3.1. Climate, Meteorological, and Bioclimatic Data Following the Administrative Boundaries
The annual average dry bulb temperature (DBT) over the years varied from 14.3 °C to 29 °C, with an average value of 23.2 °C and a standard deviation of 3.1 °C (
Figure 5a). The Mann–Kendall results showed that 15% of the municipalities (812) had an increasing trend throughout the 15-year annual records (
Figure 5b). The north region showed the lowest increase (11 municipalities), and the southeast region had the highest occurrence (416 municipalities). The results showed that the southeast region encompasses more than 50% of the municipalities, with an increasing trend, while the other regions had between 12.1% and 15.9%, except for the north region (1.4%).
Over the 15-year record, the mean variation was 0.5 °C, representing a significant temperature increase in 3288 municipalities, approximately 60% of the Brazilian municipalities (
Figure 5c). Out of them, 44 municipalities showed temperature variations from 1 °C to 1.2 °C. Except for two municipalities in the state of Paraná, all the municipalities with temperature variations above 1 °C were in the state of São Paulo. Those results, supported by the Mann–Kendall analysis, state that the southeast region had the highest DBT variations in Brazilian territory. The highest variations in the north and northeast regions represented a 0.9 °C difference between 2008 and 2022. Therefore, it can be concluded that most of the Brazilian territory has annual averages of at least 22 °C and a temperature increase between 0.4 °C and 0.7 °C.
The average of the maximum and minimum annual DBT results were mainly related to the altitude and the latitude since the locations close to the Equator and with lower altitudes had the highest records, while the locations close to and below the Tropic of Capricorn and with higher altitudes had the lowest ones (
Appendix A—
Figure A1a,c). The lower average maximum DBT mostly occurred in the high-altitude regions, represented by the blue-shaded region in
Figure A1a. Regarding the trends (
Appendix A—
Figure A1b,d), the Mann–Kendall results showed that all regions presented an increasing trend for the maximum DBT (
Appendix A—
Figure A1b).
The combined records of the average maximum and minimum DBT resulted in an average spread of 6.3 °C with a standard deviation of 2.5 °C for the Brazilian territory. The north region had the lowest spread (3.7 °C), followed by the northeast (4.2 °C), the central–west (5.5 °C), the southeast (6.7 °C), and the south (10.3 °C). Then, the southeast region had results closer to the country average, and the south showed the highest difference.
The Brazilian territory showed an average relative humidity (RH) of 70.7%, with the lowest average value of 50.7% and the highest relative humidity of 89.9% considering the monthly time series of 15 years (
Figure 6a). The annual relative humidity occurrences above 80% happened in regions with dense vegetation, especially the Amazon rainforest. The lowest RH encompasses the Semi-Arid hot climate, which also shows temperature variation above 0.4 °C. The central–west region had the lowest average RH (65.2%), and the south region had the highest average (76.4%). Regarding the regions’ distribution, the south had a standard deviation of 4.1 percentage points (pp), the lowest among the five regions, and the northeast had the highest (8.9 pp). The highest variation in the northeast region is probably related to the environmental differences between the coastal and continental areas.
Figure 6a shows that the coastal regions have higher RH, while the continental area, especially the Semi-Arid region, has lower RH.
The variation in the relative humidity over the 15 years starts from a 5.6 percentage point (pp) decrease to a 3 pp increase (
Figure 6c). The combined results of the Mann–Kendall test (
Figure 6b) and the cumulative average also show overlaps in the southeast region, representing not only a linear RH reduction over time but in an overview of the territory. The RH variation in the southeast, northeast, and north regions matches the municipalities with the highest temperature increase.
The highest RH corresponded to the cities in the north region, surrounded by the Amazon rainforest. On the other hand, the lowest RH values corresponded to the Semi-Arid region, with the highest occurrences in the northeast region. Most of the Brazilian territory presented a maximum reduction of 2 pp. This same reduction represents most of the north, northeast, southeast, and south regions. The central–west region results showed that almost 60% of its territory had no variations or an increase below 2 pp.
The distribution of the maximum and minimum RH records showed that most of Brazil presents humid and dry conditions, except for the Semi-Arid region which, despite showing a significant variation between the maximum and minimum records, mainly presented the lowest values. On the contrary, the Amazon rainforest region had the highest values. The Mann–Kendall results also show many municipalities with trends to average maximum RH decrease, especially in the north and central–west regions (
Appendix A—
Figure A2). Then, following the same pattern as the average DBT and RH, the region defined by the Cerrado and Caatinga biomes (including the Semi-Arid region) had the lowest minimum RH values.
The combined Mann–Kendall results for the average maximum and minimum RH show that most municipalities with a trend presented a reduction in the RH spread but did not become drier, with an exception for some municipalities in the north (three municipalities), southeast (seventy-six municipalities), and south (fifty-four municipalities). The municipalities of the south with a trend to become drier are the majority, according to the Mann–Kendall results (61.4%).
Regarding the RH spread, the average in the Brazilian territory was 25.7 pp with a relative deviation of 44.8%. Again, the southeast region had the results closest to the Brazilian territory average with the same average spread (25.7 pp). The central–west region had the highest spread (43.6 pp), and the south region showed the lowest spread (17.2 pp). The northeast and the north regions presented intermediary values, respectively, 24.9 pp and 32.4 pp.
Over the 15 years considered in this study, the Brazilian municipalities presented precipitation levels varying from 296 mm to 3798 mm (
Figure 7a). The average annual precipitation was 1302 mm, with a standard deviation of 500 mm. The results show that despite a high variation in the annual precipitation levels, most of the territory is located on an interval that comprehends levels ranging from 1000 mm to 2000 mm. Municipalities in the Amazon rainforest region had higher precipitation volumes. The lowest precipitation followed the same pattern of the highest temperature, temperature variation, lowest relative humidity, and relative humidity variation. The Semi-Arid region showed the lowest precipitation levels, averaging 692 mm.
The Mann–Kendall results show a significant decrease in the average annual PPT for the central–west and southeast regions and some occurrences in the other regions (
Figure 7b). The north region was the only one presenting a linear increase trend for PPT. Regarding the precipitation (PPT) variation over the 15 years, most of the Brazilian territory had a change in the precipitation levels, varying from a reduction of 59% to an increase of 33% (
Figure 7c). The highest reduction in precipitation occurred in the northeast and some municipalities of the southeast regions. The cities with increased precipitation levels showed a stable annual temperature without significant variations but slightly increased humidity levels.
The average minimum and maximum PPT summarized the seasonality in the Brazilian territory (
Appendix A—
Figure A3a,c). The Amazon rainforest region, followed by the south region, had the highest minimum PPT with averages of 19.3 mm and 38.8 mm, respectively. The results showed that 5556 locations, representing approximately 99.8% of the Brazilian municipalities, have dry months (below 100 mm), varying from 0.4 mm to 99 mm. The remaining 12 municipalities, with all months within the threshold, varied from 100 mm to 142 mm. The average maximum PPT results showed that most locations had a maximum precipitation between 200 mm and 400 mm. The average maximum PPT varied from 82.2 mm to 729 mm. Thus, 23 locations can be classified as dry throughout the year since the precipitation levels are below 100 mm. The cities with dry conditions during the year are within the Semi-Arid region. As previously mentioned, they had the lowest precipitation levels, followed by low relative humidity and higher DBT.
The Mann–Kendall results for the average maximum and minimum point to a larger decreased trend for both quantities and a slight trend for PPT increase over the territory (
Appendix A—
Figure A3b,d). The results show that the decrease trends prevail in regions with high annual average PPT and higher averages for the maximum and minimum quantities. For the increase trends, the situation is similar for the maximum PPT but, for the minimum records, the results show occurrences in regions with the lowest records.
The comparison between ERA5-Land and the Brazilian Solar Atlas annual average from monthly records resulted in an average difference of −3 kWh/m
2 varying from −25 kWh/m
2 to 12 kWh/m
2. The difference also presented a standard deviation of 4.2 kWh/m
2 and a RMSE of 5.1 kWh/m
2, proving that the ERA5-Land is a reliable source for an annual analysis based on the average of monthly integrals (
Figure 8b). The average results by region showed that the northeast records from ERA5-Land best matched the Brazilian Solar Atlas with an average underestimation of 0.1 kWh/m
2. The north region had the highest difference, with an overestimated 6.6 kWh/m
2. The ERA5-Land records overestimated the average GHI for the remaining regions based on the monthly integrals varying from 3.1 kWh/m
2 to 5.6 kWh/m
2. The results also indicated that 1579 municipalities (28.3%) had an average GHI difference higher than the RMSE. The central–west region had the lowest number of municipalities above the RMSE threshold, while the southeast region had the highest, encompassing 226 and 426, respectively. A total of 71.6% of the Brazilian municipalities showed a difference within the RMSE threshold, therefore showing a high correlation between ERA5-Land and the Brazilian Solar Atlas. Only 191 (12.1%) presented absolute values above 10 kWh/m
2 from the locations with higher differences.
The annual average from monthly integrals of GHI varied from 117 kWh/m
2 to 185 kWh/m
2, with an average value of 156 kWh/m
2 and a standard deviation of 13 kWh/m
2 (
Figure 8a). Regarding the psychrometric relations, the highest values occurred in the cities with the lowest relative humidity since these variables can be directly correlated with the cloud cover that impacts the amount of radiation the land surface receives. The same happens in the opposite direction since the Amazon rainforest regions had high humidity levels, increasing the precipitation and cloud cover, directly affecting the region’s radiation values. The Brazilian territory had most municipalities (2834) with radiation levels between 150 kWh/m
2 and 170 kWh/m
2, representing 47.3% of the territory. The north and south regions had 68.4% and 74.1% of their territory with radiation levels between 150 kWh/m
2 and 170 kWh/m
2. The central–west and southeast regions had 96.7% and 61.2% of their territory with radiation levels between 150 kWh/m
2 and 170 kWh/m
2. The northeast had 53.6% of its territory with radiation levels above 170 kWh/m
2.
The results from the classification using data from 1950 to 1990 and 2008 to 2022 (
Table 3) showed that the tropical climate area decreased by 3.1 pp, the dry climate increased by 2.6 pp, and the subtropical climate increased by 0.5 pp. Thus, the Brazilian experienced a shift towards hotter and drier conditions.
Figure 9 shows that the classification using the ERA5-Land database differed from Alvares et al. [
1] on the territory occupied by each climate group and the classification of some municipalities in the Hot Desert-Arid group (BWh). Since the Köppen–Geiger classification is based on air temperature and precipitation, the regions most affected by the temperature and precipitation variation had the most significant changes in the climate classification. Considering the 15-year data, the tropical climates, represented by group A, occupy the highest portion of the Brazilian territory (82%). The subtropical climate, represented by group C, occupies the second position, with 9.4% of the territory. Group B, representing the dry climate, covers 8.6% of the Brazilian territory.
The results for the first trimester indicate that design strategies that rely only on ventilation (E and G) are the majority in reaching thermal comfort in most of Brazil (
Figure 10). The onset of autumn increases the need for heating solutions in the southern region and parts of the southeast. The temperature reduction generally increases the municipalities’ thermal comfort, particularly in the southeast and south regions. The winter solstice represents another turning point for the bioclimatic design requirements in Brazil. The third trimester shows the highest number of municipalities with heating demand. However, thermal comfort has the highest territorial occurrence, since the heating demand comes from small municipalities, and the thermal comfort is achieved in municipalities that occupy larger territories. The last trimester closes the annual cycle and shows an increasing demand for ventilation strategies while a significant reduction for heating strategies.
The NVP based on the DBT records reinforced the thermal comfort occurrence over the Brazilian territory as they were based on the adaptive model with an 80% acceptance from ASHRAE 55 (
Figure 11a). The Brazilian territory comprised an average NVP of 274 days, and 75% of the year within thermal comfort according to the outdoor conditions. Urupema-SC had the lowest thermal comfort occurrence and, consequently, the lowest potential based only on DBT records (14 days). On the other hand, 695 municipalities (12.5%) of the north and northeast regions presented an average potential for all days of the year. The south region had the lowest average DBT NVP (160 days), and the north region had the highest average potential (360 days). The central–west and northeast regions also presented a potential above 300 days, and the southeast region—an average of 235 days.
While the temperature threshold allowed for a high ventilation potential in most of the Brazilian territory (
Figure 11a), the WBT threshold, which is related to humidity levels, resulted in a lower potential for natural ventilation (
Figure 11b). On average, humidity constraints limited ventilation for 88 days, with a standard deviation of 79 days, affecting 1171 municipalities (21%). The WBT analysis revealed that the regions with the highest potential for the DBT threshold had the lowest potential for the humidity constraints, and vice versa for the regions with the lowest potential for the temperature constraints. The north region had the lowest potential, averaging 7 days, and the south region had the highest potential of 179 days. For the WS threshold, all municipalities showed a potential for all days of the year, since none exceeded the wind speed upper limits.
The final assessment combined the temperature, humidity, and wind speed threshold (
Figure 11c), resulting in an average of 22 days suitable for natural ventilation across Brazil with a high standard deviation (27 days). A total of 1317 municipalities did not present natural ventilation potential for the entire year, mainly encompassing the north and northeast regions and some coastal municipalities in the southeast and the south regions. The central–west region had the highest average (57 days), while the north had the lowest potential (6 days).
The heating and cooling analysis from Givoni was complemented by quantifying the average annual degree days along the 15-year records. The territorial distribution in
Table 4 shows that the south region had the most even distribution for the heating demand, with a relative deviation of 51%. The north region had a more even distribution for the cooling demand, with an average relative deviation of 24%.
The HDD
18 (
Appendix A—
Figure A4a) reinforced the heating demand by the highest average of 219 K d in the south, a significant demand in the southeast region (84 K d), and a smaller portion in the central–west region (27 K d). The CDD results indicated that, in addition to passive cooling strategies, artificial methods are also necessary for several locations in Brazil, with maximum demand from 1246 K d of CDD
26 to 1970 K d of CDD
24. The results also revealed that the north region had the highest demand, followed by the northeast, central–west, southeast, and south regions. The step analysis (
Appendix A—
Figure A4b–d) showed a considerable reduction for each region from 24 °C to 26 °C base temperature. For all the regions, the results showed a significant cooling demand depending on the setpoint defined for the degree days analysis.
3.2. Bioclimatic Zones Assessment
The reclassification of the Brazilian municipalities based only on ERA5-Land data showed that 539 municipalities (9.6%) presented a different classification from Silva Machado et al. [
11]. Most of the changes were influenced by the RH records from the ERA5-Land database, resulting in changes within the same macrozone but migrating from a humid (A) to a dry (B) condition.
3.2.1. Climate and Meteorological Assessment
Considering the 15-year annual average for each municipality,
Figure 12 showed that the average DBT among zones varied from 16.7 °C (1R) to 27.6 °C (6B) with a standard deviation varying from 0.3 °C (6A) to 0.9 °C (1R). The DBT cumulative average variation (variation between the 2008 and 2022 average) showed that zones 4A, 5A, and 6A had the lowest average variation (0.3 °C), while zone 1M had the highest variation among the bioclimatic zones (0.7 °C). Despite varying from 0.3 °C to 0.7 °C among zones, the standard deviation was low for all twelve zones, ranging between 0.1 °C and 0.2 °C. Following the average DBT distribution, the average maximum and minimum DBT showed a similar distribution among the twelve zones. Zone 1R had the lowest averages for the minimum (11.2 °C) and maximum DBT (21.6 °C), while zone 6B had the highest maximum (30.1 °C) and zone 6A had the highest minimum (25.7 °C). Those results showed that the zones had a homogenous distribution regarding the average DBT, as stated by Silva Machado et al. [
11], and the average maximum and minimum DBT.
Regarding the annual average RH, the Brazilian bioclimatic zones show a clear separation between humid and dry locations (
Figure 13), as described in the method presented in
Section 2.1 and corroborating with the climatic analysis from
Section 3.1. Zones 1 and 2 showed overlaps between the M and R classes, differing from the other bioclimatic zones. Zone 6B showed the lowest average (58.4%), and zone 1R had the highest average RH (79.3%). The driest zone (6B) also had the highest average reduction (2 pp), while zones 6A and 2R had the lowest average reduction (0.1 pp). The standard deviation also shows that zone 5A had the highest variation dispersion (4.8 pp), and zone 2R had the highest homogeneity (2.1 pp). The average maximum RH varied from 78% (6B) to 86.8% (5A), while the average minimum varied from 39.7% also in zone 6B to 71.4% (1R). Zones 6B and 5A presented the highest dispersion for the average maximum and minimum, respectively, presenting standard deviations of 5.5 pp and 11.2 pp. Zone 2R had the lowest deviation for the average maximum and minimum with standard deviations of 1.6 pp and 3.7 pp.
The annual average PPT indicated that all zones have overlaps (
Figure 14), different from the previous results for the average DBT and RH. The annual average PPT varied from 897 mm for zone 6B to 1760 mm in zone 1M. The results corroborate the rules to define the Brazilian bioclimatic zones, especially for zone 6B, which has the lowest RH levels and directly impacts the PPT records. Regarding the PPT variation, most zones showed an average reduction varying from 1.4% (1M) to 24.4% (6B). Zones 1R and 2R were the only ones showing an average increase in precipitation levels, respectively, of 1% and 7.1%. Zones 5B and 6B also showed the highest deviations for the average PPT distribution, mainly representing a reduction over the 15-year records and reaching almost 60%. The average maximum PPT was highly homogeneous among the zones, varying from 197 mm (4A) to 367 mm (6A). Zone 5A also shows the highest average maximum PPT (729 mm), and zone 5B shows the lowest average maximum (82 mm). For the average minimum PPT, zone 6B had the lowest average, with no precipitation, and zones 1M, 1R, and 2R had the highest levels (40 mm). The standard deviation also pointed to a high variation for zone 5A (21 mm) and no variation for zone 6B. The average maximum and minimum analysis depicted the monthly precipitation profile for the bioclimatic zones, showing that zones with the lowest humidity levels also had lower precipitation throughout the 15-year records.
The updated Köppen–Geiger classification for Brazil showed that zones 2M, 3A, 3B, and 4A had the highest variability among the classes, with climates from classes A, B, and C. The results also pointed to regions with lower diversity with only two classes for zones 1R, 2R, and 6B. Despite the different classes within zones, it is possible to distinguish a prevailing class for zone 1M (Cfa = 52.2%), 1R (Cfb = 69.6%), 2M (Cfa = 64.3%), 2R (Cfa = 99.5%), 3B (Aw = 59.8%), 4B (Aw = 73.7%), 5B (Aw = 52.6%), 6A (Aw = 84.8%), and 6B (BSh = 52.3%). For zone 3A, the prevailing classes were Aw (44.5%) and Cfa (30.5%). Zone 4A had classes Aw (47.1%) and As (29.2%) as the predominant. Finally, zone 5A presented a more even distribution between classes Aw (36.8%), As (26.7%), and Am (23.6%). This pattern resulted from the method used for bioclimatic zoning, which combines both climatic and performance parameters, resulting in a diverse distribution within zones.
3.2.2. Bioclimatological Assessment
The findings for January to March indicated that daytime ventilation could improve thermal comfort across all zones. The results also showed that some municipalities from zones 1M, 1R, 2M, and 3A also required thermal mass associated with solar heating, while the zones with the highest temperatures, zones 5 and 6, also required strategy G of thermal mass and evaporative cooling. Zone 4B had the highest number of locations where no strategy met the 7-day threshold apart from thermal comfort.
For April, all of the bioclimatic zones required daytime ventilation, and zones 1 and 2 kept the requirement for heating strategies. Zone 1R also showed a requirement for solar heating (B). May showed a similar pattern but required artificial heating for zone 1R and solar heating combined with thermal mass for zones 3 and 4. May also showed no demand for ventilation strategies from zones 1 and 2R. June followed the requirements of May, but artificial heating was added to zones 1M and 2R, and ventilation strategies were removed from zone 2M. The coldest zones, represented by groups 1 and 2, also showed a reduction in thermal comfort and an increase for zones 4 to 6. Zone 3B also increased for April and May, but June presented a decrease.
July has the same recommendations as June, except for the combination of thermal mass with solar heating strategies and thermal comfort (CD) for zone 2R. For the CD combination, zone 2R increased from a single location in June to 299 municipalities in July. August, however, starts showing thermal comfort conditions in zone 1M and a general thermal comfort increase in all the other zones except for zone 1R. The results also show a reduction in the demand for heating strategies in all zones, especially for zones 1 and 2. For September, most municipalities did not require artificial heating (A), except for Urupema-SC. The results also showed a demand for strategy H (thermal mass with evaporative cooling) for seven municipalities of zone 6B.
The results showed that October required heating strategies from zones 1M to 3A, mainly defined by thermal mass combined with solar heating (C), except for 27.1% of the municipalities from zone 1R that still required strategy B. October also showed increased demand for daytime ventilation from zones 2M to 6B. November showed a similar pattern to October but required daytime ventilation for 76.3% of the municipalities from zone 1R. Finally, December showed that all bioclimatic zones required daytime ventilation (E) for most of their municipalities, which mainly combined thermal mass with solar heating (C) for zones 1 and 2. For zones 3 to 6, daytime ventilation was mainly combined with strategy G or with days within thermal comfort (D).
This present paper provides a detailed analysis confirming the need for those strategies and reinforcing the ventilation potential through the NVP results. The results showed that locations with dry classification “B” of zones 3 to 6 had the best outcomes for the NVP (
Figure 15). Additionally, the colder municipalities within zone 2 had a higher potential than those in zone 1.
Zones 1R, 1M, and 2R have the highest heating demand following the bioclimatic distribution (
Figure 16). This study found an average greater than zero in zones 3 and 4 but significantly lower than the demand from the coldest zones, ranging between 2 K d and 52 K d of HDD
18. The cooling demand increased from zone 1 to 6, following the daily average DBT that delimitates each of the 12 zones. The results also demonstrated that zones 1M to 4B had the highest reduction from CDD
26 to CDD
24, with an average of 79%. Zones 5A to 6B had a lower average reduction but still showed an average cooling demand reduction ranging from 49% to 66%.
3.2.3. Representative Locations and Building Performance Simulation
Using the Brazilian bioclimatic zoning methodology presented in
Section 2.2 and the ERA5-Land daily records, the 5567 Brazilian municipalities used in this study were classified to identify the representative locations (
Table 5). After applying the Kolmogorov–Smirnov test, 12 representative locations were identified with an annual average DBT varying from 16.7 °C in São Marcos-RS (1R) to 27.7 °C in Bela Vista do Piauí-PI (6B). Colônia Leopoldina-AL (4A) and Pio XII-MA (6A) had the lowest DBT variation throughout the 15-year records (0.2 °C), while Agudos-SP (3B), Bela Vista do Piauí-PI (6B), Ibicaré-SC (1M), and São Marcos-RS (1R) had the highest DBT variation (0.8 °C). Olímpia-SP (4B) showed a 0.7 °C variation between 2008 and 2022 and was the only one with an “increase” result for the Mann–Kendall trend analysis. Bela Vista do Piauí-PI (6B) showed the lowest average RH (52.9%), and Fonte Boa-AM (5A) had the highest average RH (88.9%). Regarding the 15-year variation, Muqui-ES (3A) had the highest RH decrease (3.6 pp), while Chiapetta-RS (2R) and Pio XII-MA (6A) showed an increase in the RH, respectively, of 0.2 pp and 1.1 pp. The Mann–Kendall results also showed that Agudos-SP (3B), Fonte Boa-AM (5A), Três Barras-SC (1M), Olímpia-SP (4B), and São Marcos-RS (1R) had a “decrease” trend for the RH records. Therefore, the results indicated that Olímpia-SP (4B) was the only location with significant DBT and RH Mann–Kendall analysis results.
The heating demand was highest in zones 1 and 2, while zones 3 and 4 had significantly lower demands. The cooling demand was lowest in zones 1M to 2M, followed by zones 2R to 3B. Zone 4 locations had a higher demand for cooling, and zones 5 and 6 showed the highest cooling demand at 24 °C base temperature.
The PHFT results showed that the construction set with the lowest U-value (U
low) delivered the best average PHFT, independent of the solar absorptance (
Figure 17), for the representative locations for the cold zones (1M, 1R, 2M, and 2R). From zones 3 to 6, the most insulated construction set gradually delivered intermediary to worst PTHP, while U
medium (intermediary U-value) with a low solar absorptance had the highest averages. For zones 3 and 6, U
high, with a solar absorptance of 0.3, also delivered good PHFT results, and U
medium, with a solar absorptance of 0.7, had the lowest PHFT levels.
Considering the average results from the 15-year simulation outputs for each construction set and solar absorptance combination, Três Barras-SC (1M) had the best results with an 88.4% PHFT for Ulow with low solar absorptance, and Fonte Boa-AM (5A), for Ulow with the highest absorptance, had the worst results (PHFT of 4.4%). Fonte Boa-AM (5A) had the greatest increase in PHFT, rising by 30.8 percentage points when transitioning from Ulow with a solar absorptance of 0.7 (PHFT of 4.4%) to Umedium with a solar absorptance of 0.3 (PHFT of 35.2%). On the other hand, Agudos-SP (5A) showed the lowest variation, increasing from 52% for Uhigh with a solar absorptance of 0.7 to 61.7% with the same construction set, but with a solar absorptance of 0.3.
The analysis by construction set revealed that highly insulated configurations performed best in zones 1 and 2, ranging from 69.9% to 88.2%. On average, an increase of 16.6 percentage points was found in the representative locations of colder climates. For zones 3 to 6, the construction set with an intermediary U-value delivered the best results. The average PHFT from zones 3 and 4 varied from 39.9% to 62.9%, with an average improvement of 13.9 percentage points. The hottest zones, 5 and 6, had the lowest PHFT levels, from 31.5% to 51.9%, but they had the highest improvement in PHFT levels, with an average of 18.6 percentage points.
Considering a multi-year analysis based on the cumulative average, all the locations resulted in a reduction in the PHFT when comparing 2008 and 2022, independently of the building envelope configuration, except São Marcos-RS, whose cumulative average varied of 1 pp with Umedium and Uhigh with low absorptance and 0.2 pp with Umedium with an intermediary absorptance. Bela Vista do Piauí-PI (6B), a hot and dry climate, had the highest variation with a reduction of 6.6 pp. The most insulated building envelope resulted in the highest average reduction of PHFT (9.6 pp), and the intermediary components (Umedium) had the lowest reduction (6.7 pp).
The insulated construction (U
low) delivered the best performance regardless of solar absorptance, removing all the heating demand for the representative locations from zones 2M, 3, and 4 (
Figure 18). In the remaining locations, the heating demand significantly decreased for all the years.
The 15-year average results showed that São Marcos-RS (1R) had the highest heating demand (1833.6 kWh/year) when combining U
low with the lowest solar absorptance but also presented the highest absolute reduction (1553.4 kWh/year) for the combination of U
low with a high solar absorptance. Among the other seven locations, Muqui-ES (3A) had the lowest average demand (2.8 kWh/year) and accomplished a 100% reduction when using U
low. As expected, the construction set with a high U-value presented the highest heating demand, while the most insulated resulted in a lower or even no demand for heating. In the climatic data analysis from
Section 3.1, it was found that there was a predominant heating process in Brazil’s 15-year series. As a result, the heating demand over this simulation period decreased the cumulative average, even for the building envelope configurations that allow for a higher heating exchange (U
medium and U
high).
The representative locations from zones 3 and 4 also showed a low heating demand. The 15-year average HDD
18 from
Figure 16 showed that for zones 3, the average value is 51 K d with an average maximum of 221 K d. Zone 4’s average value is 9 K d with an average maximum of 122 K d. The combined analysis of HDD
18 and simulation results points out a necessity of a review of the NBR 15575 to adjust the threshold for CgTa calculation.
The 15-year summary (
Figure 19) indicated that Pio XII-MA (6A) had the highest cooling demand when combining high U-value and solar absorptance (13,070.7 kWh/year). However, it also showed a significant reduction in cooling demand when using the insulated building envelope (U
low) with a low absorptance (4171.3 kWh/year). São Marcos-RS had the lowest cooling demand (1214.3 kWh/year) when using the intermediary U-value construction with high solar absorptance and achieved a 73% reduction using a high insulated construction with low absorptance. The results from the average construction set showed that the ones with high U-value resulted in the highest demand, while the highly insulated construction presented the lowest demand, with an average reduction of 29%. Colônia Leopoldina-AL (5A) and Pio XII-MA (6A) showed an exception since the intermediary building envelope with low absorptance delivered the best results.
Regarding cooling demand, all locations presented an increasing trend, unlike the PHFT, which mostly indicated a decreasing scenario with some increases based on the building envelope configuration. The multi-year analysis also showed that Bela Vista do Piauí-PI (6B) also had the highest variation for the cooling energy demand, with an average increase of 415.5 kWh/year on the cumulative average. On the opposite side, Chiapetta-RS (2R), with a cold climate, had the lowest increase (218 kWh/year). The building envelope also influenced the magnitude of the increase, as the highly insulated configuration indicated the lowest average increase (738 kWh/year), while the building envelope with a high U-value had the highest average increase in the building demand (966.4 kWh/year).