Contents lists available atScienceDirect
Ecological Engineering
journal homepage:www.elsevier.com/locate/ecoleng
Assessment of recycled or locally available materials as green roof substrates
Mert Eksi
a,⁎, Orhan Sevgi
b, Serdar Akburak
b, Hüseyin Yurtseven
c,
İlker Esin
daIstanbul University-Cerrahpasa, Faculty of Forestry, Landscape Architecture Department, 34473 Bahcekoy– Sariyer, Istanbul, Turkey bIstanbul University-Cerrahpasa, Faculty of Forestry, Department of Soil Science and Ecology, Istanbul, Turkey
cIstanbul University-Cerrahpasa, Faculty of Forestry, Department of Surveying and Cadastre, Istanbul, Turkey dIstanbul University-Cerrahpasa, Vocational School of Forestry Irrigation Program, Istanbul, Turkey
A R T I C L E I N F O Keywords: Green roof Substrates Recycled material Local material Plant selection A B S T R A C T
Importance of green roofs are growing due to their several benefits in urban areas. To provide a sustainable
green roof system and reduce the negative environmental effects of green roof construction, recycled or locally
available materials plays a crucial role. A study was conducted to evaluate the potential of four recycled
ma-terials (crushed concrete, crushed bricks, sawdust, and municipal waste compost) andfive locally available
materials in Istanbul (lava rock, pumice, zeolite, perlite and sheep manure) as green roof substrates over one year period. Twelve-substrate mixture were prepared at the site by mixing six inorganic mixtures (crushed concrete, crushed bricks, lava rock, pumice, zeolite, perlite) with two organic amendments (municipal waste compost and sawdust-sheep manure mixture) with a ratio of 8:2 by volume. Specific measurements such as plant
growth index, chlorophyllfluorescence, plant coverage ratio and survival of plant taxa were performed on five
native plant species, Allium schoenoprasum, Helichrysum italicum, Sedum lydium, Stachys thirkei and Thymus vul-garis. At the end of the 58 week study period, pumice and perlite-based substrates amended with municipal compost outperformed remaining substrate mixtures in terms of plant growth, plant stress, chemical and phy-sical properties. Moreover, performance of substrate mixtures consist of concrete, crushed bricks, lava rock, zeolite amended with sawdust and manure were adequate in some cases, which could be preferred as a green
roof substrate by admitting suggestions offered in the study.
1. Introduction
Green roofs are mainly implemented in urban areas due to their various benefits to mitigate the negative effects of urbanization such as cooling building envelope (Castleton et al., 2010;Fioretti et al., 2010; Butler and Orians, 2011;Schindler et al., 2019), reducing and delaying runoff (VanWoert et al., 2005;Berndtsson et al., 2006;Dunnett et al., 2008a; Liu et al., 2019), decreasing air pollution (Johnston and Newton, 2004; Currie and Bass, 2008), and carbon sequestration (Getter and Rowe, 2009a). Selection of growth substrates and vegeta-tion strongly influence the benefits provided by green roofs.
Green roof substrates influence water holding capacity, nutrient and water retention, drainage, lightness, albedo, stability and so on (Bates et al., 2015;Eksi and Rowe, 2016;Matlock and Rowe, 2016;Vannucchi et al., 2018). However, carbon footprint of materials in green roof substrates during production and transportation has to be considered along with green roof construction process. Recent studies demonstrate that recycled, locally available or locally sourced materials could lessen
the impact of green roof construction to the environment (Molineux et al., 2009;Mickovski et al., 2013;Rowe et al., 2016).
Plant selection is an essential part of the green roof design as well. Using native plant species in local or recycled substrates has the po-tential on increase the benefits of green roof systems and their con-tribution to urban ecology which will also increase the biodiversity and survival potential of plant species (Lundholm et al., 2010;Caneva et al., 2015). Plant survival and success mostly depend on suitable plant se-lection along with the proper substrate composition and adaptation to the environmental conditions.
Therefore, intensive amount of work has been done in around the world since early 2000's to adapt green roofs to local climate conditions (Moran et al., 2003;VanWoert et al., 2005;Durhman et al., 2006, 2007; Gibbs et al., 2006; Saiz et al., 2006; Getter and Rowe, 2008; Getter et al., 2009; Lundholm et al., 2010; MacIvor and Lundholm, 2011; MacIvor et al., 2011) by integrating recycled or locally available ma-terials (Molineux et al., 2009;Bates et al., 2015;Eksi and Rowe, 2016; Krawczyk et al., 2017; Chen et al., 2018) and native plant species
https://doi.org/10.1016/j.ecoleng.2020.105966
Received 27 January 2020; Received in revised form 23 June 2020; Accepted 30 June 2020
⁎Corresponding author.
E-mail addresses:merteksi@istanbul.edu.tr(M. Eksi),osevgi@istanbul.edu.tr(O. Sevgi),sakburak@istanbul.edu.tr(S. Akburak),
huseyiny@istanbul.edu.tr(H. Yurtseven),iesin@istanbul.edu.tr(İ. Esin).
Available online 15 July 2020
0925-8574/ © 2020 Elsevier B.V. All rights reserved.
(Williams et al., 2010;Nektarios et al., 2011;Papafotiou et al., 2013; Bretzel et al., 2016).
Substrate composition of green roofs based on target vegetation, green roof type, and other considerations (Ampim et al., 2010), which consist of two main components; organic and inorganic. Inorganic portion of the substrate constitutes the main component of green roof substrates, often consists of lightweight and stable aggregates and usually responsible for substrate stability, drainage and water retention. Organic portion, that constitutes the smaller part of the substrate mostly responsible for plant growth, biomass, nutrient reserve and in common practice not recommended higher than 20% by volume in the substrate (Rowe et al., 2006;Ondoño et al., 2015).
Roof gardens has been known and constructed in Turkey since 1980's. In the past decade, green roof industry has shown remarkable improvements in Turkey followed by numerous green roof applications on new commercial and residential buildings. Most of current roofs are extensive, usually covered with Sedum species and constructed by using some imported materials. Installation techniques and knowledge be-hind these projects mostly depend on some foreign guidelines. Nevertheless, several innovative green roofs were also constructed in recent years. Turkey has potential due to location, availability of local materials, existence of native plant species and skillful and well-edu-cated professionals. However, there are some deficiencies in terms of research and technical guidelines.
Green roof guidelines based on research performed in Northern countries, especially in Germany (FLL, 2008) are not always applicable in regions with hotter or drier climates, such as Mediterranean, espe-cially for shallow or unirrigated green roof systems.
Moreover, use of alternative materials such as recycled or locally available materials could lessen the impact of green roof construction on the environment by reducing manufacturing and transportation process and lower carbon footprint. Thus, there is a need for research to trial local species or substrate materials.
In this point of view, this study aims to evaluate several recycled or locally available materials along with five native herbaceous plant species to determine their potential for use in green roof substrates in Mediterranean region.
2. Materials and methods
The study was conducted at the Istanbul University-Cerrahpasa Green Roof Research Project (IUCGRP) site between June 2016 to July 2017. Study area is located at the Northern part of Istanbul in Bahçeköy – Sariyer Region, 41.10°N, 28.59°E, surrounded by open spaces in the north and south sides with a parking lot in the west.
Istanbul is located in a transitional climatic zone and exhibits sev-eral microclimates. Southern half of the city experiences warm Mediterranean climate (Köppen climate classification ‘Csa’ and ‘Csb’), where northern half experiences humid subtropical climate (Cfa) and oceanic climate (Cfb) with cooler temperatures in both winter and summer, higher humidity and more rains due to cooler Black Sea and northerly colder air masses and higher concentration of vegetation (Kottek et al., 2006;Ezber et al., 2007;Öztürk et al., 2017).
On an openfield, crates measuring 60 × 80 cm (inner dimensions 55.5 × 75 cm; 0,05 m3in volume) were placed on metal benches and each crate replicated a typical extensive green roof. Roofing layers were installed directly in the plastic crates and included moisture retention fleece (SSM45, Onduline Avrasya A.S., Istanbul), a plastic drainage mat (Maxidrain 25, NetYapi, Istanbul),filter sheet (SF Filter Sheet, Onduline Avrasya AS, Istanbul) and growing substrate. Experimental setup con-sist of 39 crates containing 12 substrate mixtures and a control treat-ment, each replicated 3 times. Crates were randomly distributed at the study area andfilled to 8 cm depth for each of the substrate blends. Each crate was set at 1% slope along with the benches below and a drainage hole was drilled from the lower side of the slope to allow excess water to drain.
Substrate mixtures were homogeneously hand mixed at the site with the addition of 80% of inorganic and 20% of organic materials by vo-lume. Locally available inorganic materials tested in the study were lava rock (tuff; 5–10 mm), zeolite (clinoptilolite; 1.6–3 mm), perlite (ex-panded; 3–5 mm) and pumice (3–8 mm). Recycled inorganic materials were crushed bricks (burnt clay bricks; 5–10 mm) and crushed concrete (10–20 mm) from demolition waste. Organic portion of the substrates composed of municipal waste compost as recycled material and sawdust -sheep manure mixture in 8:2 ratio by volume as locally available material (Fig. 1).
Table 1 Substrate physical and chemical properties at initiation and end of the study. Characteristic > Bulk Density (dry weight basis) Organic matter Nitrogen EC Maximum WHC pH Silt-clay content In fi ltration rate Unit > g/lt % mg/L g (KCl)/L % Vol pH Mass (%) mm/min Stage > Initial Final Initial Final Initial Final Initial Final Initial Final Initial Final Single Single Substrate mixtures Co +C 1314.5 1307.9 39.9 33.4 57.6 23.0 0.7 0.2 22.1 24.3 8.2 8.0 0.8 192.0 Aa Aa BCa BCa BCa Fb Ba Bb Ga Ha a b Co +S/M 1292.7 1295.5 41.1 30.3 28.2 14.9 0.4 0.2 18.1 26.3 7.9 7.9 0.1 192.0 Aa Aa BCa BCDa Fa Gb DEa Bb Gb GHa a a Control 812.9 840.1 9.1 11.0 49.0 45.3 0.2 0.1 43.5 42.7 7.6 7.8 4.3 12.0 Ca Ca Fb Ha CDEa Ca Ea EFb Fa Fa b a Lr +C 669.9 705.7 27.0 20.3 50.9 31.2 0.3 0.1 59.9 59.9 7.9 7.8 2.4 67.5 Da Da Da FGb BCDa CDEFb DEa Deb Ea Ea A A Lr +S/M 675.4 706.4 19.8 14.8 47.7 30.9 0.2 0.1 60.4 63.8 7.9 7.3 1.4 50.0 Da Da Ea GHa CDEa CDEFb DEa Fb Ea Ea a b Pe +C 182.6 156.2 35.3 39.6 38.3 37.5 2.0 0.2 242.7 271.0 7.8 7.6 19.8 30.0 Fa Ga Ca Aa Ea CDb Aa Ab Ba Ba a b Pe +S/M 145.4 147.6 27.2 28.2 26.9 28.4 0.6 0.1 358.9 369.2 8.0 7.2 6.3 28.3 Fa Ga Da CDEa Fa DEFa BCa Bb Aa Aa a b Pu +C 448.7 441.4 23.2 12.9 65.1 35.6 0.7 0.1 116.5 123.5 7.9 7.7 5.6 25.1 Ea Fa DEa Hb Ba CDEb Ba CDb Ca Ca a b Pu +S/M 416.3 437.3 21.0 11.4 43.7 29.2 0.3 0.1 125.5 106.3 7.7 7.2 3.0 33.3 Ea Fa Ea Hb CDEa DEFb DEa EFb Ca CDb a b Cb +C 1099.9 1100.9 27.5 27.3 52.8 34.3 0.4 0.1 36.3 29.3 8.0 7.8 2.0 54.4 Ba Ba Da DEa BCDa CDEb DEa DEb Fa Gb a b Cb +S/M 1134.3 1148.1 22.2 23.4 40.6 24.5 0.3 0.1 38.7 31.3 7.5 7.5 1.6 50.0 Ba Ba DEa EFa DEa EFb DEa EFb Fa Gb a a Ze +C 639.5 614.8 39.4 33.7 149.5 122.9 0.4 0.1 89.9 90.9 7.6 7.9 1.2 56.5 Da Ea BCa BCa Aa Aa CDa BCb Da Da b a Ze +S/M 639.5 648.5 47.5 34.6 133.1 71.8 0.2 0.1 94.7 94.7 7.5 7.7 1.2 77.5 Da DEa Aa ABb Aa Bb Ea DEFb Da Da b a FLL Guidelines ⁎ – ≤ 40 ≤ 80 ≤ 2.5 30 –65 6.0 –8.5 < 10% by Mass 60 –400 NCR-13 1998 1 Loss-on-ignition 2 Kjeldahl Method 3 Saturated Paste Method 4 ISO 11461:2001 5 NCR-13 1998 1 ASTM D6913/D6913M-17 6 Analysis performed by Istanbul University Faculty of Forestry Soil Ecology Laboratory, Istanbul, Turkey. WHC stands for Water Holding Capacity. Mean separation in rows and columns by least signi fi cant di ff erence (P = .05). Uppercase letters in colums denote comparsions between substrate treatments (n = 3) in same week. Lowercase letters in columns denote comparisons over time (initial and fi nal) within individual substrate blends (n = 3). Co: Concrete; Control: Zincolite; Lr: Lava rock; Pe: Perlite; Pu: Pumice; Cb: Crushed bricks; Ze: Zeolite; +C: Compost and S/M: Sawdust and manure blend. ⁎ Forschungsgesellschaft Landschaftsentiwicklung Landschaftsbau, 2008. FLL Guidelines are for single course extensive green roofs. 1 Brown, J. R. (1998). Recommended chemical soil test procedures for the North Central Region (No. 1001). Missouri Agricultural Experiment Station, U niversity of Missouri –Columbia. 2 Loss-on-ignition pp. 57 –58 from Recommended Chemical Soil Test Procedures for the North Central Region; J.R. Brown; North Central Regional Research Publication No. 221; Rev ised January 1998, pp. 57 –58. 3 Kjeldahl, J., Neue Methods zur Bestimmung des Sticksto ff s in Organischen Korpern, Z. Anal. Chem. 22:366 –382 (1883). 4 Saturated Paste Method, pp. 60 –61 from Recommended Chemical Soil Test Procedures for the North Central Region; J.R. Brown; North Central Regional Research Publication No. 221; Rev ised January 1998. 5 ISO 11461:2001 Standart: Soil quality — Determination of soil water content as a volume fraction using coring sleeves: Gravimetric method. 6 ASTM D6913/D6913M-17 Standard Test Methods For Particle-Size Distribution (Gradation) Of Soils Using Sieve Analysis.
Materials were obtained from several resources where sawdust was produced from scots pine logs in Istanbul University-Cerrahpasa Faculty of Forestry sawmill and municipal compost was produced from re-sidential waste and litter by ISTAC (Istanbul Environmental Management Industries, Istanbul Metropolitan Municipality). Recycled mineral materials including crushed bricks obtained from a brick com-pany in Turkey (Isiklar Construction Materials Inc., Istanbul) and cru-shed concrete was produced from demolition waste by ISTAC (Istanbul Environmental Management Industries, Istanbul Metropolitan Municipality). Locally available materials were obtained from several companies around Istanbul, including lava rock and pumice (Agac ve Peyzaj A.S., Istanbul Metropolitan Municipality), perlite and sheep manure (Yesil Vadi Nursery, Istanbul) and zeolite (Rota Mining Inc., Istanbul). The commercial substrate zincolite (Onduline Avrasya A.S., Istanbul; Zinco Gmbh, Germany) consisted of crushed brick and clay (40–45% by volume), pumice (40–45% by volume) and organic matter (10–15% by volume) which served as control (Fig. 1).
At the initiation and the end of the study samples were taken by using soil steel cores (100 cm3) from each plot and each substrate mixture was analyzed to determine particle size distribution, bulk density, maximum water-holding capacity (WHC), pH, soluble salts and nutrient content in laboratory setting (Istanbul University-Cerrahpasa Faculty of Forestry Soil and Ecology Lab).
Five herbaceous plant species tested in the study. These were Allium schoenoprasum, Helichrysum italicum, Sedum lydium, Thymus vulgaris and Saxifraga x arendsii. At the 17th week of the study (27 September 2016), Saxifraga x arendsii seedlings failed to survive and were replaced with Stachys thirkei seedlings collected from the open hills of Belgrade Forest with southern exposure. Stachys thirkei seedlings planted in a perlite-organic matter mixture untilfinal planting. Therefore, data obtained from Saxifraga x arendsii were ignored during study. Rest of the plant species were obtained from Nergis Peyzaj (Nergis Peyzaj Nursery, Yalova, Turkey) except Stachys thirkei. Allium schoenoprasum was in plastic pots (10,5 × 8 × 9,5 cm; upper diameter x lower diameter x height) and Helichrysum italicum was in smaller plastic pots with di-mensions of 7x6x9 cm. Sedum lydium and Thymus vulgaris were in seed trays (5x5x6 cm x 48 / perflat). Plant species were planted bare-rooted into the substrate mixtures on 1 June 2016, 7.5 cm from crate edges with three plants in a row and 10.0 cm apart. Each row was spaced 10 cm from another, resulting infive rows. Plants were randomly dis-tributed in each crate and replicated three times, resulting 15 plants per crate and 39 crates, which were randomly distributed at the site as well. Field measurements such as plant growth index (PGI), chlorophyll fluorescence (Fv/fm) and substrate volumetric moisture content (VMC) were collected initially at the time of planting (week 0, 1 June 2016), then once every three weeks for the duration of the study.
Substrate moisture content (VMC) were recorded at four points in each plot by inserting a Theta probe (ML2x; Delta-T Devices, Ltd., Cambridge, UK) with 6.0 cm rods into the substrate. The Theta probe instrument has a range of 0.0–1.0, with accuracy of ± 0.01 m3
.m−3. Plant growth index (PGI) was calculated for each plant by mea-suring plant height and width in two directions to form a growth index [(L × W × W)/3] (Monterusso et al., 2005;Rowe et al., 2006) at the time of planting. All plots were fertilized on the day of planting with controlled release fertilizer (Osmocote Exact, 15 N + 9P + 11K2O + 2MgO 5-month release, Everris International BV) at a rate of 6 g. per crate (11 g/m2) and watered tofield capacity. Further irrigation was performed to all plots during the following 15 days for plant establishment. After this point, no supplemental ir-rigation provided.
Raito of variablefluorescence to maximum fluorescence (Fv/fm) was measured during study to detect stress levels of plants. Chlorophyll fluorescence measurements are indirect measure of plant stress (Maxwell and Johnson, 2000) and normal Fv/fm values for healthy plants range from 0.700 to 0.800 and Fv/fm values below 0.600 are an indication of a plant stress (Ritchie, 2006).
Weather data was continuously recorded at the study site by an automated weather station (DeltaOhm HD2003 Three axis Ultrasonic Anemometer, Delta OHM S.r.L., Padova/Italy, measurement accuracy ± 1 °C) and precipitation measurements were collected using a rain gauge (DeltaOhm HD 2003 tipping bucket, measurement accu-racy ± 1%). Weather data recorded during study period were compared to climate norms of Istanbul between 1950 and 2015 were obtained from the Turkish State Meteorological Service National Weather Service (Turkish State Meteorological Service, 2016).
All data were checked for normality prior to analysis of variance by using the Kolmogorov-Smirnov test (Minitab, Inc., State College, PA). Significant differences among plots were analyzed by One-Way ANOVA tests using Fisher's LSD comparison (Little and Hills, 1978;Bousselot et al., 2010;Butler and Orians, 2011). Data transformation (the natural log transformation) was applied to nitrogen values inTable-1to sta-bilize the variance and normalize the data set (Underwood, 1997). Original means presented in the study. Data was analyzed using Minitab®16.2.2 (Minitab Inc., State College, PA) and Microsof Excel® 2016.
3. Results
3.1. Weather conditions
Mean ambient temperature of the study period was recorded as 15.03 °C and total precipitation during the growing season was 836.5 mm. Ambient air temperature and total precipitation were very close to TMS climate norms (Turkish State Meteorological Service, 2016). Except winter months experienced lower ambient temperatures. The warmest month of the study period was August 2016 (24.4 °C) and the warmest week of the growing season was recorded between 1 and 6 August 2016 (week 9) with an mean temperature of 26.1 °C. The longest dry period during the study lasted 20 days from 22 July to 12 August 2016 (Fig. 2).
3.2. Substrate physical and chemical properties
Particle size distribution of concrete, lava rock and brick based substrate blends were coarser than zeolite, perlite and pumice based substrates, which mostly consist offiner particles (Table 1andFig. 3). Commercial substrate could also interpreted as a coarser substrate to a lesser degree.
According to the obtained data (Table 1), percentage of organic matter content in zeolite, concrete and perlite based substrates was higher. On the contrary, organic matter content of commercial sub-strates was significantly lower. Pumice-based subsub-strates possessed greatest change in organic matter content over time with a rate of −45%.
There was a negative relationship between the WHC (water holding capacity) of the mixtures and porosity where WHC increased where porosity decreases. Porosity and infiltration rate were in positive rela-tion which caused leaching of N in coarser substrate mixtures.
At the initiation of the study, compost amendment strongly related to higher electrical conductivity (EC) due to its relatively higher salinity (2.69 g/L). By the end of the study, EC values of the substrate mixtures, an overall indicator of salinity, significantly decreased and nearly all reached the same values due to leaching out overtime. Perlite-compost (Pe-C) experienced the highest decrease in EC values with a rate of −90.6%. This was followed by pumice-compost (Pu-C) and concrete-compost (Co-C) with rates of −83.5% and − 79.6% respectively. Change in EC levels very limited in control treatments with a rate of −37.5%.
3.3. Substrate volumetric moisture content
Fig. 2. Weather conditions during the study period (1 June 2016–15 July 2017). (A) Monthly mean, maximum and minimum air temperature and climate norms, (B) monthly total precipitation, (C) weekly ambient mean air temperature and rainfall during the experiment. Local weather data obtained from the Istanbul
University-Cerrahpasa Green Roof Research Project (IUCGRP) site. Long-term climate data (1950–2015) derived from Turkish State Meteorological Service National Weather
measurements were always under the WHC (water holding capacity), but in positive relationship. VMC (volumetric moisture content) was generally lower in coarser substrates, crushed bricks, lava rock, and commercial substrate where greater percentage of the water drains immediately due to their porosity and infiltration rate. Exceptions were observed in concrete based substrates where differences were observed between field measurements and maximum water holding capacity (WHC) measurements in laboratory setting. This was due to its very coarse structure, which limits proper penetration of rods of the Theta
probe during field measurements. Therefore, field measurements sometimes taken from finer particles in the substrates which mostly tend to retain more water.
VMC of perlite was generally greater than other substrates, reached up to 0.4 m3 m-3 throughout the study period but moisture level in this substrate never reached the WHC. Moreover,fluctuation of the water content in the substrate was very quick especially after rainfall and during dry periods. Similarly, pumice had significant VMC with a peak level at 0.32 m3 m-3 during study. All treatments experienced a sharp Fig. 3. Particle size distribution of the substrate mixtures in comparison with FLL Guidelines and control substrate (control represents commercial substrate; grey field represents left and right curve limits for single layer substrates of FLL Guidelines).
decrease in VMC during dry periods, especially on week 11 (17 Aug 2016) due to evaporation and lack of supplementary irrigation. Exception was zeolite, which demonstrated more stable pattern in dry periods, ranged between 0.07 and 0.28 m3 m-3. Moreover, compost amendment had a positive impact on substrate volumetric moisture content (VMC) due to its higher water retention ability (Fig. 4.).
3.4. Plant growth index
Growth of A.schoenoprasum was greater in substrates, containing municipal waste compost (Fig. 5). In winter, growth of A.schoenoprasum plants were interrupted due to its growth pattern, where shoots dried-out and sprdried-out up in spring. Compost amended perlite, pumice and crushed concrete treatments positively influenced growth of Allium plants. Least growth were observed in zeolite and crushed brick based substrates.
Fig. 5. Mean absolute Plant Growth Index (PGI) over time offive plant species on substrate mixtures (x-axis represents weeks after initiation and y-axis represents PGI values).
Over the study period, PGI values of H.italicum plants were lower in treatments containing crushed bricks, zeolite and concrete. By week 36 (3 April 2017) with the arrival of spring differences on plant growth for H.italicum plants became evident, more specifically on compost amended substrates.
S.lydium steadily increased in size during study period regardless for substrate type and almost tripled their growth in all treatments. Decrease on plant growth index mostly depended on seasonal condi-tions. However, plants growing in perlite-compost (Pe-C) treatments were significantly larger those grown in remaining treatments.
Growth index of S.thirkei were distinctively greater in compost-amended substrates. After week 47, shoots of S.thirkei started to sprout up and differences among substrate mixtures became evident. Growth was relatively higher in pumice, perlite and concrete based substrates. Growth index of T.vulgaris plants were significantly larger in pumice-compost, zeolite-compost and control treatments than those grown in remaining substrates. Growth levels were lower in substrates amended with sawdust and manure mixture.
3.5. Plant survival
Among plant species, regardless of the substrate type, survival rate of A.schoenoprasum and S.lydium were 100%. Survival rate of S.thirkei ranged between 89 and 100% depending on substrate type. However, plant mortality was prominent on Helichrysum and Thymus plants grown in perlite and lava rock treatments between weeks 14 and 17 of the study (28 august 2016 and 20 September 2016). A dry period that lasted 18 days resulted with decrease on substrate volumetric content. This period was interrupted by a rainy day (20 September 2016) with a total precipitation of 4.1 mm and a sudden 6.9 °C decrease was re-corded on ambient air temperature. Daily temperaturefluctuation for lava rock and perlite had reached up to 39.1 °C and 37.3 °C respectively. According to obtained data, there is a positive relationship between substrate temperature fluctuation and plant mortality on Thymus and Helichrysum species (Fig. 6).
3.6. Chlorophyllfluorescence
The largest drop on mean Fv/fm values occurred at week 27 (9 December 2016) for all plant species and continued until week 36 (8 February 2016). This was due to sudden decrease on ambient air tem-perature, which was firstly measured below 0 °C. During this week,
precipitation amount was not very high, mostly recorded as snow, and night temperatures below 0 °C. Another significant change in Fv/fm levels ob-served between weeks 47 and 51 mostly due to seasonal conditions such as increase on air temperature, lack of precipitation and lower volumetric substrate moisture.
Throughout the study, Fv/fm values for Allium plants growing in commercial substrate (control) were statistically higher than remaining treatments and no indication of stress were observed. In contrast, de-spite its higher growth rate, Allium plants in perlite and pumice based substrates statistically tend to have lower Fv/fm values than remaining substrates.
Fv/fm values of H.italicum plants were above threshold of stress (0.600) throughout the study period. The exceptions were occurred during week 27 in pumice-compost (PueC) (0.536), zeolite-sawdust, manure (Ze-S/M) (0.575) along with lava rock-sawdust, and manure (Lr-S/M) (0.589) during week 51.
Significant decreases on Fv/fm values of S.lydium were detected in 3 points of the study period; after transplanting between weeks 3 and 6, sharp decrease on air temperature at week 27 and in summer between weeks 56 and 58. Fv/fm levels correspond to PGI values where plants growing in commercial substrate (control), lava rock-compost (Lr-C) and perlite-compost (Pe-C) had higher Fv/fm values.
Fv/fm values T.vulgaris in coarser substrates were lower as well as the growth levels. T.vulgaris plants in commercial substrate (control) demonstrated the most stable pattern in terms of Fv/fm despite their intermediate growth in this substrate(data not shown).
4. Discussion
This study investigated the potential for use of recycled or locally available materials as green roof substrates. Despite several drawbacks, there are promising aspects provided by these materials.
Water retention, drainage and substrate layers had the greatest negative environmental impact in green roof construction due to em-bodied energy during production and transportation process (Molineux et al., 2009; Bozorg Chenani et al., 2015). In that point, alternative substrates such as recycled or locally available materials could lessen the impact of green roof construction on the environment and bring economic benefits as they can be locally sourced (Molineux et al., 2009) and carbon footprints would be relatively lower because of their natural occurrence. However, implementing those materials to the green roof system may bring some challenges.
Green roof design can be influenced by many factors from building structure to wind direction. Nevertheless, it is possible to integrate all of the tested materials into green roof design with proper precautions under limited conditions. Therefore, recommendations for green roof substrates were expressed inTable 2. Thesefindings could be diversi-fied by mixing several materials together to improve shortcomings of those materials.
Among inorganic materials tested in the study, perlite and pumice-based substrates obtain favorable environment for plant growth. Addition of perlite had a significant effect on water holding capacity of substrates (Panayiotis et al., 2003;Getter and Rowe, 2006;Sutton et al., 2012) and porosity (Panayiotis et al., 2003), which resulted with least drought stress (Ntoulas et al., 2013). However, despite its many ad-vantages such as water retention ability and influence on plant growth, heat expanded perlite has an increased carbon footprint due to thermal treatment during processing.
In addition, wind load might have a negative effect on stability of perlite due to its lighter structure, disintegrate over time (Panayiotis et al., 2011) and higher thermal conductivity. As physiological per-formances of plants are highly dependent on temperature and extreme air and substrate temperatures (Kazemi and Mohorko, 2017), higher thermal conductivity of perlite and lava rock strongly influenced by ambient air temperature and negatively influenced plant growth.
On the contrary, pumice may be a better option due to its relatively higher bulk density, lower thermal conductivity and relatively lower environmental impact as its natural occurrence compared to perlite. Ntoulas et al. (2013)found that pumice participation was beneficial to turf grass drought tolerance in Mediterranean climate. Similarly, ac-cording to Bousselot et al. (2012), pumice has an influence on im-proving water holding capacity and promotes relatively high water absorption capacities (Wang et al., 2017) which corresponds tofindings of this study.
In terms of waste materials tested in the study, crushed concrete, promoted root aeration and drainage during the study period due to their coarser structure but failed to comply desired water retention and weight restrictions. Conversely, angular and tight structure of crushed bricks resulted with low water retention, heavy weight and lack of porosity, which limited root aeration. Another drawback come along with crushed bricks might be higher heavy metal concentration (Ye et al., 2013). Despite some disadvantages, both materials have a po-tential to reduce the environmental effects of the manufacturing and transportation process because of their reusability after completing their life span. Certainly, use of those materials would be limited due to their higher bulk density and weight restrictions, especially on existing buildings. Moreover prior to use in green roofs, these materials has to be grinded into smaller particles to comply with proper particle dis-tribution.
As infiltration rate were in positive relationship with particle size or pore radius (Hallett and Bengough, 2013), leaching of nitrogen in coarser substrates were higher which reached up to 60% in concrete
based substrates due to their very coarse structure which followed by pumice, crushed bricks and lava rock treatments. On the contrary, change in nitrogen levels were very limited in the substrates provided fine particles to the mixture, or lower infiltration rate, higher silt and clay content. Thus, change in nitrogen levels were limited in control (commercial substrate) and perlite treatments ranged between 2 and 5%. This was supported byCaravaca et al. (1999)who concluded that amounts of C and N that may be associated to clay and silt size reactions are affected by soil texture (Hassink, 1996). Zeolite-based substrates tend to retain more nitrogen compared to remaining substrate mixtures during study period. Nitrogen adsorption ability of zeolites were in-vestigated in several publications (Polat et al., 2005; Ippolito et al., 2011), which also reduces leaching of inorganic N (Witter and Lopez-Real, 1988;Ippolito et al., 2011) and increases AEC (Anion exchange capacity) (Ampim et al., 2010) due to its microstructure to absorb and retain NH4+and K+cations (Nektarios et al., 2011). This could explain the higher nitrogen retention of zeolite based substrates. However, in-fluence of nitrogen in zeolite-based substrates to the plant growth were limited. Possibly, this was due to itsfine but even and stiff structure, which limits the root aeration and plant development. However, zeolite have a potential to use in green roof substrates by using appropriate grain sizes. Moreover, as a locally available material, carbon footprint would be relatively lower.
Commercial substrate tested in the study mainly consist of a re-cycled material, crushed bricks, which is a better option rather than heat-expanded materials. This has been widely used around the world, especially in extensive roofs and performance of this substrate mostly met the expectations during the study.
Substrate volumetric moisture content (VMC) influenced by en-vironmental conditions (rainfall, air temperature, ET), substrate type and organic matter. Substrates containing greater amount offine par-ticles, increased the water storage capacity of the substrate (Thuring et al., 2010;Nektarios et al., 2011). In addition, compost amendment positively influenced water retention ability of the substrate regardless of the inorganic material.
Regarding organic matter content of the substrates, it is known that amount of organic matter incorporated with aggregate stability, in-filtration, soil aeration and soil strength in numerous studies (Whitbread, 1995;FAO, 2017), usually improves cation exchange ca-pacity (CEC) (Caravaca et al., 1999;Eksi et al., 2015) and helps to es-tablish vegetation on green roofs (Graceson et al., 2014). However, we were unable to determine the influence of organic matter throughout the study period. This may be explained byfindings of Nagase and Dunnett (2011) study, where influence of organic matter varies de-pending on wet or dry regimes, and increased organic matter do not promote increased growth in the dry regime.
Among organic amendments tested in the study, compost amend-ment more effective in increasing water retention (Paradelo et al., 2019) and positively influenced plant growth (Kotsiris et al., 2013) due to higher nitrogen content compared to sawdust and manure mixture. Table 2
Classification of substrate materials in terms of application needs.
Conditions Substrate materials
Concrete Lava rock Perlite Pumice Bricks Zeolite Control
Windy areas + + + + + +
Areas with weight restrictions + +
Dry areas / arid climates + + +
Insufficient drainage + + +
Insufficient irrigation + + +
Aesthetical concerns + + + +
Budget limitation + + +
Plants– need root aeration + + +
Plants– sensitive to substrate temperature + + +
On the other hand, as previously reported in some studies (Chelinho et al., 2019) high salinity of municipal waste compost may cause in-appropriate environment to the plant species. However, leaching over time resulted with lower EC among substrate mixtures.
Moreover, nitrogen content of the sawdust tested in the study was lower due to its origin (softwood) which immobilize much less nitrogen than hardwoods (Blunt, 1988). Nevertheless, municipal waste compost have potential to reduce the waste amount transferred to the landfills, addition of sawdust could be regarded as recycling or upcycling process of waste material, and manure could be evaluated as low impact natural amendment.
Regarding the substrate depth, 8 cm substrate depth was adequate for plant species tested in the study. Certainly, depending on weight restrictions deeper substrates would be a better choice in Mediterranean climate to increase the water retention abilities in dry periods, reduce the temperaturefluctuations and promote plant growth (Thuring et al., 2010;Ekşi and Rowe, 2019) by using proper substrate materials.
Among plant species tested in the study, survival rate of Allium plants were 100% despite various growing rates in substrate blends. It is already known that Allium species are well-adapted to extensive green roofs and several studies investigated their salt tolerance (Whittinghill and Rowe, 2011), coverage rate (Köhler, 2006) and survival rate on unirrigated treatments (Whittinghill et al., 2013;Van Mechelen et al., 2014;Gabrych et al., 2016). As reported byVandegrift et al. (2019)this adaptation ability is due to their desiccation prevention mechanisms, water storage in bulbs, minimal leaf area and self-seeding properties.
Survival of aromatic xerophytes such as Helichrysum species on ex-tensive Mediterranean green roofs under limited irrigation and sub-strate depth in Mediterranean climate have been investigated by a number of researchers (Papafotiou et al., 2013;Caneva et al., 2015; Vestrella et al., 2015). In similar climate conditions and substrate depth, similar results were obtained inPapafotiou et al. (2013)study where Helichrysum species (H.italicum and H.orientale) found suitable for use in extensive or semi-intensive green roofs in Mediterranean climate under limited irrigation and substrate depth (7.5 cm). In our study, growth rate of H.italicum plants were limited in substrate mix-tures with lower nitrogen and volumetric moisture content. Moreover, despite their higher growth rate in lava rock-compost (LreC), survival rate of H.italicum plants were 11% and the mean survival rate was 64.1%. Conversely, survival rates were 100% in zeolite-based substrates despite its lower growth rate.
Sedum species are popular in green roof systems due to their strong adaptation abilities. Several studies investigated the adaptation abilities of Sedum species on green roofs (Kirschstein, 1997; Lassalle, 1998; Rowe and Getter, 2008;Getter and Rowe, 2009b) and popular species such as S.album, S.acre, S.reflexum ve S.spirium. Those were also tested in previous studies in Istanbul climate (Eksi, 2012; Ekşi and Uzun, 2016). Even though it is a native plant in Mediterranean area, there is no sufficient information related to Sedum lydium regarding its perfor-mance on green roofs. Survival rate of S.lydium was 100% but organic amendments in the substrate mixtures clearly influenced plant growth, which was greater in substrates containing municipal waste compost and lower in substrates containing sawdust and sheep manure mixture. It was concluded that S.lydium could easily be adapted to the green roofs in Mediterranean region regardless of the substrate type due to its rapid coverage ability and resistance.
Performance of Thymus vulgaris during study period was adequate. Distinctive properties of this plant was attractiveness for bees and other pollinators, which can be evaluated as a pollinator. Drought tolerance was sufficient. It should be planted in substrates with finer particles because of its limited growth in coarser substrates.
Stachys thirkei have not been previously reported in green roof studies. However, some studies investigated another Stachys species (S.byzantina) for its cooling potential (Blanusa et al., 2013; Vaz Monteiro et al., 2017), colonization and survival abilities (Dunnett
et al., 2008b). Thus, we assume that S.thirkei would be able to provide benefits of S.byzantina due to their genus identity. Adaptation and survival rate of S.thirkei was very well, which also have a potential to contribute to urban biodiversity due to its native character in Medi-terranean region. Findings of the studyfindings are applicable on green roofs in Mediterranean or similar climate conditions in the USA, Aus-tralia or Southern Europe where plants experience frequent drought and heat stress on unirrigated roofs. Native plant selection will also increase the biodiversity and survival potential of plant species in dry periods compared to monocultures, which will also contribute to cli-mate change mitigation and adaptation strategies according to Lundholm et al. (2010). AsButler et al. (2012) pointed out, native plants are better adapted, provide greater environmental benefit, but 79% of the species found in open habitats are currently not used on green roofs (Van Mechelen et al., 2014), which also applies to green roofs in Istanbul.
5. Conclusions
This study has demonstrated the influence of recycled or locally available materials on plant growth over a period over a thirteen-month period. Among substrates tested in the study, significant decreases ob-served in the C and N contents and EC levels relative to the initial va-lues. Porosity and particle size distribution of the substrate mixtures had a strong influence on water retention, infiltration and plant survival in dry periods.
Pumice and municipal waste compost mixture showed acceptable physical and chemical properties and positively influenced plant growth, similar to the commercial substrate and somewhat better in some aspects such as locality and reduced carbon emissions. Furthermore, perlite-based substrates demonstrated similar results to pumice, but some drawbacks highlighted in the study may limit its usage. Thermal conductivity of lava rock provided unfavorable en-vironment for plant species dependent on substrate temperatures along with perlite, which is not suitable for use in terrestrial climate condi-tions. Crushed bricks was the least suitable substrate from every aspect. Zeolite performed well with regard to nutrient and water retention, but failed to support plant growth due to particle distribution. Finally, as heaviest and coarsest material in the study, concrete, proved to be a good substrate for the growth of Allium species.
The plant species tested in the study are good candidates for green roofs in Mediterranean region. However, substrate depth was not suf-ficient for Helichrysum species. Thus, deeper substrates would allow more options for plant selection, moisture retention and plant survival during drought periods. Moreover,floristic diversity has to be taken under consideration during green roof design. Therefore, further re-search still needed to determine the long-term effects of the substrate materials and native plants on green roof systems in Istanbul and si-milar climatic regions and to develop local green roof guidelines in the future.
Declaration of Competing Interest
The authors declare that they have no known competingfinancial interests or personal relationships that could have appeared to influ-ence the work reported in this paper.
Acknowledgements
Funding for this study provided by TUBITAK (The Scientific and Technological Research Council of Turkey) with grant no: 215O854 and Scientific Research Project Coordination Unit of Istanbul University (Cerrahpasa), grant no: FYL-2017-23711. We express our appreciation to Yesil Vadi Nursery (Tarabya, Istanbul) for providing some of the plants and some of the substrate materials for the study. We would like
to thank to Atila GURSES and Onduline Avrasya A.S. (Istanbul) along with Netyapi Ltd. Sti. (Istanbul) for providing green roof layers and materials. The authors acknowledge the support of Rota Madencilik for providing zeolite, ISTAC A.S.(Istanbul Environmental Management Industries, Istanbul) for providing municipal waste compost, crushed concrete and lava rock, Isiklar Building Products (Istanbul) for pro-viding crushed bricks and Nergis Peyzaj Ltd. (Yalova) for propro-viding plant material. The authors acknowledge assistance from students in Istanbul University-Cerrahpasa Landscape Architecture Department; Eslem Sura Cagaoluglu, Yasemin Irem Candan, Pavel Iokhim, Halmirza Tuna, Sumeyya Akdag, Hatice Ezgi Sakizci, Betul Sisdag, Arzu Ozdemir, Zeliha Taslicay, Emine Aydin, Busenaz Bingul, Zuleyha Ece Yildiz and Nigar Kucuk. Authors would like to thank Prof.Dr. Bradley Rowe from Michigan State University for his invaluable support.
References
Ampim, P.A.Y., Sloan, J.J., Cabrera, R.I., Harp, D.A., Jaber, F.H., 2010. Green Roof growing Substrates: Types, Ingredients, Composition and Properties. J. Environ. Hortic. 28, 244–252.
Bates, A.J., Sadler, J.P., Greswell, R.B., Mackay, R., 2015. Effects of recycled aggregate growth substrate on green roof vegetation development: a six year experiment. Landsc. Urban Plan. 135, 22–31.https://doi.org/10.1016/j.landurbplan.2014.11. 010.
Berndtsson, J.C., Emilsson, T., Bengtsson, L., 2006. The influence of extensive vegetated roofs on runoff water quality. Sci. Total Environ. 355, 48–63.https://doi.org/10. 1016/j.scitotenv.2005.02.035.
Blanusa, T., Vaz Monteiro, M.M., Fantozzi, F., Vysini, E., Li, Y., Cameron, R.W.F., 2013. Alternatives to sedum on green roofs: can broad leaf perennial plants offer better “cooling service”? Build. Environ. 59, 99–106.https://doi.org/10.1016/j.buildenv. 2012.08.011.
Blunt, B.R., 1988. Media and Mixes for Container-Grown Plants. Springer Netherlandshttps://doi.org/10.1007/978-94-011-7904-1.
Bousselot, J.M., Klett, J.E., Koski, R., 2010. Moisture content of extensive green roof growing media and growth response of 15 plant species during dry down. Hortscience 45S, S93.
Bousselot, J.M., Klett, J.E., Kosk, R.D., 2012. Evaluating a Natural Zeolite as an Amendment for Extensive Green Roof Substrate. Environ. Hortic. 30, 201–206. Bozorg Chenani, S., Lehvävirta, S., Häkkinen, T., 2015. Life cycle assessment of layers of
green roofs. J. Clean. Prod. 90, 153–162.https://doi.org/10.1016/j.jclepro.2014.11. 070.
Bretzel, F., Vannucchi, F., Romano, D., Malorgio, F., Benvenuti, S., Pezzarossa, B., 2016. Wildflowers: from conserving biodiversity to urban greening—a review. Urban For. Urban Green. 20, 428–436.https://doi.org/10.1016/j.ufug.2016.10.008. Butler, C., Orians, C.M., 2011. Sedum cools soil and can improve neighboring plant
performance during water deficit on a green roof. Ecol. Eng. 37, 1796–1803.https:// doi.org/10.1016/j.ecoleng.2011.06.025.
Butler, C., Butler, E., Orians, C.M., 2012. Native plant enthusiasm reaches new heights: Perceptions, evidence, and the future of green roofs. Urban For. Urban Green. 11, 1–10.https://doi.org/10.1016/j.ufug.2011.11.002.
Caneva, G., Kumbaric, A., Savo, V., Casalini, R., 2015. Ecological approach in selecting extensive green roof plants: a data-set of Mediterranean plants. Plant Biosyst. 149, 374–383.https://doi.org/10.1080/11263504.2013.819819.
Caravaca, F., Lax, A., Albaladejo, J., 1999. Organic matter, nutrient contents and cation exchange capacity infine fractions from semiarid calcareous soils. Geoderma 93, 161–176.https://doi.org/10.1016/S0016-7061(99)00045-2.
Castleton, H.F., Stovin, V., Beck, S.B.M., Davison, J.B., 2010. Green roofs; building energy savings and the potential for retrofit. Energy Build. 42, 1582–1591.https://doi.org/ 10.1016/j.enbuild.2010.05.004.
Chelinho, S., Pereira, C., Breitenbach, P., Baretta, D., Sousa, J.P., 2019. Quality standards for urban waste composts: the need for biological effect data. Sci. Total Environ. 694, 133602.https://doi.org/10.1016/j.scitotenv.2019.133602.
Chen, C.F., Kang, S.F., Lin, J.H., 2018. Effects of recycled glass and different substrate materials on the leachate quality and plant growth of green roofs. Ecol. Eng. 112, 10–20.https://doi.org/10.1016/j.ecoleng.2017.12.013.
Currie, B.A., Bass, B., 2008. Estimates of air pollution mitigation with green plants and green roofs using the UFORE model. Urban Ecosyst. 11, 409–422.https://doi.org/10. 1007/s11252-008-0054-y.
Dunnett, N., Nagase, A., Booth, R., Grime, P., 2008a. Influence of vegetation composition on runoff in two simulated green roof experiments. Urban Ecosyst. 11, 385–398. https://doi.org/10.1007/s11252-008-0064-9.
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, 373–384.https://doi.org/10.1007/s11252-007-0042-7. Durhman, A.K., Bradley Rowe, D., Rugh, C.L., 2006. Effect of watering regimen on
chlorophyllfluorescence and growth of selected green roof plant taxa. HortScience
41, 1623–1628.
Durhman, A.K., Rowe, D.B., Rugh, C.L., 2007. Effect of substrate depth on initial growth, coverage, and survival of 25 succulent green roof plant taxa. HortScience 42, 588–595.
Eksi, M., 2012.İstanbul ’ daki Başlıca Çatı Bahçelerinin Yapım Esasları Açısından
Değerlendirilmesi Assessment of Major Roof Gardens in Istanbul in terms of Construction Principles. 62. pp. 149–157.
Eksi, M., Rowe, D.B., 2016. Green roof substrates: effect of recycled crushed porcelain and foamed glass on plant growth and water retention. Urban For. Urban Green. 20, 81–88.https://doi.org/10.1016/j.ufug.2016.08.008.
Ekşi, M., Rowe, D.B., 2019. Effect of substrate depth and type on plant growth for ex-tensive green roofs in a mediterranean climate. J. Green Build. 14, 29–44.https:// doi.org/10.3992/1943-4618.14.2.29.
Ekşi, M., Uzun, A., 2016. Assessment of Green Roof Systems in Terms of Water and Energy Balance. Istanbul Univ, Fac For.
Eksi, M., Rowe, D.B., Fernández-ca, R., Cregg, B.M., 2015. Urban Forestry & Urban Greening Effect of substrate compost percentage on green roof vegetable production. Urban For. Urban Green. 14, 315–322.https://doi.org/10.1016/j.ufug.2015.03.006. Ezber, Y., Lutfi Sen, O., Kindap, T., Karaca, M., 2007. Climatic effects of urbanization in Istanbul: a statistical and modeling analysis. Int. J. Climatol. A J. R. Meteorol. Soc. 27, 667–679.https://doi.org/10.1002/joc.
FAO, 2017. Soil Organic Carbon: The Hidden Potential. Food and Agriculture Organization of the United Nations, Rome, Italy.https://doi.org/10.1038/nrg2350. Fioretti, R., Palla, A., Lanza, L.G., Principi, P., 2010. Green roof energy and water related performance in the Mediterranean climate. Build. Environ. 45, 1890–1904.https:// doi.org/10.1016/j.buildenv.2010.03.001.
FLL (Forschungsgesellschaft Landschaftsentwicklung Landschaftsbau) (Society of Landscape Development and Landscape Design), 2008. Guidelines for the Planning, Construction, and Maintenance of Green-roof Sites. FLL, Bonn, Germany. Gabrych, M., Kotze, D.J., Lehvävirta, S., 2016. Substrate depth and roof age strongly
affect plant abundances on sedum-moss and meadow green roofs in Helsinki, Finland. Ecol. Eng. 86, 95–104.https://doi.org/10.1016/j.ecoleng.2015.10.022.
Getter, K.L., Rowe, D.B., 2006. The role of extensive green roofs in sustainable devel-opment. HortScience 41, 1276–1285.https://doi.org/10.17776/csj.30292. Getter, K.L., Rowe, D.B., 2008. Media depth influences Sedum green roof establishment.
Urban Ecosyst. 11, 361–372.https://doi.org/10.1007/s11252-008-0052-0. Getter, K.L., Rowe, D.B., 2009a. Carbon sequestration potential of extensive green roofs.
Sci. Technol. 1–13.
Getter, K.L., Rowe, D.B., 2009b. Substrate depth influences sedum plant community on a green roof. HortScience 44, 401–407.
Getter, K.L., Rowe, D.B., Cregg, B.M., 2009. Solar radiation intensity influences extensive green roof plant communities. Urban For. Urban Green. 8, 269–281.https://doi.org/ 10.1016/j.ufug.2009.06.005.
Gibbs, J., Luckett, K., Jost, V., Morgan, S., 2006. Evaluating performance of a green roof system with different growing mediums, Sedum species and fertilizer treatments. In: Reg. Green Roof.
Graceson, A., Hare, M., Monaghan, J., Hall, N., 2014. Use of inorganic substrates and composted green waste in growing media for green roofs. Biosyst. Eng. 124, 1–7. https://doi.org/10.1016/j.biosystemseng.2014.05.007.
Hallett, P.D., Bengough, A.G., 2013. Managing the soil physical environment for plants. In: Soil Cond. Plant Growth. Oxford Blackwell Publ, pp. 238–268.
Hassink, J., 1996. Preservation of Plant Residues in Soils Differing in Unsaturated pro-tective Capacity. Soil Sci. Soc. Am. J. 60, 487–491.https://doi.org/10.2136/ sssaj1996.03615995006000020021x.
Ippolito, J.A., Tarkalson, D.D., Lehrsch, G.A., 2011. Zeolite soil application method affects inorganic nitrogen, moisture, and corn growth. Soil Sci. 176, 136–142.https://doi. org/10.1097/SS.0b013e31820e4063.
Johnston, J., Newton, J., 2004. Building Green a Guide to Using Plants on Roofs, Walls and Pavements 121.
Kazemi, F., Mohorko, R., 2017. Review on the roles and effects of growing media on plant performance in green roofs in world climates. Urban For. Urban Green. 23, 13–26. https://doi.org/10.1016/j.ufug.2017.02.006.
Kirschstein, C., 1997. Die du Sedum-arten, abgeleitet aus der bedeutung der wurzel-saugspannung-teil 1. Stadt und Grun 46, 252–256.
Köhler, M., 2006. Long-term vegetation research on two extensive green roofs in Berlin. Urban Habitats 4, 3–26.
Kotsiris, G., Nektarios, P.A., Ntoulas, N., Kargas, G., 2013. An adaptive approach to in-tensive green roofs in the Mediterranean climatic region. Urban For. Urban Green. 12, 380–392.https://doi.org/10.1016/j.ufug.2013.05.001.
Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World map of the Köppen-Geiger climate classification updated. Meteorol. Z. 15, 259–263.https://doi.org/10. 1127/0941-2948/2006/0130.
Krawczyk, A., Domagała-Świątkiewicz, I., Lis-Krzyścin, A., Daraż, M., 2017. Waste silica as a valuable component of extensive green-roof substrates. Pol. J. Environ. Stud. 26, 643–653.https://doi.org/10.15244/pjoes/64791.
Lassalle, F., 1998. Wirkung von trockenstreßauf xerophile pflanzen. Stadt und Grün 47, 437–443.
Little, T.M., Hills, F.J., 1978. Agricultural Experimentation: Design and Analysis. Wiley. Liu, W., Feng, Q., Chen, W., Wei, W., Deo, R.C., 2019. The influence of structural factors on stormwater runoff retention of extensive green roofs: new evidence from scale-based models and real experiments. J. Hydrol. 569, 230–238.https://doi.org/10. 1016/j.jhydrol.2018.11.066.
Lundholm, J., MacIvor, J.S., MacDougall, Z., Ranalli, M., 2010. Plant species and func-tional group combinations affect green roof ecosystem functions. PLoS One 5. https://doi.org/10.1371/journal.pone.0009677.
MacIvor, J.S., Lundholm, J., 2011. Performance evaluation of native plants suited to extensive green roof conditions in a maritime climate. Ecol. Eng. 37, 407–417. https://doi.org/10.1016/j.ecoleng.2010.10.004.
MacIvor, J.S., Ranalli, M.A., Lundholm, J.T., 2011. Performance of dryland and wetland plant species on extensive green roofs. Ann. Bot. 107, 671–679.https://doi.org/10.
1093/aob/mcr007.
Matlock, J.M., Rowe, D.B., 2016. The suitability of crushed porcelain and foamed glass as alternatives to heat-expanded shale in green roof substrates: an assessment of plant growth, substrate moisture, and thermal regulation. Ecol. Eng. 94, 244–254.https:// doi.org/10.1016/j.ecoleng.2016.05.044.
Maxwell, K., Johnson, G.N., 2000. Chlorophyllfluorescence - a practical guide. J. Exp. Bot. 51, 659–668.https://doi.org/10.1093/jxb/51.345.659.
Mickovski, S.B., Buss, K., McKenzie, B.M., Sökmener, B., 2013. Laboratory study on the potential use of recycled inert construction waste material in the substrate mix for extensive green roofs. Ecol. Eng. 61, 706–714.https://doi.org/10.1016/j.ecoleng. 2013.02.015.
Molineux, C.J., Fentiman, C.H., Gange, A.C., 2009. Characterising alternative recycled waste materials for use as green roof growing media in the U.K. Ecol. Eng. 35, 1507–1513.https://doi.org/10.1016/j.ecoleng.2009.06.010.
Monterusso, M.A., Bradley Rowe, D., Rugh, C.L., 2005. Establishment and persistence of Sedum spp. and native taxa for green roof applications. HortScience 40, 391–396. Moran, A., Hunt, B., Jennings, G., 2003. A North Carolinafield study to evaluate
greenroof runoff quantity, runoff quality, and plant growth. World Water Environ. Resour. Congr. (2003), 1–10.https://doi.org/10.1061/40685(2003)335. Nagase, A., Dunnett, N., 2011. The relationship between percentage of organic matter in
substrate and plant growth in extensive green roofs. Landsc. Urban Plan. 103, 230–236.https://doi.org/10.1016/j.landurbplan.2011.07.012.
Nektarios, P.A., Amountzias, I., Kokkinou, I., Ntoulas, N., 2011. Green roof substrate type and depth affect the growth of the native species Dianthus fruticosus under reduced irrigation regimens. HortScience 46, 1208–1216.
Ntoulas, N., Nektarios, P.A., Charalambous, E., Psaroulis, A., 2013. Zoysia matrella cover rate and drought tolerance in adaptive extensive green roof systems. Urban For. Urban Green. 12, 522–531.https://doi.org/10.1016/j.ufug.2013.07.006. Ondoño, S., Martínez-Sánchez, J.J., Moreno, J.L., 2015. The inorganic component of
green roof substrates impacts the growth of Mediterranean plant species as well as the C and N sequestration potential. Ecol. Indic. 61, 739–752.https://doi.org/10.1016/j. ecolind.2015.10.025.
Öztürk, M.Z., Çetinkaya, G., Aydın, S., 2017. Köppen-Geiger İklim Sınıflandırmasına Göre Türkiye’nin İklim Tipleri. J. Geogr. 35, 17–27.https://doi.org/10.26650/ JGEOG295515.
Panayiotis, N., Panayiota, T., Ioannis, C., 2003. Soil Amendments Reduce Roof Garden Weight and Influence the Growth Rate of Lantana. HortScience 38, 618–622. Panayiotis, P.A., Kastritsis, S., Ntoulas, N., Tsiotsiopoulou, P., 2011. Substrate
amend-ment effects on potted plant production and dry weight partition of Lantana camara. HortScience 46, 864–869.
Papafotiou, M., Pergialioti, N., Tassoula, L., Massas, I., Kargas, G., 2013. Growth of native aromatic xerophytes in an extensive Mediterranean green roof as affected by sub-strate type and depth and irrigation frequency. HortScience 48, 1327–1333. Paradelo, R., Basanta, R., Barral, M.T., 2019. Water-holding capacity and plant growth in
compost-based substrates modified with polyacrylamide, guar gum or bentonite. Sci. Hortic. (Amsterdam). 243, 344–349.https://doi.org/10.1016/j.scienta.2018.08.046. Polat, E., Demir, H., Onus, A.N., 2005. The Effects on Yield and Quality of Different Level of Zeolite in Lettuce (Lactuca satiıva var. longifolia) growing. Akdeniz Üniversitesi Ziraat Fakültesi Derg. 18, 95–99.
Ritchie, G. a, 2006. Chlorophyll Fluorescence : What Is it and What Do the Numbers Mean ? The Physiological Basis of Chlorophyll Fluorescence. Consultant. pp. 34–43. Rowe, D.B., Getter, K.L., 2008. Media depth influences Sedum green roof establishment.
Urban Ecosyst. 11, 361–372.https://doi.org/10.1007/s11252-008-0052-0. Rowe, D.B., Monterusso, M.A., Rugh, C.L., 2006. Assessment of heat-expanded slate and
fertility requirements in green roof substrates. Horttechnology 16, 471–477. Rowe, D.B., Matlock, J., Eksi, M., 2016. Use of recycled crushed porcelain for green roof
substrates. In: CitiesAlive: 14th Annual Green Roof and Wall Conference. Green Roofs
for Healthy Cities, Washington, DC, pp. 1–16.
Saiz, S., Kennedy, C., Bass, B., Pressnail, K., 2006. Comparative life cycle assessment of standard and green roofs. Environ. Sci. Technol. 40, 4312–4316.https://doi.org/10. 1021/Es0517522.
Schindler, B.Y., Blaustein, L., Vasl, A., Kadas, G.J., Seifan, M., 2019. Cooling effect of Sedum sediforme and annual plants on green roofs in a Mediterranean climate. Urban For. Urban Green. 38, 392–396.https://doi.org/10.1016/j.ufug.2019.01.020. Sutton, R., Rowe, D.B., Macdonagh, P., Acomb, G., Hawke, R., 2012. 1R : Plant
Performance Research 1 PLANT PERFORMANCE FOR 21ST CENTURY GREEN ROOF ECOSYSTEMS. CitiesAlive! 10th Annu. Green Roof Wall Conf. 2012 Conf. Proc. 1–15. Thuring, C.E., Berghage, R.D., Beattie, D.J., 2010. Green roof plant responses to different substrate types and depths under various drought conditions. Horttechnology 20, 395–401.
Turkish State Meteorological Service, 2016. TMS, 2016 [WWW Document]. Clim. Norm. Istanbul, Sariyer. URL https://www.mgm.gov.tr/veridegerlendirme/il-ve-ilceler-istatistik.aspx?k=A&m=ISTANBUL (accessed 11.10.18).
Underwood, A.J., 1997. Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variance. Univ. Cambridge.
Van Mechelen, C., Dutoit, T., Hermy, M., 2014. Mediterranean open habitat vegetation offers great potential for extensive green roof design. Landsc. Urban Plan. 121, 81–91.https://doi.org/10.1016/j.landurbplan.2013.09.010.
Vandegrift, D.A., Rowe, D.B., Cregg, B.M., Liang, D., 2019. Effect of substrate depth on plant community development on a Michigan green roof. Ecol. Eng. 138, 264–273. https://doi.org/10.1016/j.ecoleng.2019.07.032.
Vannucchi, F., Pini, R., Scatena, M., Benelli, G., Canale, A., Bretzel, F., 2018. Deinking sludge in the substrate reduces the fertility and enhances the plant species richness of extensive green roofs. Ecol. Eng. 116, 87–96.https://doi.org/10.1016/j.ecoleng. 2018.02.027.
VanWoert, N.D., Rowe, D.B., Andresen, J.A., Rugh, C.L., Xiao, L., 2005. Watering regime and green roof substrate design affect Sedum plant growth. HortScience 40, 659–664. https://doi.org/10.2134/jeq2004.0364.
Vaz Monteiro, M., Blanuša, T., Verhoef, A., Richardson, M., Hadley, P., Cameron, R.W.F., 2017. Functional green roofs: Importance of plant choice in maximising summertime environmental cooling and substrate insulation potential. Energy Build. 141, 56–68. https://doi.org/10.1016/j.enbuild.2017.02.011.
Vestrella, A., Savé, R., Biel, C., 2015. Performance of two simulated green roofs in the mediterranean area. J. Agric. Sci. 7, 59–75.https://doi.org/10.5539/jas.v7n11p59. Wang, X., Tian, Y., Zhao, X., 2017. The influence of dual-substrate-layer extensive green roofs on rainwater runoff quantity and quality. Sci. Total Environ. 592, 465–476. https://doi.org/10.1016/j.scitotenv.2017.03.124.
Whitbread, A., 1995. Soil organic matter: its fractionation and role in soil structure. In: In: Lefroy, R.D.B, Blair, G.J., Craswell, E.T. (Eds.), Soil Organic Matter Management for Sustainable Agriculture 56. ACIAR Proceedings, pp. 124–130.
Whittinghill, L.J., Rowe, D.B., 2011. Salt tolerance of common green roof and green wall plants. Urban Ecosyst. 14, 783–794.https://doi.org/10.1007/s11252-011-0169-4. Whittinghill, L.J., Bradley Rowe, D., Cregg, B.M., 2013. Evaluation of vegetable
pro-duction on extensive green roofs. Agroecol. Sustain. Food Syst. 37, 465–484.https:// doi.org/10.1080/21683565.2012.756847.
Williams, N.S.G., Rayner, J.P., Raynor, K.J., 2010. Green roofs for a wide brown land: Opportunities and barriers for rooftop greening in Australia. Urban For. Urban Green. 9, 245–251.https://doi.org/10.1016/j.ufug.2010.01.005.
Witter, E., Lopez-Real, J., 1988. Nitrogen losses during the composting of sewage sludge, and the effectiveness of clay soil, zeolite, and compost in adsorbing the volatilized ammonia. Biol. Wastes 23, 279–294.
Ye, J., Liu, C., Zhao, Z., Li, Y., Yu, S., 2013. Heavy metals in plants and substrate from simulated extensive green roofs. Ecol. Eng. 55, 29–34.https://doi.org/10.1016/j. ecoleng.2013.02.012.
