Monensina sódica, atendimento de exigência protéica via suplementos e predição do consumo de bovinos de corte em pastejo

Abstract

Beef cattle represents a fundamental part of Brazilian agribusiness, where the majority is represented by animals raised exclusively on pasture, with the need to search for technologies that make it more profitable and competitive. In this sense, this thesis was elaborated from three studies based on beef cattle raised in tropical pasture systems. In the first study, the objective was to evaluate the influence of protein-energy supplementation provided at the level of 1.5% of body weight, meeting different levels of protein requirement via supplement, with and without the inclusion of the additive sodium monensin. Five non-castrated Nellore cattle were used, with initial body weight (BW) of 260.0 ± 10 kg distributed in a 5 x 5 Latin square design, in a 2 x 2 + 1 factorial scheme, being: two levels of care of crude protein requirements via an energy protein supplement (70 and 100%); with or without inclusion of sodium monensia (100 mg / animal / day) and; mineral supplementation (negative control). There was no interaction between monensin and the level of crude protein served. Animals that received protein-energy supplementation showed lower consumption of DM from pasture (P <0.10), but higher consumption of total DM (P <0.10), than animals that received only mineral supplementation. The consumption of the other constituents of the diet was higher (P <0.10) when protein-energy supplementation was offered. The presence of monensin in supplements did not alter the consumption of any dietary constituents (P> 0.10), nor did it impact (P> 0.10) on the digestibility and dietary concentration of digestible organic matter (MOD), which were higher (P <0.10) when supplements were formulated to meet 100% of the crude protein dietary requirements. The animals that received a protein-energy supplement showed greater retention and efficiency in the use of dietary nitrogen (P <0.10) than the animals that consumed only mineral supplementation, which were not affected (P> 0.10) by the presence of monensin in the supplement. With the exception of excretion of fecal N (P> 0.10), greater compliance with the requirements of CP provides greater retention and efficiency of use of N (P <0.10). The second and third study aimed to develop and evaluate models for prediction of consumption of dry pasture (DMIpasture) and total (DMItotal) for beef cattle reared in tropical conditions using food supplementation. To build the models, 804 individual DMI observations were used, from 23 thesis or dissertation studies, carried out in Brazil, between 2006 and 2019, totaling 104 means of treatments. Three prediction models for DMIpasture and DMItotal were adjusted: Model I (animal-related variables), Model II (Animal variables plus food supplement) and Model III (Animal, supplement and pasture variables). The models proposed for DMIpasture showed no difference (P> 0.05), for the root of the mean square of the prediction error. The incorporation of predictive variables of the food supplement (Percentage of body weight offered and consumption of crude protein via supplement, kg / day) with animal variables (Average metabolic weight and body weight gain) Model II, did not improve the predictive capacity. The inclusion of variables associated with the animal, supplement and pasture (Crude protein) in Model III improved the accuracy and precision, providing a higher coefficient of determination and coefficient of correlation and agreement than the other models, also proved to be more accurate and precise than the one proposed by Azevedo et al., (2016) and Minson and McDonald (1987), which presented precision and accuracy inferior to the proposed Models I, II and III. Thus, it was concluded that the DMIpasture can be estimated with greater precision and accuracy with the Model III (Animal + Supplement + Pasture). For DMItotal the proposed Models showed similarity (P> 0.05), for the root of the mean square of the prediction error. The incorporation of predictive variables of the food supplement (Percentage of body weight offered and consumption of crude protein via supplement, kg / day) with the animal variables (Average metabolic weight and body weight gain) Model II, improved the predictive capacity of the model. The inclusion of variables associated with pasture (Crude protein) Model III did not improve the accuracy and precision. Among the Models proposed, Model II (Animal + Supplement) presented a low mean bias, in addition to a higher correlation and agreement coefficient than the other models, indicating that it is the most adequate model for predicting the DMItotal. In addition, it proved to be more accurate and accurate than the Models proposed by Azevedo et al., (2016), which had less predictive capacity, with a higher mean bias and low correlation and agreement coefficient. Thus, it was concluded that the DMItotal can be estimated with greater precision and accuracy with Model II (Animal + Supplement).