Genotypes differ in their response to variation in environments experienced by pigs on farm

Both animals and plants have the ability to respond to changes in their environment, which is called environmental sensitivity or phenotypic plasticity (Bradshaw, 1965). Different genotypes do not respond in the same way to different environments leading to genotype by environment interactions (GxE). There are two types of GxE: scaling effects or re-ranking for individuals. As different scaling effects for different traits can also results in re-ranking of genotypes, both are important for animal breeding and selection (Strandberg, 2006). Two traditional methods to analyse GxE are the interaction-term model and the multiple-trait model. Both methods assume that distinct environments exist. When the production environment can be described as a continuous gradient, reaction norm models are a viable alternative (Kolmodin et al. 2002). The environment is quantified on a continuous scale in reaction norm models instead of being classified into groups in the traditional GxE methods. The environment is not only the space surrounding the animal; it also includes temperature, floor space, air, nutrition, feeding, vaccination etc. As the observed performance phenotype is the result of all environmental factors, it is common to use the mean performance of animals raised in the same environment as a proxy for the sum of all environmental factors affecting performance of animals (e.g. Calus and Veerkamp, 2003; Knap and Su, 2008; Hammami et al., 2009). Reliable estimates of the parameters of reaction norm models can only be obtained when there is sufficient variation in the independent parameter, a descriptor of the environment, and when progeny of sires are distributed across environments to ensure that sires have a good representation across a wide range of environments. It was the aim of this study to quantify variation in environmental descriptors within and between herds and to explore genotype by environment interactions.