Evaluation of different methods for modelling stormwater quality

Abstract

Alongside the ever-increasing urban exploitation, the consideration of stormwater quality become more and more important. Since the beginning of the year 2000, water quality in the European Union is regulated by the European Water Framework Directive. This directive includes regulation of reciving water quality. The investigation of, and planning regarding stormwater can be facilitated by the use of models. In this study, different methods for modelling stormwater quality are evaluated. A market survey of available stormwater models in Sweden and internationally are performed. Then, a method description of the models StormTac, MIKE URBAN and SEWSYS are implemented. Each of these three models represents different methods for estimating stormwater quality. StormTac is based on static calculations of the land-use specific method, whereas the other two models are based on dynamic calculations using a pollutant source-based method in SEWSYS and a physical process-based method in MIKE URBAN. A theoretical comparison of the methodology of these three models is provided. A case study with monitoring data is then applied to StormTac and MIKE URBAN, in order to compare the simulated values with monitoring values. The outcomes from these two models turned out to fall into the measured minimum- and maximum-interval. The stormwater quantity, however, was not as well matched with the measured values. The total runoff volume simulated in StormTac was approximately three times higher than the measured volume and around three times lower in MIKE URBAN. Based on this study, StormTac is recommended for simulating long-term event mean concentrations (EMC´s) of pollutants. StormTac is a user-friendly model and it predicts EMC´s close to the measured EMC´s. StormTac can, however, not be used to investigate acute effects due to stormwater quality.