Abstract Paved surfaces, increased precipitation intensities in addition to limited capacity in the sewer systems, cause a higher risk of combined sewer overflows (CSOs). Sustainable drainage systems (SUDS) offer an alternative approach to mitigate CSO by managing the stormwater locally. Seven SUDS scenarios, developed based on the concept of effective impervious area reduction, have been implemented in the Grefsen catchment using the Mike Urban model. This study evaluated the hydrological performance of two SUDS controls (i.e. green roof (GR) and rain garden (RG)) modules of the model and the effect of the SUDS scenarios on the CSOs using event-based and continuous simulations. The Nash–Sutcliffe efficiency (NSE) along with flow duration curves (FDCs) has been used for evaluating the model performance. Event-based evaluations revealed the superior performance of the RG in reducing CSOs for larger precipitation events, while GRs were proven to have beneficial outcomes during smaller events. The study illustrated another way of assessing the continuous simulations by employing the FDCs. The FDCs were assessed against a discharge threshold at the outlet (which authorities can set as design criteria) of the catchment in terms of the extent, each scenario reduced occurrence and duration of outflow that invokes flow in the overflow pipe.

Abstract. Green roofs are increasingly popular measures to permanently reduce or delay storm-water runoff. The main objective of the study was to examine the potential of using machine learning (ML) to simulate runoff from green roofs to estimate their hydrological performance. Four machine learning methods, artificial neural network (ANN), M5 model tree, long short-term memory (LSTM) and k nearest neighbour (kNN), were applied to simulate storm-water runoff from 16 extensive green roofs located in four Norwegian cities across different climatic zones. The potential of these ML methods for estimating green roof retention was assessed by comparing their simulations with a proven conceptual retention model. Furthermore, the transferability of ML models between the different green roofs in the study was tested to investigate the potential of using ML models as a tool for planning and design purposes. The ML models yielded low volumetric errors that were comparable with the conceptual retention models, which indicates good performance in estimating annual retention. The ML models yielded satisfactory modelling results (NSE >0.5) in most of the roofs, which indicates an ability to estimate green roof detention. The variations in ML models' performance between the cities was larger than between the different configurations, which was attributed to the different climatic characteristics between the four cities. Transferred ML models between cities with similar rainfall events characteristics (Bergen–Sandnes, Trondheim–Oslo) could yield satisfactory modelling performance (Nash–Sutcliffe efficiency NSE >0.5 and percentage bias |PBIAS| <25 %) in most cases. However, we recommend the use of the conceptual retention model over the transferred ML models, to estimate the retention of new green roofs, as it gives more accurate volume estimates. Follow-up studies are needed to explore the potential of ML models in estimating detention from higher temporal resolution datasets.

Lined permeable pavements (LPPs) are types of sustainable urban stormwater systems (SUDs) that are suitable for locations with low infiltration capacity or shallow groundwater levels. This study evaluated the hydrological performance of an LPP system in Norway using common detention indicators and flow duration curves (FDCs). Two hydrological models, the Storm Water Management Model (SWMM)-LID module and a reservoir model, were applied to simulate continuous outflows from the LPP system to plot the FDCs. The sensitivity of the parameters of the SWMM-LID module was assessed using the generalized likelihood uncertainty estimation methodology. The LPP system was found to detain the flow effectively based on the median values of the detention indicators (peak reduction = 89%, peak delay = 40 min, centroid delay = 45 min, T50-delay = 86 min). However, these indicators are found to be sensitive to the amount of precipitation and initial conditions. The reservoir model developed in this study was found to yield more accurate simulations (higher NSE) than the SWMM-LID module, and it can be considered a suitable design tool for LPP systems. The FDC offers an informative method to demonstrate the hydrological performance of LPP systems for stormwater engineers and decision-makers.

No abstract