Hydrological modelling of green urban drainage systems : Advancing the understanding and management of uncertainties in data, model structure and objective functions

Abstract: The use of green urban drainage systems such as green roofs, swales and pervious areas has in recent years become a popular option to reduce flood risk and water quality problems in a more sustainable way than with traditional pipe-based drainage systems. Computer models are valuable tools for the management of such systems. While uncertainties associated with these models have been investigated for pipe-based systems, their adaptation and application to green areas requires re-examination of these uncertainties, as additional hydrological processes become relevant and new opportunities for model calibration arise. The overall aim of this thesis is to contribute to understanding and reducing of uncertainties in the mathematical modelling of green urban drainage studies. Specific topics adressed are field measurements, data processing, data selection, model structures and objective functions.Weighing-bucket precipitation sensors were confirmed on multiple occasions to be accurate to within ±1% of accumulated precipitation. A new signal processing method was able to convert accumulated precipitation to noise-free 1-minute rainfall rates that reproduced total rainfall volumes with only minor errors.Area-velocity flow sensors were tested and their measurement uncertainties quantified in laboratory experiments for flow rates up to 9 L s-1. Total flow rate uncertainty was ±0.34 L s-1 in optimal conditions (flat pipe), increasing to 0.60 and 0.83 L s-1 for pipe slopes of 2% and 4% respectively. In the presence of an upstream obstacle the uncertainty was 2 to 3 times larger, although in the case of no pipe slope this could be reduced to the same as the optimal conditions by increasing water levels in the pipe.Three different urban drainage models for green areas were compared using long-term simulations of synthetic catchments with different soil types and depth. In all models surface runoff formed a significant component of the annual water balance for some soil profiles, while the models reacted differently to changes in soil type an depth. Inter-model variation was large compared to the variation between different soil profiles.Four different models were tested for the simulation of runoff from two full-scale green roofs. More complex models showed better performance in reproducing observed runoff, while the magnitude and source of model predictive uncertainties varied between the models. It was also found that for all models calibration periods with high inter-event variability in terms of runoff retention provided more information in the calibration process.The use of soil water content observations (SWC) was investigated for the calibration of a detailed model of an urban swale. SWC observations were found to be useful for improving the identifiability of certain model parameters and the model predictions of SWC, and for setting the initial SWC in simulations. Different approaches to combining SWC and outflow observations were compared, revealing that the precision and reliability of model predictions could in some cases be improved by using a different way of determining which parameter sets to use for the generation of uncertain model predictions.The influence of calibration data selection was investigated using a model of a small green urban catchment. Performance of the model when calibrated using different sets of events varied significantly. Two-stage calibration strategies (where first small rainfall events were used to calibrate impervious area parameters, followed by using larger events to calibrate green area parameters) showed good performance especially in terms of runoff volume and peak flow. Finally it was found that the benefits of the two-stage calibration were greater when using a model with a low spatial resolution than with a high spatial resolution.For the same catchment tests were also carried out of an objective function that explicitly allows for timing errors, rather than comparing only simulated and observed values for the same time step. Model predictions generated using this objective function were equally reliable, but more precise and therefore of more practical value.Finally, drawing upon the practical experience from working with different models and drainage systems an overview is provided of the applicability of the modelling techniques used in this thesis for different models and what features may be desirable to add to models to improve this.