Effects of Urban Layout, Façade Orientation, and Façade Height on Photosynthetically Active Radiation (PAR) Availability in a Dense Residential Area: A Dynamic Analysis in Shanghai

Abstract

Photosynthetically Active Radiation (PAR) is critical for sustaining plant growth in the ground and on building surfaces, but how to accurately predict PAR availability in a complex urban environment can be a challenge. Using an advanced ray-tracing software (Radiance 4.0) and local weather data, this study presents a dynamic analysis of the effects of layout, façade orientation and height on PAR availability in four high density residential areas in Shanghai city, China. A metric system was also adopted using three light level requirements of outdoor plants (low, medium, high light levels). Key findings included: (1) the urban layout with the highest ratio of building height to north–south facing adjacent building separation achieved the higher levels of PAR availability for low/medium light level plants and the lower levels of PAR availability for high-light plants for middle and low façades and the ground, while high façades in all layouts could see similar PAR availability for all plants. (2) The PAR availability for low/medium-light plants decreased with the increasing façade height, while the PAR availability for high-light plants showed the opposite trend. (3) The north façade and its ground had higher levels of PAR availability for low/medium-light plants and lower levels of PAR availability for high-light plants than other façades. (4) All layouts offered more opportunities to apply high-light and medium-light plants at façades and the ground.

1. Introduction

1.1. Benefits of Greenery Systems in Urban Areas

With rapid population growth and urbanization in the 21st century, cities worldwide are increasingly transformed into high density environments, often at the expense of natural ecosystems [1]. These densely urbanized areas have given rise to critical challenges, including excessive energy consumption, air pollution, and water scarcity, all of which adversely affect human health and wellbeing [2,3]. Therefore, the need for sustainable and liveable cities has become critical. In recent years, as an effective solution to mitigate the negative impacts of high-density urban environments, a greening approach (i.e., the use of plants on and around buildings) has received increasing attention. The urban greening system is the incorporation of green spaces and elements into urban environments and infrastructure, such as streets, public spaces, building roofs, and façades [4,5]. The benefits of urban greening have been highlighted in research studies [4,5,6,7,8]. For urban climates, the urban greening mitigates the urban heat island effect through shading surfaces and facilitating evapotranspiration (for example, two types of green roofs, with 100 mm and 200 mm soil thickness, were found to produce a significant reduction in heat gain of 59% and 96%, respectively) [4], while vegetation can also improve air quality through capturing particulate matter (PM2.5/PM10) and sequestering carbon dioxide [6]. For the urban ecology, by providing habitats for the urban wildlife, these greening systems can enhance biodiversity and support ecological balance in a city [7]. For urban dwellers, greenery systems and the access to nature can promote human physical and mental health [5,8]. Given these advantages, the application of greening systems in urban areas has been applied to improve environment quality and human health, especially in residential areas.

1.2. Applications of Greening Systems in Residential Areas

As one critical component of urban ecosystems, the greening system in residential areas has attracted attention when initiating a sustainable urban design. Given the limited availability of public spaces in high density residential areas, plants are typically installed on building façades, roofs, and the narrow spaces between buildings. Typical applications include vertical greening system (e.g., Bosco Verticale in Milan, One Central Park in Sydney [9,10,11]), green roofs (e.g., Palazzo Verde in Antwerp, Dakparken Anton and Gerard in Eindhoven [12,13]), and ground green spaces [14,15]. The vertical greening system can be categorized into two primary types based on construction methods—green façades and living walls [9]. For green façades, climbing plants (e.g., Parthenocissus, Hydrangea) are planted in the ground and can grow directly on and up the wall [9,10]. Other plants (e.g., shrubs, grass, perennial plants, succulent plants) can be grown in pre-installed containers, which are then developed into modular systems that are attached to walls [9,11]. Green roofs are roofs where different kinds of vegetation/plants grow on engineered substrates [12] and which have four categories, intensive, semi-intensive, single-course extensive, and multi-course extensive, based on the plant type and substrate depth [13]. The intensive and semi-intensive green roofs have deep soil layers and diverse plant species, while single-course and multi-course extensive green roofs use shallow substrates and limited plants species [12,13]. Ground green spaces encompass all vegetated areas at the urban level and are implemented at varying scales [14], including roadside greenery and vegetation barriers along streets, parks and meadows, greenways and corridors (such as green trails for walking/cycling), recreational and urban gardening facilities (such as community gardens) [15]. The design and installation of these greening systems may depend on the availability of solar radiation in a given complex urban environment.

1.3. Photosynthetically Active Radiation (PAR) and Greenery Systems

Light is an essential factor for all plant life cycle stages, from seed germination and vegetative growth to flowering and reproduction [16]. Solar radiation (sunlight) serves as the primary light source for plants. However, plants selectively absorb photosynthetically active radiation (PAR), the 400–700 nanometre (nm) wavelength range of the electromagnetic spectrum that drives the photosynthesis process and fundamentally determines plants growth [17]. PAR is quantified as photosynthetic photon flux density (PPFD), which is defined as the number of photons (light particles) in the 400–700 nm range received by a surface for a specified amount of time [18]. In controllable facilities, such as greenhouses and chambers, PAR has been widely adopted as a key metric for assessing and optimizing plant growth conditions, ensuring that artificial lighting systems meet the spectral and intensity requirements for plant growth [17]. Different types of indoor plants require different PPFDs for growth. Based on needs and tolerances for light, three common types of plants include: low-light plants (538–2690 lux), medium-light plants (2690–10,760 lux), and high-light plants (>10,760 lux) [19]. However, in an urban environment, PAR availability remains understudied. Unlike controlled indoor environments, PAR availability in urban environments is complex and is influenced by site latitude, time of day and obstructions (e.g., buildings and vegetation density) [20]. These constraints can reduce photosynthetic efficiency and impair the ecological performance of greening systems [20]. Therefore, investigating the availability of PAR is necessary when establishing urban greening systems and relevant facilities for growing plants, particularly in urban residential areas.

1.4. Solar Availability in Residential Areas

Solar availability in residential areas has been studied across two spatial scales: ground-level spaces between buildings and building envelopes (roofs and façades) [20,21,22,23]. For the ground-level spaces, studies explored the effects of urban density and building height [24,25]. A high level of urban density can negatively affect solar availability by increasing shading from neighbouring buildings [26,27]. Taller buildings correlate with reduced solar radiation reaching ground surfaces due to the increased vertical obstruction [28]. Studies of building envelope focus on the effects of urban layout and building morphology (e.g., orientation, height, shape) [24]. For the layout, façade-level solar exposure is sensitive to urban form, especially horizontal (e.g., street grid) and vertical (e.g., skyline irregularity) spatial configurations [29]. For the building morphology, increasing building height may reduce incident solar radiation on vertical façades [27]. Street and building orientations can significantly influence the solar distribution across façades [30,31]. In addition, the effects of urban geometry on solar availability are affected by seasonal and geographic factors. For example, in winter in high-latitude cities, low solar altitude angles intensify shading, drastically reducing solar access to both ground and vertical surfaces [27].

1.5. Research Gaps and the Present Study

Given the discussions above, some research gaps can be identified. First, PAR availability has mainly been applied to assess indoor plant growth conditions [17], while the application of PAR availability in terms of the evaluation of urban greening systems has not been widely studied, especially in urban residential areas. Second, early studies focused on the estimation of solar radiation across entire building surfaces [27,29], neglecting PAR variability at vertically stratified positions (e.g., ground, mid-level, and upper façade positions). Third, previous studies applied general urban models and did not include the dynamic analysis of interactions between PAR availability and various building configurations (e.g., building heights, spacing distances, orientations). This type of study is especially in Chinese cities with specific climates and complicated planning constraints.

The present study explored the effects of layout, orientation, and façade height on PAR availability and distributions in a high-density residential area in Shanghai, China. Using advanced ray-tracing simulation software (Radiance) and local weather data, PAR values were simulated at the four façades of each studied building, together with the ground between two adjacent buildings, in four typical residential layouts. A metric system, based on the requirements of outdoor plant for solar radiations, was developed to evaluate the PAR availability. The results could ultimately be developed into evidence-based design guidelines for planning a biophilic city and promoting ecological resilience and public health in rapidly urbanizing regions.

2. Materials and Methods

2.1. Location and Climate

The location studied was Shanghai city in eastern China (Latitude: 31.23° N, Longitude: 121.47° E). Shanghai had a humid subtropical climate with an average annual temperature of 18 °C [32]. The highest average monthly temperature is 31.1 °C, while the lowest average monthly temperature is −5.6 °C. Total annual sunshine hours of Shanghai was 2134.7 [32].

2.2. Urban Models Studied

Four urban models (identified as layouts L1, L2, L3, and L4) were studied in terms of PAR availability (Figure 1). The selection of these urban models was based on two aspects: (1) They were typical residential urban layouts recommend by a local planning regulation in Shanghai [33]. (2) Their layouts and geometries were defined to meet the requirements (environmental performance, fire and structure safety, and transport system) of local and national regulations for the design of residential buildings [33,34,35]. The characteristics of these layouts included the distance between two north–south facing adjacent building façades (D1), and the distance between two east–west facing adjacent building façades (D2), the dimensions of a unit building (height H, width W, and length L), and the ratio of building height to building spacing (H/D1) (see details in

Table 1. Dimensions of four urban layouts and their unit buildings, and the heights of four calculation positions at façades and the ground.

Name of Layout	Distance		Building				H/D1	Height of Calculation Position Above the Ground (m)
	D1 (m)	D2 (m)	L (m)	W (m)	H (m)	Total Number of Floors		P1	P2	P3	P4
L1	35.6	17.2	45.3	12.4	20.9	6	0.59	0.05	1.41	8.5	15.6
L2	36.6	23	22.5	36.6	37.6	11	1.02	0.05	1.52	16.8	32.1
L3	53.4	35.7	29.4	16.2	55	18	1.02	0.05	1.52	27.5	53.5
L4	51.6	38.6	26.3	15.4	98.4	33	1.91	0.05	1.49	49.2	96.9

). Given the planning regulations [34,35], the building dimensions and the distances between buildings defined in these models may lead to four high density residential areas. According to Figure 1a,b, for the unit building at each layout, four façades, South (S), North (N), East (E), and West (W), were assessed in this study.

As displayed in Figure 1b,c and Table 1, for the unit building studied in each layout, the PAR availability was calculated at its centre façade position and the ground nearby, i.e., position P1 (aligned with the centre façade, with the height of 0.05 m above the ground and the distance of 0.5 m to the façade), position P2 (at the centre façade of ground floor), position P3 (at the centre façade of middle floor), and position P4 (at the centre façade of top floor). For each layout, a total of 16 positions were calculated to indicate the PAR availability. These represented possible positions for locating vegetation systems on façades and in the ground.

2.3. PAR and Lighting Metrics of Plants

Generally, PAR is quantified in terms of Photosynthetic Photon Flux Density (PPFD, µmol/m²/s) [36].

Table 2. Light and PAR requirements of three types of plants [19,37].

Type	Illuminance (Foot-Candles a)	Illuminance (lux)	PPFD (µmol/m²/s)
Low-light plants	[50–250)	[538–2690)	[9.95–49.76)
Medium-light plants	[250–1000)	[2690–10,760)	[49.76–199.06)
Highlight plants	>1000	>10,760	>199.06

^a: Conversion from footcandle to lux: 1 footcandle = 10.76 lux.

gives the light and PAR requirements of plants [19,37], while

Table 3. Examples of three types of plants which can be applied at façades and the ground in a residential area [19,37].

Type	Name
Low-light plants (LLP)	Chinese evergreen (Aglaonema), Dumb cane (Dieffenbachia), Dracaena, English ivy (Hedera helix), Philodendron, Homalomena, Pothos (Epipremnum), Arrowhead plant (Syngonium).
Medium-light plants (MLP)	Ferns; Begonias (Begonia), Spider plant (Chlorophytum), Schefflera (Schefflera), Grape ivy (Cissus), Croton (Codiaeum), Jade plant (Crassula), Flame violet (Episcia) and Peperomia (Peperomia)
High-light plants (HLP)	Cacti and succulents, Orchid cactus (Epiphyllum), Hibiscus, Ti plant (Cordyline), Gardenia (Gardenia) and Caladium

lists typical outdoor plants, which could be planted at positions on building façades and in the ground in a residential area. The illuminance (lux) (visual part of the solar radiation spectrum) was converted to PPFD (µmoles/m²/s) using the algorithm given in [36,38]:

$$\mathrm{P}\mathrm{P}\mathrm{F}\mathrm{D}=0.0185\times \mathrm{I}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}$$

(1)

According to Table 2 and Table 3, three PPFD ranges for sustaining the growth of plants were given [19,37]: (i) low level (LP) (9.95–49.76 µmoles/m²/s), for example, Aglaonema, Dieffenbachia; (ii) medium level (MP) (49.76–199.06 µmoles/m²/s), for example, Ferns, Begonia; (iii) high level (HP) (≥199.06 µmoles/m²/s), for example, Cacti and succulents, Gardenia. A very low PPFD (<9.95 µmol/m²/s) might not be able to sustain the healthy growth of plants. Thus, a PAR metric in this study was applied based on the three PPFD ranges (i, ii, iii).

2.4. Assessment of PAR Availability and Numeric Simulation

The assessment of PAR availability in the four urban layouts were conducted using the following steps:

• First, the daylight illuminances (lux) at specific positions on the façades and the ground (Figure 1) were simulated using the lighting package DAYSIM/RADIANCE [39] and the weather data of Shanghai (source: energyplus.net/weather-region/asia_wmo_region_2/CHN, accessed on 10 January 2025). The simulation period was a year (365 days), and the daytime period was considered as 06:00–18:00 (step: 1 h). A total of 4757 values of illuminance was achieved at a specific position.
• Second, the illuminance dataset was converted to the PPFD dataset using Equation (1).
• Third, the PAR availability at these positions could be indicated by the percentage of time over a year when the PPFD levels fall within each of the three PPFD ranges, such as: APT_LP (the percentage when PPFD falling within the low range (LP)), APT_MP (the percentage when PPFD falling within the medium range (MP)), APT_HP (the percentage when PPFD falling within the high range (HP)).

For each layout, the centre building was selected for simulation, and there were sixteen points analyzed, including 12 façade positions and four ground positions (Figure 1b,c). Based on the recommendations of daylighting estimation and typical materials applied in building surfaces and external ground [40,41], the reflectance of surfaces of buildings and the external ground were set as 0.4 and 0.2, respectively.

The validation studies of RADIANCE and DAYSIM simulation (solar and daylight) have been conducted for over 20 years, including field measurements under various sky conditions [42,43] and analysis in various complex urban environments [44]. For the simulation using RADIANCE and DAYSIM, the ambient settings can determine how indirect light ray (diffuse interreflections) is calculated, significantly affecting accuracy and computation time [39,45]. This study adopted the recommended settings [39] according to the complexity of the urban models studied, such as 1000 (Ambient Divisions), 5 (Ambient Bounce), 20 (Ambient Super-Samples), 300 (Ambient Resolution), and 0.1 (Ambient Accuracy). In addition, a convergence analysis [45] was conducted to further verify the simulations (see Appendix A).

3. Results

This section shows the analysis of the impact of layout, façade height, and orientation on the annual percentages of time in terms of three PPFD ranges: APT_LP, APT_MP, APT_HP.

3.1. Effects of Layout and Façade Height on PAR Availability at the Building

Figure 2 indicates variations of APT_LP, APT_MP, APT_HP at the south façade according to three façade positions (P2, P3, and P4) and four layouts (L1, L2, L3, L4).

For LP (Figure 2a): at P2, the L4 had a higher APT_LP value compared with other layouts, while the other three layouts achieved similar APT_LP values. At P3, similarly, the highest APT_LP value can be found with the L4 and other layouts L2, L3, and L4 achieved the same APT_LP value of 10%. However, at P4, there were no big differences in APT_LP values between the four layouts. For the three façade positions, all layouts can see the range of APT_LP value between 5% and 20%. For MP (Figure 2b), a similar general trend as the LP can be found: at P2 and P3, significantly higher APT_MP values were found in L4, whereas all layouts showed similar APT_MP values at P4. All APT_MP values at the three positions of all layouts ranged from 25% to 40%. For HP (Figure 2c), at P2 and P3, L4 has much lower APT_HP values than the other three layouts (P2: 25%, P3: 34%), while L1, L2 and L3, delivered similar APT_HP values. At P4, the APT_HP values in the four layouts were similar and within a range of 48% to 49%. The range of all APT_HP values at the three positions in the four layouts was 25~50%. Figure 2 also shows the varying trends of APT_LP, APT_MP, and APT_HP according to three façade heights. In general, both APT_LP and APT_MP decreased with the increasing façade height, while the APT_HP showed an opposite trend (P4 > P3 > P2).

Table 4. Relative differences of PPFD percentages (R_{APT_P}) of two façade positions (P3, P4), taking the position P2 as the reference (south façade, four layouts: L1, L2, L3, L4).

Annual PPFD Percentage	RAPT_P (%)
	L1		L2		L3		L4
	P3	P4	P3	P4	P3	P4	P3	P4
APTLP	−9	−27	−17	−42	−23	−38	−19	−56
APTMP	−6	−21	−6	−16	−3	−16	−13	−29
APTHP	8	26	13	29	14	30	36	92

lists the relative differences of PPFD percentages of P3 or P4 ($R_{APT\_P}$), taking P2 as the reference, which can be calculated by the equation:

$${\mathrm{R}}_{\mathrm{A}\mathrm{P}\mathrm{T}\_\mathrm{P}}={\displaystyle \frac{{\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{j}}-{\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{P}2}}{{\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{P}2}}}\times 100\mathrm{\%}$$

(2)

where ${\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{j}}$ is the value of APT_LP, APT_MP, or APT_HP at P3 or P4 (i = LP, MP, HP, j = P3, P4), ${\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{P}2}$ is the value of APT_LP, APT_MP, or APT_HP at P2 (i = LP, MP, HP).

• For APT_LP, taking P2 as the reference, P3 achieved large reductions (absolute R_{APT_P} > 10%) for layouts L2, L3 and L4, whilst significantly larger decreases in this value (absolute R_{APT_P} > 20%) were achieved at P4 for all layouts.
• For APT_MP, taking P2 as the reference, in layouts L1, L2 and L3, a small reduction was found for P3 (absolute R_{APT_P} < 10%), while P4 showed higher reductions in this value (absolute R_{APT_P} > 10%). L4 showed the large decrease in APT_MP,with the position moving from P2 to P3 and from P2 to P4 (absolute R_{APT_P} > 10%).
• For APT_HP, P3 delivered significantly higher values than P2 for layouts of L2, L3, and L4 (R_{APT_P} > 10%), while there were large increases in APT_HP when moving from P2 to P4 (R_{APT_P} > 25%) for all layouts.

Figure 3 shows the varying trends of APT_LP, APT_MP, APT_HP on the north façade with three positions (P2, P3, and P4) and four layouts (L1, L2, L3, L4). For LP (Figure 3a), at P2, L4 exhibited higher APT_LP than the other layouts (relative difference ≥ 10%), while at P3, a large difference in PPFD percentage can be found between L3 and L4 (relative difference = 15%). At P4, all layouts achieved the same APT_LP of around 9%. For the three positions, all layouts showed a range of APT_LP from 5% to 15%. For MP (Figure 3b), at P2 and P3, L4 achieved a significantly higher APT_MP than other layouts (relative difference ≥ 10%). All layouts had a similar APT_MP value of around 33% at P4. All positions in the four layouts had the APT_MP range from 30% to 45%. For HP (Figure 3c), L4 delivered the lowest APT_HP compared to the other three layouts at P2 and P3. At P2, L1 had a significantly higher APT_HP (relative difference > 10%) compared to the other layouts, while no big difference was found between L2 and L3 (relative difference < 10%). In addition, at P4, no clear differences of APT_HP were found between the four layouts (relative difference < 10%). The APT_HP values at the three positions for all four layouts ranged from 20% to 45%. Based on Figure 3, for APT_LP and APT_MP, there was a significant drop with the positions moving from P2 to P4. For APT_HP, a significant increase with an increasing height was found for all layouts (P4 > P3 > P2).

Table 5. Relative differences of PPFD percentages (R_{APT_P}) of two façade positions (P3, P4), taking the position P2 as the reference (north façade, four layouts: L1, L2, L3, L4).

Annual PPFD Percentage	RAPT_P (%)
	L1		L2		L3		L4
	P3	P4	P3	P4	P3	P4	P3	P4
APTLP	0	−25	−8	−31	−8	−25	−13	−40
APTMP	3	−11	−8	−16	−8	−18	−5	−21
APTHP	−3	21	17	37	18	43	24	86

gives the relative differences of PPFD percentages of P3 or P4 (R_{APT_P}), taking P2 as the reference. For APT_LP, the differences between P2 and P3 were relatively small for L1, L2, and L3 (absolute R_{APT_P} < 10%), while a larger difference of this value between the two positions was found in L4 (absolute R_{APT_P} = 13%). For APT_MP, small differences between P3 and P2 were seen in all layouts (absolute R_{APT_P} < 10%), while this value at P4 was significantly lower than at P2 (absolute R_{APT_P} > 10%). For APT_HP, P3 had relatively higher values than P2 in L2, L3, and L4 (R_{APT_P} > 15%), while P4 had significantly higher values in all layouts (absolute R_{APT_P} > 20%).

Figure 4 shows the variation of APT_LP, APT_MP, and APT_HP on the east façade with three positions (P2, P3, P4) and four layouts (L1, L2, L3, L4). For APT_LP (Figure 4a), L4 had the highest values at P2 and P3, while the other three layouts had similar values (relative difference < 10%). At P4, the same APT_LP of around 7% was found in all layouts. The range of APT_LP at the three positions in all layouts was 5~15%. For APT_MP (Figure 4b), like the APT_LP, the significantly higher values were achieved at P2 and P3 for L4 compared with the other layouts. The APT_MP for the four layouts had small difference (relative difference < 10%) at P4. For all positions, the four layouts showed an APT_MP range from 25% to 45%. For APT_HP (Figure 4c), L4 showed a much lower values than the other layouts at P2 and P3. A significantly higher APT_HP was achieved at P2 in L1. In addition, small differences of APT_HP were found at P4 between all layouts (relative difference < 10%). For the three positions, the range of APT_HP in all layouts was 24~50%. According to Figure 4, both values of APT_LP and APT_MP showed a decrease with the increasing height from P2 to P4, whilst a clear increase in APT_HP was found as P4 > P3 > P2.

Table 6. Relative differences of PPFD percentages (R_{APT_P}) of two façade positions (P3, P4), taking the position P2 as the reference (east façade, four layouts: L1, L2, L3, L4).

Annual PPFD Percentage	RAPT_P (%)
	L1		L2		L3		L4
	P3	P4	P3	P4	P3	P4	P3	P4
APTLP	−9	−36	−17	−42	−25	−42	−13	−53
APTMP	−3	−18	−11	−23	−11	−24	−12	−32
APTHP	5	29	21	44	21	45	33	100

indicates the relative differences of PPFD percentages at P3 or P4 (R_{APT_P}), taking P2 as the reference. In L2, L3 and L4, P3 and P4 received the significantly lower values of APT_LP and APT_MP than P2 (absolute R_{APT_P} > 10%), and the significantly higher values of APT_HP than P2 (R_{APT_P} > 20%). However, L1 cannot see big differences of APT_LP, APT_MP, APT_HP between P2 and P3 (absolute R_{APT_P} < 10%).

Figure 5 illustrates the variation of APT_LP, APT_MP, APT_HP on the west façade for the three positions (P2, P3, and P4) and the four layouts (L1, L2, L3, L4). For APT_LP (Figure 5a), at P2, L4 had higher values compared with other layouts, while other layouts achieved similar values of around 11%. At P3, there was a clear difference in APT_LP values only between L4 and L2 (L2 > L4, relative difference > 10%). All layouts had similar APT_LP values (around 9%) at P4. In addition, the range of APT_LP at the three positions of all layouts was 9~15%. For APT_MP (Figure 5b), at P2 and P3, the highest value was found in L4, while similar values were found in the other three layouts. In addition, no big differences in APT_MP at P4 could be found in any of the layouts (relative difference < 10%). The three positions in the four layouts show an APT_MP range from 25% to 40%. For APT_HP (Figure 5c), much lower values were found at P2 and P3 in L4 compared with the other layouts. At P4, the APT_HP values in the four layouts were similar, with a range of 46% to 48%. For the three positions in all four layouts, the range of APT_HP was from 25% to 50%. According to Figure 5, the values of APT_LP and APT_MP decreased with the increasing height from P2 to P4. The values of APT_HP showed an opposite trend (P4 > P3 > P2).

Table 7. Relative differences of PPFD percentages (R_{APT_P}) of two façade positions (P3, P4), taking the position P2 as the reference (west façade, four layouts: L1, L2, L3, L4).

Annual PPFD Percentage	RAPT_P (%)
	L1		L2		L3		L4
	P3	P4	P3	P4	P3	P4	P3	P4
APTLP	0	−18	−17	−25	−8	−25	−14	−36
APTMP	0	−22	−9	−24	−12	−21	−13	−33
APTHP	0	26	14	34	14	31	32	84

gives the relative differences of PPFD percentages at P3 or P4 (R_{APT_P}), taking P2 as the reference. For APT_LP, taking P2 as the reference, P4 achieved significantly lower values (absolute R_{APT_P} > 10%) for all layouts, while P3 showed significantly lower values (absolute R_{APT_P} > 10%) only for L2 and L4. For APT_MP, P4 had significantly lower values than P2 (absolute R_{APT_P} > 20%) for all layouts, while only L3 and L4 showed big differences between P2 and P3 (absolute R_{APT_P} > 10%). For APT_HP, P3 and P4 delivered significantly higher values than P2 for layouts of L2, L3, and L4 (R_{APT_P} > 10%). In addition, there was no difference in APT_HP between P2 and P3 for L1.

3.2. Effect of Layouts on PAR Availability at the Ground

Table 8. Annual percentages of time of three PPFD ranges (APT_LP, APT_MP, APT_HP) at the ground positions (P1) near four building façades (S, N, E, W) in four layouts (L1, L2, L3, L4).

Facade	Layout	APTLP (%)	APTMP (%)	APTHP (%)
S	L1	8	25	49
	L2	10	25	47
	L3	11	26	45
	L4	13	30	38
N	L1	9	33	40
	L2	9	37	35
	L3	9	37	35
	L4	12	44	24
E	L1	10	34	39
	L2	11	35	36
	L3	11	36	35
	L4	12	40	28
W	L1	10	30	41
	L2	11	32	38
	L3	11	31	39
	L4	13	36	31

indicates the values of APT_LP, APT_MP, APT_HP at the ground positions (P1) near the four building façades (S, N, E, W) in the four layouts (L1, L2, L3, L4). For APT_LP and APT_MP, L4 delivered relatively higher values than the layouts of L1, L2, and L3 at the ground of each façade, while L1, L2, and L3 did not show any big differences of the two values at the same positions. L1 delivered the largest APT_HP values at the ground near each façade, while the lowest values were for L4. L2 and L3 achieved the medium APT_HP values in between. In general, for all façades in the four layouts, the ranges of APT_LP, APT_MP, and APT_HP were 8~15%, 25~44%, and 24~49%, respectively.

3.3. Effect of Orientations on PAR Availability at the Building and the Ground

Taking the south façade or the ground near the south façade as the reference, the relative differences of PPFD percentages of the other three façades or the ground near the other three façades (${\mathrm{R}}_{\mathrm{A}\mathrm{P}\mathrm{T}\_\mathrm{O}}$) can be calculated by the equation:

$${\mathrm{R}}_{\mathrm{A}\mathrm{P}\mathrm{T}\_\mathrm{O}}={\displaystyle \frac{{\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{j}}-{\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{s}\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{h}}}{{\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{s}\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{h}}}}\times 100\mathrm{\%}$$

(3)

where ${\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{j}}$ is the value of APT_LP, APT_MP, or APT_HP at one specific façade position or the ground near the façade (i = LP, MP, HP, j = north, east, west), ${\mathrm{A}\mathrm{P}\mathrm{T}}_{\mathrm{i},\mathrm{s}\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{h}}$ is the value of APT_LP, APT_MP, or APT_HP at the south façade or the ground near the south façade (i = LP, MP, HP).

Table 9. Relative differences of APT_LP (R_{APT_O}) at three positions (P2, P3, P4) of three façades (N, E and W) in four layouts (L1, L2, L3, L4), taking the south façade as the reference.

Position	RAPT_O (%)
	L1			L2			L3			L4
	N	E	W	N	E	W	N	E	W	N	E	W
P2	9	0	0	8	0	0	−8	−8	−8	−6	−6	−13
P3	18	0	10	20	0	0	10	−10	10	0	0	−8
P4	10	−13	13	29	0	29	13	−13	13	29	0	29

displays the relative differences of APT_LP (R_{APT_O}) at three positions of three façades (north, east, and west). For L1, P2 does not show clear differences of APT_LP between the south façade and the other façades (R_{APT_O} < 10%), while both P3 and P4 had higher APT_LP for the north and west façades compared to the south façade (R_{APT_O} ≥ 10%). Compared with the south façade, the east façade achieved the same APT_LP values at P2 and P3, but a much lower APT_LP value at P4 (absolute R_{APT_O} > 10%). For L2, no clear differences in APT_LP could be found at P2 for all of the façades, at P3 between the south, east, and west façades, and at P4 between the east and south façades (R_{APT_O} < 10%), while P3 and P4 showed significant differences of APT_LP between the south and north façades, and P4 had big differences in APT_LP between the west and south façades (R_{APT_O} > 20%). For L3, at P2 and P3, other façades showed small differences in APT_LP compared with the south façade (R_{APT_O} ≤ 10%), while relatively bigger differences of APT_LP were found between the south façade and the other façades (absolute R_{APT_O} > 10%). For L4, significant differences of APT_LP were found at P2 between the south and west façades, at P4 between the north and south façades, and at P4 between the west and south façades (absolute R_{APT_O} > 10%).

Table 10. Relative differences of APT_MP (R_{APT_O}) at three positions (P2, P3, P4) of three façades (N, E and W) in four layouts (L1, L2, L3, L4), taking the south façade as the reference.

Position	RAPT_O (%)
	L1			L2			L3			L4
	N	E	W	N	E	W	N	E	W	N	E	W
P2	9	0	−3	23	13	10	29	19	10	13	8	5
P3	19	3	3	21	7	7	28	14	3	24	9	6
P4	23	4	−4	23	4	0	27	8	4	26	4	0

gives the relative differences of APT_MP (R_{APT_O}) at three positions of three façades (north, east, and west). For L1, significant differences of APT_MP were found at P3 and P4 between the north and south façades (R_{APT_O} > 15%). For L2 and L3, all positions at the north façade had significant differences of APT_MP from the south façade (R_{APT_O} > 20%), while no big differences of APT_MP were found at all positions between the west and south façades (R_{APT_O} ≤ 10%). On the east façade, the P2 in both L2 and L3 and the P3 in L3 displayed relatively big differences of APT_MP when compared with the south façade (R_{APT_O} > 10%). For L4, significant differences in APT_MP were achieved at all positions of the north façade only (R_{APT_O} > 10%).

Table 11. Relative differences of APT_HP (R_{APT_O}) at three positions (P2, P3, P4) of three façades (N, E and W) in four layouts (L1, L2, L3, L4), taking the south façade as the reference.

Position	RAPT_O (%)
	L1			L2			L3			L4
	N	E	W	N	E	W	N	E	W	N	E	W
P2	−11	0	0	−21	−11	−8	−26	−13	−8	−16	−4	0
P3	−20	−2	−7	−19	−5	−7	−23	−7	−7	−24	−6	−3
P4	−15	2	0	−16	0	−4	−18	−2	−6	−19	0	−4

shows the relative differences of APT_HP (R_{APT_O}) at the three positions of three façades. Compared with the south façades, the north façades in all layouts showed significant differences of APT_HP at three positions (absolute R_{APT_O} > 10%), while the west façades in all layouts and the east façades in L1 and L4 had no big differences in APT_HP at three positions (absolute R_{APT_O} < 10%). For L2 and L3, the east façades gave relatively big differences in APT_HP at P2 only (absolute R_{APT_O} > 10%).

Table 12. Relative differences of APT_LP, APT_MP, APT_HP (R_{APT_O}) at the ground position (P1) near three façades (N, E, W) in four layouts (L1, L2, L3, L4), taking the ground position (P1) near the south façade as the reference.

Annual PPFD Percentage	RAPT_O (%)
	L1			L2			L3			L4
	N	E	W	N	E	W	N	E	W	N	E	W
APTLP	13	25	25	−10	10	10	−18	0	0	−8	−8	0
APTMP	32	36	20	48	40	28	48	44	24	47	33	20
APTHP	−18	−20	−16	−26	−23	−19	−26	−26	−17	−37	−26	−18

demonstrates the relative differences of APT_LP, APT_MP, and APT_HP (R_{APT_O}) at the ground position (P1) near three façades (north, east, west), taking the ground position (P1) near the south façade as the reference.

• APT_LP: three façades in L1 and the north façade in L3 had significant differences of APT_LP compared with the south façade (absolute R_{APT_O} > 10%), while no big differences of APT_LP was found at any of the façades in L2 and L4, and at the east and west façades in L3 (absolute R_{APT_O} ≤ 10%).
• APT_MP: the four layouts achieved significantly higher differences of APT_MP between the south façade and each of other three façades (R_{APT_O} ≥ 20%).
• APT_HP: the four layouts showed relatively higher differences of APT_HP between the south façade and each of the other three façades (absolute R_{APT_O} ≥ 15%).

4. Discussion

Based on the results mentioned above (Section 3.1, Section 3.2 and Section 3.3), several key findings can be drawn and discussed as follows.

First, for the top floor of all building façades, no significant impact of layout was found on the PAR availability with the three PPFD ranges (LP, MP, HP). This could indicate that the layout would not influence the selection of plant type at the top floor. According to Figure 6 and Figure 7, at this floor (P4), all layouts had similar low-level obstructions, indicating similar daylight levels (mainly direct sunlight and skylight).

Second, for the middle and ground floors of all building façades and the external ground surfaces, the urban layout L4 achieved the relatively higher PAR availability with MP and LP and the lower PAR availability with HP than the other layouts. However, these positions generally showed small differences of all types of PAR availability between the three urban layouts: L1, L2, and L3. As shown in Table 1 and Figure 6 and Figure 7, L4 had the largest ratio of building height to building spacing and the highest level of obstruction at middle and low façade positions and the ground. Therefore, compared with the other layouts, L4 may deliver more indirect light and less direct light at these positions, resulting in a relatively lower daylight level.

Third, for all building façades in each layout, increasing the façade height reduced the PAR availability with MP and LP but increased the PAR availability with HP. Consequently, the façade height may determine the selection of plant type. Given Figure 6 and Figure 7, when moving from the ground to the top floor, the deceasing level of obstruction can clearly increase the total daylight level, resulting in more opportunities to apply the high-light plants.

Fourth, significant effects of orientation can be found for all types of PAR availability at various positions in all layouts. Compared with the south, east and west façades, the north façade generally received the higher PAR availability with LP and MP and the lower PAR availability with HP. However, there were generally no big differences for all types of PAR availability between the east and west façades for all layouts. At the ground, the south façade achieved the higher PAR availability with HP and the lower PAR availability with MP than other three façades. Given Figure 6 and Figure 7 and the location of Shanghai, the sky can deliver significantly higher direct daylight levels on the south façade than the north façade, even though they have the same obstructions, while similar obstructions of east and west façades would give rise to similar daylight levels.

Finally, for the four layouts, the PAR availability with LP was generally smaller on all façade and ground positions than the PAR availability with MP and HP, while the top floor saw a higher level of the PAR availability with HP than the PAR availability with MP. As shown in Figure 6 and Figure 7, all layouts generally had low and medium levels of obstructions across the building façades and at various ground locations. In addition, the location and climate conditions in Shanghai (humid subtropical climate) can lead to a high level of solar availability over a year [32,33,34,35]. These factors may help create more opportunities to apply the medium-light and high-light plants in these layouts.

Following the findings above,

Table 13. Recommendations of the rank of plants (HLP, MLP, LLP) which can be applied at four positions (P1, P2, P3, P4) in terms of façade orientation (S, N, E, W) and layouts (L1, L2, L3, L4).

Layout	Facade	Position
		P1	P2	P3	P4
L1	S	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
	N	HLP > MLP > LLP	HLP = MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP
	E	HLP > MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
	W	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
L2	S	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
	N	HLP = MLP > LLP	MLP > HLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP
	E	HLP = MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
	W	HLP > MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
L3	S	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
	N	HLP = MLP > LLP	MLP > HLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP
	E	HLP = MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
	W	HLP > MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP	HLP > MLP > LLP
L4	S	HLP > MLP > LLP	MLP > HLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP
	N	MLP > HLP > LLP	MLP > HLP > LLP	MLP > HLP > LLP	HLP > MLP > LLP
	E	MLP > HLP > LLP	HLP > MLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP
	W	MLP > HLP > LLP	MLP > HLP > LLP	HLP = MLP > LLP	HLP > MLP > LLP

Note: HLP: high-light plants, MLP: medium-light plants, LLP: low-light plants (see Table 3). ‘A > B’ means that A has the priority. ‘A = B’ means that both A and B can be applied at this position and no one has the priority.

presents recommendations for the rank of plant types (HLP, MLP, LLP) which could be applied at various façade and ground positions in the four layouts (L1, L2, L3, L4). For all layouts, HLP and LLP have the highest and the lowest potential to be applied at façade and ground positions, respectively.

The limitations of this study include the following: (1) this study was conducted in Shanghai, which has a special humid subtropical climate and specific urban planning regulations. The findings may not be generalizable to cities with difference climates or urban layouts. (2) The study examined only four typical urban layouts with buildings having uniform heights and orientations. Real-world urban environments have complex and mixed layouts and irregular building geometries, which were not considered. (3) Values of reflectance of building surfaces and ground were fixed as 0.4 and 0.2, respectively, which may not account for variations in real-world materials.

Suggested future work might include the following: (1) more analysis of cities with various climates and urban layouts to improve the generalization of the findings. (2) Validation of simulation results with field measurements in residential areas and PAR availability in hybrid layouts, combining high- and low-rise buildings, could be further explored. (3) The impact of real reflectance values of various materials of façades on PAR availability should be investigated.

5. Conclusions

Using advanced ray-tracing simulation software (Radiance) and local weather data, this study presents a dynamic analysis of PAR availability in a very dense residential area in Shanghai city, China. Effects of layout, façade orientation and façade height on PAR availability in buildings were analyzed using a metric system developed based on the type of outdoor plants. Three ranges of PAR availability (PPFD) for sustaining the growth of plants were applied: low-light plants (9.95–49.76 µmoles/m²/s), medium-light plants (49.76–199.06 µmoles/m²/s), and high-light plants (>199.06 µmoles/m²/s). Four typical urban layouts were assessed (layouts 1–4). Key findings included the following: (1) compared with layouts (1, 2, 3), layout 4 achieved the higher levels of PAR availability for low/medium-light plants and the lower levels of PAR availability for high-light plants at middle and low building façades and on the ground, while high building façades for all layouts can see similar PAR availability for all plants. (2) The PAR availability for low/medium-light plants decreased with the increasing façade height, while the PAR availability for high-light plants showed an opposite trend. (3) North façades and the ground had the higher levels of PAR availability for low/medium-light plants and the lower levels of PAR availability for high-light plants compared with the south, east, and west façades. (4) All layouts may provide a higher potential to apply high-light and medium-light plants at façades and on the ground than the low-light plants.