An Urban Air Quality Assessment Based on a Meteorological Perspective

doi:10.21203/rs.3.rs-2593289/v1

An integrated approach to understanding all measured pollutants with multi-discipline in different time scales and understanding the mechanisms hidden under low Air Quality conditions are essential for tackling potential air pollution issues. Sivas is a province located in the Eastern region of Türkiye with a developing industry and a growing population, and its climate is known as arid and cold. Since the end of 2016, the province has started to replace the habit of using coal for domestic heating and industry with natural gas. Despite this development, especially in Fine Particulate Matter (PM_2.5), the province reaches very high pollution values every winter, threatening public health. In this study, the air pollution of Sivas province was analysed with meteorological and PM_2.5data over six years. It was found that the winter period (January-February-November-December) of every year except 2019 -which has missing data- is the period with the highest air pollution in the province. In addition, the fact that the days exceeding the daily PM_2.5 limit values in 2016, 2017, 2020 and 2021 were also seen in the spring and summer months caused the study to focus on additional emission sources such as long-range dust transport and road vehicles. The year 2017, which has the highest values, was analysed in detail. Pollution periods with the highest episodes in 2018 were analysed with the HYSPLIT and DREAM models. As a result of the study, the average PM_2.5values in 2017 were found to be 31.66 ± 19.2 µg/m³ and a correlation of -0.49 between temperature and PM_2.5. As a result of model outputs, it was observed that the inversion is intensely observed in the province, which is associated with an increase of PM_2.5during the episodes. Especially on days with daily average PM_2.5values above 100 µg/m³, dust transport from the Arabian Peninsula is observed. Additionally, analysis of Planetary Boundary Layer (PBL) data and PM pollution revealed a significant negative correlation (r=-0.61). It was noted that pollution persisted over the province during low PBL hours, particularly during high PM_2.5 concentration periods.

Urban Air Quality

Meteorological Parameters

Dust Transport

PM2.5

Air pollution is defined as the coexistence of one or more pollutants that change the natural composition of the atmosphere and harm the health of living things in higher-than-normal concentrations (World Health Organization, 2021). Pollutants can be anthropogenic, such as the use of industrial machinery, power plants, and cars, or of various natural origins, such as volcanic eruptions and forest fires (Manisalidis et al., 2020). According to the World Health Organization (2022), the significant air pollutants are; particulate matter, ground-level ozone, carbon monoxide, sulfur oxides and nitrogen oxides. Particulate pollution includes respirable particles called PM₁₀, usually 10 micrometers (µm) and smaller in diameter, and fine respirable particles called PM_2.5, usually 2.5 micrometers (µm) and smaller (Environmental Protection Agency, 2022).

Increasing particulate matter pollution in the rapidly developing and growing world negatively affects human health and climate. Moreover, when the pollution concentration exceeds the critical level, it adversely affects many body systems, especially the respiratory and cardiovascular systems of sensitive people such as children, the elderly and those with chronic diseases (Çapraz et al. 2016, 2017; Kelly and Fussell 2015; Ghorani- Azam et al., 2016).

Many scientific studies have proven that the spatial and temporal changes of pollutants that reduce AQ are significantly affected by meteorological events and the dynamic structure of the atmosphere. (Galindo et al., 2011; Unal et al., 2011; Fortelli et al., 2016; Lana et al., 2016; Özdemir et al., 2020; Yavuz et al., 2022). Zhang (2019), in his article examining the dynamic effects of meteorological conditions on air pollution in Beijing, found that the most important meteorological factors are excessive wind speed (dynamic) and sunshine duration (thermal), respectively. Moreover, according to the study, average wind speed and precipitation capacity do not affect AQ. He et al. (2017) found that more than 70% of the variance of daily pollutant concentrations in China can be explained by meteorological conditions, and fine particle (PM_2.5) is more easily affected by meteorological conditions than coarse particle (PM₁₀). Air pollutants can also be carried in the atmosphere depending on meteorological conditions. (Querol et al., 2004; Salvador et al., 2008; Reizer and Juda-Rezler, 2016; Matthias et al., 2018; Ma et al., 2021). Especially from the desert surface by the wind, large amounts of aerosols are transported by crossing the continents and oceans every year.

Many AQ studies have been conducted in Istanbul, the province with the densest population in Türkiye (Ozdemir, 2014, 2019; Flores et al., 2020; Efe et al., 2022), and relatively fewer studies are available for other cities of Türkiye. Tayanc et al. (2022) statistically analysed the temporal and spatial variability of hourly CO, NO₂, NO_x, PM₁₀, PM_2.5, and SO₂ measured at five different stations in Konya province between 2008 and 2018. They found that the pollutant with the highest value in Konya was particulate matter. On the other hand, Türkiye is under the influence of desert dust originating from Africa and the Middle East, which are the two most important dust sources in the world. Therefore, desert dust transport may increase particulate matter concentrations in Türkiye (Kabatas et al., 2018; Özdemir 2019). Çapraz and Deniz (2020) studied particulate matter concentration during Saharan dust episode. They found that PM₁₀ concentrations increased significantly in Istanbul during the dust event, while PM_2.5 concentrations did not increase significantly.

The literature shows that Türkiye and especially the Central Anatolia region is a developing region that is exposed to high PM pollution. Even if the studies address sources of the PM pollution of the region, they are insufficient to fully define its characteristics. In this study, Sivas was determined as the study area. Sivas is a province that is under the influence of air pollution caused by domestic heating in winter due to the effect of meteorological and topographic conditions. Since Sivas, which has a high pollution potential in this sense, this study examined hourly AQ data which obtained from AQ monitoring stations (AQMS) between the years 2016-2021. Data analyzes were performed after increasing the data quality by concentrating on the raw data. The effect of weather conditions on particulate matter pollution in Sivas was investigated by combining hourly meteorological data from the same stations. Days exceeding the limit values were detected. Besides, a case study analysis with two three-dimensional atmospheric chemistry transport AQ models of a dust transportation that the province was exposed to during a certain period were also run for AQ assessment and to understand the external factors causing pollution with emission sources in the region.

1.1. Study Area

Sivas province is located between 35,83° and 38,23° east longitudes and 38,53° and 40,26° north latitudes. Most of its lands are in the upper Kızılırmak region of the Central Anatolian Region, and the other parts are in the Black Sea Region and Eastern Anatolia region (see Fig. 1.). Sivas is the second largest province in Türkiye after Konya, in terms of area which is 28,567 m². The provincial area starts on the high plateaus of Central Anatolia and rises to the east; ends with mountainous terrain. The average altitude is 1285 meters. 46.2% of the provincial area is covered by plateaus, 47.6% by mountains and 6.2% by plains. (Öztürk et al., 2017). According to Turkish Statistical Institute (TURKSTAT) the population of Sivas Province in 2022 is 636,121. The population living in towns and villages is 153.143, and the population living in provincial and district centers is 482,978. Although there is no significant increase in the total population over the years, the ratio of the population living in the center to the population living in the countryside is in an increasing trend. (TURKSTAT, 2023a).

Increasing urbanization and various local pollution sources negatively affect the AQ in Sivas. In 2021, the number of road motor vehicles Sivas was 170,639, more than half of which was cars. It has been determined that the number of independent houses in Sivas Central District is 151.917, 72% (108,666) of them use natural gas for heating, 27% (41,236) coal, 1% (28,015) fuel system In 2020, 82,472 tons of coal was used for heating purposes in Sivas. (TURKSTAT, 2023b).

1.2. PM_2.5 Monitoring

The Ministry of Environment, Urbanization, and Climate Change developed two different AQ Monitoring Stations (AQMS) in the province center that provide PM_2.5 data. Because the data quality was fairer than other AQMS, only İstasyon Kavşağı AQMS data (39.7472° N, 37.0130° E) were used for obtain hourly PM_2.5concentrations. İstasyon Kavşağı AQMS, which was established to measure traffic-related pollution, carry out PM₁₀, PM_2.5, NOx and CO measurements. AQMS is located at a transition point in the province, where the people travel across the district through this location. Moreover, it is close to business centers, which contribute to the traffic jam in every day of the week. Within the scope of the National AQ Monitoring Network, which was established to improve AQ, hourly data for the years 2016–2021 and daily, monthly, seasonal and annual concentrations were calculated from the stations that started to monitor continuously at the end of 2015. PM_2.5 continuous measurement was made with the MP101-09 model of the Environnement brand. The measurement method is based on radioactive radiation. Fine particles are measured by passing a certain amount of air over a glass fiber filter and holding it. There is a Carbon 14 source with low emission (1.33MBq) inside the device. Filter advance occurs automatically during measurement. The lifetime of the filter varies between 4 months and 1 year (Environment SA, 2023)

For six years the high data quality obtained for İstasyon Kavşağı AQMS and the data count was over 80%. Seasonal data availability for station, were as follows: winter 83% spring 93%, summer 88%, autumn 83%, per year. The limit values determined by the European Union were applied on the data in order to better analyze the AQ.

1.3. Meteorological Station Data

Meteorological factors play a crucial role in AQ by affecting the concentration, dispersion and lifetime of pollutants in the atmosphere. Meteorological data were obtained from the Turkish State Meteorology Service (TSMS) at Sivas Meteoroloji Station (39.744° N, 37.002° E) located approximately 1 km southwest of the İstasyon Kavşağı. Hourly meteorological data includes temperature (°C), pressure (hPa), wind speed (ms^− 1), solar radiation (Wm^− 2), precipitation (mm), relative humidity (%), sunshine hours between 2016 and 2021 on Table 1.

Table 1

Information of the field study and mean meteorological conditions between 2016–2021 (‘‘±’’ represent the standart deviation).
Coordinates	39.557° N 37.305° E
Average temperature (°C)	10.6 ± 9.9
Maximum temperature (°C)	39.2
Minimum temperature (°C)	-22.9
Relative humidity (%)	59.4 ± 21.8
Total precipitation (mm)	394.4 ± 56.8
Precipitation days	103 ± 10.8
Wind speed (m/s)	1 ± 0.9
Daily sunshine hours	7 ± 4.2

Sivas has continental climate characteristics. According to the TSMS, it is in the 'Dsb' climate type in the Köppen climate classification. Summers are dry, cool and short. Winter months are severely cold, long and snowy. The winter months are freezing cold, with average temperatures around 0°C. In this respect, Sivas is the coldest province of the Central Anatolia Region. The average temperature in summer is above 20°C (TSMS, 2023). Hourly meteorological data measured between 2016 and 2021 reflected the climate characteristics of Sivas (Table 1). The annual mean temperature is 10.6°C. In the measuring range, the highest recorded temperature was 39.2°C on 15 August 2019, and the lowest was − 22.9°C on 11 February 2020. Precipitation occurs in winter, spring and autumn seasons. Precipitation is usually in the form of snow due to the low temperature in winter. The annual average precipitation is 394.4 mm. 16% of precipitation falls in autumn, 38% in spring, 34% in winter and 12% in summer. The number of rainy days is 103. The wettest month is May and the driest month is August with 66.2 mm and 4.1 mm respectively. Wind was very low, relative humidity was moderate.

1.4. Modelling Applications

1.4.1. HYSPLIT Model

The Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) was used in this study which is one of the most widely and easy to use model for calculating atmospheric orbit and dispersion (Stein et al., 2015). HYSPLIT is a tool that was initially created in the 1980s (Draxler, 1982; Draxler and Hess, 1998) and is used for a wide range of tasks, including the calculation of straightforward air parcel trajectories and simulations of intricate transport, dispersion, chemical transformation, and deposition on scales ranging from 1 to 1000. Kilometer-scale Lagrangian particle/puff model. A fixed three-dimensional grid serves as the reference frame for the simulation of advection and diffusion, while a moving reference frame follows the air parcels as they move from their initial points. (Zheng et al., 2021).

1.4.2. DREAM Model

The Dust Regional Atmospheric Model (DREAM) was developed to simulate and/or forecast the cyclical behavior of mineral dust aerosols in the atmosphere and solves the powder mass continuity partial differential nonlinear equation of the Euler type. DREAM is fully integrated as one of the governing equations in the NCEP/Eta atmospheric model. The atmospheric dust cycle's primary mechanisms are all simulated using the concentration equation. During model integration, surface powder injection fluxes are calculated across model points designated as desert. The dust aerosol is driven by atmospheric pattern variables once it is released into the atmosphere: turbulent parameters early in the process when dust is lifted from the ground to the upper levels; model winds later in the process when dust is removed from the sources; and finally, by the wind speed and direction. (Nickovic et al., 2001; Nickovic, 2005; Pérez et al., 2006).

1.4.3. NASA AIRS PBL Model

A variation on the Atmospheric Infrared Sounder (AIRS) Level 2 (L2) and Level 3 (L3) support items is PBLH. Data from infrequent satellite raster observations with a spatial resolution of approximately 50 50 km are included in the AIRS L2 package. By grating and averaging L2 boreholes (i.e., quality scan) successfully taken in one day and one month into 1° 1° lat/long grid cells, AIRS L3 products are produced (Ding et al., 2020; Tian et al., 2020). The information from the ascending/day and descending/night trajectories is gridded and averaged individually. AIRS PBLH is the pressure in hPa of the layer profile relative humidity (relative to the liquid phase of water) determined in the support product pressure layer grid at the level with the biggest gradient (Martins et al., 2010).

1.5. Statistical Analysis

The nearly six years PM_2.5 data from the AQMS for the province of Sivas were used for statistical analysis. The daily means with at least 75% of valid hourly data were kept after the hourly measurements were averaged to get daily values. The remaining data after the elimination was then examined using exploratory data analysis and presented as time series and boxplots. For the purposes of this study's trend analysis dataset, the entire year was separated into four seasons: spring (March, April, and May), summer (June, July, and August), fall (September, and October), and winter (November, December, January, and February). Statistics were used to analyze both the PM_2.5 concentration data and the meteorological data. The five meteorological factors of temperature, relative humidity, wind speed, wind direction, of Sivas were examined according to the time series. For the particular analysis, the values for the study's standard deviation from the data were added. After that, descriptive analyses were carried out to compile participant characteristics and differences as well as the overall central tendency and variability of 6-year average PM_2.5 concentrations. The association of PM_2.5 without each of the researched comorbidities was also looked at in sensitivity analyses. The coefficient of determination (r) in a statistical model represents the proportion of the variance in the dependent variable that the independent variables collectively explain. The correlation method was used to calculate the relationship between PM_2.5 and meteorology results.

The obtained data were used for statistical analysis and plots using the open-source statistical programming language R under the version of 4.0.0. (R Core Team, 2019). Data from the Sivas AQMS and PBL values were examined, as well as meteorological station data.

The air pollution and meteorology data of the Sivas region for both 6 years (Fig. 2.) and 2017, which were examined in more detail, were examined together at the monthly and 6-year PM_2.5 analyses for Istasyon Kavsagi station, the first thing that stands out is that the values that increase in the winter months are at the minimum level in the summer months, especially at the end of 2016 and the beginning of 2017, the pollution values reached the maximum (it is seen that December data is missing in many years).

Table 2

Statistical summaries of daily PM_2.5 concentrations (µg/m³).
	Minimum	Maximum	Mean	Standart deviation	1st quartile	Median	3rd quartile	Number of not available data (day)	Number of exceedances –EU (day)
2016	2.4	121.37	31.15	19.73	18.87	24.71	38.44	26 (7%)	173 (7%)
2017	9.15	115.92	31.66	19.2	18.45	24.89	37.79	16 (4%)	171 (47%)
2018	6.75	77.56	26.91	13.12	17.37	22.56	33.06	23 (6%)	134 (37%)
2019	9.24	81.08	24.89	12.24	16.23	20.95	30.75	173 (47%)	71 (19%)
2020	3.37	137.03	24.92	16.01	14.24	20.1	30.39	43 (12%)	116 (47%)
2021	5.7	66.13	22.01	11.8	12.86	18.41	28.58	13 (4%)	113 (31%)

The fact that pollution values change in the same month in the same periods every year, even if the maximum values change except for 2019, which has missing data, shows the effect of seasonality on Sivas. The annual limit values for Istasyon Kavsagi AQMS in Sivas province were 31.7 ± 8, 31.6 ± 8, 26.5 ± 6, 24.9 ± 3, 25 ± 5 and 22 ± 4 µg/m³, respectively (Table 2.). Boxplot analyses showed that the mean value (red dot) was always higher than the median value (purple line) when looking at the daily mean values for 6 years. This analysis can also be explained by the fact that for each year the data between the median and Q3 (between Q2 and Q3) are higher data number than the data between the median and Q1 (between Q1 and Q2). The fact that most of the daily average pollution data fall between 75% and 50% of the data indicates that the seasonal extreme increases in particulate matter have a regular increase rather than outliers (Fig. 3.).

Hourly measuerements of PM_2.5 showed the effect of seasonality on Sivas province. is seen more strongly on morning period (03.00 to 06.00 AM). Daily average PM_2.5 values, which started in November-December of the previous year and exceeded the daily PM_2.5 limit value, continued until June 2017, and limit value exceedances were observed on spring and some summer days in addition to the temperature-dependent seasonality. Although this also strengthens the opinion that the main reason for PM_2.5 exceedances is seasonality due to temperature (direct effect of domestic heating), the limit value exceedances in relatively months with high temperature reveal the problem of Sivas province on the basis of PM_2.5 which can be seen on Fig. 4.

2.1. Analysis of PM_2.5 vs Meteorological Parameters

The relationship between temperature and PM_2.5 in Sivas province, where the temperature frequently decreased to minus degrees in winter, it is seen when the hourly values are examined both daily (weekdays - weekends and every hour of an average day) and monthly changes for 2017 are examined together on Fig. 5., it is seen that although the temperature values start to increase from February and reach the highest level in August (from 0 to 2 in Normalised level), PM_2.5 values do not move directly negatively with temperature. Contrary to the studies where seasonal pollution is common, PM_2.5 values in Sivas province for 2017 increase in February in direct proportion to the temperature and even reach their highest values (from 1.2 to 2 in Normalised level). In addition, both temperature and PM_2.5 follow a similar pattern according to the 24-hour analyses every day of the week, with temperature starting to increase from 12 AM, reaching its highest values at 1 PM and then decreasing again to reach its minimum value at 12 AM on the following day, but PM_2.5 does not act completely opposite to this pattern as expected and continues to increase especially between 10 − 1 PM when the temperature is the highest, although it reaches its highest values between 9–11 PM when the temperature is the lowest (the time interval when domestic heating is the highest), the reason for this increase at noon may be due to inversion.

The correlations of meteorology and Particulate Matter (2.5 µm and 10 µm) values for Istasyon Kavsagi AQMS in 2017 are examined and showed on Fig. 6., it is seen that the most relevant correlation value is between temperature and humidity due to both very cold winter season and arid climate (r = -0.79), another relevant correlation value is between PM₁₀ and PM_2.5 (r = 0.69), but in addition to these, no clear correlation was found between meteorological parameters in Sivas province, where very low precipitation and dry weather prevail (r = -0.44 between humidity and wind speed, r = -0.37 between temperature and cloudiness). In addition, partial negative correlations were found between PM_2.5 and temperature and between PM₁₀ and temperature in Sivas province (r= -0.44 and r= -0.34, respectively), where negative correlation would be expected to be the highest in a region where seasonality is effective.

As mentioned, this can be explained by the fact that particulate matter pollution increases proportionally with temperature at noon (although the rest of the day tends in the opposite direction), and the fact that PM_2.5 has a higher correlation value than PM₁₀ value suggests that PM₁₀ pollution is higher from pollutant sources independent of temperature (such as road vehicles and big point sources such as industry).

2.2. PBL and PM_2.5 Relationship

PBL is an important parameter in AQ which shows the mixing height during a day. Air pollutants disperse through the mixing height which directly affects the AQ of a region. Lower PBL values indicate a possible stable atmosphere and also an inversion event. In this study, the data of PBL and PM_2.5 daily mean values for 2017 were analyzed to investigate the relationship between PBL and PM_2.5 pollution in Sivas. Sivas region, where the air temperature is -10 or even − 20 degrees in winter, is prone to inversion events. Results showed that on Fig. 7. (left) the highest PM_2.5 and the lowest PBL values were observed in winter months which may show the possible effects of inversion events. PM_2.5 pollution in Sivas region increased to a maximum daily average of 120 ug/m³ during low PBL levels in winter months. However, low PBL values are not sufficient to explain high PM_2.5 values, since PM_2.5 pollution also increases especially in the winter months due to the consumption of coal for heating. Thus, pearson correlation test was applied to understand the correlation between PBL and PM_2.5 values throughout the year and it can be also seen on Fig. 7. (right). In the correlation test, the normalized (log10) values of both datasets were taken as a basis and thus the margin of error due to the units was eliminated. The correlation between them was r = -0.61 which shows that there is a significant negative correlation between lower PBL values and PM_2.5. Therefore, we can say that low mixing height values significantly affect high PM_2.5 pollution levels in the province during a year .

2.3. Long-Range Transport Analysis with HYSPLIT and DREAM Models

The selected year of dust storm period which is 15 to 16 Octover 2018, that there were abnormally high PM_2.5 values (Fig. 8.) and when this increase period was compared with other meteorological parameters, it was seen that the most likely option was an external dust transport. Although Sivas province has an arid climate and bowl-shaped geography, it is exposed to dust transports that cause particulate matter exceedances throughout the year.

Contrary to expectations, this transport comes from two directions, not just one, as shown by the HYSPLIT Backward Trajectories model run for 500–1500 and 3000 meters on October 16, 2018 (Fig. 9.). The particulate matter transports coming from the Arabian Peninsula and entering Türkiye via Syria come from 1600 km away in 2 days, and in addition to these, the starting level is ground level. It is clear that the transport from Africa comes from 2400 km away in 3 days with the highest altitude of 7500 meters.

The outputs of the DREAM-8 model for the date of October 15, 2018 were examined. It is worth noting that since Türkiye is located in the GMT + 3 time zone, the map given by the model outputs for each map should be considered by adding + 3 at the AQMS time. Model outputs are considered as 3 different types of maps. The first of these is the dust concentration at ground level (a, b, c, d on Fig. 10.) (µg/m³), the other is the dust accumulation at the level of 700 mb which is around 3 km above of the surface (e, f, g, h on Fig. 10.) (g/m³) and dry dust deposition (i,j,k,l on Fig. 10.) (mg/m³). Maps of the dust concentration showing ground level PM_2.5 pollution and transport patterns of PM_2.5. However, the most important feature of the second and third types of maps is that as indicated in the HYSPLIT model, they show the dusts transported in the upper levels of the atmosphere and the regions where these dusts deposited to the ground level and cause pollution. DREAM-8 model map outputs clearly show the PM_2.5 pollution carried over Arabian Peninsula and Syria at both ground and 700 mb level on 15 January 2018. In addition to these, the ground level maps showed that ground level winds have a very important effect on the progression of particulate matter pollution towards Sivas province, which is the study area.

Pattern of dust episode at ground level, deposition at 700 mb level and extreme PM_2.5 concentration of AQMS data progresses in a correlated manner over time at 15–16 October 2018. The air parcel which moved towards the Sivas province has a high dust concentration of 700 mb and the dry deposition value of the province is at max on 15 October 2018 12Z (e and i on Fig. 10.). It is seen that both the dust concentration at the level of 700 mb and the dry deposition values in this province decreased over time and approached zero around 15 October 2018 21Z, when the pollution reached its highest values (d, h and l on Fig. 10.). The analysis of these outputs showed that the dust pollution that came to the study area on 15–16 October 2018 was not only transported from the ground level, but also from the 700 mb level. As a result, of the dust particles which transported from South of Türkiye increased the PM_2.5 concentration values well above the average values in the mentioned time period (100–150 µg/m³ per hour). However, unlike the results of the HYSPLIT model, the DREAM-8 model results did not observe a transport originating from Africa and the Sahara Desert. Among the main reasons for this are the decrease in the intensity of the winds blowing towards Türkiye over the Mediterranean and/or the collapsed of the transported dust particles to other regions by dry deposition before they reach the Sivas province.

In air pollution and especially particulate matter analysis, it is impossible to ignore the climate, meteorological characteristics and urban habits of the people. Sivas is a successful example in representing the Northern cities of Türkiye with its location, climate and developing province structure. While the population approaching 900 thousand and the continuously increasing number of vehicles and industry represent this development, the fact that the province started to use natural gas instead of coal for domestic heating and industry since 2016 is an important detail to show the effect of fuel type-related pollution in cities. Sivas represents the meteorological characteristics of the Central-Eastern Anatolia region with very cold winters and dry summers, and the province is exposed to dust transports and inversion from time to time. In this study, the PM_2.5 analysis of Sivas province in the last 6 years has been carried out and both the relationship with meteorology and the reasons for the periods with high PM_2.5 concentrations have been analysed with various methods and data. The results of the study showed similar results with similar studies conducted in different regions in different time periods. As in Hu et al., in 2017 studied which conducted and based on the 15-day PM_2.5 pollution episode in January 2015, PM_2.5 concentrations which increased to 3 times than the average concentrations (80 µg/m³), especially when traffic and meteorological parameters are dense and highly effective on air pollution. In measurements made for Beijing, 200 µg/m³ average concentrations were found for PM_2.5. Ma et al., conducted a similar study in 2017 but includes a 2-year analysis instead of an episode for meteorological parameters like temperature and precipitation. The time periods in which air pollution concentrations in the study area exceeded 100 µg/m³ were examined, and the results showed that periods of dense air pollution were associated with low precipitation and low temperature in winter, and calm air and above-average temperature values in summer. In addition to the periodicity of meteorological effects, Guo et al., as a result of their analysis in 2022, revealed a direct relationship between high pollution concentrations in urban air and low PBL levels between 2015 and 2017, and associated all 40 episodes of high air pollution concentration (specially PM) with PBL levels for winter months. In the study, especially the year 2017, which has the highest values, was examined in detail and correlated with meteorological data. In addition, an episode in 2018 was analysed using the HYSPLIT and DREAM-8 transport model and it was understood that the PM_2.5 concentration, which increased during this period, increased due to dust transport. The convection was seen to originate from the ground and 700 mb (3000 m) levels, and it was deduced that the 700 mb level convection affected the province air by causing dry deposition over the Sivas region with strong winds. The Sahara Desert transport that was observed in the HYSPLIT model but not in the DREAM-8 model is the primary distinction between the two models. Additionally, NASA's PBL data and PM_2.5 pollution for the AQMS region were examined in the study, and it was discovered that there was a bad correlation between them (r=-0.61). It has been observed that these pollutants cannot be transported to the upper atmosphere during the day and increase their concentrations by remaining in the province air, especially in Sivas, where there is intense PM_2.5 pollution from the emission sources during the winter months when PBL is at its lowest level. This is true even though PBL alone is not enough to be the cause of air pollution. The study has important findings both in terms of analysing the PM_2.5 pollution of Sivas and showing the progression trend of this pollution in Anatolian cities with cold and arid climates, and it was found that Sivas was exposed to high PM_2.5 pollution between 2016–2021 (except 2019 with missing data), with the main reason being seasonality and domestic heating. Although no high correlation value was found between temperature and PM_2.5 concentration in the correlation analyses, even the values that are almost 3 times higher than the average every year in January, February, November and December. This finding also demonstrates that domestic emission sources related to the winter have a significant impact on PM concentrations in the province of Sivas, as do year-round pollutant sources like industry and vehicles.

Sivas stands out among the Central Anatolian cities as an important example with its geography, urbanization, and climate. This study stands out as the first and most thorough study in this area in terms of analysis of PM_2.5 concentration. It was made using the ground measurement station data of Sivas, various convection models, and atmospheric mixing height values. The study's findings demonstrated that, despite the harsh winters in Sivas, pollution persists throughout the year and is transported into areas with high pollution concentrations. It is anticipated that the study will serve as a good reference point for research in this area that includes more in-depth analyses.

Acknowledgements

The authors would like to thank the Ministry of Environment, Urbanization and Climate Change of Türkiye for providing AQ data.

Baltaci, H. (2017). Spatial and temporal variation of the extreme Saharan dust event over Turkey in March 2016. Atmosphere, 8(2), 41.
Çapraz, Ö., & Deniz, A. (2021). Particulate matter (PM10 and PM2. 5) concentrations during a Saharan dust episode in Istanbul. Air Quality, Atmosphere & Health, 14(1), 109-116.
Çapraz, Ö., Deniz, A., & Doğan, N. (2017). Effects of air pollution on respiratory hospital admissions in İstanbul, Turkey, 2013 to 2015. Chemosphere, 181, 544-550.
Çapraz, Ö., Efe, B., & Deniz, A. (2016). Study on the association between air pollution and mortality in İstanbul, 2007–2012. Atmospheric Pollution Research, 7(1), 147-154.
Draxler, R. R. (1982). Measuring and modeling the transport and dispersion of kRYPTON-85 1500km from a point source. Atmospheric Environment (1967), 16(12), 2763-2776.
Draxler, R. R., & Hess, G. D. (1998). An overview of the HYSPLIT_4 modelling system for trajectories. Australian meteorological magazine, 47(4), 295-308.
Ding, F., Savtchenko, A., Hearty, T. J., Wei, J., Theobald, M., Vollmer, B., et al. (2020). Assessing the impacts of two averaging methods on AIRS Level 3 monthly products and multiyear monthly means. Journal of Atmospheric and Oceanic Technology, 37, 1027–1050.
Efe, B., Öztaner, Y. B., Deniz, A., & Unal, A. (2022). Analysis of air pollutants in Kagithane valley and Istanbul Metropolitan Area. Air Quality, Atmosphere & Health, 1-15.
Environment, SA. Available online: http://www.environnement-sa.com/ (accessed on 01 January 2023).
Environmental Protection Agency (USEPA). (2022). https://www.epa.gov/pm-pollution/particulate-matter-pm-basics. Accessed on 17 January 2023.
Flores, R. M., Özdemir, H., Akkoyunlu, B. O., Ünal, A., & Tayanç, M. (2020). Seasonal variation of carbonaceous PM2. 5 in an Istanbul traffic site. Atmospheric Pollution Research, 11(12), 2110-2118.
Fortelli, A., Scafetta, N., & Mazzarella, A. (2016). Influence of synoptic and local atmospheric patterns on PM10 air pollution levels: a model application to Naples (Italy). Atmospheric Environment, 143, 218–228.
Galindo, N., Varea, M., Gil-Moltó, J., Yubero, E., & Nicolás, J. (2011). The influence of meteorology on particulate matter concentrations at an urban Mediterranean location. Water, Air, & Soil Pollution, 215(1), 365-372.
Ghorani-Azam, A., Riahi-Zanjani, B., & Balali-Mood, M. (2016). Effects of air pollution on human health and practical measures for prevention in Iran. Journal of research in medical sciences: the official journal of Isfahan University of Medical Sciences, 21.
Guo, Q., Wu, D., Yu, C., Wang, T., Ji, M., & Wang, X. (2022). Impacts of meteorological parameters on the occurrence of air pollution episodes in the Sichuan basin. Journal of Environmental Sciences, 114, 308-321.
He, J., Gong, S., Yu, Y., Yu, L., Wu, L., Mao, H., ... & Li, R. (2017). Air pollution characteristics and their relation to meteorological conditions during 2014–2015 in major Chinese cities. Environmental pollution, 223, 484-496.
Hu, D., Wu, J., Tian, K., Liao, L., Xu, M., & Du, Y. (2017). Urban air quality, meteorology and traffic linkages: Evidence from a sixteen-day particulate matter pollution event in December 2015, Beijing. Journal of Environmental Sciences, 59, 30-38.
Kabatas, B., Pierce, R. B., Unal, A., Rogal, M. J., & Lenzen, A. (2018). April 2008 Saharan dust event: Its contribution to PM10 concentrations over the Anatolian Peninsula and relation with synoptic conditions. Science of the Total Environment, 633, 317-328.
Kelly, F. J., & Fussell, J. C. (2015). Air pollution and public health: emerging hazards and improved understanding of risk. Environmental geochemistry and health, 37(4), 631-649.
Laña, I., Del Ser, J., Padró, A., Vélez, M., & Casanova-Mateo, C. (2016). The role of local urban traffic and meteorological conditions in air pollution: A data-based case study in Madrid, Spain. Atmospheric Environment, 145, 424-438.
Ma, T., Duan, F., He, K., Qin, Y., Tong, D., Geng, G., ... & Ma, Y. (2019). Air pollution characteristics and their relationship with emissions and meteorology in the Yangtze River Delta region during 2014–2016. Journal of Environmental Sciences, 83, 8-20.
Ma, Y., Wang, M., Wang, S., Wang, Y., Feng, L., & Wu, K. (2021). Air pollutant emission characteristics and HYSPLIT model analysis during heating period in Shenyang, China. Environmental Monitoring and Assessment, 193(1), 1-14.
Manisalidis, I., Stavropoulou, E., Stavropoulos, A., & Bezirtzoglou, E. (2020). Environmental and health impacts of air pollution: a review. Frontiers in public health, 14.
Matthias, V., Arndt, J. A., Aulinger, A., Bieser, J., Denier van der Gon, H., Kranenburg, R., ... & Quante, M. (2018). Modeling emissions for three-dimensional atmospheric chemistry transport models. Journal of the Air & Waste Management Association, 68(8), 763-800.
Mentese, S., & Ogurtani, S. Ö. (2022). Spatial and temporal look at ten-years air quality of Istanbul city. International Journal of Environmental Science and Technology, 19(2), 925-938.
Nickovic, S., Kallos, G., Papadopoulos, A., & Kakaliagou, O. (2001). A model for prediction of desert dust cycle in the atmosphere. Journal of Geophysical Research: Atmospheres, 106(D16), 18113-18129.
Nickovic, S. (2005, January). Distribution of dust mass over particle sizes: impacts on atmospheric optics. In Forth ADEC Workshop–Aeolian Dust Experiment on Climate Impact (pp. 26-28).
Ozdemir, H. (2019). Mitigation impact of roadside trees on fine particle pollution. Science of the Total Environment, 659, 1176-1185.
Ozdemir, H., Pozzoli, L., Kindap, T., Demir, G., Mertoglu, B., Mihalopoulos, N., ... & Unal, A. (2014). Spatial and temporal analysis of black carbon aerosols in Istanbul megacity. Science of the total environment, 473, 451-458.
Özdemir, E. T. (2019). Investigations of a southerly non-convective high wind event in Turkey and effects on PM10 values: a case study on April 18, 2012. Pure and Applied Geophysics, 176(10), 4599-4622.
Özdemir, E. T., Çapraz, Ö. & Deniz A. (2020). Analysis of Air Pollutants at Extreme Pressure Values in İstanbul, Turkey. 1 st International Conference on Applications of Air Quality in Science and Engineering (ICAAQSE 2020), 10-12 Feb 2020, Kuwait, 571-575.
Öztürk, M. Z., Çetinkaya, G., & AYDIN, S. (2017). Köppen-Geiger iklim sınıflandırmasına göre Türkiye’nin iklim tipleri. Coğrafya Dergisi, (35), 17-27.
R Team. CR Core Team R: A Language and Environment for Statistical Computing. https://www.r-project.org/ (2019)
Querol, X., Alastuey, A., Ruiz, C. R., Artiñano, B., Hansson, H. C., Harrison, R. M., ... & Schneider, J. (2004). Speciation and origin of PM10 and PM2. 5 in selected European cities. Atmospheric environment, 38(38), 6547-6555.
Reizer, M., & Juda-Rezler, K. (2016). Explaining the high PM10 concentrations observed in Polish urban areas. Air Quality, Atmosphere & Health, 9(5), 517-531.
Pérez, C., Nickovic, S., Baldasano, J. M., Sicard, M., Rocadenbosch, F., & Cachorro, V. E. (2006). A long Saharan dust event over the western Mediterranean: Lidar, Sun photometer observations, and regional dust modeling. Journal of Geophysical Research: Atmospheres, 111(D15).
Salvador, P., Artíñano, B., Querol, X., & Alastuey, A. (2008). A combined analysis of backward trajectories and aerosol chemistry to characterise long-range transport episodes of particulate matter: the Madrid air basin, a case study. Science of the total environment, 390(2-3), 495-506.
Stein, A. F., Draxler, R. R., Rolph, G. D., Stunder, B. J., Cohen, M. D., & Ngan, F. (2015). NOAA’s HYSPLIT atmospheric transport and dispersion modeling system. Bulletin of the American Meteorological Society, 96(12), 2059-2077.
Tayanc, M., Sezen, I., Ünal, A., Flores, R. M., & Karanfil, S. (2022). A holistic approach to the air quality of Konya City, Turkey. Air Quality, Atmosphere & Health, 1-15.
Tian, B., Manning, E., Roman, J., Thrastarson, H., Fetzer, E., & Monarrez, R. (2020). AIRS version 7 Level 3 product user guide. Jet Propulsion Laboratory, California Institute of Technology.4
Toffoli, A., Bitner‐Gregersen, E. M., Osborne, A. R., Serio, M., Monbaliu, J., & Onorato, M. (2011). Extreme waves in random crossing seas: Laboratory experiments and numerical simulations. Geophysical Research Letters, 38(6).
Turkish State Meteorological Service (TSMS). 2023. Climate of Turkey Available from http://www.meteor.gov.tr Accessed on 17 January 2023
Turkish Statistical Institute (TURKSTAT) (2023a) Address based population registration system results. In: Address Based Popul. Regist. Syst. Results 2020. data.tuik.gov.tr/Bulten/Index?p=Adrese-Dayali-Nufus-Kayit-Sistemi-Sonuclari-2020–37210. Accessed on 17 January 2023
Turkish Statistical Institute (TURKSTAT). (2023b). https://www.tuik.gov.tr/. Accessed on 17 January 2023.
Unal, Y. S., Toros, H., Deniz, A., & Incecik, S. (2011). Infuence of meteorological factors and emission sources on spatial and temporal variations of PM10 concentrations in Istanbul metropolitan area. Atmospheric Environment, 45(31), 5504–5513.
World Health Organization. (‎2021)‎. WHO global air quality guidelines: particulate matter (‎PM2.5 and PM10)‎, ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide. World Health Organization. https://apps.who.int/iris/handle/10665/345329. License: CC BY-NC-SA 3.0 IGO
WHO. (2022). https://www.who.int/health-topics/air-pollution. Accessed on 23 Jun 2022
Yavuz, V., Özen, C., Çapraz, Ö., Özdemir, E. T., Deniz, A., Akbayır, İ., & Temur, H. (2022). Analysing of atmospheric conditions and their effects on air quality in Istanbul using SODAR and CEILOMETER. Environmental Science and Pollution Research, 29(11), 16213-16232.
Zhang, Y. (2019). Dynamic effect analysis of meteorological conditions on air pollution: A case study from Beijing. Science of the Total Environment, 684, 178-185.
Zheng, Y., Jiang, F., Feng, S., Cai, Z., Shen, Y., Ying, C., ... & Liu, Q. (2021). Long-range transport of ozone across the eastern China seas: a case study in coastal cities in southeastern China. Science of The Total Environment, 768, 144520.

No competing interests reported.

An Urban Air Quality Assessment Based on a Meteorological Perspective

Status:

Version 1

Abstract

Figures

Introduction

1. Material And Methods

1.1. Study Area

1.2. PM_2.5 Monitoring

1.3. Meteorological Station Data

1.4. Modelling Applications

1.4.1. HYSPLIT Model

1.4.2. DREAM Model

1.4.3. NASA AIRS PBL Model

1.5. Statistical Analysis

2. Results And Discussions

2.1. Analysis of PM_2.5 vs Meteorological Parameters

2.2. PBL and PM_2.5 Relationship

2.3. Long-Range Transport Analysis with HYSPLIT and DREAM Models

3. Conclusion

Declarations

Acknowledgements

References

Additional Declarations

Status:

Version 1

An Urban Air Quality Assessment Based on a Meteorological Perspective

Status:

Version 1

Abstract

Figures

Introduction

1. Material And Methods

1.1. Study Area

1.2. PM2.5 Monitoring

1.3. Meteorological Station Data

1.4. Modelling Applications

1.4.1. HYSPLIT Model

1.4.2. DREAM Model

1.4.3. NASA AIRS PBL Model

1.5. Statistical Analysis

2. Results And Discussions

2.1. Analysis of PM2.5 vs Meteorological Parameters

2.2. PBL and PM2.5 Relationship

2.3. Long-Range Transport Analysis with HYSPLIT and DREAM Models

3. Conclusion

Declarations

Acknowledgements

References

Additional Declarations

Status:

Version 1

1.2. PM_2.5 Monitoring

2.1. Analysis of PM_2.5 vs Meteorological Parameters

2.2. PBL and PM_2.5 Relationship