The influence of green roof features on their carbon and nutrient stocks

The influence of green roof features on their carbon and nutrient stocks | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The influence of green roof features on their carbon and nutrient stocks Laure Steenaerts, François Rineau, Tom Artois, Bernard Bosman, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6856018/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 17 Jan, 2026 Read the published version in Urban Ecosystems → Version 1 posted 9 You are reading this latest preprint version Abstract Extensive green roofs can be implemented to counteract the adverse effects of nature loss in urban areas by providing ecosystem services, such as biodiversity increase, storm water management and temperature regulation. Green roofs also sequester carbon (C) and retain other nutrients, improving urban air and rainwater quality. This study examined how green roof characteristics (i.e. green roof age, substrate depth, fertilization, vegetation type and composition) affect total carbon (TC), total nitrogen (TN) and total phosphorus (TP) stocks in the green roof substrate, as well as two important nitrogen fluxes (mineralization and nitrification). We hypothesized that vegetation type ( Sedum -only vs. diverse vegetation), substrate depth and fertilization would be the main characteristics affecting TC, TN and TP substrate stocks and nitrogen-fluxes. Twelve extensive green roofs in three cities in Flanders, Belgium, were sampled across four seasons. Results showed that green roofs have a low C sequestration potential compared to natural ecosystems. Roofs with diverse vegetation had higher TC and TN substrate stocks, particularly those with mosses and herbs. Fertilization had no significant effect on vegetation or TC, TN, and TP substrate stocks, while substrate depth had a significant effect only on TP substrate stocks. Overall, green roofs offer limited C sequestration and nutrient retention, although optimizing substrate composition and increasing plant species richness could enhance these benefits. This study highlights the potential for improving green roof performance through better design and management practices. Green roofs Carbon sequestration Nutrient retention Vegetation diversity Figures Figure 1 Figure 2 1. Introduction Green infrastructures, such as green roofs and green walls, are interesting options to integrate nature into urban ecosystems, contributing to the effort of mitigating global warming and counteracting biodiversity loss. Among these green infrastructures, extensive green roofs are commonly used due to their light-weighted substrate and low-growing vegetation. On extensive green roofs, the vegetation mainly consist of species of Sedum , hereafter referred to as succulents, frequently combined with annual herbs (Ampim, Sloan, Cabrera, Harp, & Jaber, 2010 ; Buffam & Mitchell, 2015 ; Manso, Teotónio, Silva, & Cruz, 2021 ). Extensive green roofs have been investigated for their potential to provide several ecosystem services in an urban context, such as increasing local biodiversity, mitigating heat island effects, and buffering stormwater (Jamei, Chau, Seyedmahmoudian, & Stojcevski, 2021 ; Wooster, Fleck, Torpy, Ramp, & Irga, 2022 ; Zhang, Lin, Zhang, & Ge, 2021 ). Another important ecosystem service is the potential for atmospheric carbon dioxide uptake via carbon (C) sequestration in the green roof substrate and vegetation. Green roofs are often promoted for their C sequestration potential (Getter et al., 2009 ; Aitkenhead-Peterson et al., 2011 ; Kuronuma et al., 2018 ; Shafique et al., 2020 ), but some studies found C sequestration to be rather low (Whittinghill et al., 2014 ; Agra et al., 2017 ). In addition, succulents have rather low photosynthetic levels (Starry, Lea-Cox, Kim, & van Iersel, 2014 ), and can switch to the crassulacean acid metabolism (CAM) in times of drought and heat, which are common conditions on green roofs. This reduces their C sequestration potential in comparison to C 3 and C 4 plants (Berndtsson et al., 2006 ; Getter et al., 2009 ; Agra et al., 2017 ). On the other hand, succulents have a lower decomposition rate than C 3 and C 4 plants, which improves their C sequestration potential (Berndtsson, et al., 2006 ; Johnson et al., 2016 ; Agra et al., 2017 ). Over the past years, research has focused on the C sequestration potential of green roofs in different set-ups, mostly focusing on one or two green roof characteristics, such as substrate composition and vegetation type (e.g. Shafique et al., 2020 ). For example, green roofs with a more diverse vegetation, containing herb species in addition to the typical succulents, have a higher litter input because annual herbs have a higher turnover rate compared to perennial succulents (Berndtsson, et al., 2006 ; Johnson et al., 2016 ; Agra et al., 2017 ). Indirectly, the presence of annual herbs could increase the TC substrate stocks, because roofs with such vegetation typically have a deeper substrate and are often fertilized once or twice a year, which promotes root system development and plant productivity, resulting in more below- and aboveground biomass and litter inputs (Johnson et al., 2016 ). This variation in experimental set-up produces contrasting results in literature. Although some studies show that green roofs can sequester a considerable amount of TC in plants and soil (Getter et al., 2009 ; Li and Babcock, 2014 ; Kuronuma et al., 2018 ; Shafique et al., 2020 ), other studies present results where Sedum- covered green roofs have a minor C sequestration potential compared to intensive green roofs or other plant systems on ground-level (Bouzouidja et al., 2018 ; Whittinghill et al., 2014 ). In addition, only few studies monitored green roofs over a longer period to investigate the effect of age on TC substrate stocks and found that older roofs had higher TC substrate stocks (Getter et al., 2009 ; Mitchell et al., 2021 ). Studies that investigate the relative importance of multiple green roof features on the biogeochemistry or ecosystem services of green roofs are lacking. Another potential ecosystem service green roofs provide is mitigating water pollution by retaining nutrients from atmospheric depositions, such as nitrogen (N), thereby lowering N concentrations in surface waters. Research has shown that green roofs can act both as sources and sinks of N, depending on vegetation and substrate characteristics and management practices (Berndtsson et al., 2006 ; Teemusk and Mander, 2007 , 2011 ; Wang et al., 2017 ; Karczmarczyk et al., 2018 ; Todorov et al., 2018 ). Ideally, a green roof would immobilize all added nutrients via atmospheric deposition (nitrogen species) and fertilization (nitrates, phosphates), implying no nutrient load for the city’s water system. Fertilization is an issue for their nutrient retention capacity, considering the already high atmospheric N deposition in many urban areas. Full nutrient recycling is, therefore, very unlikely in thin extensive green roofs that are often fertilized and are easily saturated with rainwater. This results into the direct runoff of excessive nutrients, such as nitrates and phosphates, via drainage water (Berndtsson et al., 2006 ; Clark and Zheng, 2014 ; Todorov et al., 2018 ). This N release from the green roof system is highly dependent on two particular N cycle fluxes: N mineralization and nitrification (De Neve, 2017 ). These microbial processes convert organic matter into plant-available inorganic N, such as NH 4 + and NO 3 − . High rates of N-mineralization and nitrification in the green roof substrate could result in an excess of inorganic N. In combination with the already high inorganic N input via atmospheric deposition and possible fertilization, N leaching and, consequently, eutrophication in downstream waters is plausible (Buffam & Mitchell, 2015 ; Mitchell, Hamilton, Uebel-Niemeier, Hopfensperger, & Buffam, 2018 ). In contrast to C stocks, previous research on nutrients (N and phosphorus (P)) has mainly focused on nutrient runoff rather than nutrient stocks and fluxes (Wang et al., 2017 ; Karczmarczyk et al. , 2018; Todorov et al., 2018 ). Again, vegetation type seems to be a major determinant of nutrient runoff from green roofs, where more diverse vegetation roofs have a higher nutrient retention potential. This effect can be attributed to differences in plant productivity and different plant nitrate requirements (Johnson et al., 2016 ). Indirectly, a more diverse vegetation could also increase the soil microbial diversity, thereby altering nutrient cycling (Hoch et al., 2019 ; Rumble & Gange, 2017 ). Fertilization is another potentially important determinant of TN and TP runoff. There is, however, little known about how fertilization affects the TN and TP stocks and whether the potential benefits of fertilization, i.e. more plant growth and coverage, outweigh the chances of nutrient leaching (Clark & Zheng, 2013 , 2014 ; Emilsson, Czemiel Berndtsson, Mattsson, & Rolf, 2007 ). The goal of this study is to investigate the effect of multiple green roof characteristics (i.e. substrate depth, substrate composition, vegetation type, green roof age and fertilizer application) onto the main features of the C, N and P cycle (TC stocks, nitrification,…). In addition, we examine two fluxes of the N-cycle, i.e. N-mineralization and nitrification rates, which have not been measured on green roofs so far. We hypothesize that green roofs will build up some TC in the substrate, and consequently older roofs will have higher TC substrate stocks compared with younger roofs. Furthermore, we hypothesize that vegetation characteristics, such as composition, biomass, and cover, determine TC and nutrient substrate stocks: we expect higher TC, TN and TP substrate stocks, when annual herb species are present, and when biomass and cover increase. In addition, we hypothesize that a green roof with a more diverse vegetation would have higher N-mineralization and nitrification rates. Finally, we hypothesize that fertilization is one of the main determinants of TC, TN and TP substrate stocks. This effect of fertilization could be positive or negative, depending on the ability of the substrate to retain nutrients. 2. Material and methods 2.1 Study site and green roof characteristics Twelve extensive green roofs were selected in three cities in Flanders: Hasselt (50° 56′ N, 5° 20′ E), Antwerp (51° 13′ N, 4° 24′ E), and Ghent (51° 3′ N, 3° 42′ E). During the sampling year (2019), yearly minimum and maximum temperature were 7°C and 16°C on average (Table 5 in supplementary material). The average total rainfall in 2019 in Flanders was 743 ± 12 mm, evenly distributed along the year (Table 6 in supplementary material). The selected green roofs were aged 4 to 15 years at the time of sampling, sized 25 to 777 m 2 , with a substrate depth between 4.5 and 12 cm; some roofs were fertilized once a year in spring with an organic controlled release fertilizer (Table 1 ). Five roofs were planted with only species of Sedum , referred to as “ Sedum -only roofs”, and seven roofs were planted with species of Sedum and a mix of annual herb species, named “diverse-vegetation roofs” (Table 1 ). On average for diverse-vegetation roofs 31% of the surface area was covered with Sedum , 20% with herbs and 31% with mosses. The remaining percentage (18%) was bare substrate (based on picture analyses by eye, see Fig. 3 in supplementary material). The substrate of these extensive green roofs consisted mainly of different lava stone fractions, expanded clay, crushed bricks, pumice stone and a small portion of compost to provide organic matter. The exact initial substrate composition of the examined green roofs was not known. Table 1 Characteristics of the studied extensive green roofs: roof ID, location, roof area, construction year, vegetation type, fertilization (Y = fertilized, N = not fertilized), average substrate depth in centimeters. Roof ID Location Area (m 2 ) Construction year Vegetation type Fertilization (Y/N) Average substrate depth (cm) 1 Ghent 25 2014 Diverse N 7.0 2 Ghent 110 2005 Sedum- only N 6.0 3 Ghent 588 2013 Sedum -only Y 5.0 4 Ghent 76 2015 Diverse N 8.0 5 Hasselt 432 2015 Diverse Y 11.0 6 Hasselt 108 2012 Sedum -only Y 5.0 7 Hasselt 175 2004 Diverse N 8.0 8 Hasselt 225 2015 Diverse Y 12.0 9 Antwerp 280 2008 Sedum -only Y 4.5 10 Antwerp 708 2014 Sedum -only Y 6.0 11 Antwerp 777 2009 Diverse Y 8.5 12 Antwerp 312 2015 Diverse Y 8.5 2.2 Sample collection and analyses All green roofs were examined seasonally, i.e. in April (spring; 2019), July (summer; 2019), October (autumn; 2019), and January (winter; 2020). Every roof was divided into virtual plots of 1m 2 , per sampling date (summer, autumn, winter) four plots per roof were randomly chosen. In spring, only three plots per roof were selected and roof 11 was not accessible. In the center of every plot, a subplot of 25 cm × 25 cm was defined. Pictures were taken to estimate the vegetation cover (see Fig. 3 ), the aboveground biomass was cut off and stored in plastic bags and the upper 6 cm of the substrate (or less in case the substrate was thinner) was collected. Samples were kept on ice and stored at 4°C until further analyses. One week after sampling, roots were carefully picked out of the substrate with a forceps and were rinsed with demineralized water. Roots as well as the aboveground biomass were oven-dried at 70°C for seven days after which the dry weight was measured. After weighing, the aboveground biomass was ground to fine powder with a mixer mill MM 200 (Retsch GmbH, Haan, Germany) and analyzed for total carbon (TC) and total nitrogen (TN) by dry combustion, based on the Dumas-method using an elemental analyzer (Flash 2000 CN Soil Analyser, Interscience, Louvain-la-Neuve, Belgium). Substrate samples, which were stored at 4°C, were processed 7 days after sampling. They were thoroughly mixed manually and divided into three subsamples for further analyses. Since it was practically not possible to take intact soil cores and the first 6 cm of the substrate was structurally homogeneous, one fresh subsample was used to estimate bulk density by drying a known volume of soil at 70°C for 7 days and weighing it afterwards (Blake, 1965 ). Another fresh subsample was passed through a 4-mm sieve and stored another 7 days (14 days after sampling) at 4°C. This sample was used to measure the net N-mineralization and net nitrification rate with an aerobic incubation method (Hart, Start, Davidson, & Firestone, 1994 ). The water holding capacity (WHC) and gravimetric water content were measured (oven-drying at 70°C for 7 days)) before incubation to adjust all samples to 60% WHC during incubation. For 28 days, 15 g soil was incubated at a constant temperature of 20°C in the dark and was kept at 60% WHC by weekly addition of distilled water (Hart et al., 1994 ). Before and after incubation, N-NO 3 and N-NH 4 were extracted with 1M KCl (1:5, w:v) (Allen, 1989 ) and analyzed colorimetrically with a continuous flow analyzer (AutoAnalyser 3, Bran + Leubbe, Germany). Net N-mineralization and nitrification rate were calculated as the net increase in inorganic N (N-NO 3 + N-NH 4 ) and nitrate (N-NO 3 ) during the 28-day incubation period. The third fresh subsample was passed through a 2-mm sieve and oven-dried at 70°C for 7 days. After drying, 10 g of soil was used to measure pH in deionized water (1:5, w:v) and 1M KCl (1:5, m:v) (Houba, van de Lee, Novazamsky, & Walinga, 1989 )). The rest of the dried soil was ground to pass a 0.25-mm sieve in an ultra-centrifugal mill (Model ZM 200, Retsch GmbH, Haan, Germany) and analyzed for TC and TN. Total phosphorus (TP) was measured with a continuous flow analyzer (San + + Skalar) after digestion with H 2 SO 4 -salicylic acid-H 2 O 2 and selenium (Novozamsky, Houba, van Eck, & van Vark, 1983 ). TC, TN, TP per m 2 were calculated by multiplying the raw values from the analyses (percentages for TC and TN and mg kg − 1 for TP) with the substrate depth and bulk density. 3. Statistical analyses 3.1 Carbon sequestration We investigated seasonal changes in substrate C stocks and used this as a proxy for C sequestration. The effect of season on substrate and aboveground biomass TC stocks was investigated through linear mixed models (LMM) (lme4 package; Bates et al., 2015 ) with season as a fixed factor and roof ID (1–12) as a random factor to account for repeated measures on the same roof. Assumptions of heteroscedasticity and normality of residuals were tested with Levene’s test (car package, Fox and Weisberg, 2019 ) and Shapiro-Wilk’s test, respectively. If assumptions were not met the dependent variable was log-transformed. Outliers were detected and removed by using “outlierstest” (car package; Fox and Weisberg, 2019 ) In case of a significant effect, post-hoc tests were caried out with Tukey HSD (multcomp package; Hothorn, Bretz and Westfall, 2008 ). 3.2 Effect of green roof characteristics on TC, TN, TP substrate stocks TC, TN, and TP substrate stocks were used as dependent variables. For each dependent variable, two LMMs (lme4 package; Bates et al., 2015 ) where made to avoid multicollinearity (VIF (Variance Inflation Factor): car package, Fox and Weisberg, 2019 ) between vegetation type ( Sedum -only vs. diverse vegetation) and percentage of plant cover. In the first model, vegetation type was the fixed factor and in the second model, fertilization, roof age, substrate depth, aboveground biomass, belowground biomass, succulent coverage, herb coverage and moss coverage were fixed factors. To account for correlations within one roof and through seasons, roof ID and season were added as crossed random factors. Assumptions of normality and heteroscedasticity were tested, and outliers were detected and removed as described above. Dependent variables were log-transformed if assumption were not met. The second model was also checked for multicollinearity with VIF (car package, (Fox & Weisberg, 2019 ). Preliminary analyses showed that roof 11 deviated strongly from all other roofs, therefore the analyses were done on the dataset without roof 11. Furthermore, to check whether the significant effect of vegetation type on TC, TN and TP substrate stocks was not correlated with fertilization, the same models were made with only fertilized roofs in the dataset (4 Sedum -only roofs, 4 diverse vegetation roofs). Additionally, to determine if significant differences in TC substrate between vegetation types could be caused by a difference in biomass or by a difference in C content between perennial succulents and annual herb species, three LMMs (lme4 package; Bates et al., 2015 ) were set up with aboveground biomass, belowground biomass and TC of the aboveground biomass as dependent variable and vegetation type as fixed factor. Again, with roof ID and season as crossed random factors. Model assumptions were tested as described above. 3.3 Effect of fertilization on green roof vegetation To assess the effect of fertilization on total plant coverage, succulent coverage, herb coverage, moss coverage, belowground and aboveground biomass, LMMs (lme4 package; Bates et al., 2015 ) were created, with season and roof ID as crossed random factors. For total plant, succulent, herb and moss coverage a generalized linear mixed model (GLMM) (lme4 package; Bates et al., 2015 ) was specified with a binomial distribution. Overdispersion was tested by giving all samples an ID number and comparing AIC values of the models with and without ID number. No model turned out to be overdispersed. Other assumptions (heteroscedasticity, normality) were tested as described above, and dependent variables were log-transformed when necessary. 4. Results 4.1 Total carbon stocks The investigated roofs showed in general rather low TC stocks. The total carbon (TC) stock of the aboveground biomass was significantly different between seasons ( p < 0.001, Type II Wald Chi squared test). A Tukey post-hoc test showed that the TC stocks in aboveground biomass are higher in summer than in spring ( p < 0.001) and winter ( p = 0.004) (Fig. 1 b). Plants (aboveground biomass) accumulated TC between spring and summer with an increase from 49 g TC m − 2 to 114 g TC m − 2 . After summer, TC stock in plants decreased, with average values of 67 g TC m − 2 and 62 g TC m − 2 , in autumn and winter, respectively (Fig. 1 b). Belowground biomass was very low, with an average maximum of 31 g dry weight m − 2 in autumn (Fig. 4 in supplementary material) The TC stock in the substrate of the measured green roofs declined over the period of one year (Fig. 1 a) with significant differences between spring-winter ( p = 0.011) and summer-winter ( p < 0.001). 4.2 Difference in substrate TC, TN and TP stocks and net N mineralization between Sedum -only and diverse vegetation roofs The average TC substrate stock of the twelve investigated green roof ranged between 700–2900 g C m − 2 (Table 2 ). Sedum -only roofs contained significantly ( p < 0.001) less substrate TC compared to diverse-vegetation roofs, ranging from 700–1000 g TC m − 2 and 1400–2900 g TC m − 2 , respectively (Table 2 , Fig. 2 a). Roof 12 showed a relatively low TC content for a diverse-vegetation roof. The significant difference of substrate TC between the two vegetation types persisted when only fertilized roofs ( p = 0.005) (n = 8; 4 Sedum -only roofs, 4 diverse-vegetation roofs) were considered (Fig. 5 in supplementary material). Linear mixed models with aboveground biomass ( p = 0.117), belowground biomass ( p = 0.920) and TC of the aboveground biomass ( p = 0.067) as dependent variables and vegetation type as independent variable, showed no significant effects (Fig. 6 in supplementary material). The average TN and TP content of the substrates were 40–100 g TN m − 2 and 40–100 g TP m − 2 (Table 2 ). Similar to the TC values, Sedum -only roofs had significantly less TN ( p < 0.001) in the substrate compared to diverse-vegetation roofs (Fig. 2 b) and, although not significant, their substrate TP stock ( p = 0.360) was also lower than in diverse-vegetation roofs (Fig. 2 c). When only fertilized roofs were included in the model, TN ( p = 0.025) remained significantly higher in diverse-vegetation roofs, while the difference in TP remained statistically non-significant ( p = 0.114) (Fig. 5 in supplementary material). Furthermore, the N-mineralization rate was highly variable among and within roofs, with values ranging from 0.16–1.44 mg N kg soil − 1 day − 1 (Table 2 ). Similarly, the net nitrification varied strongly among and within roofs, with values of 0.18–1.45 mg N kg soil − 1 day − 1 (Table 2 ). Table 2 Average values and standard deviation of substrate TC, TN, TP, N-mineralization, net nitrification and pH per roof across all seasons (n = 16: values ± S.D.). Values of roofs with diverse vegetation are in bold. TC ( g m − 2 ) TN ( g m − 2 ) TP ( g m − 2 ) N – mineralization (mg N kg − 1 soil day − 1 ) Net nitrification (mg N kg − 1 soil day − 1 ) pH Roof 1 1448 ± 641 90 ± 34 40 ± 8 0.22 ± 0.25 0.21 ± 0.24 6.22 ± 0.59 Roof 2 924 ± 779 58 ± 43 70 ± 12 0.28 ± 0.32 0.29 ± 0.3 6.04 ± 0.11 Roof 3 743 ± 164 48 ± 12 49 ± 9 0.57 ± 0.72 0.57 ± 0.74 6.02 ± 0.14 Roof 4 1382 ± 559 72 ± 27 68 ± 17 0.16 ± 0.29 0.18 ± 0.28 6.06 ± 0.15 Roof 5 2451 ± 496 85 ± 19 124 ± 19 0.71 ± 0.64 0.72 ± 0.67 6.5 ± 0.18 Roof 6 862 ± 507 54 ± 27 74 ± 8 1.2 ± 1.28 1.45 ± 1.66 5.61 ± 0.25 Roof 7 1922 ± 807 103 ± 46 78 ± 10 0.6 ± 0.39 0.62 ± 0.39 5.78 ± 0.34 Roof 8 2883 ± 293 90 ± 10 137 ± 12 0.18 ± 0.14 0.19 ± 0.16 6.69 ± 0.08 Roof 9 1047 ± 206 38 ± 9 80 ± 18 0.48 ± 0.39 0.51 ± 0.4 5.81 ± 0.14 Roof 10 714 ± 306 46 ± 22 48 ± 10 1.44 ± 1.18 1.53 ± 1.25 5.64 ± 0.31 Roof 11 5142 ± 527 167 ± 29 147 ± 21 2.06 ± 1.57 2.11 ± 1.56 5.9 ± 0.19 Roof 12 833 ± 255 62 ± 17 43 ± 10 0.61 ± 0.31 0.64 ± 0.28 6.56 ± 0.29 4.3 The effect of various green roof characteristics on TC, TN, TP stocks and green roof vegetation There was no significant relationship between substrate TC and the percentage of herb cover ( p = 0.676; Table 3 ) (Fig. 2 a). The percentage moss cover, however, had a significant positive effect on substrate TC ( p = 0.016) and TN ( p = 0.007). There was no significant increase in substrate TC with roof age ( p = 0.616). Furthermore, there was a significant influence of the aboveground biomass on TN stocks ( p = 0.019), with more TN with increasing biomass. The TP stock in the upper 6 cm of the substrate was significantly higher in roofs with greater substrate depths ( p = 0.012). Fertilization had no significant effect on TC, TN and TP substrate stocks (Table 3 ), nor on vegetation cover or biomass (Table 4 ). Table 3 Overview of Chi-square distribution values, p-values of LMMs with substrate TC, TN, TP as dependent variables and roof age in 2019 (years), roof depth (cm), aboveground biomass (g m − 2 ), belowground biomass (g m − 2 ), Sedum coverage (%), herb coverage (%), moss coverage (%) as fixed factors (roof 11 excluded). Dependent variable Fixed factor Chi-squared p-value TC Moss coverage 5.803 0.016 Aboveground biomass 3.006 0.083 Fertilization 1.723 0.189 Coverage of succulents 0.401 0.527 Roof depth 0.253 0.615 Roof age 4.448 0.616 Herb coverage 0.174 0.676 Belowground biomass 0.019 0.890 TN Moss coverage 7.396 0.007 Aboveground biomass 5.474 0.019 Coverage of succulents 2.426 0.119 Fertilization 1.624 0.203 Herb coverage 1.206 0.272 Belowground biomass 1.102 0.294 Roof age 6.883 0.332 Roof depth 0.138 0.710 TP Roof depth 6.304 0.012 Aboveground biomass 3.755 0.052 Herb coverage 1.938 0.164 Coverage of succulents 1.422 0.233 Belowground biomass 0.631 0.427 Fertilization 0.626 0.429 Moss coverage 0.078 0.780 Roof age 2.154 0.905 Table 4 Chi-squared and p-values from (G)LMM-models with ‘fertilization’ as fixed factor Dependent variable Chi-squared p-value Total plant coverage 0.059 0.808 Aboveground biomass 0.280 0.597 Belowground biomass 0.775 0.379 Coverage of succulents 0.250 0.617 Herb coverage 0.034 0.854 Moss coverage 0.892 0.345 5. Discussion 5.1 Total carbon stocks and carbon storage potential We measured TC values between 0.8–3 kg C m − 2 in the green roof substrate. These values are comparable with other studies, in which variation in concentration is explained by green roof properties, such as substrate type, substrate depth, and vegetation composition (Bouzouidja et al., 2018 ; Getter et al., 2009 ; Whittinghill et al., 2014 ). Compared to natural ecosystems (e.g. shrublands, grasslands, forests), where TC stocks are typically an order of magnitude higher (Smith et al., 2004 ; Whittinghill et al., 2014 ), green roofs contain relatively low TC stocks. Substrate TC stocks showed a decreasing trend from spring to winter, with significantly lower values in winter compared to summer and spring, suggesting a net loss of substrate TC. Losses of plant TC via litter and root exudates, have earlier been shown to become fixed in the green roof substrate (Kuronuma et al., 2018 ). In this study, however, we did not observe an increase in substrate TC in autumn when leaf litter is expected to be highest. The green roof substrate was the biggest pool of carbon, with 20–30 times more TC per square meter compared to the vegetation, making the potential annual input of plant TC difficult to detect. Furthermore, we observed no significant effect of green roof age on green roof TC substrate stocks, indicating extensive green roofs have a small potential to sequester C, even at the decadal time scale. Literature shows in general that green roofs can sequester C, however, it is often mentioned that Sedum -covered green roofs have a minor C sequestration potential in comparison with intensive green roofs or other in ground-systems (Bouzouidja et al., 2018 ; Getter et al., 2009 ; Kuronuma et al., 2018 ; Li & Babcock, 2014 ; Shafique et al., 2020 ; Whittinghill et al., 2014 ). Plant health and growth affected by a lot of parameters, such as soil temperature, soil moisture and soil nitrogen content (Mikan et al., 2002 ; Sinsabaugh et al., 2002 ; Xu, Baldocchi and Tang, 2004 ; Guntiñas et al., 2013 ). Moreover, green roofs are artificial environments, which are often exposed to extreme conditions, such as extreme temperatures and periods of drought (Oberndorfer et al., 2007 ) and in addition surrounding building structures influence shading, sun, wind, and rain exposure (Mitchell et al., 2021 ; Regan et al., 2014 ). All these factors can influence plant growth and thereby TC storage capacity. This makes every green roof unique and possibly determines their capacity to store TC. In conclusion, our findings suggest that extensive green roofs have a limited C storage potential. To optimize this minimal C storage potential, it is important to optimize vegetation growth and engineer the substrate to facilitate efficient C storage. In this optimization process it is essential to consider the various factors that determine the local environment of a specific green roof. 5.2 Effect of vegetation types 5.2.1 Carbon We tested the hypothesis that green roofs with a more diverse vegetation have higher substrate TC stocks compared to Sedum -only roofs. We observed that the substrate of diverse-vegetation roofs contained twice as much TC compared to Sedum -only roofs, albeit with a high variability in the results. It is plausible that higher TC stocks in diverse vegetation resulted from higher biomass production, due to a higher productivity of annual herbs species compared to succulents (Agra et al., 2017 ; Oberndorfer et al., 2007 ). There was, however, no significant difference in aboveground ( p = 0.490 ) and belowground biomass ( p = 0.492 ) between the two types of vegetation, suggesting that there was not more production of biomass on diverse-vegetation roofs. It is fact that the higher turnover rate of annual herbs compared to perennial succulents is expected to yield higher productivity per unit biomass (Agra et al., 2017 ; Oberndorfer et al., 2007 ). However, this is not supported by our data that indicate that the contribution of annual biomass production as a source of C input in the substrate of extensive green roofs is negligible and could not have resulted in a doubling of the total substrate C. We therefore speculate that difference in the initial substrate composition possibly caused this difference substrate C, as the substrate of diverse-vegetation roofs has in general a higher organic matter content to promote plant growth compared to Sedum- only roofs. This is in line with the findings of Shafique et al. ( 2020 ), who concluded that the substrate composition is vital for C capture and storage in green roof substrates. To clarify this, future research should focus on a more long-term follow-up of newly installed green roofs with these two vegetation types. 5.2.2. Nitrogen and phosphorus In addition to the effect of vegetation on the TC substrate stocks, we also investigated the hypothesis that diverse-vegetation roofs will have higher TN and TP substrate stocks and higher N-mineralization and nitrification rates. We observed that the TN concentrations in the substrate of diverse vegetation roofs are two times higher when compared with the substrate of Sedum -only roofs. Johnson et al. 2016 showed that a higher plant species richness enhances TN retention in green roof substrates. The green roofs with diverse vegetation in our study did not have a higher species richness compared to Sedum -only roofs (Table 7 in supplementary material). Consequently, our findings do not align with those of Johnson et al. 2016 , who demonstrated that green roofs have the potential to retain nitrogen. It must be noticed that, just as for TC the initial substrate composition could also play a role in the amount of TN in the substrate. However, we also measured older green roofs (> 4 years old), of which is hypothesized that TN present in the initial substrate would already be leached or consumed. Table 7 Number of plant species that were identified during four monitoring periods (May, June, August and September 2019) Roof ID Number of plant species 1 40 2 39 3 22 4 22 5 NA 6 NA 7 33 8 42 9 15 10 17 11 27 12 27 Furthermore, we observed that N-mineralization and nitrification rates are highly variable within and among roofs and could not be explained by vegetation type. To our knowledge there are no studies that measured actual N-mineralization and nitrification on green roofs. We suggest that other parameters causing spatial heterogeneity on green roofs, such as difference in sun and wind exposure and moisture content but also the small-scale spatial heterogeneity of the substrate itself, could be steering this high variability (Regan et al., 2014 ). Furthermore, monitoring N-fluxes in combination with the assessment of microbial and plant communities as well as other abiotic factors such as pH, bulk density, water holding capacity, …, seems necessary to unravel the N-fluxes in green roofs in more detail. In contrast to the TN substrate stocks, the TP substrate stocks did not significantly differ between diverse-vegetation roofs and Sedum -only roofs. Remarkably, TP substrate stocks are as high as TN substrate stocks, which indicates that the substrate is extremely P-rich. Research on TP stocks of green roofs is limited and mainly focuses on PO 4 3− leaching, showing that green roofs are often sources of phosphates (Karczmarczyk et al. 2018 ). Again, initial substrate composition could be the main determining factor of TP concentrations in the substrate, where TP concentrations are not only determined by the available PO 4 3− but also by mineral bound phosphorus, which is likely very high and similar in the measured green roof substrates. 5.3 The effect of other green roof characteristics on TC, TN, TP substrate stocks and green roof vegetation Besides the initial vegetation type, other green roofs parameters such as plant cover, plant biomass, substrate depth, roof age and maintenance practices can affect the TC, TN and TP retention capacity and the development of green roof vegetation (Bouzouidja et al., 2018 ; Clark & Zheng, 2013 ; Dusza et al., 2017 ; Hoch et al., 2019 ). We hypothesized that fertilization would be the main determinant of TC, TN and TP substrate stocks. Our results show moss coverage has a significant effect on the TC and TN substrate stocks. The investigated green roofs had an average moss cover of 31%. Mosses have a lower decomposition rate compared to vascular plants, possibly causing a more persistent litter layer which increases TC of the soil (Kasimir, He, Jansson, Lohila, & Minkkinen, 2021 ). Furthermore, mosses have high N retention capacities (Ayres, Van Der Wal, Sommerkorn, & Bardgett, 2006 ), which suggest that mosses would retain more N and not release it in the substrate. However, the harsh conditions on green roofs, such as drought and heat, could cause a higher litter input from mosses than in other environments, which would increase substrate TN. Although mosses are often esthetically not appreciated, they might be beneficial in terms of substrate TC and TN stocks. In addition, aboveground biomass showed a significant influence on TN substrate stocks, with more TN in the substrate with increasing biomass. We expected however to see the same effect for TC, as plant litter is a source of TN and TC. Nevertheless, we assume plant litter input on green roofs is minimal due to the low resilient vegetation. Therefore, this effect is likely to a higher nitrogen retention capacity, both by plants and soil particles, whereas this is less the case for carbon (Johnson et al., 2016 ). For TP substrate stock there only was a significant positive effect of substrate depth. In this study only the upper 6cm of all green roofs were sampled. Hereby the positive effect of substrate depth indicates that deeper roofs had more TP in the upper 6cm of the substrate. From this we conclude that a deeper green roof substrate promotes the build-up of P substrate stocks. A high TP substrate stock can imply that the organic P is not mineralized by present micro-organisms, thereby causing a phosphorus pool with low plant-availability. Furthermore, inorganic P could possibly attach to the mineral fraction of the green roof substrate, such as lava stones and expanded clay, contributing to a higher TP substrate stock (Karczmarczyk et al., 2018 ). It is however unclear, how these explanations are in relation with substrate depth. Another likely explanation is that P present in the soil solution is prone to leaching (Mitchell, Matter, Durtsche, & Buffam, 2017 ), whereby thinner substrates are more quickly saturated (van Seters, Rocha, Smith, & MacMillan, 2009 ) and will lose more P. There was no effect of herb of Sedum coverage and the TC, TN or TP substrate stocks, although plant density or coverage is a main determinant of carbon and nutrient cycling according to the study of Buffam and Mitchell, 2015 . These results, together with our result that diverse vegetation roofs have higher TC and TN substrate stocks compared to Sedum -only roofs, suggest that the presence or absence of annual herb species is more important than the actual percentage herb coverage. Further, we expected that green roofs will build up some TC in the substrate over their lifetime and that N and P substrate stocks will not be related with roof age, given their high susceptibility to leaching. Our results show no relation between roof age and TC, TN or TP substrate stocks. The age of the investigated roofs varied between 4 and 15 years during the sampling period. Other studies show that green roofs can build up TC and TN in their substrate (Köhler & Poll, 2010 ; Mitchell et al., 2021 ; Schrader & Böning, 2006 ). Comparing green roof age with TC, TN and TP substrate stocks is however far from straightforward. Besides green roof age, there are numerous green roof characteristics to consider, which influence the retention capacity and leaching susceptibility of the green roof substrate. A study that monitors green roofs throughout their lifespan and systematically monitors all green roof characteristics would be valuable in gaining further insight into the carbon and nutrient dynamics of green roofs over time. Last, there was no significant effect of fertilization on TP substrate stocks, nor on TN substrate stocks. This could indicate that excessive nutrient input via fertilizers is quickly lost by leaching (Clark & Zheng, 2013 , 2014 ). Moreover, our results showed no effect of fertilization on coverage or biomass, while one of the main reasons for the application of fertilizer is to increase the esthetical value of a green roof by enhancing plant cover and greenness. Clark et al., 2014, do show that fertilizer application helps to establish plant growth and roof coverage. This suggests that fertilization application is not useful once a green roof is already fully covered, which could be an indication of green roofs being in a steady-state after 4 years of age (Mitchell et al., 2021 ), with constant pools and fluxes. 6. Conclusion In conclusion, we found that extensive green roofs covered with mainly succulent species have a low potential to sequester carbon, likely because of the low and slow growing vegetation and the porous and artificial substrate. We showed that adding annual herb species was, probably indirectly, beneficial for TC substrate stocks and the closely linked N substrate stocks. Moreover, the presence of mosses contributed most to higher TC and TN substrate stocks, while they are usually given a low esthetical value. Furthermore, our results indicate that fertilizer application is not useful once extensive green roofs reach a certain age, which could be very important when considering eutrophication in downstream waterways. Therefore, we presume that when extensive green roofs are given the chance to develop as natural pioneer ecosystems, they might reach a steady state, where maintenance and fertilization is not necessary. To promote a more natural development of extensive green roofs, locally adapted plant species could be integrated into green roof designs. Although the use of natural soil is often not feasible due to its weight, alternative substrates that mimic natural soil properties while remaining lightweight, such as biochar-based mixtures, could be explored. Considering that current artificial substrates like expanded clay and lava stones are derived from finite resources, such alternatives may offer a more sustainable solution. Moreover, using local vegetation in combination with more soil-like, yet lightweight substrates may enhance the ability of green roofs to deliver ecosystem services aligned with local climate conditions and biodiversity. Additionally, we propose that the initial substrate composition could be the main factor driving TC, TN and TP substrate stocks. Adding extra C to substrate during the installation could be a way to store external C, such as compost or biochar, since the carbon in the soils did not decline with age and is apparently stable. For this to succeed, it is imperative that the extra C is ecologically produced with minor CO 2 emissions. This would be an interesting path for future research. Furthermore, comprehensively mapping all possible fluxes and stocks, as well as the various factors having an impact on the fluxes and stocks, and monitoring these over an extended period, ideally even throughout the entire lifespan of a green roof, could clarify the many uncertainties that currently still exist. Declarations Competing interests The authors declare no conflicts of interest. Funding The author L.S. and the research were financed by Research Foundation - Flanders (FWO) via a Strategic Basic Research project (S002818N: “EcoCities: Green roofs and walls as a source for ecosystem services in future cities”). Author Contribution L.S., F.R., I.J., E.S. and T.A contributed to the study conception and design. Samples were taken by L.S and T.V.D. Analyses were performed by L.S, B.B and M.P-E. All plant species were identified, and coverage percentages were estimated by C.V.M. Statistical analyses were carried out by L.S. The first draft of the paper was written by L.S. and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Acknowledgments This study was supported by the SBO project S002818N "EcoCities: Green roofs and walls as a source for ecosystem services in future cities”. Data Availability The research data is available upon request. References Agra, H., Klein, T., Vasl, A., Kadas, G., & Blaustein, L. (2017). Measuring the effect of plant-community composition on carbon fixation on green roofs. Urban Forestry and Urban Greening , 24 , 1–4. https://doi.org/10.1016/j.ufug.2017.03.003 Aitkenhead-Peterson, J. A., Dvorak, B. D., Volder, A., & Stanley, N. C. (2011). Chemistry of growth medium and leachate from green roof systems in south-central Texas. Urban Ecosystems , 14 (1), 17–33. https://doi.org/10.1007/s11252-010-0137-4 Allen, S. E. (1989). Chemical analysis of Ecological Materials (2nd editio). Oxford, London: Blackwell Scientific Publications. 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. Journal of Environmental Horticulture , 28 (4), 244–252. https://doi.org/10.24266/0738-2898-28.4.244 Ayres, E., Van Der Wal, R., Sommerkorn, M., & Bardgett, R. D. (2006). Direct uptake of soil nitrogen by mosses. Biology Letters , 2 (2), 286–288. https://doi.org/10.1098/rsbl.2006.0455 Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software , 67 (1), 1–48. https://doi.org/10.18637/jss.v067.i01 Berndtsson, J. C., Emilsson, T., & Bengtsson, L. (2006). The influence of extensive vegetated roofs on runoff water quality. Science of the Total Environment , 355 (1–3), 48–63. https://doi.org/10.1016/j.scitotenv.2005.02.035 Blake, G. R. (1965). Bulk density. In Methods of Soil Analysis, Part 1: Physical and Mineralogical Properties, Including Statistics of Measurement and Sampling (9th ed.). https://doi.org/10.2134/agronmonogr9.1.c30 Bouzouidja, R., Rousseau, G., Galzin, V., Claverie, R., Lacroix, D., & Séré, G. (2018). Green roof ageing or Isolatic Technosol’s pedogenesis? Journal of Soils and Sediments , 18 (2), 418–425. https://doi.org/10.1007/s11368-016-1513-3 Buffam, I., & Mitchell, M. E. (2015). Nutrient cycling in green roof ecosystems. In Green roof ecosystems (pp. 107–137). https://doi.org/10.1007/978-3-319-14983-7 Clark, M. J., & Zheng, Y. (2013). Plant nutrition requirements for an installed sedum-vegetated green roof module system: Effects of fertilizer rate and type on plant growth and leachate nutrient content. HortScience , 48 (9), 1173–1180. Clark, M. J., & Zheng, Y. (2014). Effect of fertilizer rate on plant growth and leachate nutrient content during production of sedum-vegetated green roof modules. HortScience , 49 (6), 819–826. De Neve, S. (2017). Organic matter mineralisation as a source of nitrogen. In F. Tei (Ed.), Advances in Research on Fertilization Managment of Vegetable Crops (pp. 65–83). https://doi.org/10.1007/978-3-319-53626-2 Dusza, Y., Barot, S., Kraepiel, Y., Lata, J. C., Abbadie, L., & Raynaud, X. (2017). Multifunctionality is affected by interactions between green roof plant species, substrate depth, and substrate type. Ecology and Evolution , 7 (7), 2357–2369. https://doi.org/10.1002/ece3.2691 Emilsson, T., Czemiel Berndtsson, J., Mattsson, J. E., & Rolf, K. (2007). Effect of using conventional and controlled release fertiliser on nutrient runoff from various vegetated roof systems. Ecological Engineering , 29 (3), 260–271. https://doi.org/10.1016/j.ecoleng.2006.01.001 Fox, J., & Weisberg, S. (2019). An {R} Companion to Applied Regression (p. Third Edition). p. Third Edition. Retrieved from https://socialsciences.mcmaster.ca/jfox/Books/Companion/ Getter, K. L., Rowe, D. B., Robertson, G. P., Cregg, B. M., & Andresen, J. A. (2009). Carbon Sequestration Potential of Extensive Green Roofs. Environmental Science & Technology , 43 (19), 7564–7570. https://doi.org/10.1021/es901539x Guntiñas, M. E., Gil-Sotres, F., Leirós, M. C., & Trasar-Cepeda, C. (2013). Sensitivity of soil respiration to moisture and temperature. Journal of Soil Science and Plant Nutrition , 13 (2), 445–461. https://doi.org/10.4067/S0718-95162013005000035 Hart, S. C., Start, J. M., Davidson, E. A., & Firestone, M. K. (1994). Nitrogen mineralization, immobilization, and nitrification. In R. Weaver, J. Angle, P. Bottomley, D. Bezdicek, S. Smith, & A. Tabatabai (Eds.), Methods of Soil Analysis. Part 2. Microbial and Biochemical Properties. (pp. 985–1018). https://doi.org/10.2136/sssabookser5.2.c42 Hoch, J. M. K., Rhodes, M. E., Shek, K. L., Dinwiddie, D., Hiebert, T. C., Gill, A. S., … McGuire, K. L. (2019). Soil microbial assemblages are linked to plant community composition and contribute to ecosystem services on urban green roofs. Frontiers in Ecology and Evolution , 7 (JUN). https://doi.org/10.3389/fevo.2019.00198 Hothorn, T., Bretz, F., & Westfall, P. (2008). Simultaneous Inference in General Parametric Models. Biometrical Journal , 50 (3), 346–363. Houba, V. J. G., van de Lee, J. J., Novazamsky, I., & Walinga, I. (1989). Soil and plant analysis, a series of syllabi. Part 5: Soil Analysis Procedures . Wageningen: Agricultural University, Wageningen. Jamei, E., Chau, H. W., Seyedmahmoudian, M., & Stojcevski, A. (2021). Review on the cooling potential of green roofs in different climates. Science of the Total Environment , 791 , 148407. https://doi.org/10.1016/j.scitotenv.2021.148407 Johnson, C., Schweinhart, S., & Buffam, I. (2016). Plant species richness enhances nitrogen retention in green roof plots. Ecological Applications , 26 (7), 2130–2144. https://doi.org/10.1890/15-1850.1 Karczmarczyk, A., Bus, A., & Baryla, A. (2018). Phosphate leaching from green roof substrates-Can green roofs pollute urban water bodies? Water , 10 . https://doi.org/10.3390/w10020199 Kasimir, He, H., Jansson, P. E., Lohila, A., & Minkkinen, K. (2021). Mosses are Important for Soil Carbon Sequestration in Forested Peatlands. Frontiers in Environmental Science , 9 (September), 1–19. https://doi.org/10.3389/fenvs.2021.680430 Köhler, M., & Poll, P. (2010). Long-term performance of selected old Berlin green roofs in comparison to younger extensive green roofs in Berlin. Ecological Engineering , 36 , 722–729. Kuronuma, T., Watanabe, H., Ishihara, T., Kou, D., Toushima, K., Ando, M., & Shindo, S. (2018). CO2 Payoffof extensive green roofs with different vegetation species. Sustainability (Switzerland) , 10 (7), 1–12. https://doi.org/10.3390/su10072256 Li, Y., & Babcock, R. W. (2014). Green roofs against pollution and climate change. A review. Agronomy for Sustainable Development , 34 (4), 695–705. https://doi.org/10.1007/s13593-014-0230-9 Manso, M., Teotónio, I., Silva, C. M., & Cruz, C. O. (2021). Green roof and green wall benefits and costs: A review of the quantitative evidence. Renewable and Sustainable Energy Reviews , 135 (July 2020). https://doi.org/10.1016/j.rser.2020.110111 Mikan, C. J., Schimel, J. P., & Doyle, A. P. (2002). Temperature controls of microbial respiration in arctic tundra soils above and below freezing. Soil Biology and Biochemistry , 34 (11), 1785–1795. https://doi.org/10.1016/S0038-0717(02)00168-2 Mitchell, M. E., Emilsson, T., & Buffam, I. (2021). Carbon, nitrogen, and phosphorus variation along a green roof chronosequence: Implications for green roof ecosystem development. Ecological Engineering , 164 (March), 106211. https://doi.org/10.1016/j.ecoleng.2021.106211 Mitchell, M. E., Hamilton, T. L., Uebel-Niemeier, C., Hopfensperger, K. N., & Buffam, I. (2018). Nitrogen cycling players and processes in green roof ecosystems. Applied Soil Ecology , 132 (June), 114–125. https://doi.org/10.1016/j.apsoil.2018.08.007 Mitchell, M. E., Matter, S. F., Durtsche, R. D., & Buffam, I. (2017). Elevated phosphorus: dynamics during four years of green roof development. Urban Ecosystems , 20 (5), 1121–1133. https://doi.org/10.1007/s11252-017-0664-3 Novozamsky, V. J. G., Houba, R., van Eck, R., & van Vark, W. (1983). A novel digestion technique for multi-element plant analysis. Communications in Soil Science and Plant Analysis , 14 , 239–249. Oberndorfer, E., Lundholm, J., Bass, B., Coffman, R. R., Doshi, H., Dunnett, N., … Rowe, B. (2007). Green Roofs as Urban Ecosystems: Ecological Structures, Functions, and Services. BioScience , 57 (10), 823–833. https://doi.org/10.1641/B571005 Regan, K. M., Nunan, N., Boeddinghaus, R. S., Baumgartner, V., Berner, D., Boch, S., … Marhan, S. (2014). Seasonal controls on grassland microbial biogeography: Are they governed by plants, abiotic properties or both? Soil Biology and Biochemistry , 71 , 21–30. https://doi.org/10.1016/j.soilbio.2013.12.024 Rumble, H., & Gange, A. C. (2017). Microbial inoculants as a soil remediation tool for extensive green roofs. Ecological Engineering , 102 , 188–198. https://doi.org/10.1016/j.ecoleng.2017.01.025 Schrader, S., & Böning, M. (2006). Soil formation on green roofs and its contribution to urban biodiversity with emphasis on Collembolans. Pedobiologia , 50 , 347–356. Shafique, M., Xue, X., & Luo, X. (2020). An overview of carbon sequestration of green roofs in urban areas. Urban Forestry and Urban Greening , 47 (May 2019), 126515. https://doi.org/10.1016/j.ufug.2019.126515 Sinsabaugh, R. L., Carreiro, M. M., & Repert, D. A. (2002). Allocation of extracellular enzymatic activity in relation to litter composition, N deposition, and mass loss. Biogeochemistry , 60 (1), 1–24. https://doi.org/10.1023/A:1016541114786 Smith, J. E., Heath, L. S., & Woodbury, P. B. (2004). How to estimate forest carbon for large areas from inventory data. Journal of Forestry , 102 (5), 25–31. https://doi.org/10.1093/jof/102.5.25 Starry, O., Lea-Cox, J. D., Kim, J., & van Iersel, M. W. (2014). Photosynthesis and water use by two Sedum species in green roof substrate. Environmental and Experimental Botany , 107 , 105–112. https://doi.org/10.1016/j.envexpbot.2014.05.014 Teemusk, A., & Mander, Ü. (2007). Rainwater runoff quantity and quality performance from a greenroof: The effects of short-term events. Ecological Engineering , 30 , 271–277. https://doi.org/10.1016/j.ecoleng.2007.01.009 Teemusk, A., & Mander, Ü. (2011). The Influence of Green Roofs on Runoff Water Quality: A Case Study from Estonia. Water Resources Management , 25 , 3699–3713. https://doi.org/10.1007/s11269-011-9877-z Todorov, D., Driscoll, C. T., Todorova, S., & Montesdeoca, M. (2018). Water quality function of an extensive vegetated roof. Science of the Total Environment , 625 , 928–939. https://doi.org/10.1016/j.scitotenv.2017.12.085 van Seters, T., Rocha, L., Smith, D., & MacMillan, G. (2009). Evaluation of green roofs for runoff retention, runoff quality, and leachability. Water Quality Research Journal of Canada , 44 (1), 33–47. https://doi.org/10.2166/wqrj.2009.005 Wang, H., Qin, J., & Hu, Y. (2017). Are green roofs a source or sink of runoff pollutants? Ecological Engineering , 107 (3888), 65–70. https://doi.org/10.1016/j.ecoleng.2017.06.035 Whittinghill, L. J., Rowe, D. B., Schutzki, R., & Cregg, B. M. (2014). Quantifying carbon sequestration of various green roof and ornamental landscape systems. Landscape and Urban Planning , 123 , 41–48. https://doi.org/10.1016/j.landurbplan.2013.11.015 Wooster, E. I. F., Fleck, R., Torpy, F., Ramp, D., & Irga, P. J. (2022). Urban green roofs promote metropolitan biodiversity: A comparative case study. Building and Environment , 207 (PA), 108458. https://doi.org/10.1016/j.buildenv.2021.108458 Xu, L., Baldocchi, D. D., & Tang, J. (2004). How soil moisture, rain pulses, and growth alter the response of ecosystem respiration to temperature. Global Biogeochemical Cycles , 18 (4), 1–10. https://doi.org/10.1029/2004GB002281 Zhang, S., Lin, Z., Zhang, S., & Ge, D. (2021). Stormwater retention and detention performance of green roofs with different substrates: Observational data and hydrological simulations. Journal of Environmental Management , 291 (April). https://doi.org/10.1016/j.jenvman.2021.112682 Additional Declarations No competing interests reported. Supplementary Files SUPPLEMENTARYMATERIAL.docx Cite Share Download PDF Status: Published Journal Publication published 17 Jan, 2026 Read the published version in Urban Ecosystems → Version 1 posted Editorial decision: Revision requested 03 Nov, 2025 Reviews received at journal 30 Oct, 2025 Reviews received at journal 16 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers agreed at journal 31 Aug, 2025 Reviewers invited by journal 25 Aug, 2025 Editor assigned by journal 23 Jun, 2025 Submission checks completed at journal 23 Jun, 2025 First submitted to journal 09 Jun, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6856018","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":508380618,"identity":"b7ea1a90-f9aa-40d8-945c-430f7fd2426c","order_by":0,"name":"Laure Steenaerts","email":"data:image/png;base64,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","orcid":"","institution":"Hasselt University","correspondingAuthor":true,"prefix":"","firstName":"Laure","middleName":"","lastName":"Steenaerts","suffix":""},{"id":508380619,"identity":"38d90301-a617-444e-b4f6-33a68d258e3f","order_by":1,"name":"François Rineau","email":"","orcid":"","institution":"Hasselt University","correspondingAuthor":false,"prefix":"","firstName":"François","middleName":"","lastName":"Rineau","suffix":""},{"id":508380620,"identity":"748dccf1-1a8c-44b0-aa05-3d11aa19f81f","order_by":2,"name":"Tom Artois","email":"","orcid":"","institution":"Hasselt University","correspondingAuthor":false,"prefix":"","firstName":"Tom","middleName":"","lastName":"Artois","suffix":""},{"id":508380621,"identity":"e70031eb-fb5f-459d-954a-ffb8aab171a3","order_by":3,"name":"Bernard Bosman","email":"","orcid":"","institution":"University of Liège","correspondingAuthor":false,"prefix":"","firstName":"Bernard","middleName":"","lastName":"Bosman","suffix":""},{"id":508380622,"identity":"ce819830-0b7f-4766-ab0d-e8a95676fdd5","order_by":4,"name":"Monique Carnol","email":"","orcid":"","institution":"University of Liège","correspondingAuthor":false,"prefix":"","firstName":"Monique","middleName":"","lastName":"Carnol","suffix":""},{"id":508380623,"identity":"85c49ea2-7840-4277-b0c6-3ad924f552af","order_by":5,"name":"Miguel Portillo-Estrada","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Miguel","middleName":"","lastName":"Portillo-Estrada","suffix":""},{"id":508380624,"identity":"26d27d60-faf8-4210-9b73-5f9ced9d1cd6","order_by":6,"name":"Eric Struyf","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Eric","middleName":"","lastName":"Struyf","suffix":""},{"id":508380625,"identity":"2c85f426-9988-40bd-9d65-d264e4fb740b","order_by":7,"name":"Thomas Van Dijck","email":"","orcid":"","institution":"Hasselt University","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"Van","lastName":"Dijck","suffix":""},{"id":508380626,"identity":"74e3df87-ed99-49f3-bb41-fa0ee658dfe6","order_by":8,"name":"Carmen Van Mechelen","email":"","orcid":"","institution":"Hogeschool PXL","correspondingAuthor":false,"prefix":"","firstName":"Carmen","middleName":"Van","lastName":"Mechelen","suffix":""},{"id":508380627,"identity":"63adcd3b-3bbc-4c6e-ac37-f0a9524db5c4","order_by":9,"name":"Ivan Janssens","email":"","orcid":"","institution":"University of Antwerp","correspondingAuthor":false,"prefix":"","firstName":"Ivan","middleName":"","lastName":"Janssens","suffix":""}],"badges":[],"createdAt":"2025-06-09 15:53:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6856018/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6856018/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11252-025-01898-x","type":"published","date":"2026-01-17T16:30:45+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":90397840,"identity":"bcc007c7-615c-4701-ac24-e0c1b12f6061","added_by":"auto","created_at":"2025-09-02 09:43:47","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":57148,"visible":true,"origin":"","legend":"\u003cp\u003eAverage \u003cstrong\u003e(a) \u003c/strong\u003esubstrate TC (g m\u003csup\u003e−2\u003c/sup\u003e) \u003cstrong\u003e(b)\u003c/strong\u003e TC of aboveground biomass (g m\u003csup\u003e-2\u003c/sup\u003e) per season. Boxplots show the 25\u003csup\u003eth\u003c/sup\u003e, 50\u003csup\u003eth\u003c/sup\u003e and 75\u003csup\u003eth\u003c/sup\u003e percentiles, with whiskers reaching 2.5% and 97.5% of the distribution, with the average as a dot inside the box and outliers outside the range. P-values between seasons after Tukey post hoc tests (lmer-model).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6856018/v1/96089a047a014013ecf8745f.png"},{"id":90397841,"identity":"f98fcfc7-0054-42e1-a05f-7a473c01ef53","added_by":"auto","created_at":"2025-09-02 09:43:47","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":50249,"visible":true,"origin":"","legend":"\u003cp\u003eAverage (\u003cstrong\u003ea\u003c/strong\u003e) TC (g m\u003csup\u003e-2\u003c/sup\u003e), (\u003cstrong\u003eb\u003c/strong\u003e) TN (g m\u003csup\u003e-2\u003c/sup\u003e), and (\u003cstrong\u003ec\u003c/strong\u003e) TP (g m\u003csup\u003e-2\u003c/sup\u003e) in green roof substrate of Sedum-only roofs and diverse-vegetation roofs (fertilized + non fertilized). Boxplots show the 25\u003csup\u003eth\u003c/sup\u003e, 50\u003csup\u003eth\u003c/sup\u003e and 75\u003csup\u003eth\u003c/sup\u003e percentiles, with whiskers reaching 2.5% and 97.5% of the distribution, with the average as a dot inside the box and outliers outside the range. P-values between vegetation types after Type II Wald Chisquare tests (lmer-model).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6856018/v1/ee6a42a63407e4ceb8cd3111.png"},{"id":100616131,"identity":"0f8cdb3f-ed2a-477a-aaa9-00e3c5033a23","added_by":"auto","created_at":"2026-01-19 17:40:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1545692,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6856018/v1/ca90e877-711f-4d4d-bd27-53f7d060d838.pdf"},{"id":90397842,"identity":"6facd2da-14bc-449a-b53c-4c98016bd81d","added_by":"auto","created_at":"2025-09-02 09:43:47","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1412626,"visible":true,"origin":"","legend":"","description":"","filename":"SUPPLEMENTARYMATERIAL.docx","url":"https://assets-eu.researchsquare.com/files/rs-6856018/v1/dfad62ca44e1338aca50ee5c.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"The influence of green roof features on their carbon and nutrient stocks","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eGreen infrastructures, such as green roofs and green walls, are interesting options to integrate nature into urban ecosystems, contributing to the effort of mitigating global warming and counteracting biodiversity loss. Among these green infrastructures, extensive green roofs are commonly used due to their light-weighted substrate and low-growing vegetation. On extensive green roofs, the vegetation mainly consist of species of \u003cem\u003eSedum\u003c/em\u003e, hereafter referred to as succulents, frequently combined with annual herbs (Ampim, Sloan, Cabrera, Harp, \u0026amp; Jaber, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Buffam \u0026amp; Mitchell, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Manso, Teot\u0026oacute;nio, Silva, \u0026amp; Cruz, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Extensive green roofs have been investigated for their potential to provide several ecosystem services in an urban context, such as increasing local biodiversity, mitigating heat island effects, and buffering stormwater (Jamei, Chau, Seyedmahmoudian, \u0026amp; Stojcevski, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Wooster, Fleck, Torpy, Ramp, \u0026amp; Irga, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhang, Lin, Zhang, \u0026amp; Ge, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Another important ecosystem service is the potential for atmospheric carbon dioxide uptake via carbon (C) sequestration in the green roof substrate and vegetation. Green roofs are often promoted for their C sequestration potential (Getter et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Aitkenhead-Peterson et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Kuronuma et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Shafique et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), but some studies found C sequestration to be rather low (Whittinghill et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Agra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). In addition, succulents have rather low photosynthetic levels (Starry, Lea-Cox, Kim, \u0026amp; van Iersel, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), and can switch to the crassulacean acid metabolism (CAM) in times of drought and heat, which are common conditions on green roofs. This reduces their C sequestration potential in comparison to C\u003csub\u003e3\u003c/sub\u003e and C\u003csub\u003e4\u003c/sub\u003e plants (Berndtsson et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Getter et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Agra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). On the other hand, succulents have a lower decomposition rate than C\u003csub\u003e3\u003c/sub\u003e and C\u003csub\u003e4\u003c/sub\u003e plants, which improves their C sequestration potential (Berndtsson, et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Johnson et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Agra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Over the past years, research has focused on the C sequestration potential of green roofs in different set-ups, mostly focusing on one or two green roof characteristics, such as substrate composition and vegetation type (e.g. Shafique et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). For example, green roofs with a more diverse vegetation, containing herb species in addition to the typical succulents, have a higher litter input because annual herbs have a higher turnover rate compared to perennial succulents (Berndtsson, et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Johnson et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Agra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Indirectly, the presence of annual herbs could increase the TC substrate stocks, because roofs with such vegetation typically have a deeper substrate and are often fertilized once or twice a year, which promotes root system development and plant productivity, resulting in more below- and aboveground biomass and litter inputs (Johnson et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). This variation in experimental set-up produces contrasting results in literature. Although some studies show that green roofs can sequester a considerable amount of TC in plants and soil (Getter et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Li and Babcock, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Kuronuma et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Shafique et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), other studies present results where \u003cem\u003eSedum-\u003c/em\u003ecovered green roofs have a minor C sequestration potential compared to intensive green roofs or other plant systems on ground-level (Bouzouidja et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Whittinghill et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). In addition, only few studies monitored green roofs over a longer period to investigate the effect of age on TC substrate stocks and found that older roofs had higher TC substrate stocks (Getter et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Mitchell et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Studies that investigate the relative importance of multiple green roof features on the biogeochemistry or ecosystem services of green roofs are lacking.\u003c/p\u003e\u003cp\u003eAnother potential ecosystem service green roofs provide is mitigating water pollution by retaining nutrients from atmospheric depositions, such as nitrogen (N), thereby lowering N concentrations in surface waters. Research has shown that green roofs can act both as sources and sinks of N, depending on vegetation and substrate characteristics and management practices (Berndtsson et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Teemusk and Mander, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Karczmarczyk et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Todorov et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Ideally, a green roof would immobilize all added nutrients via atmospheric deposition (nitrogen species) and fertilization (nitrates, phosphates), implying no nutrient load for the city\u0026rsquo;s water system. Fertilization is an issue for their nutrient retention capacity, considering the already high atmospheric N deposition in many urban areas. Full nutrient recycling is, therefore, very unlikely in thin extensive green roofs that are often fertilized and are easily saturated with rainwater. This results into the direct runoff of excessive nutrients, such as nitrates and phosphates, via drainage water (Berndtsson et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Clark and Zheng, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Todorov et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). This N release from the green roof system is highly dependent on two particular N cycle fluxes: N mineralization and nitrification (De Neve, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). These microbial processes convert organic matter into plant-available inorganic N, such as NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e. High rates of N-mineralization and nitrification in the green roof substrate could result in an excess of inorganic N. In combination with the already high inorganic N input via atmospheric deposition and possible fertilization, N leaching and, consequently, eutrophication in downstream waters is plausible (Buffam \u0026amp; Mitchell, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Mitchell, Hamilton, Uebel-Niemeier, Hopfensperger, \u0026amp; Buffam, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In contrast to C stocks, previous research on nutrients (N and phosphorus (P)) has mainly focused on nutrient runoff rather than nutrient stocks and fluxes (Wang et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Karczmarczyk \u003cem\u003eet al.\u003c/em\u003e, 2018; Todorov et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Again, vegetation type seems to be a major determinant of nutrient runoff from green roofs, where more diverse vegetation roofs have a higher nutrient retention potential. This effect can be attributed to differences in plant productivity and different plant nitrate requirements (Johnson et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Indirectly, a more diverse vegetation could also increase the soil microbial diversity, thereby altering nutrient cycling (Hoch et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Rumble \u0026amp; Gange, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Fertilization is another potentially important determinant of TN and TP runoff. There is, however, little known about how fertilization affects the TN and TP stocks and whether the potential benefits of fertilization, i.e. more plant growth and coverage, outweigh the chances of nutrient leaching (Clark \u0026amp; Zheng, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2013\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Emilsson, Czemiel Berndtsson, Mattsson, \u0026amp; Rolf, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe goal of this study is to investigate the effect of multiple green roof characteristics (i.e. substrate depth, substrate composition, vegetation type, green roof age and fertilizer application) onto the main features of the C, N and P cycle (TC stocks, nitrification,\u0026hellip;). In addition, we examine two fluxes of the N-cycle, i.e. N-mineralization and nitrification rates, which have not been measured on green roofs so far. We hypothesize that green roofs will build up some TC in the substrate, and consequently older roofs will have higher TC substrate stocks compared with younger roofs. Furthermore, we hypothesize that vegetation characteristics, such as composition, biomass, and cover, determine TC and nutrient substrate stocks: we expect higher TC, TN and TP substrate stocks, when annual herb species are present, and when biomass and cover increase. In addition, we hypothesize that a green roof with a more diverse vegetation would have higher N-mineralization and nitrification rates. Finally, we hypothesize that fertilization is one of the main determinants of TC, TN and TP substrate stocks. This effect of fertilization could be positive or negative, depending on the ability of the substrate to retain nutrients.\u003c/p\u003e"},{"header":"2. Material and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Study site and green roof characteristics\u003c/h2\u003e\u003cp\u003eTwelve extensive green roofs were selected in three cities in Flanders: Hasselt (50\u0026deg; 56\u0026prime; N, 5\u0026deg; 20\u0026prime; E), Antwerp (51\u0026deg; 13\u0026prime; N, 4\u0026deg; 24\u0026prime; E), and Ghent (51\u0026deg; 3\u0026prime; N, 3\u0026deg; 42\u0026prime; E). During the sampling year (2019), yearly minimum and maximum temperature were 7\u0026deg;C and 16\u0026deg;C on average (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e5\u003c/span\u003e in supplementary material). The average total rainfall in 2019 in Flanders was 743\u0026thinsp;\u0026plusmn;\u0026thinsp;12 mm, evenly distributed along the year (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e6\u003c/span\u003e in supplementary material).\u003c/p\u003e\u003cp\u003eThe selected green roofs were aged 4 to 15 years at the time of sampling, sized 25 to 777 m\u003csup\u003e2\u003c/sup\u003e, with a substrate depth between 4.5 and 12 cm; some roofs were fertilized once a year in spring with an organic controlled release fertilizer (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Five roofs were planted with only species of \u003cem\u003eSedum\u003c/em\u003e, referred to as \u0026ldquo;\u003cem\u003eSedum\u003c/em\u003e-only roofs\u0026rdquo;, and seven roofs were planted with species of \u003cem\u003eSedum\u003c/em\u003e and a mix of annual herb species, named \u0026ldquo;diverse-vegetation roofs\u0026rdquo; (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e1\u003c/span\u003e). On average for diverse-vegetation roofs 31% of the surface area was covered with \u003cem\u003eSedum\u003c/em\u003e, 20% with herbs and 31% with mosses. The remaining percentage (18%) was bare substrate (based on picture analyses by eye, see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003e in supplementary material). The substrate of these extensive green roofs consisted mainly of different lava stone fractions, expanded clay, crushed bricks, pumice stone and a small portion of compost to provide organic matter. The exact initial substrate composition of the examined green roofs was not known.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCharacteristics of the studied extensive green roofs: roof ID, location, roof area, construction year, vegetation type, fertilization (Y\u0026thinsp;=\u0026thinsp;fertilized, N\u0026thinsp;=\u0026thinsp;not fertilized), average substrate depth in centimeters.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof ID\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLocation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eArea (m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eConstruction year\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eVegetation type\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eFertilization (Y/N)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eAverage substrate depth (cm)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGhent\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e7.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGhent\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e110\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2005\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eSedum-\u003c/em\u003eonly\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e6.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGhent\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e588\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2013\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eSedum\u003c/em\u003e-only\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGhent\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e8.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHasselt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e432\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e11.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHasselt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e108\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2012\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eSedum\u003c/em\u003e-only\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHasselt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e175\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2004\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e8.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHasselt\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e225\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e12.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAntwerp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e280\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2008\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eSedum\u003c/em\u003e-only\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e4.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAntwerp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e708\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eSedum\u003c/em\u003e-only\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e6.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAntwerp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e777\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e8.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAntwerp\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e312\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2015\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDiverse\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eY\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e8.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Sample collection and analyses\u003c/h2\u003e\u003cp\u003eAll green roofs were examined seasonally, i.e. in April (spring; 2019), July (summer; 2019), October (autumn; 2019), and January (winter; 2020). Every roof was divided into virtual plots of 1m\u003csup\u003e2\u003c/sup\u003e, per sampling date (summer, autumn, winter) four plots per roof were randomly chosen. In spring, only three plots per roof were selected and roof 11 was not accessible. In the center of every plot, a subplot of 25 cm \u0026times; 25 cm was defined. Pictures were taken to estimate the vegetation cover (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e3\u003c/span\u003e), the aboveground biomass was cut off and stored in plastic bags and the upper 6 cm of the substrate (or less in case the substrate was thinner) was collected. Samples were kept on ice and stored at 4\u0026deg;C until further analyses.\u003c/p\u003e\u003cp\u003eOne week after sampling, roots were carefully picked out of the substrate with a forceps and were rinsed with demineralized water. Roots as well as the aboveground biomass were oven-dried at 70\u0026deg;C for seven days after which the dry weight was measured. After weighing, the aboveground biomass was ground to fine powder with a mixer mill MM 200 (Retsch GmbH, Haan, Germany) and analyzed for total carbon (TC) and total nitrogen (TN) by dry combustion, based on the Dumas-method using an elemental analyzer (Flash 2000 CN Soil Analyser, Interscience, Louvain-la-Neuve, Belgium).\u003c/p\u003e\u003cp\u003eSubstrate samples, which were stored at 4\u0026deg;C, were processed 7 days after sampling. They were thoroughly mixed manually and divided into three subsamples for further analyses.\u003c/p\u003e\u003cp\u003eSince it was practically not possible to take intact soil cores and the first 6 cm of the substrate was structurally homogeneous, one fresh subsample was used to estimate bulk density by drying a known volume of soil at 70\u0026deg;C for 7 days and weighing it afterwards (Blake, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e1965\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAnother fresh subsample was passed through a 4-mm sieve and stored another 7 days (14 days after sampling) at 4\u0026deg;C. This sample was used to measure the net N-mineralization and net nitrification rate with an aerobic incubation method (Hart, Start, Davidson, \u0026amp; Firestone, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1994\u003c/span\u003e). The water holding capacity (WHC) and gravimetric water content were measured (oven-drying at 70\u0026deg;C for 7 days)) before incubation to adjust all samples to 60% WHC during incubation. For 28 days, 15 g soil was incubated at a constant temperature of 20\u0026deg;C in the dark and was kept at 60% WHC by weekly addition of distilled water (Hart et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1994\u003c/span\u003e). Before and after incubation, N-NO\u003csub\u003e3\u003c/sub\u003e and N-NH\u003csub\u003e4\u003c/sub\u003e were extracted with 1M KCl (1:5, w:v) (Allen, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e1989\u003c/span\u003e) and analyzed colorimetrically with a continuous flow analyzer (AutoAnalyser 3, Bran\u0026thinsp;+\u0026thinsp;Leubbe, Germany). Net N-mineralization and nitrification rate were calculated as the net increase in inorganic N (N-NO\u003csub\u003e3\u003c/sub\u003e\u0026thinsp;+\u0026thinsp;N-NH\u003csub\u003e4\u003c/sub\u003e) and nitrate (N-NO\u003csub\u003e3\u003c/sub\u003e) during the 28-day incubation period.\u003c/p\u003e\u003cp\u003eThe third fresh subsample was passed through a 2-mm sieve and oven-dried at 70\u0026deg;C for 7 days. After drying, 10 g of soil was used to measure pH in deionized water (1:5, w:v) and 1M KCl (1:5, m:v) (Houba, van de Lee, Novazamsky, \u0026amp; Walinga, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1989\u003c/span\u003e)). The rest of the dried soil was ground to pass a 0.25-mm sieve in an ultra-centrifugal mill (Model ZM 200, Retsch GmbH, Haan, Germany) and analyzed for TC and TN. Total phosphorus (TP) was measured with a continuous flow analyzer (San\u0026thinsp;+\u0026thinsp;+\u0026thinsp;Skalar) after digestion with H\u003csub\u003e2\u003c/sub\u003eSO\u003csub\u003e4\u003c/sub\u003e-salicylic acid-H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e and selenium (Novozamsky, Houba, van Eck, \u0026amp; van Vark, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1983\u003c/span\u003e). TC, TN, TP per m\u003csup\u003e2\u003c/sup\u003e were calculated by multiplying the raw values from the analyses (percentages for TC and TN and mg kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for TP) with the substrate depth and bulk density.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Statistical analyses","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Carbon sequestration\u003c/h2\u003e\u003cp\u003eWe investigated seasonal changes in substrate C stocks and used this as a proxy for C sequestration. The effect of season on substrate and aboveground biomass TC stocks was investigated through linear mixed models (LMM) (lme4 package; Bates et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) with season as a fixed factor and roof ID (1\u0026ndash;12) as a random factor to account for repeated measures on the same roof. Assumptions of heteroscedasticity and normality of residuals were tested with Levene\u0026rsquo;s test (car package, Fox and Weisberg, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) and Shapiro-Wilk\u0026rsquo;s test, respectively. If assumptions were not met the dependent variable was log-transformed. Outliers were detected and removed by using \u0026ldquo;outlierstest\u0026rdquo; (car package; Fox and Weisberg, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) In case of a significant effect, post-hoc tests were caried out with Tukey HSD (multcomp package; Hothorn, Bretz and Westfall, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Effect of green roof characteristics on TC, TN, TP substrate stocks\u003c/h2\u003e\u003cp\u003eTC, TN, and TP substrate stocks were used as dependent variables. For each dependent variable, two LMMs (lme4 package; Bates et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) where made to avoid multicollinearity (VIF (Variance Inflation Factor): car package, Fox and Weisberg, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) between vegetation type (\u003cem\u003eSedum\u003c/em\u003e-only vs. diverse vegetation) and percentage of plant cover. In the first model, vegetation type was the fixed factor and in the second model, fertilization, roof age, substrate depth, aboveground biomass, belowground biomass, succulent coverage, herb coverage and moss coverage were fixed factors. To account for correlations within one roof and through seasons, roof ID and season were added as crossed random factors. Assumptions of normality and heteroscedasticity were tested, and outliers were detected and removed as described above. Dependent variables were log-transformed if assumption were not met. The second model was also checked for multicollinearity with VIF (car package, (Fox \u0026amp; Weisberg, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Preliminary analyses showed that roof 11 deviated strongly from all other roofs, therefore the analyses were done on the dataset without roof 11. Furthermore, to check whether the significant effect of vegetation type on TC, TN and TP substrate stocks was not correlated with fertilization, the same models were made with only fertilized roofs in the dataset (4 \u003cem\u003eSedum\u003c/em\u003e-only roofs, 4 diverse vegetation roofs).\u003c/p\u003e\u003cp\u003eAdditionally, to determine if significant differences in TC substrate between vegetation types could be caused by a difference in biomass or by a difference in C content between perennial succulents and annual herb species, three LMMs (lme4 package; Bates et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) were set up with aboveground biomass, belowground biomass and TC of the aboveground biomass as dependent variable and vegetation type as fixed factor. Again, with roof ID and season as crossed random factors. Model assumptions were tested as described above.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Effect of fertilization on green roof vegetation\u003c/h2\u003e\u003cp\u003eTo assess the effect of fertilization on total plant coverage, succulent coverage, herb coverage, moss coverage, belowground and aboveground biomass, LMMs (lme4 package; Bates et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) were created, with season and roof ID as crossed random factors. For total plant, succulent, herb and moss coverage a generalized linear mixed model (GLMM) (lme4 package; Bates et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) was specified with a binomial distribution. Overdispersion was tested by giving all samples an ID number and comparing AIC values of the models with and without ID number. No model turned out to be overdispersed. Other assumptions (heteroscedasticity, normality) were tested as described above, and dependent variables were log-transformed when necessary.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Total carbon stocks\u003c/h2\u003e\u003cp\u003eThe investigated roofs showed in general rather low TC stocks. The total carbon (TC) stock of the aboveground biomass was significantly different between seasons (\u003cem\u003ep\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;0.001, Type II Wald Chi squared test). A Tukey post-hoc test showed that the TC stocks in aboveground biomass are higher in summer than in spring (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and winter (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.004) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). Plants (aboveground biomass) accumulated TC between spring and summer with an increase from 49 g TC m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e to 114 g TC m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e. After summer, TC stock in plants decreased, with average values of 67 g TC m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e and 62 g TC m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e, in autumn and winter, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). Belowground biomass was very low, with an average maximum of 31 g dry weight m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e in autumn (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e in supplementary material) The TC stock in the substrate of the measured green roofs declined over the period of one year (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003ea) with significant differences between spring-winter (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.011) and summer-winter (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\u003cp\u003e\u003cb\u003e4.2 Difference in substrate TC, TN and TP stocks and net N mineralization between\u003c/b\u003e \u003cb\u003eSedum\u003c/b\u003e\u003cb\u003e-only and diverse vegetation roofs\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe average TC substrate stock of the twelve investigated green roof ranged between 700\u0026ndash;2900 g C m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). \u003cem\u003eSedum\u003c/em\u003e-only roofs contained significantly (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) less substrate TC compared to diverse-vegetation roofs, ranging from 700\u0026ndash;1000 g TC m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e and 1400\u0026ndash;2900 g TC m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Roof 12 showed a relatively low TC content for a diverse-vegetation roof. The significant difference of substrate TC between the two vegetation types persisted when only fertilized roofs (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.005) (n\u0026thinsp;=\u0026thinsp;8; 4 \u003cem\u003eSedum\u003c/em\u003e-only roofs, 4 diverse-vegetation roofs) were considered (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e in supplementary material). Linear mixed models with aboveground biomass (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.117), belowground biomass (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.920) and TC of the aboveground biomass (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.067) as dependent variables and vegetation type as independent variable, showed no significant effects (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003e in supplementary material).\u003c/p\u003e\u003cp\u003eThe average TN and TP content of the substrates were 40\u0026ndash;100 g TN m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e and 40\u0026ndash;100 g TP m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Similar to the TC values, \u003cem\u003eSedum\u003c/em\u003e-only roofs had significantly less TN (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) in the substrate compared to diverse-vegetation roofs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003eb) and, although not significant, their substrate TP stock (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.360) was also lower than in diverse-vegetation roofs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). When only fertilized roofs were included in the model, TN (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.025) remained significantly higher in diverse-vegetation roofs, while the difference in TP remained statistically non-significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.114) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e in supplementary material). Furthermore, the N-mineralization rate was highly variable among and within roofs, with values ranging from 0.16\u0026ndash;1.44 mg N kg soil\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e day\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Similarly, the net nitrification varied strongly among and within roofs, with values of 0.18\u0026ndash;1.45 mg N kg soil\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e day\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eAverage values and standard deviation of substrate TC, TN, TP, N-mineralization, net nitrification and pH per roof across all seasons (n\u0026thinsp;=\u0026thinsp;16: values\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;S.D.). Values of roofs with diverse vegetation are in bold.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTC (\u003cem\u003eg m\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;2\u003c/em\u003e\u003c/sup\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTN (\u003cem\u003eg m\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;2\u003c/em\u003e\u003c/sup\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTP (\u003cem\u003eg m\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;2\u003c/em\u003e\u003c/sup\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN \u0026ndash; mineralization (mg N kg\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;1\u003c/em\u003e\u003c/sup\u003e soil day\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;1\u003c/em\u003e\u003c/sup\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNet nitrification\u003c/p\u003e\u003cp\u003e(mg N kg\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;1\u003c/em\u003e\u003c/sup\u003e soil day\u003csup\u003e\u003cem\u003e\u0026minus;\u0026thinsp;1\u003c/em\u003e\u003c/sup\u003e)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003epH\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 1\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1448\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e641\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e90\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e34\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e40\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e8\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.22\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.25\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.21\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.24\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e6.22\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.59\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e924\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;779\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e58\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e70\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e0.28\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e0.29\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e6.04\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.11\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e743\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;164\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e48\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e49\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e0.57\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e0.57\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e6.02\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1382\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e559\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e72\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e27\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e68\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e17\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.16\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.29\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.18\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.28\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e6.06\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.15\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 5\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e2451\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e496\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e85\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e19\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e124\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e19\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.71\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.64\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.72\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.67\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e6.5\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.18\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof 6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e862\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;507\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e54\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e74\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e1.2\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e1.45\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e5.61\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.25\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 7\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1922\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e807\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e103\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e46\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e78\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e10\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.6\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.39\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.62\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.39\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e5.78\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.34\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 8\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e2883\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e293\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e90\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e10\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e137\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e12\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.18\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.14\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.19\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.16\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e6.69\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.08\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof 9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e1047\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;206\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e38\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e80\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e0.48\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e0.51\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e5.81\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof 10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e714\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;306\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e46\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e48\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e1.44\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e1.53\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e5.64\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;0.31\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 11\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e5142\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e527\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e167\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e29\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e147\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e21\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e2.06\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e1.57\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e2.11\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e1.56\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e5.9\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.19\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRoof 12\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e833\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e255\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e62\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e17\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e43\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e10\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.61\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.31\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003e0.64\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.28\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e6.56\u003c/b\u003e\u0026thinsp;\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;\u003cb\u003e0.29\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003e4.3 The effect of various green roof characteristics on TC, TN, TP stocks and green roof vegetation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThere was no significant relationship between substrate TC and the percentage of herb cover (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.676; Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e3\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). The percentage moss cover, however, had a significant positive effect on substrate TC (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.016) and TN (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.007). There was no significant increase in substrate TC with roof age (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.616). Furthermore, there was a significant influence of the aboveground biomass on TN stocks (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.019), with more TN with increasing biomass. The TP stock in the upper 6 cm of the substrate was significantly higher in roofs with greater substrate depths (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.012). Fertilization had no significant effect on TC, TN and TP substrate stocks (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e3\u003c/span\u003e), nor on vegetation cover or biomass (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eOverview of Chi-square distribution values, p-values of LMMs with substrate TC, TN, TP as dependent variables and roof age in 2019 (years), roof depth (cm), aboveground biomass (g m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e), belowground biomass (g m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e), Sedum coverage (%), herb coverage (%), moss coverage (%) as fixed factors (roof 11 excluded).\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDependent variable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFixed factor\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eChi-squared\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMoss coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5.803\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.016\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAboveground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.006\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.083\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFertilization\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.723\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.189\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoverage of succulents\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.401\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.527\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoof depth\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.253\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.615\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoof age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.448\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.616\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHerb coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.174\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.676\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBelowground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.019\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.890\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMoss coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.396\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.007\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAboveground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5.474\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.019\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoverage of succulents\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.426\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.119\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFertilization\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.624\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.203\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHerb coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.206\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.272\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBelowground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.102\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.294\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoof age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6.883\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.332\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoof depth\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.138\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.710\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoof depth\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6.304\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.012\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAboveground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.755\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.052\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHerb coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.938\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.164\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoverage of succulents\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.422\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.233\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBelowground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.631\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.427\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFertilization\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.626\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.429\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMoss coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.078\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.780\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRoof age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.154\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e0.905\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eChi-squared and p-values from (G)LMM-models with \u0026lsquo;fertilization\u0026rsquo; as fixed factor\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDependent variable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChi-squared\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal plant coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.059\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.808\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAboveground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.280\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.597\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBelowground biomass\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.775\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.379\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCoverage of succulents\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.250\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.617\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHerb coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.034\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.854\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMoss coverage\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.892\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.345\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e5.1 Total carbon stocks and carbon storage potential\u003c/h2\u003e\u003cp\u003eWe measured TC values between 0.8\u0026ndash;3 kg C m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e in the green roof substrate. These values are comparable with other studies, in which variation in concentration is explained by green roof properties, such as substrate type, substrate depth, and vegetation composition (Bouzouidja et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Getter et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Whittinghill et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Compared to natural ecosystems (e.g. shrublands, grasslands, forests), where TC stocks are typically an order of magnitude higher (Smith et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Whittinghill et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), green roofs contain relatively low TC stocks.\u003c/p\u003e\u003cp\u003eSubstrate TC stocks showed a decreasing trend from spring to winter, with significantly lower values in winter compared to summer and spring, suggesting a net loss of substrate TC. Losses of plant TC via litter and root exudates, have earlier been shown to become fixed in the green roof substrate (Kuronuma et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In this study, however, we did not observe an increase in substrate TC in autumn when leaf litter is expected to be highest. The green roof substrate was the biggest pool of carbon, with 20\u0026ndash;30 times more TC per square meter compared to the vegetation, making the potential annual input of plant TC difficult to detect.\u003c/p\u003e\u003cp\u003eFurthermore, we observed no significant effect of green roof age on green roof TC substrate stocks, indicating extensive green roofs have a small potential to sequester C, even at the decadal time scale. Literature shows in general that green roofs can sequester C, however, it is often mentioned that \u003cem\u003eSedum\u003c/em\u003e-covered green roofs have a minor C sequestration potential in comparison with intensive green roofs or other in ground-systems (Bouzouidja et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Getter et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Kuronuma et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Li \u0026amp; Babcock, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Shafique et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Whittinghill et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e\u003cp\u003ePlant health and growth affected by a lot of parameters, such as soil temperature, soil moisture and soil nitrogen content (Mikan et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Sinsabaugh et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Xu, Baldocchi and Tang, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Gunti\u0026ntilde;as et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Moreover, green roofs are artificial environments, which are often exposed to extreme conditions, such as extreme temperatures and periods of drought (Oberndorfer et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) and in addition surrounding building structures influence shading, sun, wind, and rain exposure (Mitchell et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Regan et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). All these factors can influence plant growth and thereby TC storage capacity. This makes every green roof unique and possibly determines their capacity to store TC.\u003c/p\u003e\u003cp\u003eIn conclusion, our findings suggest that extensive green roofs have a limited C storage potential. To optimize this minimal C storage potential, it is important to optimize vegetation growth and engineer the substrate to facilitate efficient C storage. In this optimization process it is essential to consider the various factors that determine the local environment of a specific green roof.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e5.2 Effect of vegetation types\u003c/h2\u003e\u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\u003ch2\u003e5.2.1 Carbon\u003c/h2\u003e\u003cp\u003eWe tested the hypothesis that green roofs with a more diverse vegetation have higher substrate TC stocks compared to \u003cem\u003eSedum\u003c/em\u003e-only roofs. We observed that the substrate of diverse-vegetation roofs contained twice as much TC compared to \u003cem\u003eSedum\u003c/em\u003e-only roofs, albeit with a high variability in the results. It is plausible that higher TC stocks in diverse vegetation resulted from higher biomass production, due to a higher productivity of annual herbs species compared to succulents (Agra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Oberndorfer et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). There was, however, no significant difference in aboveground (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.490\u003c/em\u003e) and belowground biomass (\u003cem\u003ep\u0026thinsp;=\u0026thinsp;0.492\u003c/em\u003e) between the two types of vegetation, suggesting that there was not more production of biomass on diverse-vegetation roofs. It is fact that the higher turnover rate of annual herbs compared to perennial succulents is expected to yield higher productivity per unit biomass (Agra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Oberndorfer et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). However, this is not supported by our data that indicate that the contribution of annual biomass production as a source of C input in the substrate of extensive green roofs is negligible and could not have resulted in a doubling of the total substrate C. We therefore speculate that difference in the initial substrate composition possibly caused this difference substrate C, as the substrate of diverse-vegetation roofs has in general a higher organic matter content to promote plant growth compared to \u003cem\u003eSedum-\u003c/em\u003eonly roofs. This is in line with the findings of Shafique et al. (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), who concluded that the substrate composition is vital for C capture and storage in green roof substrates. To clarify this, future research should focus on a more long-term follow-up of newly installed green roofs with these two vegetation types.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section3\"\u003e\u003ch2\u003e5.2.2. Nitrogen and phosphorus\u003c/h2\u003e\u003cp\u003eIn addition to the effect of vegetation on the TC substrate stocks, we also investigated the hypothesis that diverse-vegetation roofs will have higher TN and TP substrate stocks and higher N-mineralization and nitrification rates. We observed that the TN concentrations in the substrate of diverse vegetation roofs are two times higher when compared with the substrate of \u003cem\u003eSedum\u003c/em\u003e-only roofs. Johnson et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e showed that a higher plant species richness enhances TN retention in green roof substrates. The green roofs with diverse vegetation in our study did not have a higher species richness compared to \u003cem\u003eSedum\u003c/em\u003e-only roofs (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e in supplementary material). Consequently, our findings do not align with those of Johnson et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e, who demonstrated that green roofs have the potential to retain nitrogen. It must be noticed that, just as for TC the initial substrate composition could also play a role in the amount of TN in the substrate. However, we also measured older green roofs (\u0026gt;\u0026thinsp;4 years old), of which is hypothesized that TN present in the initial substrate would already be leached or consumed.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eNumber of plant species that were identified during four monitoring periods (May, June, August and September 2019)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"2\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRoof ID\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNumber of plant species\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e39\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e22\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e22\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNA\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNA\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e33\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e17\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFurthermore, we observed that N-mineralization and nitrification rates are highly variable within and among roofs and could not be explained by vegetation type. To our knowledge there are no studies that measured actual N-mineralization and nitrification on green roofs. We suggest that other parameters causing spatial heterogeneity on green roofs, such as difference in sun and wind exposure and moisture content but also the small-scale spatial heterogeneity of the substrate itself, could be steering this high variability (Regan et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Furthermore, monitoring N-fluxes in combination with the assessment of microbial and plant communities as well as other abiotic factors such as pH, bulk density, water holding capacity, \u0026hellip;, seems necessary to unravel the N-fluxes in green roofs in more detail.\u003c/p\u003e\u003cp\u003eIn contrast to the TN substrate stocks, the TP substrate stocks did not significantly differ between diverse-vegetation roofs and \u003cem\u003eSedum\u003c/em\u003e-only roofs. Remarkably, TP substrate stocks are as high as TN substrate stocks, which indicates that the substrate is extremely P-rich. Research on TP stocks of green roofs is limited and mainly focuses on PO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e3\u0026minus;\u003c/sup\u003e leaching, showing that green roofs are often sources of phosphates (Karczmarczyk et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Again, initial substrate composition could be the main determining factor of TP concentrations in the substrate, where TP concentrations are not only determined by the available PO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e3\u0026minus;\u003c/sup\u003e but also by mineral bound phosphorus, which is likely very high and similar in the measured green roof substrates.\u003c/p\u003e\u003cp\u003e\u003cb\u003e5.3 The effect of other green roof characteristics on TC, TN, TP substrate stocks and green roof vegetation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBesides the initial vegetation type, other green roofs parameters such as plant cover, plant biomass, substrate depth, roof age and maintenance practices can affect the TC, TN and TP retention capacity and the development of green roof vegetation (Bouzouidja et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Clark \u0026amp; Zheng, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Dusza et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Hoch et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). We hypothesized that fertilization would be the main determinant of TC, TN and TP substrate stocks.\u003c/p\u003e\u003cp\u003eOur results show moss coverage has a significant effect on the TC and TN substrate stocks. The investigated green roofs had an average moss cover of 31%. Mosses have a lower decomposition rate compared to vascular plants, possibly causing a more persistent litter layer which increases TC of the soil (Kasimir, He, Jansson, Lohila, \u0026amp; Minkkinen, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Furthermore, mosses have high N retention capacities (Ayres, Van Der Wal, Sommerkorn, \u0026amp; Bardgett, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2006\u003c/span\u003e), which suggest that mosses would retain more N and not release it in the substrate. However, the harsh conditions on green roofs, such as drought and heat, could cause a higher litter input from mosses than in other environments, which would increase substrate TN. Although mosses are often esthetically not appreciated, they might be beneficial in terms of substrate TC and TN stocks.\u003c/p\u003e\u003cp\u003eIn addition, aboveground biomass showed a significant influence on TN substrate stocks, with more TN in the substrate with increasing biomass. We expected however to see the same effect for TC, as plant litter is a source of TN and TC. Nevertheless, we assume plant litter input on green roofs is minimal due to the low resilient vegetation. Therefore, this effect is likely to a higher nitrogen retention capacity, both by plants and soil particles, whereas this is less the case for carbon (Johnson et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFor TP substrate stock there only was a significant positive effect of substrate depth. In this study only the upper 6cm of all green roofs were sampled. Hereby the positive effect of substrate depth indicates that deeper roofs had more TP in the upper 6cm of the substrate. From this we conclude that a deeper green roof substrate promotes the build-up of P substrate stocks. A high TP substrate stock can imply that the organic P is not mineralized by present micro-organisms, thereby causing a phosphorus pool with low plant-availability. Furthermore, inorganic P could possibly attach to the mineral fraction of the green roof substrate, such as lava stones and expanded clay, contributing to a higher TP substrate stock (Karczmarczyk et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). It is however unclear, how these explanations are in relation with substrate depth. Another likely explanation is that P present in the soil solution is prone to leaching (Mitchell, Matter, Durtsche, \u0026amp; Buffam, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), whereby thinner substrates are more quickly saturated (van Seters, Rocha, Smith, \u0026amp; MacMillan, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) and will lose more P.\u003c/p\u003e\u003cp\u003eThere was no effect of herb of \u003cem\u003eSedum\u003c/em\u003e coverage and the TC, TN or TP substrate stocks, although plant density or coverage is a main determinant of carbon and nutrient cycling according to the study of Buffam and Mitchell, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2015\u003c/span\u003e. These results, together with our result that diverse vegetation roofs have higher TC and TN substrate stocks compared to \u003cem\u003eSedum\u003c/em\u003e-only roofs, suggest that the presence or absence of annual herb species is more important than the actual percentage herb coverage.\u003c/p\u003e\u003cp\u003eFurther, we expected that green roofs will build up some TC in the substrate over their lifetime and that N and P substrate stocks will not be related with roof age, given their high susceptibility to leaching. Our results show no relation between roof age and TC, TN or TP substrate stocks. The age of the investigated roofs varied between 4 and 15 years during the sampling period.\u003c/p\u003e\u003cp\u003eOther studies show that green roofs can build up TC and TN in their substrate (K\u0026ouml;hler \u0026amp; Poll, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Mitchell et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Schrader \u0026amp; B\u0026ouml;ning, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Comparing green roof age with TC, TN and TP substrate stocks is however far from straightforward. Besides green roof age, there are numerous green roof characteristics to consider, which influence the retention capacity and leaching susceptibility of the green roof substrate. A study that monitors green roofs throughout their lifespan and systematically monitors all green roof characteristics would be valuable in gaining further insight into the carbon and nutrient dynamics of green roofs over time.\u003c/p\u003e\u003cp\u003eLast, there was no significant effect of fertilization on TP substrate stocks, nor on TN substrate stocks. This could indicate that excessive nutrient input via fertilizers is quickly lost by leaching (Clark \u0026amp; Zheng, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2013\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Moreover, our results showed no effect of fertilization on coverage or biomass, while one of the main reasons for the application of fertilizer is to increase the esthetical value of a green roof by enhancing plant cover and greenness. Clark et al., 2014, do show that fertilizer application helps to establish plant growth and roof coverage. This suggests that fertilization application is not useful once a green roof is already fully covered, which could be an indication of green roofs being in a steady-state after 4 years of age (Mitchell et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), with constant pools and fluxes.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eIn conclusion, we found that extensive green roofs covered with mainly succulent species have a low potential to sequester carbon, likely because of the low and slow growing vegetation and the porous and artificial substrate. We showed that adding annual herb species was, probably indirectly, beneficial for TC substrate stocks and the closely linked N substrate stocks. Moreover, the presence of mosses contributed most to higher TC and TN substrate stocks, while they are usually given a low esthetical value. Furthermore, our results indicate that fertilizer application is not useful once extensive green roofs reach a certain age, which could be very important when considering eutrophication in downstream waterways. Therefore, we presume that when extensive green roofs are given the chance to develop as natural pioneer ecosystems, they might reach a steady state, where maintenance and fertilization is not necessary. To promote a more natural development of extensive green roofs, locally adapted plant species could be integrated into green roof designs. Although the use of natural soil is often not feasible due to its weight, alternative substrates that mimic natural soil properties while remaining lightweight, such as biochar-based mixtures, could be explored. Considering that current artificial substrates like expanded clay and lava stones are derived from finite resources, such alternatives may offer a more sustainable solution. Moreover, using local vegetation in combination with more soil-like, yet lightweight substrates may enhance the ability of green roofs to deliver ecosystem services aligned with local climate conditions and biodiversity. Additionally, we propose that the initial substrate composition could be the main factor driving TC, TN and TP substrate stocks. Adding extra C to substrate during the installation could be a way to store external C, such as compost or biochar, since the carbon in the soils did not decline with age and is apparently stable. For this to succeed, it is imperative that the extra C is ecologically produced with minor CO\u003csub\u003e2\u003c/sub\u003e emissions. This would be an interesting path for future research. Furthermore, comprehensively mapping all possible fluxes and stocks, as well as the various factors having an impact on the fluxes and stocks, and monitoring these over an extended period, ideally even throughout the entire lifespan of a green roof, could clarify the many uncertainties that currently still exist.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThe author L.S. and the research were financed by Research Foundation - Flanders (FWO) via a Strategic Basic Research project (S002818N: \u0026ldquo;EcoCities: Green roofs and walls as a source for ecosystem services in future cities\u0026rdquo;).\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eL.S., F.R., I.J., E.S. and T.A contributed to the study conception and design. Samples were taken by L.S and T.V.D. Analyses were performed by L.S, B.B and M.P-E. All plant species were identified, and coverage percentages were estimated by C.V.M. Statistical analyses were carried out by L.S. The first draft of the paper was written by L.S. and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThis study was supported by the SBO project S002818N \u0026quot;EcoCities: Green roofs and walls as a source for ecosystem services in future cities\u0026rdquo;.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe research data is available upon request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAgra, H., Klein, T., Vasl, A., Kadas, G., \u0026amp; Blaustein, L. (2017). Measuring the effect of plant-community composition on carbon fixation on green roofs. \u003cem\u003eUrban Forestry and Urban Greening\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e, 1\u0026ndash;4. https://doi.org/10.1016/j.ufug.2017.03.003\u003c/li\u003e\n\u003cli\u003eAitkenhead-Peterson, J. A., Dvorak, B. D., Volder, A., \u0026amp; Stanley, N. C. (2011). Chemistry of growth medium and leachate from green roof systems in south-central Texas. \u003cem\u003eUrban Ecosystems\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 17\u0026ndash;33. https://doi.org/10.1007/s11252-010-0137-4\u003c/li\u003e\n\u003cli\u003eAllen, S. E. (1989). \u003cem\u003eChemical analysis of Ecological Materials\u003c/em\u003e (2nd editio). Oxford, London: Blackwell Scientific Publications.\u003c/li\u003e\n\u003cli\u003eAmpim, P. A. Y., Sloan, J. J., Cabrera, R. I., Harp, D. A., \u0026amp; Jaber, F. H. (2010). Green Roof Growing Substrates: Types, Ingredients, Composition and Properties. \u003cem\u003eJournal of Environmental Horticulture\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(4), 244\u0026ndash;252. https://doi.org/10.24266/0738-2898-28.4.244\u003c/li\u003e\n\u003cli\u003eAyres, E., Van Der Wal, R., Sommerkorn, M., \u0026amp; Bardgett, R. D. (2006). Direct uptake of soil nitrogen by mosses. \u003cem\u003eBiology Letters\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(2), 286\u0026ndash;288. https://doi.org/10.1098/rsbl.2006.0455\u003c/li\u003e\n\u003cli\u003eBates, D., Maechler, M., Bolker, B., \u0026amp; Walker, S. (2015). Fitting Linear Mixed-Effects Models Using lme4. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e, \u003cem\u003e67\u003c/em\u003e(1), 1\u0026ndash;48. https://doi.org/10.18637/jss.v067.i01\u003c/li\u003e\n\u003cli\u003eBerndtsson, J. C., Emilsson, T., \u0026amp; Bengtsson, L. (2006). The influence of extensive vegetated roofs on runoff water quality. \u003cem\u003eScience of the Total Environment\u003c/em\u003e, \u003cem\u003e355\u003c/em\u003e(1\u0026ndash;3), 48\u0026ndash;63. https://doi.org/10.1016/j.scitotenv.2005.02.035\u003c/li\u003e\n\u003cli\u003eBlake, G. R. (1965). Bulk density. In \u003cem\u003eMethods of Soil Analysis, Part 1: Physical and Mineralogical Properties, Including Statistics of Measurement and Sampling\u003c/em\u003e (9th ed.). https://doi.org/10.2134/agronmonogr9.1.c30\u003c/li\u003e\n\u003cli\u003eBouzouidja, R., Rousseau, G., Galzin, V., Claverie, R., Lacroix, D., \u0026amp; S\u0026eacute;r\u0026eacute;, G. (2018). Green roof ageing or Isolatic Technosol\u0026rsquo;s pedogenesis? \u003cem\u003eJournal of Soils and Sediments\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(2), 418\u0026ndash;425. https://doi.org/10.1007/s11368-016-1513-3\u003c/li\u003e\n\u003cli\u003eBuffam, I., \u0026amp; Mitchell, M. E. (2015). Nutrient cycling in green roof ecosystems. In \u003cem\u003eGreen roof ecosystems\u003c/em\u003e (pp. 107\u0026ndash;137). https://doi.org/10.1007/978-3-319-14983-7\u003c/li\u003e\n\u003cli\u003eClark, M. J., \u0026amp; Zheng, Y. (2013). Plant nutrition requirements for an installed sedum-vegetated green roof module system: Effects of fertilizer rate and type on plant growth and leachate nutrient content. \u003cem\u003eHortScience\u003c/em\u003e, \u003cem\u003e48\u003c/em\u003e(9), 1173\u0026ndash;1180.\u003c/li\u003e\n\u003cli\u003eClark, M. J., \u0026amp; Zheng, Y. (2014). Effect of fertilizer rate on plant growth and leachate nutrient content during production of sedum-vegetated green roof modules. \u003cem\u003eHortScience\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e(6), 819\u0026ndash;826.\u003c/li\u003e\n\u003cli\u003eDe Neve, S. (2017). Organic matter mineralisation as a source of nitrogen. In F. Tei (Ed.), \u003cem\u003eAdvances in Research on Fertilization Managment of Vegetable Crops\u003c/em\u003e (pp. 65\u0026ndash;83). https://doi.org/10.1007/978-3-319-53626-2\u003c/li\u003e\n\u003cli\u003eDusza, Y., Barot, S., Kraepiel, Y., Lata, J. C., Abbadie, L., \u0026amp; Raynaud, X. (2017). Multifunctionality is affected by interactions between green roof plant species, substrate depth, and substrate type. \u003cem\u003eEcology and Evolution\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(7), 2357\u0026ndash;2369. https://doi.org/10.1002/ece3.2691\u003c/li\u003e\n\u003cli\u003eEmilsson, T., Czemiel Berndtsson, J., Mattsson, J. E., \u0026amp; Rolf, K. (2007). Effect of using conventional and controlled release fertiliser on nutrient runoff from various vegetated roof systems. \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(3), 260\u0026ndash;271. https://doi.org/10.1016/j.ecoleng.2006.01.001\u003c/li\u003e\n\u003cli\u003eFox, J., \u0026amp; Weisberg, S. (2019). \u003cem\u003eAn {R} Companion to Applied Regression\u003c/em\u003e (p. Third Edition). p. Third Edition. Retrieved from https://socialsciences.mcmaster.ca/jfox/Books/Companion/\u003c/li\u003e\n\u003cli\u003eGetter, K. L., Rowe, D. B., Robertson, G. P., Cregg, B. M., \u0026amp; Andresen, J. A. (2009). Carbon Sequestration Potential of Extensive Green Roofs. \u003cem\u003eEnvironmental Science \u0026amp; Technology\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(19), 7564\u0026ndash;7570. https://doi.org/10.1021/es901539x\u003c/li\u003e\n\u003cli\u003eGunti\u0026ntilde;as, M. E., Gil-Sotres, F., Leir\u0026oacute;s, M. C., \u0026amp; Trasar-Cepeda, C. (2013). Sensitivity of soil respiration to moisture and temperature. \u003cem\u003eJournal of Soil Science and Plant Nutrition\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(2), 445\u0026ndash;461. https://doi.org/10.4067/S0718-95162013005000035\u003c/li\u003e\n\u003cli\u003eHart, S. C., Start, J. M., Davidson, E. A., \u0026amp; Firestone, M. K. (1994). Nitrogen mineralization, immobilization, and nitrification. In R. Weaver, J. Angle, P. Bottomley, D. Bezdicek, S. Smith, \u0026amp; A. Tabatabai (Eds.), \u003cem\u003eMethods of Soil Analysis. Part 2. Microbial and Biochemical Properties.\u003c/em\u003e (pp. 985\u0026ndash;1018). https://doi.org/10.2136/sssabookser5.2.c42\u003c/li\u003e\n\u003cli\u003eHoch, J. M. K., Rhodes, M. E., Shek, K. L., Dinwiddie, D., Hiebert, T. C., Gill, A. S., \u0026hellip; McGuire, K. L. (2019). Soil microbial assemblages are linked to plant community composition and contribute to ecosystem services on urban green roofs. \u003cem\u003eFrontiers in Ecology and Evolution\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(JUN). https://doi.org/10.3389/fevo.2019.00198\u003c/li\u003e\n\u003cli\u003eHothorn, T., Bretz, F., \u0026amp; Westfall, P. (2008). Simultaneous Inference in General Parametric Models. \u003cem\u003eBiometrical Journal\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(3), 346\u0026ndash;363.\u003c/li\u003e\n\u003cli\u003eHouba, V. J. G., van de Lee, J. J., Novazamsky, I., \u0026amp; Walinga, I. (1989). \u003cem\u003eSoil and plant analysis, a series of syllabi. Part 5: Soil Analysis Procedures\u003c/em\u003e. Wageningen: Agricultural University, Wageningen.\u003c/li\u003e\n\u003cli\u003eJamei, E., Chau, H. W., Seyedmahmoudian, M., \u0026amp; Stojcevski, A. (2021). Review on the cooling potential of green roofs in different climates. \u003cem\u003eScience of the Total Environment\u003c/em\u003e, \u003cem\u003e791\u003c/em\u003e, 148407. https://doi.org/10.1016/j.scitotenv.2021.148407\u003c/li\u003e\n\u003cli\u003eJohnson, C., Schweinhart, S., \u0026amp; Buffam, I. (2016). Plant species richness enhances nitrogen retention in green roof plots. \u003cem\u003eEcological Applications\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(7), 2130\u0026ndash;2144. https://doi.org/10.1890/15-1850.1\u003c/li\u003e\n\u003cli\u003eKarczmarczyk, A., Bus, A., \u0026amp; Baryla, A. (2018). Phosphate leaching from green roof substrates-Can green roofs pollute urban water bodies? \u003cem\u003eWater\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e. https://doi.org/10.3390/w10020199\u003c/li\u003e\n\u003cli\u003eKasimir, He, H., Jansson, P. E., Lohila, A., \u0026amp; Minkkinen, K. (2021). Mosses are Important for Soil Carbon Sequestration in Forested Peatlands. \u003cem\u003eFrontiers in Environmental Science\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(September), 1\u0026ndash;19. https://doi.org/10.3389/fenvs.2021.680430\u003c/li\u003e\n\u003cli\u003eK\u0026ouml;hler, M., \u0026amp; Poll, P. (2010). Long-term performance of selected old Berlin green roofs in comparison to younger extensive green roofs in Berlin. \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e, 722\u0026ndash;729.\u003c/li\u003e\n\u003cli\u003eKuronuma, T., Watanabe, H., Ishihara, T., Kou, D., Toushima, K., Ando, M., \u0026amp; Shindo, S. (2018). CO2 Payoffof extensive green roofs with different vegetation species. \u003cem\u003eSustainability (Switzerland)\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(7), 1\u0026ndash;12. https://doi.org/10.3390/su10072256\u003c/li\u003e\n\u003cli\u003eLi, Y., \u0026amp; Babcock, R. W. (2014). Green roofs against pollution and climate change. A review. \u003cem\u003eAgronomy for Sustainable Development\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(4), 695\u0026ndash;705. https://doi.org/10.1007/s13593-014-0230-9\u003c/li\u003e\n\u003cli\u003eManso, M., Teot\u0026oacute;nio, I., Silva, C. M., \u0026amp; Cruz, C. O. (2021). Green roof and green wall benefits and costs: A review of the quantitative evidence. \u003cem\u003eRenewable and Sustainable Energy Reviews\u003c/em\u003e, \u003cem\u003e135\u003c/em\u003e(July 2020). https://doi.org/10.1016/j.rser.2020.110111\u003c/li\u003e\n\u003cli\u003eMikan, C. J., Schimel, J. P., \u0026amp; Doyle, A. P. (2002). Temperature controls of microbial respiration in arctic tundra soils above and below freezing. \u003cem\u003eSoil Biology and Biochemistry\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(11), 1785\u0026ndash;1795. https://doi.org/10.1016/S0038-0717(02)00168-2\u003c/li\u003e\n\u003cli\u003eMitchell, M. E., Emilsson, T., \u0026amp; Buffam, I. (2021). Carbon, nitrogen, and phosphorus variation along a green roof chronosequence: Implications for green roof ecosystem development. \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e164\u003c/em\u003e(March), 106211. https://doi.org/10.1016/j.ecoleng.2021.106211\u003c/li\u003e\n\u003cli\u003eMitchell, M. E., Hamilton, T. L., Uebel-Niemeier, C., Hopfensperger, K. N., \u0026amp; Buffam, I. (2018). Nitrogen cycling players and processes in green roof ecosystems. \u003cem\u003eApplied Soil Ecology\u003c/em\u003e, \u003cem\u003e132\u003c/em\u003e(June), 114\u0026ndash;125. https://doi.org/10.1016/j.apsoil.2018.08.007\u003c/li\u003e\n\u003cli\u003eMitchell, M. E., Matter, S. F., Durtsche, R. D., \u0026amp; Buffam, I. (2017). Elevated phosphorus: dynamics during four years of green roof development. \u003cem\u003eUrban Ecosystems\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(5), 1121\u0026ndash;1133. https://doi.org/10.1007/s11252-017-0664-3\u003c/li\u003e\n\u003cli\u003eNovozamsky, V. J. G., Houba, R., van Eck, R., \u0026amp; van Vark, W. (1983). A novel digestion technique for multi-element plant analysis. \u003cem\u003eCommunications in Soil Science and Plant Analysis\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e, 239\u0026ndash;249.\u003c/li\u003e\n\u003cli\u003eOberndorfer, E., Lundholm, J., Bass, B., Coffman, R. R., Doshi, H., Dunnett, N., \u0026hellip; Rowe, B. (2007). Green Roofs as Urban Ecosystems: Ecological Structures, Functions, and Services. \u003cem\u003eBioScience\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e(10), 823\u0026ndash;833. https://doi.org/10.1641/B571005\u003c/li\u003e\n\u003cli\u003eRegan, K. M., Nunan, N., Boeddinghaus, R. S., Baumgartner, V., Berner, D., Boch, S., \u0026hellip; Marhan, S. (2014). Seasonal controls on grassland microbial biogeography: Are they governed by plants, abiotic properties or both? \u003cem\u003eSoil Biology and Biochemistry\u003c/em\u003e, \u003cem\u003e71\u003c/em\u003e, 21\u0026ndash;30. https://doi.org/10.1016/j.soilbio.2013.12.024\u003c/li\u003e\n\u003cli\u003eRumble, H., \u0026amp; Gange, A. C. (2017). Microbial inoculants as a soil remediation tool for extensive green roofs. \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e102\u003c/em\u003e, 188\u0026ndash;198. https://doi.org/10.1016/j.ecoleng.2017.01.025\u003c/li\u003e\n\u003cli\u003eSchrader, S., \u0026amp; B\u0026ouml;ning, M. (2006). Soil formation on green roofs and its contribution to urban biodiversity with emphasis on Collembolans. \u003cem\u003ePedobiologia\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e, 347\u0026ndash;356.\u003c/li\u003e\n\u003cli\u003eShafique, M., Xue, X., \u0026amp; Luo, X. (2020). An overview of carbon sequestration of green roofs in urban areas. \u003cem\u003eUrban Forestry and Urban Greening\u003c/em\u003e, \u003cem\u003e47\u003c/em\u003e(May 2019), 126515. https://doi.org/10.1016/j.ufug.2019.126515\u003c/li\u003e\n\u003cli\u003eSinsabaugh, R. L., Carreiro, M. M., \u0026amp; Repert, D. A. (2002). Allocation of extracellular enzymatic activity in relation to litter composition, N deposition, and mass loss. \u003cem\u003eBiogeochemistry\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e(1), 1\u0026ndash;24. https://doi.org/10.1023/A:1016541114786\u003c/li\u003e\n\u003cli\u003eSmith, J. E., Heath, L. S., \u0026amp; Woodbury, P. B. (2004). How to estimate forest carbon for large areas from inventory data. \u003cem\u003eJournal of Forestry\u003c/em\u003e, \u003cem\u003e102\u003c/em\u003e(5), 25\u0026ndash;31. https://doi.org/10.1093/jof/102.5.25\u003c/li\u003e\n\u003cli\u003eStarry, O., Lea-Cox, J. D., Kim, J., \u0026amp; van Iersel, M. W. (2014). Photosynthesis and water use by two Sedum species in green roof substrate. \u003cem\u003eEnvironmental and Experimental Botany\u003c/em\u003e, \u003cem\u003e107\u003c/em\u003e, 105\u0026ndash;112. https://doi.org/10.1016/j.envexpbot.2014.05.014\u003c/li\u003e\n\u003cli\u003eTeemusk, A., \u0026amp; Mander, \u0026Uuml;. (2007). Rainwater runoff quantity and quality performance from a greenroof: The effects of short-term events. \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e, 271\u0026ndash;277. https://doi.org/10.1016/j.ecoleng.2007.01.009\u003c/li\u003e\n\u003cli\u003eTeemusk, A., \u0026amp; Mander, \u0026Uuml;. (2011). The Influence of Green Roofs on Runoff Water Quality: A Case Study from Estonia. \u003cem\u003eWater Resources Management\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e, 3699\u0026ndash;3713. https://doi.org/10.1007/s11269-011-9877-z\u003c/li\u003e\n\u003cli\u003eTodorov, D., Driscoll, C. T., Todorova, S., \u0026amp; Montesdeoca, M. (2018). Water quality function of an extensive vegetated roof. \u003cem\u003eScience of the Total Environment\u003c/em\u003e, \u003cem\u003e625\u003c/em\u003e, 928\u0026ndash;939. https://doi.org/10.1016/j.scitotenv.2017.12.085\u003c/li\u003e\n\u003cli\u003evan Seters, T., Rocha, L., Smith, D., \u0026amp; MacMillan, G. (2009). Evaluation of green roofs for runoff retention, runoff quality, and leachability. \u003cem\u003eWater Quality Research Journal of Canada\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(1), 33\u0026ndash;47. https://doi.org/10.2166/wqrj.2009.005\u003c/li\u003e\n\u003cli\u003eWang, H., Qin, J., \u0026amp; Hu, Y. (2017). Are green roofs a source or sink of runoff pollutants? \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e107\u003c/em\u003e(3888), 65\u0026ndash;70. https://doi.org/10.1016/j.ecoleng.2017.06.035\u003c/li\u003e\n\u003cli\u003eWhittinghill, L. J., Rowe, D. B., Schutzki, R., \u0026amp; Cregg, B. M. (2014). Quantifying carbon sequestration of various green roof and ornamental landscape systems. \u003cem\u003eLandscape and Urban Planning\u003c/em\u003e, \u003cem\u003e123\u003c/em\u003e, 41\u0026ndash;48. https://doi.org/10.1016/j.landurbplan.2013.11.015\u003c/li\u003e\n\u003cli\u003eWooster, E. I. F., Fleck, R., Torpy, F., Ramp, D., \u0026amp; Irga, P. J. (2022). Urban green roofs promote metropolitan biodiversity: A comparative case study. \u003cem\u003eBuilding and Environment\u003c/em\u003e, \u003cem\u003e207\u003c/em\u003e(PA), 108458. https://doi.org/10.1016/j.buildenv.2021.108458\u003c/li\u003e\n\u003cli\u003eXu, L., Baldocchi, D. D., \u0026amp; Tang, J. (2004). How soil moisture, rain pulses, and growth alter the response of ecosystem respiration to temperature. \u003cem\u003eGlobal Biogeochemical Cycles\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(4), 1\u0026ndash;10. https://doi.org/10.1029/2004GB002281\u003c/li\u003e\n\u003cli\u003eZhang, S., Lin, Z., Zhang, S., \u0026amp; Ge, D. (2021). Stormwater retention and detention performance of green roofs with different substrates: Observational data and hydrological simulations. \u003cem\u003eJournal of Environmental Management\u003c/em\u003e, \u003cem\u003e291\u003c/em\u003e(April). https://doi.org/10.1016/j.jenvman.2021.112682\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"urban-ecosystems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ueco","sideBox":"Learn more about [Urban Ecosystems](https://www.springer.com/journal/11252)","snPcode":"11252","submissionUrl":"https://submission.nature.com/new-submission/11252/3","title":"Urban Ecosystems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Green roofs, Carbon sequestration, Nutrient retention, Vegetation diversity","lastPublishedDoi":"10.21203/rs.3.rs-6856018/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6856018/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eExtensive green roofs can be implemented to counteract the adverse effects of nature loss in urban areas by providing ecosystem services, such as biodiversity increase, storm water management and temperature regulation. Green roofs also sequester carbon (C) and retain other nutrients, improving urban air and rainwater quality. This study examined how green roof characteristics (i.e. green roof age, substrate depth, fertilization, vegetation type and composition) affect total carbon (TC), total nitrogen (TN) and total phosphorus (TP) stocks in the green roof substrate, as well as two important nitrogen fluxes (mineralization and nitrification). We hypothesized that vegetation type (\u003cem\u003eSedum\u003c/em\u003e-only vs. diverse vegetation), substrate depth and fertilization would be the main characteristics affecting TC, TN and TP substrate stocks and nitrogen-fluxes.\u003c/p\u003e\u003cp\u003eTwelve extensive green roofs in three cities in Flanders, Belgium, were sampled across four seasons. Results showed that green roofs have a low C sequestration potential compared to natural ecosystems. Roofs with diverse vegetation had higher TC and TN substrate stocks, particularly those with mosses and herbs. Fertilization had no significant effect on vegetation or TC, TN, and TP substrate stocks, while substrate depth had a significant effect only on TP substrate stocks. Overall, green roofs offer limited C sequestration and nutrient retention, although optimizing substrate composition and increasing plant species richness could enhance these benefits. This study highlights the potential for improving green roof performance through better design and management practices.\u003c/p\u003e","manuscriptTitle":"The influence of green roof features on their carbon and nutrient stocks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-02 09:43:42","doi":"10.21203/rs.3.rs-6856018/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-03T10:02:40+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-30T14:49:57+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-16T15:13:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"33323396461783422397441895221427679787","date":"2025-10-02T20:53:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"249391621253539067758533915222916797579","date":"2025-08-31T10:20:24+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-25T12:24:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-24T02:44:30+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-23T23:07:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"Urban Ecosystems","date":"2025-06-09T15:42:38+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"urban-ecosystems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ueco","sideBox":"Learn more about [Urban Ecosystems](https://www.springer.com/journal/11252)","snPcode":"11252","submissionUrl":"https://submission.nature.com/new-submission/11252/3","title":"Urban Ecosystems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"34a22757-a44f-4d1e-b110-8873280c5c86","owner":[],"postedDate":"September 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-01-19T17:06:49+00:00","versionOfRecord":{"articleIdentity":"rs-6856018","link":"https://doi.org/10.1007/s11252-025-01898-x","journal":{"identity":"urban-ecosystems","isVorOnly":false,"title":"Urban Ecosystems"},"publishedOn":"2026-01-17 16:30:45","publishedOnDateReadable":"January 17th, 2026"},"versionCreatedAt":"2025-09-02 09:43:42","video":"","vorDoi":"10.1007/s11252-025-01898-x","vorDoiUrl":"https://doi.org/10.1007/s11252-025-01898-x","workflowStages":[]},"version":"v1","identity":"rs-6856018","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6856018","identity":"rs-6856018","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}