A multiobjective response surface approach for improved water quality planning in lakes and reservoirs

摘要: [1] Improved data collection techniques as well increasing computing power are opening up new opportunities for the development of sophisticated models that can accurately reproduce hydrodynamic and biochemical conditions water bodies. While model complexity is considered a virtue scientific purposes, it definite disadvantage management (engineering) limits applicability to what-if analysis over few, priori defined interventions. In recent past, this has become significant limitation, particularly considering advances in quality rehabilitation technologies (e.g., mixers or oxygenators) which many design parameters have be decided. paper, novel approach toward integrating science-oriented engineering-oriented improving planning presented. It based on use few appropriately designed simulations complex process-based iteratively identify multidimensional function (response surface) maps interventions into objective function. On basis response surface (RS), greater number quickly evaluated corresponding Pareto front approximated. Interesting points then selected simulated using original model, thus obtaining decision-objective samples refine RS approximation. The demonstrated Googong Reservoir (Australia), periodically affected by high concentrations manganese cyanobacteria. Results indicate improvements could observed simply changing location two installed 2007. Furthermore, also suggests best an additional pair mixers.