Degree Name

Doctor of Philosophy


Department of Biological Sciences


The issue of animal welfare is one of increasing importance in many agricultural industries, particularly those based on the intensive management of livestock. While there have been investigations into the welfare status of poultry and pigs kept under intensive conditions, there have been no such studies for intensively managed beef cattle. This study was instigated to develop a set of welfare indices for the beef cattle feedlot industry. For indices of welfare to be effective they need to be based on a clear ethical framework, pertinent to a model for assessing welfare, scientifically credible and publicly acceptable, i.e., reasonably understandable to the wider community. Moral contractualism was chosen as an appropriate ethical framework for the study. Methods for assessing animal welfare can be divided into three approaches: (1) examinations of the human-animal interaction, (2) measurements of the mental state or feelings of the animal, and (3) measurements of the physical state of the animal. This study employed the third approach. In this approach, cattle were assumed to have reduced welfare if they were under a significant state of distress. This includes entering apre-pathological state. This is characterised by physiological changes such as elevated hypothalamus-adrenal-axis ( H P A ) activity in conjunction with depressed immune functions A pre-pathological state is defined in physiological terms, but may well produce behavioural changes. If these could be determined, they would have potential as an aid to the field diagnosis of animals with poor welfare in commercial feedlot situations. Because little is known about the the behaviour of feedlot cattle, three preliminary studies were performed which examined certain behaviours that might be pertinent to identifying distressed cattle The first two preliminary behavioural trials were conducted in a large commercial feedlot. These investigated lying behaviour and agonistic and afnliative behaviour. The results of the first trial showed that cattle have a consistent diurnal pattern of lying in commercial feedlot conditions, and that different background environments had no affect on this pattern. The results of the second trial showed that agonistic and afnliative encounters occur frequently in the first 21 days of feedlot life, that there is a clear structure of winners and losers, and a major emphasis on the "head push" as an agonistic behaviour. A third trial was run in an experimental feedlot enclosure. It developed the observational and statistical methodology for determining spatial relationships and grid square occupation for cattle in a feedlot environment. The results of this trial indicated that cattle do have preferences for certain areas in a feedlot. The experimental part of the study was then initiated. It measured the behavioural physiological responses of matched groups of cattle to three different environments: (1) pasture, this was regarded as a control, but not with behavioural observations, because group behaviour changes to accommodate different environments, (2) "normal" feedlot, this was characterised by a stocking density of 12 square metres/head and a firm, dry, pen substrate and (3), a "stressed" feedlot which had a stocking density of six square metres/head and a pen substrate which was constantly wet. Inclusive of the physiological responses were two physical measurements of adrenal glands and a panel of 20 immune variables, many of which had not been quantified before on cattle. T w o experimental trials were performed. The results of the experimental trial showed that there was a significantly higher adrenal mass and a higher adrenal cortical volume (strongly indicating increased H P A activity) in both feedlot groups compared to the pasture group. But only two out of the 20 immune variables: serum IgA, and the g a m m a delta T-lymphocyte W C 1 , showed decreased activity in the feedlot groups in comparison to the pasture group. It was concluded that support from other immune variables was required before it could be stated unequivocally that the immune system was compromised and thus that pre-pathological states existed in the feedlot groups. Because pre-pathological individual animals could not be identified, it was not possible to use the behavioural data for determining behavioural correlates. However, the study has provided new information for pasture and feedlot cattle ethograms. Some main findings are: (1) cattle have a very strong motivation for lying behaviour and will adjust their pattern of lying to accommodate chronic wet and crowded conditions, (2) despite high stocking densities, cattle are observed to be by themselves far more than with other animals within one steer's length of them, indicating that the use of a steer's length as a measure for determining nearest neighbours should be revised, (3) cattle in all treatments had preferences with regard to the occupation of grid space, (4) the backgrounds of groups of cattle affects their rates of agonistic behaviours. The conclusions of the main experimental trial are that, although pre-pathological states could not be clearly demonstrated, there were some physiological adjustments taking place to normal and "stressed" feedlot conditions. This suggests that if extra stressors, such as the social mixing of unfamiliar cattle, and/or bacterial or viral pathogens were to be introduced then pre-pathological states could occur, followed in some cases, by pathological states. Further work in this area would benefit by conducting more pre-treatment sampling to clearly establish the baseline for the immune parameters and to "focus" the immune system on a challenge (such as provided by vaccination) midway through the treatments. In conclusion, this study has determined on an ethical framework and a model for assessing welfare for beef cattle in feedlots. It has provided four physiological variables (relative adrenal weight, adrenal index, serum IgA and W C 1 lymphocytes) which are suitable candidates as physiological variables in further welfare studies. It has also provided n e w findings on the behaviour of feedlot cattle, which have potential for their management and for further welfare investigations.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.