Agricultural sector has a key role in poverty reduction and improving food security. One of the important challenges in the agriculture sector is to feed the increasing global population. Agriculture has both significant positive and negative impacts on environment and is a significant contributor to pollution which in turn restricts expansion of the agriculture production . Climate change has also attracted attentions in agricultural-environmental interactions. Bayesian networks are relatively well recognized and employed in different types of environmental modeling efforts. In this study a Bayesian network model was designed to investigate the relationship between agro-economic-environmental indicators and Environmental Performance Index (EPI) in the EU countries compared to Middle East (ME) countries that face many challenges for sustainable development. Scarcity of water, population growth and degradation of natural ecosystems are a few of these challenges. In this study, we showed the relationships between influential variables on EPI based on expert interview and available data. The results indicated land productivity was directly affected by organic agriculture area. It is found that with expansion of the organic and conservation agriculture area, productivity can be increased in EU Countries. Our findings also showed that with decreasing N2O and CH4 emissions, EPI in EU countries increased but it decreased in ME countries. Therefore, EU countries have been able to improve the EPI and achieve sustainable development objectives due to improved agricultural practices and optimum applications of pesticides and fertilizer. However, ME countries did not achieve such status. Modeling of agroenvironmental indicators can help policymakers to evaluate the changes of agro-ecosystem and apply them even at international scales.

Keywords