Sensitivity of dairy cattle farming to the economic environment within the context of climate variations

Context and challenges

Dairy cattle farms, which play a central role in agriculture and the food industry, face significant environmental challenges. Because of their contribution to greenhouse gas (GHG) emissions, they are obliged to take climate change mitigation measures. In France, agriculture accounts for 18% of GHG emissions, 59% of which are attributable to livestock farming, primarily cattle. This means that farm management practices must evolve to reduce these impacts without compromising productivity. These changes affect various aspects of livestock farming, such as milk and meat production, as well as the preservation of biodiversity. While it is essential to understand the synergies and trade-offs between these different factors, the wide diversity of livestock systems makes it difficult to generalize findings. In fact, intensive and extensive practices respond differently depending on the context. Finally, climatic and economic uncertainties further complicate the sustainable management of dairy cattle farms.

Goals

The aim of this project is to explore the consequences of economic and climatic variations on production, environmental performance, and the ecosystem services provided by livestock farms. A statistical modeling approach will be used to formalize the variety of complex interdependencies among variables within livestock farm systems and to describe the synergies and trade-offs that arise when implementing measures aimed at reducing their environmental impacts.

1. The first step of the project will involve developing the conceptual framework for the study by defining the subject of study and by targeting climatic, economic, and ecosystem service indicators associated with the descriptive variables of breeding farms.
2. The second step will involve compiling a list of descriptive databases of dairy cattle farms and indicators of the contextual factors listed above, at the national level in France. The compatibility of these data and indicators will then be examined.
3. The third stage of the project will focus on exploring statistical modeling approaches for the defined system, based on the available data. Multivariate analyses will be conducted to explore the data in order to identify potential correlations between variables, and to identify those variables that are important for the subsequent analysis. Learning models (e.g., random forests and deep learning) that account for nonlinear dependencies among variables will also be considered for use. Similarly, copula-based multivariate modeling approaches will be explored.

This project should also encourage reflection on how to define the vulnerability of dairy cattle breeding farms in order to help assess the multiple risks they face. This reflection can be enriched by interviews with farmers to identify hazards and uncertainties not accounted for in the data, as well as the adaptation strategies they are considering to address these risks. Feedback from farmers on how they prioritize the services provided by breeding farms can also be gathered. Such qualitative information can supplement the information contained in the data.

INRAE structures

Non-INRAE partner