Temporal fluctuations of PM_2.5 and PM₁₀, population exposure, and their health impacts in Dezful city, Iran

Abstract

Morbidity and mortality impacts of particulate matter (PM) are globally important health critical parameters. In this ecological-descriptive study, the health impact of PM₁₀ and PM_2.5 associated with there temporal variations in Dezful city were assessed from 2013 to 2015. AirQ+ software handles the PM air pollutants by addressing impact evaluation and life table evaluation. We used a new method to analysis fine particles feature by using regular daily observations of PM₁₀. In this method, relationship between PM_2.5 and PM₁₀ mass concentrations were analyzed and calculated. The annual average concentrations of PM₁₀ were 147.1, 114.3 and 158.8 μg/m³, and the annual average concentration of PM_2.5 were 57.8, 50.7 and 58.2 μg/m³ in 2013, 2014 and 2015, respectively. PM₁₀ also had obvious diurnal variations with highest hourly concentrations in 13:00 and 22:00 but the lowest concentrations often occurred in 05:00 and 16:00. Unexpectedly, in weekends the concentration of PM pollutants appeared to have increased from 18:00 to midnight. The daily based analysis showed that there are 147 dusty days in the study period during which the most severe dusty day occurred in 2014. Over the study period, mean levels of PM₁₀ and PM_2.5 in both conditions were higher in 2015 compare to 2013 and 2014, which probably is due to higher frequency of dust storms in 2015. Hence, during 2015 and 2013 they're were higher morbidity and mortality compare to 2014 due to exposure to higher polluted air with PMs in all cases except lung cancer (LC).

Introduction

According to many epidemiologic studies conducted in the field of air pollution and community health, it is well accepted that exposure to air pollution is associated with a wide range of acute and chronic health effects from minor health problems to death [8]. Unfortunately, due to the expansion of urbanization and industrialization, the decline in air quality is one of the major and growing concerns of developed and developing countries [10, 30]. Air pollution is mainly caused by both human activities and natural events. Human is the major contributing to air pollution. They negatively impact the environment by burning fuels, road traffic, industrial processes, etc. [38]. Air pollutants are series of pollutants that directly and indirectly affect human health, animal and the environment [5, 6, 9, 36]. The US Environmental Protection Agency (US EPA) has selected six major pollutants as “criteria air pollutants”, which include carbon monoxide, nitrogen dioxide, particulate matter (PM), sulfur dioxide, lead and ozone [14, 28]. Today, many Iranian cities face the problem of poor air quality and dust phenomenon [12, 19, 20, 32, 39]). Many of the health effects associated with PM have been reported [3, 4], including deaths, lung cancer, hospitalization due to cardiovascular and respiratory diseases, physician office visit, deterioration of respiratory symptoms, school and workplace absenteeism, activity limitation, acute and chronic bronchitis, and so on [33, 34]. PM is a mixture of liquid and solid particles with different sizes and chemical compositions. There are two types of PM particles. PM₁₀ refers to particles equal or smaller than 10 μm and PM_2.5 are particles smaller than 2.5 μm. PM_2.5 threatens health more than PM₁₀, because the possibility of deposition of smaller particles in the lower abdomen of the lung is greater. Studies have shown that long-term contact with PM leads to a decrease in life expectancy. This decline in life expectancy is mainly due to death from respiratory diseases and lung cancer [2].

The AirQ+ model provided by the World Health Organization (WHO) is the most reliable method for assessing the adverse effects of exposure to airborne contaminants on human health. This software uses the data processed by Excel to estimate the relative risk of occurrence and attributable components, and also displays the result as morbidity and mortality [13]. It is a special software which enables users to estimate the potential effects of exposure to air pollutants on human health in a particular area at a specific time [30]. Recently, AirQ models have been applied for analyzing the associated health impacts of PMs in various cities of Iran. However such assessment did not performed for Dezful city, which has a frequent exposure to dust storms. We aimed to investigate temporal profile of particles and their associated health impacts in Dezful city using AirQ+ software from 2013 to 2015.

Material and methods

Study area

Dezful city in accordance with Iran 2016 census (www.amar.org.ir), has a population of 443,971 and has an area of 4762 km², is located at the southwest of Iran. It is located at geographical coordination of 32° 22´ N latitude and 48° 32′ E longitude. Figure 1 presents the monitoring sites, geographical location of Air Quality Monitoring Station (AQMS), and Meteorological station (MS) in Dezful city area. The residence age data profile is shown in Table 1 (Statistical Center of Iran).

Fig. 1.

Urban distribution and the location of AQMS and MS in Dezful city

Table 1.

Baseline incidence of different categories of morbidity and mortality for babies, adults and all ages (All NC: all natural cases; COPD: chronic obstructive pulmonary diseases; LC: lung cancer; IHD: ischemic heart diseases; ALRI: acute lower respiratory infections; HARD: hospital admission respiratory diseases; HACVD: hospital admission cardiovascular diseases) [15]

BI*	All Ages	0–5	>25	>30
Mortality
All NC				113.8
COPD				11.6
LC				0.7
IHD			5.6
Strokes			16.7
ALRI		3.0
Morbidity
HARD	1260
HACVD	436

^*BI per 10⁵ persons

Methods of the study

The input atmospheric pollutants data required for models simulation were collected from the Khuzestan Department of Environment, the concentrations of PMs, dusty and non-dusty days data were obtained from Khuzestan Meteorological Organization, morbidity and mortality data were obtained from Statistical Center of Iran and health information data were obtained from the Ministry of Health. These data were obtained in volumetric base from 2013 to 2015.

The adverse health effects associated with PMs inhaled was analyzed using the AirQ+ models. The PMs were selected as pollutant in AirQ+ analysis. As the collected data was not in same model units, there was a conflict between the model and the collected data. In order to solve this problem, the data was processed using Microsoft Excel software and the conversion between volumetric and weight units (temperature and pressure correction), processing (average) and filtering were performed. In order to use PM₁₀ concentration data in the standard conditions (STP), for non-standard temperatures and pressures, Eq. (1) can be used.

$$\frac{P_1{V}_1}{T_1}=\frac{P_2{V}_2}{T_2}$$ 1

where; P₁, V₁, and T₁ are the initial pressure, volume and absolute temperature in the non-standard conditions, respectively. P₂, V₂, and T₂ are the pressure, volume and absolute temperature in the standard conditions, respectively. The mean daily concentrations of PM₁₀ were applied in the study.

Data analysis

The AirQ+ software is a valid and reliable tool for estimating the long and short-term effects of air pollutants including PMs, which has been introduced by WHO. In this research, the AirQ+ models were used to evaluate the effects of fine particulate matters on the health of Dezful citizens from 2013 to 2015. The models for determining the health effects are mostly statistical-epidemiological type, which integrate air quality data at concentration intervals with epidemiological parameters such as relative risk, baseline incidence (BI), etc. and displays the outcome in terms of morbidity and mortality (Table 1). In running the software, linear-log calculation method was used to assess respiratory diseases mortality and all-cause mortality attributable to long-term exposure to PM_2.5. The integrated exposure-response function (IER) was applied to estimate cause-specific mortality in adults (IHD, stroke, COPD, LC) and ALRI for infants [11, 16, 24]. The relative risk (RR) values were calculated by the following formula:

$${\mathrm{RR}}_{\text{linear}log}=\mathrm{e}{{{}^{\mathrm{\beta }\left\{\right[ln\left(\mathrm{X}+1\right)ln(\mathrm{X}}}_0}^{+1\left)\right]\}}$$ 2

In IER function:

$${\mathrm{RR}}_{\mathrm{IER}}=1+\mathrm{\alpha }\left\{1–exp.\left[-\mathrm{\gamma }{\left(\mathrm{X}–{\mathrm{X}}_0\right)}^{\mathrm{\gamma }}\right]\right\}$$ 3

where X (μg m⁻³) is the measured annual PM_2.5 concentration and X₀ is the reference level (10 μg m⁻³ and 0 μg m⁻³) concentration indicate the change in RR for X concentration. α, γ, β show the overall shape of the nonlinear concentration-response relationship [11, 24]. The number of deaths attributable to annual levels of PM_2.5 was calculated by the following equation:

$$\mathrm{E}=\mathrm{BI}\times \mathrm{PAF}$$ 4

where E is the rate of the health impact attributable to PM_2.5 exposure and BI is the baseline incidence of the considered health endpoints in an exposed group and PAF is the population attributable fraction [13, 21, 25]. PAF was computed with the following equation:

$$\mathrm{PAF}=\mathrm{RR}–1/\mathrm{RR}$$ 5

Result and discussion

Relationship between PM₁₀ and PM_2.5 concentrations

In order to apply the data obtained in this study to the daily PM₁₀ measurement results, PM₁₀ concentration ranges and the ratio of PM_2.5/PM₁₀ were used. The data was assess from 2013 to 2015 with similar geographical (Ahvaz, Iran) and climate conditions with good agreement of R² = 0.89. Finally, y = 4.5442x^-0.473 equation was applied on Dezful PM₁₀ concentration for calculating hourly PM_2.5 values. The obvious relationship between PM₁₀ and PM_2.5 concentrations is shown in Fig. 2. These results allow for the effective use of the daily PM₁₀ data to calculate PM_2.5.

Fig. 2.

Fittest function between different ratios of PM_2.5/PM₁₀ according to available PM data from the nearest city (Ahvaz) (a) and hourly frequency of PM₁₀ in Ahvaz through 36 classes (b)

Temporal variations of PM₁₀

To delineate the characteristics of the day-to-day changes in PM₁₀ concentration during a week, the average values of the observed PM₁₀ concentrations were calculated for each day of the week over all the days during the period 2013–2015. Figure 3 shows diurnal and weekly variations in PM₁₀ mass concentration in Dezful during study. As it can be seen in Fig. 3, variation of PM₁₀ concentration in the diurnal period observed between the minimum and maximum values at 6:00 local time (LT) by an average amount of 110 μg/m³ and at 20:00 LT by an average amount of 192 μg/m³, respectively. Thus, air pollution level increases in working hours of factories, offices The data also showed that particle concentration has increased on weekends which is closely intertwined with road traffic, increased leisure-time travel, lack of monitoring of polluted industries. Weekend trend is lower than weekday trend between 02:00–16:00. We assumed that the phase of the weekly PM₁₀ cycle is associated with changes in the atmospheric circulation that might be triggered by the accumulation of PM₁₀ through diabetic heating of lower troposphere. We observed similar results to those from previous. For instance, a study in Ahvaz city showed that maximum and minimum values of PM₁₀ occurred in July (420.5 μg/m³) and January (154.6 μg/m³), respectively. Diurnal profile of PM₁₀ in Ahvaz exhibited a peak between 8:00–11:00 [27] which had higher concentration than Dezful. From March 2013 to February 2014 in Southwest China, the daily average concentrations of PM₁₀ and PM_2.5 were 156.6 and 99.5 μg m⁻³, respectively, which exceeded both the Chinese ambient air-quality standards for PM and the guidelines of WHO [23].

Fig. 3.

Diurnal variation of PM₁₀ based on weekend, weekday and cumulative mass concentrations in Dezful

We exploit week-to-week variation to understand temporal patterns in PM₁₀ data. As shown in Fig. 4, the minimum daily value of PM₁₀ was recorded on Saturday and Thursday by 135 and 136 μg/m³, respectively, while the highest daily mean of PM₁₀ were observed on midweek or Monday (157 μg/m³). Average of maximum hours of PM₁₀ observed as the highest value on Wednesday (433 μg/m³) whereas mean of minimum hours of PM₁₀ occurred on Thursday (the beginning day of weekend for getting enough rest of it) by 63 μg/m³. It seems that during the early part of a week the anthropogenic aerosols are gradually accumulated in the lower troposphere. Due to promoted ventilation during the latter part of the week, aerosol concentrations reduce in the boundary layer. A weekly average of PM₁₀ level in Dezful is 143 μg/m³ which is lower than daily average guideline values of the National Ambient Air Quality Standards (150 μg/m³) and 2.88 times higher than WHO air quality guidelines (50 μg/m³). In china during 2001–2005, the PM₁₀ concentrations at 29 sampling stations show significant weekly cycles with the largest values around midweek and smallest values in weekend [18]. During 2013 to 2014 in Southwest China, the highest concentrations of particles were observed on Mondays and the lowest on Thursdays. Weekend effects were also obvious, which were mainly attributed to human activities [23]. Other studies confirmed our findings, in Northwest Germany On weekday, by decreasing the traffic by approximately 99% during late-night hours, the PM₁₀ concentration was reduced by 12% of the daily mean value. A correlation between PM₁₀ and the particle number concentration was found for each weekday and traffic contributes a constant amount of particles in a daily and weekly cycle [17, 26].

Fig. 4.

Weekly variations of PM₁₀ pollutant during 2013 and 2015 in Dezful with WHO and NAAQS guideline and standard

According to Fig. 5 the highest concentration of PM₁₀ (247 μg/m³) was observed in July and the lowest (71 μg/m³) was observed in November. The coarse particle concentrations increased after school beginning and decreased after New Year (Nowruz). Lower values during the wet season are understood in terms of convective ventilation and wet removal. February and summer are two periods of time which dust storms happened frequently. This may be because of the trapping of PM emissions due to poor dispersion conditions. In Northwest China for the period of 2001–2007, higher values were shown in November, December and January to March. The maximum monthly average PM₁₀ concentrations appear in December (271 μg/m³) followed by March (245 μg/m³) while it is low during summer months (May to October) with monthly average PM₁₀ concentrations below 150 μg/m³ [40]. The highest and the lowest monthly average PM₁₀ concentrations at the urban stations in Athens from 2001 to 2010 are observed during the autumn/winter and the summer months, respectively [37].

Fig. 5.

Monthly and occasional average variations of PM₁₀ through the whole study period in Dezful (BSB/ASB: Before and After School Beginnings and BNY/ANY: Before and After Persian New Year)

Distribution of particulate matters in dusty and non-dusty days

Farthermore, dust storms, which produce anomalously high PM₁₀ concentrations of natural origin, are more frequent in spring. We defined criteria to distinct between normal and dusty days based on hourly average concentration of PM₁₀. Exceeding from 200 μg/m³ was considered as dust storm days [22, 27] and lower than this value could be supposed as normal days. Table 2 represents distribution of dusty days over the period of the study. Based on this table, there was 147 dusty days out of three years of the study. Transporting large amounts of particulates, pollutants, and biological materials for long distances. During transport, some of the atmospheric aerosols undergo chemical modifications. It should be noted that heavy metals as well as cations and anions are transported by particles can cause significant cardiovascular effects and oxidative stress. PM₁₀ and PM_2.5 concentration ranged from 119 μg/m³ (in June 2015) to 2077 μg/m³ (in July 2014) and 52 (in June 2015) to 240 μg/m³ (in July 2014). More than half of the July’s days in the study period was dusty, that’s why PM₁₀ level rose to peak in July (Fig. 5). The air stability conditions and regular temperature inversions provided air accumulation over the Dezful city by limiting the dilutions and dispersions. According to previous studies, Saudi Arabia, Iraq and Kuwait, are the main sources of dust storms in the Middle East [19]. In Saudi Arabia a consistent and systematic pattern decreasing in the order: spring (February–April, 53%) followed by summer (May–July, 30%) and fall and winter (September–January, 18%). The spring Peak of PM₁₀ is coinciding with dust events that commonly occur during spring in Riyadh. In contrast, the lowest PM concentration in winter can be attributed to the absence of dust events during winter [31]. The results of a study in Sistan and Baluchestan Province (southeast Iran) showed that large quantities of transported dust that strongly dependent on the duration of the dust events, and secondarily, on the wind speed and distance from the source region. Daily PM₁₀ levels during intense dust storms rise up to 2000 μg m⁻³, even reaching to 3094 μg m⁻³, while the monthly mean PM₁₀ variation shows extreme values (>500 μg m⁻³) for the period June to October [35]. From March 2008 to February 2009, The majority of PM₁₀ episodes were attributed to the intrusion of dust to Jeddah urban air: The USEPA 24-h average concentration of PM₁₀ (150 μg/m³) was exceeded in 38 days at the two selected air quality monitoring sites in Jeddah [1].

Table 2.

Annual and monthly distribution of dusty days regarding with statistical PM₁₀ and PM_2.5 analyses

Time	PM10				PM2.5				No
	Min	Ave	Max	SD	Min	Ave	Max	SD
2013	137	312	645	136	58	90	137	20	51
2014	181	413	2077	505	67	94	240	45	13
2015	119	344	1842	281	52	87	206	23	83
Jan	910	1131	1353	6	131	135	139	6	2
Feb	154	597	1842	499	62	107	206	39	14
Mar	–	–	–	–	–	–	–	–	–
Apr	169	236	349	56	64	78	95	9	8
May	186	223	269	31	68	75	80	4	9
Jun	119	312	645	144	52	88	137	22	31
Jul	151	317	2077	276	61	88	240	27	48
Aug	167	286	469	106	62	85	115	18	11
Sep	137	366	584	134	58	93	115	16	11
Oct	152	225	453	86	62	73	96	10	10
Nov	–	–	–	–	–	–	–	–	–
Dec	241	343	479	122	81	94	110	15	3

The other view of PM₁₀ and PM_2.5 mass concentrations was shown in Table 3 during non-dusty and all days. In general, mean levels of PM₁₀ and PM_2.5 in both conditions were higher in 2015 than 2013 or 2014, this probably due to higher frequency of dust in this year (Table 2). However minimum and average concentrations of coarse and fine particles in 2014 were lower than 2013 and 2015, while the most severe dusty day occurred in 2014.

Table 3.

Statistical analysis of PM₁₀ and PM_2.5 during non-dusty (NDD) and cumulative (CD) days for each year

Year		PM10				PM2.5
		Min	Ave	Max	SD	Min	Ave	Max	SD
Non-dusty days	2013	17.8	101.7	215.0	53.6	20.4	49.0	76.8	15.4
	2014	11.2	95.5	231.8	42.4	15.1	47.9	76.4	12.0
	2015	18.2	102.7	305.2	51.0	20.5	49.5	82.2	12.7
Cumulative days	2013	17.8	147.1	645.4	116.9	20.4	57.8	137.1	23.6
	2014	11.2	114.3	2077.3	146.0	15.1	50.7	239.6	19.1
	2015	18.2	158.8	1841.8	175.0	20.5	58.2	206.0	22.4

Health enhancing by dust removing

The deterioration of ambient air quality due to anthropogenic activities has significant impacts on human health. Long and short-term health effects of mean PM_2.5, given in Table 3, are shown in Table 4 in both non-dusty and cumulative days. Higher concentration of PM_2.5 caused worse health condition. Therefore, 2015 and 2013 repotted higher morbidity and mortality than 2014 during both situations of exposure to polluted air in all cases except lung cancer. This is probably because of low baseline incidence of lung cancer mortality (0.7). In general, eliminating dust over Dezful caused 81, 11, 4, 19 and 4 lower deaths for all NC, COPD, IHD, strokes and ALRI (Table 4). Similarly, it decreased number of HARD and HACVD from 970 to 832 and 165 to 141.

Table 4.

Long term and short term health endpoints attributed to PM_2.5 during non-dusty (NDD) and cumulative (CD) days through low, mean and high relative risk

Year	Levels	Long term (Mortality)												Short term (Morbidity)
		All NC	RR	COPD	RR	LC	RR	IHD	RR	Stroke	RR	ALRI	RR	HARD	RR	HACVD	RR
2013 NDD	Lower	122	1.165	17	1.090	1	1.070	46	1.870	97	1.490	7	1.280	73	1.019	9	1.007
	Mean	179	1.264	37	1.220	3	1.290	60	2.560	192	2.860	10	1.460	279	1.076	47	1.036
	High	229	1.365	56	1.370	4	1.480	74	4.030	245	5.860	13	1.660	561	1.166	85	1.066
2014 NDD	Lower	118	1.160	17	1.090	1	1.070	46	1.859	96	1.480	7	1.270	71	1.018	9	1.007
	Mean	175	1.256	37	1.219	3	1.289	60	2.539	191	2.827	10	1.439	271	1.074	46	1.035
	High	224	1.353	55	1.360	4	1.479	74	3.985	245	5.750	13	1.638	546	1.161	82	1.064
2015 NDD	Lower	123	1.168	17	1.090	1	1.075	46	1.875	98	1.495	7	1.285	74	1.019	9	1.007
	Mean	181	1.268	37	1.220	3	1.295	60	2.565	193	2.875	10	1.465	282	1.077	48	1.036
	High	232	1.370	56	1.370	4	1.485	74	4.050	246	5.895	13	1.670	567	1.168	86	1.067
2013 CD	Lower	147	1.206	21	1.110	1	1.090	48	1.938	106	1.558	8	1.330	89	1.023	11	1.008
	Mean	214	1.333	42	1.250	3	1.338	62	2.668	200	3.096	12	1.558	339	1.094	58	1.044
	High	272	1.464	60	1.408	4	1.548	76	4.364	249	6.332	15	1.826	676	1.207	103	1.082
2014 CD	Lower	126	1.173	18	1.097	1	1.080	46	1.880	100	1.507	7	1.290	76	1.020	9	1.007
	Mean	186	1.277	38	1.227	3	1.300	60	2.584	194	2.911	10	1.477	290	1.080	49	1.038
	High	238	1.383	56	1.370	4	1.497	75	4.098	246	5.958	13	1.687	583	1.174	88	1.069
2015 CD	Lower	148	1.208	21	1.110	1	1.090	48	1.940	106	1.560	8	1.332	90	1.023	11	1.008
	Mean	216	1.336	42	1.250	3	1.340	62	2.692	201	3.106	12	1.562	341	1.095	58	1.045
	High	273	1.469	60	1.410	4	1.552	76	4.374	249	6.348	15	1.834	681	1.209	104	1.083

Additionally, results of a study in Tehran showed that long-term exposure to ambient PM_2.5 contributed to between 24.5% and 36.2% of mortality from cerebrovascular disease (stroke), 19.8% and 24.1% from ischemic heart disease (IHD), 13.6% and 19.2% from lung cancer (LC), 10.7% and 15.3% from chronic obstructive pulmonary disease (COPD), 15.0% and 25.2% from acute lower respiratory infection (ALRI), and 7.6% and 11.3% from all-cause annual mortality during 2006 to 2015 [11, 36]. According to a similar study results in Iran, the numbers of 145, 98, 76, and 73 of HARD cases were estimated for Ahvaz, Isfahan, Shiraz, and Tehran which were responsible for 12, 8, 6, and 6% of the total HARD, respectively [29]. Also, about 4.9% of hospital visits in Ilam for COPD in can be attributed to PM₁₀ concentrations over 10 μg/m³ [7].

Conclusion

In this study impacts of airborne particles concentration on public health was assessed in Dezful city from 2013 to 2015. To calculate health impacts AirQ+ model was used. In this method, the relationship between concentration of PM₁₀ and PM_2.5 was established and applied to daily PM₁₀ observations. Besides, the highest annual average concentrations of PM₁₀ and PM_2.5 were recorded in 2015. Due to diurnal variation of PM₁₀ concentration, there was also an increase in PM₁₀ concentration on weekends. Regarding to dusty days, the highest monthly PM level was found in July. Long and short-term health effects of mean PM_2.5 were also estimated. Their were higher morbidity and mortality in 2015 and 2013 compare to 2014. We assumed that pollution materials accumulated before dust outbreaks mix with dust particles, cause high concentration of pollution species exacerbating health problems. Our results were compatible with the other studies and showed that PM₁₀ concentration exceeded WHO guidelines which affected the health of Dezful citizens. Thus, control procedures and consistent assessments are required to reduce atmospheric PM pollutions significantly. Additionally, health promotion programs motivate decision-makers, political parties and institutions of government to mitigate air pollution.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.