Plant establishment on a green roof under extreme hot and dry conditions: The importance of leaf succulence in plant selection

https://doi.org/10.1016/j.ufug.2015.11.004Get rights and content

Highlights

  • We evaluated green roof plant performance under extreme hot and dry conditions.

  • Only succulent plants remained alive after 42 weeks with limited rainfall.

  • Survival was related to leaf succulence making this a useful trait for plant selection.

  • Irrigation will be needed for non-succulent plants in hot and dry climates.

Abstract

Plant selection for extensive green roofs has largely been based on cool, temperate climate research. However, as green roof implementation in hotter and drier climates increases, there is a need to evaluate plant performance under these climatic conditions. Succulents have been shown to be successful in hot and dry green roofs, although survival differs between species and the role of leaf succulence in survival has not been fully explored. For non-succulent plants, habitats with conditions similar to green roofs (‘habitat templates’) have been used to select plants, although few studies have discussed the performance of these selections under green roof conditions. Therefore, we evaluated establishment of 32 plant species on an unirrigated extensive (125 mm deep) green roof in Melbourne, Australia over a 42 week period (from winter through summer into autumn). Plants were selected on the basis of life-form, succulence, appropriate habitat templates and/or successful use on green roofs internationally. Climatic conditions during the experiment were often extreme, with evaporation regularly exceeding rainfall and a hot and dry summer (mean maximum air temperature 35 °C and 80.6 mm total rainfall), leading to roof temperatures of 65 °C. After 42 weeks, only succulent plants remained alive and only three of the succulent species had 100% survival. Survival was positively related to the degree of leaf succulence (g H2O leaf area cm−1) making this a useful trait for plant selection for unirrigated green roofs in hot, dry climates. The failure of most species, despite being chosen from appropriate habitats, demonstrates the need to evaluate potential plants on green roofs under extreme climatic conditions. Supplementary irrigation may be essential to sustain non-succulent species during extreme weather in hot and dry climates.

Introduction

In cities worldwide, green roofs are increasingly being constructed because of their multiple environmental, social and economic benefits, including improved stormwater management, increased longevity of roof membranes, reductions in building heat loads, contributions to wildlife habitat provision and reduced noise and air pollution (Oberndorfer et al., 2007, Rowe, 2011). The vegetation used in lightweight, shallow extensive green roofs is generally restricted to low growing and shallow rooted perennial species with multiple tolerances to drought, heat, cold, wind and pests and diseases (Snodgrass and Snodgrass, 2006).

Plant selection for extensive green roofs has largely been based around experience, observation and research from temperate regions of Europe and North America (Dvorak and Volder, 2010, Oberndorfer et al., 2007) where drought periods are measured in weeks rather than months, and where ambient air temperatures rarely exceed 35 °C (Lundholm et al., 2010, Nagase and Dunnett, 2010). However, as green roof implementation in hotter and drier climates is expanding, there is a need to evaluate plant establishment on green roofs in these regions (Benvenuti and Bacci, 2010, Bousselot et al., 2011, Papafotiou et al., 2013, Van Mechelen et al., 2014). This is important as plant failure has been identified as a major barrier to developing successful green roofs in these regions, particularly where the green roof industry is still in its infancy (Williams et al., 2010) and plant growth conditions on green roofs are often harsher than in green roof modules at ground level or in pot-based experiments (Dvorak and Volder, 2010). Differences in climate and plant availability mean it is not appropriate to assume that the same plant species will be universally suitable (Williams et al., 2010).

Traditionally two approaches for plant selection have been used: the plant trait approach and the habitat template approach. The plant trait approach uses traits such as succulence to select drought tolerant plants for green roofs. To date, the most common plants used on green roofs worldwide are succulents, principally Sedum species (Getter and Rowe, 2008, Monterusso et al., 2005, Snodgrass and Snodgrass, 2006). They achieve drought tolerance through low water use, leaf water storage (Farrell et al., 2012) and often Crassulacean Acid Metabolism (CAM) (Butler et al., 2011, Durhman et al., 2006). CAM leads to greater water use efficiency through opening of stomata at night for CO2 uptake, greatly reducing water lost through transpiration per unit CO2 fixed (Sayed, 2001). Leaf water storage, or succulence, (expressed as g water cm−2 leaf area) enables plants to survive dry conditions by providing water that can be used to maintain plant function when water is unavailable for uptake by roots (von Willert, 1992). Generally, plants with greater succulence are able to survive longer periods without water (von Willert, 1992), however, many common green roof Sedum species such as Sedum acre and Sedum album exhibit reduced succulence, enabling these species to also tolerate winter frosts and sub-zero temperatures in temperate climates (Osmond et al., 1975, Teeri et al., 1981). Consequently, Sedum species with low succulence may be less suitable for green roofs in hot and dry climates than species with greater succulence (Williams et al., 2010). The habitat template approach is based on selecting plants from habitats analogous to extensive green roofs, such as rocky barrens or outcrops (Farrell et al., 2013, MacIvor and Lundholm, 2011), prairies or grasslands (Sutton et al., 2012), and annual and perennial wildflowers from agricultural systems and roadside environments (Benvenuti, 2014). For example in an analysis of species from shallow soil habitats in southern France, Van Mechelen et al., (2014) identified 28 plant species as being potentially suitable for hot and dry Mediterranean green roofs. However, as demonstrated in a number of green roof experiments, theoretical suitability does not guarantee survival in hot and dry conditions (Benvenuti and Bacci, 2010, Vestrella et al., 2015, Zhang et al., 2014), confirming the need to evaluate new potential species in green roofs. Differences in the performance and survival of these plants under water limitation often reflect differences in plant traits (Lundholm et al., 2015) and physiological strategies for dealing with water stress (Farrell et al., 2013, Raimondo et al., 2015).

While researchers have evaluated plant establishment through summer on green roofs with species chosen using the habitat template approach, few have looked at extreme conditions, where elevated temperatures and rainfall deficits of several months create extremely hostile conditions for plant survival. Further, although succulence has been used extensively as a trait to select plants for green roofs, the influence of the degree of succulence on plant survival has not been investigated. Therefore, we evaluated plant establishment over 42 weeks in an extensive green roof in Melbourne, Australia. Thirty two plant species were selected on the basis of leaf succulence (succulent species only), appropriate green roof habitat templates or use on green roofs internationally. The objectives of the study were to (1) compare and evaluate plant health, vigour and survival of the 32 species and (2) explore the effects of differences in life-forms and degree of leaf succulence on plant establishment.

Section snippets

Study location and climate

A 20 m2 extensive green roof was constructed on a 6.6 m high brick building at the Burnley Campus of The University of Melbourne, Australia (37°47′S; 144°58′E). Melbourne's climate is characterised by cool, wet winters and dry, hot summers. Melbourne's mean annual rainfall is 648.4 mm (1855 to 2015), with 153.9 mm falling in summer months (Bureau of Meteorology, 2014). Mean annual temperatures range between 10.2 and 19.9 °C with maximum mean temperatures in January (25.9 °C) and minimum mean

Climatic conditions and green roof temperatures

Climatic conditions over the 42 week experimental period are shown in Fig. 2, Fig. 3. Spring (September to November) maximum temperatures were generally below 25 °C, consistent with mean monthly maximum temperatures (see Section 2.1 for long-term climate information). However, rainfall over spring was un-seasonally low (80.2 mm, Fig. 2); compared to the long-term average of 163 mm (Bureau of Meteorology, 2014). Low rainfall coupled with high evaporative demand during spring meant that there were

Discussion

Climatic conditions on the green roof were extreme, average maximum summer temperatures were 35 °C and evaporation exceeded rainfall (CWI < 0.5) for 33 of the 42 weeks. The gradual decline in health and vigour of all plants in the experiment is not surprising and compares to other green roof plant studies where plant growth declined with increasing dryness and heat (Dvorak and Volder, 2013, Nagase and Dunnett, 2010). Succulents were the only species to survive the entire experiment, however not

Conclusions

Approaches to plant selection for green roofs have typically been based around matching similar habitats to a green roof, i.e. habitat template (Lundholm, 2006, Van Mechelen et al., 2014); or identifying plants with suitable traits such as succulence (Farrell et al., 2013). The failure of most species in this experiment to survive during establishment, despite being chosen from appropriate dryland habitats, demonstrates the need to also evaluate plants under field conditions for green roof use

Acknowledgements

We thank Burnley technical staff Sasha Andrusiak, Ruth Mitchell, Nick Osborne and Ross Payne for their help with this experiment and colleague Dr. Steven Livesley for his review of the manuscript.

References (48)

  • D.B. Rowe et al.

    Effect of green roof media depth on Crassulacean plant succession over seven years

    Landscape Urban Plann.

    (2012)
  • C. Van Mechelen et al.

    Plant trait analysis delivers an extensive list of potential green roof species for Mediterranean France

    Ecol. Eng.

    (2014)
  • D.A. White et al.

    Managing productivity and drought risk in Eucalyptus globulus plantations in south-western Australia

    For. Ecol. Manage.

    (2009)
  • N.S.G. Williams et al.

    Green roofs for a wide brown land: opportunities and barriers for rooftop greening in Australia

    Urban For. Urban Greening

    (2010)
  • D. Wolf et al.

    Water uptake in green roof microcosms: effects of plant species and water availability

    Ecol. Eng.

    (2008)
  • H. Zhang et al.

    Effect of substrate depth on 18 non-succulent herbaceous perennials for extensive green roofs in a region with a dry spring

    Ecol. Eng.

    (2014)
  • Australian Standard

    AS 3743-2003 Potting Mixes

    (2003)
  • Australian Standard

    AS 4419-2003 Soils for Landscaping and Garden Use

    (2003)
  • S. Benvenuti et al.

    Initial agronomic performances of Mediterranean xerophytes in simulated dry green roofs

    Urban Ecosyst.

    (2010)
  • J.M. Bousselot et al.

    Moisture content of extensive green roof substrate and growth response of 15 temperate plant species during dry down

    HortScience

    (2011)
  • Bureau of Meteorology

    Climate statistics for Australian Locations

    (2014)
  • C. Butler et al.

    Plasticity in CAM-C3 photosynthesis in eight species of green roof Sedum

  • CSIRO et al.

    Climate Change in Australia: Technical Report 2007 Canberra

    (2007)
  • A.K. Durhman et al.

    Effect of watering regimen on chlorophyll fluorescence and growth of selected green roof plant taxa

    HortScience

    (2006)
  • Cited by (85)

    View all citing articles on Scopus
    View full text