Assessing light performance of vertical greenery shading in tropical climate

Natural light is abundant in tropical climates and advantageous for incorporating daylighting into building designs. However, this daylight intensity often leads to excessive brightness indoors, specifically in high-rise buildings with glass façades. In addressing sustainability concerns, incorporating greenery outside glass façades can effectively shade and alleviate eyestrain for building occupants. Therefore, this study aimed to investigate the effectiveness of plant leaves in reducing the high light intensity on glass facades. An experiment was conducted using the Vernonia elliptica plant, which thrives in medium to high sunlight in tropical climates. Three different leaf area indexes (LAI) were examined in this study as the independent variables, while light illuminance and luminance served as the dependent variables. Two identical box models measuring 1 m × 1 m × 1 m were utilized for the experiment. The two models were oriented towards the west and north, representing intense and longer light exposure. The first, the base case, featured a glass façade without any other additional element, whereas the other incorporated greenery on its glass façade. The obtained results indicated that the impact of leaf density on illuminance and luminance is significant, specifically when the LAI is doubled. It was also found that denser foliage with longer leaf strands produced better results, specifically at low altitudes. These results can be used to implement vertical greenery shading in high-rise office buildings.


Introduction
The tropical climate along the equator offers a significant advantage in terms of abundant sunlight, which is beneficial for implementing daylighting in buildings.However, this advantage poses a challenge for medium to high-rise office buildings with glass curtain walls dominating their facades [1].The penetration of sunlight often results in discomfort glare for office users, specifically during the dry season when the sky is clear or partially cloudy [2][3][4][5].A survey on discomfort glare in tropical skies with partially cloudy or intermediate skies dominance has found that the higher ratio of window luminance to background luminance was the most influential variable [5].To solve this problem, it is, thus, necessary to install sun-shading systems to lower the window luminance.In addition to sustainability concerns, vertical greenery systems (VGS) have emerged as a potential alternative for sun-shading, surpassing artificial solutions.VGS functions as a solar tracker [6], with its leaves naturally seeking sunlight.Each leaf possesses mechanisms for transmission, absorption, and reflection [7][8][9], and the presence of different plants with numerous layers of leaves [10,11] forms a forest-like shading effect for the earth [12].This approach mitigates the glare issue and fosters a connection between humans and nature [13,14], and direct contact with nature can improve the overall quality of life [15].Implementing VGS in office areas not only enhances visual connection with nature but also has a positive impact on productivity [16].
By incorporating greenery as a view, glare is reduced, thus relieving eyestrain and promoting the health and well-being of office users [17,18].Accordingly, it was found that vegetation serves as an effective shadow factor that can be compared to various artificial barriers [19,20].While the thermal parameters, heat transfer, and energy balance of vertical greenery have already been studied and formulated by various investigators [21][22][23] there is still a need to explore the specific mechanism through which plants filter the abundant incoming light on glass facades [24].This study focuses on leaf area index (LAI) as a quantitative variable to investigate its performance in shading sunlight.Furthermore, Vernonia elliptica plants, which thrive in tropical climates (nparks.com),were used for this purpose.The hanging strands of these plants, with denser leaves at the top, provide effective shading since the highest brightness levels usually get into buildings from the upper part of windows [25,26].

Plant organs and radiation interception
Plants can be categorized based on different components, including individual organs, whole plants/pure stands, and plant communities [10].Among the organs of a plant, its leaf plays a crucial role, serving as the primary receiver of radiation.Accordingly, certain wavelengths of radiation are essential for the process of photosynthesis as well as the overall growth and development of the plant.It is important to note that the interception of radiation depends on factors such as the area density, angle, and dispersion of the leaf [11].Leaf area density refers to the total leaf area per unit volume (m 2 /m 3 ).
In contrast, leaf angle encompasses the azimuth and inclination of leaves, categorized as planophile, plagiophile, erectophile, and extremophile.Leaf dispersion represents the interception of light at a certain leaf area density.Additionally, foliage thickness, expressed as leaf area index (LAI), denotes the age and lifecycle of the plant, which can depend on the season and its upkeep.LAI represents the total area per unit of ground covered [27].In the case of horizontal surfaces, an LAI of 3 indicates full radiation interception [11].In a study conducted on Shibataea Kumasasa plants, however, it was suggested that an LAI of 3 may be too dense to meet the daylight requirements in a drawing studio [28].

Leaf absorption, transmission, and reflection, as well as their correlation with leaf area index
The dominant parts of the solar radiation spectrum that reach the earth include ultraviolet, visible light, and near-infrared waves within 300 to 2500 nm.Among these spectrums, plant leaves primarily transmit visible and near-infrared wavelengths [29].However, it has been discovered that leaves do not absorb the range of 750-1250 nm (near IR wavelength) due to their chlorophyll content, which explains why leaf reflectance and transmittance are high in this region [9].Following LAI, it has been found that light transmission is inversely proportional to LAI, absorptance is proportional to LAI, low LAI reflects the visible and near IR, and high LAI reflects low visible light and high near IR [12].
Accordingly, there is a correlation between the number of leaf layers, transmittance, and reflectance.As the number of leaf layers increases, more light is absorbed or reflected, and less is transmitted.But only short wavelengths are transmitted at certain layers, while the longer ones are not transmitted but reflected [7].

Method
This study used two identical box models with dimensions of 1 m × 1 m × 1 m.Other experiments have been conducted where the results obtained from identical boxes but with different dimensions or surfaces were compared [13,19].This experiment using a model was the first step of a larger study toward the performance of biotic shading.In this first step, the objective was to find the influence of foliage thickness on different Leaf Area Index (LAI) on light performance.One of the vertical surfaces of the box was an opening made of clear glass, measuring 5 mm in thickness and 1 m × 1 m in dimension.Besides representing the usual module of room opening, this 1 m × 1 m dimension makes it more straightforward to get the LAI of the plant because LAI is the reference of foliage thickness counted by comparing the total leaf area to the 1 m2 area enclosed in.
Regarding the limitation, even though the result cannot be generalized, this study that describes the shading mechanism and the trend of its light performance can be used as a reference for a real room at the next step.To ensure that light penetration occurred solely through the glass opening without any internal surface reflection, all surfaces inside the box, except for the glass opening, were painted black.
Contrary to white, yellow, and blue pigments, which reflect visible light and near IR, black pigment absorbs all the light wavelengths, including near IR, to store heat [30].In this study, two scenarios were considered, the first as a base case with no vegetation and the second with vegetation placed at 15 cm from the front glass opening (see figures 1-3).The three different Leaf Area Index (LAI) values in this study, which correspond to the typical thickness of Vernonia plants and range from thinner to denser foliage, represent the independent factor of foliage thickness.LAI is typically used for horizontal surfaces, which are usually uniform.Still, in this study, VGS was applied to a vertical surface whose upper part exhibited thicker leaves than the lower part.Therefore, a manual leaf counting method [31], and the image j software [32], were utilized to study the area covered from the upper to lower parts.To minimize the influence of the soil and planter boxes, they were placed above the glass opening, allowing the shade to originate solely from the plants themselves.The LAI values were obtained by adjusting the position and quantity of planter boxes.The total leaf area was taken from the average leaf area dimension from the biggest to the smallest leaf samples and then multiplied by the total number of leaves counted manually.By dividing the stem length by the typical spacing between leaves along the stem, one can calculate the average number of leaves per stem.Image J software calculates the percentage of leaves in the box's upper, middle, and lower thirds, each taken from 1/3 of its 1-meter height.
The detailed properties of the plants can be found in table 1.The dependent factors in this study were illuminance and luminance.Illuminance measurements were conducted to assess the adequacy of daylight throughout the day.In contrast, luminance measurements were used to determine the intensity and brightness of incoming light based on different sun altitudes, such as 30°for low, 50°for mid, and 70°for high sun altitudes.Following this, multiple measurement points were established for data collection, and they include Lkc, which was used to measure luminance at the center of the glass surface, E1 for measuring the illuminance in the center of the box, respectively, and E0, for measuring the outdoor illuminance.For data collection, hobo data logger U12-012 and hobo pendant devices were employed to measure indoor and outdoor illuminance at 15-minute intervals throughout the day.Additionally, the Konica LS-150 instrument was used to measure the luminance at different sun altitudes, including 30°(low), 50°(mid), and 60-70°(high).Two orientations were selected for this study: the west, which represents the highest sunlight intensity, and the north, which represents the longest duration of sunlight exposure throughout the year.The measurement period for the west orientation spanned from May 11 to May 28, 2022, while the north orientation extended from July 18 to August 17 of the same year.These periods were selected to coincide with the summer solstice and the location's proximity to the equator.Furthermore, the chosen periods considered plants' thickness and leaves' dimensions at the time of measurement.Sun position data, including azimuth and altitude, were obtained from Planetcalc.com based on Surabaya's specific position and time, which were at 7°14' 57' South Latitude and 112°4 5' 3' East Longitude.

Plant properties
The Vernonia elliptica plants used in this study were cultivated in planter boxes with a well-balanced composition of soil and aggregate.These plants were carefully maintained starting in October 2021.During the measurement period, the age of the plants ranged from 6-12 months.On average, the dimensions of the leaves were approximately 1.5 cm in width and 3.5 cm in length.From observations, it was found that the Vernonia elliptica leaves spread out from the center of the strand, resulting in variations in azimuth, and the average leaf inclination was approximately 27°, indicating that the plants can be categorized as planophile.Accordingly, in this study, three different LAI values were employed: 0.6; 1.16, and 1.7.These values were selected because they correspond to the foliage thickness that the plants could attain at their age during the measurement time and represent the thickness according to their growing cycle.The LAI 0.,6 represents the six months to a year, the LAI  1.,16 represents 1-1.5 years, and the LAI 1.,7 represents the 1.5-2 years of the plant age (see figure 3).By maintaining the soil with regular watering, suitable fertilizer, and stem cutting regularly, this Vernonia plant will reach its maximum growth at four years old with a height of about 3-4 meters; then, it usually needs changing with the new plants because of the falling leaf phase.The actual room was usually 4 m high from floor to floor; since the upper window position brings the most glaring point, the upper and lower parts can be inserted by shading of plants, and the middle part of the windows can be exposed to bring up the daylighting as well as the view outside.

Illuminance of west orientation
The measurement for LAI 0.6 was conducted from May 11 to 16, 2022, and for representation, figure 4, comprising data from May 13, 2022, will be used to describe the light conditions at that specific time.During this period, the outdoor illuminance exhibited fluctuations throughout the day, ranging from 1895 lux at 6 am to a peak of 28933.5 lux at 9:30 am, then gradually decreasing to 516.7 lux at 5 pm.Following this, the bare glass façade allowed an average of 26.21% of outdoor daylight illuminance to penetrate, but at 3:15-3:45 pm, it unexpectedly exhibited an illuminance value of 125.7%.This occurrence suggests that the measurement point might have been subject to reflections from the glass surface or direct sunlight during that particular time.
Conversely, the presence of vegetation resulted in a smoother illuminance graph.The illuminance ranged from 7.9% to 39.79% of the outdoor illuminance, providing an average daylight of 14.6% to the indoor area.The average outdoor illuminance recorded was 13,500 lux, while the bare glass averaged 2,660 lux (excluding the direct light penetration from 3:15-3:45 pm).Also, the LAI 0.6 condition yielded an average illuminance of 1475 lux.
While analyzing the obtained data, it was determined that LAI 0.6 exhibited the highest illuminance reduction at 2:45 PM compared to other time intervals throughout the day.
The measurement for LAI 1.16 was conducted from May 18 to 22, 2022, and for representation, figure 5, comprising data from May 21, 2022, will be used to describe the light conditions at that specific time.Similar to previous measurements, the outdoor illuminance exhibited fluctuations throughout the day.It ranged from 2497.2 lux at 6 am, reaching its peak at 23422.4 lux at 09:15 am, gradually decreasing to 1184 lux at 5 pm.Notably, the bare glass façade allowed an average of 47.1% of daylight illuminance to penetrate, but at 2:30-2:45 pm, it surprisingly exhibited an illuminance value of 131.5%.This indicates that the measurement point was exposed to direct sunlight or glass reflections during that specific time interval.
On the other hand, for the condition with vegetation, the illuminance ranged from 5.15% to 21.14% of the outdoor illuminance.The average outdoor illuminance recorded during this period was 11950 lux, while the bare glass yielded an average of 4550 lux.It is also important to note that the LAI 1.16 condition resulted in an average illuminance of 1400 lux.Analyzing the data further, it was found that the LAI 1.16 condition exhibited a maximum reduction in illuminance at 3 PM compared to other time intervals during the day.
The LAI 1.7 measurement was carried out from May 24th to May 28th, 2022, and for representation, figure 6, comprising data from May 25th, 2022, will be used to describe the light conditions at that specific time.The outdoor illuminance exhibited fluctuations during this period, ranging from 2669.5 lux at 6 am to a peak of 26178 lux at 10 am, then decreasing to 721.2 lux at 5 pm.Furthermore, the bare glass façade allowed for illuminance levels ranging from 14.7% to 67.56%, with an average of 31.57%daylight.On the other hand, the presence of vegetation resulted in outdoor illuminance levels ranging from 5.48% to 15.25%.It is important to note that the average outdoor illuminance was measured at 9680 lux, while the bare glass exhibited an average of 3924 lux, and the LAI 1.7 averaged 1290 lux.The LAI 1.7 demonstrated the most significant reduction in illuminance according to time, at 5 PM.
To determine the illuminance reduction solely attributed to the plants, it is necessary to subtract the illuminance reduction caused by the glass facade.For west orientation buildings, the LAI 0.6 measurement resulted in an illuminance reduction ranging from 7.3% to 12.4%, with an average reduction of 8.96%.However, LAI 1.16 achieved the highest illuminance reduction, at 34.3%.LAI 1.7, also demonstrated a similar average illuminance reduction value, which was 38% (figure 7).These results show a significant difference in the obtained values for LAI 0.6 and LAI 1.16, while that of LAI 1.16 and LAI 1.7 was relatively small despite the same 0.56 point difference in the LAI values.This non-linear relationship between less light and the difference in LAI is in line with a study by [7,9,33].That study found that overlapping leaves mostly block longer wavelengths of light while letting shorter wavelengths pass through, which means that light absorption changes less for denser LAI values.Following this, an analysis of leaf shading overlapping with the glass surface was performed to investigate other potential factors that might influence this result.Previous studies have shown that the shading effect of the glass surface's enclosure affects light transmission reduction and leaf density [34].
Figure 8 shows the lighter blue color, representing the surface of LAI 1.16 leaves not covered by LAI 0.6.From the observation, it was found that the difference in leaf shading surface area between LAI 1.1 and LAI 0.6 was 18.95%.Similarly, the lighter red represents the portion of LAI 1.7 leaves that LAI 1.16 did not cover.It was also revealed that the difference between LAI 1.7 and LAI 1.16 leaf shading surfaces was 32.08%.The study of this leaf shade overlap showed that the thickness of the leaves, especially at the top (see table 1 for the percentage of leaves), had a bigger effect on the experiment's results for lowering the amount of light coming in than the total enclosed surface area.Therefore, further measurements were required to obtain more reliable and meaningful data.Regarding the optimum illuminance reduction concerning time, the results showed that higher LAI values are associated with greater reductions in illuminance, particularly during periods of lower sun position.

Luminance of west orientation
In the case of west orientation buildings, the luminance levels were measured at 11:30, 13:30, and 15:30.These specific time points were selected to examine the penetration of sunlight at different sun altitudes, such as low (30°), mid (50°), and high (60°−70°).The average luminance values obtained from these measurements are presented in table 2.
Figure 9 shows that the luminance percentage (%) decreases as the density increases.This trend is evident for both the 60-70°and 50°sun altitudes, where there was a linear decrease in luminance within a range of 8%-12% for every 0.56 point difference in LAI.The shallower line on the graph indicates that at a 50°sun altitude, the middle portion of the glass displayed almost constant leaf overlap across all LAI values, resulting in minimal differences in luminance.On the other hand, at a 30°sun altitude, a higher decrease in luminance ranging from 15%-22% was observed.This result can be attributed to the varying quantities of leaves enclosing the glass's upper, middle, and lower parts.In the case of LAI 0.6, there was almost no enclosed foliage at the lower part, and thus, the foliage failed to block the sunlight from striking the area at a lower sun altitude.Meanwhile, at the upper part, the 60-70°sun altitude, the densest leaves were found to block sunlight effectively (figure 10).
To validate this result, another experiment was conducted with LAI 3.4 from March 21st to March 24th, 2023, and the findings are documented in table 3.With LAI 3.4, the shading factor averaged 49.64%, and the luminance significantly decreased, specifically at an altitude of 50°, which was halved compared to LAI 1.7.

Illuminance of north orientation
When measuring the west orientation, the best illuminance reduction was seen at LAI 1.16 compared to LAI 0.6.On average, there was little difference between LAI 1.16 and LAI 1.7, so a different method was used for the north orientation.The measurement was first conducted by comparing LAI 1.16 to bare glass in this case.Subsequently, measurements were taken simultaneously between plants with LAI 0.6 and LAI 1.7, compared to LAI 1.16 as the base case.
The first measurement for the north orientation was conducted from the 20th to the 26th, 2022.For representation, figure 11, comprising data from July 20, 2022, will be used to describe the light conditions at that specific time.The outdoor illuminance exhibited fluctuations ranging from 3272.2 lux at 6 am to a peak of 46844.8lux at 7 am, followed by a decrease to 538.2 lux at 5:15 pm.Accordingly, the average measured outdoor illuminance was 22,000 lux.The bare glass façade allowed for outdoor illuminance levels ranging from 4.4 to  45.8%, with an average of 2600 lux.On the other hand, when vegetation LAI 1.16 was present, the illuminance ranged from 1.9% to 17.4%, averaging at 1000 lux.The highest difference in illuminance between the bare glass and LAI 1.16 was observed at 12:30 PM, with a percentage difference of 28.4%, whereas the lowest difference was    and 13 show data from August 3rd, 2022, which was used to show the light at that time.The observation found that the outdoor illuminance in this case also exhibited fluctuations, ranging from 882.6 lux at 6 am to a peak of 46844.8lux at 2 pm, followed by a decrease to 559.7 lux at 5:15 pm.The average outdoor illuminance measured was 12000 lux.When comparing the illuminance of LAI 1.16 to the outdoor illuminance, it ranged from 3.3% to 16.2%, with an average of 1100 lux.On the other hand, the illuminance of LAI was 0.6 compared to the outdoor illuminance, which ranged from 7.1% to 37.0%, averaging 2350 lux.
The highest difference between LAI 0.6 and LAI 1.16 was observed at 12.15 pm, with a percentage difference of 20.82%, while the smallest was 3.8%, which occurred at 2 pm when the outdoor illuminance was at its highest.On average, the illuminance of LAI 0.6 was between 9.6% and 12.2% higher compared to LAI 1.16.
The measurement of LAI 1.16 in comparison to LAI 1.7 was conducted from August 11th to August 17th, 2022.For representation, figures 14 and 15, comprising data from August 12, 2022, were used to describe the light at that specific time.Accordingly, the outdoor illuminance exhibited fluctuations, ranging from 839.6 lux at 6 am to a peak of 154312.1 lux at 1:30 pm, followed by a decrease to 247.6 lux at 5:30 pm.When comparing the illuminance of LAI 1.16 to the outdoor illuminance, it ranged from 2.07% to 17.6%, with an average of 1800 lux.On the other hand, the illuminance of LAI was 1.7 compared to the outdoor illuminance, which ranged from 1.1% to 10.3%, averaging at 1000 lux.The highest difference between LAI 1.7 and LAI 1.16 was observed at 4 pm, and the percentage difference was 7.96%.The smallest difference was 1% and was found at 1:30 pm when the outdoor illuminance was at its highest.On average, the illuminance of LAI 1.7 was 0.8% to 14% lower compared to LAI 1.16.
In the case of the North orientation, despite the outdoor illuminance exhibiting significant fluctuations and reaching its peak at noon, the illuminance inside both boxes remained low.The LAI 0.6 brought about an average illuminance reduction of only 2.9%, while the LAI 1.16 reduced 13.8%.Additionally, LAI 1.7 achieved an average illuminance reduction of 17.7% (figure 16).These results indicate a lower percentage of illuminance reduction for the north orientation compared to the west.This disparity can be attributed to the path of the sun, which starts from the east in the morning, reaches its highest at midday, and then moves to the west in the afternoon.This shows that the sunlight passes over the façade surface and the vertically hanging leaves of the  plant.Notably, no incoming sunlight directly strikes the glass or the plant surface; only the uniform light penetrates through (figure 17).In the case of LAI 0.6, there was almost no significant illuminance reduction inside the box compared to the bare glass.Even for LAI 1.7, the densest leaves on the upper part only had a significant influence at noon.This shows that a high LAI is unnecessary for the North orientation.However, the data shows that higher LAI values lead to better illuminance reduction for low-altitude angles.
The results show that the maximum illuminance reduction for LAI 0.6 was observed at 12:15 PM, while that of LAI 1.16 and LAI 1.7 was at 12:45 PM and 4 PM, respectively.

Luminance of north orientation
For the north orientation, the luminance measurements were conducted at different time intervals to study the impact of various sun altitudes.The measurements were taken from 07:45 to 08:00 and from 15:00 to 15:15 for the low sun altitude (30°), from 09:15 to 09:45 and 13:15 to 13:45 for the mid-altitude (50°), and from 11:15 to 11:30 for the high sun altitude (60°−70°).The average luminance results are shown in table 4.
Figure 18 shows that as the density increases, there is a decrease in the luminance percentage.In the case of LAI 0.6, the luminance reduction percentage was insignificant for all sun altitudes.However, for altitudes of 60-70°, every increase in LAI results in a luminance reduction ranging from 3.96% to 4.22%.For altitudes of 50°F

Figure 1 .
Figure 1.Layout and section of the box model and the point of measurement.

Figure 2 .
Figure 2. Inside the box model: left = bare glass; right = glass plus vegetation.

Figure 3 .
Figure 3. Bare glass and three leaf area indexes from the front side.

Figure 4 .
Figure 4. Illuminance graph of LAI 0.6 versus Without vegetation versus Outdoor.

Figure 5 .
Figure 5. Illuminance graph of LAI 1.16 Versus without vegetation Versus outdoor.

Figure 6 .
Figure 6.Illuminance graph of LAI 1.7 Versus without vegetation versus outdoor.

Figure 7 .
Figure 7. Average illuminance reduction percentage according to LAI, West orientation.

Figure 10 .
Figure 10.Diagram of West orientation sun path versus the box model.

Figure 16 .
Figure 16.Average illuminance reduction percentage according to LAI, North orientation.

Figure 17 .
Figure 17.Diagram of North orientation sun path versus the box model.

Table 2 .
Average luminance of different LAI versus bare glass, West orientation.

Table 3 .
Result summary of West orientation.