A herd modelling approach to determine the economically and environmentally most interesting dietary amino acid level during the fattening period

Abstract

For a sustainable pork production, emission of pollutants from pig farms has to be reduced as much as possible whilst economical results have to be maximised. To evaluate different feeding strategies in terms of economic performance and environmental impact (through nitrogen excretion), we simulated herd performance using the prediction model InraPorc. A population of 1000 virtual pigs was generated. Performance of these pigs was simulated using different feeding strategies varying in digestible lysine to NE ratio in the diet (85 to 115% of the mean population requirement) and number of diets used (i.e. a 1-phase feeding strategy, a 2-phase strategy or a continuous multiphase strategy where the lysine/NE ratio was changed daily according to requirement). Diets were formulated on a least-cost basis using two contexts of ingredient prices: September 2007 (P1) or March 2009 (P2). Simulations were ended at a mean population body weight of 112 kg. Average daily gain increased while feed-to-gain ratio and duration of growth decreased with increasing lysine content. Maximum performance was achieved for a supply between 105 and 115% of the mean population requirement for lysine/NE. Increasing the lysine/NE ratio from 85 to 100% for the 1-phase strategy and to 105% for multiphase strategies reduced total feed cost in both economic contexts. The financial return (carcass payment minus feed and labour costs) increased with increasing lysine/ NE ratio and was maximal with the multiphase strategy and a lysine/NE ratio corresponding to 110% of the mean population requirement. When using a lysine/NE ratio greater than the mean population requirement, multiphase feeding strategies reduced nitrogen excretion by 3 to 16% compared with 1-phase strategy. The relationship between financial return and nitrogen excretion depended on the feeding strategy. Using multiphase strategies allows to better match the nutrient supply to the requirement. Based on feed cost alone, it allows optimizing economic return while reducing nitrogen excretion. Apart from the interest of including the variation between animals, stochastic simulation modelling can be helpful in multiple criteria evaluation of feeding practices.