Effects of trace elements on benthic macroinvertebrate distribution in the sediments of two rivers in the Olifants River Basin, South Africa

Abstract Many freshwater ecosystems are facing ecological challenges such as influx of trace elements from human activities, and sediments behave as a sink for trace elements in these environments. Trace element contamination in river sediments is a major environmental concern due to its devastating effect on benthic organisms. The aim of this study was to assess the effect of trace elements in sediments on the distribution of benthic macroinvertebrates in the Blyde and Mohlapitsi rivers of the Olifants River Basin, South Africa. Sediment samples and benthic macroinvertebrates were collected from six sites along each river. The results showed variations in the concentration of the trace elements in the sediment and the distribution of benthic macroinvertebrates among the different sites and between the two rivers. Both rivers recorded high taxa richness and sensitive taxa were well represented in most parts of the rivers, an indication that the rivers are not highly disturbed. There were insignificant correlations between most of the trace elements and the abundance of the benthic macroinvertebrates, though some taxa showed significant correlations with the trace elements. Despite the low impact of the trace elements on the distribution and diversity of benthic macroinvertebrates, ecological health of the rivers and structural distribution of macroinvertebrates may be affected in the future due to increasing human activities in the catchments. It is therefore recommended that effluents from human activities should be controlled to prevent a further degradation of the rivers to protect macroinvertebrates and other aquatic biota.


Introduction
Many surface water ecosystems are being polluted due to human activities such as mining, industries and agriculture, coupled with domestic waste and vehicle emissions (Lai et al. 2013, Wang et al. 2020, Li et al. 2022).The pollutants from these activities include trace elements, which can alter the chemical and biological characteristics of the aquatic system beyond their natural self-purification capacity.In many developing countries, trace element contamination is aggravated by inadequate environmental management practices and their enforcement (Ferreira et al. 2017, Santos et al. 2017, Addo-Bediako et al. 2018).Due to their prevalence and toxicity, trace elements in aquatic ecosystems pose a serious environmental threat (Xu et al. 2018, Addo-Bediako 2020, Venkateswarlu and Venkatrayulu 2020).Their toxicity in aquatic ecosystems is dependent on their bioavailability (Addo-Bediako and Malakane 2020).
River sediments have a strong adsorption capacity for pollutants, and therefore, contamination in the sediments is often greater than in the water column (Addo-Bediako 2022).Thus, sediments can behave as sinks of trace elements that are then released into the water column and/or accumulate in plant and animal tissues before entering food chains (Oremo et al. 2019, Balzani et al. 2021).Sediments have been used in studying trace element accumulation in rivers because they usually contain historical evidence of natural and anthropogenic fluxes of the elements.
Sediments serve as a habitat for various benthic macroinvertebrates and many other organisms (Shuman et al. 2020).Macroinvertebrates are known to be the most suitable organisms for biomonitoring as they have low mobility, and are in contact with both sediments and the water column, thus being exposed to pollutants in both compartments, and show a wide range of tolerance to pollutants (Rico-S anchez et al. 2022).They are therefore potential bioindicators of not only water quality but also sediment quality (Pallottini et al. 2015).Sediments contaminated with trace metals pose a threat to benthic macroinvertebrates and may also affect other organisms in higher trophic levels.Trace element contamination can reduce benthic macroinvertebrate species richness, as well as density, growth and reproduction (Qu et al. 2010).The effects of trace elements on macroinvertebrates are highly variable among taxa and may affect ecosystem function (Qu et al. 2010).Benthic macroinvertebrates live in the benthic substrate, and are the main material exchangers across the sediment-water interface (Bere et al. 2016).They have a wide range of sensitivities to contamination by metals (Iwasaki et al. 2018), and they are often used in assessing the ecological impacts of metals in streams and rivers (Namba et al. 2020).
The Olifants River Basin is continuously being polluted due to industrialization, farming activities, transportation, urbanization and domestic wastes, and it is now one of the most polluted river systems in South Africa (Ashton and Dabrowski 2011).There is, therefore, increased concern regarding the long-term impact of water pollution on the aquatic ecosystem and the health of communities in the basin, especially those still reliant on untreated water from the river and its tributaries.The aim of the study therefore, was to assess the effects of trace elements contamination on benthic macroinvertebrate communities in two rivers in the Olifants River System, the Blyde and Mohlapitsi rivers.The study used multivariate analysis and other metrics to determine: (i) the importance of trace elements in the sediments in explaining benthic macroinvertebrate distribution; (ii) the relative importance of different trace elements on macroinvertebrate composition; and (iii) the macroinvertebrate families that are tolerant and those that are sensitive to trace element contamination.

Study Area
Two rivers of the Lower Olifants sub-catchment were selected for the study, the Blyde and Mohlapitsi Rivers.The Blyde and Mohlapitsi rivers serve as important tributaries of the Olifants River, as they provide water of good quality to it (Ashton and Dabrowski 2011).The Blyde River sub-catchment lies partly on an escarpment and as a result experiences considerably higher rainfall than the other sub-catchments in the Olifants River Basin, with mean annual precipitation sometimes exceeding 1000 mm.However, mean annual precipitation is generally less than 600 mm outside the escarpment.Most of the rain occurs during the summer months (between November and February).The average maximum and minimum temperatures for summer are 30.4 and 19.7 C, and winter temperatures are 25.3 and 11.5 C (Raven 2004).The Mohlapitsi River sub-catchment has mean annual rainfall of about 500-600 mm (Jogo and Hassan, 2010), and temperatures from an average monthly maximum and minimum of 30.2 C and 18.0 C for January to 22 C and 5.2 C for June, respectively (Nell and Dreyer 2005).
The rivers navigate different land uses including human settlements, forested area, mixed agriculture and nature reserves.The activities in the Blyde River catchment include lodges and agricultural fields in the upstream, mainly citrus, while the middle is characterized by mixed farming and settlements, and the lower part is characterized by nature reserves.In the upper region of the Mohlapitsi River, it flows through a nature reserve, while the middle is characterized by mixed farming and human settlements and lower courses are characterized by nature reserve/livestock grazing.The climate of the basin is warm to hot sub-tropical, with seasonal rainfall occurring during the summer months (October to March), peaking in January.Six sampling sites were selected along each river to cover the different land uses in catchments.In the Blyde River, the sites were BS1, BS2 (upstream), BS3, BS4 (midstream), and BS5, BS6 (downstream).In the Mohlapitsi River, the sites were MS1, MS2 (upstream), MS3, MS4 (midstream), and MS5, MS6 (downstream) (Figure 1).

Sampling of sediments
Surface sediment samples were collected during summer and winter seasons of 2018 and 2019.The samples were collected at a depth of between 5 cm and 10 cm using a hand trowel.At each site, five sub-samples were mixed together, forming a composite sample (Bervoets and Blust 2003).The samples were placed in nitric acid pre-treated containers and transported to the laboratory.The purpose of acidification (usually nitric or sulfuric) is to fix a sample so that it will not change until it is analyzed.Prior to chemical analysis at an accredited (ISO 17025) laboratory in Pretoria, South Africa, the samples were frozen at À20 C. The samples were stored in acid-washed polypropylene pre-weighed vials and dried at 60 C for 24 h.The samples were then sieved through a 2-mm nylon sieve to remove any stones and coarse debris.A 0.1 g of each sediment sample was digested with 8 ml of 68% nitric acid (HNO 3 ) and 3 ml of 40% hydrochloric acid (HCl).The digested sample was then passed through a membrane filter and the concentrations of the elements were analyzed using inductively coupled plasma optical emission spectrometry (ICP-OES) (Perkin Elmer, Optima 2100 DV).Analytical accuracy was determined using certified standards (De Bruyn Spectroscopic Solutions 500 MUL20-50STD2) and recoveries were within 10% of certified values.

Sampling of benthic macroinvertebrates
Seasonal samples of benthic macroinvertebrates were collected at the selected sites and seasons (summer and winter) of each of the rivers.The samples were collected covering mud, sand, gravel and stones biotopes.Samples were collected using a hand-held kick net (dimensions 30 Â 30 cm, mesh size 500 lm), according to Dickens and Graham (2002) method, whereby the substrate was disturbed by kicking sediment and stones to free macroinvertebrates.Sampling was repeated three times at each site.The macroinvertebrates were then separated from organic and mineral matter and were emptied into labelled 500-mL polyethylene bottles and preserved in 70% ethanol and transported to the laboratory.The macroinvertebrates were identified up to the family level, with the aid of a stereomicroscope (Leica EZ4) and magnifying glass, using guides by Gerber and Gabriel (2002).

Statistical analysis
One-way ANOVA was performed to determine significant difference among the sites of the rivers for the benthic macroinvertebrates.Significant ANOVAs (p < 0.05) were followed by Tukey HSD post hoc analysis to identify differences between study sites.The Shannon Weiner index was used to compare the macroinvertebrate diversity at the selected sites.The following macroinvertebrate metrics were also used to assess the environmental integrity of the two rivers: Ephemeroptera-Plecoptera-Trichoptera (EPT), abundance, taxa richness, Shannon-Wiener diversity and evenness.To assist the interpretation of the changes in community profiles, Spearman correlation coefficients between the abundance of macroinvertebrates and trace element concentrations were calculated.Differences of benthic macroinvertebrate communities in relation to environmental variables (trace elements) were analyzed using multivariate statistical analyses.

Trace elements in the sediments
There were variations in the concentrations of trace elements recorded in the sediments of the two rivers (Table 1).The concentrations of As, Cu, and Ni were higher in the Blyde River than the Mohlapitsi River, whiles the concentrations of Al, Cr, Fe, Mn, Pb and Zn were higher in the Mohlapitsi River than in the Blyde River.In the Blyde River, the highest concentrations of Al (46 196 mg kg À1 ), As (107.6 mg kg À1 ), Cu (274.3 mg kg À1 ), Fe (48221 mg kg À1 ), Pb (16.1 mg kg À1 ) and Zn (75.7 mg kg À1 ) were recorded at BS3, and the highest concentrations of Cr (108 mg kg À1 ), Mn (1298.9mg kg À1 ) and Ni (281.7 mg kg À1 ) at BS5.In the Mohlapitsi River, the highest concentrations of As (21.1 mg kg À1 ), Cu (62.8 mg kg À1 ) and Pb (15.2 mg kg À1 ) were recorded at MS1, Mn (3221 mg kg À1 ) at MS2 and MS5, Fe (64 991 mg kg À1 ) at MS3, Cr (745 mg kg À1 ), Ni (124 mg kg À1 ) and Zn (701 mg kg À1 ) recorded at MS5, and Al (44 518 mg kg À1 ) at MS6.The concentration of As exceeded the CCME (Canadian Council of Ministers of the Environment) (2012) recommended value of 5.9 mg kg À1 at all sites in both rivers, except at MS5.The concentrations of Cr and Cu exceeded the recommended values of 37.5 and 35.7 mg kg À1 , respectively, at all sites of the two rivers.However, the concentrations of Pb and Zn were below the recommended values of 35 and 123 mg/kg, respectively, in both rivers (Table 1).The concentration of Al was below the average shale value at all sites of both rivers, As concentration was only below the average shale value at MS3 and MS5, the Cr concentration was below the average shale value at BS1, BS2, BS4 and BS6, Cu concentration was below the average shale value at BS1, MS3 and MS4, the Fe concentration was above the average shale value at BS3, MS3, MS4 and MS5, Mn concentration was less than the average shale value at BS1, BS2, BS6 and MS6, Ni concentration was less than the average shale value at MS3, Pb and Ni concentrations were below the shale values at all sites of both rivers (Table 1).The percentage composition of each trace element is shown in Figure 2a and 2b, showing variation in the distribution of the elements in both rivers.

Structural composition of macroinvertebrate
A total of 17 674 individual macroinvertebrates belonging to 34 families and 11 orders were recorded in the Blyde River, and a total of 26 240 macroinvertebrates belonging to 63 families and 11 orders/class were recorded in the Mohlapitsi River.The total number of taxa was between 25 and 32 at the different sites of the Blyde River and between 32 and 44 at the different sites of the Mohlapitsi River (Table 2).The most abundant family in the Blyde River were the Hydropsychidae (3423), Caenidae (2396), Baetidae (2207) and Simuliidae (2206), and in the Mohlapitsi River, the most dominant taxa were Thiaridae (16661), Baetidae (2527) and Planorbidae (2096).In the Blyde River, the highest Shannon's diversity (H' ¼ 1.06) and evenness (E ¼ 0.31) were found at BS6, and the lowest Shannon's diversity (H' ¼ 0.89) was at BS1 and the lowest evenness (E ¼ 0.27) at BS3.In the Mohlapitsi River, the highest Shannon's diversity (H' ¼ 1.07) and evenness (E ¼ 0.29) were found at MS2, and the lowest Shannon's diversity (H' ¼ 0.41) and evenness (E ¼ 0.12) were recorded at MS3 (Table 3).The composition of the EPT taxa differed among the different sites in both rivers.In the Blyde River, there was no significant difference in the EPT taxa (ANOVA: p > 0.05) among the sites, but in the Mohlapitsi River, there was a significant difference in the taxa among the sites (ANOVA: p < 0.05).There was a significant difference observed among the sites in the Blyde River (ANOVA, F ¼ 3.6, p < 0.005).Tukey's post-hoc analysis revealed significant differences between BS1 and BS5 (p < 0.01), BS3 and BS5 (p < 0.01) and BS4 and BS5 (p < 0.005).However, there was no significant difference among the sites in the Mohlapitsi River (ANOVA, F ¼ 0.5, p > 0.05) (Table 3).

Structural composition of macroinvertebrates in relation to environmental variables
Generally, the correlations between trace elements and abundance of macroinvertebrates were insignificant and only a few taxa had significant correlations with the trace elements.
In the Mohlapitsi River, the following relationships were significant, Al negatively affected the abundance of Naucoridae (r¼ À0.928, p < 0.001), As had a negative effect on abundance of Elmidae and Thiaridae ((r¼ À0.986, p < 0.001 and r¼ À0.928, p < 0.001,  In the Blyde River, the projection of the trace element concentrations and the macroinvertebrate assemblages showed a total variance of 70.2% (Figure 3).The first axis, explaining 48.5% of variance, and positively related to As and Cu.The second axis, explaining 21.7% of variance, positively related to Cr and Fe.Most of the taxa that were negatively correlated with the trace elements were distributed at the lower part of the NMDS ordination (Figure 3).In general, most of the taxa especially the sensitive taxa were distributed at the reverse direction of the trace element contamination.In the Mohlapitsi River, the trace element concentrations and the macroinvertebrate assemblages had a total variance of 58.68%.The first axis, explaining 30.84% of variance, and was negatively related to As and Cu.The second axis, explaining 27.84% of variance, and was positively related to Fe and Cr.Similarly, in the Mohlapitsi River, most of the taxa were negatively correlated with the trace elements, and were distributed at the lower part of the NMDS ordination (Figure 4).

Trace elements in the sediments
In freshwater ecosystems, weathering and erosion contribute significantly to natural pollution, due to run-off water leaching from soil and rock, both of which contain certain amounts of trace elements (Kolarova and Napi orkowski 2021).The Olifants River and its tributaries are currently under pressure from human activities in the catchments, and are being impacted by pollutants including trace elements.In this study, the concentrations of trace elements, especially As, Cr and Cu were high and exceeded the recommended values in many sites (CCME (Canadian Council of Ministers of the Environment) 2012), whilst the concentrations of Pb and Zn were below the recommended guideline values (CCME (Canadian Council of Ministers of the Environment) 2012), and therefore pose no serious threat in the study areas.
There were variations between the rivers and sites due to the fact that different types of activities resulting in trace element pollution in sediments are not necessarily the same for different geographical locations.The high concentrations of As, Cu, Mn and Ni in the Blyde River, especially in the midstream (BS3) could be attributed to many agricultural activities, such as commercial citrus and maize farming, which use fertilizers and pesticides (Islam et al. 2020, Kumar et al. 2021).In the Mohlapitsi River, the high concentrations of Cr, Cu and Mn could be from fertilizers and fungicides (Khan et al. 2020, Li et al. 2022).

Structural composition of macroinvertebrate
In both rivers, most of the tolerant taxa were recorded in areas with high disturbances.Generally, there was no significant difference in EPT taxa among the sites in the Blyde River (range from 8 to 10); however, there was a significant difference in EPT among the sites in the Mohlapitsi River (6-15).The midstream sites, MS3 and MS4 had the lowest EPT taxa.This could be related to the increase in pollution loads and disturbances.The EPT taxa are generally considered to be sensitive to environmental stress (Suhaila andChe Salmah 2017, Rico-S anchez et al. 2022), such as pollution and removal of riparian vegetation which can lead to alterations in the longitudinal flow profile and homogeneity of habitats within the systems.This may cause reduction in diversity of macroinvertebrate composition between different habitats (Hepp et al. 2013, Do Amaral et al. 2015).Furthermore, when there are changes in the environment, sensitive taxa are lost but lead to an increase of more tolerant taxa (Azrina et al. 2006, Al-Shami et al. 2011).Thus, more tolerant taxa to environmental changes are able to use a variety of habitats, including disturbed sites (Do Amaral et al. 2015).
In the Blyde River, the Shannon-Wiener's diversity index indicated a maximum at the downstream sites with less anthropogenic activities, since these sites are located in a nature reserve, but minimum values in the upstream (with domestic waste from the resort and agricultural activities).On the contrary, in the Mohlapitsi River, the highest Shannon-Wiener's diversity index was at the upstream (MS1 and MS2), with less anthropogenic activities and minimum values in the midstream (MS3 and MS4), where there are human settlement and sand mining activities.The highest diversity index in the downstream of the Blyde River and in the upstream of the Mohlapitsi River shows diversified taxa composition of macroinvertebrates in these parts of the rivers whereas, lowest diversity index in the upstream of the Blyde River and midstream of the Mohlapitsi River is an indication of low diversity of organisms.
Diversity values between 1.0 and 3.0 indicate moderate pollution and values less than 1.0 indicate heavy pollution (Sharma et al. 2016).Thus, BS4, BS5 and BS6 of the Blyde River, and MS1 and MS2 of the Mohlapitsi River fall under moderate pollution.In the Blyde River, the minimum evenness was in the upstream and maximum evenness in the downstream, while in the Mohlapitsi River, the maximum evenness was at upstream sites (MS1 and MS2) and the minimum evenness at midstream sites (MS3 and MS4).The low diversity and evenness in the upstream of the Blyde River and the midstream of the Mohlapitsi River is an indication of stress (pollution) from the anthropogenic activities.Study by Malakane et al. (2020) in the Blyde River found deteriorating water quality in terms of physicochemical parameters in the upstream of the river and had effect on the distribution and diversity of macroinvertebrates.Similarly, in the Mohlapitsi River, a recent study found an impact of deteriorating water quality (physicochemical parameters) in the midstream on the distribution of macroinvertebrates (Raphahlelo et al. 2022).Thus, the low abundance and diversity of macroinvertebrates in non-protected areas could be due to human activities that may be disturbing the habitat or putting additional pollutants in the river.

Structure composition of macroinvertebrates in relation to environmental variables
The correlations between many of the trace elements and the abundance of macroinvertebrates were insignificant in both rivers.This could be due to the fact that the concentrations of the trace elements were lower than the concentrations critical to affect the abundance of most of the macroinvertebrates.In the Blyde River, only two taxa, Philopotamidae (Tricoptera) and Leptophlebiidae (Ephemeroptera) had a significant positive correlation with trace elements.Studies have also found some Ephemeroptera and Trichoptera taxa to be highly tolerant of trace elements (Beasley andKneale 2003, Wright andRyan 2016).It has also been found that some taxa of Ephemeroptera and Trichoptera are among the first groups to recover in rivers contaminated with trace elements (Bere et al. 2016).However, the concentrations of Al, Fe and Ni, Pb were found to have significantly negative effect on the following macroinvertebrate taxa, Oligochaeta (Al) Gyrinidae and Platycnemididae (Fe and Ni), and Aeshnidae (Pb) in the Blyde River.Other studies have found strong negative correlations between trace elements and macroinvertebrate total abundance, species richness, reflecting the negative effects of trace elements on stream ecosystems (Qu et al. 2010).
In the Mohlapitsi River, 12 taxa correlated significantly with increasing concentrations of trace elements.The tolerance of macroinvertebrates including sensitive taxa from Trichoptera, especially Hydropsyche, to trace elements has been well reported (Clements et al. 2000).However, the abundance of the following taxa, Naucoridae (Al), Elmidae and Thiaridae (As), Hirudinae, Psephenidae, Syrphidae, Planorbidae, and Thiaridae (Cu), Naucoridae (Pb), and Lestidae (Zn) were significantly decreased with increasing concentration of trace elements.This shows that different macroinvertebrate taxa respond differently to different levels of trace elements, because of differences in tolerance to contamination (Dallas andDay 2004, Bonada et al. 2006).Despite the fact that macroinvertebrates respond differently to different concentrations of trace elements, more taxa responded negatively to increasing concentrations of trace elements than the taxa that responded positively to trace element concentrations (Qu et al. 2010).Contaminated sediments may pose a threat to the aquatic biota in the benthic environment, exposing them to toxic chemical concentrations.This may cause changes in taxa composition and abundance in disturbed areas with high tolerant taxa richness, and high sensitive taxa richness in areas with less disturbances (Rico-S anchez et al. 2022).

Conclusion
The aim of the study was to assess the effects of trace elements contamination on benthic macroinvertebrate communities in two rivers in the Olifants River System, the Blyde and Mohlapitsi rivers.There were variations in the concentrations of trace elements and distribution of macroinvertebrates in both rivers.Comparisons of the benthic macroinvertebrates among the different sites and between the two rivers suggest that the abundance of Aeshnidae, Gyrinidae, Oligochaeta, and Platycnemididae was significantly decreased by high concentrations of trace elements in the Blyde River, and in the Mohlapitsi River, the abundance of Elmidae, Hirudinae, Lestidae, Naucoridae, Planorbidae, Psephenidae, Syrphidae, and Thiaridae was significantly decreased with increasing concentration of trace elements.Though, most of the taxa were not affected by trace element contamination, a long-term exposure to trace elements could have adverse effects on benthic macroinvertebrate diversity and ecological health of the affected rivers.It is therefore recommended that uncontrolled effluents from human activities should be prevented in order to prevent a further degradation of the rivers in the basin to protect the aquatic biota.

Figure 1 .
Figure 1.Map showing the locations of the sampling sites of the Blyde and Mohlapitsi rivers.

Figure 2 .
Figure 2. (a) Composition of trace metal concentrations in the sediments of the Blyde River and (b) composition of trace metal concentrations in the sediment of the Mohlapitsi River.

Figure 3 .
Figure 3. Non-metric multidimensional scaling (NMDS) ordination for the Blyde River, based on trace element concentrations and macroinvertebrate assemblage data.

Figure 4 .
Figure 4. Non-metric multidimensional scaling (NMDS) ordination for the Mohlapitsi River, based on trace element concentrations and macroinvertebrate assemblage data.

Table 1 .
Mean trace element concentrations (mg kg À1 ) in the sediments of the Blyde and Mohlapitsi rivers with average shale and Canadian Council of Ministers of the Environment (CCME) reference values.

Table 2 .
Results of the abundance, diversity and evenness of macroinvertebrates in the different sites of Blyde and Mohlapitsi rivers.

Table 3 .
Analysis of variance (ANOVA) results of benthic macroinvertebrate distribution among the different sites of the Blyde and Mohlapitsi rivers.