A good welfare protocol has to be representative over long‐term situations, which is referred to consistency over time. The aim of this study was to evaluate temporal consistency of behavioural indicators of dairy cattle welfare in the Welfare Quality protocol. This was done by visiting eight dairy farms for three times by two observers.
Animal welfare has more attention than ever before. Concerns of consumers about the welfare of livestock, have led to an urgent need for strategies to assess and improve animal welfare. The Welfare Quality protocol (WQ) is an example of a welfare assessment method, based on animal based indicators and aimed for a high validity, feasibility and reliability which is crucial for detecting the risk of poor welfare on farms. A good welfare protocol has to be representative over long‐term situations, which is referred to consistency over time. Several assessing methods are animal‐based and are already studied with regard to consistency over time. Behavioural indicators are not yet studied with regard to dairy cattle welfare and the expectations are that behaviour like head butts and displacements, rubbing and social licking have potential to be consistent over time. The other behaviours are uncommon in dairy cattle or their expression could be influenced by changes in the herd or housing conditions. So the aim of this study was to evaluate temporal consistency of behavioural indicators of dairy cattle welfare in the Welfare Quality protocol. This was done by visiting eight dairy farms for three times by two observers. Following the approach of the WQ protocol the observers measured on the same time, the same animals on continuous behavioural observations (head butts, displacements, chasing, chasing up, fighting, social licking, tongue rolling, rubbing, coughing and sneezing), lying down movements with potential collision percentages and herd scans. Besides this, resource‐checklists and interviews were done to obtain extra information about changes in management, herd and housing conditions.
In general no significant differences were found between the farm visits for the continuous behavioural observations (except for head butts). Correlations coefficients between farm visit 1 and 2 and between visit 1 and 3 show some similarity over time, but some indicators show low and moderate correlations between the farm visits which indicates no consistency. This variation in correlations coefficients could be explained due to large variations in the observations and too few observations of uncommon behavioural indicators, like fighting, chasing, chasing up, sneezing and tongue rolling. Also for time lying down movements a lot of variations were found which were evidenced by low correlation coefficients. Herd scans show the most similarity over time and is suggested to be consistent over time. To evaluate temporal consistency it was concluded that most behavioural indicators did not show significant differences over time. However, because of the small sample size and relatively low correlation coefficients, it could not be concluded that consistency over time was high. To obtain more reliable results for consistency over time, it is recommended to study these behavioural indicators in a larger sample size and observing behaviour by video recording to prevent observers from disturbing the observed herd.
Student: JH Oevermans
Supervisors: dr ir M de Vries, ing F Steenstra