The pollution levels of heavy metals are generally low in natural aquatic ecosystems (Komijani et al., 2021). However, with the rapid growth of human population and accelerated urbanization, heavy metal pollution in water typically occurs because of the irrational exploitation of natural resources and many industrial and agricultural activities (Nguyen et al., 2017; Luo et al., 2022; Peng et al., 2022). Heavy metals in water generally originate from geological background deposition, surface runoff, and atmospheric dry–wet deposition (Li et al., 2015; Liu et al., 2018; Miranda et al., 2021; Peng et al., 2022). When heavy metals enter water, they gradually accumulate in sediments through suspended particulate matter deposition and chemical precipitation. Dissolved heavy metals could be assimilated by hydrobionts (Chaturvedi et al., 2018; Faisal et al., 2022; Li et al., 2014a, b; Peng et al., 2022). When environmental conditions such as pH, redox potential, and salinity change, heavy metals in sediments are released into the overlying water through a series of physical, chemical, and biological processes. Heavy metals directly and indirectly produce toxic and harmful effects on aquatic organisms and subsequently affect human health through biological enrichment and food chain amplification process, which is essential for hydrobionts and human health (Wang et al., 2013, Abou Zakhem et al. Zuo et al., 2016, Arisekar et al., 2022). Therefore, studying heavy metals in sediments is critical for environmental safety and biological health.

Heavy metals in sediments have been extensively studied in developed and developing countries, such as China and India (Li et al., 2014a, b; Omer et al., 2016; Tang et al., 2018; Wu et al., 2023; Zhang et al., 2017), and relatively underdeveloped African countries have received less attention. Despite the low levels of industrial and agricultural production in Africa, heavy metal concentrations in rivers and soils have recently increased as a result of industrial mining, fossil burning, and the extensive use of agricultural fertilizers and fungicides (Semu & Singh, 1995, Yabe et al., 2010; Mungai et al., 2016; Mwelwa et al., 2023). Meanwhile, because of ineffective implementation of environmental regulations, unsupported management of aquatic resources, and a marginal awareness of sewage discharge control, the problem of heavy metal pollution cannot be ignored.

Lake Tanganyika is located south of the equator (3°–9°S). It is the second deepest lake in the world (maximum depth of 1470 m) and the sixth largest lake in the world (approximately 32900 km2). This is a nearly closed system with permanently stratified water. Atmospheric deposition, precipitation, and river inputs are the major external sources of input (Coulter, 1991). Lake Tanganyika is home to many endemic species and is an important source of water and animal proteins in the lakeside region. In addition to the two major inflow rivers, the Rusizi River in Burundi and Malagarasi River in Tanzania, there are many small rivers along the lake. With the drastic increase in population and the development of industrial and agricultural activities, the rivers, bays, and harbors along Lake Tanganyika have become polluted. In particular, urban rivers located northeast of Lake Tanganyika are strongly affected by human activities, and environmental problems such as soil erosion and water pollution are severe (Coulter, 1992; Vandelannoote et al., 1996). Studies on heavy metals in Lake Tanganyika have primarily focused on fish and sediments. The heavy metal content of fish in Lake Tanganyika was lower than that in other lakes observed during the 1990s (Vandelannoote et al., 1996). The Pb and Fe contents in the bivalves were elevated in Lake Tanganyika, and the heavy metal contents in nearshore sediments were greater than those in the offshore sediments (Sindayigaya et al., 1994). A positive correlation has been observed between the heavy metal content of soil and tobacco leaves and the use of fertilizers and fungicides in Tanzania along Lake Tanganyika (Chale, 2002). Previous studies have primarily focused on the content and distribution of heavy metals in the sediments of Lake Tanganyika and its watersheds (Chale, 2002; Coulter, 1992; Sindayigaya et al., 1994; Vandelannoote et al., 1996). However, owing to the lack of the monitoring of pollutants in the water and sediments of inflow rivers, determining the status and ecological risks of heavy metal pollution in inflow rivers is difficult. The objectives of this study were to (1) investigate the concentrations of heavy metals in the surface sediments of the inflow rivers in northeastern Lake Tanganyika, (2) determine the status and potential ecological risk of heavy metal pollution, and (3) identify the sources of heavy metal pollution using multivariate statistical techniques. The results are of practical significance for protecting the aquatic life and ecosystem of the lake and ensuring the safety of residents along the lake.

1 Geographical Setting

The study area was located in the municipality of Bujumbura in the Republic of Burundi and Kigoma in the United Republic of Tanzania. Both Bujumbura and Kigoma belong to central-eastern Africa and have tropical dry and wet season climates. The dry season is distinct from the rainy season, which generally lasts from October to April, whereas the dry season lasts from May to September. The temperature in Bujumbura ranges from 19 to 29 °C, with a mean temperature of 24 °C. The temperature of Kigoma varies from 19 to 28 °C, with an average temperature of 23.5 ℃. The annual precipitation rates in Bujumbura and Kigoma were 855 mm and 977 mm, respectively.

Bujumbura is the largest city along Lake Tanganyika, with the highest population density of 3276 persons/km2, which contains Burundi’s main industries, such as chemical, paper, and dairy industries. During recent years, ecological deterioration in Bujumbura and its surrounding areas has been aggravated, and environmental problems have become increasingly prominent. Kigoma, an inland area in western Tanzania, is an industrial laggard with an agricultural economy that grows cassava, maize, beans, bananas, potatoes, coffee, and palm trees. However, with economic development and rapid population growth, coupled with the fragile nature of the region, soil erosion around Kigoma is becoming severe, and environmental pressure continues to increase (Bigirimana, 2012). In this study, the surface sediments of the inflow rivers in northeastern Lake Tanganyika were selected as research subjects. The distribution characteristics of heavy metals and their potential ecological risks to surface sediments from inflow rivers were studied.

2 Materials and Methods

2.1 Sampling and Measurement Methods

In March 2014, 16 samples were collected from the tributaries, main streams, and estuarine wetlands of the Luiche and Malagarasi rivers in Tanzania, along with three main rivers in Bujumbura (Fig. 1, Table 1). Surface sediments of inflow rivers were scraped to about 1–2 cm depth using a Peterson sampler.

Fig. 1
figure 1

Study area with sampling sites for the inflow rivers in the northeastern quadrant of Lake Tanganyika

Full size image
Table 1 Essential information regarding the 16 surface sediments of the inflow rivers in the northeastern quadrant of Lake Tanganyika
Full size table

The samples were completely mixed and sealed in polyethylene ziplocked bag. All samples were stored at 4℃ and transported to the laboratory for pretreatment and analysis. All samples were air-dried and then sieved through a 100 mm mesh to remove stones and sundried portions. Approximately 0.12 g of the ground samples were placed in a Teflon breaker. The HCl, HNO3, HF, and HClO4 were added in sequence. All the samples were heated on an electric heating plate at 160–180 °C for approximately 2.5 h, and then heated at 220 °C until they no longer emitted smoke. After cooling, 1 mL HNO3, 0.5 mL HCl, and ultra-pure water were added to dissolve the residue. The Zn, Fe, and Mn contents were determined via inductively coupled plasma-atomic emission spectrometry (ICP-AES, Leeman Labs, Profile DV). The Cu, Cd, Pb, and Cr concentrations were measured via inductively coupled plasma mass spectrometry (ICP-MS, Agilent 7700x). Data quality control was performed by using a procedural blank, duplicates and standard reference material GBW07309 after every tenth sample. No contamination was detected in laboratory procedural blank. Accuracy, expressed as recovery of the reference material, was between 92 and 106% for the metals.

2.2 Data Analysis

The indices for geoaccumulation (Igeo) and enrichment factor (EF) have been employed to assess heavy metals within sediments (Zhang et al., 2009; Varol & Şen, 2012). The concentrations of heavy metals in the sediments of the inflow rivers in the northeastern Lake Tanganyika were assessed by the Igeo and EF values.

The Igeo introduced by (Műller, 1979) was used to determine and define heavy metal contamination within the sediments for rivers inflowing in the northeastern quadrant of Lake Tanganyika. Igeo is defined by the following equation: Igeo = \({\text{log}}_{2}(\frac{Cn}{1.5Bn})\), where Cn is the measured concentration for heavy metal “n” and Bn is the background value of the same. The factor 1.5 is a correction value, resulting from lithogenic effects, for the background matrix. The background values used were those observed by (Taylor & McLennan, 1995): 35 mg/kg for Cr, 20 mg/kg for Ni, 25 mg/kg for Cu, 71 mg/kg for Zn, 0.098 mg/kg for Cd, 35 mg/kg for Fe, 20 mg/kg for Pb, and 600 mg/kg for Mn.

Fe is one of the most abundant elements on the earth, and is commonly used for normalization purpose (Table 2). EF is expressed as follows (Salomons & Forstner, 1984; Sinex & Wright, 1988): EF = \(\frac{\left(\frac{Me}{Fe}\right)sample}{\left(\frac{Me}{Fe}\right)Background}\), where (Me/Fe)sample is the heavy metal to Fe ratio in the samples, and (Me/Fe)sample is the natural background value of the heavy metal to Fe ratio. The background values of the heavy metals were the same as those used in Igeo. The classification of Igeo and EF is presented in Table 3 (Han et al., 2006; Műller, 1979).

Table 2 Comparison of heavy metal concentration in inland water sediments of Africa (mg/kg)
Full size table
Table 3 Classification of heavy metals evaluated with geoaccumulation index (Igeo) and enrichment factor (EF)
Full size table

3 Results and Discussions

3.1 Concentrations and Spatial Variations of Heavy Metals

The concentrations of heavy metals in the surface sediments of the inflow rivers in northeastern Lake Tanganyika are shown in Fig. 2. The concentrations (mg/kg) of heavy metals such as Cr (5.05–69.29, mean of 22.11), Ni (1.99–39.56, mean of 11.99), Cu (5.08–64.43, mean of 18.42), Zn (6.19–48.39, mean of 21.18), Cd (0.02–0.11, mean of 0.05), Pb (2.64–18.54, mean of 6.58), Fe (10.74–94.05, mean of 45.25), and Mn (156.03–1714.90, mean of 747.84) exhibited wide variations.

Fig. 2
figure 2

Variations in the concentration of heavy metals in the surface sediments of the inflow rivers within in the northeastern quadrant of Lake Tanganyika

Full size image

The spatial distribution of heavy metals, including Cr, Ni, Cu, Zn, Cd, Pb, Fe, and Mn, in the surface sediments transported by inflow rivers in the northeastern quadrant of Lake Tanganyika at different sites is shown in Fig. 2. Heavy metal concentrations show significant spatial variations along northeastern Lake Tanganyika. High concentrations of Zn, Pb, and Cd were observed in the rivers of the Burjumbura city (TL1-2), whereas low values were observed in the Luiche River (TL5 and TL6). By analyzing the relationship between heavy metals and land types (Table 1), we observed that the average contents of Zn, Pb, and Cd decreased within estuary wetland and grassland with increasing distance from urban areas. This indicated that the heavy metals of the inflow rivers are closely related to the land use type. The heavy metal content in the inflow river sediments was higher in the areas with high human activity owing to runoff input. The Cu content was high in sampling sites TL5 and TL6. After field investigation, the TL5 and TL6 sites were generally located in secondary grassland, which belongs to fallow farmland. The surrounding woodland primarily comprises palm forest. A large area of tall and densely planted palms has the characteristics of forest and is classified as forestland. However, it is a perennial oil crop that is fertilized during the process of planting and growing. These farmlands and palm forests may be responsible for the high Cu content in the inflow river sediments. The Fe, and Mn contents were significantly different in different regions, with the highest content at the TL5 site and lowest content at the TL7 site. Compared with background values (Műller, 1979), the Fe content at 10 sampling sites was greater than the background value, and the Mn contents at the four sampling sites were higher than the background value.

3.2 Assessment of Heavy Metals Contamination

3.2.1 Geoaccumulation Index (Igeo)

The Igeo of heavy metals in the surface sediments of the inflow rivers in northeastern Lake Tanganyika is presented in Table 4. Based on the classification of Igeo values, the concentrations of Zn, Cd, and Pb were defined as unpolluted, however, Cr, Ni, Cu, Fe, and Mn exhibited significant spatial variations in the northeastern Lake Tanganyika. Cr was unpolluted to moderately polluted at the TL3 site, whereas it was unpolluted at the other sites. Ni was unpolluted to moderately polluted at the TL7 site, whereas it was unpolluted at the other sites. Cu was unpolluted to moderately polluted at the Tl6 and TL7 sites, whereas it was unpolluted at the other sites. Mn was unpolluted to moderately polluted at the TL2, TL5, TL6, and TL16 sites, whereas it was unpolluted at the other sites. Relatively high pollution level was observed in an urban area and estuary wetland, with low pollution levels in woodland and grassland away from the urban area.

Table 4 Heavy metal contamination of the surface sediments evaluated with geoaccumulation index (Igeo) and enrichment factor (EF) for inflow rivers within the northeastern quadrant of Lake Tanganyika
Full size table

3.2.2 Enrichment Factor (EF)

The EF values of heavy metals in the surface sediments of the inflow rivers in northeastern Lake Tanganyika are presented in Table 4. Based on the classification of EF values, the Pb, and Fe concentrations were only slightly enriched in the northeastern quadrant of Lake Tanganyika. Cr and Cd were moderately enriched in the TL1 and TL2 sites. Moderately severe enrichment was observed at the TL7 site, whereas minor enrichment was observed at other sites. Ni and Cu were greatly enriched at the TL7 site, with minor enrichment at the other sites. Zn was moderately enriched in the TL7 site, and minor enrichment was observed at other sites. Mn exhibited moderate enrichment in the TL1, TL2, and TL5 sites, moderately severe enrichment in the TL16 site, and minor enrichment at other sites. Mn had moderate enrichment at the TL8 site and moderately severe enrichment at the TL4, TL6, TL7, TL9, TL10, and TL13 sites. Severe Mn enrichment was observed at TL5, TL11, and TL12 sites. In addition, very severe Mn enrichment was observed at TL1, TL14, TL15, and TL16 sites and extremely severe enrichment at TL2 and TL3 sites. Thus, there is a significant spatial variation in the enrichment levels of heavy metals in northern Lake Tanganyika. Cr, Ni, Cu, Zn, and Cd were minor in woodland and grassland farther from urban areas, whereas high enrichment levels were observed in the town area and the estuary wetland.

3.3 Multivariate Statistical Analysis

3.3.1 Pearson’s Correlations Analysis

The Pearson’s coefficient of determination is an effective method for determining inter-elemental associations and can provide more details about sedimentary components (Guo et al., 2014; Wang et al., 2021). Pearson’s correlation analysis results of heavy metals in the surface sediments of the inflow rivers of northeastern Lake Tanganyika are presented in Table 5. A significant positive correlation existed between the pairs of heavy metals, such as Cu–Ni (0.813), Zn–Cr (0.875), Zn–Ni (0.675), Cd–Cr (0.649), Cd–Zn (0.835), Pb–Cr (0.730), Pb–Zn (0.762), Pb–Cd (0.715), and Mn–Fe (0.809). In addition, there was a weak positive correlation between Cd, and Pb and a weak negative correlation between Fe and Cr, Ni, Cu, Zn, Cd, and Pb. These results indicated that Cr, Zn, Cd, and Pb may have similar input sources and distribution characteristics. Ni, Fe, and Mn have the same source and geochemical characteristics. Cu was not significantly correlated with other heavy metal elements.

Table 5 Pearson’s correlation matrix for heavy metals in the surface sediments of inflow rivers within the northeastern quadrant of Lake Tanganyika
Full size table

3.3.2 Principal Component Analysis (PCA) and Cluster Analysis (CA)

PCA is a useful method for combining many variables into several potential components forming the basis of multivariate data (Ziegler & Murray, 2007; Yao et al., 2012). The scores for heavy metals in the surface sediments of the inflow rivers in northeastern Lake Tanganyika are presented in Table 6. The results showed that the heavy metals Cr, Ni, Cu, Zn, Cd, Pb, Fe, and Mn were dominated by three principal components (F1, F2, and F3) that accounted for 89.879% of the total variance. F1 accounted for 46.500% of the total variance of three principal components and had high loadings for Cr (0.870), Zn (0.990), Cd (0.853), and Pb (0.745). F2 accounted for 24.322% of the total variance of three principal components with a positive Fe (0.797) and Mn (0.811) loading and a negative Ni (-0.418) loading. F3 accounted for 19.058% of the total variance of three principal components with a high loading of Cu (0.714). These results also suggest that F1, F2, and F3 have different geochemical behaviors and sources. Previous studies have confirmed that Cd is primarily derived from phosphate fertilizers, agrochemicals, and pesticides. Ni and Mn originate from the weathering of parent materials and subsequent pedogenesis. Pb is used as a tracer of vehicles, agrochemicals, and industrial wastes, whereas Cu is a marker of paint and metal industries (Li & Zhang, 2010). Our studies indicate that Ni, Fe, and Mn primarily originate from natural sources, Cu from metal industries, and Cr, Zn, Cd, and Pb may originate from natural sources and anthropogenic inputs, including agricultural and industrial inputs, which are mixed because of transport processes. Because Cr, Zn, Cd, and Pb were slightly and moderately enriched, we conclude that the heavy metals Cr, Zn, Cd, and Pb might have originated from a mixture of natural sources and anthropogenic inputs, including agricultural and industrial inputs. Ni, Fe, and Mn primarily originated from natural sources, and Cu originated from anthropogenic inputs such as vanadium iron ore mining and iron and steel smelting.

Table 6 Component matrixes for heavy metal concentration (principal component analysis with varimax rotation) in the surface sediments of inflow rivers within the northeastern quadrant of Lake Tanganyika
Full size table

The purpose of CA is to identify groups of similar sampling sites based on similarities within a class and dissimilarities between classes (Zhang et al., 2009). The CA results were generated using SPSS software (version 17.0) and showed two major clusters: cluster 1 (TL1, TL3-4, and TL7-5) and cluster 2 (TL2, TL5-6, and TL16) (Fig. 3). The land types of the sites in cluster 1 were artificial surface and arable land, the heavy metals of this cluster are influenced by the anthropogenic impact. The sites in cluster 2 were grass and wetland, the heavy metal concentrations correspond to the natural sources.

Fig. 3
figure 3

Dendrogram showing the clustering of sampling sites according to Ward’s method using the squared Euclidean distance

Full size image

As discussed above, this study commenced by examining the source of heavy metal pollution, utilizing sediment from the river flowing into Lake Tanganyika as the focal point of research. Through a comprehensive analysis of the heavy metal content in the sediment of Lake Tanganyika, the study investigated the current status of heavy metal pollution and its associated ecological risks, ultimately uncovering the origins of such pollution. This study posited that a comprehensive assessment of metal contamination cannot rely solely on metal concentrations. It was recommended that a holistic approach incorporating sediment standard criteria, enrichment factor, and geoaccumulation index should be employed to more effectively evaluate the fate and transport of metals originating from human activities and their consequent environmental effects on lake sediments.

4 Conclusions

  1. (1)

    The heavy metal concentrations within surface sediments transported by the inflow rivers within the northeastern quadrant of Lake Tanganyika exhibited significant spatial variation. The correlation analysis between heavy metal content in surface sediments and land use structure shows that the heavy metal content is highest near urban areas, followed by estuarine wetlands and forest grasslands, which shows that the heavy metal content is closely related to human activities. The heavy metal content in river surface sediments is higher in areas with intense human activity.

  2. (2)

    According to Igeo and EF, the concentrations of Zn, Cd, and Pb were defined as unpolluted, however, Cr, Ni, Cu, Fe, and Mn exhibited significant spatial variations in the northeastern quadrant of Lake Tanganyika. The relatively high pollution level was observed in the town area and estuary wetland, with low pollution levels in woodland and grassland away from the town. The concentrations of Pb, and Fe were only slightly elevated in the northeastern quadrant of Lake Tanganyika. The EF levels of Cr, Ni, Cu, Zn, and Cd were minor in woodlands and grasslands away from the town, whereas high enrichment levels were observed in the town area and estuary wetland.

  3. (3)

    PCA and correlation analysis (CA) revealed that the heavy metals Cr, Zn, Cd, and Pb originate from a mixture of natural sources and anthropogenic inputs, including agricultural and industrial inputs. Ni, Fe, and Mn primarily originate from natural sources, whereas Cu originate from anthropogenic inputs such as vanadium iron ore mining and iron and steel smelting. These results would help to establish nonpoint sources of heavy metal pollution and provide valuable information for water conservation strategies in the Lake Tanganyika Basin.