Title: Nature as a template for a new concept of extensive green roofs
Other Titles: Natuur als sjabloon voor een nieuw concept van extensieve groendaken
Authors: Van Mechelen, Carmen; S0168104
Issue Date: 10-Mar-2015
Abstract: In an era of urbanization, biodiversity is under pressure more than ever. Biodiversity is considered a major driver of ecosystem functioning and the provision of ecosystem services. Green roofs, a prime example of urban novel ecosystems, offer habitats and can hence mitigate some biodiversity loss in cities. Apart from biodiversity, green roofs also offer other ecosystem services, such as thermal regulation, stormwater management, and aesthetic and amenity value. Here we focused on extensive green roofs (substrate depth < 20 cm) as these can be applied widely and are more durable (i.e. less maintenance, self-sustaining).
In southern Europe (Mediterranean), the performance of (extensive) green roofs is rather low, probably due to the elevated temperatures and summer drought. One may expect that plant stress on green roofs will further increase as a result of climate change. Irrigation could help plant growth and survival. However, irrigation is often perceived as an unsustainable practice, as water is already a limiting factor in many regions and climate change will lead to an even more severe water scarcity during summer. Therefore, research is needed to select plant species suitable for Mediterranean (unirrigated) extensive green roofs, and to adapt green roof design to meet the requirements of the selected plant species. More northern countries with colder climates (e.g. temperate maritime climate) will also face higher temperatures and erratic precipitation events as a result of climate change. The green roof industries located in these regions will hence also benefit from the outcome of such research.
The main goal of this thesis was to elaborate and test a new concept for extensive green roof design, comprising both plant selection and design elements. The work is based on the habitat template theory, which states that natural habitats with similar characteristics as extensive green roofs should be targeted when searching for suitable plant species. Mediterranean regions are a hotspot of biodiversity and contain many habitats that match to some extent the conditions on extensive green roofs (e.g. shallow, free draining, nutrient poor and calcareous soils, high temperature fluctuations, windy). We hence hypothesized that it would be possible to find potential plant species for use on extensive green roofs. Because of practical reasons we selected the southern part of France as study region.
During fieldwork in southern France during the spring of 2011, 372 potential plant species were recorded in 20 natural open Mediterranean habitats. Variation in species composition in relation to environmental factors was analyzed, and the vegetation could be classified into four major vegetation types. When comparing the potential species with a list of species commonly used on green roofs in northwestern Europe, it was found that 79% is currently not used on green roofs. Results also show that annual species comprise an important part of Mediterranean vegetation; however this life form is currently rarely used on green roofs. As annual species provide many interesting features, especially for regions with prolonged drought, we concluded that this life form should be considered for application on green roofs. Given the large species pool found during the field work, we developed a screening procedure to guide the green roof industry towards the most appropriate species for use on Mediterranean extensive green roofs. Both functional plant traits (specifically related to drought tolerance and self-regulation as these are of prime importance for survival on green roofs) and utilitarian characteristics were used. After statistical analysis, the most important traits were incorporated into a multi-criteria screening tool. This tool also used exclusion criteria (e.g. trees, no stress tolerance) to delete the unsuitable species. It was illustrated on the species found in Mediterranean habitats, which resulted in 34 high potential species not yet used on green roofs. Interestingly, 35% of these species had an annual life form, confirming our previous recommendation to include annual species in green roof vegetation.
As the fieldwork and screening procedure only provided theoretical evidence of the potential of species from Mediterranean habitats for use on extensive green roofs, we tested 18 potential species during a two-year experiment on unirrigated extensive green roofs. Experiments were conducted both in Avignon, France (representative for the Mediterranean climate) and Heverlee, Belgium (representative for the temperate maritime climate). The experimental setup included different substrate compositions (substrate depth 5 or 10 cm; presence/absence of a water retention layer) and exposition (exposed or sheltered). Weather conditions greatly affected vegetation development in both locations. Overall, the deepest substrate depth of 10 cm combined with a water retention layer resulted in the highest cover, species abundance and biodiversity values throughout the experiment. The exposition proved to be a less important factor for species performance. Both bryophytes and weeds established spontaneously over the two years, especially in the sheltered plots.
Plant growth and survival on green roofs during drought remains challenging as the substrate is supposed to be light weight, shallow and relatively low in nutrients. In horticulture, the positive effects of organic matter and arbuscular mycorrhizal fungi (AM fungi) are acknowledged, but evidence of the effects on green roofs is still scarce. We hence examined how AM fungi and organic matter affect early growth and survival of plants in experimental green roofs. The experiment was conducted both in a greenhouse and on a rooftop in Heverlee, Belgium. A seed mix of commonly used green roof plants and an inoculum consisting of Glomus species were used. Organic matter positively affected plant height, dry weight and cover in the greenhouse. No significant effect of AM fungi on initial plant growth and survival was found, although microscopic evidence of AM fungi - root associations was demonstrated. We concluded that a small increase in organic matter can already stimulate plant growth on green roofs. More time is probably needed for the AM fungi - root associations to result in positive effects on plant growth and survival under drought.
As stated earlier, irrigation can also help plant growth and survival on green roofs, although this often has a negative connotation in sustainable water management. As information on appropriate green roof irrigation is still very scarce, we reviewed the existing literature and drew attention to alternative water sources and water conservation strategies. Water can be conserved by adapting irrigation requirements (i.e. by targeting the green roof vegetation or materials), by using alternative irrigation sources (e.g. gray water and rainwater harvesting) or by controlling and monitoring irrigation regimes. Some form of irrigation is essential during establishment and the first growing season on all types of green roofs and in all considered climates. Afterwards, irrigation is necessary on green roofs in (semi)-arid climates, and advised in small amounts in the other climates. Minimum irrigation specifications for different climates were proposed, based on the deficit irrigation theory. If green roofs are designed carefully, then irrigation practices can be sustainable in the long term and contribute to better urban life quality.
In a final study we focused on the biodiversity and ecosystem services aspect of extensive green roof systems. There is a growing consensus that ecosystem services can be predicted through functional and phylogenetic diversity of plant communities. We analyzed the initial plant composition of commercial extensive green roof systems in terms of their functional and phylogenetic diversity. Clustering techniques grouped the systems in three green roof types that differ in species richness and composition, substrate depth, installation method and functional and phylogenetic diversity. Correlation between functional diversity and species richness was strongly positive, suggesting that species-rich systems might offer more ecosystem services. We proposed two methods to select the species combination with maximal functional diversity. The described approach is generic in nature and can be used for plant selection for a broad range of novel ecosystems.
We conclude that natural habitats in the Mediterranean region can definitely inspire us as a source for development and improvement of extensive green roof design, whether this is for the current Mediterranean climate itself or for other climates under predicted climate change. Appropriate vegetation choice is essential, as well as the design in terms of sustainable irrigation techniques, appropriate substrate depth and composition, and water retention possibilities. Finally suggestions for further research were made.
Description: This PhD manuscript is a result of a collaboration between the KU Leuven and the Université d'Avignon et des Pays de Vaucluse.
Table of Contents: Chapter 1: Introduction
Chapter 2: Mediterranean open habitat vegetation offers great potential for extensive green roof design
Chapter 3: Plan trait analysis delivers an extensive list of potential green roof species for Mediterranean France
Chapter 4: Vegetation development on different extensive green roof types in a Mediterranean and temperate maritime climate
Chapter 5: Effects of addition of organic matter and arbuscular mycorrhizal fungi on initial green roof plant development
Chapter 6: Adapting green roof irrigation practices for a sustainable future
Chapter 7: Functional and phylogenetic diversity as a framework for novel ecosystem design, the example of extensive green roofs
Chapter 8: General discussion
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Division Forest, Nature and Landscape Research

Files in This Item:
File Status SizeFormat
PhD final 2.pdf Published 7719KbAdobe PDFView/Open Request a copy

These files are only available to some KU Leuven Association staff members


All items in Lirias are protected by copyright, with all rights reserved.