Monitoramento da nutrição de pequenos ruminantes na caatinga cearense, utilizando a espectroscopia NIR

• Ano de defesa: 2018
• Autor(a) principal: Gonçalves, Juliete de Lima
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Zootecnia; Programa de Pós-Graduação em Zootecnia; UFPB
• Palavras-chave em Português: Análise multivariada; Caprinos; Massa de forragem; Ovinos; CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA::NUTRICAO E ALIMENTACAO ANIMAL
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/14260

Brazil has great potential for pasture farming, due to the territorial expansion and the diversity of vegetation. Therefore, it is necessary to know the quality of the selected diet and the consumption by small ruminants in grazing to make the correction of the feed and thus improve the productive indexes. The objective of this study was to develop models with the NIRS technique, so that they can be used to monitor the quality of the diet of small ruminants in the pasture of caatinga. The nutrient intake and digestibility, as well as the pasture evaluation, were also evaluated in the different strata over the evaluated periods. The experiment was conducted at the Família Agrícola Dom Fragoso School, in Independência, CE, in an area of 35 ha of pasture of caatinga, where an evaluation of pasture, diet quality and consumption of goats and sheep was carried out in May 2014 to April 2015. For the evaluation of the botanical composition, phytosociological parameters, soil and plant cover, forage mass, definitions of visual patterns and evaluation of the bromatological composition of the different strata, the experimental area was divided into transects, totaling 246 evaluated points. It was identified 81 species, distributed in 32 families, showing the diversity of plants present in the area. The climatic conditions, mainly related to rainfall distribution, in the different periods and strata evaluated affected the availability of forage mass, soil and plant cover, frequency of forage species of the herbaceous stratum, as well as the nutritional value of the pasture, obtaining better results in the rainy period. The indices for monitoring the native pasture of caatinga bring relevant information from the pasture area and, thus, constitute a tool that contributes to the better management of the forage resources. To determine the bromatological composition of the selected diet and the consumption of small ruminants, five goats and five sheep were used fistulated in the rumen, where the extrusa was collected, for six days, monthly, throughout the experimental period. Fecal collection bags were used to collect fecal samples, and the dry matter intake was obtained by the ratio between fecal production during a 24 hour period and the inverse of the in vitro digestibility of the dry matter of the extrusa. From the dry matter consumption the other nutrients were calculated, multiplying the amount of dry matter consumed by the percentage of each of the nutrients of the extrusa. The quality of the goats and sheep diet was affected by the evaluated period, which reflected in nutrient consumption and digestibility. The dry matter intake xvii influenced the consumption of other nutrients in both species, goats and sheep, and periods. The in vitro digestibility of the dry matter and in vitro digestibility of the organic matter of the sheep diet were superior to the goats, but they were able to maintain the body weight. The high levels of crude protein in the diet of goats and sheep in the rainy season are linked in part to the fraction of acid detergent fiber, leaving it unavailable to the animal. The Perten® NIR equipment DA 7250 was used to collect spectra of fresh and dry feces samples without grinding, which makes a reading in the spectral range from 950 to 1650 nm, with a spectral range of 5 nm. In order to obtain spectra of pre-dried and milled samples, two different models of NIR devices, Perten® DA 7250 and FOSS® 5000 NIR were used, with a reading in a range of 1100 to 2500 nm in the near infrared region and resolution of 2 nm. 711 samples of feces were scanned, 336 samples of goats and 375 of sheep, where models were developed with composite and individual samples, using different mathematical treatments, exploratory analysis and the use of PLS regression for the calibration models, validation and prediction of the parameters of crude protein (CP) and in vitro digestibility of organic matter (IVDMO), with fecal samples in different processes. NIR spectroscopy, combined with chemometrics techniques, allowed the construction of calibration, validation and prediction models able of quantifying the parameters of CP and IVDMO of the diet of goats and sheep on grazing in the caatinga, with precision and accuracy, and the models with individual samples present greater prediction and precision errors than models with composite samples. The use of PCA and SIMCA showed that it would be possible to include both species, sheep and goats, periods (rainy, rainy/dry transition, dry, dry/rainy transition) and variation of PB and IVDMO parameters in the development of a global model. The types of processing of the samples generated different spectra and showed that the drying and grinding procedures can improve the performance of the models, as well as the use of the tool of selection of the wavelengths of more important waves. The NIRS technology can be used as a tool to monitor the diet of small ruminants in caatinga rangeland.