Radiocarbon Concentration Measurements in Tree Leaves near SOCOCIM (Rufisque, Senegal), A Cement Factory

Radiocarbon content in biogenic samples is widely used to study the variation of atmospheric CO 2 due to anthropogenic activities. A total of 20 samples of several types of tree leaves, were analyzed for this study. Sampling was carried out at the end of the rainy season in 2017 from the surrounding of the SOCOCIM cement factory in Rufisque town. Rufisque is located on the pe-ninsula of Cape Verde, 25 km east of Dakar, where it is the «south gate» of the agglomeration. Reference samples of five different species were collected during the same period (2017) from a clean zone. The 14 C method was used for the determination of Δ 14 C values. The data show that the 14 C concentration in the studied sites was significantly lower than the clean area, due to the release of anthropogenic CO 2 . To estimate the Suess effect, the fossil fuel fraction was determined based on equations of mass balance for CO 2 concentration, stable isotopic composition of carbon, and 14 C concentration. The results show that selected locations are affected differently according to their distance from the factory and the wind direction.


Introduction
Radiocarbon ( 14 C) is produced in the atmosphere by a reaction of neutrons with atmospheric 14 N that produces 14 C, which is rapidly oxidized into CO 2 and then exchanges with different carbon reservoirs. Natural processes such as solar activ-ity, Earth's magnetic field, ocean circulation, and rates of exchange between carbon reservoirs all affect the 14 C content. Besides these natural variations, human activities also have an impact on atmospheric 14 C concentration. Two anthropogenic effects are recorded by atmospheric 14 C: first, the Suess effect, which is the addition of carbon dioxide by fossil fuel combustion; and secondly, the increase in 14 C concentration in the atmosphere because of atmospheric nuclear weapons testing [1]. Both anthropogenic factors occur on a global scale, however there are also local Suess effects due to local fossil-fuel sources. So, there are regional discrepancies that are related to the emission zone proximity as well as geographic location [2]. If we compare the global 14 C atmospheric record to regional or local signals, then it is possible to relate this to regional and local changes in human activities such as industrial activity, traffic, domestic use, and land use. [3] highlighted an example of traffic-derived CO 2 in the atmosphere of an urban forest. In this context, the 14 C method has been widely used during the last decades, in various applications such as archaeology [4], forensic studies [5], hydrology [6] [7], and geology, and has gained great interest in ecology and environmental studies [8] [9]. During photosynthesis, tree rings, leaves, and short-lived plants assimilate carbon from the air, and provide changes in atmospheric 14 C concentration. Due to high dead carbon emissions in industrialized and urban areas, 14 C concentrations are diluted [10] [11]. The determination of the 14 C content in atmospheric CO 2 or from biosphere materials makes it possible to estimate the excess CO 2 or the total emission of carbon dioxide of fossil origin [12]. This is based on the differences between radiocarbon concentration in a reference site supposed "clean area" and industrial or urban area. In Senegal we are interested in determining the local variations of 14 C concentration due to fossil fuel combustion caused by the different sources of pollution, such as the transport sector, energy stations, and industries. This work aims to calculate the pollution data obtained in the vicinity of a cement factory (SOCOCIM) situated in Rufisque (33˚37'54"N, 35˚26'70"E). In mining, quarrying, crushing, grinding, and calcining generate large amounts of pollutants, mainly CO 2 [13]. This cement factory located general the cement plant undertakes various processes such as in the town of Rufisque and is created in the 1948's (before independence) however with the expansion of the city the dwellings have moved closer to the site so this could represent a risk to the environment and therefore to the local population. To evaluate this potential pollution a reference site has been chosen in a rural Village of Ngazobil (14˚12'N, 16˚52'W) which is far away from the cement factory and should not be affected by the CO 2 emissions. Accelerator Mass Spectrometry (AMS) was used to quantify the 14 C concentration of the sample material. Measurements have been performed at the ETH Zurich Laboratory of Ion Beam Physics (Switzerland) using a MICADAS instrument. The radiocarbon isotopic ratio (Δ 14 C) and δ 13 C were determined in 20 samples of several species from tree leaves collected at the two sites. Estimation of the fossil fuel fraction was carried out based on equations of mass balance for CO 2 concentra-Open Journal of Air Pollution tion, stable isotopic composition of carbon, and 14 C concentration [14].

Samples
For this study a total of 20 samples were collected around the cement factory of Rufisque ( Figure 1). In the Colobane district, leaves from Azadirachta indica, Albicia le bec and Calotropis procera were collected. Leaves from Prosopis chilensis were collected opposite the factory, 50 m from the beach. In a salad field in front of the plant, leaves from Prosopis chilensis, peltiferum, Arkin sonia, cordial, Khay, Lecenia were collected. In a distance of only 10 m from the cement factory samples of Pocéa, aubergine and Goumelia leaves were collected.
These species were chosen because they are widely grown in this region. Thus, the data obtained from this study can be compared to other Δ 14 C determined in different zones in the region. All samples were collected in the same period, at the end of the vegetation season in 2017, to avoid possible seasonal variations of 14 C concentration [15]. As reference, leaves from Ngazobil ( Figure 1) a site 101.82 km away from the cement factory located in the municipality of Joal-Fadiouth, in the department of Mbour have been chosen. Five tree leaves samples (Kaya Senegalaisis, Ziziphus mucronata, Albizia, Terminalia catapa, Faderbia albidia) were collected from this clean zone in the same sampling period.

Sampling Site
The climate of the study sites is characterized by the maritime trade winds from the Azores high, from north to north-east, it is constantly humid in winter. The sea trade winds are constantly wet, cool, or even cold in winter. Also, the 'harmattan, of direction East dominant, finishing branch of the continental trade-wind Sahelian, is characterized by a great drought linked to its long continental course, at last the monsoon, comes from the trade-wind resulting from the Anticyclone of Saint Helena in the South Atlantic. It enters the country during the summer in a south-east -north-west direction. It is marked by a low thermal amplitude, but with temperatures generally higher than those of the maritime trade-winds. The rains that fall come from weakened grain lines. They are very weakened. They are very localized in time, usually occur from July to October. However the times when they start and stop are very fluctuating [16] [17].
Samples were collected from rural areas distributed in the vicinity of a cement factory. The selected sites are distinguished by low population density and agricultural land. They are relatively far from large urban cities, about 25 km from the capital Dakar. The Suess effect could thus be attributed to the potential influence of the cement factory. The first location is situated to the west of the cement factory, while the others are distributed to the north and northeast. Figure   1 presents a map of Senegal showing the location of the studied region and the selected reference zones. Table 1 lists the coordinates of sampling locations, including reference areas, as well as their distance from the factory. Open Journal of Air Pollution

Chemical Treatment
Sample material was purified using an acid-base-acid (ABA) protocol. At first the samples were treated with a 0.5 M HCl solution at 60˚C for about 2 hours. Following a washing step in deionized water, they were introduced into 0.1M NaOH solution at 60˚C for one hour. After washing again, the materials were introduced into a 0.5 M HCl solution at 60˚C for one and a half hour. Finally, the cleaned materials were washed and dried. From the purified materials a subsample containing approximately 1 mg of carbon was taken, packed in an Al capsule, and introduced into the automated combustion and graphitization system AGE [18], where a catalytic reduction on 4 mg of iron powder of the produced CO 2 gas with hydrogen gas took place at 650˚C. Finally, the graphite produced was pressed into Al cathodes which can be introduced into the ion source of the ETHZ AMS instrument.

Measurements and Calculations
Analysis of samples has been conducted in routine measurements campaigns at The δ 13 C values as given in Table 2 are the result of the AMS measurements.
They are representative for fractionation effects which may occur during sample preparation and measurement procedure and are used to extrapolate these fractionation effects on the measured Δ 14 C (‰) (calculated from Equation (2)) and are the basis of the applied fractionation correction. They cannot be used to assess the δ 13 C of the original sample material.

Estimation of Fossil Fuel Component
To estimate the local Suess effect in the studied area, the fraction F(%) of fossil-fuel derived CO 2 that was incorporated by the plant material can be calcu-

Results and Discussion
Sampling was done at two sites. The first is the reference site or clean air site which is a village called Ngazobil (N 14˚47'39.9"; Wo 17˚02'49.8"). Ngazobil is located at the Atlantic Ocean coastline, surrounded by sea waters that often cause winds, and the local air conditions are mostly dominated by seabreeze.
The only main activity is fishing, so no industrial complexes and traffic could be responsible for any increase in local fossil CO 2 release. Thus, we can regard Ngazobil as free of urban pollution.  (Synal pers. communication). This is in nice agreement with our reference value. All Δ 14 CO 2 data at the Rufisque site are lower than the Ngazobil average value (Figure 2). We observe Δ 14 CO 2 values between 1.6‰ and 15.3‰. This clearly indicates the impact of the local cement production and the lower the Δ 14 CO 2 is, the higher is the impact of fossil CO 2 releases. In the district of Colobane located 60 m from the cement factory we observe Δ 14 C values of 11.96‰, 15.31‰, and 11.88‰, respectively. Even if these relatively high values remain lower than those obtained at the reference site. The lowest Δ 14 C (1.57‰) value was observed at a location within a approx. 50 m range from the leaves of a prosopis chilensis tree. In a salad field at the same distance to the factory, samples SOCO06, SOCO07, SOCO08 and SOCO10 gave the Δ 14 C values of 12.00‰, 13.38‰, 12.65‰, 9.86‰ and 12.46‰ respectively. Although these 14 C concentrations are relatively close to the reference value, the impact of the fossil CO 2 is significant. On the other hand, four samples (SOCO17, SOCO19, SOCO20 and SOCO22) were taken at a distance of 10m from the factory. They fall into two distinct groups, giving values of Δ 14 C 12.63‰, 12.19‰ and 5.99‰, 5.02‰, respectively. The first group shows a week, the second a rather strong impact from the CO 2 emissions. By using Equation (3), these results are converted into fraction fossil fuels F(%.) as shown in Table 3 demonstrating the degree of the Suess effect for each location   due to anthropogenic 14 CO 2 emissions from the cement factory. In general, there is no clear trend between distance from the factory and the observed 14 C depletion a week tendency indicating higher impact at lower distances to the factory may be suggested by the data set. However, very local effect as observed at sampling points SOCO20/22 and SOCO05 may overrule a general tendency. It is remarkable that for these samples between 1% -1.5% of the organic carbon has originated from the fossil fuel emissions of the cement factory.
So far, sampling was carried out on a single season (2017). Confirmation of the values over a longer time period would be helpful to draw more solid conclusions. For an assessment of the Suess effect, these values should be compared with those obtained at points around the alleged source of pollution.

Conclusion
This study falls within the general framework of the determination of air pollution due to fossil CO 2 by the various cement factory installed in Senegal. The SOCOCIM cement factory, which is the oldest in Senegal, was the subject of a collection of samples to determine the excess CO 2 due to the anthropogenic ef-