Urbanization replaces once permeable surfaces with relatively impervious ones, thereby degrading the natural hydrologic cycle. Impervious surfaces intensify stormwater runoff in terms of overall mass and temporal response, especially under torrential rainfalls. Since such runoff could become massive in volume and concentrated in time, they place significant stress on the urban drainage system and increase the risks of combined sewage overflow and flooding, which could introduce a range of deleterious consequences to cities and surrounding natural habitats.
In sustainable urban stormwater management like the Low Impact Development, green roof presents an on-site source-reduction measure that mimics the pre-development hydrologic functions of storing and gradually releasing precipitation. Green roof can retain and detain stormwater as well as delay and suppress peak discharge. However, green roof stormwater studies have largely been conducted in non-tropical regions of the world. Since green roof’s quantitative hydrologic performance can be much influenced by local meteorological conditions, the degree to which such findings can be generalized to other climates, such as Hong Kong’s humid subtropical regime, calls for investigation. Moreover, substrate depth has long been regarded as an influential factor in green roof stormwater retention, but two recent studies have provided contradictory results. The objectives of this study are: 1) To evaluate green roof stormwater mitigation performance and potentials in Hong Kong for the first time; 2) To investigate systematically the effect of substrate depth on quantitative hydrologic performance; 3) To identify factors that affect green roof performance; 4) To develop a holistic conceptualization of the various system water storage spaces within a green roof system, for a better understanding of their role in stormwater mitigation.
Using small-scale (1.1 m2) raised green roof plots placed on an actual urban rooftop, the effect of 40 mm soil, 40 mm soil + 40 mm rockwool, 80 mm soil, and 80 mm soil + 40 mm rockwool on stormwater mitigation performance relative to control were analyzed. Three core performance indicators (percent retention, peak delay, and peak reduction) were employed to evaluate green roof performance.
The results suggest that, while the retention performance of the studied green roofs under Hong Kong’s heavy rainfall regime seems to be less effective, remarkable peak reduction and peak delay were observed even when the green roof system has reached full moisture-storage capacity. Such findings are in line with the proposed Green-roof System Capacity model that green roof serves as an effective buffer that regulates water flow through the system.
No statistical significance was found between substrate-depth treatments, despite the higher performance across all three indicators for treatment 80. However, satisfactory peak performance of the 40-mm thin substrate suggests that green roof can be applied even on existing buildings that have limited loading capacity. Pertinent meteorological factors were identified. All in all, extensive green roof remains as an effective and promising alternative mitigation strategy to urban stormwater management in Hong Kong with potential application to other tropical areas. / Li Ka Shing Prize, The Best MPhil Thesis in the Faculties of Architecture, Arts, Business & Economics, Education, Law and Social Sciences (University of Hong Kong), 2012-2013. / published_or_final_version / Geography / Master / Master of Philosophy
| Identifer | oai:union.ndltd.org:HKU/oai:hub.hku.hk:10722/207562 |
| Date | January 2013 |
| Creators | Wong, Kwan-lam, 王韵琳 |
| Publisher | The University of Hong Kong (Pokfulam, Hong Kong) |
| Source Sets | Hong Kong University Theses |
| Language | English |
| Detected Language | English |
| Type | PG_Thesis |
| Rights | Creative Commons: Attribution 3.0 Hong Kong License, The author retains all proprietary rights, (such as patent rights) and the right to use in future works. |
| Relation | HKU Theses Online (HKUTO) |
Page generated in 0.026 seconds