Bayesian Networks for the management of greenhouse gas emissions in the British agricultural sector

Recent years have witnessed a rapid rise in the development of deterministic and nondeterministic models to estimate human impacts on the environment. An important failing of these models is the difficulty that most people have understanding the results generated by them, the implications to their way of life and also that of future generations. Within the field, the measurement of greenhouse gas emissions (GHG) is one such result. The research described in this paper evaluates the potential of Bayesian Network (BN) models for the task of managing GHG emissions in the British agricultural sector. Case study farms typifying the British agricultural sector were inputted into both, the BN model and CALM, a Carbon accounting tool used by the Country Land Business Association (CLA) in the UK for the same purpose. Preliminary results show that the BN model provides a better understanding of how the tasks carried out on a farm impact the environment through the generation of GHG emissions. This understanding is achieved by translating the emissions information into their cost in monetary terms using the Shadow Price of Carbon (SPC), something that is not possible using the CALM tool. In this manner, the farming sector should be more inclined to deploy measures for reducing its impact. At the same time, the output of the analysis can be used to generate a business plan that will not have a negative effect on a farm’s capital income.