Modeling heat stress changes based on wet-bulb globe temperature in respect to global warming

Abstract

Background

This ecological study aims to model the trend of changes in exposure of outdoor workers to heat stress in outdoors in the coming decades with the use of the Wet-Bulb Globe Temperature (WBGT), Hadley Coupled Atmosphere- Ocean General Circulation Model, version 3 (HADCM3), and Long Ashton Research Station Weather Generator (LARS-WG) in Tehran, Iran, considering the climate change and the global warming.

Methods

The hourly values of environmental parameters including minimum and maximum air temperature, relative humidity, precipitation and radiation related to Prakash , Shahriar and Damavand cities were obtained from the Meteorological Organization of Iran. These data were recorded during 1965 to 2015. The climate modeling was done for 2011–2030, 2046–2065, and 2080–2099.

Results

The minimum and maximum air temperatures in the different months of the year in the three studied cities show an increasing trend. Our finding shows that the WBGT will be increased by 2099. In Pakdasht, this index will be close to the danger zone in the coming years, especially in 2080–2099.

Conclusions

All the results obtained indicate an increase in risk of heat stress in outdoor workplaces, given the global warming.

Background

Exposure to heat stress in workplaces has become one of the concerns for workers and health professionals [1]. Estimates suggest that about 16 million Iranian workers are engaged in two million occupational units, of which 45% account for the service workforce, while 30% and 25% of them are working in the agricultural and industrial sectors, respectively. They are exposed to different health hazardous agents in different ways. Thus, it can be stated that the number of workers engaged in outdoor jobs is much higher (about three times more) than those of indoors [2].

The outdoor workers, who are exposed to direct sunlight for long time a day, may also be subjected to hot environments in some jobs such as asphalting, surface mines, brick kilns, petrochemicals. Evidently, these workers are at a high risk of heat stress [3]. Besides inappropriate weather conditions, their jobs usually involve high physical activities [4].

Short-term exposure to extreme heat (acute exposure) can lead to an increase in deep body temperature and heat-related illnesses such as mild rash, muscle cramps, heat exhaustion and heat shocks [5]. Moreover, the reported effects of occupational chronic exposure to heat include cardiovascular diseases [6], mental health issues [7] and kidney disorders [8]. Working in hot environments also increases the risk of injury and occupational accidents [9]. The increased body temperature and decreased body fluids can result in physical fatigue, irritability, lethargy, impaired judgment, loss of consciousness, decreased agility, loss of concentration and coordination, and blurry vision; all of them may adversely affect workers’ performance, safety and efficiency [10-13].

Climate is affected by increased levels of greenhouse gases in the atmosphere. This leads to an increase in the Earth’s temperature which is referred to as global warming. According to numerous studies, there has been a positive trend in global temperature [14]. Climate change includes global warming and the “side effects” of warming like heat waves, which are putting many people such as outdoor workers at risk of heat stress, heat-related illness and even death, especially in the warm seasons [15, 16]. Today, researchers tend to focus on predicting climate change in order to estimate its global and regional side effects [17].

Measuring and computing the climate parameters such as temperature and precipitation is necessary in climate change studies; besides, heat stress indices have been used to determine the heat stress levels in workplaces [18]. Approximately 45 to 60 indices exist for the assessment of heat stress. One of the most applicable and reliable indices is the Wet-Bulb Globe Temperature (WBGT) [4]. This index was introduced by Minard and Yaglue in 1957 and established in 1989 as the ISO-7243 standard [19, 20].

This ecological study aims to model the trend of changes in exposure of outdoor workers to heat stress outdoors in the coming decades with the use of the WBGT, Hadley Coupled Atmosphere- Ocean General Circulation Model, version 3 (HADCM3), and Long Ashton Research Station Weather Generator (LARS-WG) in Tehran, Iran, considering climate change and global warming.

HadCM3 is a coupled atmosphere - ocean model presented by Gordon et al. [21]. This model has been implemented at the Hadley Center for Climate Prediction and Research (HCCPR), UK. HadCM3 is introduced as a grid-point model with a horizontal resolution of 2.5 degrees in latitude by 3.75 degrees in longitude. In this model, the simulations are based on a 360-day year and 30-day months. The high resolution of the ocean component is the most important advantage of the model. Another advantage is the good coordination between atmosphere and ocean components. The IPCC has used HadCM3’s general circulation model data to prepare the third report [22]. In spite of many advantages, the HadCM3 model has a low spatial resolution. In order to overcome this problem, the outputs need to be scaled up before being used in a climate-change impact assessment [23]. Additionally, in order to further address the issue in general atmospheric circulation models, statistical downscaling and dynamic downscaling are used. Statistical downscaling methods require less parameters than dynamic ones and have higher process speeds. It is also possible to apply them without using supercomputing, which is why they are very much considered in climatic studies. One of the models used in statistical methods is the LARS-WG model. This model is extensively used in predicting and modeling future climate change [24].

Methods

Study area

Covering an area of ​​18,814 square kilometers, the Tehran Province is located in the north of the central plateau of Iran (35.7117 N 51.4070E). This Province has 16 towns. Tehran is located between a mountainous region and a plain. Two factors play important role in the climate of Tehran, including Alborz mountain range and western humid winds [25].

The Tehran Province can be divided into three climatic zones:

• A.North highlands climate

In the southern range, the central Alborz Mountains are located at an altitude of over 3,000 m with humid and semi-humid climate and very cold long winters. The most prominent town for this zone is Toochal.

• BFoothill climate

The climate is specific to an altitude of 2,000 m above sea level, with semi-humid weather and cold and relatively long winters. Damavand, Firouzkooh, Ab Ali, Lavasan and Taleghan enjoy this climate.

• CSemi-arid and arid climates

With short winters and hot summers, this climate is specific to altitudes less than 1,000 m. The lower the height, the drier the environment. Varamin, Shahriar and Southern Karaj are affected by this climate [20].

Therefore, given the high frequency of outdoor workplaces such as surface mines, agriculture industry and construction sites, Pakdasht, Shahriar and Damavand cities have been investigated. These cities are shown in Table 1.

Table 1.

Location of the stations studied in the Tehran Province [21]

No.	Station	Location (deg N)	Latitude (deg E)	Longitude	Height above sea level (m)
1	Pakdasht	Southeast	35.47	51.67	1200
2	Shahriar	West	35.65	51.06	1160
3	Damavand	North East	35.72	52	2300

Collecting the data

Hourly values of environmental parameters including minimum and maximum air temperature, relative humidity, precipitation and solar radiation at a height of 2 meters were obtained from stations of the Meteorological Organization of Iran. These data included data recorded in the Pakdasht, Shahriar and Damavand meteorological stations during 1965 to 2015.

Hadley Coupled Atmosphere - Ocean General Circulation Model (HadCM3)

In order to predict the climate in the coming decades, HadCM3 was used. General circulation models are the most reliable tools for investigating the effects of the climate-change phenomenon (atmospheric response to an increase in the greenhouse-gas concentrations) on different systems. These models are able to model climatic parameters for a long period of time using scenarios approved by the Intergovernmental Panel on Climate Change) IPCC((A1B.B1, A2, A1) [26].

These predictions are based on a variety of scenarios, each of which has various hypotheses about population growth, economic development, technology development, living standards and existing energy production options, which are also referred to as the emission scenarios. The scenario used in this study is known as A1B. In this scenario, emphasis is on the balanced use of various energy sources, low population growth and rapid growth of new technologies [27]. The LARS-WG5 model was used to downscale the HadCM3 output.

Long Ashton Research Station Weather Generator (LARS-WG)

Racsko et al. [28] developed the original version of LARS model in Budapest as part of an agricultural risk project, and then Semenov and Barrow improved the model [29]. LARS is a stochastic generator of random meteorological data, which is used to generate daily data of precipitation, daily radiation and maximum and minimum daily temperatures at a station under present and future climate conditions [22]. In LARS, modeling the precipitation and its probability are performed by semi-empirical probability distribution and Markov chain. The modeling of radiation and temperature are based on semi-empirical probability distribution and Fourier series, respectively [30]. This model is highly precise in generating weather data (air temperature and precipitation) as reported for 18 stations of Canada [31] and 20 stations in different regions of UK [32, 33]. Data is generated by the model in three phases: calibration, evaluation and generation of weather data.

In order to generate data, the characteristics of each station including name, location, altitude, and daily meteorological data file must be first provided as input to the model. Then, data is analyzed and the output file contains the statistical characteristics of the data including simulated monthly averages during the entire study period. In this study, the climate modeling was done for 2011–2030, 2046–2065, and 2080–2099. The generated data in the periods mentioned above were used for minimum and maximum temperatures.

Calculating WBGT

The WBGT was calculated according to [34].

$$\mathit{\text{WBGT}}=0.567\times t_a+3.94+0.393\times E$$ 1

$$E=\frac{\mathit{RH}}{100}\times 6.105\times {\mathrm{e}}^{17.27\times \frac{t_a}{237.7+t_a}}$$ 2

where, t_a: average air temperature (^oC), RH: Relative humidity (%), E: Vapor Pressure of Water (hpa).

Owing to the fact that in the LARS model, the relative humidity parameter is not predicted, the correlation coefficient between the average air temperature and relative humidity in the meteorological data from 1965 to 2015 was calculated. Then, a linear regression equation was used as a dependent variable in order to calculate the relative humidity (Table 2). Data was analyzed using SPSS 20 and Excel 2013.

Table 2.

Pearson correlation coefficients (r) and regression relations between air temperature (t_a) and relative humidity (RH) for 1965–2015 data at the studied meteorological stations

City	r	Linear regression equation
Pakdasht (Doushan Tapeh)	0.801	RH= -1.4571 × ta + 69.383
Shahriar (Karaj)	0.748	RH= -1.5061 × ta + 68.318
Damavand (Firuzkuh)	0.665	RH= -1.2747 × ta + 61.441

Results

The output results of the model for the three cities over three periods are presented in Tables 3, 4 and 5. As can be seen in the Tables 3, 4 and 5, it is evident that over the entire year for the three cities, the minimum and maximum air temperatures will increase in the coming years. Among all three cities are located in the Tehran Province, the temperature increase in Pakdasht will be greater than the other two cities. In Pakdasht, the changes in the maximum air temperature indicated an average increase up to 3–4 °C by 2099. These changes will reach the maximum level in July and August. For the minimum air temperatures, the results showed that the highest change (4–4.5 °C) will occur in July and August by 2099. In Shahriar, changes in the maximum temperatures also indicated an average increase up to 3–4 °C by 2099, which will reach the maximum level in July and August (4.3 °C). The results indicate the maximum change (4.2 °C) by 2099 in July and August for minimum air temperatures. In Damavand, the average change in maximum and minimum air temperatures by 2099 is approximately 2.65 °C. These changes will reach the maximum level in July and August (4 °C).

Table 3.

Predictions for minimum and maximum air temperatures (°C) based on the LARS and HADCM3 models for the coming years in different months of the year for Pakdasht

Month	ta	2011–2030	2046–2065	2080–2099
January	Min	1	2.17	3.1
	Max	8.81	9.96	11.3
February	Min	2.44	3.6	3.9
	Max	11.35	12.49	13.1
March	Min	7.34	8.8	9.7
	Max	16.53	18	18.8
April	Min	12.82	14.6	15.84
	Max	23.44	25.2	26.4
May	Min	18.15	19.8	21.8
	Max	29	30.64	32.7
June	Min	23.52	25.1	26.8
	Max	34.98	36.5	38.2
July	Min	26.67	28.83	30.64
	Max	37.96	40.2	42
August	Min	25.4	27.7	30
	Max	36.98	39.3	41.62
September	Min	21.65	23.3	24.6
	Max	33.3	34.95	36.15
October	Min	14.77	16.1	17.2
	Max	25.13	26.4	27.62
November	Min	8.05	9.47	10.52
	Max	17.16	18.6	19.5
December	Min	3.26	4.84	5.84
	Max	10.88	12.46	13.5

Table 4.

Predictions for minimum and maximum air temperatures (°C) based on the LARS and HADCM3 models for the coming years in different months of the year for Shahriar

Month	ta	2011–2030	2046–2065	2080–2099
January	Min	-2	-1	0.1
	Max	6.16	7.16	8.51
February	Min	− 0.04	0.95	1.4
	Max	9.46	10.47	11.02
March	Min	4.81	6.1	7.35
	Max	15.65	16.92	18.55
April	Min	9.45	10.94	12.02
	Max	21.42	22.92	24.26
May	Min	13.9	15.42	17.15
	Max	28	29.6	31.2
June	Min	17.75	19.4	21.37
	Max	33.8	35.45	37.24
July	Min	20.27	22.33	24.5
	Max	36.16	38.22	40.25
August	Min	19.92	22	24.11
	Max	35.5	37.5	39.8
September	Min	16.23	17.76	19.5
	Max	31.37	32.9	34.3
October	Min	11.9	12.93	14.2
	Max	24	25.05	26
November	Min	5.54	6.7	7.5
	Max	14.33	15.48	16.49
December	Min	0.7	1.9	2.84
	Max	8.16	9.4	10.61

Table 5.

Predictions for minimum and maximum air temperatures (°C) based on the LARS and HADCM3 models for the coming years in different months of the year for Damavand

Month	ta	2011–2030	2046–2065	2080–2099
January	Min	-8.84	-7.75	-6.8
	Max	2.51	3.6	4.5
February	Min	-7	-5.9	-5.2
	Max	4.73	5.8	6.2
March	Min	-1.8	− 0.5	0.6
	Max	11.26	12.6	13.65
April	Min	2.3	3.86	5
	Max	16.83	18.37	19.55
May	Min	6.72	8.1	9.8
	Max	22.36	23.72	25.46
June	Min	10.52	11.7	13.3
	Max	27	28.3	29.8
July	Min	14.44	16.3	18.2
	Max	30.2	32	33.9
August	Min	13.26	15.26	17.58
	Max	29.83	31.82	34.1
September	Min	8.82	10.4	11.82
	Max	26.2	27.8	29.11
October	Min	2.88	3.96	5.3
	Max	19.7	20.73	22.1
November	Min	-2.33	-1.12	− 0.26
	Max	11.57	12.8	13.55
December	Min	-6.24	-4.85	-3.9
	Max	6.06	7.5	8.4

According to Figs. 1, 2 and 3, it is obvious that in all months of the year at the studied cities, the average temperature will increase in the coming years. In addition, this increase in temperature will be greater in the summer than other seasons. The highest increase in temperature is reported in June and July. The results indicate an increase of 3 °C by 2099 in the average temperature.

Fig. 1.

Intra-annual variation of the monthly mean air temperature over three future periods for Pakdasht

Fig. 2.

Intra-annual variation of the monthly mean air temperature over three future periods for Shahriar

Fig. 3.

Intra-annual variation of the monthly mean air temperature over three future periods for Damavand

Furthermore, the average WBGT was investigated in the cities over different months; the results are shown in Figs. 4, 5 and 6. In Pakdasht, this index will be close to the danger zone in the coming years, especially in 2080–2099. In July and August, WBGT will increase by 3.2 °C in all three cities. In Damavand, WBGT will trend up to 2.7 °C in August. Among the cities studied, Pakdasht has the highest WBGT. The results also indicated that the average monthly WBGT index in Pakdasht in June, July and August is mainly classified in the hot region (extreme caution, 24–28 °C) and even in the danger zone. The average monthly WBGT in Shahriar in July and August was found in the hot region. Also, in Damavand, the average monthly WBGT index is located in the warm region (warn, 18–24 °C) in June, July, August and September.

Fig. 4.

Intra-annual variation of the WBGT over three future periods for Pakdasht

Fig. 5.

Intra-annual variation of the WBGT over three future periods for Shahriar

Fig. 6.

Intra-annual variation of the WBGT over three future periods for Damavand

Discussion

Climate change and global warming are great environmental problems today, affecting all people on Earth. Given the rising global temperature in the 21st century due to the climate change, the rise of heat stresses is becoming more important, especially in outdoor workers such as miners, road builders, farmers and urbanists. They are usually forced to use the personal protective equipment like shoes, helmets, respiratory masks, and earmuff. Wearing this protective equipment and doing the heavy physical works, especially in the warm season, can increase the risk of heat strain [35]. Given the climate change and global warming, this study aimed to model the heat stress for outdoor workers based on the WBGT over the coming decades using HADCM3 and LARS-WG in the Tehran Province. Three major cities of the Tehran Province with climate diversity were studied, in which there are several outdoor workers, especially in agriculture, open surface mines and construction sites [20].

According to the output results of the LARS weather generator in the A1B scenario, the minimum and maximum air temperatures in the different months of the year in the three studied cities showed an increasing trend. The temperature will increase significantly in summer compared to the other seasons. Additionally, the increase in temperature will be much more evident in July. The results show an increase of 4 °C by 2099 compared to previous years. All obtained results indicated an increase in temperature, given the global warming. In consistent with present study, the other research used the weather data (1988–2005) [25] to model the Tehran’s climatic pattern using the LARS model. The weather was forecasted for 2010–2039. In Tehran, the average monthly temperature would increase by about 0.2 °C. The highest increase is estimated in January (about 0.8 °C).

Petkova et al. [36] predicted the effect of heat-related mortality in three major Northeast US cities (Boston, New York, and Philadelphia) using AR5 models. The heat-related mortality rates were higher in New York City in 2020, 2050, and 2080, followed by those in Philadelphia and Boston. This indicated significant vulnerabilities and adaptability challenges [36].

Ghorbani [37] in the study entitled “assessing spatial and seasonal pattern in climate change, temperatures across Iran” which was based on the data from the National Center for Environmental Prediction (NECP) under the A1B scenario, pointed out that temperatures will change in Iran. These changes will occur throughout Iran, but the rate of change will not be the same in all regions. The results also showed that in all seasons of the year, the air temperature will be rising over the coming years. The temperature will reach the highest rates in the summer (with an increase of up to 0.04 °C per year). In winter, most of the regions in Iran will witness a temperature rise of 0.02–0.04 °C per year [37].

In a study by Masuodian [38], it was concluded that over the last half of 21th century, the night time and daily temperatures of Iran will increase by about 3° and 1°, in 100 years, respectively. The rising trend of temperature has been seen mainly in warm and low-altitude locations.

According to general atmospheric circulation models and scenarios applied by the IPCC, it is anticipated that the Earth’s temperature in 2100 would be 1.1–4.6 °C warmer than 1900, accompanied by a change in precipitation [39]. Several studies have shown that extreme temperature changes during the year leads to various illnesses and even deaths. For example, Nastos and Matzarakis [40] investigated the role of the climate in the mortality rate in Athens in a 10-year period, using heat-stress indices, including the Universal Thermal Climate Index (UTCI) and the Physiologically Equivalent Temperature (PET). They showed that there was a statistically significant relationship between air temperature, UTCI, PET and mortality rate [40]. Certainly, extreme heat conditions will lead to physiological limitations. On the other hand, parts of the world that are excessively populated will be uninhabitable because of climate change, if global temperature increases at least up to 7 °C [2].

In the present study, the WBGT index was applied to evaluate the heat stress among outdoor workers. Despite its simplicity, WBGT is the most widely used index for estimating the heat stress all over the world. WBGT is well associated with physiological parameters at high temperatures [3]. Our findings showed that the WBGT index has an increasing trend up to 2099; this increase is much more significant in Pakdasht.

Mohraz et al. [17] investigated the past and future trends of heat stress in outdoor workplaces of Tehran, based on time series modeling. They showed a significant increasing trend in average annual WBGT (1961–2009). In addition, the model estimated an increase of 1.55 °C in the WBGT by 2050. The results also indicated that the monthly average WBGTs in March and October are in the hot region (caution, 18–24 °C). Using WBGT, the increase in the heat stress threshold for 15 regions of the world has been studied during the period from 1973 to 2003, given the climate change. The results suggest an increased trend in all regions except the northeastern United States and northeastern Australia [41]. In this study, the Hadcam3 model was used to determine regional variations of average summer temperatures, relative humidity and WBGT under the A1B scenario for 2020 and 2050. The results showed that the WBGT for all regions will rise, but the variation in temperature and relative humidity will depend on the climatic conditions prevalent in the studied regions [41]. The results of these studies are in good agreement with those of the present study.

Our results indicates that for the summer, especially July and August, the WBGT index will increase by 3.2 °C. In this regard, Maeda et al. [42] pointed out that heat-related mortality rates were higher in July and August than other months of the year.

Climate change and the gradual increase in global temperature are certainly some of the biggest environmental and health challenges. The richest energy consumers, who have the greatest responsibility for greenhouse-gas emissions, play an important role in global warming and climate change. In contrast, poorer countries are more vulnerable to damages caused by climate change. Studies indicate a gradual increase in global temperature because of human activity. Temperature is estimated to increase up to 2.5-7 °C in this century. This is a serious warning effect that needs to be taken into account [43]. Outdoor workers can be the most vulnerable people because of climate change. Excessive heat along with different job requirements such as clothing, workload, metabolism, long exposure hours and personal protective equipment may lead to high occurrence of heat strain and other complications associated with skin rashes, muscle cramps, heat exhaustion and death in warm seasons [44]. Obviously, neglecting the heat stress in workplaces can have a negative effect on the ability to work and can reduce the physical and mental performance, beside a wide range of heat -related complications and illnesses [9].

Conclusions

According to the results, it can be concluded that exposure to heat stress will be increasing in Tehran within specific periods especially in warm months due global warming. These changes can have adverse effects on the health of individuals, especially vulnerable people such as the elderly, children and patients. Heat stress can also affect the level of health, safety and productivity of outdoor workers with high physical load. In the present study, the changes in the Wet-bulb Globe Temperature over the coming years (up to 2099) were measured and modelled, taking into account global warming and using a general circulation models, the LARS model and A1B. For future studies, it is recommended to use several climate models and to take into account several scenarios in order to compare the results of the models and accurately assess the changes for future plans.

Funding information

This study has been financially supported by the Institute for Environmental Research, Tehran University Medical Sciences (Grant No. 94-01-46-28540).

Compliance with ethical standards

Conflict of interest

The authors confirm that there is no conflict of interests.