Optimizing effluent trading and risk management schemes considering dual risk aversion for an agricultural watershed

Increasing amounts of wastewater are discharged to water bodies, with a risk of exceeding their capacity to cope with such loads. Nutrient discharge forms two types of risk, i.e., economic and excess effluent risks. In this study, a Bayesian dual risk aversion based stochastic programming (BDRSP) is proposed for selecting optimal effluent trading and multi-risk management schemes. The BDRSP framework includes uncertain simulation for nutrient loading, optimization techniques for optimal trading planning, copulas for disclosing spatial correlation of nutrient pollution as well as TOPSIS for selecting optimal risk management schemes. BDRSP is applied to a real case of Daguhe watershed, China for planning of a NH3-N trading system. Trading ratios are estimated based on ratio between environmental damages at the watershed outlet that emission discharges in two sources. Optimal effluent trading scheme is obtained considering random pollutant loading and the associated dual risk. The spatial pattern of nutrient pollution risk is identified based on joint probability distributions and the related joint exceedance probability of different locations with copulas. Optimal dual risk management schemes are generated considering system benefit, unit revenue as well as NH3-N loading and its spatial pattern. Risk management schemes under high economic and excess effluent risk control levels (i.e. 0.85<1, 0.4<0.6 and 0.85<1, 0.8<1) are recommended.

Automated summary

This study proposes a Bayesian dual risk aversion based stochastic programming method for selecting optimal effluent trading and multi-risk management schemes. By considering uncertain nutrient loading and spatial correlation, the study obtains the optimal trading scheme and risk management scheme, providing guidance for water protection and management.