Joint analytical hierarchy and metaheuristic optimization as a framework to mitigate fertilizer-based pollution.

The emission of nitrogenous pollution from agricultural lands in form of ammonia volatilization, leaching, runoff, N2O emissions, etc. is still a serious challenge to which agricultural sector faces. In this context, a vast number of decision support systems have been developed and tested to find the best nitrogen application rate. These models are highly dependent on crop simulation models, mathematical and regression models, evolutionary algorithms and artificial intelligent, GIS-based models, etc., while in most cases have ignored to be interfered with regional and national regulations established by experts in the field. In this study, a new framework combining analytical hierarchy (AHP)/modified AHP methods (MAHP) plus metaheuristic optimization techniques has been suggested to find the best nitrogen application rate considering regional capacities and requirements. To reach the objectives of the present study a three yield field experiment was conducted upon which crop yield, nitrogen use efficiency, nitrogen uptake, soil nitrate, ammonia volatilization, N2O emissions, and N leaching were monitored or measured. Using the results from the field experiments and a survey from local experts, the models were developed. AHP-assisted optimization model could cause some biases in the final results due to its intrinsic nature which avoids direct pairwise comparison among indicators (so called sub-criteria) under two different main-criteria. On the contrary, MAHP-assisted model could well reflect the concerns of experts and notably decrease hotspot pollution. Such decision support system can satisfy both farmers and environmentalists' need because of the created high profit and low environmental pollution, while saving resources and ensuring a sustainable production system.