Effects of a Saharan Dust Episode on Emergency Attendances for Respiratory Diseases in İstanbul

Abstract

Saharan dust events have an important effect on the air quality of Turkey due to their significant contribution to particulate matter concentrations. These events likely impact public health in urban areas. However, there is no available information on the health effects of Saharan dust in Turkey. On Sunday 1 February 2015, İstanbul experienced an episode of extreme Sahara dust event, which increased the particulate matter concentrations greatly compared to the average values of the city. In this study, we examined the relationship between particulate matter (PM10, PM2.5) concentrations and emergency attendances for asthma, chronic obstructive pulmonary disease (COPD), and acute bronchitis on the episode day to better understand the association between an extreme dust event and emergency attendances for respiratory health in the city. Analyses showed that there was no significant effect of the Saharan dust event on emergency attendances for asthma and COPD in İstanbul compared to average emergency attendance numbers of the city. However, emergency attendances for acute bronchitis significantly increased on the episode day. This study revealed, extreme Saharan dust events can considerably increase the risk of visiting hospital for acute bronchitis in İstanbul during a severe dust episode.

1. Introduction

Large quantities of dust from arid lands and deserts can be transported thousands of kilometers by winds. Therefore, the impact of dust storms can be observed many hundreds of kilometers downwind from the emission source. Ambient concentrations of particulate matter during dust transportation events can be extremely high due to millions of tons of mineral dust and many people are likely to inhale a higher than usual concentration of particulate matter (PM) during these events. Studies suggest that dust storms can increase the risk of respiratory health effects, such as emergency department visits, hospital admissions, and mortality [1, 2, 3, 4, 5].

Sahara dust is the largest, naturally occurring source of particulate matter in The World. Due to the proximity to the Sahara Desert, Sahara dust transportation have an important effect on the air quality of Turkey and can lead to high PM levels that can exceed the limit values [6, 7, 8, 9]. These events likely impact public health in urban areas in Turkey. However, there is no available information on the health effects of Saharan dust in Turkey so far.

On Sunday, 1 February 2015, İstanbul experienced a serious Sahara dust transport event caused by strong southwesterly winds. During this episode, the sky turned dusty orange, ferry traffic was interrupted, and over 100 flights were canceled by Turkish Airlines. Daily average PM₁₀ concentration was recorded at 325,1 μg/m³, which is an extremely high level of ambient PM₁₀ not normally encountered in the city. This episode provided an opportunity to assess the health effects during an extreme dust event that significantly affected daily city life. We conducted an analysis to determine the effects of this episode on the emergency department visits for asthma, COPD, and acute bronchitis in İstanbul by using air quality and health data. Synoptic analysis and NASA satellite maps are also used to show the dust transportation from the Sahara Desert as a dust source region.

2. Methodology

2.1 Study area

Located between Black Sea and Sea of Marmara, İstanbul is the largest urban area of Turkey with a population of more than 16 million. The city forms one of the largest urban agglomerations in Europe. İstanbul is separated into Asian and European parts by Bosphorus strait which is approximately 30 km in length.

2.2 Air quality data

Hourly air pollution data (PM₁₀ and PM_2.5) were used from the database of Republic of Turkey Ministry of Environment and Urbanization, the government agency in charge of collection of air pollution data in Turkey. The hourly concentrations for each pollutant were obtained from the 9 fixed-site air quality monitoring stations of Marmara Clean Air Center Monitoring Network (Kağıthane, Esenyurt, Silivri, Ümraniye, Mecidiyeköy, Şirinevler, Üsküdar and Kandilli). To evaluate the weather conditions on the episode days, hourly meteorological data (temperature and relative humidity) were obtained from the Air Quality Monitoring Stations where meteorological measurements are also made (Table 1). Daily means of the concentrations calculated from the hourly data of the pollutants and weather variables were used to represent the daily reading for İstanbul.

	Mean + SD	Min	P(25)	P(50)	P(75)	Max	Episode; Day
Number of emergency attendances
Acute bronchitis (n = 154,182)	369 ± 200	51	228	337	471	1259	829
Asthma (n = 65,592)	157 ± 59	52	122	155	186	779	153
COPD (n = 25,240)	60 ± 24	15	42	59	74	217	33
Air pollution concentrations (μg/m3)
PM10	52.7 ± 24.7	17.0	35.3	45.6	60.5	325.1	325.1
PM2.5	25.5 ± 14.6	5.6	15.4	20.9	31.4	105.0	70.6
Weather
Temperature (°C)	15.2 ± 7.1	−3.8	9.5	15.1	22.0	30.0	17,4
Humidity (%)	79.5 ± 10.7	45.9	72.7	80.2	86.9	100	50,6

Table 1.

Summary statistics of number of emergency attendances, air pollutant concentrations, and weather conditions in İstanbul, Turkey (2014–2017).

2.3 Satellite data

In this study, Hybrid Single-Particle Lagrange Integrated Trajectory Model developed by Air Resources Laboratory of National Oceanic and Atmospheric Administration (NOAA) was used to detect the trajectories showing dust transportation during the episode. Region coordinates are marked on the map in HYSPLIT model and past trajectories showing dust transportation on the maps are obtained. In order to visually examine the high particulate matter concentrations during the episode day, Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Aqua satellite images were also used in the study.

2.4 Health data

Data on the number of daily respiratory hospital admissions of the 21 public hospitals in İstanbul from January 1, 2014, to December 31, 2017 (1445 days) were obtained from the database of Republic of Turkey Ministry of Health. The daily respiratory hospital admission counts in our study were selected for seven different respiratory outcomes according to the International Classification of Disease, Tenth Revision (ICD-10) by the World Health Organization: Acute bronchitis (J20), COPD (J44), and asthma (J45-J46).

In Turkey, admissions for state hospitals (except emergency services) on weekdays are made by appointment taken via the internet. Weekend services of state hospitals are limited to emergencies services only. In this study, due to the Saharan dust episode that occurred on Sunday, all the hospital admissions made on the day of the episode are attendances to emergency services.

3. Results

On 01 February 2015, a cyclone centered on Italy caused a southerly wind event on the eastern Mediterranean (Figure 1). Sahara Desert sands were lifted by strong winds, passed over Aegean Sea, and reached İstanbul. Dust load transport and its path can be seen well on NASA Earth Data imaginary as a continuous dust flow from North Africa to the Black Sea (Figure 2). HYSPLIT 48 hr. air-mass back trajectories at different elevations (Figure 3) indicate the movement of air masses from Saharan Desert in North Africa passing through Mediterranean to İstanbul.

Figure 1.

Mean sea-level pressure (hPa) map of Europe showing weather fronts on 01 February 2015 [10]. İstanbul is marked with a red dot.

Figure 2.

Saharan dust transport on 01 February 2015 [11].

Figure 3.

HYSPLIT backward air mass trajectories originating from North Africa and arriving in Istanbul on 01 February 2015 [12].

The stations of air quality monitoring network in İstanbul measured high concentrations of PM₁₀ (325.1 μg/m³) and PM_2.5 (70.6 μg/m³) on February 1, 2015, which were good matched with the NASA satellite image and HYSPLIT model trajectories. Particulate matter size distribution was dominant at a 3–7.2 μm size fraction on the episode day [13]. On February 2, dust transportation was ceased, and precipitation started. Consequently, particulate matter concentrations decreased with wet and dry deposition. Figure 4 shows the significant contribution of Saharan dust to ambient concentrations of particulate matter during the episode. Descriptive statistics for daily mean pollutant levels, temperature, and humidity over a 4-year period between 2014 and 2017 are presented in Table 1, showing comparisons of concentrations on the dust storm day (01/02/15) with the average numbers between 2014 and 2017. The PM₁₀ concentration recorded on the episode day (325.1 μg/m³), which greatly exceeded the European union daily standard of 50 μg/m³, is the highest concentration of daily average PM₁₀ observed between 2014 and 2017 in İstanbul. Yearly the average PM₁₀ concentration was 52.6 μg/m³ between 2014 and 2017 in the city, respectively. Considering the yearly average, it could be inferred that ambient PM₁₀ concentration increased more than 6 times during the dust episode.

Figure 4.

Daily mean values of PM₁₀ and PM_2.5 measured in İstanbul between 20/01/2015 and 15/02/2015 [9].

The PM_2.5 concentration recorded on the episode day was 70.6 μg/m³. Yearly average PM_2.5 concentrations between 2014 and 2017 were 25.5 μg/m³ (Table 1). Although a high PM_2.5 value is measured on the episode day, PM_2.5 concentration did not rise considerably compared to the very high PM₁₀ value (Figure 4). Low atmospheric pressure, warm air, and strong winds were observed during the episode. The daily average air temperature on the day of the event was 17.4°C and the daily average wind speed (6.5 m/s) was also very high compared to the average temperature (6.1°C) and wind speed values (2.6 m/s) of the city (Table 1).

Descriptive statistics for daily mean emergency department visits for three different respiratory diseases are presented in Table 1, showing comparisons of the ED visits on the dust episode day (01/02/15) with the average ED admissions between 2014 and 2017. Because the Saharan dust episode happened on Sunday, all hospital applications on the episode day are made for emergency departments. According to our results, the dust storm episode had no significant effect on the emergency department visits for asthma and COPD; ED visits for asthma (n = 153) and COPD (n = 33) on the episode day were below the average values of ED visits (avg. = 157 and avg. = 60) made between 2014 and 2017 (Table 1). However, the dust storm episode had a significant effect on the emergency department visits for acute bronchitis (Figure 5). Elevated Saharan dust levels exacerbated acute bronchitis in patients, as assessed by the increasing number of ED admissions (n = 829) on 01/02/15. According to results, more than twice of the average ED visits (avg. = 369) have been made on the episode day (Table 1).

Figure 5.

Hospital applications for asthma, COPD, and acute bronchitis in İstanbul between 20/01/2015 and 15/02/2015.

4. Discussion and conclusion

In this study, the relationship of a severe Saharan dust episode with the emergency department visits for asthma, chronic obstructive pulmonary disease (COPD), and acute bronchitis in İstanbul was examined. Investigation of the impact of a Saharan dust episode on respiratory diseases has not been done previously in Turkey, yet. Our results showed that the dust episode did not significantly affect ED visits for asthma and COPD as there were no substantial changes on the emergency visits that could be linked to an extreme increase in dust concentration. However, the dust storm was significantly associated with increased ED visits for acute bronchitis as there was a large increase in ED visits on the episode day compared to average levels of the city.

Many studies stated that dust storms can increase the risk of emergency department visits, respiratory hospitalizations, or admissions for respiratory diseases. A study conducted in Guadeloupe Island over 1 year found that PM₁₀ and PM_2.5–10 pollutants contained in the Saharan dust increased the risk of visiting the health emergency department for children with asthma [14]. In Rome, Alessandrini et al. [1] investigated the residents hospitalized during Saharan dust days between 2001 and 2004. They found that the effect of PM_2.5–10 on respiratory hospitalizations was higher during dust days compared with dust-free days. No effect of PM_2.5 was detected. Saharan dust also increased the effect of PM₁₀ on cerebrovascular diseases. Extremely high levels of particulate matter were recorded, with daily average levels of coarse matter (<10 μm) peaking over 11,000 μg/m³ and fine (<2.5 μm) over 1600 μg/m³ d. in Sydney, Australia. The dust storm period was associated with large increase in asthma emergency department visits, and respiratory emergency department visits. There was no significant increase in cardiovascular emergency department visits or hospital admissions [14]. According to a study in Taiwan, Asian dust storms increase cardiopulmonary emergency visits in Taipei when ambient PM₁₀ concentrations are above 90 μg/m³. Compared to their pre-dust periods, emergency visits for ischemic heart diseases, cerebrovascular diseases, and COPD during high dust events are increased by 0.7 cases (35%), 0.7 cases (20%), and 0.9 cases (20%) per event, respectively [15].

However, several studies did not find any link between dust episodes and negative health outcomes. Prospero et al., found no substantial changes in pediatric asthma attendances that could be linked to short-term surges in dust concentrations from African dust carried in the Atlantic Trade Winds in Barbados reporting on a 2-year study [16]. In a study undertaken in Saudi Arabia, where major sandstorms are frequent, there was no correlation between the average daily PM₁₀ and PM_2.5 levels and the average number of children presenting with acute asthma per day, their daily asthma score, or admission rate [17]. The study of Menendez et al. [18] conducted in Gran Canaria (Spain) found no statistically significant relations between emergency patients with respiratory diseases and elevated Saharan dust levels.

Considering the health effects of particulate matter, it is suggested to investigate the effects of dust transport episodes on human health in future studies.

Conflict of interest

“The authors declare no conflict of interest.”