Human Wellness in the Cape Fear River Basin Based on CAFO Data

Abstract

In the U.S, animal farms have moved to an industrial scale resulting in concentrated animal feeding operations (CAFOs) that manage to house thousands of live animals at high densities. Even though CAFOs have remarkably increased the production of animal agriculture, the outcomes related to their activity present possible health and wellness metrics risks to nearby communities. North Carolina has the highest density of swine CAFO activity in the U.S. and the entire world. In this work, we aimed to study the impacts on North Carolina communities and develop predictive models to predict the effects of potential future CAFOs.We analyzed how these variables relate to each other and CAFO abundance to apply classical machine learning models. We developed two groups of models. Group A models predict the areas of likely CAFO expansion and Group B models predict the effects on certain wellness metrics in those areas. Group A models can narrow down the areas of concern and allows us to apply group B models. Results of group B models predict changes in the wellness metrics if certain levels of CAFO development were to occur. The developed models prove effective in the objectives outlined. Additionally, the models could prove an effective tool when considering the expansion of CAFOs into currently unaffected areas.