ABSTRACT This study aimed at identifying the best gender division pattern for swine at nursery stage, through the evaluation of behavioral and environmental parameters. Three treatments were established to achieve our objective: single-sex housing of 26 surgically castrated male pigs; mixed-sex housing of 13 castrated males and 13 females; and single-sex housing of 26 females. Environmental and behavioral data have been recorded. It was used an ethogram to make an inventory of behaviors and that was used in multivariate analysis. Bite addiction differed statistically among treatments, and the lowest occurrence was in mixed-sex housing. Results of the multivariate analysis suggested that regardless of the treatment chosen, relative humidity is positively associated with the production of gases and dry-bulb temperature with airspeed. Mixed-sex housing had multivariate relations suggesting less aggressive behaviors over food resources, being an example of social interaction that improves swine welfare.

ABSTRACT Using geostatistics, this study aimed to analyze environmental conditions for wean-to-finish swines at 3 pm, during the most sensitive phases to thermal stress (growth/ finishing and finishing). To this end, dry-bulb temperature, relative humidity, air speed, ammonia and carbon dioxide concentrations were measured at 81 equidistant points within the barn. Descriptive statistics was used to classify attribute variability, generating mean, median, variation coefficient, asymmetry and kurtosis, through the normality of Kolmogorov-Smirnov (KS). For that, it was used Minitab 15 software. Moreover, semivariograms were used to assess spatial dependence using the GS+ software, and through Surfer, spatial distribution maps were designed. Dry-bulb temperature and relative humidity showed a more homogeneous distribution whether compared to the other attributes. Furthermore, critical points were observed within the facility, with temperatures and low humidity, indicating the need for fogging equipment and fans to improve thermal comfort and animal welfare.

RESUMO O “wean to finish” é um sistema de criação de suínos, no qual os animais permanecem em uma mesma instalação do desmame até à fase de abate. Objetivou-se estudar o comportamento dos animais em alojamento exclusivo de 26 machos suínos (T1), misto (T2), contendo 13 machos e 13 fêmeas, e o tratamento único de fêmeas (T3), com 26 animais. Para isso, utilizou-se da análise multivariada, por meio da análise dos gráficos de componentes principais, por ser um artifício interessante utilizado por diversos pesquisadores para identificar quais variáveis são fortemente ou não correlatas. Assim, indica-se que o aumento da temperatura ambiental predispõe ao comportamento deitado em todos os tratamentos avaliados. No tratamento exclusivo de suínos machos castrados cirurgicamente, o comportamento de monta relaciona-se com vício, já o agrupamento misto, monta associa-se com lúdico. Todavia, fêmeas suínas alojadas unicamente não tendem a praticar o comportamento de monta.

ABSTRACT The "wean to finish" is a pig housing system, in which animals remains in the same building from the weaning to slaughter. The objective was to study the behavior of animals in unique accommodation of 26 male pigs (T1), mixed (T2), containing 13 males and 13 females, and the only treatment of females (T3) with 26 animals. For this, we used multivariate analysis, by analyzing the principal components, as an interesting tool used by many researchers to identify which variables are strongly correlated or not. Thus, we observed the main trends and characteristics in each treatment. The increase in air temperature has caused the behavior of lying down in every treatment. Only in the male treatment, the mount behavior is related to addiction, however in the mixed treatment the mount is related to a playful behavior. Only female pigs’ treatment did not show any mount behavior.

O "wean to finish" é um sistema de criação de suínos que ainda é novo no Brasil, e não existem muitos estudos a respeito do ambiente onde estes animais são criados. Assim, o objetivo deste trabalho foi avaliar o ambiente térmico e aéreo por meio da técnica da geoestatística em galpão de suínos criados no sistema "wean to finish", na fase de terminação, nos horários das 09h, 12h e 15h. A variabilidade dos atributos foi baseada na estatística descritiva dos dados de temperatura, umidade relativa, velocidade do ar, concentrações de amônia e dióxido de carbono, obtendo-se: média, mediana, coeficiente de variação, assimetria e curtose, com o teste de normalidade de Kolmogorov-Smirnov, através do programa estatístico Minitab 15. Para a determinação da existência ou não da dependência espacial, utilizou-se do exame de semivariograma, através do programa GS+, e, na elaboração dos mapas de distribuição espacial das variáveis, foi utilizado o programa Surfer. Os atributos de velocidade do ar e as concentrações dos gases NH3 e CO2 apresentaram alto coeficiente de variação devido aos dados serem heterogêneos, pois a instalação apresentava ventilação natural. O uso dos mapas de krigagem permitiu observar a diferença espacial da distribuição das variáveis avaliadas com diferentes números de animais/baia e em diferentes horários.

The "wean to finish" is a new pig production system in Brazil. To date, few research have investigated the environmental conditions of this pig farming system. The aim of this study was to evaluated the thermal environmental conditions and gases concentration using geostatistical techniques of a wean-to-finish barn during finishing phase at 9AM, 12PM e 3PM. The data variability was evaluated by the descriptive analysis, of the temperatures, relative humidity, air velocity, ammonia and carbon dioxide concentration. The results showed the average, median, coefficient of variation, asymmetry and curtosis and the Kolmogorov-Smirnov Test with the software Minitab 15. To determine the spatial dependence, the semivariogram exam was done, with the software GS+. The air velocity and gases concentration showed the highest coefficient of variation. This has happened because of the heterogeneity of the data caused by the natural ventilation. The Krigging maps showed the spacial differences of the pens with different numbers of animals in different hour of the day.