1 Introduction

Developed by the Atmospheric Studies Group, CALPUFF is considered one of the most powerful environmental modelling programs. The fact that CALPUFF can be run for any location worldwide and for any modelling period has attracted researchers wishing to simulate the dispersion of selected pollutants or model pollutant concentration levels associated with investigating geophysical and meteorological conditions.

In recent years, a tremendous amount of research has been published on CALPUFF modelling and different pollutant concentrations (Liao et al. 2014; Feng et al. 2014; Wu and Nelson 2014). Abdul-Wahab et al. (2013) focused on an examination of the impact of geophysical and meteorological conditions on the dispersion of nitrogen dioxide (NO2) emitted from a proposed refinery located in Canada. Abdul-Wahab et al. (2014a) also employed CALPUFF to examine the effect of meteorological conditions on the dispersion of an accidental release of hydrogen sulfide (H2S). CALPUFF software has also been applied for modelling greenhouse gas emissions from the steady and non-steady state operations of a combined cycle power plant located in Ontario, Canada (Abdul-Wahab et al. 2014b). Examining the dispersion of vinyl chloride (C2H3Cl) from a renewable energy facility located in Ontario, Canada was the main aim of a study by Obaid et al. (2014). Other case studies, for which CALPUFF software was used to study concentrations for multiple pollutants include Ghannam and El-Fadel (2013), Tian et al. (2013) and Abdul-Wahab et al. (2015).

Salalah, considered Oman’s second largest city and the capital of Dhofar region, experiences a monsoon season from mid-June to mid-August. During this season, Salalah’s surroundings and its landscape is transformed completely from brown to shiny green misty pastures. Due to Salalah’s monsoon season that attracts tourists, vehicle activities increased drastically over the past years. This certainly triggered health related concerns in the long run since vehicular traffic is considered as one of the major source of CO, NOx, and CO2. The high levels of these pollutants cause damage to both human health and the environment.

Despite the fact that considerable amount of studies have taken into consideration the vehicle emissions and the dispersion of line source-emitted pollutants (Oettl et al. 2003; Hong-di and Wei-Zhen 2012; Rosemont Copper Company 2011), no studies have focused on investigating the trends of line source-emitted pollutants for Salalah for the past years. Thus, this paper focuses on employing CALPUFF modelling software for estimating the trends for the dispersion of CO, NOx, and CO2. As the number of cars in Salalah has increased significantly within the past years, the main objective of this study was to assess the concentration levels of CO, NOx, and CO2 from 2004 to 2014. The developed knowledge from this study will provide a strong evidence to convince Salalah’s decision-makers to take the necessary action regarding traffic jams to reduce environmental stresses and also plan vehicle type, routing, and city planning now before the problem becomes out of control in the future.

2 Materials and methods

2.1 Area of study description

Located in the southern region of the Sultanate of Oman, Salalah is Oman’s second largest city and the capital of the Dhofar region. Figure 1 represents the location and terrain map of Salalah and the more specific location for the case study. The city generally has an arid climate, although summers are cooler than in more northern or inland parts of Oman. Additionally, Salalah is one of the only places on the Arabian Peninsula, along with Yemen, that experiences a monsoon season (Salalah, http://en.wikipedia.org/wiki/Salalah). From mid-June to mid-August, monsoon clouds from the Indian Ocean bring a constant drizzle to the area, transforming Salalah’s surroundings into an oasis of misty pastures (Salalah, http://www.lonelyplanet.com/oman/dhofar/salalah). During this time, Salalah’s landscape is transformed from brown and dry to shiny, green and lush. As a result, tourists from other parts of Oman as well as from all over the globe arrive to Salalah to enjoy the scenery.

Fig. 1
figure 1

Location and terrain map of Salalah region on the map of Oman

Full size image

From 2004 to 2014, a significant increase in the number of tourists resulted in an increase in the number of vehicles in Salalah. Due to this rise in vehicle numbers, traffic jams were recorded as well (Fig. 2). The street of interest in the current study is divided into three main sections, with each representing an active area within Salalah. Based on Google Earth satellite images of the specified street, statistical data (Table 1) were collected in order to estimate the average number of vehicles passing through this street’s three sections. Data collection was conducted for 2004–2006, 2008, 2009 and 2012–2014. The days were selected based on image availability and the maximum number of vehicles recorded within the provided days. According to the data collected (Fig. 3), there was a significant increase in the number of vehicles in all three street sections. As expected, Section 1 had the highest number of vehicles for many reasons:

  • Section 1 is the longest section of road;

  • It intersects with a large number of streets and has the highest accessibility from other exits and entrances as compared to Sections 2 and 3;

  • Section 1 connects a large number of residential areas in Salalah (Fig. 2);

  • Section 1 is the main road connecting Salalah’s city center and its industrial region;

  • It connects Central Salalah with West Salalah (Uwqad), Salalah Port, and the Free Zone.

  • As the acquisition times in all images were in the morning, most industrial activities would logically be in Section 1.

Fig. 2
figure 2

Case study street location with major sections in Google maps

Full size image
Table 1 Statistical data collection for average number of vehicles entering the domain for the three sections
Full size table
Fig. 3
figure 3

The trend of number of vehicles entering the domain for the three sections

Full size image

Section 2, as demonstrated in Fig. 3, had the lowest record of activity since the exits and entrances connecting this road are limited. However, Section 3 also had a high rate of vehicle use since most of the governmental buildings are located in Al-Saadah, which is in the northeast of Salalah.

For this study, the three sections of the selected street were considered, and the total number of vehicles for each selected day was calculated. It was assumed that the number of vehicles extracted from Google Earth images was for a 5-min time period, which is approximately the time required for each batch of vehicles to be replaced by another batch. Table 1 lists the hourly variation in the number of vehicles for the selected years associated with the days where the maximum number of vehicles was recorded. Figure 4 represents the recorded variation starting from 2004, with approximately 3708 vehicles, and ending in 2014 with a total of approximately 12,996 vehicles.

Fig. 4
figure 4

Hourly recorded variation in vehicle numbers from 2004 to 2014

Full size image

2.2 CALPUFF modelling system

CALPUFF modelling system, also known as CALPro software, is a meteorological and air quality modelling software designed by the Atmospheric Studies Group (ASG) and got approved by U.S. Environmental Protection Agency (EPA). Taking into account the area of interest’s geography and climate factors as coastal and wind conditions, dispersion and concentration levels of air pollutants that are emitted from various sources such as landfills, traffic, and industrial plants can be predicted and investigated by employing CALPUFF software. In addition to a pre-processing package, the software mainly consists of three dominant components: CALMET, CALPUFF, and CALPOST. Additional information related to CALPro software can be found in Scire et al. (2000) and Abdul-Wahab and Fadlallah (2014). As a start, the meteorological grid shared information for Salalah was inserted into a common file, before running the CALPUFF pre-processors, using the Identify Shared Information module listed in Table 2. For this study, starting at 00h00 till 23h00, the model was simulated for selected modelling periods (i.e. 25–27/05/2004, 26–28/09/2005, 2–4/08/2006, 21–23/08/2008, 20–22/06/2009, 4–6/09/2012, 6–8/11/2013, 21–23/01/2014). Considering the line source emission rates shown in Table 4 and the data files produced by CALMET, CALPUFF was run for the pollutants CO, NOx, and CO2 for the selected modelling periods. For the CALPUFF modelling system, the fractional convergence criteria for numerical slug sampling integration and numerical area source integration were set at 1.0E−04 and 1.0E−6, respectively. Concentration levels simulation results and plume trajectories were produced by activating CALPOST. For a visualized simulation of the pollutants plume trajectories and concentration levels, CALVIEW was operated. The selected layer for assessing the effect of CO, NOx, and CO2 on humans distributed within the desired area of interest was the first layer, which is in the range of 0–20 m.

Table 2 Meteorological domain of study model input information
Full size table

2.3 Upper air and surface meteorological data

The upper air meteorological data information was extracted from the listed radiosonde stations on the National Oceanic and Atmospheric Administration’s (NOAA) official website (https://esrl.noaa.gov/raobs/). For a representation of a full time period, 12-h interval data were obtained from the upper air climate station for the selected years associated with the modelling days mentioned earlier. The extracted data were processed in a compatible format that can be run using CALPro’s READ62 program to produce UP.DAT file suitable as an input to CALMET.

On the other hand, the surface meteorological data used in this study were obtained from Oman’s Directorate General of Meteorology and Air Navigation (www.met.gov.om). The hourly meteorological data consist of the following information: precipitation (mm), temperature (°C), station pressure (mbar), wind direction (°), wind speed (m/s), relative humidity (%), cloud cover (tenths), and cloud height (ft). For a full period similar to that of the upper air data period, the obtained data were prepared in a format that could be run using CALPro’s SMERGE program to produce a SURF.DAT file suitable for input into CALMET. The extracted information related to upper air and surface stations is summarized in Table 3.

Table 3 Upper air and surface stations information
Full size table

2.4 Emission data

In order to calculate the emission rates for a line source, emission factors are required. As vehicles come in different types, emission rates differ according to vehicle specifications such as engine capacity, fuel type, vehicle speed, vehicle weight, and so on. Numerous environmental institutes and research papers have concentrated on calculating emission rates. Abdul-Wahab and Fadlallah (2014) made a comparison between various research studies in terms of CO, NOx, and CO2 emission rates. A conclusion was reached based on the fact that Muscat suffers from the absence of information related to vehicle emissions; thus, the emission factors set by the Transport Research Laboratory (TRL) were compared to those found through other studies. In the current study, an Excel sheet developed by the TRL which categorizes vehicles based on certain specifications (e.g., weight, engine capacity, etc.) was used (TRL, https://www.gov.uk/government/publications/road-vehicle-emission-factors-2009). Assuming an average vehicle speed of 34.5 km/h with the following specifications: petrol fuel type, pre-euro emission standard, < 1400 cc engine capacity, and < 2.5 t vehicle weight, CO, NOx, and CO2 emission factors were determined to, respectively, be 15.74, 1.29 and 182.35 g/veh km.

Moving to the calculation related to emission rates, Table 4 demonstrates the calculation of CO, NOx, and CO2 emission rates by taking into consideration the number of vehicles passing through the specified street for the modelling days selected for the study. CALPUFF line source input parameters, including the street coordinates with respect to the located origin, emission height, car width, building dimensions, and the separation between each building, are listed in Table 5.

Table 4 Line source emission rate calculations for the desired pollutants
Full size table
Table 5 CALPUFF line source input parameters
Full size table

2.5 CALPUFF operation

Figure 5 represents CALPUFF operation flow diagram. Simulations were run on a 16-bit Windows XP computer with an Intel Pentium 4 3.4 GHz processer, 1 GB RAM. As a start, the identified shared information common file that includes meteorological and geophysical data for the domain of study, listed in Table 2, were processed before proceeding with CALMET. The geophysical data: terrain, land use, and coastline data were obtained from the ASG official website (www.src.com/datasets/datasets_main.html). The extracted geophysical data were employed to activate terrain (TERREL) and land use (CTGPROC) processing packages. TERREL.DAT and LU.DAT are the resultant data set files from the previous process. These files are essential in running geophysical (MAKEGEO) processing package producing GEO.DAT. As previously mentioned, the meteorological data for both surface and upper air data were extracted and processed with an assistance of SMERGE and READ62 respectively producing SURF.DAT and UP.DAT data files. All the previously produced data files with an extension (.DAT) are considered as CALMET inputs. For this study, starting at 00h00 till 23h00, the model was simulated for the selected modelling periods. Considering the line source emission rates shown in Table 4 and the data files produced by CALMET, CALPUFF was run for the pollutants CO, NOx, and CO2 for the selected modelling periods. Concentration levels simulation results and plume trajectories were produced by activating CALPOST. For a visualized simulation of the pollutants plume trajectories and concentration levels, CALVIEW was operated. The selected layer for assessing the effect of CO, NOx, and CO2 on humans distributed within the desired area of interest was the first layer, which is in the range of 0–20 m.

Fig. 5
figure 5

CALPUFF operation flow diagram

Full size image

3 Results and discussions

According to the CALPUFF-simulated terrain map of Salalah (Fig. 6), the domain consists of various elevation ranges since Salalah’s terrain is mountainous. Additionally, as Salalah is a coastal city, the ocean has a significant influence over its meteorological conditions. Figure 7 represents a wind rose for each of the selected modelling days investigated in this study. Each wind rose shows the distribution of wind direction and speed at the study location. The wind speed ranges are indicated by colored bands, while the frequency of the wind that blows from specific directions is demonstrated by the length of the spoke. A spoke with multi-coloured bands represents various ranges of wind speed which occurred in the indicated direction of the spoke. The analysis of the simulation results is divided into two main parts: results analysis based on the selected modelling periods, and emissions trends for the pollutants CO, NOx, and CO2.

Fig. 6
figure 6

Terrain map of Salalah domain of study

Full size image
Fig. 7
figure 7

Salalah windrose for the selected simulation periods

Full size image

3.1 Modelling periods

3.1.1 Period 1 (May 26, 2004)

For May 26, 2004, wind blew from the southwest (SW) and the south-southwest (SSW), with gusts predominantly from the SSW (Fig. 7). For the majority of the day, the wind from the SSW had a speed ranging from 1.8 to 3.3 m/s. As for the wind blowing from the SW, the most frequent wind speed ranged from 3.3 to 5.4 m/s. Table 6 lists the day’s top five CO, NOx, and CO2 concentration levels with their corresponding allowable limits. For the three investigated pollutants, the maximum concentration occurred (− 0.5, 0.5 km) away from the defined origin at 05h00 LST. According to CO simulation results for this modelling period, the maximum concentration recorded was 22,542 µg/m3. This concentration level was within the criterion limits assigned by U.S. EPA standards. As for nitrogen oxides (NOx), the top five recorded concentration levels exceeded the limits set by the EPA. At 05h00 LST, two of the top recorded concentrations occurred simultaneously but in different locations close to the study area’s center. On the other hand, allowable CO2 concentrations set by the Ontario Ministry of the Environment and Climate Change (MOE) emission standards for a 0.5-h average period were exceeded by simulation results.

Table 6 Line source allowable and top 5 average CO, NOx, and CO2 concentrations simulated on May 26, 2004; September 27, 2005; August 3, 2006; August 22, 2008 from 00:00 to 23:00
Full size table

3.1.2 Period 2 (September 27, 2005)

On September 27, 2005, wind blew from the west, east, SW, and SSW (Fig. 7).The dominant wind direction was from the SW, with a speed ranging from 5.4 to 8.5 m/s. The same speed was recorded for wind from the SSW direction, which displayed the second highest gust frequency. Table 6 illustrates the top five maximum concentration levels of CO, NOx, and CO2 along with their allowable concentrations. The maximum concentration occurred (−4.5, −0.5 km) from the origin at 03h00 LST. As for the remaining peak values, they all occurred at 04h00 LST in multiple locations within the study domain. The highest recorded CO concentration, 5533 µg/m3, was located (−4.5, −0.5 km) from the origin, and was far below the EPA-assigned value for CO concentration levels. As for NOx, the five recorded concentration levels exceeded the EPA’s recommended limit for 1-h NO2 average concentration standards. Similar to what was observed for the dispersion of NOx, CO2 concentrations exceeded the concentration limits set by Ontario’s MOE emission standards.

3.1.3 Period 3 (August 3, 2006)

From Fig. 7, it can be noticed that for the modelling day August 3, 2006, wind blew from the SW and southeast (SE), but it gusted predominantly from the south-southeast (SSE). The wind speed ranged from 1.8 to 3.3 m/s. As for the wind from the SE, which was the second most frequent direction from which wind blew during the specified period, the most frequent wind speed ranged from 0.5 to 1.8 m/s. Table 6 displays the allowable and top five average CO, NOx, and CO2 concentrations simulated on August 3, 2006 and shows that the maximum concentrations of CO were within the prescribed limits. On the other hand, both NOx, and CO2 clearly exceeded allowable concentrations. The maximum concentrations were recorded (−1.5, −0.5 km) from the origin at 20h00 LST. For NOx, the two highest concentrations occurred at the same time but in different locations near the domain origin. As for CO2, the maximum concentration level was 305,661.4 µg/m3, which is far beyond the criterion limit of 63,000 µg/m3.

3.1.4 Period 4 (August 22, 2008)

The wind rose associated with the modelling day August 22, 2008 shows that wind blew in directions ranging from the SW to the SE (Fig. 7). Specifically, most of the gusts blew from the SSW and SSE. The wind predominantly blew from the SSW direction with speeds ranging from 3.3 to 5.4 m/s for the majority of the day. The wind from the SSE direction blew second most frequently, with the most frequent wind speed ranging from 1.8 to 3.3 m/s. Table 6 lists the maximum five concentration levels for CO, NOx, and CO2 against their respective criterion limits on August 22, 2008. For the three pollutants, the maximum concentrations were recorded from 19h00 to 22h00 LST. The highest concentration occurred at 22h00 LST, and the second and fourth highest concentrations occurred at 21h00 LST. The third highest concentration occurred at 19h00 LST, and the fifth highest concentration occurred at 20h00 LST. It is worth mentioning that four of the maximum concentrations occurred at the same coordinate (3.5, 0.5 km), which was close to the origin of the study domain. Although the maximum concentration of CO reached 19,217 µg/m3, it was still within the criterion limits. In contrast, the NOx, and CO2 maximum concentration levels clearly exceeded the allowable concentration standards.

3.1.5 Period 5 (June 21, 2009)

The wind rose associated with the modelling day June 21, 2009 demonstrates that wind only blew in directions ranging from the west to the south (Fig. 7). Specifically, most of the wind blew from the SW to the SSW. For the majority of the day, the wind came from the SSW with a speed ranging from 5.4 to 8.5 m/s. The second most frequent direction from which the wind came, the SW, blew from 1.8 to 3.3 m/s, while the second most noticeable wind speed ranged from 5.4 to 8.5 m/s. Table 7 shows the top five maximum concentrations of CO, NOx, and CO2 for a 1-h average for June 21, 2009. The simulation results demonstrate the fact that all maximum concentrations occurred at 05h00 LST. The highest 1-h average concentration occurred 1.5 km east and 0.5 km north of the center of the street. The second highest concentration was located 3.5 km west and 0.5 km north of the center of the street. The third highest concentration occurred 2.5 km west and 0.5 km north of the street. The fourth highest concentration occurred at a location 0.5 km east and north of the defined street. Finally, the fifth highest concentration was 1.5 km west and north of the selected domain origin. Based on CO simulation results, the maximum concentrations were within the criterion limits for all levels. As for NOx, the maximum concentration for a 1-h period reached within the simulated model was found to be approximately 2644.4 µg/m3. According to U.S. EPA standards, the top five concentrations exceeded the allowable concentrations for NOx. As for CO2, all top five concentrations were significantly greater than 63,000 µg/m3, which is the allowable concentration limit assigned by the Ontario MOE’s emission standards.

Table 7 Line source allowable and top 5 average CO, NOx, and CO2 concentrations simulated on June 21, 2009; September 5, 2012; November 7, 2013; January 22, 2014 from 00:00 to 23:00
Full size table

3.1.6 Period 6 (September 5, 2012)

The wind rose associated with the modelling day of September 5, 2012 (Fig. 7), shows that the wind blew from the NW, the south and the SW. The wind predominantly came from the SSW with a speed that varied between 1.8–3.3 and 3.3–5.4 m/s for the majority of the day. As for the wind from the SW, which was the second most frequent direction from which the wind blew, the most frequent wind speed ranged from 1.8 to 3.3 m/s. Table 7 lists the allowable concentration according to Ontario’s MOE and the top five average concentrations of CO, NOx, and CO2 for September 5, 2012. According to the simulation results, the maximum concentration occurred 3.5 km west and 0.5 km north of the defined origin at 05h00 LST. The first and the third highest 1-h average concentrations all occurred at the same location, 2 h apart from each other. Similar to the previous simulation analysis, CO concentrations were below the EPA’s threshold concentration levels. As for NOx, the maximum concentration was five times greater than the allowable concentration set by EPA standards. CO2 concentrations, on the other hand, were far beyond the criterion limits. The maximum value was 168,093.1 µg/m3 compared to 63,000 µg/m3 as an allowable a 0.5-h average period concentration.

3.1.7 Period 7 (November 7, 2013)

Based on Fig. 7, it can be observed that for the modelling period November 7, 2013, wind blew from directions ranging from the south to the NNW, with the three most predominant gust directions being the SSE, SE, and the north. The wind from the SSE and SE directions had a speed ranging from 3.3 to 5.4 m/s for the majority of the day. As for the wind from the north, the most frequent wind speeds were between 1.8 and 3.3 m/s. Table 7 summarizes the top five concentration levels of CO, NOx, and CO2 for November 7, 2013, with their allowed concentrations. The maximum concentration of the three pollutants occurred at 19h00 LST, 2.5 km west and 0.5 km south of the domain origin. CO concentrations were within the limits set by the U.S. EPA. Similarly, CO2 levels were also below the allowed 0.5-h average concentrations. On the other hand, nitrogen oxides simulation results showed a violation in concentration levels. Occurring at 19h00 LST, the maximum concentration was 365.01 µg/m3, which is higher than the criterion limit.

3.1.8 Period 8 (January 22, 2014)

The wind rose on January 22, 2014 (Fig. 7) shows that wind blew in directions ranging from the WNW to the SE, although it blew predominantly from the north with a speed between 5.4 and 8.5 m/s for the majority of the day. For the remainder of the day, the wind speed varied between the ranges (3.3–5.4 m/s) and (8.5–10.8 m/s). Table 7 lists the top five CO, NOx, and CO2 concentration levels on this particular day with their corresponding allowable limits. For the three investigated pollutants, the maximum concentration occurred at (−2.5, −0.5 km) away from the defined origin at 03h00 LST. According to CO simulation results for this modelling period, the maximum concentration recorded was approximately 20,291 µg/m3, which is within the criterion limits set by U.S. EPA. On the other hand, the maximum concentration of NOx for a 1-h period reached within the simulated model was found to be 1699.1 µg/m3. According to the U.S. EPA’s standards, the top five concentrations exceeded the allowable limits for NOx. For CO2, the maximum concentration level was approximately 288,128.4 µg/m3, which is beyond the criterion limit for a 0.5-h period (63,000 µg/m3).

3.2 Emission trends

Table 8 lists the maximum concentrations of CO, NOx, and CO2 with their corresponding criterion limits for the simulation periods associated with this study. For more visualization regarding the trends of these line source emitted pollutants for the past 10 years, Fig. 8 summarizes the comparison between allowed and simulated concentration trends of CO, NO2, and CO2.

Table 8 Maximum highest concentrations of CO, NOx, and CO2 with their corresponding criterion limits
Full size table
Fig. 8
figure 8

Comparison between allowed and simulated concentration trends of CO, NOx, and CO2

Full size image

3.2.1 Carbon monoxide (CO)

Starting with the first modelling period, May 26, 2004, the maximum CO concentration reached 22,542 µg/m3. This concentration significantly decreased the next modelling day, September 27, 2005, to 5533 µg/m3. Figure 9 represents the plume trajectories of the highest 1-h CO concentration levels, which dropped due to wind blowing from the land toward the sea. The simulated wind vectors showed the winds shifting to the southwest. Comparing this value to the concentration level simulated on August 3, 2006, a tremendous increase of around 21,590 µg/m3 can be observed. Although the number of vehicles increased through the 10-year study, a slight decrease occurred in the concentration of CO on August 22, 2008. For the next year, the concentration of carbon monoxide increased to 31,581 µg/m3. Based on the visualized trend, this value is the maximum concentration level which occurred in the past 10 years. From September 5, 2012, to January 22, 2014, the trend was that of fluctuating in the concentration levels, which started at approximately 11,834 µg/m3, decreased to 4359.1 µg/m3, and finally settled with a value of 20,291 µg/m3 in 2014. All these fluctuations in concentration levels were due to the wind direction associated with each modelling day. Despite the fact that the concentration levels listed for CO were high in general, these concentration levels were within the criterion limits assigned by the U.S. EPA.

Fig. 9
figure 9

Plume trajectories of the highest 1-h CO concentration levels for the selected simulation periods

Full size image

3.2.2 Nitrogen dioxide (NO2)

On May 26, 2004, the maximum concentration of NO2 reached 1887.5 µg/m3. This concentration level had decreased four-fold by the next modelling day, September 27, 2005, to 463.29 µg/m3. Figure 10 represents the plume trajectories of the highest 1-h NOx concentrations. A notable drop occurred due to the wind blowing from the land toward the sea, and simulated wind vectors shifted to the southwest. Comparing this value to the concentration level simulated on August 3, 2006, it can be observed that the concentration of NO2 significantly increased, to 1802.5 µg/m3. Because the number of vehicles increased in Salalah over the study period, a slight decrease occurred in the concentration of NO2 on August 22, 2008. However, by the next modelling day, nitrogen dioxide concentration levels had increased to 2644.4 µg/m3. Based on that trend, this value is the maximum concentration level that occurred within the past 10 years. Between September 5, 2012, and January 22, 2014, the trend showed a fluctuation in concentration levels, with NO2 concentrations of 991.33 µg/m3 in 2012, which decreased to 365.01 µg/m3 in 2013, and finally settled with a value of 1699.1 µg/m3 in 2014. Further examination of Table 8 and Fig. 8 illustrates that all the simulated 1-h average NOx concentrations were found to significantly exceed their respective 1-h criterion limits.

Fig. 10
figure 10

Plume trajectories of the highest 1-h NOx concentration levels for the selected simulation periods

Full size image

3.2.3 Carbon dioxide (CO2)

On the first modelling day, May 26, 2004, the simulated maximum concentration of CO2was found to be 320,061.7 µg/m3. On the next modelling day, September 27, 2005, the concentration of CO2 significantly decreased to 78,560.8 µg/m3 (Fig. 11). The reason behind this significant drop is related to the breeze that comes toward the sea from the land. Comparing this result with the results of the simulation period on August 3, 2006, it is clear that CO2 concentration levels significantly increased to 305,661.4 µg/m3. A slight decrease occurred in the concentration of CO2 on August 22, 2008, compared to the increase in the number of vehicles associated with the modelling period. On the next modelling day, CO2 concentration levels increased to 448,402.2 µg/m3, which were the maximum concentration levels observed within the 10-year study. Between September 5, 2012, and January 22, 2014, there was a fluctuation in concentration levels. In 2012, the concentration of 168,093.1 µg/m3 decreased to 61,894.8 µg/m3 in 2013, and finally settled with a value of 288,128.4 µg/m3 in 2014. The simulation results related to CO2 in Table 8 and Fig. 8 demonstrate that all the concentration levels associated with the previously mentioned simulation periods significantly exceeded their respective 0.5-h criterion limits except for the simulation results related to November 7, 2013, where the concentrations of CO2 were below the allowable concentrations assigned by the Ontario MOE.

Fig. 11
figure 11

Plume trajectories of the highest 1-h CO2 concentration levels for the selected simulation periods

Full size image

4 Conclusions

Employing software models in assessing the effect of the geophysical and meteorological conditions on the dispersion of line source emissions in the environment is consequential. Based on that, the influence of three vehicle pollutants, CO, NOx, and CO2 on Salalah’s environment was investigated for a 10-year period. The analysis of the simulation results was divided into two main sections: a study of each of the selected modelling days and an assessment of the trends in the concentration levels of CO, NOx, and CO2 within a 10-year period. In the first part of the study, the meteorological conditions for each of the selected modelling periods were different and, based on this fact, the variation in concentration levels, coordinates, and timing where maximum concentrations were recorded were noticeable. In the second part of the analysis, the concentration trends of CO, NOx, and CO2 for the 10-year study period were investigated. From the results of this study, June 21, 2009, was determined to be the day with the highest 1-h average of CO and NOx concentrations and 0.5-h average CO2 concentration within the past 10 years. November 7, 2013, was determined to have had the lowest 1-h average of CO and NOx concentrations and 0.5-h average CO2 concentration. On June 21, 2009, the highest 1-h average CO and NOx concentrations and 0.5-h average CO2 concentration were determined to, respectively, be 31,581, 2644.4, and 4488,402.2 µg/m3. These NOx and CO2 concentrations significantly exceeded the EPA’s NOx criterion limit of 188.2 µg/m3 and the Ontario MOE’s 0.5-h criterion of 63,000 µg/m3. Based on the simulation results associated with the case study, almost all of the maximum concentrations originated close to the center of the domain of study. Because Salalah is a coastal city, the wind gusts moving toward the land from the sea significantly affect the dispersion of CO, NOx, and CO2. Based on that geographical feature, the three pollutants accumulated close to the center of the study street.

The trend of vehicle-emitted pollutants indicates that their concentrations may even exceed the concentrations associated with this study and disperse close to the defined street. During the day, in places where concentrations of CO, NOx, and CO2 are extremely high, citizens of Salalah who are working or people who are engaging in tourist activities close to the street might experience serious adverse CO, NOx, and CO2 effects. Due to the side effects that might occur from the emission of high concentrations of CO, NOx, and CO2, it is recommended to construct flyovers instead of roundabouts in order to decrease traffic jams.