Espectroscopia do infravermelho próximo (nirs) na estimativa da produção de metano entérico e composição química de forrageiras tropicais.

Abstract

Ruminants contributes to the production of food in the world and these herbivores, through enteric fermentation, produce methane, a potent greenhouse gas known for its high capacity to absorb infrared radiation The amount of methane produced by ruminants can be manipulated through the diets ingested by them, in the insertion of foods that promote changes in the fermentation profile in order to decrease methanogenesis. Therefore, it is essential to know the chemical composition, apparent digestibility of the foods used, as well as the estimated methane production that they may have. Near-infrared spectroscopy (NIRS) is an alternative technique to conventional methods for evaluating the nutritional value of foods, since it is a fast method, which does not require large structures and specific chemical solutions. The objective was to approach the use of NIRS technology in forages, and also to estimate the in vitro and in vivo enteric methane production by the NIRS technique. In the experiment I, the objective was to evaluate the in vitro digestibility and enteric methane production of two tropical forages in different years, investigating the possibility of using a calibration model to validate databases of different years and/or species. In the experiment II the aim of this study was to evaluate the digestibility and methane production of SF6 (sulfur hexafluoride) in samples of ruminal fluid, ruminal contents and moist and dried and ground feces in male sheep. The spectra of the representative samples of the experiments, being tropical forages in experiment I and feces, liquid and ruminal content in experiment II, were collected in the equipment NIRFlex N-500 FT-NIR Spectrometer Technical Datasheet Cell N500-001. Multivariate analyzes were performed through the NIRCal software, aiming an adequate adjustment to all the characteristics and consequent validation of the calibration curves developed with high robustness and, later, Pearson correlations between results from the laboratory and predicted in the NIRS. With the present studies, it was concluded that it is possible to obtain satisfactory calibration and validation parameters in most of the variables related to the chemical composition and digestibility of forage, so there is no need to develop different calibration curves per cultivar or year, because from a "global" curve it is possible to develop robust evaluation models, but in prediction equations of products generated during the IVDMD, such as methane gas, it is recommended to carry out more conventional laboratory analysis, aiming to increase accuracy of the models. The prediction of in vivo methane emitted, dry matter intake and digestibility through the analysis of wet or dry and ground faeces present high parameters and robust model, indicating that it can be done.