Event-based calibration of build-up parameters in conceptual water quality models using turbidity data

Build-up/wash-off water quality models are frequently used to assess stormwater runoff quality. In this work, we explore the possibility of using continuous turbidity data to calibrate and validate such models under realistic conditions in two case studies located in Flanders, Belgium. By using such rich continuous data, a much larger number of rainfall events can be used for calibration than it is feasible with grab samples. The approach shows that the build-up rate used in these conceptual models varies widely during different rainfall events between 100 and 1350 kg/ha/y in the first case study and between 100 and 1750 kg/ha/y in the second case study. The time dependence of the build-up rates from different catchment types and its correlations with site-specific information/external conditions, such as wind speed, temperature and traffic volume are investigated. The modelled build-up rates are lower during storm events in weekend days compared to week days, consistent with traffic measurements. Furthermore, emission factors for different vehicle types previously published in literature can be used to estimate the build-up rate for streets. In one case study a strong negative correlation between wind speed and build-up rate was observed (r = −0.89). Overall, our study shows that continuous turbidity data can be used to study the impact of external conditions on the build-up rate in conceptual water quality models.