The Tanzania Zambia Mafuta (TAZAMA) pipeline started operation in 1968; by the year 1973 it had experienced 100 spills at different locations along its route (TAZAMA 2016). Oil spillage in the soil contributes to the addition of heavy metals, some of which are very hazardous to flora and fauna inhabiting a particular soil (Gordon et al., 2018). Moreover, oil pollutants present a major threat to many ecosystems (Xie et al., 2018). The effects of pollution are extended to many flora and fauna since some of oil pollutants including heavy metals can persist in soils for decades (Buskey et al., 2016; Pennings et al., 2014). Studies have documented effects of crude oil pipelines leakages on soil health and plants biological diversity (Allison et al., 2017; Asadirad et al., 2016; Oriaku et al., 2017). Crude oil spillage from pipelines is experienced along pipelines during times of construction, operation and maintenance (Vaezi and Verma, 2018).

Mikumi National Park is an important ecosystem as it shares boundary with Selous Game Reserve – a world heritage site. Crude oils contain hazardous pollutants including heavy metals and hydrocarbons. These pollutants are very toxic and are furthermore not easily degraded once in soil (Bai et al., 2019).

Despite the importance of the global ecosystem in Mikumi, the sensitivity of the general infrastructure element running through the area and the possible pollutants from oil, little is known about the amount of heavy metals that have been released to the environment and its spatial distribution in Mikumi National Park. Therefore, this study was conducted to examine the spatial distribution of heavy metals across the segment of the TAZAMA pipeline through the Mikumi National Park (MINAPA). Moreover, no Environmental Impact Assessment was conducted during the construction of the pipeline and hence no baseline information is available (TAZAMA, 2016). To cover that gap, the study involved a comparison between segments of the pipeline with history of and segments, which never experienced leakages. The following objectives were addressed during the study:

(i) What is the concentration of total heavy metals in soil in oil spilled compared to non-oil spilled segments of the pipeline?

(ii) Does the concentration of heavy metals in soil vary significantly with distance from the edge of the pipeline?

Study area

The study was conducted in Mikumi National Park, Tanzania across the segment of the TAZAMA pipeline. Five transects across the pipeline were studies, three with known history of oil spillage and two, which never experienced oil spills (Figure 1).

Soil sampling and heavy metals analysis

Transects were set at each site perpendicular to the pipeline. Soil was sampled at a distance of 0 m, 5, 20, 35, 50, 100 and 200 m away from the edge of the pipeline to the North and South. At distance of 0 m, soil samples were taken by auger to a depth of 80 cm in spilled transects and 60 cm in non-spilled transects. More depth was possible at 0 m on oil spilled transects because there were piles of soil above the general soil surface resulting from re-covering of the sites after addressing the spillage by the TAZAMA staff. At distances 5, 20 and 35 m from the pipeline soil profiles were excavated to a depth of 1.5 m or a limiting layer, total of 30 soil profiles were excavated. At 50 m, 100 m and 200 m soil samples were also taken by auger to a depth of 60 cm. Soil samples were taken from each designated horizon described using the FAO guidelines for soil profiles description (FAO 2006)(Figure 2).

Heavy metals checked were the ones, which reflect the chemistry of the oil through the TAZAMA pipeline. Six heavy metals were analysed from the soil samples including Hg, As, Pb, Cd, Cr and Cu. Total trace elements were extracted from soil by acid digestion Nitric/Perchloric acid 5:1 as described by Stewart et al. (1974). During digestion large amount of the sample was taken, that is 100 g of soil was mixed with 50 ml Nitric/Perchloric acid and digested using hot plate until the sample became colourless. After digestion the samples were filtered through suction pump and the leaching of minerals facilitated by adding 100 ml of dionazed water. The filtrate was collected in to 250 ml flask after which metals were concentrated by removing excess water using rota vapour at 60°C. The resulting concentrated solution of 10 ml was analysed for heavy metals. Metals’ concentration were done to offset the detection limits using AAS which is 0.01-0.001 depending on metal detection limits according to the calculation below.

The analysis of the levels of heavy metals was done at the University of Dar es Salaam Tanzania, using Perkin-Elmer 3100 Atomic Absorption Spectrophotometer. For each sampling point, samples were analysed in triplicates; therefore the values presented are the means of three samples.

Data analysis

Continuous vertical variability of heavy metals was modelled using equal area spline functions to get values for each soil depth (Bishop et al., 1999). Descriptive statistics were involved in calculating the mean and range of concentration of heavy metals. Linear regression analysis was involved in determination of the rate of change concentration of heavy metal along the distances from the edge of the pipeline.

Comparison between spilled and non-oil spilled transects

All  six  studied  heavy  metals were detected in the study area. As, Pb and Cr were detected in both transects, that is, with oils spills and those without history of oil spillage to a distance of 0-35 m from the edge of the pipeline, with higher mean concentration in transects with oil spillage compared to those without (Table 1). From 50-200 m away these four metals were detected in transects with oil spillage only. Hg and Cd were detected in transects, with history of oil spillage only. Cu was detected in all transects and at all ranges of distance. In the manner of abundance metals were Cu>Pb>Cr>As>Hg>Cd. Copper was the most abundant metal of all and Cadmium was the least found.

Variation of concentration of heavy metals with distance from the edge of the pipeline

Concentration of all detected heavy metals decreased with increased distance from the pipeline to both North and South direction. The rate of change of concentration of each metal per 1 m increases in distance, at each transects in both direction (Table 2). Rate of change of concentration was statistically significant in transects with oil spills for As, Pb and Cr; in transects with no oil spillage the change was not statistically significant. Hg and Cd decreased with increased distance but the increase is not statistically significant. Cu decreased with increased distance on both transects but the increase is not statistically significant (Table 4). Total concentration of all detected heavy metals were below those established by World Health Organisation WHO (Solek-Podwika et al., 2016) and Tanzania soil quality-limits for soil contaminants in habitat and agriculture (Tanzania Bureau of Standards, 2007) maximum permissible levels (Table 3).

Crude oils vary in their Chemistry hence their heavy metals contents vary from one to another. The oil transported through the TAZAMA pipeline is a murban crude oil, which is composed of gas oil and naphtha. Total heavy metals contained in the TAZAMA pipeline are Pb, As, Hg, Cr and Cu (TAZAMA, 2016). Therefore, these were the selected  metals  which  were  checked  in sampled soils. Transects with oil spillage had higher mean heavy metals concentrations compared to those without. This suggests crude oil from the pipelines contributes to the addition of concentration of heavy metals in soils (Fei et al., 2019). Total heavy metals concentration decreased with increased distance from the pipeline in both transects. This may be due to the increased distance from the source of pollution. Same findings were reported by Sun et al. (2019) when assessing the level, source and distribution of heavy metals from a typical coal industrial city of Tangshan China. The decrease of concentration of heavy metals in both transects suggests that the pipeline leads to the increase of heavy metals with and without oil spills. According to Jasper (2012), oil leakage happens in pipelines during their check-ups and maintenance. This can lead to the emission of heavy metals to the soil even when a recorded oil spillage has not occurred. Metals like Hg and Cd were only detected in transects with oil spillage. This  may  be  due to the fact that their presence is mainly attributed by anthropogenic sources particularly petroleum (Yadav et al., 2019). Fernández-Martínez et al. (2019) suggest that Hg total concentration in soils is mainly due to petroleum activities and mines than other anthropogenic activities. Higher concentration of heavy metals in transects with oil spillage than without may be due to presence of petroleum pollutants. Cu was detected in abundance in both transects; the rate of change of Cu was insignificant in all transects at all intervals. This may be due to the fact that it is an essential micronutrient in the soil. According to Chrysargyris et al. (2019), Cu is among the essential micronutrients though its availability in excessive levels may be harmful to the plants (Wyszkowski, 2019). The concentration of heavy metals in all transects were below the WHO and TZS maximum permissible limits in soils. Moreover, the TZS limits are too general and do not specify the kind of land use for a particular soil. However, more attention should be paid to the pipeline safety and maintenance since the concentration in transects with oil spillage is higher compared to the ones without.

The level of heavy metals concentration detected in all the transects is below the WHO and the Tanzania soil quality-limits for soil contaminants in habitat and agriculture (TZS) maximum permissible limits. However, the concentration is higher  in  transects with history of oil spillage compared to  those  without.  This calls  for  more attention since  prevention  of  heavy  metals  pollution  is currently a global agenda. For the Mikumi National Park in particular, more attention should be given for the TAZAMA pipeline since it is a home for various species of flora and fauna including the endangered and critically endangered species.

The authors have not declared any conflict of interests.

The researcher expresses her gratitude to the PELIBIGO Project under EnPe (Energy and Petroleum) Norway, for funding this study.