Effectiveness of Cool and Green Roofs Inside and Outside Buildings in the Brazilian Context

Abstract

Several studies have assessed the thermal performance of green and cool roofs. However, few have comprehensively addressed Brazilian buildings and climates, considering indoor and outdoor environments. Considering three Brazilian cities, this study aims to assess the performance of green and cool roofs compared with traditional fibre cement roofs in a typical multifamily residential building. Energy consumption, thermal comfort, and outside surface temperature were assessed using computer simulation. The results show that the cool roofs performed better in cities with warmer climates (e.g., Cfa and Aw), reducing electricity consumption by up to 24.8% compared with traditional roofs. Green roofs are better suited for colder climates (e.g., Cfb), with up to 28.2% energy savings. Green roofs provided the highest percentage of thermal comfort hours in all climates. Cool and green roofs provided hourly reductions in outside roof surface temperature of up to 16.5 °C and 28.4 °C, respectively, compared with the traditional roof. This work reinforces that the choice between these two roof types for each city depends on the parameter used for comparison. Based on the relevant information applied to Brazilian buildings and representative climates presented, this work provided recommendations for urban planning policies and building regulations in Brazil.

1. Introduction

Buildings are responsible for a large proportion of energy consumption worldwide. Globally, the buildings sector is responsible for around 30% of final energy consumption [1]. In Brazil, the residential sector was responsible for 27% of electricity consumption in 2022 [2]. In this sense, research aimed at reducing energy consumption is relevant worldwide. Due to the high concentration of buildings in the urban environment, the envelope characteristics play a fundamental role in the urban thermal balance. They absorb solar and infrared radiation and dissipate part of the accumulated heat, increasing the ambient temperature [3]. These impacts can be mitigated by implementing passive strategies. In particular, the roof surfaces play an important role—they represent around 20% to 25% of urban surfaces and are characterised by the highest viewing factor for incident solar radiation [4,5].

The literature shows that passive solutions such as green and cool roofs can be effective as a response to increasing urbanisation and climate change and can reduce the energy consumption of buildings [6,7,8,9] and the urban heat island effect [10,11], in addition to increasing human well-being [12,13]. The thermal performance of green roofs is mainly associated with physical processes in the vegetation and substrate layer [14]. Cool roofs are characterised by high solar reflectance, which reduces the solar radiation absorbed by the roof surface, and high thermal emittance, which helps dissipate the heat absorbed [3]. However, it should be noted that the thermal performance of strategies implemented in the envelope is significantly affected by the climate and the characteristics of the building envelope.

Green roofs can reduce cooling loads by up to 70% and lower indoor temperatures by up to 15 °C while providing environmental benefits such as pollutant reduction and noise mitigation [15]. Additionally, green roofs have been shown to reduce heat flow into buildings by up to 93% compared with traditional roofs [16]. Cool roofs are usually more effective in summer, reducing outside roof temperatures by an average of 3.3 °C, whereas green roofs offer superior winter insulation [17]. Cool roofs also achieve significant energy savings across different climates, with average reductions between 15.0% and 35.7%, demonstrating their versatility and effectiveness [18].

Brazil currently has a population of around 200 million inhabitants and more than 90 million households [19]. The country also has a wide variety of climates and buildings with different envelopes (due to differences in materials and construction processes) compared with buildings in the USA, Europe, and Asia, where most research is carried out [20].

Although several studies have comparatively evaluated the thermal performance and thermal comfort parameters of green and cool roofs, few studies consider the Brazilian context [21,22,23] and present a broad assessment of buildings’ inside and outside environments. Given this, there is little information on the performance of green and cool roofs in Brazilian climates and buildings, especially considering representative models of actual buildings rather than prototypes or models with generic geometries. One of the main barriers to the use of green roofs, especially in developing countries such as Brazil, is that policymakers still fail to realise the positive aspects of such roofs [20]. Without proper government support, implementing cool and green roofs is challenging for the general public. In addition, most international studies compare the thermal and energy performance of green and cool roofs with traditional slab roofs with a waterproof membrane [8,13,24]. In contrast, fibre cement roofs are commonly used in Brazilian buildings, accounting for around half of the national roof tile market [25]. Since each country has different climatic and urbanisation conditions, local research is essential for implementing passive roofing solutions. Furthermore, although several standards encourage the use of thermal insulation, it is not a common practice in Brazil [26]. Therefore, it should be analysed considering the climatic context and the building’s use pattern, since its adoption can lead to room overheating.

In order to fill these gaps, this work aims to evaluate the performance of green and cool roofs compared with traditional fibre cement roofs on a typical Brazilian residential building in three climatic contexts through computer simulation. The three roofs are investigated with and without thermal insulation to address the inside and outside environments of the building. For this purpose, the inside environment of the building was assessed in two scenarios: with air-conditioning through energy consumption and with natural ventilation through the percentage of hours of thermal comfort. The outside environment was evaluated using the outside surface temperature. This work presents relevant information applied to Brazilian buildings and representative climates. In addition, this paper provides recommendations for urban planning policies and building regulations in Brazil and subsidies to designers to choose more sustainable roof typologies in terms of energy consumption.

2. Materials and Methods

Initially, a building model representative of Brazilian social housing was chosen. Then, three different climatic contexts were selected to evaluate the performance of green and cool roofs compared with traditional roofs, with and without thermal insulation. The efficiency of the roofs was assessed through computer simulation using the EnergyPlus programme, based on the energy consumption for air-conditioning, the thermal comfort of occupants, and the outside surface temperature of the roofs. EnergyPlus is a valid programme, and it is commonly used to simulate the thermal performance of buildings [27]. One of the advantages of simulation is that it allows testing different construction typologies in different climatic contexts during the building design (before construction). In addition, other work has demonstrated the validity of simulating green roofs in the Brazilian context [21,28].

Figure 1 shows a schematic of the methodology used in this work.

2.1. Case Studies

2.1.1. Building Model

The building typology used is a five-storey multifamily residential building representative of Brazilian social housing projects [29]. The choice of a multifamily residential model was because this type of building needs more energy for air-conditioning than single-family houses [4]. In addition, multifamily buildings are more common in large urban centres, where the heat island phenomenon is more pronounced. Figure S1 of the Supplementary Material shows a perspective view of the building and the floor plan of the flat used in this study. The building has five floors with a floor-plan area of 444.94 m² each, and it was simulated in an open field, i.e., without considering shading from other urban elements. The analyses were carried out in a flat on the top floor of the building, where the influence of the roof is most noticeable [4,30,31]. The flat has a floor-plan area of 38.6 m².

The input parameters defined include the geometric characteristics surveyed in [29] and are shown in Table S1 in the Supplementary Material. The internal loads due to the presence of people, the lighting system, and the equipment were defined according to Tables S2–S4 in the Supplementary Material. Figure S2 of the Supplementary Material shows the occupancy schedule for the long-term rooms (living room and bedrooms). The model was evaluated in two scenarios: natural ventilation and air-conditioning.

2.1.2. Climatic Contexts

The TMY climate files of 2007–2021 [32] of the Brazilian cities of Florianópolis, Curitiba, and Brasília were used. The air temperature, incident radiation, relative humidity, and rainfall are shown in Figure 2. Figure 3 shows the climate classification of Brazil and the location of the cities considered in this study.

These cities were selected to investigate dry and humid climatic conditions and temperate and hot summer situations. Furthermore, the climates of the selected cities encompass a substantial share of the national territory and represent distinct Brazilian bioclimatic zones. Florianópolis has a humid subtropical climate with no dry season and hot summers (Cfa), Curitiba has a humid subtropical climate with no dry season and temperate summers (Cfb), and Brasília has an arid tropical climate with dry winters (Aw) [33]. According to the NBR 15220-3 standard [34], the climate classification of Florianópolis, Curitiba, and Brasília is bioclimatic zones 3, 1, and 4, respectively.

2.1.3. Roof Typologies

Three typologies of roofs were analysed: traditional fibre cement, cool, and extensive green roof. The structural support for all roof typologies was a 12 cm concrete slab. All roof typologies were assessed in two conditions: without and with thermal insulation (expanded polystyrene). Figure S3 of the Supplementary Material shows the roof layers and the abbreviations used to identify them.

A traditional fibre cement roof (made up of synthetic fibres and cement) was considered since this roof typology is commonly used in Brazilian buildings due to its easy handling and low cost. Tables S5 and S6 of the Supplementary Material show the data of the traditional roof used in the simulation. The cool roof was considered a white fibre cement roof. The data used to simulate the cool roof was the same as that used for the traditional roof (Tables S5 and S6 of the Supplementary Material).

The Sailor model [35] implemented in Energyplus was used to model the green roof (vegetation and substrate layers). An extensive green roof was adopted, given its lower weight (usually between 49 and 98 kg/m²) [14,36,37,38]. Table S7 of the Supplementary Material shows the input data for the vegetation and substrate layers (ecoroof) considered in the simulations. The EnergyPlus default data were adopted for the remaining parameters. Irrigation was not considered in the simulations to analyse the response of green roofs in each climate. The other layers of the green roof were modelled separately. The filter, drainage, and root barrier layers are usually polymeric materials [39]. A geotextile fabric was considered as a filter [37], high-density polyethylene (HDPE) as drainage [40], and low-density polyethylene (LDPE) as a root barrier [40,41]. For the waterproofing layer, a bituminous membrane was considered. The data for these layers are shown in Table S8 of the Supplementary Material.

2.2. Computer Simulation

2.2.1. Energy Consumption for Air-Conditioning

Using the Ideal Loads Air System object class, the following output variables were obtained: annual thermal loads for cooling and annual thermal loads for heating. The setpoints adopted for the thermostat were 21 °C for heating and 26 °C for cooling. These setpoints were determined based on NBR 16401-2 [42]. The range between 21 °C and 26 °C is also commonly used in studies assessing energy needs for air-conditioning [36]. The ideal loads air system was only activated during occupancy periods in the long-term rooms (living room and bedrooms). It was considered that all windows and doors remain closed at all times (with infiltration occurring through gaps), except for the bathroom window, which remains open at all times, following NBR 15575-1 [43].

The result of the annual thermal loads (cooling and heating) in the long-term rooms was used to calculate the annual energy consumption needed for air-conditioning. The annual energy consumption was obtained by dividing the thermal loads by the energy efficiency coefficient of the air-conditioning system [44]. In the case of cooling, this coefficient was considered equal to 5.5, the minimum for the air-conditioner to have the highest level of energy efficiency in Brazil [45]. As for heating, the coefficient used was 4.47, calculated according to the energy efficiency coefficient for cooling recommended in [44].

The evapotranspiration of the green roofs was also presented to complement the analysis. EnergyPlus estimates this process by evaluating the latent heat fluxes in the vegetation and substrate layers.

2.2.2. Thermal Comfort

The thermal comfort of the occupants was assessed using the variable thermal comfort hours in the long-term rooms (living room and bedrooms) during occupancy periods. It was assumed that the occupants would accept the adaptive thermal comfort conditions of ASHRAE 55 [46]. The internal operative temperatures in the rooms and the average daily outside air temperatures (dry bulb) were obtained from the simulation of the building in the scenario with natural ventilation (Air Flow Network).

The average outside temperature was calculated using the arithmetic mean of the average daily outside air temperatures (dry bulb) over the last seven days and was used to obtain the operative temperature ranges for adaptive comfort, considering 80% satisfaction. The lower and upper limit temperatures were determined using Equations (1) and (2) [46], respectively. Above the upper limit temperature, occupants would feel discomfort from heat, and below the lower limit temperature, they would feel discomfort from cold.

$${\mathrm{L}\mathrm{L}}_{80\mathrm{\%}\mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}=0.31 \times {\mathrm{t}}_{\mathrm{a}\mathrm{o}\mathrm{u}\mathrm{t}}+14.3$$

(1)

$${\mathrm{U}\mathrm{L}}_{80\mathrm{\%}\mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}=0.31 \times {\mathrm{t}}_{\mathrm{a}\mathrm{o}\mathrm{u}\mathrm{t}}+21.3$$

(2)

where LL80%acceptability is the temperature of the lower limit of the comfort zone (°C); UL80%acceptability is the temperature of the upper limit of the comfort zone (°C); and ${\mathrm{t}}_{\mathrm{a}\mathrm{o}\mathrm{u}\mathrm{t}}$ is the average outside temperature (°C).

Once the temperature ranges in which the occupants feel thermal comfort were obtained (values between the lower and upper limits), the number of occupied hours within this range was verified, and the percentage of thermal comfort hours was defined.

Natural ventilation was considered during occupancy periods in the long-term rooms (see Section 2.1.1). The bathroom window was considered open at all times, and the windows in the other transient rooms were considered closed at all times. Internal doors were considered open, except for the bathroom door, which is permanently closed. The external door is permanently closed. In the case of closed windows and doors, infiltration occurs through gaps following NBR 15575-1 [43].

2.2.3. Outside Surface Temperature

The outside surface temperature was used to assess the influence of each roof typology in terms of urban heat islands. In the case of green roofs, the outside surface temperature was calculated considering the average of the vegetation and substrate temperatures, according to [47].

2.3. Results Analysis

The electricity used to heat and cool the long-term rooms (living room and bedrooms) was the parameter used to assess which roof typology provides the best conditions for the inside environment in the scenario with air-conditioning. The percentage of occupied hours in which there is thermal comfort was used to investigate the influence of roof typologies on the well-being of the occupants in the inside environment with natural ventilation. Finally, the outside roof surface temperature was used to assess how the roof typologies affect the outside environment. For each parameter, the performance of the cool and green roofs was simulated for one year and compared with the performance of the traditional roof. Complementary analyses were carried out using the evapotranspiration of the green roofs over the year.

3. Results and Discussions

3.1. Energy Consumption for Air-Conditioning

Figure 4 shows the annual electricity consumption for heating and cooling the building per square metre of air-conditioned room area. In Florianópolis, a city with a humid climate throughout the year and hot summers, the most efficient solutions were the cool and green roofs without insulation, reducing the annual electricity consumption by 16.7% and 14.7%, respectively, compared with the traditional roof without insulation. In Curitiba, which has the coldest climate among the cities evaluated, there is higher electricity consumption for heating and lower electricity consumption for cooling than in Florianópolis. In this climatic context, the energy consumption for air-conditioning was also lower when using cool and green roofs (a reduction of 22.3% and 28.2% compared with the traditional roof without insulation, respectively). Both typologies reduced cooling consumption by about 30% compared with the traditional roof. However, the cool roof increased heating consumption by 26.3% due to reduced solar heat gain through the roof. In comparison, the green roof reduced this consumption by 26.5% due to the insulation provided mainly by the substrate layer. In Brasília, characterised by a dry climate with mild temperatures even during winter, electricity consumption for heating can be considered null. Cool and green roofs showed the best performance, reducing annual electricity consumption by 24.8% and 21.0%, respectively, compared with traditional roofs without insulation.

Considering the typology and envelope of the building adopted in this study and the characteristics of the climates analysed, it was observed that the cool and green roofs without insulation show the best energy performance.

Regarding thermal insulation, in all the roof typologies and cities evaluated, insulation reduced the energy needed for heating when it existed. The reductions were up to 67.4% in Florianópolis and 51.6% in Curitiba, depending on the roof typology. On the other hand, on hot days, the addition of an insulation layer, despite reducing the heat flow from the roof to the interior of the building during the day, made nighttime cooling difficult, i.e., it makes it difficult for internal heat—generated by people, equipment, and lighting—to escape on summer evenings. In the traditional roof, insulation had minimal impact on the energy needed to cool the flat (variations between −4.2% and 1.3% in energy consumption, depending on the climatic context). In cool and green roofs, the insulation increased the energy needed for cooling (up to 21.2%, depending on the location.

Cool roofs reflect more heat than traditional roofs, so they retain less heat to dissipate at night. Insulation further limits any remaining heat or internal heat from escaping. Traditional roofs also benefit from the reduction in heat loss due to insulation. However, they retain more heat, allowing greater dissipation during cooler periods. In green roofs, the vegetation and substrate absorb and store solar heat during the day. When the roof is insulated, this heat cannot easily dissipate into the flat’s interior at night when temperatures are lower. Consequently, the accumulated heat may remain in the roof system, influencing the flat’s internal temperature even after the peak external heat has passed. This effect can lead to an increase in cooling load.

Thus, considering the annual electricity consumption for air-conditioning in multifamily residential buildings, thermal insulation is not recommended in Florianópolis and Brasília. In these locations, insulation was irrelevant when applied to the traditional roof and increased electricity consumption by 10.1% and 20.6% when considered in cool and green roofs. In Curitiba, insulation is recommended only for traditional roofs (reduced electricity consumption by 10.6%). In all climatic contexts, even when thermal insulation was considered, the traditional roof performed worse than cool and green roofs.

The results found when adding insulation to the green roof are similar to those observed by Vera et al. [9]. On the other hand, Silva et al. [8] reported that thermal insulation reduced the energy needed for cooling in extensive and semi-intensive green roofs but increased it in intensive green roofs, which have higher LAI and vegetation height. However, in our study, the LAI and vegetation height have higher values compared with those considered by the mentioned authors for extensive and semi-intensive green roofs, supporting that isolated green roofs with higher LAI and vegetation height do not fully take advantage of cooling and evaporation effects.

Figure S4 of the Supplementary Material shows the annual evapotranspiration results of the green roof in Florianópolis, Curitiba, and Brasília. Annually, the green roof released 1459 mm of water in Florianópolis, 1281 mm in Curitiba, and 2003 mm in Brasília through evaporation (substrate) and transpiration (vegetation). Evapotranspiration is one of the main heat dissipation mechanisms in green roofs, and the higher value in Brasília corroborates the high incidence of solar radiation, high annual temperatures, and low relative humidity of the city (see Section 2.1.2) [48,49]. These results help to explain the greater reduction in energy consumption for air-conditioning when using the green roof (compared with the traditional roof) in Brasília compared with Florianópolis. On the other hand, despite the lower evapotranspiration in Curitiba, the city showed the highest reduction in energy consumption for air-conditioning compared with the traditional roof. In nonirrigated green roofs, the amount of water for evapotranspiration is limited, especially in hot climates. In this case, the outside temperature plays an important role, and colder cities can show greater energy savings, as reported by Borràs et al. [50].

3.2. Thermal Comfort

Figure 5 shows the percentage of annual thermal comfort and thermal discomfort hours due to heat and cold in long-term rooms. Tables S9–S11 of the Supplementary Material show the cold and heat discomfort and comfort hours in each season in Florianópolis, Curitiba, and Brasília, respectively.

Considering the context of Florianópolis and roofs without insulation, the case with a green roof showed the highest percentage of thermal comfort hours (71.2% in the living room, 78.8% in bedroom 1, and 78.0% in bedroom 2). The case with a traditional roof resulted in the lowest percentage of thermal comfort hours (59.3% in the living room, 75.8% in bedroom 1, and 73.3% in bedroom 2). The case with the cool roof had an intermediate performance (67.9% of thermal comfort hours in the living room, 76.9% in bedroom 1, and 75.8% in bedroom 2) because although it reduced the discomfort hours due to heat, it increased the discomfort hours due to cold compared with the traditional roof. This behaviour is consistent with the higher solar reflectance [51,52]. Using a layer of insulation on the roof provided a slight increase (up to 4.3%) in the percentage of annual thermal comfort hours, except in the living room when considering a cool or green roof. This room is occupied during the afternoon and evening (14:00 to 21:59), and heat discomfort is prevalent. In the bedrooms, predominantly occupied at night (22:00 to 07:59), where cold discomfort predominates or the cold discomfort hours are similar to the heat discomfort hours, the insulation in all the roofs increased the annual thermal comfort hours.

Among the roof typologies without thermal insulation in Curitiba, the green roof also provided the highest percentage of annual thermal comfort hours in long-term rooms (82.2% in the living room, 69.8% in bedroom 1, and 75.6% in bedroom 2). The cool roof increased the percentage of annual thermal comfort hours in the living room from 68.0% to 77% but reduced the percentage of thermal comfort hours from 67.7% to 64.1% in bedroom 1, and from 72.8% to 69.5% in bedroom 2, compared with the case with the traditional roof. This is because, in Curitiba, discomfort in bedrooms is mainly attributed to cold conditions; in this scenario, the cool roof does not provide an advantage over the traditional roof. In this climatic context, adding insulation to the roofs resulted in a more significant increase (up to 10.4%) in the percentage of annual thermal comfort hours, except for the living room when considering green roofs. In all roof typologies, the insulation layer helped to reduce the hours of discomfort due to cold but increased the hours of discomfort due to heat.

In Brasília, considering roofs without insulation, the highest percentages of thermal comfort hours were observed for the green roof scenario (57.3% in the living room, 96.3% in bedroom 1, and 96.1% in bedroom 2). Slightly lower percentages of thermal comfort hours were found for the cool roof scenario (56.8% in the living room, 94.3% in bedroom 1, and 94.9% in bedroom 2). The traditional roof scenario showed the lowest percentages of thermal comfort hours, particularly concerning comfort in the living room. Annually, the percentages of thermal comfort with the traditional roof were 33.3% in the living room, 91.3% in bedroom 1, and 89.4% in bedroom 2. The thermal insulation increased the percentage of annual thermal comfort hours in all the long-term rooms in the case of the traditional roof (up to 4.8%) but decreased or had minimal impact in the cases of the cool and green roofs.

Green roofs provided the highest number of hours of thermal comfort in all cities in all seasons, owing to their high thermal inertia, which effectively moderates daily temperature fluctuations, corroborating literature [53,54]. In Florianópolis and Curitiba, cool roofs outperformed traditional ones during the summer, and they may also offer benefits in spring and autumn in rooms where heat discomfort is prevalent. In Brasília, where higher temperatures prevail year-round, cool roofs consistently showed strong performance across all seasons.

The thermal insulation layer reduces cold discomfort, ensuring extended thermal comfort hours during autumn and winter. However, it may increase heat discomfort due to the rooms overheating and the difficulty of dissipating heat to the external environment, especially in rooms where heat discomfort is more significant (as is the living room in this study). As a result, insulation generally impairs thermal comfort during the summer, spring, and autumn. In Brasília, insulation reduces the number of hours of thermal comfort even during the winter, a season also marked by considerable heat discomfort. Thermal insulation in the traditional roof increased the percentage of thermal comfort hours in long-term rooms in the three climatic contexts analysed. For cool and green roofs, the impact of insulation varied; in some cases, it enhanced occupants’ thermal comfort, while in others, it reduced thermal comfort. Overall, with the use of cool and green roofs, it was observed that insulation reduces the percentage of thermal comfort hours in the rooms where thermal discomfort occurs predominantly due to heat (in this case, the living room) and in climatic contexts where thermal discomfort is predominantly due to heat (the climate of Brasília).

3.3. Outside Surface Temperature

The maximum outside roof surface temperatures were recorded at different hours, on different days, or even in different months, depending on the roof typology. However, when comparing the roof typologies, the behaviour pattern was similar throughout the year. Figure 6 shows the temperature profile of the outside surface of the roofs in the three climatic contexts over three summer days, considering hourly data. At hours with the highest solar radiation, the traditional roof showed the highest outside surface temperature, followed by the cool roof, in all climate contexts. Insulation increased outside surface temperatures in the case of traditional and cool roofs and was irrelevant in the case of green roofs.

Considering the hourly outside surface temperatures of roofs without insulation over a year, the traditional roof reached the highest outside surface temperatures (maximum hourly temperatures were between 61.9 °C and 65.5 °C, depending on the location). In comparison, the green roof showed the lowest temperatures (maximum hourly temperatures between 39.1 °C and 41.4 °C). The cool roof showed intermediate outside surface temperatures, with maximum hourly temperatures ranging from 47.8 °C to 49.9 °C. Figure 7 shows the temperature profile of the traditional, cool, and green roofs without insulation over a year for Florianópolis as an example. The results corroborate with Gagliano et al. [13,37], who considered the solar reflectance of the cool roof 0.55—a value similar to the one adopted in this study (0.58)—and the vegetation LAI equal to 5.0—a value even higher than the one adopted in this study (3.0).

Using an insulation layer led to higher maximum hourly temperatures on the traditional (between 68.3 °C and 72.7 °C, depending on the location) and cool roofs (between 52.6 °C and 54.8 °C) and did not influence the outside surface temperature of the green roofs (between 38.8 °C and 41.2 °C). The increase in thermal resistance in the traditional and cool roofs hindered heat transfer through the roof in hot periods, raising the temperature of the outer surface of the roofs. Green roofs benefit from the shading and reflective effects of vegetation [13,37].

Figure 8 shows the differences between the outside surface temperatures of the cool and green roofs compared with the traditional roof without insulation, considering the average daily temperatures. The cool roof was able to reduce the average daily surface temperature by up to 4.7 °C in Florianópolis, 5.9 °C in Curitiba, and 5.3 °C in Brasília. The differences were more significant in the case of the green roof, which reduced the average daily surface temperature by 7.4 °C, 10.1 °C, and 9.5 °C in Florianópolis, Curitiba, and Brasília, respectively.

3.4. Summary of Results

Table 1. Summary of the results of energy consumption for air-conditioning, thermal comfort, and outside surface temperature for traditional, cool, and green roofs, with and without insulation, in Florianópolis, Curitiba, and Brasília.

Parameter	Roof Typology	City
		Florianópolis	Curitiba	Brasília
Heating and cooling energy in long-term rooms (kWh/m2.year)	TR00	15.0	9.1	18.1
	TR04	15.0	8.1	18.2
	CR00	12.5	7.0	13.6
	CR04	14.0	7.2	16.4
	GR00	12.8	6.5	14.3
	GR04	14.1	7.1	16.6
Thermal comfort (average percentage of annual thermal comfort hours in long-term rooms)	TR00	69.4	69.5	71.3
	TR04	72.2	76.0	72.8
	CR00	73.5	70.2	82.0
	CR04	74.1	76.1	77.2
	GR00	76.0	75.9	83.3
	GR04	74.7	77.6	77.5
Peak temperature (hourly) of the outside roof surface (°C)	TR00	62.8	65.5	61.9
	TR04	68.3	72.7	68.4
	CR00	49.1	49.9	47.8
	CR04	52.9	54.8	52.6
	GR00	41.4	39.1	40.5
	GR04	41.2	38.8	40.7

The data highlighted in blue represent the results of the roofs with the best performance in each of the parameters assessed.

summarises the results of the parameters used to evaluate the roofs in the three cities.

It can be seen that from the perspective of energy consumption for heating and cooling, the cool roof without insulation showed the best performance in warmer climates—Florianópolis (Cfa) and Brasília (Aw). The cool roof keeps the surface cooler than the traditional roof and radiates the stored heat, especially at night, reducing heat flow into the building, consistent with other studies [8,24]. On the other hand, in Curitiba, the coldest city (Cfb), the green roof without insulation provided the lowest energy consumption. These results corroborate the findings of Balvedi and Giglio [21], who considered a building with a generic geometry in Brazilian cities (climates Cfa, Aw, and Am). The authors reported the best performance of cool (reflectance equal to 0.8) and green (LAI equal to 3) roofs compared with the traditional fibre cement roof in terms of energy consumption for air-conditioning, with the cool roof showing lower energy consumption for cooling. In addition to the climatic aspect, cool and green roof design and building characteristics strongly influence the results. Gagliano et al. [13] reported that for a single-storey house located in Catania (Csa), extensive green roofs (LAI equal to 5) provided a reduction of up to 29.3% in annual energy consumption for air-conditioning compared with both traditional and cool roofs (reflectance equal to 0.55). Silva et al. [8] observed that for a room with adiabatic walls located in Lisbon (also Csa), semi-intensive (LAI equal to 2.5) and intensive (LAI equal to 5) green roofs could reduce annual energy consumption compared with traditional and cool roofs. On the other hand, extensive roofs (LAI equal to 1) would only reduce consumption on roofs without insulation.

The green roof resulted in the highest annual percentage of comfort hours in all climates. These results are consistent with the findings of Gagliano et al. [13] and Cirrincione et al. [12]. In locations with higher average annual temperatures (Florianópolis and Brasília), the green roof performed better without the thermal insulation layer. On the other hand, in the climate of Curitiba, the green roof performed better with thermal insulation.

Concerning the outside surface temperature of the roofs, the green roof showed more satisfactory behaviour in all the cities evaluated. Green roofs with lower LAIs (e.g., 2) may have behaviour similar to aged cool roofs (reflectance of approximately 0.65), while cool roofs with higher reflectance (around 0.80) can be better solutions [47]. The reflectance of 0.58 adopted in this study is conservative. However, the reflectance of these roofs decreases in the first months of application, and this aspect should be considered. The outside surface temperature of the green roof was not influenced by thermal insulation, resulting in differences in maximum temperature of less than 0.3 °C, consistent with the findings reported by Gagliano et al. [13].

The results show that green and cool roofs are solutions for reducing the outside surface temperature of the roofs, helping to mitigate the effects of heat islands and improving the quality of the outside environment in all the climatic contexts analysed. Thermal insulation is not appropriate from the perspective of heat islands, as it increases the outside surface temperature of traditional and cool roofs by hindering the heat transfer to the lower layers of the roof and is irrelevant in the case of green roofs.

3.5. Recommendations for Urban Planning Policies and Building Regulations in Brazil

To promote the application of cool and green roofs on buildings, policies could be implemented to improve the understanding of the benefits of these strategies in the inside and outside environments of buildings. On a building scale, encouraging these roof typologies can reduce energy consumption and improve occupants’ thermal comfort in new buildings and retrofit projects. At a city-wide urban planning level, policies should create a favourable environment for these passive strategies to maximise cooling effects in urban areas.

The Brazilian government has recently focused on drafting the National Urban Development Policy (PNDU). This policy, under development at the time of publication of this paper, aims to support cities in their urban policy actions. In this context, the National Agenda for Sustainable Urban Development, which will be part of the PNDU, considers the characteristics of urban spaces and the state of climate emergency as a basis for developing political projects for cities. Among the measures to mitigate greenhouse gas emissions and disaster risks, the National Agenda [55] acknowledges implementing green infrastructure, promoting sustainable construction and technologies, and protecting and recovering vegetation cover and local biodiversity. In terms of energy inefficiency in cities, despite Brazil’s clean energy matrix compared with other countries [56], adopting climate-adapted building systems to reduce energy consumption in buildings and reduce the effect of heat islands should be explored.

In Brazil, standards NBR 15575-1 [43] and NBR 15220-3 [34] aim to improve the thermal performance of buildings. NBR 15575-1 [43] establishes the procedure for classifying buildings according to the thermal performance level (minimum, intermediate, and superior). NBR 15220-3 [34] describes the method used for bioclimatic zoning for thermal performance analyses and nonstandardised construction guidelines.

Although such standards have been recently developed and updated, they still do not recommend using passive strategies for roofs other than conventional ones (clay or fibre cement tiles, with or without ceiling, insulating and reflective layers). In Brazil, some legislation requires [57,58,59] or encourages [60,61] to implement green roofs in specific situations; however, they are punctual and not widely spread. To date, there are no technical standards for the construction of green roofs, which does not provide suitable guidance for professionals.

Given that the improvement in thermal performance with the use of cool and green roofs has been demonstrated, it is recommended that urban development policies address the cool and green roofs to disseminate these strategies among professionals and the general population, as well as the new Energy Performance of Buildings Directive [62] in Europe, which considers green roofs. In future versions, the NBR 15220-3 standard [34] could recommend cool and green roofs in the design phase and as retrofit measures on pitched and flat roofs, respectively. Additionally, policymakers can extend municipal or state legislation, including financial incentives for adopting these solutions according to the climate. In general terms, green roofs could be recommended for buildings located in Cfa, Cfb, and Aw climates. Although they result in lower energy savings in Cfa and Aw climates, green roofs offer the benefits of thermal comfort and reduced outside roof temperatures, helping to mitigate the effects of urban heat islands and improving the quality of the outside environment. On the other hand, cool roofs could also be indicated for buildings in cities with Cfa and Aw climates, clarifying that the greatest benefit achieved is related to energy savings. However, some regions still lack further studies, considering Brazil’s large area and climate variability. Methodologies that help designers choose between these types of roofs are still needed, given that in specific climatic contexts, the most suitable roof typology differs depending on the parameter considered.

4. Conclusions

This study presented a comparative analysis of different roof typologies (traditional, cool, and green) in three Brazilian climatic contexts. The following general conclusions can be drawn:

-

Cool roofs are recommended to reduce energy consumption in cities with a warmer climate, such as Florianópolis and Brasília (electricity consumption can be reduced by up to 24.8% compared with the traditional roof). Green roofs showed better energy performance in Curitiba, a city with lower average annual temperatures, reducing energy consumption by 28.2% compared with the traditional roof.

-

Green roofs resulted in the highest percentages of thermal comfort hours in all climates (between 57.3% and 82.2% in the living room and between 69.8% and 96.3% in the bedrooms, depending on the climatic context).

-

Cool and green roofs can reduce the outside surface temperature compared with the traditional roof. Considering hourly temperatures, the cool roof can reduce the temperature by up to 16.5 °C, and the green roof by up to 28.4 °C, compared with the traditional roof.

-

Depending on the roof typology and the climate, thermal insulation can increase or decrease electricity consumption (variation between −10.6% and 20.6%) and the percentage of thermal comfort hours (variation between −13.7 and 8.0 percentage points). The outside surface temperature of traditional and cool roofs increased with thermal insulation, and that of the green roof was not affected. Therefore, from the perspective of the outside environment, thermal insulation is not recommended.

The results of this study confirm the positive contribution of cool and green roofs to a building’s inside and outside environments in different Brazilian climatic contexts, and some recommendations for urban planning policies and building regulations in Brazil were provided. However, it should be noted that the thermal and energy performance of cool and green roofs depends on the design characteristics adopted. Thus, the results found in this research are limited to the reference data used for the roof layers.

In the simulations for both traditional and cool roofs, the solar absorptance was assumed to remain constant after three years of degradation throughout the analysis period. This approach did not account for the gradual degradation of solar absorptance over time or its potential restoration following washing cycles. We, therefore, recommend further investigation into the feasibility of frequent washing of traditional and cool roofs to recover solar absorptance. Furthermore, only one type of green roof was evaluated in this study. Exploring different materials and layers could yield varying results. Future research could also apply the proposed methodology considering projected future climatic conditions, which may influence the outcomes.

Additionally, it was observed that the ideal roof typology based on a particular relevant parameter may not be the best solution from the perspective of another parameter. Hence, choosing the most suitable typology for each climate is challenging. Furthermore, evaluating these roofs depends on other aspects, such as life cycle analysis, costs, and social issues. Therefore, there is a need for a broader evaluation, considering other important aspects and providing a single indicator to define the roof typology with the best overall performance in different climatic contexts.