MISAR in enhancing agricultural resilience: a comprehensive approach to climate change risk management for mango farms in Sicily, Italy

Abstract

Agriculture plays a crucial role in the economy of Italy, particularly in the region of Sicily where it serves as a primary source of income. To ensure high yields, it is essential to enhance farmers’ knowledge and awareness, especially in mitigating potential risks and damages caused by climate change and managing farming processes such as soil and water preparation, fertilizer, and pesticide management. To follow the MISAR (Climate Change Risk Management by Improving the Individual and Social Awareness of Risk in Sicily) targets, this paper focuses on the importance of Information and communication technologies (ICT) in the “Mango Farms Risk Management Plan” to foster stronger connections between stakeholders and farmers in Messina. Climate change poses various hazards such as temperature fluctuations, extreme events, soil salinity, and irregular rainfall, which are expected to increase in the future. Effective decision-making for stakeholders and farmers requires efficient analytical tools, particularly for handling large data sets. The paper introduces a new architecture called ADM, which combines decision support systems (DSS), agent-based modeling (ABM), and machine learning (ML) methods to develop a comprehensive risk plan for future agricultural challenges. The ADM model in MISAR incorporates empirical information collected during the ML phase, including the reactions of mango plants to risks and determining factors like extreme temperature changes. To promote and safeguard mango cultivation and production, changes in temperature are estimated using advanced techniques such as random forest and feed-forward neural networks. Weather stations equipped with meteorological sensors are strategically placed within farms, providing direct measurements of hazards. Each station has its own credentials, allowing farmers access to the data. Furthermore, historical data analysis considers data from municipal meteorological stations and satellite sources. The model facilitates mutual communication between decision-makers and farmers, enabling farmers to monitor forecasts and report unexpected events in their respective farm areas.