrate did not have a significant effect (p > 0.05) on final live weight, average daily gain, carcass dressing percentage or intake of concentrate and straw. Dry matter and organic matter digestibility were higher (p < 0.05) for the M (75.4% and 76.6%) than for the B (71.0% and 73.1%) treatment, but with no effect on digestible organic matter intake. In general, the majority cereal in the concentrate did not affect rumen fermentation, including methane production, or the degradability of dry matter and starch. A partial substitution of maize with barley in the concentrate offered to beef calves does not seem a promising strategy to decrease the emissions of enteric methane on-farm.

326 Nutritional Grouping and Machine-Learning Techniques: Towards a Feed Efficiency Improvement in Beef Cattle Production

Nutritional grouping (NG) strategies have been proposed mainly in dairy cattle to improve economic efficiency (30-60$/cow per year) of dairy farmers. In dairy cattle, NG is made according to net energy (NE) requirements estimated from some animal variables (i.e., milk yield and dairy merit). However, NG has not been evaluated for beef cattle, likely because data are not readily available and existing protocols have been fixed for decades by the beef industry, such as allotting by body weight (BW): growing diet for 180 to 400-kg animals, and finishing diet for 400 to 550-kg animals. Therefore, we investigated an unsupervised machine learning (ML) technique in R to sort animals at the arrival according to their DM intake (DMI), NE and metabolizable protein (MP) requirements, estimated through INRA from BW and average daily gain. We utilized data from 77 growing Friesian steers allotted conventionally by BW (139±21.6 kg) in 4 pens, weighed every 3 weeks (n=9 weighings) during 190 days, and fed a high-energy diet (1.59 Mcal NE/kg DM and 89 g MP/kg DM). Results identified that estimated performances were significantly (P&lt; 0.002) different between 3 groups made from ML and between 4 pens made from BW; however, ML groups presented a minor within-group variability of NE and MP requirements (-25%CV) than pens. We also regressed energy and protein intake minus their respective requirements by the duration of the experiment observing a better fitting (R2= 0.93 vs 0.80) when allotting animals by groups than by pens, which implicates better feed efficiency if diets were formulated according to requirements established by ML groups. We demonstrated that ML techniques could help the decision-making of beef farmers to better sort animals according to their requirements and not only by BW. The next step is to formulate rations according to animal requirements to improve efficiency.

Collapse

[Yeast diversity in Bulnesia retama and Larrea divaricata canopies and associated soils]

Bush like vegetation dominates arid environments, and there is nutrients accumulation under shrub canopies and relatively unfertile soils between vegetal patches areas. Plants are one of the most common habitats for yeasts. There are many reports about yeasts inhabiting different plant components. Nevertheless, there are no reports about yeasts associated with Zygophyllaceae, an important shrub family of the Argentinean Province of Monte. The objective of this work was to analyzed yeast biodiversity of Bulnesia retama and Larrea divaricata canopies and associated soils, at Medanos Grandes of Caucete, San Juan, Argentina. Eighty seven (87) isolated yeasts were identified. From B. retama canopy and associated soil was observed a larger taxonomical diversity respect to L. divaricata. Nine (9) and ten (10) species were isolated from canopy and associated soil of B. retama, respectively. From L. divaricata canopy were 4 species and 3 species from its associated soil isolated. Identified genera were: Candida, Debaryomyces, Dekkera, Saccharomyces, Torulaspora, Sporidiobolus and Pichia. Fourteen (14) species were found at all microenvironments.