The use of green roofs is increasing in many countries because of their benefits to the urban environment. However, only a few plant selection studies for green roofs have been carried out and little information on plant performance on roof environments is available in the UK climate. As a result, only a limited range of plants such as Sedum spp. are commonly used for green roofs, especially for shallow substrate green roofs. Therefore, this thesis investigates plant selection for extensive green roofs in the UK. The work in this thesis focused on the following objectives. (1) To identify groups of plants that have potential for use on green roofs, with regard to tolerance of rooftop conditions, (2) To investigate establishment methods for diverse, attractive, flowering green roof vegetation, with attention to seedling techniques, (3) To test survival and performance of a selected range of species and cultivars from the previously identified groups (annuals and geophytes) at different substrate depths, irrigation regimes and covering plants treatments, (4) To compare green roof performance (water management and drought tolerance) between different vegetation types and drought tolerance with different percentages of organic matter in the growing substrate, (5) To investigate the performance of plants as well as their aesthetic appeal, seasonal interest over time and what is required for maintenance (weed Invasion and self-seeding). The direct sowing of perennial and grass mixtures, the use of annual plant seed mixtures and the use of geophytes could be useful techniques for the quick establishment, long flowering, their beautiful colour of flowers, cost effectiveness and providing food resources for biodiversity in an extensive green roof. Germination testing revealed that many perennial and grasses which have potential for use in extensive green roofs did not require chilling for germination and had high germination rates in spring. The results suggested that spring might be the best season for direct sowing on the roofs for quick establishment. In annual plant meadows, it was shown that a low sowing density could be better than high density to reduce competition, resulting in good individual plant growth when there was sufficient watering. However, a high sowing density was recommended for the dry conditions. For geophytes, growth, survival rate, regeneration and flowering were more successful in a deeper substrate rather than a shallow substrate. The vegetation cover by Sedum seemed to work as a protection layer and the overall emergence was encouraged with Sedum, especially in the shallow substrate. In the study of amount of water runoff from different vegetation types, it was shown that grass species may be the most effective for reduction of water runoff followed by forbs and sedums. The size and structure of plants significantly influenced the amount of water runoff, however, species richness did not affect the amount of water runoff significantly. In the study of the drought tolerance of different vegetation types, the forbs and grasses groups used in this study reached permanent wilting point after two to three weeks of no watering and they were required to be watered once a week to maintain their visually attractive forms. Sedum spp. were able to survive well and maintain good visual quality even after three weeks of no watering. There was a tendency that overall survival increased as species richness increased. The diversity in vegetation reduced the vigor potential dominant species. In the investigation of the relationship between percentage of organic matter of substrate and plant growth, it was concluded that about 10% (about 14% in total) of organic matter was the best because the plants showed stable growth regardless of the watering regime. In wet conditions, increased organic matter resulted in increased growth, whereas in the dry conditions, increased organic matter did not result in increased growth. In the investigation of plant growth and performance on an existing semi-extensive green roof it was shown that it is possible to create low-input green roofs which have long flowering and seasonal interest with a little maintenance and supplemental irrigation if appropriate plants were chosen. Plant species diversity might affect overall flowering succession and dynamic change and planting density might affect interaction between plants. In areas of high plant species diversity, there were more possibilities to have a longer flowering term, more seasonal interest and dynamic change than low plant species diversity. In areas of low planting density, individual plants generally produced the better growth than those in high planting density. Moreover, plant growth had more interaction between species in the higher planting density. The tendency was observed that the plants had better growth in the NE and the SE. Also, longer flower duration was shown in the NW whereas many species started flower from the SE. The combination of low plant species diversity and high planting density appeared to reduce weeds effectively. Using a gravel mulch in the shallow substrate could reduce the number of weeds significantly.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:500197 |
| Date | January 2008 |
| Creators | Nagase, Ayako |
| Publisher | University of Sheffield |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://etheses.whiterose.ac.uk/10325/ |
Page generated in 0.0106 seconds