Lozano-Jaramillo, M. (2019). Predicting breed by environment interaction using ecological modelling. PhD thesis, Wageningen University, the Netherlands.Animal production plays an important socio-economic role in developing countries. Livestock largely contribute to the local food production, and they represent a source of income for many smallholder farmers. Smallholder production systems are characterized by low producing indigenous breeds. To enhance food security, several breeding projects have introduced commercial breeds as an alternative to help smallholder farmers to increase the food supply. As a common practice in animal breeding, commercial breeds are introduced from a nucleus to different locations. However, no knowledge on the new environmental conditions in the new tropical environment are taken into account. The aim of this thesis was to understand how the environment plays a role in shaping the variation in breed performance across different agro-ecologies. To achieve this, I used different approaches than what it is commonly used in animal breeding, to understand the environmental variation. I use Geographic Information Systems (GIS) along with different models used in ecology to predict the performance of genotypes in changing environments. Results highlight the importance of acknowledging the environment being composed of many variables that have continuous variation when designing breeding programs or testing schemes. I recommend the application of these models to livestock to make predictions on how productivity traits respond as a function of the environment.