| dc.description.abstract | Urban development, with the introduction of piped drainage systems and increase in impervious surfaces, have altered the natural runoff patterns severely. This has resulted in increased risks of flooding and damages, and reduced ecological status of natural water bodies caused by deterioration in water quality and changes in both high- and low flow regimes. Stormwater management is aimed at counteracting these unfortunate consequences by attenuating and delaying peak flows, remove pollutants and reduce stormwater runoff volumes. This can be achieved by combining a set of methods serving different purposes, from local stormwater measures situated close to the runoff source via efficient drainage systems, centralized measures and efficient flow paths for extreme flooding events. This study focuses on one of the local measures, namely green roofs, and more specifically the widely used extensive green roofs with build-up depths up to 15 cm. Green roofs can reduce stormwater runoff volumes (retention) and attenuate and delay peak runoff rates (detention). These processes are highly climate dependent. A large amount of studies have been carried out worldwide, but few of these are carried out in comparable climates with the combination of low temperatures and high precipitation amount experienced in many Norwegian cities. This study was motivated by the need of more knowledge on extensive green roof performance in cold and wet climates.
The main objective of the work has been to study and quantify what role extensive green roofs can play as a part of stormwater management in cold and wet climates, to find the limitations and possibilities of this measure, and further to investigate tools for performance estimates and design.
The study is based on field observations (3-8 years of data) from several different configurations of extensive green roofs located in four Norwegian cities providing an opportunity to study a variation of cold and wet climates with respect to temperatures, precipitation amounts and patterns. In addition to field observations, a water balance model have been applied to study retention behavior based on 30 year time series of daily data, while a physical based model (SWMM) have been applied to replicate green roof runoff hydrographs with high time resolution (1 minute).
Green roofs were found to be efficient for small and medium precipitation events both with respect to reduction of stormwater volumes (retention), and for attenuation and delay of peak flows (detention). Green roofs can contribute in replicating predevelopment hydrology by promoting evapotranspiration and by prolonging the runoff times compared to a conventional roof. The performance show large seasonal variations with best performance in the temperate season (May through October). Peak reduction and peak delay was found to decrease with increasing peak precipitation and increasing initial soil moisture, with limited efficiency for the highest intensity events.
Retention performance of extensive green roofs in the tested climates were mainly found to be governed by the process of evapotranspiration, rather than the maximum water storage capacities. Estimates for retention capacity in terms of average volume removal or available retention capacity for a random precipitation event, was found both from field studies and from long term modelling. Rough estimates based on all observed configurations and locations in the temperate season were found to be approximately 45-55 mm/month and 5 mm per event. A green roof designed with a maximum water storage capacity of 10-15 mm was found to be sufficient for stormwater retention and detention performance in the studied climates, while 20-30 mm are recommended to promote plant welfare and reduce the risk of drought.
SWMM has provided a promising tool for modelling of green roof runoff for practitioners. The model does however still have some challenges associated with the green roof module, related to both evapotranspiration and the generality of the model parameters. The model needs improvements before it can be recommended used as a tool for estimating green roof performance. A conservative simplified approach for estimating green roof performance on design events is suggested, where the first part of the precipitation hyetograph, equivalent to the available retention capacity, is removed.
Field studies on green roof runoff water quality confirmed earlier reported studies that there are possible challenges associated with increased nutrient runoff. Longer and more studies are needed to include the large variability associated with precipitation amounts and patterns, aging of material, and especially for cold climates, the influence of the large seasonal variations.
Different local stormwater measures (green roofs, bioretention cells and detention basins) were compared in terms of long term and short term (design event) performance for both present and future climate. Green roofs and bioretention cells can provide a substantial and important contribution to volume and pollution control of stormwater runoff, also with comparable good performance in a high emission future climate scenario with increasing precipitation amounts. Green roofs and bioretention cells have limited detention performance on large design events, but if applied in series they can reduce the required downstream detention basin volumes substantially. More suitable methods for this type of analysis are needed for practitioners to be able to predict overall performance of the applied stormwater measures.
Stormwater design and performance estimates of green roof are recommended done based on overall volume reductions and an initial retention of a certain precipitation depth in the beginning of a design precipitation event. Stormwater regulations should, in addition to peak flow control, therefore include regulations for volume control and pollutant control. This would promote the use of a larger variety of stormwater measures, including green roofs, giving multiple benefits which could provide a more resilient stormwater management in our cities. | nb_NO |
