A Preliminary Assessment of Dispersion Level of SO2 in Fars Industrial Region, South of Iran, by GIS

The city of Zarghan is located 25 km northeast of Shiraz, southern Iran. Zarghan is affected by numerous pollution sources such as oil refinery, an industrial park, and Shiraz-Tehran highway. The numerous contaminating sources around Zarghan can cause serious local air pollution. Sulfur dioxide gas is an important index of air pollution in cities. Therefore, in order to control and manage Zarghan air quality, it is important to monitor sulfur dioxide concentration in the surrounding area. It is also essential to know about the contribution level of other sources of pollution as well as dispersion radius of pollutants in the area. In this study, the concentration of sulfur dioxide was measured by passive sampling at 10 different stations. These values were interpolated in other parts of the city using ArcGIS software. The results of sampling showed that the concentration of the gas was 60 µgm−3 around oil refinery. The level was 19 µgm−3 in region located about 3 km from the oil refinery. It was also demonstrated that the gas concentration was not higher than the standard limit within residential area. On the other hand, the role of the local highway and industrial park was not significant in contaminating air in urban areas.


Introduction
Sulfur dioxide is one of the most important pollutants in urban areas which have caused numerous cases of respiratory diseases and deaths in many communities. Some studies show that there is a significant relationship between daily mortality and exposure to sulfur dioxide. Studies on asthma have not yet determined the safe limit of SO 2 not causing respiratory malfunctioning. In fact, even very low concentrations of sulfur dioxide can cause respiratory capacity reduction [1,2]. Maantay declared that people living in industrialized area of the Bronx and New York City were more susceptible for asthma (>66%) due to sulfur dioxide [3].
Previous studies have revealed that oil refineries were one of the major sources of emitting SO 2 around and in the city [4]. Many other studies showed the effects of highways on dispersing SO 2 around nearby cities [5]. Industrial parks could also be considered as other sources of the SO 2 pollutant. The size of the industry, its distance from the neighboring town, and direction of the wind were among the most important parameters causing air pollution in cities. Many cities are being affected by a series of air pollution sources. Having a thorough knowledge on these sources is an important tool in controlling and managing them.
Geographic Information System (GIS) software is a powerful tool to determine the contribution levels of such sources. So far, many studies on controlling air pollution have been carried out using this software [6,7]. In order to monitor geographical dispersion of a pollutant in an area, it is required that sample pollutant is to be examined at different stations simultaneously. Due to low cost and easy operation of diffusive method, this technique has been widely used to monitor air pollution with high precision in large scales. Diffusive sampling is a powerful tool to take samples 2 Journal of Environmental and Public Health of atmospheric gas using molecular diffusion with controlled speed that has no need to pump air during sampling [8]. Many studies on the accuracy of passive sampling of SO 2 have shown that this method had an acceptable error rate compared to other methods of sampling [9]. Adame et al. [10] used K-mean cluster technique to measure daily cycles for SO 2 pollutant at different air quality regimes in a heavily industrialized area in Spain. Four optimal cluster numbers were obtained for the daily patterns of SO 2 . While two of the clusters showed a low mixing ratio, the others exceeded the thresholds limits of air quality [10]. Bhanarkar et al. used a comprehensive emission inventory and dispersion modeling approaches to determine the contribution of SO 2 from different types of pollution sources in Jamshedpur in India. Although the first approach showed that industrial sources account for 77% of the total emissions of SO 2 , the second approach demonstrated that more than 50% of SO 2 was emitted from industrial sources [11]. Studying the effect of emission sources and meteorological conditions on SO 2 pollution in Mongolia was done by Luvsan and his colleagues using multiple regression models. Data revealed that SO 2 concentrations in industrialized area increased with the decrease of wind speed and temperature and with the increase of relative humidity [12]. SO 2 distribution to air quality in Hong Kong was assessed by the multilayer, non-steady-state puff dispersion model. The results showed that the contribution of the power plant located in marine sources regions is significant during both summer and winter times [13].
Since the industrial regions of Zarghan (northeast of Shiraz) are affected by numerous air pollution sources, suitable monitoring systems are urgently needed that can rapidly and reliably detect and quantify polluting sources. Therefore, the main objectives of this study were to (i) determine preliminary assessment of the dispersion level of SO 2 using passive sampling and GIS software and (ii) assess the contribution level of generating sources of SO 2 in urban areas.

Zarghan Climatic Conditions.
In order to study the dispersion level of SO 2 in Zarghan, knowing the information about the local climatic condition is very essential. Zarghan average annual temperature is 16 ∘ C, and the total average annual rainfall is 330.2 mm. More than 56% of the precipitation falls during winter, and the number of rainy days is 48 days. The average annual humidity is 42%. Based on Domarten climate classification, the city is located on the semiarid area. Monthly average wind speed at the time of air sampling is 0.9 ms −1 .

Locating Air Pollutant Sources in the Study
Area. At first, the major sources of air pollution in the surrounding area of Zarghan are studied. Zarghan is located 25 km northeast of Shiraz, nearby Shiraz-Tehran highway. Many industrial complexes are located about 10 km from Zarghan among which one can name oil refinery, industrial park with more than 35 industries, ceramic industry, Loab industry, and chemical industry. Besides, due to a rocky wall of 300 m height in the east side of the town, the air flow through the town is often blocked. Each of the above sources individually could be the cause of air pollution in Zarghan.

Air Sampling for Detecting 2 .
A diffusive sampler consisted of a tube in which the adsorbent material adsorbed the pollutant. Speed-controlled pollutant entered the adsorbent tube by molecular diffusion with no need for electrical power or pumping. Therefore, SO 2 sampling was based on the adsorbent molecular diffusion. In this research, sampling tubes were sealed and carried to the lab for analysis of samples after two weeks of adsorption period. The tubes were of 20 mm diameters and made of polypropylene. To reduce the turbulence effects of the wind and rain, the sampling was taken by a metal clamp inside a fiberglass container as shown in Figure 1.
After collecting the specimens, they were sent to Pasam Company in Switzerland for the determination of SO 2 . Mixture of potassium carbonate and glycerin was used to extract SO 2 and the total amount of extracted SO 2 determined by ion chromatography.

2.4.
2 Sample Collecting Locations. According to various sources, which are believed to be effective in Zarghan air pollution, boundary conditions for creating a mesh net and the locations of sampling were determined. Geographical coordinates, names, and reasons for choosing the sampling This residential point is located at the end of the town and the farthest to the oil refinery.
7 N e a r w a y S q . 3291485.09MN 663854.37ME This is the first urban area after oil refinery, located between Zarghan and the oil refinery.  points are given in Table 1. Due to the small size of the study area, 10 points were selected and one sample was collected at each point. Moreover, it should be mentioned that statistical methods were not used in choosing the number of samples. Sampling adsorbents were installed by a crane at a height of 3 to 4 meters from the ground adjacent to electrical towers. The adsorbents were exposed to outdoor SO 2 for 17 days from the second to the nineteenth of January, 2012.

Drawing GIS Maps.
At the beginning, an image of the coverage area was obtained by Google Earth software. Geographical coordinates of 4 points of suitable dispersion were determined by the software and used in an Excel file as the ground reference ( Figure 2).
Using ArcGIS software, the image was processed as the ground reference and the result saved in TIFF format with a pixel size of 5 m. The data obtained from sampling of SO 2 were interpolated using the passive sampling method with different methods such as inverse distance weighing (IDW), natural nearest neighbor. Kriging format (TIFF) is a raster image format with a pixel size of 5 m for each specimen being prepared (with at least 10 sample points). Then, all interpolated layers were cut into the size of sample point's area to perform interpolation for all layers.

3.1.
2 Passive Sampling. Table 2 shows the results of SO 2 measurements in different stations in the industrial regions of Zarghan.

Regional Variation of
2 Concentration Using GIS Software. Due to the importance of wind direction, maps of the wind rose of the sampling season and Zarghan are shown in Figure 3. As shown in Figure 4, geographic concentration coverage area of SO 2 has been studied using different methods of nearest neighbor, IWD, and Kriging. SO 2 pollution maps were prepared using passive sample interpolation.

A Comparison of Measured 2 Concentration with the Values Recommended by the WHO Air Quality Guidelines.
WHO (World Health Organization) guideline value for 24 hours of SO 2 concentration was set as 20 gm −3 [14]. Since in this study the passive sampling method was used and the specimens were exposed to the air pollution in many of the studied stations for 17 days, the data were comparable to the average 24-hour SO 2 guideline. In general, many of the stations where samples were collected had acceptable SO 2 levels except the following 4 stations which had considerable pollution levels.
(1) The station of oil refinery with the value of 63.3 gm −3 .
(2) The station near street with the value of 29.7 gm −3 .
Journal of Environmental and Public Health 5

GIS Data Analysis.
As GIS pictures showed, among generating sources of SO 2 pollution, the oil refinery and the highway were of more importance. In general, the concentration of SO 2 was much greater around the oil refinery than near the highway. Although the prevailing wind direction at the time of sampling was not toward Zarghan, it could affect point located 3 km away. In addition the industries located in the industrial park could not have a major effect on SO 2 concentration in Zarghan. The concentration of SO 2 in the industrial park with 4000-meter distance from theoil refinery was 24 gm −3 showing a higher concentration compared to point located 3 km away.
On the other hand, SO 2 concentration at the distance of 200 meters from the highway inside the industrial park was greater than the one at station adjacent to the highway. It was concluded that the industrial park had intensified air pollution caused by the oil refinery by an amount of 5 gm −3 . Although the prevailing wind in the highway blew toward Zarghan, it could have affected only 150 meters around. The 24-hour concentration of SO 2 was just on standard border line in point located 3 km away, while it was within standard limits in Zarghan urban areas. The reason for higher level of SO 2 in point located 3 km away was its adjacency to the oil refinery.
According to the simulation of pollutants concentration in the atmosphere, using CALPUFF nonstationary Gaussian model at Rio Grande City, the concentration of CO reached 28.019 mg/m 3 , the SO 2 reached 4.8623 mg/m 3 , and NO 2 reached 40.8490 mg/m 3 [15].
Similar study showed that most areas of Dallas County only experienced small variations of SO 2 concentration over the period from 1996 to 2002 and that annual SO 2 concentration in some areas of the county greatly increased over this period. In addition, results from the source-contribution analysis indicated that the spatial-temporal variations in SO 2 concentration in Dallas County from 1996 to 2002 were not attributed to any particular type of emission source, but industrial emission sources could be the origin of extreme variations in SO 2 concentrations in some areas [16].
Application of the SIM-air modeling tool in six Indian cities estimated that 21,400 premature deaths will be in the six cities in 2020; implementation of the six interventions in the transport and brick kiln sectors can potentially save 5870 lives (27%) annually and result in an annual reduction of 16.8 million tons of carbon dioxide emissions in the six cities [17].
As seen from the images, the nearest neighbor interpolation method showed the linear effects of pollution on the highway clearly. Pictures obtained by interpolation method revealed the expansion of pollutions in the area.

Conclusions
As shown, Zarghan has been located in the green zone, and SO 2 could not be accumulated in the town despite the existence of wind. According to the wind rose map, during January, the prevailing direction was not toward Zarghan. Therefore, it would be possible to record higher level of SO 2 concentration in Zarghan if samples were collected during windy months blowing toward the city. On the contrary, the pollution due to the highway was affected by prevailing winds blowing toward the city, and, as seen in the GIS images, the highway pollution could not affect the city. It would be highly recommended to repeat the experiment during other months of the year, especially October, in which the prevailing wind direction is toward Zarghan.