An agent based decision framework to advance agricultural water management under global change

Agriculture is the major sector responsible for nearly 85 % of consumptive water use worldwide by human, where irrigation water supply is vital for coping with the inherent variability of natural hydro-climatic conditions so as to secure the production of food. However, evidenced by numerous studies, there is a strong trend of climate change, whose impact on agriculture may be multifold. For example, increase of global temperature may trigger long lasting drought events, while change in spatio-temporal distribution of precipitation would lead to less secured water availability. As the population growth is expected to continue, to meet the projected increase of food demand agricultural systems are called to adapt management strategies (e.g., diversifying crop patterns, or modifying irrigation scheduling to increase the water use efficiency) in order to squeeze more food out from a unit of water input, i.e., to follow the soft-path measures. In our work we contribute a novel decision-analytic framework to assist decision-makers in designing and assessing alternative soft adaptation solutions (e.g., distributed and participatory management, coordination mechanisms, use of medium/long term predictions) in order to improve the overall water productivity. The approach is demonstrated on Lake Como water system where a large regulated lake upstream (supply sector) is connected by an extensive cultivated area downstream (demand sector) through Muzza canal. An Agent-Based agronomic modeling framework is implemented to simulate the distributed physical environment coupled with multiple decision-making authorities, as well as their interaction amongst. Our results show that the proposed methodology is rigorous and effective for understanding the vulnerability of complex water system, evaluating alternative adaptation measures as well as assessing potential utility of technology advance in supporting agricultural practice.