Study of the association of atmospheric temperature and relative humidity with bulk tank milk somatic cell count in dairy herds using Generalized additive mixed models

作     者：Testa, Francesco Marano, Giuseppe Ambrogi, Federico Boracchi, Patrizia Casula, Antonio Biganzoli, Elia Moroni, Paolo 

作者机构：Univ Milan Dipartimento Sci Vet Salute Prod Anim & Sicurezza Via Celoria 10 I-20133 Milan Italy Univ Milan Dept Clin Sci & Community Hlth Via Vanzetti 5 I-20133 Milan Italy Univ Milan Dipartimento Med Vet Via Celoria 10 I-20133 Milan Italy Fdn IRCCS Ist Nazl Tumori Unit Med Stat & Biometry Via Venezian 1 I-20133 Milan Italy Cornell Univ Anim Hlth Diagnost Ctr Qual Milk Prod Serv 240 Farrier Rd Ithaca NY 14853 USA 

出 版 物：《RESEARCH IN VETERINARY SCIENCE》 (兽医科学研究)

年 卷 期：2017年第114卷

页      面：511-517页

核心收录：

学科分类：0906[农学-兽医学] 09[农学] 

主　　题：Bovine Somatic cell count Temperature Humidity Index Generalized Additive Models 

摘      要：Elevated bulk tank milk somatic cell count (BMSCC) has a negative impact on milk production, milk quality, and animal health. Seasonal increases in herd level somatic cell count (SCC) are commonly associated with elevated environmental temperature and humidity. The Temperature Humidity Index (THI) has been developed to measure general environmental stress in dairy cattle;however, additional work is needed to determine a specific effect of the heat stress index on herd-level SCC. Generalized Additive Model methods were used for a flexible exploration of the relationships between daily temperature, relative humidity, and bulk milk somatic cell count. The data consist of BMSCC and meteorological recordings collected between March 2009 and October 2011 of 10 dairy farms. The results indicate that, an average increase of 0.16% of BMSCC is expected for an increase of 1 degrees C degree of temperature. A complex relationship was found for relative humidity. For example, increase of 0.099%, 0.037% and 0.020% are expected in correspondence to an increase of relative humidity from 50% to 51%, 80% to 81%;and 90% to 91%, respectively. Using this model, it will be possible to provide evidence-based advice to dairy farmers for the use of THI control charts created on the basis of our statistical model.