PM2.5 and PM10 Emission Inventory from Road Dust in Abidjan City ()
1. Introduction
Air quality degradation in urban areas has become a major global concern, particularly due to its profound impacts on human health and the environment. Among the most critical air pollutants are suspended fine particles, commonly known as particulate matter (PM). These particles are classified by their size, with PM2.5 and PM10 referring to particles with a diameter less than 2.5 and 10 micrometers, respectively. Their small size allows them to penetrate deep into the respiratory system and bloodstream, leading to a range of severe health issues, including respiratory and cardiovascular diseases [1] [2]. These air pollutants are diverse sources, often distinguished between their natural and anthropogenic origins. The anthropogenic sources of atmospheric pollution are often closely related to greenhouse gases.
Air quality modeling and forecasting meet various challenges, including those related to the availability of accurate high-resolution (spatial and temporal) emission inventory data (used as input data) as well as in-situ measurements (for model calibration and validation) [3] [4]. In their review of real-time air quality forecasting, [4] identified several inaccuracies and their possible cause, which include factors related to meteorology, boundary conditions, emissions and model processes. Thus, these factors contribute to remarkable differences between simulation and forecast results and measured values. Therefore, accurate and high-resolution (temporal and spatial) emission inventories are essential to improve model performance. However, Africa and particularly West Africa suffer from a lack of accurate and high-resolution emission inventories. The few regional inventories that have been published for Africa for anthropogenic emissions, such as [5]-[7], have addressed combustion sources and some at the city level, such as [8]. These inventories cover gases (CO, NOx, SO2 and NMVOC) and aerosols (black carbon, organic carbon, PM2.5 and PM10) at regional level (10 km × 10 km horizontal resolution) and at city level (1 km × 1 km horizontal resolution) and have been developed taking into account certain sources, namely domestic fires, traffic, waste combustion, industries and thermal power plants [5], but not yet road dust emissions.
De Longueville et al. [9] found that out of 231 articles published over the last decade on the direct links between desert dust and human health, only one of these studies concerned West Africa, even though this region is very close to the Sahara, the largest contributor to global dust levels. Furthermore, Semerjian et al. [10] showed that very few studies have been conducted on road dust in Africa, given the lack of publications on the subject from these regions. Yet West Africa is one of the areas where morbidity rates are significantly higher than in other parts of the world, and where respiratory infections alone account for more than 20% of infant mortality causes [9] [11]-[13].
Haynes et al. [14] showed that very few studies exist on the chemical composition of road dust in Africa. Furthermore, these few studies [15] [16] found high levels of heavy metals such as copper, lead, zinc, and cadmium in roadside dust, which is resuspended by vehicles and therefore likely to increase the concentrations of these pollutants in the atmosphere.
Abidjan is undergoing rapid urbanization and population growth, resulting in a significant increase in road traffic. This situation is intensified by the presence of unpaved roads, which are a major source of dust. The dispersion of this dust in the ambient air increases PM2.5 and PM10 concentrations, contributing to the deterioration of the city’s air quality [17].
The development of a PM2.5 and PM10 emission inventory, specific to road dust raising, is essential to assess the impact of this source on Abidjan’s air pollution. This inventory will be spatialized at a horizontal resolution of 1 km × 1 km, and at an annual temporal resolution. It will enhance existing inventories in Abidjan and improve the PM mapping derived from modelling, providing a valuable tool for local authorities in managing and reducing pollution levels. Section 2 describes the methodology, the data, the tools, and techniques used for estimating and spatializing emissions. Finally, the third section presents the results obtained, followed by a discussion.
2. Methodology
2.1. Study Area
Abidjan, the economic capital of Côte d’Ivoire, is a major metropolis in West Africa, home to some 6.321 million inhabitants according to the latest general population and housing census in Côte d’Ivoire [18], and covers an area of 2119 km2. Located in the south-eastern part of the country, it is the main economic and urban center, with a significant concentration of industrial, commercial and service activities. The city is characterized by rapid urbanization and an expanding road infrastructure, essential for linking its various districts.
Abidjan’s road network is vast and varied, comprising freeways, paved main roads and secondary roads, some of which are unpaved. The latter are particularly prone to dust generation, especially during the dry season due to traffic, in areas with high vehicle density. The city’s geographical configuration and rapid development make it a relevant study site for the analysis of road dust emissions. Figure 1 shows a map of primary, secondary and tertiary paved roads in Abidjan district.
Figure 1. Map of primary, secondary and tertiary roads in Abidjan district.
2.2. Determination of the PMs Road Dust Emissions in Abidjan
The methodology used to derive PMs Road Dust (RD) emissions emission was based on the AP-42 method [19] that is an empirical model developed by the US Environmental Protection Agency (EPA). RD PMs emissions are then calculated using Equation (1):
(1)
where Et was a total emission of compound; Vehl a number of vehicles per type; Dl the distance traveled in a year per vehicle type, EFl,k emission factor of compound per road type. The emissions factor is calculated using Equation (2) and Equation (3) respectively for paved and unpaved road:
(2)
(3)
where EF (g/VKT) represents the particle emission factor, k (g/VKT) is the particle size multiplier (paved road: k = 0.15 for PM2.5 and k = 0.62 for PM10, unpaved road: k = 107.12 for PM2.5 and k = 732.94 for PM10), sL (g/m2) is the average silt load, W (tons) is the average weight of vehicles on the roads, P: number of rainy days with precipitation of at least 0.254 mm, S (%) is the silt content of the surface material and M (%) is the moisture content of the surface material.
Thus, road dust emission factor is calculated on the basis of several variables, such as traffic, nature of the road soil, weather, average vehicle weight and annual mileage.
Road and traffic data
Statistics on road characteristics (paved and unpaved) are essential for estimating PM emissions on these two categories of road. In this study we used the Greater Abidjan urban transport master plan report of the Greater Abidjan urban planning master plan development project (SDUGA) data
(https://openjicareport.jica.go.jp/pdf/12230645_01.pdf). The Abidjan district has a total of 1772.2 km of roads, 854.6 km of which are paved (48%) and 917.5 km unpaved (52%).
To date, there are no official statistics on the number of vehicles on the road in Côte d’Ivoire. We therefore needed to collect data on the number of vehicles, their average weight and the annual mileage covered by these vehicles for Abidjan. The fleet data were estimated by the study of [8], whose methodology is based on data from the Ministry of Transport via its technical structures, notably the Guichet Unique du Transport Terrestrial de Côte d’Ivoire (which provides the first registrations) and SICTA (the technical structure in charge of technical inspections to estimate the annual mileage traveled, fleet exits, etc.).
The average distance traveled per vehicle category per year is determined using roadworthiness test data, as in the study of [8]. These average distances are available for the years 2012 to 2019, and for the missing years in the study period, the annual average value per category was used. Thus, the average annual distances traveled by category from 2010 to 2019 are obtained.
Average weight values are taken from the literature. For each vehicle category, values are given in interval form, but average values are used for each category. We have used 1.225 tons, 1.75 tons, 20 tons, 0.2 tons and 16 tons respectively for cars, vans, trucks, motorcycles and buses.
Annual mileage traveled by vehicle category
The average annual number of kilometers traveled (VKT) per vehicle category is taken from the work of [8]. Due to the lack of available data on vehicle counts by road type (paved and unpaved), the distribution of VKT by road type (paved and unpaved) was obtained based on the assumption of an even distribution of traffic volume across these two road types. We therefore multiplied the number of VKT per vehicle category by the proportion of each type of road (48% for paved roads and 52% for unpaved roads) to obtain these VKT for paved and unpaved roads in Abidjan, respectively.
Meteorological data and other parameter
The meteorological data from the Ogimet website (http://www.ogimet.com/) providing synoptic information for the city of Abidjan (station ID: 65578) from 2010 to 2019 were used for this work. For this dataset the total precipitation for the 24-hour period from 6:00 a.m. to 6:00 a.m. is reported at 6:00 a.m. with reference code 4. We have carried over this value when it is neither 0 nor NaN; otherwise, we check the precipitation values from the previous 24 hours to see if there was an error in carrying over the cumulative total at 6:00 a.m. For this work, a day is accounted for rainy day if at least 0.254 mm was recorded. This threshold of 0.254 mm represents the minimum amount of precipitation needed for mitigating suspended particulate emissions on an hourly basis according to [19]. We then applied this threshold and retained only the days on which precipitation exceeded it.
Other useful parameters for calculating road dust emissions are soil silt content (S), surface moisture (M) and average silt load (sL). These parameters are all taken from the literature. Due to the lack of data on silt content for all of Abidjan’s municipalities, and given that most of the city’s roads (paved or unpaved) are made of clayey sand, given its availability (Kouassi et al.), we calculated an average based on the limited data available in the literature for municipalities and assumed that these average values were representative. Indeed, the silt content of the soil was estimated at 5.6% based on the values S = 7.6% ± 3.9% and S = 3.6% for the municipalities of Marcory and Cocody, respectively, located in the Abidjan district, as reported by [20]. Furthermore, the surface moisture content in Abidjan is taken from the study in [20] and estimated at M = 4%, because during the dry season, unpaved road surfaces experience extreme moisture loss due to high evaporation rates, leading to significant dust emissions and surface degradation (Kamara et al. 2025). Finally, the average silt load used in this study is taken from the study by [21] and is estimated at sL = 0.531 g/m2.
2.3. Spatial Distribution of RD PMs Emissions
RD PMs emissions, obtained for each type of road (paved and unpaved) have been distributed on map, using a road density map for the city of Abidjan, based on the shapefiles (spatial distribution) of the roads collected from https://extract.bbbike.org/ (last access: 19 July 2025) combined to road occupancy per type of vehicle. Using a methodology similar to that described in [8], emissions were spatially distributed across different road types. The gridding steps are described as follows: 1) extraction of primary, secondary, and tertiary roads from Shapefiles, 2) calculation of road density and its normalization, 3) application of vehicle occupancy weights by road category (primary, secondary, and tertiary) based on the work in [22] and finally 4) allocation of road traffic emissions by road type (paved and unpaved).
3. Results
In this section, we present the results of the inventory covering the period from 2010 to 2019. Emission factors are analyzed by vehicle category, followed by annual changes in emissions over the period 2010-2019. The number of vehicles by category and the average annual distances traveled by category from 2010 to 2019 is shown in Table 1.
Table 1. Vehicles number and the average mileage in Abidjan by category from 2010 to 2019.
Parameters |
Categories |
2010 |
2011 |
2012 |
2013 |
2014 |
2015 |
2016 |
2017 |
2018 |
2019 |
Vehicle number |
Private car |
283,740 |
294,210 |
301,288 |
320,151 |
336,034 |
354,584 |
377,824 |
405,158 |
411,469 |
428,656 |
Taxis |
50,072 |
51,919 |
53,169 |
56,497 |
59,300 |
62,574 |
66,675 |
71,497 |
72,612 |
75,645 |
Van |
55,495 |
58,079 |
61,050 |
66,925 |
71,564 |
76,555 |
82,333 |
88,043 |
91,448 |
96,213 |
Truck |
41,570 |
43,325 |
45,385 |
48,698 |
51,350 |
54,748 |
59,392 |
63,930 |
65,385 |
68,571 |
Two wheels vehicle |
43,455 |
44,863 |
46,963 |
50,929 |
54,762 |
59,249 |
64,539 |
77,654 |
75,577 |
80,082 |
Bus |
13,637 |
13,949 |
14,015 |
14,452 |
14,603 |
14,833 |
15,149 |
15,597 |
15,725 |
15,991 |
Annual mileage |
Private car |
14,650.1 |
14,650.1 |
14,367.4 |
14,999.9 |
15,073.2 |
15,103.9 |
13,705.9 |
14,650.1 |
14,650.1 |
14,650.1 |
Taxis |
56,409.3 |
56,409.3 |
61,622.3 |
64,944.8 |
54,457.3 |
58,240.4 |
42,781.6 |
56,409.3 |
56,409.3 |
56,409.3 |
Van |
19,858.6 |
19,858.6 |
20,969.8 |
20,726.8 |
20,206.6 |
20,029.3 |
17,360.6 |
19,858.6 |
19,858.6 |
19,858.6 |
Truck |
84,525.6 |
84,525.6 |
133,271 |
70,326.3 |
107,568 |
69,670.2 |
41,792.5 |
84,525.6 |
84,525.6 |
84,525.6 |
Two wheels vehicle |
14,650.1 |
14,650.1 |
14,367.4 |
14,999.9 |
15,073.2 |
15,103.9 |
13,705.9 |
14,650.1 |
14,650.1 |
14,650.1 |
Bus |
81,394.7 |
81,394.7 |
75,977.3 |
138,262 |
72,472.1 |
65,856.3 |
54,405.6 |
81,394.7 |
81,394.7 |
81,394.7 |
3.1. RD PM2.5 and PM10 Emissions Factors in Abidjan
The average emission factors for PM2.5 and PM10 from paved and unpaved roads for the period 2010-2019 are presented in Table 2. Same emission factors are used for private cars and taxis since they belong to the same vehicle category and the average weight of vehicle were used for emission calculation. Moreover, emission factors for particulate matter are classified by vehicle respective weights, and by year as rain height parameter is taken into account in Equation (2) and Equation (4). Highest emission factor is found for trucks, followed respectively by buses, vans, cars and motorcycles (TW) which exhibit the lowest value. The emission factor is strongly influenced by vehicle weight since heavier a vehicle is, the more dust it raises. This is explained by the pressure exerted by the vehicle on the road surface. On paved roads, PM2.5 EF values range from 0.01 to 1.61 g/VKT, PM10 EF values range from 4.8 to 30.4 g/VKT, while on unpaved roads, PM2.5 EF values range from 4.8 to 30.4 g/VKT, and PM10 EF values range from 32.9 to 207.9 g/VKT.
Table 2. Particulate matter (PM2.5 and PM10) emissions factors from paved and unpaved roads in Abidjan.
Categories |
PM2.5 EF (g/VKT) |
PM10 EF (g/VKT) |
Paved road |
Unpaved road |
Paved road |
Unpaved road |
Private car and taxi |
0.09 |
9.9 |
9.9 |
68.0 |
Van |
0.13 |
11.5 |
11.5 |
78.5 |
Truck |
1.61 |
30.4 |
30.4 |
207.9 |
Two wheels vehicle |
0.01 |
4.8 |
4.8 |
32.9 |
Bus |
1.28 |
27.8 |
27.8 |
190.1 |
Unpaved roads exhibit much higher (from 8 to 400 time) EF than paved road (Table 2). This may be explained by significant differences in surface conditions and the amount of dust available. Paved roads are characterized by a smooth, solid surface, with less available dust ready to be released into the atmosphere by traveling/moving vehicles. Therefore, these paved roads generate limited/less amount of fine particles to be dispersed in the air. However, unpaved roads have a crumbly surface, with a large amount of available dust on the surface due to their rapid deterioration. This facilitates the particles mobilization and released by passing vehicles. The high PM EF values found for the unpaved roads in this study, highlight the major impact of road infrastructure on particulate emissions. Therefore, the outlying districts of Abidjan, where the most of unpaved roads are located, will tend to emit more PM due to their traveling/moving vehicles (light or heavy).
3.2. Annual Vehicle Kilometers Traveled and Total Rainy Day per Year
Table 1 shows the average annual mileage used to calculate vehicle kilometers traveled (VKT) on paved and unpaved roads in Abidjan. The VKT of unpaved roads is slightly higher than that of paved roads. This could be due to factors linked to urban growth, road infrastructure structure and traffic patterns. Indeed, the city of Abidjan has undergone rapid expansion, particularly in outlying areas where unpaved roads are more numerous. These developing neighborhoods are home to a significant proportion of the population. Residents of these areas have to use unpaved roads to travel to paved roads, which increases the number of kilometers traveled on these roads. In addition, the difficulty of circulating on paved roads, often due to traffic jams and poor road conditions, can lead drivers to prefer unpaved roads as an alternative, especially cab drivers and transporters seeking to avoid traffic jams. These peripheral areas also concentrate a large proportion of informal economic activities. This includes markets and small-scale industries, leading to frequent use of unpaved roads by heavy vehicles such as trucks and minibuses for goods and passenger transport. In short, rapid urbanization, the lack of paved roads in outlying areas and the importance of economic activities in unpaved areas explain why the VKT of unpaved roads in Abidjan can be higher than that of paved roads.
In terms of rainy days per year, the total number of rainy days per year with at least 0.254 mm in Abidjan ranged from 72 in 2013 to 172 in 2017.
3.3. PM2.5 and PM10 Emissions from RD in Abidjan
Figure 2 shows the evolution of PM2.5 and PM10 emissions from paved (Figures 2(a)-(c)) and unpaved (Figures 2(b)-(d)) roads, respectively, over the period 2010 to 2019. The highest emissions of PM2.5 and PM10 are found on unpaved roads. Among the different categories of vehicles, trucks have the highest emissions over the entire study period. For example, for PM2.5, trucks emitted about 56.5 kilotons (48% of the total) in 2011 and 107.8 kilotons (51% of the total) in 2019 on unpaved road. Their emissions on the paved road reached 2.6 kilotons (72% of total) in 2010 and 4.6 kilotons (76% of total) in 2019. Moreover, the truck category PM2.5 emissions were lower in 2013 and 2016, while an increasing trend was observed from 2017 onward. This increasing trend in PM2.5 emission by truck category could be attributed to an increase in trade and also infrastructure construction. The decrease in PM2.5 emissions observed in 2013 and 2016 may be related to the low VKT in these years. All categories, with the exception of trucks, show higher PM emissions in 2013, which may be related to the low number of rainy days (days with at least 0.254 mm) recorded that year. PM10 emissions from paved and unpaved roads in Abidjan for the period 2010 to 2019 show similar trends to those for PM2.5.
In terms of main contributors, trucks are followed by private cars and taxi, which are frequently used on the city’s inland roads for public and private transport. Variations in emissions are strongly linked to VKT and EF, which in turn depend on the parameters such as vehicle weight, road type and precipitation. These results highlight the importance of road infrastructure as well as the contribution of vehicle type and their impact on road dust emissions. More paved roads, in good conditions, may contribute in limiting PM2.5 and PM10 emissions form road. Unpaved roads, without appropriate treatment, emit more PM2.5 and PM10, contributing to air quality degradation in surrounding areas.
Gridded PM2.5 and PM10 emissions from road dust in Abidjan for the year 2019 are shown in Figure 3 (a and b respectively) combining paved and unpaved road. PM2.5 and PM10 from road dust are estimated to be 217.2 kilotons and 1469.4 kilotons respectively. The vast majority of emissions are found for unpaved roads, which accounted approximately for 97.2% of the total PM2.5 and 98.3% for the total PM2.5 emitted. The present road dust emissions inventory is complementing an existing anthropogenic emission inventory for Abidjan focused on combustion sources [8].
Figure 2. PM2.5 emissions from (a) paved and (b) unpaved roads dust and PM10 emissions from (c) paved and (d) unpaved in Abidjan from 2010 to 2019.
Figure 3. Spatial distribution of (a) PM2.5 and (b) PM10 emissions from roads dust in Abidjan for year 2019.
3.4. Sensitivity Analysis and Emissions Uncertainties
A sensitivity analysis was conducted to examine how PM2.5 emission factors from unpaved roads vary around the chosen average value for silt content. The results of the sensitivity test show that, using the two extreme values for silt content, the PM2.5 emission factors for unpaved roads will fluctuate between 7 - 12.7 g/VKT, 8 - 11.5 g/VKT, and 21.3 - 38.8 g/VKT, 3.3 - 6.2 g/VKT, and 19.5 - 35.5 g/VKT, respectively, for passenger cars, vans, trucks, motorcycles, and buses. This gives us a relative variation in the PM2.5 emission factor compared to the average value, ranging from −29.8% to +27.7%.
A Monte Carlo simulation was conducted to quantify the uncertainty associated with emission estimates [26]. In this study, we assume that when the coefficient of variation (CV) is less than 30%, the distribution is normal, and that it is log-normal and asymmetric when it is greater than 30% [5]. The Monte Carlo method was then used to propagate the uncertainties related to silt content (S: 5.6 ± 2%), surface moisture (M: 4 ± 1%), and vehicle weight (±1) to obtain the uncertainty in the PM2.5 emission factor and their emissions by vehicle type for the year 2019. The simulation involved repeating the test 100,000 times to obtain the mean value and estimated standard deviations with a 95% confidence interval. The results reveal significant variability in emission factors (±30.7%), mainly due to uncertainties regarding silt content (S), moisture (M), and vehicle weight (W). The relative uncertainty in emissions remains consistent across vehicle categories (±45.1% to ±45.4%), as emissions vary linearly with the emission factor (EF). The sensitivity analysis shows that silt content dominates the uncertainty in emissions (S: 0.86), followed by vehicle weight (W: 0.07), while moisture content (M: 0.06) has a mitigating but weaker contribution.
4. Discussions
The PM2.5 and PM10 emission factors measured in our study for road dust from paved roads in Abidjan are much higher than those reported in the literature as shown in Table 3. This may be related to local factors favoring presence of more dust material on paved roads. For example, the proximity of uncovered surfaces to paved road may facilitate the transport of sand and clay on these paved roads which are then mobilsed and release into the ambient air by vehicles when traveling/moving. Moreover, [23] highlighted that paved roads experiencing construction work were emitting 5 times more PM10 than the paved roads with no construction work in Milan (Italy).
PM10 emissions are significantly higher on unpaved roads than on paved roads. This is mainly due to the fact that unpaved roads are a major source of resuspension of dust particles, particularly PM10. This observation is consistent with the study by [27] [28], who showed that dust raised by passing vehicles on unpaved surfaces is a predominant source of PM10 in both urban and rural environments. Heavy vehicles, such as trucks, play a key role in increasing PM10 emissions, particularly on unpaved roads. This finding is supported by [29], who demonstrated that heavy vehicles can significantly increase levels of suspended dust particles on unpaved roads, leading to elevated PM10 concentrations. Unpaved roads emit much more fine particulate matter due to the nature of their more friable surface, which accumulates more dust and degrades more quickly when disturbed by vehicles [28] [29].
Table 3. Comparison of PM10 emission factors obtained in several cities around the world using the USEPA method.
Road type |
Cities |
PM10 EF (g·veh−1·km−1) |
Reference |
Paved roads |
Abidjan
(Côte d’Ivoire) |
0.06 - 6.67 |
This work |
Paved roads (mean) |
Milan (Italie) |
0.013 - 0.032 |
Amato et al. [23] |
Roads with
construction work |
Milan (Italie) |
0.066 - 0.10 |
Amato et al. [23] |
Paved roads |
Braga (Portugal) |
0.033 |
Alves et al. [24] |
Unpaved roads |
Abidjan
(Côte d’Ivoire) |
33.22 - 209.62 |
This work |
Unpaved roads |
Muscatine County (US) |
444 - 795 |
Kacer et al. [25] |
PM2.5 is generally less abundant on paved roads, as PM2.5 often originates from the disintegration of PM10 and other combustion sources, rather than from the simple resuspension of road dust as shown by [30]. Therefore, results obtained in Abidjan, where road dust, especially on unpaved surfaces, is the main source of PM10 emissions, are in line with those observed in other cities (Table 3). A study by [31] in Nairobi showed that unpaved roads contributed significantly to high PM10 levels, while PM2.5 was mainly associated with exhaust emissions. Furthermore, emission levels of carbonaceous particles from traffic in Abidjan in 2019 with 2198.8 kilotons and 2062.6 kilotons respectively for BC and OC [8], which are the majority constituents of PM2.5 from fuel combustion [32]. For the same year 2019, we obtain PM2.5 emissions from road dust estimated at 217.2 kilotons, i.e., around 1/20th of carbon compound emissions. This observation is supported by the study by [30], which showed that the majority of fine particles (PM2.5) on paved roads come mainly from tire wear and incomplete fuel combustion, while road dust contributes more to PM10.
5. Conclusions
This work enabled us to draw up an inventory of fine particle emissions, in particular PM2.5 and PM10, specific to Abidjan’s road dust. The methodology main stages consisted in collecting and analyzing road network data, characterizing the traffic fleet and determining road dust emission factors for particulate matter (PM2.5 and PM10) specific to road (paved and unpaved) and vehicle (truck, private, …) types. Moreover, traffic data (VKT) and emission factors, were used to estimate PM2.5 and PM10 emissions. Emissions for 2019 are 25.1 and 6.1 kilotons for PM10 and PM2.5 respectively on paved roads, and 1444.3 and 211.1 kilotons for PM10 and PM2.5 respectively on unpaved roads. On paved roads, average emission factor values are 60.8 - 6670.3 mg/VKT for PM10 and 14.7 - 1613.8 mg/VKT for PM2.5. On unpaved roads, they rise to 33.2 - 209.6 g/VKT for PM10 and 4.8 - 30.6 g/VKT for PM2.5. The highest VKT values were obtained on unpaved roads, underling that unpaved roads are responsible for the highest levels of particulate emissions, particularly from heavy vehicles such as trucks and coaches. Furthermore, the emission factors calculated in this study were found to be significantly higher than those currently available in the literature. This difference is mainly due to the high density of unpaved roads in Abidjan. The results of this work thus reveal the importance of improving road infrastructure to reduce fine particle emissions in densely populated urban areas.
The limitations of this study include the lack of data on vehicle counts on different types of roads (paved and unpaved) and the experimental determination of certain parameters used (S, sL, and M).
These elements could help reduce uncertainties in these emission estimates and provide a better understanding of local variability in PM2.5 and PM10 concentrations based on microclimatic characteristics and economic activities. This inventory will enrich existing inventories in Abidjan (spatial resolution of 1 km × 1 km) and improve PM mapping based on modeling, thus providing a valuable tool for the development of public policies aimed at reducing air pollution levels in cities in Côte d’Ivoire.
Data Availability
Data is provided within this link https://doi.org/10.1594/PANGAEA.990397.
Acknowledgements
The authors would like to express their sincere gratitude to the IRD’s Young Associated Team for Atmospheric Physicochemistry and Impacts (JEAI PATI) and the French Research Institute for Sustainable Development (IRD) for their financial support, which contributed to the publication of this work. The JEAI PATI receives funding from the IRD for the period 2023-2025.