Comportamento de proteção materna em vacas Girolando: métodos de avaliação e suas relações com a reatividade ao manejo e produção de leite
Carregando...
Arquivos
Data
2018-09-26
Autores
Góis, Karen Camille Rocha [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Os objetivos com esta tese foram testar o uso de métodos com escores predefinidos e da avaliação qualitativa do comportamento (QBA) como potenciais indicadores do comportamento de proteção materna de vacas Girolando, bem como avaliar a relação entre a reatividade ao manejo, o comportamento de proteção materna e a produção de leite. Foram registrados em vídeos os comportamentos de 88 vacas recém paridas, durante a primeira aproximação do manejador à vaca com bezerro, para o manejo de desinfecção do umbigo e identificação do bezerro. Foram adaptados dois métodos baseados em escalas com escores visuais pré-definidos, o escore composto de proteção materna (CPM) e o escore de proteção materna (EPM), além da adaptação da QBA. Para avaliar a confiabilidade dos métodos, foi utilizada uma equipe de observadores (n=10) para testar o CPM e o EPM e outra equipe (n=8) para testar a QBA. Para avaliar a reatividade ao manejo, foram utilizados: escore composto de reatividade no tronco de contenção (REA), escore de reatividade na preparação para a ordenha (REP) e escore de reatividade na colocação das teteiras (REC). Foi realizada a mensuração dos picos (P) e médias (M) de produção de leite no primeiro (1) e segundo (2) mês pós-parto de 60 vacas recém-paridas pertencentes a dois grupos genéticos, ¾HG= 3/4Holandês+1/4Gir (n=36) e ½HG= 1/2Holandês+1/2Gir (n=24). As confiabilidades inter e intraobservador de CPM e EPM foram calculadas pelo coeficiente Kappa (k). Utilizando a análise de componentes principais foi possível identificar três componentes principais (CP) da QBA. As confiabilidades inter e intraobservador dos CP foram analisadas calculando os coeficientes de Kendall (W). Para avaliar se houve diferença nas médias de ordem dos CP em função de CPM e EPM, foi utilizado o teste de Kruskal-Wallis seguido do teste “post hoc” de Dunn. Coeficientes de correlação de Spearman ponderados foram estimados para avaliar a associação entre os métodos de avaliação do comportamento. Foi avaliado o efeito das variáveis comportamentais das vacas sobre a produção de leite utilizando o PROC GLIMMIX do SAS®. As confiabilidades intraobservador variaram de 0,50 a 1,00 para CPM e de 0,50 a 0,87 para EPM. Por sua vez, as confiabilidades interobservador variaram de 0,34 a 1,00 para CPM e de 0,55 a 0,89 para EPM. A confiabilidade intraobservador do QBA foi muito forte para os oito observadores no CP1 (W=0,71 a 0,89) e CP2 (W=0,74 a 0,90). No CP3, variou de forte a muito forte (W=0,63 e 0,84). A confiabilidade interobservador nas três avaliações variou de forte a muito forte em CP1 (W=0,70 a 0,73), CP2 (W=0,71 a 0,77) e CP3 (W=0,55 a 0,73). Houve uma relação entre os métodos com escore (EPM e CPM) e os dois componentes principais extraídos da QBA (CP1 e CP2). Houve efeito de EPM e CP1 sobre M1 e P1, com produções mais baixas para as vacas categorizadas como “alto”. Concluímos que CPM, EPM e QBA se mostraram métodos válidos e confiáveis para a avaliação do comportamento de proteção materna de vacas Girolando. Vacas que protegeram fortemente suas crias podem gerar diversas implicações de manejo, oferecendo maior risco aos trabalhadores, além de apresentarem menor produção de leite nos primeiros dois meses pós-parto.
The objectives of this thesis were to test the use of methods with predefined scores and qualitative behavior assessment (QBA) as potential indicators of the protective maternal behavior of Girolando cows, as well as to evaluate the relationship between reactivity, protective maternal behavior and milk production. Behaviors of 88 newly calved cows were recorded in videos during the first handling of their calves for navel disinfection and identification. Two methods based on scales with predefined visual scores were used, the protective maternal composite score (CPM), and the protective maternal score (EPM), and the adaptation of QBA was performed. To evaluate the reliability of the methods, one observer’s team (n=10) were used to test the CPM and EPM and another team (n=8) to test the QBA. In order to evaluate the cow reactivity, the reactivity score inside the cattle crush (REA), the reactivity score during the milking preparation (REP) and the reactivity score during the teat cups placement (REC) were used. Measurements of peaks (P) and averages (M) milk yield in the first (1) and second (2) months postpartum were performed on 60 cows of two genetic groups, ¾ HG = 3/4 Holstein + 1/4 Gyr (n = 36) and ½HG = 1/2 Holstein + 1/2Gyr (n = 24). The inter and intraobserver reliability of CPM and EPM were calculated by Kappa coefficient (k). Using the Principal component analysis, it was possible to identify three principal components (CP) in the QBA. The inter and intraobserver reliability of CP was analyzed by calculating the Kendall coefficients (W). The Kruskal-Wallis test with the Dunn “post hoc” test was used to evaluate if CP1 and CP2 were related to CPM and EPM. Weighted spearman's correlation coefficients were estimated to evaluate the association among the three behavioral methods used. The effect of cows’ behavioral variables on milk yield was evaluated using the PROC GLIMMIX of SAS®. The intraobserver reliability for CPM ranged from 0.50 to 1.00, while EPM ranged from 0.50 to 0.87. The CPM interobserver reliability ranged from 0.34 to 1.00, while for EPM ranged from 0.55 to 0.89. The QBA intraobserver reliability was very strong for eight observers in CP1 (W=0.71 to 0.89) and CP2 (W=0.74 to 0.90). In the CP3, it ranged from strong to very strong (W=0.63 to 0.84). The interobserver reliability in the three evaluations ranged from strong to very strong in CP1 (W=0,70 a 0,73), CP2 (W=0,71 a 0,77) and CP3 (W=0,55 a 0,73). There was a relationship between the scoring methods (EPM and CPM) and the two principals components extracted from QBA (CP1 and CP2). There was an effect of EPM and CP1 on M1 and P1, with lower milk yields for cows categorized as “high”. In conclusion, CPM, EPM and QBA were shown to be applicable and reliable methods to evaluate the protective maternal behavior of Girolando cows. Cows that strongly protect their offspring have different handling implications, promove greater risk to the workers, as well as lower milk production in the first two months postpartum.
The objectives of this thesis were to test the use of methods with predefined scores and qualitative behavior assessment (QBA) as potential indicators of the protective maternal behavior of Girolando cows, as well as to evaluate the relationship between reactivity, protective maternal behavior and milk production. Behaviors of 88 newly calved cows were recorded in videos during the first handling of their calves for navel disinfection and identification. Two methods based on scales with predefined visual scores were used, the protective maternal composite score (CPM), and the protective maternal score (EPM), and the adaptation of QBA was performed. To evaluate the reliability of the methods, one observer’s team (n=10) were used to test the CPM and EPM and another team (n=8) to test the QBA. In order to evaluate the cow reactivity, the reactivity score inside the cattle crush (REA), the reactivity score during the milking preparation (REP) and the reactivity score during the teat cups placement (REC) were used. Measurements of peaks (P) and averages (M) milk yield in the first (1) and second (2) months postpartum were performed on 60 cows of two genetic groups, ¾ HG = 3/4 Holstein + 1/4 Gyr (n = 36) and ½HG = 1/2 Holstein + 1/2Gyr (n = 24). The inter and intraobserver reliability of CPM and EPM were calculated by Kappa coefficient (k). Using the Principal component analysis, it was possible to identify three principal components (CP) in the QBA. The inter and intraobserver reliability of CP was analyzed by calculating the Kendall coefficients (W). The Kruskal-Wallis test with the Dunn “post hoc” test was used to evaluate if CP1 and CP2 were related to CPM and EPM. Weighted spearman's correlation coefficients were estimated to evaluate the association among the three behavioral methods used. The effect of cows’ behavioral variables on milk yield was evaluated using the PROC GLIMMIX of SAS®. The intraobserver reliability for CPM ranged from 0.50 to 1.00, while EPM ranged from 0.50 to 0.87. The CPM interobserver reliability ranged from 0.34 to 1.00, while for EPM ranged from 0.55 to 0.89. The QBA intraobserver reliability was very strong for eight observers in CP1 (W=0.71 to 0.89) and CP2 (W=0.74 to 0.90). In the CP3, it ranged from strong to very strong (W=0.63 to 0.84). The interobserver reliability in the three evaluations ranged from strong to very strong in CP1 (W=0,70 a 0,73), CP2 (W=0,71 a 0,77) and CP3 (W=0,55 a 0,73). There was a relationship between the scoring methods (EPM and CPM) and the two principals components extracted from QBA (CP1 and CP2). There was an effect of EPM and CP1 on M1 and P1, with lower milk yields for cows categorized as “high”. In conclusion, CPM, EPM and QBA were shown to be applicable and reliable methods to evaluate the protective maternal behavior of Girolando cows. Cows that strongly protect their offspring have different handling implications, promove greater risk to the workers, as well as lower milk production in the first two months postpartum.
Descrição
Palavras-chave
Avaliação qualitativa, Defesa materna, Comportamento, Maternal defense, Qualitative behavior