Abstract

In this study, the authors designed and applied a new irrigation method called Capillary Wicking Irrigation (CWI), which used microfiber fabric as the source material of irrigation. At present, the effects of CWI on soil moisture, plant growth and surface temperature of a green roof with rain storage are not clear. An experiment was conducted on a green roof in Guangzhou. The authors set three transparent plexiglass containers (A, B and C) with a side length of 1.5 m as an experimental frame on the roof. The authors put ‘steering wheel’ microfiber CWI in containers A and C, which were planted with Sedum lineare Thunb and Fittonia verschaffeltii, respectively. Container B with no CWI was planted with Sedum lineare Thunb. Results indicated that CWI could increase soil water content and make the variation of soil water content gentle in the containers on the roof. The green roof with rain storage had the function of heat preservation in winter and cooling effect in summer, especially for the green roof with CWI. Compared with container B, container A gave better plant growth, for ‘steering wheel’ microfiber CWI can basically provide automatic and suitable water supply for the plant. Therefore, CWI is an effective infiltration irrigation technique for roof greening.

INTRODUCTION

With the acceleration of China's urbanization process, both urban pluvial flooding and water shortage are joining forces to restrict urban development. On the one hand, the increase of impervious area and the decrease of river and lake area greatly weaken the city's ability to retain rainfall runoff, so that the rainwater can quickly gather into a large amount of surface runoff (Bengtsson 2005), which can easily result in urban waterlogging. On the other hand, if the future climate becomes drier, the provision of a secure water supply will become more pressing (Mortazavi-Naeini et al. 2015). In order to cope with these problems, China has set off a construction boom in the sponge city (Che et al. 2015). Sponge city refers to a city that can regulate and store rainwater like a sponge, releasing and utilizing the stored water when needed. As an important part of the sponge city, the green roof has received much attention and become a research hotspot.

The green roof generally consists of a water storage layer, drainage layer, filter layer, soil matrix layer and vegetation layer (Berndtsson 2010; Shafique et al. 2018). Roof greening in the city helps to weaken runoff (Voyde et al. 2010; Stovin et al. 2012; Fassman-Beck et al. 2013; Speak et al. 2013; Li & Babcock 2014; Versini et al. 2015), save energy (Niachou et al. 2001; Castleton et al. 2010; D'Orazio et al. 2012; Chen 2013), increase evapotranspiration (Ouldboukhitine et al. 2012), mitigate the heat-island effect (Susca et al. 2011; Kolokotsa et al. 2013), reduce noise and air pollution (Van Renterghem & Botteldooren 2009; Rowe 2011), and beautify the environment (Brenneisen 2006; Fernandez-Canero & Gonzalez-Redondo 2010; Xiao et al. 2014). Due to the heat consumption effect of evapotranspiration (Mueller et al. 2015), a green roof can bring cooling. Xiao et al. (2014) pointed out that a green roof could effectively reduce the surface temperature of a building in summer and increase the surface temperature of a building in winter. However, the plant in a green roof often suffers from water stress during a long dry period (Qin et al. 2016). Moreover, literature on the irrigating green roof with rain storage is scarce in roof greening research. Hence, how to select the proper irrigation way is a key factor that determines the successful development of the green roof. This study applied a new irrigation method called Capillary Wicking Irrigation (CWI) to a green roof with rain storage.

So far, sub-surface infiltration irrigation is one of the most advanced water-saving irrigation technologies in the world. In many countries with advanced water-saving technology, sub-surface infiltration irrigation is widely used in greenhouses, orchards and greenbelt (Howes & Abrahams 2003; Kalfountzos et al. 2007). The development and technology research of sub-surface infiltration irrigation equipment is still in its initial stage, which is not mature enough. The blockage of sub-surface infiltration irrigation prevents its development. Capillary Wicking Irrigation (CWI) is a novel infiltration irrigation method, in which water is applied slowly by microfiber wicking infiltration based on capillarity, which can solve the problem of blockage and reduce the ineffective water consumption of soil evaporation. The capillary wicking is often made of cotton, linen, polyester and other materials. In addition to capillarity, the soil moisture moves toward the root of the plant by water potential gradient and plant transpiration. Microfiber CWI technology absorbs water from the rain storage layer below through microfiber capillarity, and it overcomes water gravity with the help of capillary force, and absorbs water into the upper soil layer for plant use. The capillarity of microfiber CWI is similar to that of ‘sweating’, which slowly and continuously infiltrates soil. The amount of infiltration irrigation is induced by soil–plant coupling, and the water supply realizes the automatic water supply (Tan et al. 2014). At the same time, CWI can help to store a part of the precipitation on the roofs and then decrease direct surface runoff to reduce the impact of urban flooding. On the one hand, roof soil can absorb precipitation; on the other hand, the CWI system can store rainwater and apply it. Thus it is expected to slow down waterlogging to a certain extent and control precipitation runoff within a certain range. CWI has solved the problem of pipeline blockage in infiltration irrigation (Wang 2018). In addition, compared with drip irrigation and sprinkling irrigation, the CWI system does not need to be pressurized, so it will not produce adverse effects such as power consumption and head loss. In addition, CWI will not lead to soil hardening, which has a broad application prospect. Nowadays, CWI has been studied in field experiments in a vineyard in Ningxia, China, which showed that under the same amount of irrigation water, compared with furrow irrigation, CWI was more effective in reducing water stress on grapes, thus increasing yield, soluble solids content, sugar–acid ratio and anthocyanin content (Sun et al. 2018). Compared with straight capillary wicking and annular capillary wicking, the ‘steering wheel’ capillary wicking gave higher irrigation uniformity (Huang et al. 2018). In this study, we adopted ‘steering wheel’ microfiber CWI on a green roof with rain storage. However, research on CWI is scarce and the effects of CWI on a green roof with rain storage are not yet clear. The primary objective of this study is to evaluate the effects of ‘steering wheel’ microfiber CWI on soil moisture, plant growth and surface temperature of a green roof with rain storage.

MATERIALS AND METHODS

Experimental site and design

The pilot green roof was installed on Di Huan Building at the campus of the Sun Yat-Sen University (SYSU), Haizhu District, Guangzhou, China. The study site in Guangzhou has a subtropical monsoon climate. The average annual air temperature is 22 °C. The highest monthly temperature in the year is in July, with an average monthly temperature of 28.7 °C. The lowest monthly temperature in the year is in January, with an average monthly temperature of 12.5 °C. Guangzhou is rich in precipitation, with an average annual precipitation of 1,695.9 mm. There is abundant rainfall from April to September, but there is often dry weather during other months.

In this study, we used a plexiglass frame with the Capillary Wicking Irrigation (CWI) system to store roof rainwater. There were three transparent plexiglass containers (A, B and C) and every container had a side length of 1.5 m. Figure 1 shows the ‘steering wheel’ microfiber Capillary Wicking Irrigation (CWI) system. The bottom layer of the plexiglass container was a rainwater storage layer, the middle layers were an infiltration layer and a soil matrix layer, and the uppermost layer was the vegetation layer. The water storage layer thickness was set to 15 cm, and four drainage holes were provided at a height of 15 cm to remove excess rain. In order to enable the water storage layer to support the weight of the soil matrix layer, the storage layer was filled with ceramsite (Figure 2(a)). The infiltration layer with a thickness of 5 cm was located below the soil matrix layer and above the rainwater storage layer. The infiltration layer was made of 150 g/m2 non-woven fabric, and the upper and lower ends of the grid plate were tightly wrapped with non-woven fabric to play the role of two layers of filtration (Figure 2(a)). Non-woven fabrics have good corrosion resistance, water permeability, and antimicrobial properties, and do not need to be replaced during long-term use. The grid plate, made of glass fiber reinforced plastic (GFRP), was used to separate the soil matrix layer and water storage layer. The GFRP grid flap had parameters of 5 cm × 5 cm × 5 cm (length × width × height) (Figure 2(a)). The culture soil was used in the soil matrix layer with a thickness of 20 cm. The culture soil is loose and breathable, does not accumulate water, has good water seepage performance, and the soil pH value is between 5.6 and 7.0, which is suitable for much vegetation. In this experiment, we adopted the ‘steering wheel’ type capillary wicking infiltration irrigation method, and the capillary wicking was made of microfiber material. We put the ‘steering wheel’ microfiber capillary wicking with non-woven fabric outside (Figure 2(b)) in containers A and C. Both container A and container C had nine such infiltration devices. Container B had no infiltration device. The function of the outer non-woven fabric was to separate the capillary wicking from the soil particles as much as possible to minimize clogging. For the vegetation layer, we planted Sedum lineare Thunb (Figure 2(c)) in container A and container B, and Fittonia verschaffeltii (Figure 2(d)) in container C. Sedum lineare Thunb and Fittonia verschaffeltii are more and more widely used in the green roof, especially in South China with the subtropical monsoon climate. Sedum lineare Thunb is suitable for growing at high temperature and humidity, so it is suitable for living in Guangzhou, China (Ma 2018). Fittonia verschaffeltii has great waterlogging tolerance, and it likes a suitable temperature and wet environment (Shi 2017).

Figure 1

‘Steering wheel’ microfiber Capillary Wicking Irrigation (CWI) system on the roof.

Figure 1

‘Steering wheel’ microfiber Capillary Wicking Irrigation (CWI) system on the roof.

Figure 2

(a) Grid plate and water storage layer filled with ceramsite in experimental container, (b) ‘steering wheel’ microfiber capillary wicking with non-woven fabric outside, (c) Sedum lineare Thunb and (d) Fittonia verschaffeltii on the roof.

Figure 2

(a) Grid plate and water storage layer filled with ceramsite in experimental container, (b) ‘steering wheel’ microfiber capillary wicking with non-woven fabric outside, (c) Sedum lineare Thunb and (d) Fittonia verschaffeltii on the roof.

The experiment began on June 28, 2017, and the seedling transplanting time was November 15, 2017. After transplanting, every experimental container was watered with 3,000 ml per week. On December 10, 2017, the water storage layer of every experimental container was filled with water, and then artificial irrigation was stopped.

Measurement and sampling

A small meteorological station monitoring the meteorological elements in this study was located beside the experimental site. The monitored elements included temperature, solar radiation, rainfall, etc. The use of an automatic meteorological station improved the reliability of observational data to some extent. To measure the soil water content in the three containers, three S-SMC-M005 soil moisture sensors (HOBO company, USA) were placed in every experimental container. Every soil moisture sensor was buried at 5 cm soil depth and the data-recording interval was 10 min. This experiment measured the surface temperature using the MI-210 infrared temperature-measuring instrument (Apogee company, USA). The measurement accuracy is ±0.3 °C at −10–65 °C, and the angle of view is 22° half-angle. We held the temperature-measuring instrument at a vertical height of 20 cm above the surface of the plant in the container to measure the surface temperature of the pilot green roof. Nine points were uniformly selected for surface temperature measurement in every container. Meanwhile, we also held the temperature-measuring instrument at a vertical height of 20 cm above the surface of the ordinary roof to measure the surface temperature of the ordinary roof. The surface of the ordinary roof was made of concrete and cement. Nine points were uniformly selected for surface temperature measurement of the ordinary un-greened roof. We measured the surface temperature at 13:00 once a week. During the growth stage of Sedum lineare Thunb and Fittonia verschaffeltii, the plant height of the sampled plants was measured by a ruler. When the plant height was measured, the plant was in a natural state, and the longest stem height was taken as the plant height value. We also measured the maximum canopy diameter of Sedum lineare Thunb by a ruler once a week. Six plants were uniformly selected in every container for plant growth measurement.

Statistical analysis

The data of repeated soil water content tests were treated by the formula of average value, as follows: 
formula
(1)
where is the average of the soil water content; is the effectively measured soil water content value; and n is the actual number of measurement points. The data of repeated surface temperature tests were treated by the formula of average value, as Equation (2) shows: 
formula
(2)
where (°C) is the average of the surface temperature of the green roof or ordinary roof; (°C) is the effectively measured temperature value; and n is the actual number of measurement points.

The data analysis of variance (ANOVA) using SPSS software was performed to determine significant differences among different treatments. The least significant difference (LSD) at the 5% level was used in the multiple comparisons.

RESULTS AND DISCUSSION

Precipitation and soil water content

A total of 15 rainfalls in Guangzhou were recorded at the meteorological station from December 10, 2017, to March 22, 2018 (Table 1). The maximum precipitation was 77.4 mm on January 6, 2018, and this rainfall lasted for 22 h. During this stage in winter, the amount of precipitation ranged from 0.4 mm to 77.4 mm, and the rainfall intensity ranged from 0.3 mm/h to 12.8 mm/h. In addition, a total of 13 rainfalls in Guangzhou were recorded at the meteorological station from June 6 to July 26 in 2018 (Table 2). The maximum precipitation was 228.8 mm on June 8, 2018, and this rainfall lasted for 17.5 h. During this stage in summer, the amount of precipitation ranged from 7.4 mm to 228.8 mm, and the rainfall intensity ranged from 2.4 mm/h to 13.1 mm/h. Whether in winter or summer, the rainfall statistical characteristic parameters varied widely. Therefore, the statistical rainfall events had a certain representation.

Table 1

The rainfall in Guangzhou from December 10, 2017, to March 22, 2018

Rainfall eventStart timeEnd timePrecipitation (mm)Rainfall intensity (mm/h)Rainfall duration (h)
2017/12/16 05:04 2017/12/16 05:24 0.4 0.8 0.5 
2017/12/28 12:44 2017/12/28 13:44 0.6 0.6 
2018/01/06 00:24 2018/01/06 07:54 2.0 0.3 7.5 
2018/01/06 11:34 2018/01/07 09:44 77.4 3.5 22 
2018/01/08 02:24 2018/01/08 12:24 6.0 0.6 10 
2018/01/08 17:14 2018/01/08 22:44 6.2 1.1 5.5 
2018/01/09 06:44 2018/01/09 09:34 6.4 2.6 2.5 
2018/01/30 02:24 2018/01/31 04:54 7.2 0.3 26 
2018/02/22 07:24 2018/02/22 17:14 4.6 0.5 10 
10 2018/03/07 22:14 2018/03/08 06:24 14.0 1.8 
11 2018/03/08 10:04 2018/03/08 10:44 0.6 0.9 0.7 
12 2018/03/15 08:54 2018/03/15 10:54 1.2 0.6 
13 2018/03/19 05:14 2018/03/19 06:24 4.4 4.4 
14 2018/03/19 13:14 2018/03/19 15:04 10.4 5.2 
15 2018/03/19 23:14 2018/03/20 00:14 12.8 12.8 
Rainfall eventStart timeEnd timePrecipitation (mm)Rainfall intensity (mm/h)Rainfall duration (h)
2017/12/16 05:04 2017/12/16 05:24 0.4 0.8 0.5 
2017/12/28 12:44 2017/12/28 13:44 0.6 0.6 
2018/01/06 00:24 2018/01/06 07:54 2.0 0.3 7.5 
2018/01/06 11:34 2018/01/07 09:44 77.4 3.5 22 
2018/01/08 02:24 2018/01/08 12:24 6.0 0.6 10 
2018/01/08 17:14 2018/01/08 22:44 6.2 1.1 5.5 
2018/01/09 06:44 2018/01/09 09:34 6.4 2.6 2.5 
2018/01/30 02:24 2018/01/31 04:54 7.2 0.3 26 
2018/02/22 07:24 2018/02/22 17:14 4.6 0.5 10 
10 2018/03/07 22:14 2018/03/08 06:24 14.0 1.8 
11 2018/03/08 10:04 2018/03/08 10:44 0.6 0.9 0.7 
12 2018/03/15 08:54 2018/03/15 10:54 1.2 0.6 
13 2018/03/19 05:14 2018/03/19 06:24 4.4 4.4 
14 2018/03/19 13:14 2018/03/19 15:04 10.4 5.2 
15 2018/03/19 23:14 2018/03/20 00:14 12.8 12.8 
Table 2

The rainfall in Guangzhou from June 6 to July 26 in 2018

Rainfall eventStart timeEnd timePrecipitation (mm)Rainfall intensity (mm/h)Rainfall duration (h)
2018/06/06 04:30 2018/06/06 10:00 13.4 2.4 5.5 
2018/06/07 02:00 2018/06/07 15:30 74.8 5.5 13.5 
2018/06/08 00:30 2018/06/08 18:00 228.8 13.1 17.5 
2018/06/12 18:00 2018/06/13 12:00 154.1 8.6 18 
2018/06/22 03:00 2018/06/22 05:00 12.0 6.0 
2018/06/23 06:00 2018/06/23 09:30 23.2 6.6 3.5 
2018/06/24 05:00 2018/06/24 08:30 19.1 5.4 3.5 
2018/06/25 16:00 2018/06/25 19:00 16.4 5.5 
2018/07/02 12:30 2018/07/02 14:00 10.6 7.1 1.5 
10 2018/07/06 14:30 2018/07/06 17:30 7.4 2.5 
11 2018/07/07 14:00 2018/07/07 18:30 26.9 6.0 4.5 
12 2018/07/13 13:00 2018/07/13 15:30 18.2 7.3 2.5 
13 2018/07/14 09:30 2018/07/14 16:30 38.3 5.5 
Rainfall eventStart timeEnd timePrecipitation (mm)Rainfall intensity (mm/h)Rainfall duration (h)
2018/06/06 04:30 2018/06/06 10:00 13.4 2.4 5.5 
2018/06/07 02:00 2018/06/07 15:30 74.8 5.5 13.5 
2018/06/08 00:30 2018/06/08 18:00 228.8 13.1 17.5 
2018/06/12 18:00 2018/06/13 12:00 154.1 8.6 18 
2018/06/22 03:00 2018/06/22 05:00 12.0 6.0 
2018/06/23 06:00 2018/06/23 09:30 23.2 6.6 3.5 
2018/06/24 05:00 2018/06/24 08:30 19.1 5.4 3.5 
2018/06/25 16:00 2018/06/25 19:00 16.4 5.5 
2018/07/02 12:30 2018/07/02 14:00 10.6 7.1 1.5 
10 2018/07/06 14:30 2018/07/06 17:30 7.4 2.5 
11 2018/07/07 14:00 2018/07/07 18:30 26.9 6.0 4.5 
12 2018/07/13 13:00 2018/07/13 15:30 18.2 7.3 2.5 
13 2018/07/14 09:30 2018/07/14 16:30 38.3 5.5 

Figure 3 shows that the soil water content of container A was higher than that of container B. The variations of soil water content in containers A and C showed a gentle trend, whereas container B gave a drastic change (Figure 3(a)–3(c)). The reason was that the experimental containers with ‘steering wheel’ microfiber CWI (containers A and C) could still infiltrate water upward into the soil without precipitation, so that the soil did not suffer from water shortage for a long time. During rainfall, the soil evaporation was negligible, so the water potential gradient inside the soil and the amount of infiltration of the capillary wicking was greatly reduced or became zero, and the soil water was supplemented by rainwater, whereas the soil water content of the experimental container without CWI (container B) was only slowly replenished by the evaporation of water from the rainwater storage layer, so the soil water content was low and the response of the soil water content to precipitation was drastic. Compared with winter, the variations of soil water content in containers A and B became gentler in summer (Figure 3(a), 3(b), 3(d) and 3(e)). Compared with container A, the variation of soil water content in container B showed a sharper trend in summer (Figure 3(d) and 3(e)). In the several rainfall events, the soil water content increased negatively (Figure 3) due to the shortage of rainwater in the storage layer, the small amount of precipitation and the great evaporation intensity in the dry period.

Figure 3

The variation of soil water content before and after rainfall in containers on the roof: (a, b, c) from December 10, 2017, to March 22, 2018; and (d, e) from June 6 to July 26 in 2018.

Figure 3

The variation of soil water content before and after rainfall in containers on the roof: (a, b, c) from December 10, 2017, to March 22, 2018; and (d, e) from June 6 to July 26 in 2018.

Figure 4 shows that in the same drought period, all the soil water contents decreased. From January 9 to January 29, the soil water contents in containers A and C (with ‘steering wheel’ microfiber CWI) decreased by 21% and 10%, respectively, whereas the soil water content in container B (with no CWI) decreased by 71%. From July 15 to July 24, the soil water content in container A (with ‘steering wheel’ microfiber CWI) decreased by 34.9%, whereas the soil water content in container B (with no CWI) decreased by 64.1%. Compared with container B, containers A and C gave higher soil water contents during the drought stage (Figure 4(a)), which could be interpreted as being for two reasons. One is that containers A and C had ‘steering wheel’ microfiber CWI, which could absorb water in the rainwater storage layer to increase soil water content. The other is that there might be no evaporation from containers A and C, because the surface temperatures of containers A and C were higher than that of container B (Figure 7(a)). For container A, the soil water contents in the subsequent statistical days was in a relatively stable state (Figure 4(a) and 4(b)). The results indicated that the ‘steering wheel’ microfiber CWI can keep the soil water content in a relatively stable state during the drought stage. When the transpiration of plants consumed soil water, under the influence of the water potential gradient, the ‘steering wheel’ microfiber capillary wicking would absorb water in the rainwater storage layer, and supply and replenish water into the matrix layer, so that the soil water content of the matrix layer would be stable again. This process was repeated again and again in the soil, and the macroscopic performance was that the soil water content remained relatively stable. Therefore, the ‘steering wheel’ microfiber CWI can basically provide automatic and suitable water supply for the plant on a green roof with rain storage, which is progress in irrigation science and roof greening.

Figure 4

Effects of Capillary Wicking Irrigation (CWI) on soil water content in containers on the roof for the same drought time: (a) from January 9 to January 29 in 2018; and (b) from July 15 to July 24 in 2018. Values are means ± SE (n = 3). The different lower-case letters mean significant differences (P < 0.05) among treatments by least significant differences of Duncan's new multiple range test.

Figure 4

Effects of Capillary Wicking Irrigation (CWI) on soil water content in containers on the roof for the same drought time: (a) from January 9 to January 29 in 2018; and (b) from July 15 to July 24 in 2018. Values are means ± SE (n = 3). The different lower-case letters mean significant differences (P < 0.05) among treatments by least significant differences of Duncan's new multiple range test.

Plant growth

Plant height and canopy diameter are important indicators reflecting plant growth. The growth of plants on a green roof also indirectly impacts evapotranspiration, roof surface temperature, heat-island effects and the ability to mitigate urban pluvial flooding. For Sedum lineare Thunb in container A, the highest plant height reached 16.7 cm, the lowest 14.9 cm, and the growth rate ranged from 338.2% to 421.9% (Figure 5(a)). The highest canopy diameter reached 29.1 cm, the lowest 16.1 cm, and the growth rate ranged from 106.4% to 243.2% (Figure 6). For Sedum lineare Thunb in container B, the highest plant height reached 10.8 cm, the lowest 5.8 cm, and the growth rate ranged from 75.7% to 237.5% (Figure 5(b)). The highest canopy diameter reached 15.1 cm, the lowest 11.2 cm, and the growth rate ranged from 55.5% to 122.1% (Figure 6). The average plant height of Sedum lineare Thunb in container A with CWI was 97.1% higher than that of Sedum lineare Thunb in container B without CWI (Figure 5(a) and 5(b)). For Fittonia verschaffeltii in container C, the actual height did not change (Figure 5(c)). When we observed again in March, the Fittonia verschaffeltii was all apoptotic. Sedum lineare Thunb has strong vitality and can grow on the thinner soil matrix. Under the condition of long-term fertilization and irrigation, the plant height of Sedum lineare Thunb reached 13.8 cm after three months in southern China (Liang et al. 2014). Zhao & Kuang (2005) reported that in northern China, the highest plant height of Sedum lineare Thunb reached 13 cm after overwintering without artificial irrigation. In our roof experiment, the highest plant height of Sedum lineare Thunb reached 16.7 cm with CWI.

Figure 5

Effects of Capillary Wicking Irrigation (CWI) on plant height in (a) container A, (b) container B and (c) container C on the roof.

Figure 5

Effects of Capillary Wicking Irrigation (CWI) on plant height in (a) container A, (b) container B and (c) container C on the roof.

Figure 6

Effects of Capillary Wicking Irrigation (CWI) on canopy diameter of Sedum lineare Thunb in container A and container B on the roof.

Figure 6

Effects of Capillary Wicking Irrigation (CWI) on canopy diameter of Sedum lineare Thunb in container A and container B on the roof.

Figure 7

Effects of Capillary Wicking Irrigation (CWI) on roof surface temperature during (a) winter and (b) summer. Vertical bars are standard errors and values are means ± SE (n = 9).

Figure 7

Effects of Capillary Wicking Irrigation (CWI) on roof surface temperature during (a) winter and (b) summer. Vertical bars are standard errors and values are means ± SE (n = 9).

The ‘steering wheel’ microfiber CWI can better provide the vegetation with the moisture needed for growth. For container A (with ‘steering wheel’ microfiber CWI), the water supply was relatively uniform, and the growth of the Sedum lineare Thunb in different positions was basically similar, and the plant height and canopy diameter were also increased by the supply of water. For container B (with no CWI), the soil water was mainly derived from natural rainfall and evaporation from the water storage layer. In container B, the water distribution at each position was not the same, so that the growth of the Sedum lineare Thunb in each position was not completely consistent. Compared with B, container A gave higher plant height and canopy diameter due to the ‘steering wheel’ microfiber CWI being able to increase soil water content, improve irrigation uniformity and promote transpiration of plants. Sedum lineare Thunb is suitable for planting on the roof. However, Fittonia verschaffeltii is more demanding on the growth environment and is not suitable for outdoor planting, especially on the roof. In this experiment, it was speculated that the main reason for the death of Fittonia verschaffeltii was that the Fittonia verschaffeltii suffered from sunlight stress, temperature stress and so on.

Surface temperature

Soil evaporation and vegetation transpiration of the green roof with rain storage changed the original hydrometeorology of the roof. Compared with an ordinary (non-greening) roof, the surface temperature of the green roof with rain storage changed greatly, especially in the role of cooling in summer and warming in winter. Figure 7(a) shows that the surface temperature of container A was almost the highest among all treatments. Except for 3 January 2018, the average surface temperature of container A was significantly higher than those of containers B and C, which could be related to the specific heat capacity of water. The variations of surface temperature in container A showed a gentler trend, compared with container B. In winter, compared with ordinary roof surface temperature, the mean surface temperatures of the three containers A, B and C were maximally enhanced by 7.6, 2.8 and 5.5 °C, respectively. DeNardo et al. (2005) indicated that the maximum surface temperature of a green roof averaged 6 °C higher in winter based on roof temperature data. The average surface temperatures in the three experimental containers in winter were, in most cases, higher than the ordinary roof surface temperature, which indicated that the green roof had the function of heat preservation in winter, which was consistent with the finding of Xiao et al. (2014), who reported that a green roof effectively reduced a building heating load in winter.

Figure 7(b) shows that the surface temperature of container A was lower than that of container B, except for 30 June and 20 July 2018. Containers A and B gave a lower surface temperature than an ordinary roof. In summer, compared with ordinary roof surface temperature, the mean surface temperatures of containers A and B were maximally reduced by 7.8 and 5.6 °C, respectively. Our roof experiment indicated that the green roof had the function of a cooling effect in summer, which was consistent with the findings of some researchers (Zhao & Xue 2008; Tang et al. 2010; Parizotto & Lamberts 2011). Jim & Peng (2012) indicated that the cooling effect of a green roof was more obvious when the solar radiation was larger and the soil moisture content was higher in summer. Onmura et al. (2001) reported that a green roof mainly relied on plant transpiration and soil surface evaporation to dissipate heat.

The different treatments produced different rates of vegetation coverage in the order of container A > container B > container C. In addition, container A (with ‘steering wheel’ microfiber CWI) had better effects of heat preservation in winter and cooling in summer than container B (with no CWI). The better the growth of the vegetation, the higher the surface temperature of the green roof in winter and the lower the surface temperature of the green roof in summer. At the same time, this showed that the ‘steering wheel’ CWI device could change the surface temperature of the green roof with rain storage by increasing soil water content, so that the green roof with rain storage had more obvious effects of heat preservation in winter and cooling in summer, thus it was more conducive to mitigating the urban heat-island effect.

CONCLUSIONS

Before and after rainfall, the ‘steering wheel’ microfiber CWI made the variation of soil water content on a green roof with rain storage gentle. The ‘steering wheel’ microfiber CWI significantly increased the soil water content during the drought stage. The growth of Sedum lineare Thunb in container A (with ‘steering wheel’ microfiber CWI) was better than in container B (with no CWI), for the ‘steering wheel’ microfiber CWI basically provided automatic and suitable water supply for the plant of the green roof with rain storage. Furthermore, the ‘steering wheel’ microfiber CWI improved irrigation uniformity and promoted transpiration of the plants. Sedum lineare Thunb is suitable for planting on the roof, whereas Fittonia verschaffeltii is not suitable for outdoor planting. In winter, compared with ordinary roof surface temperature, the mean surface temperatures of three containers A, B and C were maximally enhanced by 7.6, 2.8 and 5.5 °C, respectively, which indicated that the green roof had the function of heat preservation in winter. In summer, compared with ordinary roof surface temperature, the mean surface temperatures of containers A and B were maximally reduced by 7.8 and 5.6 °C, respectively, which indicated that the green roof had a cooling effect in summer. The better the growth of the vegetation, the more obvious the effects of heat preservation in winter and cooling in summer on the green roof. It is useful and beneficial to apply this new irrigation method, called ‘steering wheel’ microfiber CWI, to the green roof with rain storage.

ACKNOWLEDGEMENTS

We acknowledge, with gratitude, the financial support from the National Natural Science Foundation of China (31470707 and 31270748), and the Shenzhen Science and Technology Project (JCYJ20150331160617771 and JC201005280681A). We would like to thank Lunhang Ma, Xinxin Liu, Lingna Yang and Honggang Yu for their participation in this roof experiment.

REFERENCES

REFERENCES
Bengtsson
L.
2005
Peak flows from thin sedum-moss roof
.
Nordic Hydrology
36
(
3
),
269
280
.
Brenneisen
S.
2006
Space for urban wildlife: designing green roofs as habitats in Switzerland
.
Urban Habitats
4
,
27
36
.
Castleton
H. F.
,
Stovin
V.
,
Beck
S. B. M.
&
Davison
J. B.
2010
Green roofs; building energy savings and the potential for retrofit
.
Energy and Buildings
42
(
10
),
1582
1591
.
Che
W.
,
Zhao
Y.
&
Li
J. Q.
2015
Considerations and discussions about sponge city
.
South Architecture
2015
(
4
),
104
107
(in Chinese)
.
DeNardo
J. C.
,
Jarrett
A. R.
,
Manbeck
H. B.
,
Beattie
D. J.
&
Berghage
R. D.
2005
Stormwater mitigation and surface temperature reduction by green roofs
.
Transactions of the ASAE
48
(
4
),
1491
1496
.
D'Orazio
M.
,
Di Perna
C.
&
Di Giuseppe
E.
2012
Green roof yearly performance: a case study in a highly insulated building under temperate climate
.
Energy and Buildings
55
,
439
451
.
Fassman-Beck
E.
,
Voyde
E.
,
Simcock
R.
&
Hong
Y. S.
2013
4 living roofs in 3 locations: does configuration affect runoff mitigation?
Journal of Hydrology
490
,
11
20
.
Fernandez-Canero
R.
&
Gonzalez-Redondo
P.
2010
Green roofs as a habitat for birds: a review
.
Journal of Animal and Veterinary Advances
9
,
2041
2052
.
Huang
S. Y.
,
Zhang
Q. T.
&
Cheng
D. P.
2018
Effects of capillary wicking irrigation on soil moisture, temperature and plant growth of green roof with rain storage
. In:
Proceedings of the 2018 Meeting of the CFCCE'18 (Conference on Flood Catastrophes in a Changing Environment)
.
Jim
C. Y.
&
Peng
L. L. H.
2012
Weather effect on thermal and energy performance of an extensive tropical green roof
.
Urban Forestry and Urban Greening
11
(
1
),
73
85
.
Kalfountzos
D.
,
Alexiou
I.
,
Kotsopoulos
S.
,
Zavakos
G.
&
Vyrlas
P.
2007
Effect of subsurface drip irrigation on cotton plantations
.
Water Resources Management
21
(
8
),
1341
1351
.
Li
Y.
&
Babcock
R. W.
2014
Green roof hydrologic performance and modeling: a review
.
Water Science and Technology
69
(
4
),
727
738
.
Liang
J. J.
,
Deng
G. Z.
,
Liang
Z. Z.
,
Wang
J.
,
Mao
J. Y.
&
Huo
H.
2014
Study on the breeding of Sedum lineare Thunb and roof greening technology
.
Southern Horticulture
25
(
6
),
19
21
(in Chinese)
.
Ma
L. Y.
2018
Experiment and Simulation of Sedum Lineare Green Roof in Guangzhou Area
.
South China University of Technology, Guangzhou, China
.
Mortazavi-Naeini
M.
,
Kuczera
G.
,
Kiem
A. S.
,
Cui
L.
,
Henley
B.
,
Berghout
B.
&
Turner
E.
2015
Robust optimization to secure urban bulk water supply against extreme drought and uncertain climate change
.
Environmental Modelling and Software
69
,
437
451
.
Mueller
N. D.
,
Butler
E. E.
,
McKinnon
K. A.
,
Rhines
A.
,
Tingley
M.
,
Holbrook
N. M.
&
Huybers
P.
2015
Cooling of US Midwest summer temperature extremes from cropland intensification
.
Nature Climate Change
6
(
3
),
317
322
.
Niachou
A.
,
Papakonstantinou
K.
,
Santamouris
M.
,
Tsangrassoulis
A.
&
Mihalakakou
G.
2001
Analysis of the green roof thermal properties and investigation of its energy performance
.
Energy and Buildings
33
(
7
),
719
729
.
Onmura
S.
,
Matsumoto
M.
&
Hokoi
S.
2001
Study on evaporative cooling effect of roof lawn gardens
.
Energy and Buildings
33
(
7
),
653
666
.
Ouldboukhitine
S.-E.
,
Belarbi
R.
&
Djedjig
R.
2012
Characterization of green roof components: measurements of thermal and hydrological properties
.
Building and Environment
56
,
78
85
.
Qin
H. P.
,
Peng
Y. N.
,
Tang
Q. L.
&
Yu
S. L.
2016
A HYDRUS model for irrigation management of green roofs with a water storage layer
.
Ecological Engineering
95
,
399
408
.
Rowe
D. B.
2011
Green roofs as a means of pollution abatement
.
Environmental Pollution
159
(
8–9
),
2100
2110
.
Shafique
M.
,
Kim
R.
&
Rafiq
M.
2018
Green roof benefits, opportunities and challenges – a review
.
Renewable and Sustainable Energy Reviews
90
,
757
773
.
Shi
X. M.
2017
Studies on Endurance of Three Kinds of Color-Leafed Ground Cover Plants under Waterlogging Stress
.
Sichuan Agricultural University
,
Chengdu, China
(in Chinese).
Speak
A. F.
,
Rothwell
J. J.
,
Lindley
S. J.
&
Smith
C. L.
2013
Rainwater runoff retention on an aged intensive green roof
.
Science of the Total Environment
461–462
,
28
38
.
Stovin
V.
,
Vesuviano
G.
&
Kasmin
H.
2012
The hydrological performance of a green roof test bed under UK climatic conditions
.
Journal of Hydrology
414–415
(
11
),
148
161
.
Sun
Y. B.
,
Zhang
Q. T.
,
Wang
Y.
,
Zhou
Y. L.
,
Qu
Z.
&
Wang
Z. P.
2018
Influence research on infiltration irrigation by annular capillary wicking to growing development and fruit quality of grape in arid region
.
Agricultural Research in the Arid Areas
36
(3),
1
8
.
Susca
T.
,
Gaffin
S. R.
&
Dell'Osso
G. R.
2011
Positive effects of vegetation: urban heat island and green roofs
.
Environmental Pollution
159
,
2119
2126
.
Tan
S. L.
,
Zhang
Q. T.
&
Qiu
G. Y.
2014
Infiltration irrigation technology helps to save water resources
.
High-Technology & Industrialization
5
,
66
69
.
Tang
M. F.
,
Zheng
S. K.
&
Yang
Z. J.
2010
Thermal performance and energy analysis of green roof
.
Journal of Civil, Architectural & Environmental Engineering
32
(
2
),
87
90
.
Van Renterghem
T.
&
Botteldooren
D.
2009
Reducing the acoustical façade load from road traffic with green roofs
.
Building and Environment
44
(
5
),
1081
1087
.
Versini
P.-A.
,
Ramier
D.
,
Berthier
E.
&
de Gouvello
B.
2015
Assessment of the hydrological impacts of green roof: from building scale to basin scale
.
Journal of Hydrology
524
,
562
575
.
Wang
X. L.
2018
A new type of emitters test on the uniformity of water outflow from the infiltration irrigation pipe capillary drainage belt
.
Water Saving Irigation
3
,
39
41 + 45
.
Xiao
M.
,
Lin
Y. L.
,
Han
J.
&
Zhang
G. Q.
2014
A review of green roof research and development in China
.
Renewable and Sustainable Energy Reviews
40
,
633
648
.
Zhao
D. G.
&
Kuang
X. M.
2005
Preliminary report on overwintering test of Sedum lineare thunb in Beijing
.
Grassland and Turf
2005
(
1
),
71
72
(in Chinese)
.
Zhao
D. G.
&
Xue
W. C.
2008
Effect of light roof greening on saving electric power consumption
.
Acta Agriculturae Shanghai
24
(
1
),
99
101
.