Untersuchungen zur Erfassung und Genetik von Verhaltensmerkmalen beim Schwein unter Praxisbedingungen

Abstract

Public and political concern for livestock increases and due to new laws housing systems of pigs will change in the future. It is desired to increase well-being of animals, for example, by group housing of pigs at the different stages. However, beside the positive effects for animal welfare group housing of animals also leads to agonistic interactions within the group resulting in stress. Furthermore, due to structural changes herd sizes increase and the handleability of pigs gains importance under these conditions to reduce levels of stress at handling for stockmen and animals. The breeding of less-aggressive and calm pigs might be a solution for these problems. The selection for handleability and aggressiveness in breeding programmes requires methods for recording traits related to handleability and aggressiveness under commercial farm conditions to collect large sample sizes. Furthermore, associations with other relevant traits have to be investigated. The aim of the present study was the development of behaviour tests and checking the ability of these tests for use in breeding programmes. Especially the feasibility of the tests under commercial conditions for recording large sample sizes was tested. First of all, the environment effect of housing was investigated. Therefore, the behaviour of 126 rearing pigs housed in different rearing housing systems (10 pigs per pen, dry feeder vs. 30 pigs per pen, dry feeder and wet dry feeder) were compared using different behaviour tests (novel arena test, novel object test, weighing behaviour) as well as skin lesions to show effects of marginal environmental enrichment on behaviour. Behaviour patterns in the novel arena test were not different between pigs reared in the different environments, but the enriched housed pigs showed significantly more activity, for example, on the scale (p < 0.0001) and less skin lesions (p = 0,0074) compared to barren housed animals (chapter 3). The use of skin lesions as an indicator for individual aggressiveness was investigated. Therefore, associations among agonistic behaviour of sows post mixing and skin lesions recorded 10 weeks post mixing were analysed. Our results showed that there was no association between the frequency of being aggressor of agonistic interactions post mixing and skin lesions (p > 0.1). However, sows being receivers of agonistic interactions had more skin lesions (p = 0.0435). Number and severity of skin lesions seem not to be useful as an indicator for individual aggressiveness. Furthermore, more aggressive sows had better reproductive performances (e.g. being aggressor of agonistic interaction vs. total born piglets: rp = 0.20) and sows with higher scores for skin lesions had reduced reproductive performances (skin lesion scores front vs. total born piglets: rp = –0.28). An unfavourable association between aggressive behaviour and reproduction is shown (chapter 4). Finally, associations among backtest, weighing behaviour, skin lesions, and daily gains were analysed. Therefore, genetic parameters were estimated as well as phenotypic correlations were calculated. Behaviour observations from birth to slaughtering of 976 fattening pigs were recorded. The estimated heritabilities were in the most cases useful for breeding purposes (e.g. backtest 2: h² = 0.36 ± 0.08; finishing pig scale score: h² = 0.20 ± 0.07). However, associations were only found between backtest 1 and backtest 2 (rp = 0.30; rg = 0.84 ± 0.11) and between backtest 2 and rearing pigs scale score (rp = 0.13; rg = 0.57 ± 0.21). Due to comparably weak associations towards weighing behaviour, the lack of further associations towards other behaviour traits as well as the high labour and time input for the procedure, the backtest seems not to be useful for breeding issues. There were also no further relationships between the other behaviour traits based on the overall low genetic and phenotypic correlations. The individual reaction of animals appears to be strongly affected by the situation (test) and there was little consistency across contexts in behaviour of pigs. Regarding the results of the present study (e.g. heritabilities), the integration of behaviour traits into breeding programmes is overall possible. First of all, the selection of behaviour traits related to aggressiveness or handleability is important. Backtest as well as skin lesions are not related to aggressiveness or only marginally to handleability and therefore are not useful as indicators for breeding calm and less-aggressive pigs. For example, weighing behaviour or direct observations of agonistic interactions seem to be more appropriate. However, positive correlations among aggressions and weight gain as well as reproductive performance show unfavourable associations towards economically important traits.