This study aims to evaluate the outdoor environment conditions (synoptic weather conditions) on extreme events of human thermal discomfort in the city of Pelotas, Brazil, between 18th January and 20th August, 2019. Digital thermohygrographs were installed outside the residences in order to record of air temperature data, dew point temperature and air relative humidity. Five different macro-regions of the city were analyzes applying two different biometeorological indices. For the summer months, the Humidex Index was used, and for fall and winter, the wind chill index was used. The findings showed that the afternoons during the summer months were the most thermally uncomfortable, while the nights presented the greatest sensations of comfort. In the winter months, the opposite occurred: afternoons with more comfortable periods and nights with more thermally uncomfortable periods were observed. It was also noted that in the situation of extreme heat stress, in the summer, there was the influence of the South Atlantic Subtropical Anticyclone (SASA) on the region, which, in association with the low humidity of the air on the continent, favored, then, the extreme discomfort sensations. Likewise, situations of extreme discomfort due to cold that were also associated with low air humidity, and high atmospheric pressure systems (not necessarily the SASA), which caused drops in temperature and dry air.
Research Article
Analysis of thermal discomfort associated with synoptic conditions in the city of Pelotas, extreme south of Brazil
https://doi.org/10.21203/rs.3.rs-2947906/v1
This work is licensed under a CC BY 4.0 License
Journal Publication
published 08 Jan, 2024
Read the published version in International Journal of Biometeorology →
You are reading this latest preprint version
Effective temperature
Humid subtropical climate
Southern Brazil
Thermal discomfort
Human Biometeorology is the science that studies the interactions between atmospheric processes and living organisms, with the main objective of evaluating the impact of climate variations on human health, proposing to answer questions of how weather and climate can affect the well-being of all human beings (Hoppe 1999; International Society of Biometeorology 2021).
Among the areas covered by Human Biometeorology is thermal comfort, which is defined as the state of equilibrium in which the body finds itself when the loss of energy from the body to the external environment is equivalent to the loss of energy produced by the organism (Fanger 1970). Some environmental factors such as temperature, humidity, wind speed and personal factors such as level of physical activity and insulation of clothing influence thermal comfort, making it subjective, as it is a perceived sensation that varies from person to person (Nedel 2008).
In order to represent how much meteorological variables impact the human body, several studies have analyzed thermal comfort through simplified biometeorological indices (Suping et al. 1992; Hoppe 1999; Nedel 2008; Gobo and Galvani 2012; Nedel 2021 among others), which relate only environmental factors. Some of the simplified biometeorological indices found in the literature are: (a) effective temperature (ET) (Missenard 1948); (b) discomfort index (DI) (Giles et al. 1990); (c) wind chill index (WCI) (Suping et al. 1992), (d) humidex (HU) (Masterton and Richardson 1979); (e) heat index (HI) (Steadman 1979), among others.
Extreme conditions of discomfort due to high or low temperatures lead to serious consequences for human health, and may cause the manifestation of certain diseases, mainly in children and the elderly. Sudden changes in atmospheric weather also contribute to changes in the quality of air breathed, especially when the location is under the influence of a cold and dry air mass, making it difficult to disperse pollutants into the atmosphere (Nedel 2008).
In view of the aforementioned, this study becomes important for evaluating the influence of weather conditions on human health in the city of Pelotas, seeking a better understanding of comfort and thermal discomfort in occurrences of extreme temperature and humidity events associated with synoptic conditions, as well as how to mitigate health damage caused by them. Therefore, the objective of this study is to evaluate the extreme cases of human thermal discomfort in the city of Pelotas for the summer, fall and winter periods in the year 2019 and their relationship with the synoptic systems active during these events from the Analysis of data according to the seasons, evaluating the most appropriate indices for summer, fall and winter; Identification of the thermal comfort range of the five main macro-regions of the city of Pelotas; Relate the influence of synoptic conditions operating in the region to extreme cases of thermal comfort.
The area of this study is the city of Pelotas, which has an estimated population of 343,836 inhabitants (IBGE 2021) and is located in the southern portion of the state of Rio Grande do Sul (RS), in the extreme south of Brazil (31°46'19'' S, 52°20'34'' W), as shown in Fig. 1.
The urban perimeter is characterized by diversified topographies in the different Macro-regions (MRs) of the city: Areal, Central Region, Fragata, Laranjal and Três Vendas. The Central and Laranjal MRs have altitudes close to mean sea level, while the others have varied altitudes, being fundamental controls for the specific climate of the city.
Figure 1 Map of the Urban Zone of the municipality of Pelotas by macro-regions (MRs) and location of meteorological sensors
The data used in this study is part of the Project “Association between Weather Conditions and Physical Activity Levels in Children and Adults” in association with the COORTE (2015) project, developed by the Center for Epidemiological Research (CPE), at the Federal University of Pelotas (UFPel), which included the participation of 4-year-old children.
Air temperature, dew point temperature and relative humidity data from 9 am, 3 pm and 9 pm were used for the period from January 18 to August 20, 2019, obtained through five external meteorological sensors model HOBO Datalogger U23-001 located in different neighborhoods (See Supplementary Figure S1), characterizing the five main MRs in the city of Pelotas, shown in Fig. 1. The installation position of these sensors was suitable for each region, and the São Gonçalo and Barragem MRs were disregarded due to the lack of complete data on the sensors in these locations.
Due to the model of the sensors used for in situ data collection not measuring the variable wind speed, this data was obtained from the Conventional Meteorological Station (EMC) of the National Institute of Meteorology (INMET), situated in Capão do Leão (-31.80° S, -52.41° W), data is available at (https://portal.inmet.gov.br), at 9 am, 3 pm and 9 pm (12 UTC, 18 UTC and 00 UTC). The EMC was chosen because it is the closest station to the in situ data collection.
In order to obtain more adequate and consistent results, two different biometeorological indices were used, since the months studied correspond to different seasons. One of the thermal comfort indices used was the Wind Chill Index (WCI) presented in Eq. 1 (1), developed by Suping et al (1992). The WCI is an adaptation of the Effective Temperature (ET) of Missenard (1937) where the effect of the variable wind speed is considered. This index was applied for the fall months between April, May and June (AMJ) and for the winter months July and August (JA). Where, T is the air temperature (°C), RH the relative humidity (%) and WS the wind speed (m/s).
$$WCI=37- \frac{37-T}{0.68-0.0014RH+\frac{1}{1.76+ {1.4WS}^{0.75}}}-0.29T\left( 1-\frac{RH}{100} \right)$$
1
Gobo and Galvani (2012) showed that WCI inadequately represents thermal comfort in spring and summer months. For this reason, for the hottest months (January (J), February (F) and March (M)) the Humidex (HU) presented in Eq. 2 (2) was applied, which considers only the variables temperature and relative humidity of the air, given that the wind speed is negligible from high temperature values. It is also necessary to calculate the vapor pressure, using Eq. 3 (3), which includes the dew point temperature (DPT) variable. Where, HU is the Humidex temperature (°C), T is the air temperature (°C), e is water vapor pressure (WVP), DPT is the dew point temperature in K.
$$HU=T+\left(0.5555\right)\left(e-10\right)$$
2
$$e=6.11expexp \left(5417.7530\right) \left(\left(\frac{1}{273.15}\right)-\left(\frac{1}{DPT}\right)\right)$$
3
It is important to note that the biometeorological indices (WCI and HU) used in this study are simplified indices, that is, they represent human thermal comfort considering only meteorological variables, without involving personal variables.
The State of Rio Grande do Sul (RS) does not have a classification of thermal comfort zones suitable for the climate of the State, for this reason, the Perceived Equivalent Temperature (PET) classification levels, adapted by Monteiro and Alucci (2009) presented in Table 1, will be used. This classification is suitable for external environments in the city of São Paulo, and has already been successfully used in previous studies on Rio Grande do Sul (RS) (Gobo and Galvani 2012), where both biometeorological indices used will be analyzed through this classification.
Table 1 Distribution of comfort and thermal discomfort zones.
Synoptic Analysis
The synoptic analysis of the most uncomfortable thermal events, due to cold and heat, was carried out using data from the ERA5 reanalysis, the fifth generation of global climate reanalysis (Hersbach et al. 2020) belonging to the European Center For Medium-Range Weather (ECMWF), which has a resolution of 0.25° (~ 30 km). To that end, the variables of mean sea level pressure (MSLP), air temperature at 2 meters, wind at 10 meters, specific humidity at the level of 1.000 hPa and wind current lines at 850 hPa and 250 hPa were analyzed.
The results of the calculations performed with the biometeorological indices (HU and WCI) already mentioned, corresponding to the summer (JFM), fall (AMJ) and winter (JA) seasons in the morning, afternoon and night (9 am, 3 pm and 9 pm), in the five macro-regions (MRs) of the city of Pelotas/RS.
Through Fig. 2, it is possible to verify the distribution of thermal sensations perceived during the summer, fall and winter seasons in the five macro-regions of the city of Pelotas. In the summer months, the HU was used, and in the fall and winter months, the WCI was used, which showed very different behaviors. It was observed that in summer (Fig. 2a), there is a certain uniformity of thermal sensations between the macroregions, except for Areal MR, which presents an “outlier” (very high value), indicating the greatest thermal stress due to heat. The medians in all MRs were similar, as well as the maximum and minimum values of thermal sensation. The data distribution was above 30°C, indicating that the summer was predominantly marked by stress due to high temperatures. Central MR presented the lowest maximum values among the data series, considering Table 1.
Figure 2 Boxplot of thermal sensations for each season of the year and for each macro-region. Namely a) summer, b) fall and c) winter. The diagram shows the median (black line inside the box) and the minimum and maximum values. Data considered outliers is represented as (°)
The analyses for the fall season, shown in Fig. 2b, present outliers for minimum and maximum sensations, that is, periods of greater stress due to cold and heat, except for Laranjal MR, which presented only an outlier for minimum sensations, that is, more stressful values by cold. A uniform behavior in the concentration of data is observed. The medians also showed similar behavior. The Central MR showed greater outliers for the thermal sensation of heat and the MR with the least outliers (discomfort due to very cold weather) were observed in the Três Vendas MR. Thus, the behavior of thermal sensation during fall suggests that, although there are days (and periods of the day) with a sensation of thermal comfort, there is a strong indication of thermal stress due to cold, including outliers associated with negative temperatures sensations.
During the winter in Fig. 2c, the data distribution is a little smaller in the Laranjal MR. The Areal and Central MRs presented outliers for maximum thermal sensation. The winter season intensifies even more the thermal discomforts due to cold, as well as in the fall, lower values of thermal sensations were obtained in the Três Vendas MR with negative thermal sensations.
Through Fig. 3a, it was possible to observe that in the summer months, all MRs had similar behavior, with few variations. In relation to the fall and winter months, this difference was greater, which may be associated with the location of each sensor and the environment in which it is installed (afforestation, soil influence, surrounding conditions, topography, etc.).
In the analyses carried out by shifts, it was observed that the summer months present greater thermal comfort at night, and greater thermal stress due to high temperature in the afternoons, which is directly associated with solar radiation. During the winter, the opposite happens, during the afternoons there are greater occurrences within the comfort range, while at night they become more thermally uncomfortable.
Figure 3 Chart of monthly averages in the period from January 18th to August 20th, from 9 am, 3 pm and 9 pm of the a) Humidex (HU) and Wind Chill Index (WCI), b) Air Temperature (T) and Relative Humidity (RH)
To justify the variations in thermal comfort between the MRs observed in Fig. 3a, we sought to evaluate the behavior of the monthly averages of temperatures and relative humidity (Fig. 3b), where it was observed that in relation to temperatures, there are no significant variations. The relative humidity variable, on the other hand, presents a greater variation in different situations, both by shifts and by MR, and an increase in amplitude was obtained as the colder months approached.
Supplementary Figure S2 shows the "Heat map" of the existing correlations between the variables thermal comfort (TC), temperature (T), relative humidity (RH), dew point temperature (DPT - only for summer period) and wind speed (WS - only for winter period) for 9 am. In Figure S2a, the correlations for the summer months are observed, where correlations above 90% and very significant between TC and the variables T and DPT, on the other hand, the RH variable mostly does not present statistical significance when compared to TC.
In Figure S2b, the correlations for the winter months shows that the wind speed (WS) presents inverse correlations with the TC, which means that the higher is the WS, the lower is the TC, while the RH showed low correlations with TC and/or did not show statistical significance.
For the 3 pm period, supplementary Figure S3a shows the correlations for the summer months, where significant correlations can be seen between all variables, with the exception of Laranjal MR where there were no significant correlations between TC and RH and DPT and RH. However, when analyzing T and RH, the variables present negative correlations, demonstrating opposite behavior, which means that the increase in T is associated with a decrease in RH. In S3b, the correlations of the variables for the winter months are observed, where the variable WS presents inverse correlations with the TC, indicating that under higher intensities of wind speed, the thermal comfort decreases. Considering the relative humidity (RH), it does not present statistical significance associated with WS, but it presents significance to all the other variables. The variable temperature (T) presents positive correlations when compared with TC and negative correlations when correlated with WS. Laranjal MR did not obtain statistical significance in any of the variables studied in relation to RH, which may be associated with the proximity of Laguna dos Patos.
In the “Heat map” for the 9 pm period in the summer months (Supplementary Figure S4a), significant correlations are observed between DPT and all the analyzed variables, showing that DPT has a greater relationship with TC in this period when compared to the other shifts. The RH presents significant correlations with the TC variable only in Laranjal and Areal MRs, while in the others it did not obtain statistical significance, possibly due to the characteristics of the location where the sensors were installed. When the variable T is analyzed, it demonstrates similar behavior to DPT, presenting significant values of correlations in all the variables analyzed, with the exception of the variable RH where it did not present statistically significant values.
For the winter months at 9 pm (supplementary Figure S4b), most of the results showed positive and significant correlations, except for the variable T when it’s correlated with the RH in the Três Vendas MR, where it did not show statistical significance. The WS variable is significant in relation to all other variables, demonstrating that its increase directly implies thermal discomfort in the city of Pelotas in all MRs in the winter period at night.
Table 2 presents the maximum and minimum values obtained for the summer months using the HU. It is observed that the maximum peaks occurred in the morning and night shifts on the Fragata MR, with 48.20°C and 42.34°C of thermal sensation, respectively. The afternoon period reached a maximum of 52.02°C at Areal MR. Minimum HU values were observed in the Central MR in the morning and afternoon, with 17.27°C and 24.37°C of thermal sensation, while at night the minimum recorded was 17.31°C in the Laranjal MR.
Table 2 Maximum and minimum values of HU, in bold the maximum and minimum values recorded in each shift for the summer months (JFM) of 2019 and maximum and minimum values of WCI thermal sensation, highlighting the maximum and minimum values recorded in each shift for the fall (AMJ) and winter (JA) months.
The same analysis was performed for the fall (AMJ) and winter (JA) months, using the WCI. Table 2 also shows the morning maximum at Laranjal MR with 27.61°C, while for the afternoon and night periods the maximums were obtained at Central MR, with values of 31.72°C and 19.91°C, respectively. Minimum WCI values were observed with − 7.74°C in the morning on the Fragata MR, -3.56°C on the Laranjal MR and − 10.19°C on the Três Vendas MR.
In Supplementary Figure S5a, the temporal variation of the Humidex (HU) was observed referring to the 9 am. The red lines represent the range of perceived thermal comfort (18°C − 26°C). In general, values of thermal stress due to predominant heat in the data series (> 26°C - above the comfort zone) were obtained in all MRs. The peak of maximum thermal sensation occurred on January 29, on the Fragata MR with a thermal sensation of 48.20°C, as previously mentioned. During this period, no thermal stress due to cold (< 18°C) was observed. Few days were observed within the comfort zone, starting to have a higher frequency at the end of March, when summer is coming to an end.
In Supplementary Figure S5b, we have the results for the 3 pm period, which represents the afternoon, which presented a similar pattern to the morning shift, with predominant thermal stress throughout the data series and without any cold discomfort. All MRs were found with HU values more aligned during this period, without major differences from each other. It is noteworthy that in this shift, values below 25°C were not obtained, unlike the morning shift, which showed a decrease in thermal stress due to heat at the end of March, with the end of summer.
Supplementary Figure S5c shows the period from 9 pm, which represents the night period. Different from the previous shifts, this one presented some changes in relation to the morning and afternoon: although a large part was still under thermal stress due to heat, there were higher values recorded within the comfort zone, mainly in the month of March. This can be explained due to the fact that at night the surface is not being heated through solar radiation, making it easier for the temperature to have a little decrease and the perceived thermal sensations to be closer to what is considered pleasant, together with the end of the summer season.
The temporal variation of the Wind Chill Index (WCI) can be observed in Supplementary Figure S6a, referring to the time of 9 am for the fall and winter months. It was noticed that for these months, the values of thermal stress due to cold are predominant in the morning, where the month of April was the one that obtained the most values in the comfort zone. The minimum thermal sensation was − 7.74°C and occurred on July 6th on the Fragata MR, and a maximum of 27.61°C on the Laranjal MR on April 1st.
In Supplementary Figure S6b, the WCI is observed for the afternoon period (3 pm) in the fall and winter months. Unlike all other analyses, the afternoon at these times of the year showed the highest occurrences of values within the comfort zone. It is observed that in the months of April to June, the great majority of the data were found predominantly within the comfort zone. From July to August the events of colder thermal sensations occurred, in view of the arrival of winter. The episode of biggest cold stress occurred on July 5 at Laranjal MR, with − 3.56°C of thermal sensation and the biggest heat stress occurred on April 2 at Central MR.
In Supplementary Figure S6c, the WCI is observed for the night period (9 pm) in fall and winter months. It is observed that in winter the thermal sensations were below 18°C, with a few exceptions. The maximum wind chill occurred on April 2nd (19.4°C) and the minimum was − 10.1°C on June 6th.
Case studies of the most thermally uncomfortable events
The occurrences of thermal (dis)comfort were analyzed through Supplementary Figure S7, which shows the histograms for the summer, fall and winter seasons in the three analyzed shifts. In histograms, the base of each of the bars represents a class and the height represents the amount or absolute frequency with which the value of each class occurs. Data analysis allows identifying that there are peaks of thermal discomfort due both cold and high temperatures in the city of Pelotas, and, thus, the sequence of days with bigger thermal discomfort recorded in the analyzed period was filtered, following the classification of Perceived Equivalent Temperature (PET), previously shown in Table 1.
The highest occurrences are observed during the summer (S7a), ranging from 31°C to 41°C. There were 118 records of sensations above 41°C, characterized by the classification as “Very Hot”. During the fall season (S7b), the largest amount of data was found between the values of 12°C to 26°C of thermal sensation, showing to be the season with the highest records within the comfort range. In winter, there is a predominance of thermal sensation between the range of 5°C to 18°C (S7c). There were 35 times of negative sensations, being classified through PET as “Very Cold”.
From a total of 645 hours analyzed from January 18 to August 20, 2019, at 9 am, 3 pm and 9 pm, it was observed that the MRs with the longest hours within the range of thermal comfort were the Central MR with 17.9% and Areal with 16.9%. The other MRs obtained percentages of 14.4% in Laranjal, 11.3% in Fragata and Três Vendas with 9.3%, this last one being the most thermally uncomfortable MR.
Synoptic Analysis - January 28, 2019
According to the dates/times found by the frequency distribution, the dates with extreme thermal discomfort in the city of Pelotas were selected. According to the meteorological sensors used in the research, on 01/28/2019 the maximum temperature of 39.94°C, relative humidity of 43.1% and the HU index showed the sensation of thermal stress of 52.02°C.
Figure 4 Fields of air temperature at 2m a), pressure (hPa) at mean sea level b), wind direction and magnitude at 10m c), specific humidity at 1.000 hPa d), streamlines at 850 hPa e), streamlines at 250 hPa f), layer thickness (500–1.000 hPa) g) and geopotential height at 500 hPa h), all referring to January 28, 2019, at 3 pm (local time)
Figure 4 shows the analyses for extreme discomfort due to high temperatures, based on the meteorological fields at 3 pm (local time) on January 28, 2019. In the air temperature field at 2m (Fig. 4a), it is possible to observe that high temperatures predominate throughout the state of Rio Grande do Sul, with temperatures ranging from 30°C to 35°C, while the maximum recorded by the sensor was 39.94°C. The mean sea level pressure (MSLP) field (Fig. 4b) shows the influence of the South Atlantic Subtropical Anticyclone (SASA) exerting anticyclonic flow over a large part of the continent, including the southern region of Brazil.
In the region of Argentina, a cyclonic system was observed, typical of the low thermal pressure active during the summer in the region, but in this case, it does not exert a direct influence on the State of Rio Grande do Sul (RS). The wind field at 10m (Fig. 4c) demonstrates the influence of the North/Northwest flow in the South region of Brazil, which is being directly aided by the SASA system, which when entering the continent (12°S 50°W) acquires characteristics continents, due to the low specific humidity prevailing in the continent (Fig. 4d). This flow transports hot and dry air from the central region of Brazil to the south, helping to warm the region.
In Fig. 4d, low values of humidity prevail throughout the southern region of Brazil and much of the national territory, which is in agreement with the other variables previously evaluated, since the day under analysis represents high temperatures and low humidity conditions.
Although low-level jets are frequent under this type of condition analyzed (north winds, thermal low in northwest Argentina), the influence of this type of phenomenon was not observed in Fig. 4e. At the level of 850 hPa, the winds are observed from the west and their intensity does not match this type of situation. In Fig. 4f, it is possible to notice the subtropical jet acting intensely, intensifying the SASA system in the Pacific, showed no significant influence for the conditions studied either.
The layer thickness field (500–1.000hPa) (Fig. 4g) demonstrates great warm advection across the continent, consistent with situations previously seen on the surface, as well as the geopotential height field at 500 hPa (Fig. 4h) does not indicate any type of vorticity advection, nor crests and troughs acting in the State.
Therefore, it is observed that the conditions of thermal discomfort in the analyzed period were associated with the action of the atmospheric blocking system, which caused a sequence of days under conditions of high temperatures and low humidity, due to the influence of a dipole in the Pacific Ocean, being intensified by the subtropical jet stream.
Synoptic Analysis - July 6, 2019
On this day, in the city of Pelotas, according to the meteorological sensor, a minimum temperature of 3.9°C and relative humidity of 88.9%. The wind speed was 5.6 m/s. The thermal sensation according to the Wind Chill Index (WCI) was − 10.19°C. The following day, July 7, freezing rain was recorded in several places of the city.
Through the air temperature field at 2m (Fig. 5a) low temperatures can be seen in all regions of the State, with temperatures varying between − 5°C and 14°C. This corroborates what is shown in Fig. 5b, where the presence of a polar air mass is observed over the region, which transports cold and dry air and causes a drop in temperatures. The wind field at 10m (Fig. 5c) indicates winds from the West/Northwest in Rio Grande do Sul.
Figure 5 Fields of air temperature at 2m a), pressure (hPa) at mean sea level b), wind direction and magnitude at 10m c), specific humidity at 1.000 hPa d), referring to July 6, 2019, at 9 am (local time). Streamlines at 850 hPa e), streamlines at 250 hPa f), layer thickness (500–1.000 hPa) g) and geopotential height at 500 hPa h), referring to July 6, 2019, at 9 pm (local time)
Air humidity (Fig. 5d) is consistent with the variables previously analyzed, since the system operating in the region has polar high characteristics, which transports colder and drier air. The impact of this variable on temperature consists in the fact that in a humid atmosphere, water vapor stores and emits heat through longwave radiation towards the surface. In dry conditions, however, this phenomenon does not occur, which allows a strong decline in air temperature (Ynoue et al. 2017).
Regarding the geopotential height field (Fig. 5h) for the region of interest, it is notable the presence of a ridge under the Argentina region (30°S 60°W), which exerts anticyclonic vorticity advection east of the ridge, located in the state of Rio Grande do Sul. This vorticity feeds the anticyclonic system acting under the study region. On the other hand, at the level of 850 and 250 hPa (Figs. 5e and 5f), no relevant characteristics that are involved in the situation acting in the wind field were observed.
Considering the appropriate indices for each season of the year in the city of Pelotas, it is concluded that most of the time the region is in situations of thermal discomfort. The analysis by shifts showed that in summer months, there is greater thermal comfort at night, and bigger thermal stress due to high temperatures in the afternoons, which may be associated with solar radiation, since during the afternoons the temperature maximums are recorded, unlike the nights, making afternoons with more uncomfortable sensations due to high temperatures and nights with sensations closer to thermal comfort. During winter, the opposite of summer happens, during the afternoons the highest occurrences of thermal sensations occur within the comfort range, while the nights are the most thermally uncomfortable.
Observing the three different periods analyzed in the winter and summer seasons for the five MRs studied, it was noted that during the summer months the TC is directly correlated with T and DPT in the three analyzed shifts. While for the winter months, the highest direct correlations of TC were with the temperature (T), and inverse with the wind speed (WS), demonstrating that the increase in the intensity of the WS directly influences the thermal discomfort in all MRs and in all the shifts. All correlations considered showed statistical significance at levels of 99% (p < 0.001) and 95% (p < 0.05), rejecting the null hypothesis for all cases, and it can be said that it has found a statistically significant difference.
In the summer, the minimum values of HU were observed at the Central MR in the morning and afternoon, with 17.27°C and 24.37°C of thermal sensation, which is somehow surprising, since the MR is located in an area with asphalt and large buildings, with a large flow of vehicles that could influence the local temperature. However, a justification for such results may be associated with the location where the sensor of this MR was installed, which was located below a water tank under shade, in a very wooded place.
Regarding the case studies, it was found that the extreme discomfort caused by high temperatures had a direct influence on SASA, which influenced the temperature conditions in the city of Pelotas. It is important to note that the maximum discomfort observed occurred with the combination of high temperature and low humidity, which is surprising, since commonly the greatest discomfort due to high temperatures occurs with the combination of high temperature and high humidity. A justification for this is due to the single time analyzed (3 pm), which ends up making a large cut in the data and, as a result, losing information from other times in the afternoon. On the other hand, extreme cold discomfort was observed in a post-frontal situation, where Rio Grande do Sul was under a polar bullish system. The low temperatures combined with the dry air, culminated in the occurrence of negative temperatures throughout the city of Pelotas.
In general, Central MR and Areal MR were the ones that obtained the highest percentage of hours within the range of thermal comfort, being the most comfortable places in the city, with percentages of 17.9% and 16.9%. As for the least comfortable, the Fragata MR and Três Vendas MR were observed, where the percentages were between 11.3% and 9.3%, respectively. Laranjal MR demonstrated about 14.4% of hours within the thermal comfort zone.
This research sought to emphasize how the different MRs in the city of Pelotas may have different characteristics and with what factors they may be associated, as well as to seek under which conditions the city experienced the biggest discomforts in 2019. As a result, this study becomes important in the future to develop mechanisms that can serve as a warning to the community and mitigate possible worsening of existing illnesses.
Acknowledgments
The authors thank the Brazilian National Council for Scientific and Technological Development (CNPq) for funding and support through the call Universal MCTI/CNPq N. 01/2016, the entire team and third parties involved in the COORTE project of the Epidemiological Research Center of Federal University of Pelotas (UFPEL) and the partial support of PRPPG/UFPel and CAPES in the publication of this article.
- Fanger PO (1970) Thermal comfort: analysis and applications in environmental engineering. New York: McGraw Hill Book Company, pp 244. https://doi.org/10.1177/146642407209200337
- Giles DB, Balafouts C, Maheras P (1990) Too hot for comfort: the heatwaves in Greece in 1987 and 1988. International Journal of Biometeorology 34:98–104. https://doi.org/10.1007/BF01093455
- Gobo JPA, Galvani E (2012) Application of the Wind Chill Index (WCI) in thermal comfort studies for the state of Rio Grande do Sul. Revista Geonorte, Edição Especial 2, 1:403 – 413.
- Hersbach H et al (2020) The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 1999-2049. https://doi.org/10.1002/qj
- Hoppe PR (1999) The physiological equivalent temperature: a universal index for the assessment of the thermal environment. International Journal of Biometeorology 43:71-75. https://doi.org/10.1007/s004840050118
- IBGE – Instituto Brasileiro de Geografia e Estatística. https://www.ibge.gov.br/ Accessed 01 June 2021.
- International Society Biometeorology. Definition and classification of biometeorology. http://biometeorology.org. Accessed 25 June 2021.
- Masterton J, Richardson FA (1979) Humidex: a method of quantifying human discomfort. Environment Canada. Downsview: atmospheric environment service, Ontario.
- Missenard A (1937) L’Homme at lê climat. Paris: Plon, 1937.
- Missenard A (1948) Equivalences thermiques des ambiances; equivalences de passage; equivalences de sejour. Chaleur set Industrie, Juillet Août.
- Monteiro LM, Alucci MP (2009) Territorios y espacios urbanos sustentables: confort ambiental en espacios abiertos. Ambiente Construído 3:1- 26. https://doi.org/10.29097/26191709.301
- Nedel AS (2008) Meteorological Conditions Favorable to the Occurrence of Respiratory Diseases in Children in the City of São Paulo Thesis, Institute of Astronomy, Geophysics and Atmospheric Sciences of São Paulo, University of São Paulo.
- Nedel AS, Alonso MF, Freitas, RAP, Trassante FC, Silva HN, Bortolli E, Medeiros MAF, Hallal PC, Vianna JCT (2021) Analysis of indoor human thermal comfort in Pelotas municipality, extreme southern Brazil. Int J Biometeorol 65:419–428. https://doi.org/10.1007/s00484-020-02015-7
- Steadman RG (1979) The assessment of sultriness. Part I: a temperature- Humidity index based on human physiology and clothing science. Journal Appl. Meteorology 18:861–873. https://doi.org/10.1175/1520-0450(1979)018<0861:TAOSPI>2.0.CO;2
- Suping Z, Guanglin M, Yanwen, W, JI L (1992) Study of the relationships between weather conditions and the marathon race, and of meteorotropic effects on distance runners. International Journal of Biometeorology 36:63-68. https://doi.org/10.1007/BF01208915
- Ynoue RY et al (2017) Meteorologia: noções básicas. Oficina de Textos.
