- Email: [email protected]
The effects of substrate depth heterogeneity on plant species coexistence on an extensive green roof
Ecological Engineering 68 (2014) 184–188
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Short communication
The effects of substrate depth heterogeneity on plant species coexistence on an extensive green roof Amy Heim ∗ , Jeremy Lundholm Biology Department, Saint Mary’s University, 923 Robie St., Halifax, NS B3H 3C3, Canada
a r t i c l e
i n f o
Article history: Received 4 December 2013 Received in revised form 15 January 2014 Accepted 25 March 2014 Available online 4 May 2014 Keywords: Constructed ecosystem Soil depth variability Interspecific competition Habitat heterogeneity
a b s t r a c t Green roofs are often planted with mixtures of plant species. Studies of green roof plant community composition over time show that species diversity often declines, in part due to competition between plant species. The incorporation of substrate depth heterogeneity into green roof designs is expected to increase habitat heterogeneity, and could reduce interspecific competition among plants, but this has not been tested empirically. Two species with contrasting responses to substrate depth and water availability, Festuca rubra and Sedum acre, were planted in a rooftop substrate depth heterogeneity experiment. There were a total of four treatments: three homogeneous treatments (substrate depth 5 cm, 10 cm, and 15 cm) and one heterogeneous treatment (patches with substrate depths of 5 cm and 15 cm), with the same substrate volume and average depth as the 10 cm homogeneous treatment. By the end of the study period there was little difference in the ratio of species’ cover between the 10 cm and 5/15 cm treatment. However, there was significantly less spread of each species into areas planted with the other species in the 5/15 cm treatment compared to the 10 cm treatment, and the 5/15 cm (heterogeneous treatment) had significantly greater overall plant cover. The results suggest that, while the effects are subtle, soil depth heterogeneity could allow coexistence of species associated with different conditions for longer than homogeneous conditions, and could result in greater plant species diversity in green roof ecosystems. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Recent research indicates that the choice of plant species used in green roof vegetation can affect the provisioning of ecosystem services such as storm water retention (Dunnett et al., 2008a; MacIvor and Lundholm, 2011; Nardini et al., 2012; Nagase and Dunnett, 2012), reduction of hot-season roof temperatures (Lundholm et al., 2010), and biomass production and carbon capture (Getter et al., 2009; Song et al., 2013). Such differences in plant performance stem from anatomical, morphological and physiological traits of plant species (Farrell et al., 2012, 2013a) and can be used to engineer green roof ecosystems for optimal performance. Species mixtures have the potential to improve green roof functioning relative to monocultures (Ranalli and Lundholm, 2008; Cook-Patton and Bauerle, 2012). Combinations of functionally distinct species are predicted to enhance overall resource consumption and to increase plant biomass (Kinzig et al., 2001), which
∗ Corresponding author. Tel.: +1 585 749 0890. E-mail addresses: [email protected] (A. Heim), [email protected] (J. Lundholm). http://dx.doi.org/10.1016/j.ecoleng.2014.03.023 0925-8574/© 2014 Elsevier B.V. All rights reserved.
in turn may affect several green roof ecosystem services (Lundholm et al., 2010). In natural plant communities, sustained coexistence is expected only when species competing for resources have key differences in their resource requirements or in their responses to the abiotic environment (Chesson, 2000). While there are many possible ways in which such niche differentiation can allow greater species diversity, some of the most common mechanisms involve spatial heterogeneity in the environment. Plant species that vary in their response to environmental conditions are thought to be able to coexist if the environment is heterogeneous and the species can disperse to those microsites with optimal characteristics for their growth and survival (Lundholm, 2009). In engineered ecosystems, we can incorporate heterogeneity at the appropriate scales and plant species associated with these different conditions. Spatial heterogeneity has been incorporated into green roof designs with the purpose of increasing species diversity of arthropods and plants (Brenneisen, 2006). Green roof designers vary substrate depths and create habitat heterogeneity by adding surface features such as coarse woody debris, piles of stones or different kinds of substrate (Bates et al., 2013). These measures are thought to encourage greater species diversity in invertebrate and
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
Fig. 1. Orientation of F. rubra and S. acre in the four different treatments.
plant communities by allowing more species to find appropriate microsites than on a more homogeneous roof. While there is little published on the effects of these designed features, one empirical study has shown a positive relationship between indices of habitat heterogeneity and species diversity of invertebrates (Gedge and Kadas, 2005). A recent study comparing three types of green roofs also showed that roofs with more structural complexity in the vegetation (i.e. spatial heterogeneity generated by the plant canopy) did support more species of arthropods (Madre et al., 2013). Bates et al. (2013) observed that the survival of certain plant species is facilitated by specific microsites on green roofs where habitat heterogeneity was purposefully incorporated into the system, warranting further exploration of the role of spatial heterogeneity on green roofs. Spatial heterogeneity in substrate depth should create corresponding variability in substrate moisture and thus support both species that require wetter conditions and droughttolerant species, leading to greater overall diversity (Brenneisen, 2006; Köhler and Poll, 2010). Here we hypothesized that two species with contrasting responses to substrate depth would display a greater ability to coexist under conditions of heterogeneous substrate depth, compared with homogeneous conditions. This study involved experimentally comparing green roof systems featuring heterogeneous substrate depths with homogeneous substrate depths, while controlling for mean substrate depth. We used a grass, expected to dominate in deeper substrates, and a Sedum, expected to dominate in shallower substrates. We expected plant sizes and coverage between the two species to be more even in the heterogeneous vs. homogeneous treatments, indicating a greater likelihood of coexistence.
185
order to maintain an even substrate surface, so we could examine the effects of substrate depth alone without also changing the vertical profile of the substrate. According to previous research, we predicted that the Sedum should outperform the grass at 5 cm and the grass should outperform the Sedum at 15 cm in terms of growth (Dunnett and Kingsbury, 2004). Therefore, the mixed 5/15 cm substrate depth should decrease the competition between these two species. The volume of substrate used in the 10 cm treatment was equal to that used in the 5/15 cm treatment, and these two treatments had the same average substrate depth, allowing us to separate the potential effects of average substrate depth from those of spatial heterogeneity in substrate depth (e.g. Vivian-Smith, 1997). The treatments consisted of 24 wooden planter boxes (61 cm × 61 cm); which were 15 cm high with no base. A nurserygrade weed control fabric (Quest Home & Garden, Mississauga, ON, CA) was placed under the boxes to prevent damage to the roof. To create four different substrate depth treatments, 5 cm thick concrete slabs (length and width of 60.96 cm) were placed in the wooden boxes to manipulate substrate depth. Two concrete slabs were used for the 5 cm substrate depth, one for the 10 cm depth and no concrete slabs were used for the 15 cm substrate depth. The 5/15 cm substrate depth treatment involved four concrete slabs, each 30.48 cm by 30.48 cm with a thickness of 5 cm, placed two high diagonally across from each other in a wooden box, to create two squares with a substrate depth of 5 cm and two with 15 cm. A root barrier/water retention fleece was placed in all boxes above the concrete slabs (EnkaRetain and Drain 3111® , Colbond Inc., NC, USA). The boxes were then filled to the rim with Sopraflor X substrate purchased in 2012 (Soprema Inc., Drummondville, QC, CA) (for substrate chemistry, see supplemental data). Plant species used included Sedum acre and Festuca rubra, which were chosen due to their different drought tolerance and water usage, as observed in previous trials (MacIvor and Lundholm, 2011). Both species were harvested in May 2012 from previous experiments at Saint Mary’s and the Dartmouth Commons (S. acre only) in Dartmouth, Nova Scotia, Canada. Once harvested, plants were transplanted directly into the planter boxes until ∼25–45% cover was achieved in each quarter of the box. Initially, each quadrant contained either 1–3 individual S. acre plants (average height 7.4 cm) or 2–12 individual F. rubra plants (average height 22.6 cm). After planting, the vegetation was watered twice over a twoweek period to encourage establishment. All planter boxes were divided into four quadrants, each containing plants from one of the two species (two squares per species per planter box, with duplicates of the same species arranged diagonally). In the heterogeneous treatment (5/15 cm), S. acre was planted in the 5 cm depth quadrants, and F. rubra was planted in the 15 cm depth quadrants. Data was collected monthly throughout the 2012 and 2013 growing seasons (June–September). Cover for each species
2. Methods The study site was located on the roof of the five-story Atrium building at Saint Mary’s University in Halifax, Nova Scotia, Canada (44◦ 39 N, 63◦ 35 W) (Fig. 1). This experiment included four different substrate depth treatments: three homogeneous treatments, 15 cm, 10 cm, and 5 cm and one heterogeneous treatment, wherein half of the substrate had a depth of 5 cm, and the other half a depth of 15 cm (henceforth, the 5/15 cm treatment). We created the depth treatments by varying the depth below the substrate surface, in
Fig. 2. The spread of F. rubra and S. acre from their original 30.48 by 30.48 cm planting by the end of the study period (September 18, 2013). The cover change was compared between both species, those bars that share a letter are not significantly different.
186
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
Fig. 3. Total monthly cover F. rubra (F) + S. acre (S) during the 2012 (A) and 2013 (B) growing season. Monthly ratio of cover F. rubra (F)/S. acre (S) during the 2012 (C) and 2013 (D) growing season. For each month, the bars that share a letter are not significantly different.
was determined using photographs analyzed in ImageJ (Image Processing and Analysis in Java, http://rsbweb.nih.gov/ij/). Cover ranged from 0 to 1 and represented the proportion of the substrate surface covered by plants. Total cover for each replicate was obtained by summing the cover of both species within the replicate. Occasionally these two species overlapped within the planter box resulting in a total cover greater than 1.0. In order to determine the spread of F. rubra or S. acre outside their initial location, the point interception method (Floyd and Anderson, 1987) was used. A sampling frame was placed over the entire planter box. The number of times the aboveground biomass of S. acre or F. rubra touched the pin (diameter 3.7 cm) outside their initial quadrant was recorded. The pin interceptions were taken from 20 different locations in the planter box (5 per quadrant). 2.1. Statistical methods Relative Cover Change (RCC), total cover and ratio of cover were compared across the substrate depth treatments using 1-way ANOVAs with Tukey post hoc tests. All analyses used substrate depth treatment as the independent variable. The ratio in cover was determined by dividing the final cover of F. rubra by the final cover of S. acre. The data gathered from the point interception method (hits) was used to calculate the RCC of the spread of S. acre and F. rubra outside their original planting. There were no initial hits when calculating this migration, therefore, the formula used to calculate the RCC was modified to accommodate this: RCC = (ln(coverT2 ) − 0)/# of days). 3. Results 3.1. Cover During the 2012 growing season there was no significant difference in total cover between the 10 cm and 5/15 cm treatments. During August and September 2013 cover in the 10 cm treatment was significantly lower than the 5/15 cm treatment. The ratio of cover was only significantly lower in the 10 cm treatment compared to the 5/15 cm treatment for August and September 2012 and June and July 2013 (Fig. 2).
3.2. Spread outside of planted areas For the two treatments with an average substrate depth of 10 cm (the 10 cm and 5/15 cm treatments) the 10 cm treatment had significantly greater RCC, for both species, outside its original planting than the 5/15 cm treatment (Fig. 3). For both species, this spread was mainly caused by seedling dispersal.
4. Discussion and conclusions In this study, we wanted to determine if spatially variable substrate depths could allow more even growth and coverage between two species that were not expected to coexist in homogeneous conditions. Although we used only two species, these are representative of two broad classes of plants used on green roofs: succulents and grasses. The key comparisons in this study were between the 10 cm and 5/15 cm depth treatments, as both of these had the same mean depth, total weight, and volume of substrate. This allowed for an evaluation of the effects of substrate depth heterogeneity in isolation. The two other homogeneous substrate depth treatments (5 and 15 cm) are included for comparison, but do not have corresponding heterogeneous depth treatments. While we expected more even distributions of plant cover between the two species in the heterogeneous depth treatment compared to the uniform 10 cm depth treatment, this only occurred at the end of the first year of the study and the beginning of the second year of study. By the end of the second year, both heterogeneous and homogeneous treatments at 10 cm average depth had equivalent cover ratios, close to 1.5, compared to the 5 cm homogeneous treatment, which was strongly biased in favor of S. acre, while the 15 cm homogenous treatment was biased in favor of F. rubra. This supports many previous studies on extensive green roofs that indicate that Sedum tend to dominate in lower substrate depths, while greater depths are required for the persistence of grasses. While two growing seasons were insufficient for complete exclusion of either species in any of the treatments, previous studies show that species exclusion can happen within the first year (Rowe et al., 2012). A sevenyear study that included S. acre in mixtures with other succulents showed that it tended to drop out of mixtures in deeper substrate depths (Rowe et al., 2012), likely due to competition with larger
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
species. In another multi-year study, (Dunnett et al., 2008b) also found that S. acre had an advantage at lower depth (10 cm) in the first two years but declined in abundance after year two. A study over two decades in Germany showed that S. acre and a different species of Festuca (ovina) both persisted over the entire study on a roof with homogeneous 10 cm substrate depth (Köhler, 2006). In this study cover ratios were roughly equivalent in both 10 cm homogeneous and 5/15 cm heterogeneous treatments, and by the end of the second growing season both were biased in favor of the larger F. rubra. On the other hand, the 5/15 cm treatment showed significantly less spread of each species into the areas planted with the other species. This result is consistent with what we observed in the homogeneous 5 cm and 15 cm treatments: S. acre did poorly in the 15 cm homogeneous treatment, and had less ability to spread into 15 cm deep patches within the 5/15 cm heterogeneous treatment; F. rubra displayed the opposite pattern, a lower ability to spread into 5 cm sections in the 5/15 cm treatment. This study shows a strong effect of substrate depth, with dominance of F. rubra at mean substrate depths of 10 cm or greater, and dominance of S. acre at 5 cm. The decrease in the performance of F. rubra at 5 cm was most likely due to a lack of water, not competition, since S. acre growth was mainly observed around the edges of F. rubra. Given that F. rubra performed poorly at 5 cm depth, both when planted in the 5 cm homogeneous treatment and when spreading from 15 cm patches into 5 cm patches within the 5/15 cm treatment, results suggest that persistence of S. acre will be greater in the 5/15 cm treatment, as the 5 cm depth patches can act as refugia for the succulent species. There appeared to be a more even distribution of cover at the 5/15 cm substrate depth until August 2013. At this point both the 10 cm and 5/15 cm treatments had very similar ratios with almost 50/50 cover between the two species. The 2013 growing season had very mild weather, likely favoring the less stress-tolerant species F. rubra; it is possible that a harsher summer would have resulted in a more even distribution of the canopy cover in the 5/15 cm treatment, as was observed in the 2012 growing season. The effects of fluctuations in growing season conditions warrant further investigation in green roof plant species coexistence. Total average plant cover in 5/15 cm treatment was significantly greater than the 10 cm treatment for August and September 2013, although both treatments had the same amount of substrate. This indicates that a heterogeneous substrate depth could result in greater cover with the same amount of substrate as a homogeneous substrate depth, but this could also depend on the responses of the individual species to growing season conditions. The most common way to establish substrate depth heterogeneity on green roofs is to create areas with less depth (hollows), contrasting with areas where the substrate is mounded (Brenneisen, 2006; Köhler and Poll, 2010; Gedge et al., 2013). This creates heterogeneity not only in substrate depth, but possibly also in exposure, where mounds would be more prone to wind and desiccation, being above the main grade of the roof, and in hydrology, where water could drain off the mounded areas into the hollows. This is expected to contribute to greater diversity of invertebrates as well as plants (Brenneisen, 2006). In this study, we maintained a uniform substrate surface profile, so that we could explore the effects of substrate depth variability alone. In this study, concrete slabs were used to manipulate substrate depth. However, due to weight restrictions, this method is not feasible for many green roof systems. If desired, this kind of heterogeneity could easily be created in green roof installations by embedding lightweight objects (styrene foam blocks, wooden planks, pumice or other lightweight materials) under the substrate, creating variability in depth but
187
not surface topography. Alternate materials, including additives that enhance water retention (e.g. Farrell et al., 2013b), or different kinds of drainage layers (e.g. Savi et al., 2013) could also be used to create functional heterogeneity that might increase coexistence between plant species. While substrate depth heterogeneity had no detrimental effects on green roof vegetation structure in this study, other potential effects of such treatments should be explored including invertebrate habitat provisioning, carbon capture, and thermal performance. Finally, experiments need to be complemented with “field studies” of previously established green roofs that vary in the degree of habitat heterogeneity, plant species richness, and time since establishment, to determine which taxa are actually affected by these variables and whether heterogeneity can influence longterm coexistence.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.ecoleng.2014.03.023.
References Bates, A., Sadlera, J., Mackay, R., 2013. Vegetation development over four years on two green roofs in the UK. Urban Forest. Urban Green. 12, 98–108. Brenneisen, S., 2006. Space for urban wildlife: designing green roofs as habitats in Switzerland. Urban Habitats 4, 27–36. Chesson, P., 2000. Mechanisms of maintenance of species diversity. Annu. Rev. Ecol. Syst. 31, 343–366. Cook-Patton, S.C., Bauerle, T.L., 2012. Potential benefits of plant diversity on vegetated roofs: a literature review. J. Environ. Manage. 106, 85–92. Dunnett, N., Nagase, A., Booth, R., Grime, P., 2008a. Influence of vegetation composition on runoff in two simulated green roof experiments. Urban Ecosyst. 11 (4), 385–398. Dunnett, N., Nagase, A., Hallam, A., 2008b. The dynamics of planted and colonising species on a green roof over six growing seasons 2001–2006: influence of substrate depth. Urban Ecosyst. 11 (4), 373–384. Dunnett, N., Kingsbury, N., 2004. Planting Green Roofs and Living Walls, vol. 75. Timber Press, Portland, OR, pp. 104–114. Farrell, C., Szota, C., Williams, N.S., Arndt, S.K., 2013a. High water users can be drought tolerant: using physiological traits for green roof plant selection. Plant Soil 372, 177–193. Farrell, C., Ang, X.Q., Rayner, J.P., 2013b. Water-retention additives increase plant available water in green roof substrates. Ecol. Eng. 52, 112–118. Farrell, C., Mitchell, R.E., Szota, C., Rayner, J.P., Williams, N.S.G., 2012. Green roofs for hot and dry climates: interacting effects of plant water use, succulence and substrate. Ecol. Eng. 49, 270–276. Floyd, D., Anderson, J., 1987. A comparison of three methods for estimating plant cover. J. Ecol. 75 (1), 221–228. Gedge, D., Grant, G., Kadas, G., Dinham, C., 2013. Creating Green Roofs for Invertebrates A Best Practice Guide. Buglife – The Invertebrate Conservation Trust, Available at: http://www.buglife.org.uk/sites/default/files/Creating% 20Green%20Roofs%20for%20Invertebrates Best%20practice%20guidance 2.pdf (accessed 28.10.13). Gedge, D., Kadas, G., 2005. Green roofs and biodiversity. Biologist 52, 161–169. Getter, K.L., Rowe, D.B., Robertson, G.P., Cregg, B.M., Andresen, J.A., 2009. Carbon sequestration potential of extensive green roofs. Environ. Sci. Technol. 43 (19), 7564–7570. Kinzig, A.P., Pacala, S.W., Tilman, D., 2001. The Functional Consequences of Biodiversity. Princeton University Press, Princeton, pp. 365. Köhler, M., Poll, P.H., 2010. Long-term performance of selected old Berlin greenroofs in comparison to younger extensive greenroofs in Berlin. Ecol. Eng. 36 (5), 722–729. Köhler, M., 2006. Long-term vegetation research on two extensive green roofs in Berlin. Urban Habitats 4 (1), 3–26. Lundholm, J., MacIvor, J.S., MacDougall, Z., Ranalli, M., 2010. Plant Species and Functional Group Combinations Affect Green Roof Ecosystem Functions. Plos One. 5, 3. Lundholm, J.T., 2009. Plant species diversity and environmental heterogeneity: spatial scale and competing hypotheses. J. Veg. Sci. 20 (3), 377–391. MacIvor, S., Lundholm, J., 2011. Performance evaluation of native plants suited to extensive green roof conditions in a maritime climate. Ecol. Eng. 37 (3), 407– 417. Madre, F., Vergnes, A., Machon, N., Clergeau, P., 2013. A comparison of 3 types of green roof as habitats for arthropods. Ecol. Eng. 57 (4), 109–117.
188
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
Nagase, A., Dunnett, N., 2012. Amount of water runoff from different vegetation types on extensive green roofs: effects of plant species, diversity and plant structure. Landscape Urban Plan. 104 (3–4), 356–363. Nardini, A., Andri, S., Crasso, M., 2012. Influence of substrate depth and vegetation type on temperature and water runoff mitigation by extensive green roofs: shrubs versus herbaceous plants. Urban Ecosyst. 15 (3), 697–708. Ranalli, M., Lundholm, J., 2008. Biodiversity and ecosystem function in constructed ecosystems. Perspect. Agric. Vet. Sci. Nutr. Nat. Resour. 3 (034), 1–16.
Rowe, D.B., Getter, K.L., Durhman, A.K., 2012. Effect of green roof media depth on Crassulacean plant succession over seven years. Landscape Urban Plan. 104, 310–319. Savi, T., Andri, S., Nardini, A., 2013. Impact of different green roof layering on plant water status and drought survival. Ecol. Eng. 57, 188–196. Song, U., Kim, E., Bang, J.H., Son, D.J., Waldman, B., Lee, E.J., 2013. Wetlands are an effective green roof system. Build. Environ. 66, 141–147. Vivian-Smith, G., 1997. Microtopographic heterogeneity and floristic diversity in experimental wetland communities. J. Ecol. 85 (1), 71–82.
Contents lists available at ScienceDirect
Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng
Short communication
The effects of substrate depth heterogeneity on plant species coexistence on an extensive green roof Amy Heim ∗ , Jeremy Lundholm Biology Department, Saint Mary’s University, 923 Robie St., Halifax, NS B3H 3C3, Canada
a r t i c l e
i n f o
Article history: Received 4 December 2013 Received in revised form 15 January 2014 Accepted 25 March 2014 Available online 4 May 2014 Keywords: Constructed ecosystem Soil depth variability Interspecific competition Habitat heterogeneity
a b s t r a c t Green roofs are often planted with mixtures of plant species. Studies of green roof plant community composition over time show that species diversity often declines, in part due to competition between plant species. The incorporation of substrate depth heterogeneity into green roof designs is expected to increase habitat heterogeneity, and could reduce interspecific competition among plants, but this has not been tested empirically. Two species with contrasting responses to substrate depth and water availability, Festuca rubra and Sedum acre, were planted in a rooftop substrate depth heterogeneity experiment. There were a total of four treatments: three homogeneous treatments (substrate depth 5 cm, 10 cm, and 15 cm) and one heterogeneous treatment (patches with substrate depths of 5 cm and 15 cm), with the same substrate volume and average depth as the 10 cm homogeneous treatment. By the end of the study period there was little difference in the ratio of species’ cover between the 10 cm and 5/15 cm treatment. However, there was significantly less spread of each species into areas planted with the other species in the 5/15 cm treatment compared to the 10 cm treatment, and the 5/15 cm (heterogeneous treatment) had significantly greater overall plant cover. The results suggest that, while the effects are subtle, soil depth heterogeneity could allow coexistence of species associated with different conditions for longer than homogeneous conditions, and could result in greater plant species diversity in green roof ecosystems. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Recent research indicates that the choice of plant species used in green roof vegetation can affect the provisioning of ecosystem services such as storm water retention (Dunnett et al., 2008a; MacIvor and Lundholm, 2011; Nardini et al., 2012; Nagase and Dunnett, 2012), reduction of hot-season roof temperatures (Lundholm et al., 2010), and biomass production and carbon capture (Getter et al., 2009; Song et al., 2013). Such differences in plant performance stem from anatomical, morphological and physiological traits of plant species (Farrell et al., 2012, 2013a) and can be used to engineer green roof ecosystems for optimal performance. Species mixtures have the potential to improve green roof functioning relative to monocultures (Ranalli and Lundholm, 2008; Cook-Patton and Bauerle, 2012). Combinations of functionally distinct species are predicted to enhance overall resource consumption and to increase plant biomass (Kinzig et al., 2001), which
∗ Corresponding author. Tel.: +1 585 749 0890. E-mail addresses: [email protected] (A. Heim), [email protected] (J. Lundholm). http://dx.doi.org/10.1016/j.ecoleng.2014.03.023 0925-8574/© 2014 Elsevier B.V. All rights reserved.
in turn may affect several green roof ecosystem services (Lundholm et al., 2010). In natural plant communities, sustained coexistence is expected only when species competing for resources have key differences in their resource requirements or in their responses to the abiotic environment (Chesson, 2000). While there are many possible ways in which such niche differentiation can allow greater species diversity, some of the most common mechanisms involve spatial heterogeneity in the environment. Plant species that vary in their response to environmental conditions are thought to be able to coexist if the environment is heterogeneous and the species can disperse to those microsites with optimal characteristics for their growth and survival (Lundholm, 2009). In engineered ecosystems, we can incorporate heterogeneity at the appropriate scales and plant species associated with these different conditions. Spatial heterogeneity has been incorporated into green roof designs with the purpose of increasing species diversity of arthropods and plants (Brenneisen, 2006). Green roof designers vary substrate depths and create habitat heterogeneity by adding surface features such as coarse woody debris, piles of stones or different kinds of substrate (Bates et al., 2013). These measures are thought to encourage greater species diversity in invertebrate and
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
Fig. 1. Orientation of F. rubra and S. acre in the four different treatments.
plant communities by allowing more species to find appropriate microsites than on a more homogeneous roof. While there is little published on the effects of these designed features, one empirical study has shown a positive relationship between indices of habitat heterogeneity and species diversity of invertebrates (Gedge and Kadas, 2005). A recent study comparing three types of green roofs also showed that roofs with more structural complexity in the vegetation (i.e. spatial heterogeneity generated by the plant canopy) did support more species of arthropods (Madre et al., 2013). Bates et al. (2013) observed that the survival of certain plant species is facilitated by specific microsites on green roofs where habitat heterogeneity was purposefully incorporated into the system, warranting further exploration of the role of spatial heterogeneity on green roofs. Spatial heterogeneity in substrate depth should create corresponding variability in substrate moisture and thus support both species that require wetter conditions and droughttolerant species, leading to greater overall diversity (Brenneisen, 2006; Köhler and Poll, 2010). Here we hypothesized that two species with contrasting responses to substrate depth would display a greater ability to coexist under conditions of heterogeneous substrate depth, compared with homogeneous conditions. This study involved experimentally comparing green roof systems featuring heterogeneous substrate depths with homogeneous substrate depths, while controlling for mean substrate depth. We used a grass, expected to dominate in deeper substrates, and a Sedum, expected to dominate in shallower substrates. We expected plant sizes and coverage between the two species to be more even in the heterogeneous vs. homogeneous treatments, indicating a greater likelihood of coexistence.
185
order to maintain an even substrate surface, so we could examine the effects of substrate depth alone without also changing the vertical profile of the substrate. According to previous research, we predicted that the Sedum should outperform the grass at 5 cm and the grass should outperform the Sedum at 15 cm in terms of growth (Dunnett and Kingsbury, 2004). Therefore, the mixed 5/15 cm substrate depth should decrease the competition between these two species. The volume of substrate used in the 10 cm treatment was equal to that used in the 5/15 cm treatment, and these two treatments had the same average substrate depth, allowing us to separate the potential effects of average substrate depth from those of spatial heterogeneity in substrate depth (e.g. Vivian-Smith, 1997). The treatments consisted of 24 wooden planter boxes (61 cm × 61 cm); which were 15 cm high with no base. A nurserygrade weed control fabric (Quest Home & Garden, Mississauga, ON, CA) was placed under the boxes to prevent damage to the roof. To create four different substrate depth treatments, 5 cm thick concrete slabs (length and width of 60.96 cm) were placed in the wooden boxes to manipulate substrate depth. Two concrete slabs were used for the 5 cm substrate depth, one for the 10 cm depth and no concrete slabs were used for the 15 cm substrate depth. The 5/15 cm substrate depth treatment involved four concrete slabs, each 30.48 cm by 30.48 cm with a thickness of 5 cm, placed two high diagonally across from each other in a wooden box, to create two squares with a substrate depth of 5 cm and two with 15 cm. A root barrier/water retention fleece was placed in all boxes above the concrete slabs (EnkaRetain and Drain 3111® , Colbond Inc., NC, USA). The boxes were then filled to the rim with Sopraflor X substrate purchased in 2012 (Soprema Inc., Drummondville, QC, CA) (for substrate chemistry, see supplemental data). Plant species used included Sedum acre and Festuca rubra, which were chosen due to their different drought tolerance and water usage, as observed in previous trials (MacIvor and Lundholm, 2011). Both species were harvested in May 2012 from previous experiments at Saint Mary’s and the Dartmouth Commons (S. acre only) in Dartmouth, Nova Scotia, Canada. Once harvested, plants were transplanted directly into the planter boxes until ∼25–45% cover was achieved in each quarter of the box. Initially, each quadrant contained either 1–3 individual S. acre plants (average height 7.4 cm) or 2–12 individual F. rubra plants (average height 22.6 cm). After planting, the vegetation was watered twice over a twoweek period to encourage establishment. All planter boxes were divided into four quadrants, each containing plants from one of the two species (two squares per species per planter box, with duplicates of the same species arranged diagonally). In the heterogeneous treatment (5/15 cm), S. acre was planted in the 5 cm depth quadrants, and F. rubra was planted in the 15 cm depth quadrants. Data was collected monthly throughout the 2012 and 2013 growing seasons (June–September). Cover for each species
2. Methods The study site was located on the roof of the five-story Atrium building at Saint Mary’s University in Halifax, Nova Scotia, Canada (44◦ 39 N, 63◦ 35 W) (Fig. 1). This experiment included four different substrate depth treatments: three homogeneous treatments, 15 cm, 10 cm, and 5 cm and one heterogeneous treatment, wherein half of the substrate had a depth of 5 cm, and the other half a depth of 15 cm (henceforth, the 5/15 cm treatment). We created the depth treatments by varying the depth below the substrate surface, in
Fig. 2. The spread of F. rubra and S. acre from their original 30.48 by 30.48 cm planting by the end of the study period (September 18, 2013). The cover change was compared between both species, those bars that share a letter are not significantly different.
186
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
Fig. 3. Total monthly cover F. rubra (F) + S. acre (S) during the 2012 (A) and 2013 (B) growing season. Monthly ratio of cover F. rubra (F)/S. acre (S) during the 2012 (C) and 2013 (D) growing season. For each month, the bars that share a letter are not significantly different.
was determined using photographs analyzed in ImageJ (Image Processing and Analysis in Java, http://rsbweb.nih.gov/ij/). Cover ranged from 0 to 1 and represented the proportion of the substrate surface covered by plants. Total cover for each replicate was obtained by summing the cover of both species within the replicate. Occasionally these two species overlapped within the planter box resulting in a total cover greater than 1.0. In order to determine the spread of F. rubra or S. acre outside their initial location, the point interception method (Floyd and Anderson, 1987) was used. A sampling frame was placed over the entire planter box. The number of times the aboveground biomass of S. acre or F. rubra touched the pin (diameter 3.7 cm) outside their initial quadrant was recorded. The pin interceptions were taken from 20 different locations in the planter box (5 per quadrant). 2.1. Statistical methods Relative Cover Change (RCC), total cover and ratio of cover were compared across the substrate depth treatments using 1-way ANOVAs with Tukey post hoc tests. All analyses used substrate depth treatment as the independent variable. The ratio in cover was determined by dividing the final cover of F. rubra by the final cover of S. acre. The data gathered from the point interception method (hits) was used to calculate the RCC of the spread of S. acre and F. rubra outside their original planting. There were no initial hits when calculating this migration, therefore, the formula used to calculate the RCC was modified to accommodate this: RCC = (ln(coverT2 ) − 0)/# of days). 3. Results 3.1. Cover During the 2012 growing season there was no significant difference in total cover between the 10 cm and 5/15 cm treatments. During August and September 2013 cover in the 10 cm treatment was significantly lower than the 5/15 cm treatment. The ratio of cover was only significantly lower in the 10 cm treatment compared to the 5/15 cm treatment for August and September 2012 and June and July 2013 (Fig. 2).
3.2. Spread outside of planted areas For the two treatments with an average substrate depth of 10 cm (the 10 cm and 5/15 cm treatments) the 10 cm treatment had significantly greater RCC, for both species, outside its original planting than the 5/15 cm treatment (Fig. 3). For both species, this spread was mainly caused by seedling dispersal.
4. Discussion and conclusions In this study, we wanted to determine if spatially variable substrate depths could allow more even growth and coverage between two species that were not expected to coexist in homogeneous conditions. Although we used only two species, these are representative of two broad classes of plants used on green roofs: succulents and grasses. The key comparisons in this study were between the 10 cm and 5/15 cm depth treatments, as both of these had the same mean depth, total weight, and volume of substrate. This allowed for an evaluation of the effects of substrate depth heterogeneity in isolation. The two other homogeneous substrate depth treatments (5 and 15 cm) are included for comparison, but do not have corresponding heterogeneous depth treatments. While we expected more even distributions of plant cover between the two species in the heterogeneous depth treatment compared to the uniform 10 cm depth treatment, this only occurred at the end of the first year of the study and the beginning of the second year of study. By the end of the second year, both heterogeneous and homogeneous treatments at 10 cm average depth had equivalent cover ratios, close to 1.5, compared to the 5 cm homogeneous treatment, which was strongly biased in favor of S. acre, while the 15 cm homogenous treatment was biased in favor of F. rubra. This supports many previous studies on extensive green roofs that indicate that Sedum tend to dominate in lower substrate depths, while greater depths are required for the persistence of grasses. While two growing seasons were insufficient for complete exclusion of either species in any of the treatments, previous studies show that species exclusion can happen within the first year (Rowe et al., 2012). A sevenyear study that included S. acre in mixtures with other succulents showed that it tended to drop out of mixtures in deeper substrate depths (Rowe et al., 2012), likely due to competition with larger
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
species. In another multi-year study, (Dunnett et al., 2008b) also found that S. acre had an advantage at lower depth (10 cm) in the first two years but declined in abundance after year two. A study over two decades in Germany showed that S. acre and a different species of Festuca (ovina) both persisted over the entire study on a roof with homogeneous 10 cm substrate depth (Köhler, 2006). In this study cover ratios were roughly equivalent in both 10 cm homogeneous and 5/15 cm heterogeneous treatments, and by the end of the second growing season both were biased in favor of the larger F. rubra. On the other hand, the 5/15 cm treatment showed significantly less spread of each species into the areas planted with the other species. This result is consistent with what we observed in the homogeneous 5 cm and 15 cm treatments: S. acre did poorly in the 15 cm homogeneous treatment, and had less ability to spread into 15 cm deep patches within the 5/15 cm heterogeneous treatment; F. rubra displayed the opposite pattern, a lower ability to spread into 5 cm sections in the 5/15 cm treatment. This study shows a strong effect of substrate depth, with dominance of F. rubra at mean substrate depths of 10 cm or greater, and dominance of S. acre at 5 cm. The decrease in the performance of F. rubra at 5 cm was most likely due to a lack of water, not competition, since S. acre growth was mainly observed around the edges of F. rubra. Given that F. rubra performed poorly at 5 cm depth, both when planted in the 5 cm homogeneous treatment and when spreading from 15 cm patches into 5 cm patches within the 5/15 cm treatment, results suggest that persistence of S. acre will be greater in the 5/15 cm treatment, as the 5 cm depth patches can act as refugia for the succulent species. There appeared to be a more even distribution of cover at the 5/15 cm substrate depth until August 2013. At this point both the 10 cm and 5/15 cm treatments had very similar ratios with almost 50/50 cover between the two species. The 2013 growing season had very mild weather, likely favoring the less stress-tolerant species F. rubra; it is possible that a harsher summer would have resulted in a more even distribution of the canopy cover in the 5/15 cm treatment, as was observed in the 2012 growing season. The effects of fluctuations in growing season conditions warrant further investigation in green roof plant species coexistence. Total average plant cover in 5/15 cm treatment was significantly greater than the 10 cm treatment for August and September 2013, although both treatments had the same amount of substrate. This indicates that a heterogeneous substrate depth could result in greater cover with the same amount of substrate as a homogeneous substrate depth, but this could also depend on the responses of the individual species to growing season conditions. The most common way to establish substrate depth heterogeneity on green roofs is to create areas with less depth (hollows), contrasting with areas where the substrate is mounded (Brenneisen, 2006; Köhler and Poll, 2010; Gedge et al., 2013). This creates heterogeneity not only in substrate depth, but possibly also in exposure, where mounds would be more prone to wind and desiccation, being above the main grade of the roof, and in hydrology, where water could drain off the mounded areas into the hollows. This is expected to contribute to greater diversity of invertebrates as well as plants (Brenneisen, 2006). In this study, we maintained a uniform substrate surface profile, so that we could explore the effects of substrate depth variability alone. In this study, concrete slabs were used to manipulate substrate depth. However, due to weight restrictions, this method is not feasible for many green roof systems. If desired, this kind of heterogeneity could easily be created in green roof installations by embedding lightweight objects (styrene foam blocks, wooden planks, pumice or other lightweight materials) under the substrate, creating variability in depth but
187
not surface topography. Alternate materials, including additives that enhance water retention (e.g. Farrell et al., 2013b), or different kinds of drainage layers (e.g. Savi et al., 2013) could also be used to create functional heterogeneity that might increase coexistence between plant species. While substrate depth heterogeneity had no detrimental effects on green roof vegetation structure in this study, other potential effects of such treatments should be explored including invertebrate habitat provisioning, carbon capture, and thermal performance. Finally, experiments need to be complemented with “field studies” of previously established green roofs that vary in the degree of habitat heterogeneity, plant species richness, and time since establishment, to determine which taxa are actually affected by these variables and whether heterogeneity can influence longterm coexistence.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.ecoleng.2014.03.023.
References Bates, A., Sadlera, J., Mackay, R., 2013. Vegetation development over four years on two green roofs in the UK. Urban Forest. Urban Green. 12, 98–108. Brenneisen, S., 2006. Space for urban wildlife: designing green roofs as habitats in Switzerland. Urban Habitats 4, 27–36. Chesson, P., 2000. Mechanisms of maintenance of species diversity. Annu. Rev. Ecol. Syst. 31, 343–366. Cook-Patton, S.C., Bauerle, T.L., 2012. Potential benefits of plant diversity on vegetated roofs: a literature review. J. Environ. Manage. 106, 85–92. Dunnett, N., Nagase, A., Booth, R., Grime, P., 2008a. Influence of vegetation composition on runoff in two simulated green roof experiments. Urban Ecosyst. 11 (4), 385–398. Dunnett, N., Nagase, A., Hallam, A., 2008b. The dynamics of planted and colonising species on a green roof over six growing seasons 2001–2006: influence of substrate depth. Urban Ecosyst. 11 (4), 373–384. Dunnett, N., Kingsbury, N., 2004. Planting Green Roofs and Living Walls, vol. 75. Timber Press, Portland, OR, pp. 104–114. Farrell, C., Szota, C., Williams, N.S., Arndt, S.K., 2013a. High water users can be drought tolerant: using physiological traits for green roof plant selection. Plant Soil 372, 177–193. Farrell, C., Ang, X.Q., Rayner, J.P., 2013b. Water-retention additives increase plant available water in green roof substrates. Ecol. Eng. 52, 112–118. Farrell, C., Mitchell, R.E., Szota, C., Rayner, J.P., Williams, N.S.G., 2012. Green roofs for hot and dry climates: interacting effects of plant water use, succulence and substrate. Ecol. Eng. 49, 270–276. Floyd, D., Anderson, J., 1987. A comparison of three methods for estimating plant cover. J. Ecol. 75 (1), 221–228. Gedge, D., Grant, G., Kadas, G., Dinham, C., 2013. Creating Green Roofs for Invertebrates A Best Practice Guide. Buglife – The Invertebrate Conservation Trust, Available at: http://www.buglife.org.uk/sites/default/files/Creating% 20Green%20Roofs%20for%20Invertebrates Best%20practice%20guidance 2.pdf (accessed 28.10.13). Gedge, D., Kadas, G., 2005. Green roofs and biodiversity. Biologist 52, 161–169. Getter, K.L., Rowe, D.B., Robertson, G.P., Cregg, B.M., Andresen, J.A., 2009. Carbon sequestration potential of extensive green roofs. Environ. Sci. Technol. 43 (19), 7564–7570. Kinzig, A.P., Pacala, S.W., Tilman, D., 2001. The Functional Consequences of Biodiversity. Princeton University Press, Princeton, pp. 365. Köhler, M., Poll, P.H., 2010. Long-term performance of selected old Berlin greenroofs in comparison to younger extensive greenroofs in Berlin. Ecol. Eng. 36 (5), 722–729. Köhler, M., 2006. Long-term vegetation research on two extensive green roofs in Berlin. Urban Habitats 4 (1), 3–26. Lundholm, J., MacIvor, J.S., MacDougall, Z., Ranalli, M., 2010. Plant Species and Functional Group Combinations Affect Green Roof Ecosystem Functions. Plos One. 5, 3. Lundholm, J.T., 2009. Plant species diversity and environmental heterogeneity: spatial scale and competing hypotheses. J. Veg. Sci. 20 (3), 377–391. MacIvor, S., Lundholm, J., 2011. Performance evaluation of native plants suited to extensive green roof conditions in a maritime climate. Ecol. Eng. 37 (3), 407– 417. Madre, F., Vergnes, A., Machon, N., Clergeau, P., 2013. A comparison of 3 types of green roof as habitats for arthropods. Ecol. Eng. 57 (4), 109–117.
188
A. Heim, J. Lundholm / Ecological Engineering 68 (2014) 184–188
Nagase, A., Dunnett, N., 2012. Amount of water runoff from different vegetation types on extensive green roofs: effects of plant species, diversity and plant structure. Landscape Urban Plan. 104 (3–4), 356–363. Nardini, A., Andri, S., Crasso, M., 2012. Influence of substrate depth and vegetation type on temperature and water runoff mitigation by extensive green roofs: shrubs versus herbaceous plants. Urban Ecosyst. 15 (3), 697–708. Ranalli, M., Lundholm, J., 2008. Biodiversity and ecosystem function in constructed ecosystems. Perspect. Agric. Vet. Sci. Nutr. Nat. Resour. 3 (034), 1–16.
Rowe, D.B., Getter, K.L., Durhman, A.K., 2012. Effect of green roof media depth on Crassulacean plant succession over seven years. Landscape Urban Plan. 104, 310–319. Savi, T., Andri, S., Nardini, A., 2013. Impact of different green roof layering on plant water status and drought survival. Ecol. Eng. 57, 188–196. Song, U., Kim, E., Bang, J.H., Son, D.J., Waldman, B., Lee, E.J., 2013. Wetlands are an effective green roof system. Build. Environ. 66, 141–147. Vivian-Smith, G., 1997. Microtopographic heterogeneity and floristic diversity in experimental wetland communities. J. Ecol. 85 (1), 71–82.