Diffuse groundwater pollution is a growing concern everywhere in the world and one of the most problematic and widespread of the vast number of potential groundwater contaminants are nitrates, which often primarily comes from the intense use of fertilizer in agriculture.
       Groundwater pollution has provoked a normative and a recommendation development. In Europe the Nitrate Directive was established in 1991, and the Water Framework Directive (WFD) in 2000. The WFD states that all water bodies have to reach a good quality status by 2015. The WFD explicitly recognizes the role of economics in reaching environmental and ecological objectives. One of the elements that the WFD mentions is the cost-effectiveness analysis (CEA) as a method to obtain the most cost-effective program of measures to reach good water status.
       A hydro-economic modeling framework is developed for determining optimal management of ground water nitrate pollution from agriculture. A holistic optimization model determines the spatial and temporal fertilizer application rate that maximizes the net benefits in agriculture constrained by the quality requirements in groundwater at various control sites. Since emissions (nitrogen loading rates) are what can be controlled, but the concentrations are the policy targets, we need to relate both. Agronomic simulation models are used to obtain crop yield and nitrate leaching functions in terms of water and fertilizer use, while numerical groundwater flow and solute transport simulation models were used to develop unit source solutions that were assembled into a pollutant concentration response matrix. The integration of the response matrix in the constraints of the management model allows simulating by superposition the evolution of groundwater nitrate concentration over time at different points of interest throughout the aquifer resulting from multiple pollutant sources distributed over time and space. In this way, the modeling framework relates the fertilizer loads with the nitrate concentration at the control sites. The benefits in agriculture were determined through crop prices and crop production functions. In this way, this framework provides a practical tool for analyzing the opportunity cost of measures for reducing nitrogen loadings and assessing their effectiveness for maintaining groundwater nitrate concentration within the target levels.
       The management model was applied to a hypothetical groundwater system. Optimal solutions of fertilizer use to problems with different initial conditions, planning horizons, and recovery times were determined. The illustrative example shows the importance of the location of the pollution sources in relation to the control sites, and how both the selected planning horizon and the target recovery time can strongly influence the limitation of fertilizer use and the economic opportunity cost for meeting the environmental standards. There is clearly a trade-off between the time horizon to reach the standards (recovery time) and the economic losses from nitrogen use reductions.
       In decision-making processes, reliability and risk-aversion play a decisive role. This dissertation presents a stochastic optimization framework to incorporate the effects of the hydraulic conductivity uncertainty on the least-cost allocation of nitrogen reduction among the agriculture pollution sources in order to meet the groundwater nitrate concentration targets in a groundwater basin under heterogeneous physical conditions. Four different formulations were applied: Monte Carlo simulation with pre-assumed parameter field, Monte Carlo optimization, stacking management, and mixed-integer stochastic model with predefined reliability. These formulations were tested in an illustrative example were 100 realizations were performed for two cases with different hydraulic conductivity field variance. 
       The methodology was applied also to a real case-study, “El Salobral-Los Llanos” (within the Mancha Oriental groundwater body). The model yields the optimal fertilizer application that meets the groundwater nitrate standard for a 50 year planning horizon. The average fertilizer application has to be reduced by 39 kg/ha in order to comply with the environmental standard. This reduction implies a smaller production, which represents a forgone benefit of about 1.2 M€/year.  
Despite the necessary limitations of any modeling approach, the relevance and complexity of real-world groundwater diffuse pollution issues call for the development of integrated hydro-economic models in order to address the problem of multiple pollution sources under heterogeneous conditions, integrating the main agronomic, economic and biophysical elements of the process (including the associated uncertainty) at the groundwater basin scale. This research provides a useful methodology and tool for decision-making in the ongoing process of implementation of the Water Framework Directive and the Groundwater Daughter Directive criteria to the groundwater bodies. Further research would be required in order to extend the representation of the diversity of potential on-farm management decisions and other policy options apart from fertilizer use limitations.