Spatial distribution and ecological risk of heavy metal in surface sediment of Old Brahmaputra River, Bangladesh

ABSTRACT The study was conducted to investigate the concentrations, distributions, sources, and potential ecological risk of heavy metals in surface sediments from the Old Brahmaputra River. Sediment samples were collected from five sampling stations and analyzed with an inductively coupled plasma mass spectrometer (ICP-MS). The results showed that mean concentrations (mg/kg) of heavy metals were in order of Ni (58.82) > Cr (32.45) > Pb (21.04) > Cu (14.24) > Cd (3.81) > As (2.02). Element-specific environmental indices exposed that the ecological risks trailed in the same downward direction as Cd > Ni > Pb > As > Cu > Cr. The contamination factor (CF), enrichment factor (EF), and geo-accumulation index (Igeo) exposed all cases to a modest level of contamination except for Cd, which has a significant level of contamination. Only Ni and Cd were found to pose a high ecological risk for aquatic life based on the sediment quality guidelines. The prospective ecological risk factor and potential ecological risk both suggested low to moderate ecological risk. The river sediments were mostly uncontaminated or mildly contaminated, posing a minimal ecological risk. The study concluded that the Old Brahmaputra River is subjected to anthropogenic disturbances in its aquatic environment.


Introduction
In recent decades, exponential population growth, rapid urbanisation, haphazard industrialisation, and unplanned agriculture have provoked some serious environmental challenges [1][2][3][4]. Heavy metals are becoming a global issue that has been thoroughly investigated in recent times in emerging countries due to their abundance, permanence, cytotoxicity, long-term ecosystem mobility, and negative consequences to aquatic of toxic element pollution in the Old Brahmaputra River sediments were conducted in precise locations such as Jamalpur and Mymensingh (BAU Botanical Garden site) [45] and Narsingdi [50]. A limitation of such research is the absence of data on river flow and sediment quality in terms of heavy metals in the surface sediments of the Old Brahmaputra River in Mymensingh. Thus, there is an urgent need to comprehend the possible sources and ecological impacts of metal toxicity in the sediments of the Old Brahmaputra River in order to formulate appropriate policies for the protection and management of this watershed.

Study area and sampling sites
The study site was in the Old Brahmaputra River of Mymensingh, Bangladesh, with latitudes ranging from 24°15'00 ′′ to 25°15'00 ′′ N and longitudes ranging from 90°05'00 ′′ to 90°50'00 ′′ E ( Figure 1). The Brahmaputra River originates from the Manos Sarowar, located between the Himalayas and the Kailash Range in Tibet, China [14]. The river originates from the left bank of the Brahmaputra to the north of Bahadurabad in Jamalpur district and flows southeasterly (passing by Jamalpur and Mymensingh towns) for approximately 200 km before meeting the Meghna River at Bhairab Bazar in Kishoreganj district [51]. The major types of soils reported from the study area include 'Non-calcareous Dark Grey Floodplain Soils' and 'Non-calcareous Grey Floodplain Soils.' The climate of Mymensingh is a little cooler as it is closer to the Himalayas and sufficient to be a monsoon-influenced humid subtropical climate [47]. For this experiment, sediment samples were taken from five diverse stations located at Mymensingh City Corporation: Zainul Abedin Park (S1), Thanar Ghat (S2), Shamvuganj Bridge (S3), Kewatkhali Boat Ghat (S4), and BAU Botanical Garden (S5). Sampling stations (around 1.5 km from each other) were sensibly chosen from upstream (S1) to downstream (S5) based on the assumed sediment superiority and magnitude of effluence from earlier visits to the river side. The S1 (24°46'11.37 ′′ N, 90°24'00.29 ′′ E) was situated very adjacent to Zainul Abedin Park, and the leading pollution sources at this site mainly include agricultural runoff and waste water from the municipal drainage system. The S2 (24°45'35.44 ′′ N to 90°24'54.58 ′′ E) was dominated by electric power plants and the food and beverage industries, and therefore industrial discharge along with domestic and municipal waste waters were the major pollution sources in this sampling site. The S3 (24°44'58.46 ′′ N, 90°2 5'24.89 ′′ ) is positioned just beside an electric power plant and the food and beverage industries, whereas the industrial discharge from these industries is the core pollution source. The S4 (24°44' 6.71 ′′ N to 90°25'47.22 ′′ E) was covered by residential and agricultural areas; as a result, municipal waste water and agricultural runoff were the main pollution sources at this site. The S5 (24°43'34.80 ′′ N to 90°26'29.50 ′′ E) was located at the Bangladesh Agricultural University campus, and it was severely influenced by residential areas, agricultural land, and a water treatment plant (WTP), and therefore municipal waste water, agricultural runoff, and WTP discharge were significant sources of pollution at this site.

Sample collection and preparation
In July 2021, a total of 30 composite sediment samples were obtained from the river using the standard protocol [52]. A basic random sampling strategy was used to collect samples from multiple places along the river [53]. A portable Ekman Dredge Grab Sampler (20 × 20 × 20 cm) was used to collect samples at depths ranging from 0 to 10 cm. At each site, three composite samples weighing roughly 200 g were extracted. To avoid adulteration from the sampler's metallic portions, the upper 5 cm of each sample was detached from the midpoint of the catcher with an acid-dressed plastic spatula [3]. Every sample was prepared by deliberately combining sediments taken three times at each sampling site [53]. Samples were then instantly packed in acid-rinsed polyethylene plastic bags and put in storage at a low temperature (4°C) prior to sample preparation in the laboratory [28,54,55]. Samples were air-dried for seven days at ambient temperature in a dry, dust-free environment. Using a mortar and pestle, the dry samples were crushed into a tiny grain size and uniform mixture and then sieved through a 2 mm aperture to eliminate organic debris, pebbles, and lumps [56]. The homogeneous powdered samples were then kept in an airtight zip-lock bag in a freezer at 8°C until chemical analysis was performed [57][58][59].

Sample digestion and instrumental analysis
Sediments samples were analysed in the Laboratory of Soils Science, Department of Soil Science, Bangabandhu Sheikh Mujibur Rahman Agricultural University (BSMRAU). Sediments samples analyses were operated according to international standard analytical procedures. Examined solutions had a purity of 99.98% and were acquired from Merck Germ Merck. Ultra-pure HClO 4 and HNO 3 (1:2.5) were used to digest the samples. In a 100 mL beaker (Pyrex, Germany), about 0.5 g powdered samples were taken, and then 15 mL of di-acid mixtures were added to it. The beaker with sample and acid mixture was broiled on a hot plate for 5 h by 130°C up to it stayed 2-3 mL in a beaker. Addition of 5 mL of di-acid mixture and boiling were repeated up to the suspension reached to transparent or bright shaded. Following cooling using deionised water, the content was obtained. Then, digested materials were filtered using a filter paper (Whatman no. 41). The samples were then subjected to analysis for heavy metals with inductively coupled plasma mass spectrometer (ICP-MS, Agilent7500i, USA).

Quality control
For ICP's functions, quality control involved immersing glassware and plastics jars in 5% HNO 3 and drying them prior to usage. All assessments were carried out with deionised water and analytical grade reagents. Analytical reagent blanks and sediment reference materials were created and inserted in each set of 5 sediment samples for assessment to evaluate the precision and accuracy of the substantive methodologies used. This study employed sediment reference materials from the Inter-professional Office for Analytical Studies (BIPEA) in Paris (France). A three-step exploration was carried out. When the observed concentrations of examined heavy metals in the reference material were within two standard deviations on both sides of the average certified values in the lab control chart, the findings of the analyses were approved. To minimise any batch-specific mistakes, sediment extracts were evaluated in triplicate along with verified standard reference (NMIJ CRM 7303-a lake sediment). The results of certified and measured values were Cr (39.1 ± 2.8, 39.5 ± 2.18), Ni (21.8 ± 2.5, 21.9 ± 0.51), Cu (23.1 ± 3.1, 23.3 ± 2.16), As (8.6 ± 1.0, 8.53 ± 0.37), Cd (0.34 ± 0.02, 0.33 ± 0.05) and Pb (31.3 ± 1.1, 31.7 ± 3.11) which showed recovery rate 97-101%. The certified and measured values indicated the good precision of the instrument analysis.

Sediment quality guidelines (SQGs)
Material poisoning, or the potential dangers posed to aquatic organisms by metals poisoned sediments, is frequently assessed employing sediment quality guidelines (SQGs) [3,60,61]. The consensus-based SQGs were produced utilising a variety of ways to obtain documented freshwater sediment-quality criteria [62], and the standards are accessible in Table 1. A threshold effect level (TEL) specifies the pollutant concentrations below which no detrimental impacts on sediment-dwelling life forms are envisaged, and a probable effect level (PEL) identifies the concentration above which prospective deleterious effects on sediment-dwelling life forms can be perceived [63]. The effect range low (ERL) is the lowest dose at which adverse effects are uncommon, but harmful effects are widespread beyond the effect range median (ERM) level [15]. The lowest effect level (LEL) implies pristine to marginally impacted sediments. The bulk of sediment-dwelling organisms should be unaffected by this dose [62]. Sediments are deemed severely disturbed at the severe effect level (SEL), and when this threshold is exceeded, unfavourable effects on the dominance of sediment-dwelling organisms are envisaged [64]. The toxic effect threshold (TET) is a metric for determining how badly contaminated sediments are. Negative repercussions for sediment-dwelling species are projected if this level is surpassed [62].

Geo-accumulation index (I geo )
The I geo is broadly applied in the appraisal of metal contamination in sediments [53] simply matching actual quantities to pre-industrial background levels, and it may be derived employing Muller's algorithm [70]: Where, C n and B n is the measured and geochemical background concentration of inspected metal (n), respectively [67,65]. The factor 1.5 is applied for the probable deviations in background values because of lithological effect [71]. The I geo scale classified as seven classes, i.e. class 0 (I geo ≤ 0): uncontaminated, class 1 (0 ≤ I geo ≤ 1): uncontaminated to moderately contaminated, class 2 (1 ≤ I geo ≤ 2): moderately contaminated, class 3 (2 ≤ I geo ≤ 3): moderately to strongly contaminated, class 4 (3 ≤ I geo ≤ 4): strongly contaminated, class 5 (4 ≤ I geo ≤ 5): strongly to extremely contaminated, and class 6 (5 < I geo ): extremely contaminated.

Enrichment factor (EF)
The enrichment factor (EF) is a useful measure for determining the extent of heavy metal contamination caused by humans. Through using equation below, the EF for every component was determined to assess manmade implications on heavy metals in sediments [3]: Where, (C M and C Al ) Sample is the average content of the tested metal in sediment sample; (C M and C Al ) Background is the background content used as the reference element. The Al, Fe, Ti, Si, Sr, and K are often used as reference elements because of their low occurrence variability [72]. The content of Al was employed as a reference element in sediments in this study since it enables for differentiation across native and enhanced element contents [28,29]. The following grades of enrichment have been recommended: no enrichment (EF , 1); mild (EF , 3<3); moderate (EF = 3 − 5), moderately intense (EF = 5 − 10), severe (EF = 10 − 25), extremely severe (EF = 25 − 50), and extremely severe enrichment (EF . 50) [72,73]. Notably, EF values of <1.5 or <2 specify that the metal is totally derived from crustal ingredients or regular practices, while EF values >1.5 or 2 point out that anthropogenic sources are becoming increasingly important [74].

Contamination factor (CF) and contamination degree (CD)
To explore the contamination level of heavy metal in sediments, CF has been commonly employed by many authors previously [53,73,75]. The CF is the ratio of each metal's measured concentration in the sediment to its background concentration, as determined by Rudnick and Gao [65]. The following formula was used to compute the CF for each metal stated by Håkanson [69].
The CD was determined by the sum of all the contamination factors for all of the elements to disclose the degree of potential toxic metal in sediments [41,69]. In this study, the CD was computed by the sum of the six heavy metals in the sediments of the Old Brahmaputra River. Håkanson [69] categorised four ratings of sediments with respect to CF values e.g. low (CF < 1); moderate (1 ≤ CF < 3); considerable (3 ≤ CF < 6); and very high contamination (CF ≥ 6); and in contrast, Tomlinson et al. [76] testified four grades of sediments based on the CD ranges, i.e. low (CD < 6); moderate (6 ≤ CD < 12); considerable (12 ≤ CD < 24) and very high contamination (CD ≥ 24).

Modified degree of contamination (mCD)
The mCD evaluates every sediment sample using a unique contamination indicator [69]. The mCD can be determined using the techniques below: Where, mCD is characterise the modified contamination degree; n is the number of total metals; i is the ith pollutant; CF i is the contamination factor of ith metals. The mCD is categorised as mCD , 1.5: Nil to very low; 1.5 ≤ mCD , 2.0: low; 2.0 ≤ mCD , 4.0: moderate; 4.0 ≤ mCD , 8.0: high; 8.0 ≤ mCD , 16.0: very high; 16.0 ≤ mCD , 32.0: extremely high; mCD . 32: ultra-high contamination [60].

Pollution load index (PLI)
The PLI assesses cumulative poisoning load at multiple places using different metals in soils and sediments, and provides an assessment of the overall toxicity score within each sample site [56,77,78]. PLI was calculated for all sample sites using the given formula proposed by Tomlinson et al. [76] as the nth root of the product of the content's multiplications: A PLI 0 indicates excellence, whereas a merit 1 designates merely background contaminant intensities and gradual degradation of site integrity [76]. This quantitative indicator provides a quick and easy way to compare the severity of metal pollution, where PLI > 1 point to pollution be present, contrariwise, if PLI < 1 terms as absence metal pollution [11,73].

Toxic unit analysis ( TU S )
Prospective cytotoxic effect of toxic compounds in surface sediments is defined as the sum of toxic units ( TU S ) [39]. The toxic unit (TU), which is calculated as the proportion of the weighted content of toxic components in sediments to the probable effect level (PEL), reflects the intensity beyond which detrimental effects are supposed to happen often.
Where, C M and PEL are the quantified concentration and the probable effect levels value of specific heavy metals, respectively; TU S is the product of toxic units (TUs) for heavy metals in urban river sediments samples. When the aggregate of toxic units (TUs) for all assessed surface sediments exceeds 4, toxic metal toxicity ranges from moderate to severe [53,63].

2.4.8.
Ecological risk factor (E i r ) and potential ecological risk (PER) index In this study, the potential ecological risk index approach developed by Håkanson [69] was employed, which reveals benthic population susceptibility to hazardous substances and illustrates the PER induced through cumulative toxicity [3,11,56,75]. The potential ecological risk coefficient E i r of a particular metal and the potential ecological risk index PER of multi-metals can be determined using the formulae below: In these equations, C i f and T i f is the accumulating coefficient and the toxic-response factor of metal (i). The accumulating coefficient C i f computation equation is as follows: Where, C i m and C i n is the merit of heavy metal concentration in sediments and pre-industrial background contents of particular metal, respectively. Chen and Zhou [79] itemised four grades of ecological contamination extent for instance low (E i r , 40 or PER < 150); moderate (40 ≤ E i r , 80 or 150 ≤ PER < 300); considerable (80 ≤ E i r , 160 or 300 ≤ PER < 600); and very high (160 ≤ E i r , 320 or 600 ≤ PER) ecological risk for the sediments. The goal of this strategy is to acquire a more precise assessment of the probable dangers of heavy metal contamination in sediments at the index level, not merely the quantity of pollution extent [60].

Nemerow's integrated pollution index (NIPI)
An integrated strategy which is compatible with huge value analysing not just work of individual elements but also significance of a component with more severe pollution [11]. The Nemerow's [80] composite index was computed as follows: Where, Cn and Bn is the measured and background content of metals, respectively, and Pi av is the average score of contamination factors of explored metals and Pi max is the maximum contamination factors of single metal in a given sample. NIPI is deliberated as non-pollution (NIPI ≤ 0.7); warning (0.7 , NIPI ≤ 1); low (1 , NIPI ≤ 2); moderate (2 , NIPI ≤ 3); and high level of pollution (NIPI . 3) [3].

Multiple probable effect concentrations quality (mPECQs)
Mean mPECQ is often used to assess the toxicity of sediments when incorporating the cumulative influence of toxicants, and was computed using the formula below [62,81]: Where, Ci and PECi indicate the assessed and probable effect concentration of the ith metal, respectively, and n state the metal quantity. Merits of mPECQ are regarded as three sets: set 1 (mPECQ < 0.1): occurrence of noxiousness is deliberated to be fairly low (<25%) and sediment is viewed to be innocuous; set 2 (1 < mPECQ < 5): projected to be lethal with the deadliness occurrence of 70-75%; and set 3 (mPECQ > 5): sediment is lethal with the likelihood of more than 75% [81,82].

Toxic risk index (TRI)
For the toxic risk assessment of heavy metals in sediments, the TRI technique relies on TEL and PEL influences [25]. The TRI for a single substance was weighed spending the principle below: The subsequent procedure was casted-off to quantify the combined noxious threats of metals in sediments: Where, TRIi signifies the toxic risk index of an individual metal, Ci directs metal content in sample, n is the figure of metals and TRI denotes assimilated toxic risk. TRI can be classified as follows: no risk: TRI ≤ 5; low risk: 5 < TRI ≤ 10; moderate risk: 10 < TRI ≤ 15; considerable risk: 15 < TRI ≤ 20; and very high toxic risk: TRI > 20 [3,83].

Statistical analysis
In this study, various statistical tools including Pearson's correlation (PC) and principal component analysis (PCA) method were employed with the help of IBM SPSS Statistics 20.0 to reveal the relationships between studied heavy metals as well as to identify their plausible sources in the sediments. The PC enables to the measurement of the strength of relationships between pairs of metals [75,84]. The PCA is frequently used to reduce the data while recollecting necessary information to minimise the number of independent variables, i.e. principal components (PCs), that are interpreted to understand mutual relationships among the variables [84]. Principal component analysis (PCA), a multivariate statistical technique was used to identify potential sources of heavy metals in soil samples of three different agroforestry systems. Each measured variable's average value was used for PCA, where each main component's eigenvalue and loading value had to be more than 1 and 0.5, respectively. The spatial distribution maps were prepared using ArcGIS 14.1 software.

Spatial distribution of heavy metal concentration in sediments
Descriptive statistics of detected elements (Cr, Ni, Cu, Cd, Pb, and As) levels in sediments obtained from the Old Brahmaputra River in Bangladesh, as well as their comparison to background values (BV), upper continental crust (UCC), toxicity reference value (TRV), average shale value (ASV), toxic response factor (TRF), and probable effect concentration (PEC), are summarised in Table 1. In sediments, richness of Cr, Ni, Cu, Cd, Pb and As was fluctuated from 23.91 to 41.50, 49.02 to 78.43, 9.53 to 22.09, 0.47 to 4.92, 7.29 to 32.29, and 1.07 to 3.24 mg/kg, respectively. Average concentrations of explored elements in sediments were reduced with the succeeding descendant direction: Ni > Cr > Pb > Cu > Cd > As, and findings revealed that quantities were considerably diverse throughout the study sites. Different metropolitan activities, such as manufacturing releases, various land practices, commercial activities, urban discharges, native trash, resident density, rapid urbanisation activities, mining (excavation/dredging of riverbed), and urban runoff from Mymensingh town, could be significant sources of a wide range of metallic intensities [11,59]. Elevated levels of Ni, Cd, and Pb were perceived to be higher than their respective BV values, and they were found to be nearly half as high when compared to Cr, Cu, and As. Moreover, concentrations of Ni, Cd, Pb, and As were found to be greater than individual UCC values, while Cr and Cu concentrations were found to be lower than UCC values. On the other hand, prominently elevated amounts of Cd and Pb were identified compared to the corresponding ASV standards, and concentrations of the remaining metals were nearly half of their ASV levels. Extensive ranges of metal concentrations were perceived among the sampling stations. This might be due to the geographical, hydrological, and geological differences of the watershed, as well as geomorphological factors including differences in precipitation, land use patterns, and industrial emissions [39,53]. Variance of coefficient (VC) is a crucial way of delivering poisonous metal exposure; low and high VC levels indicate that metal excretions are caused intentionally and naturally, separately [85,86]. The VC levels of assessed metals (Table 1) demonstrated that human activities had a significant impact on metal contamination levels across the locations. The skewness values for Ni and Cu were both more than one and positive, indicating that the dispersion of these metals is right-skewed, with several low numbers and a few larger ones, and the overall mean is higher than the median values (Table 1). Furthermore, the kurtosis values for Ni, Cu, Cd, and As were higher than one and converged to one for Cr and Pb (Table 1). This suggests that Cr and Pb (negative kurtosis values) have less severe dispersion, but Ni, Cu, Cd, and As (positive kurtosis values) have more severe dispersion. When both skewness and kurtosis variables are analysed at the same time, it becomes clear that none of the six heavy metals follow a normal distribution (Table 1). Toxicity reference values (TRV) are used to investigate if trace metals in sediments are harmful to sediment-dwelling organisms [68]. The finding indicated that levels of Cr, Ni, and Cd were many times higher than the relevant USEPA-TRV standards; however, the contents of Cu, Pb, and As were lower than the TRV values. However, toxic response factors (TRF) for Cr, Ni, Cu, and Pb were several times greater, and concentrations of Cd and As were much lower than corresponding TRF values. Furthermore, only the Ni concentration exceeded the particular PEC value, and the rest of the measured metal concentrations were significantly lower than the particular PEC values. Because of the biodiversity risk posed by metals in the sediments of the studied area, some sample areas were deemed to be an environmental threat to varying degrees.
Chromium (Cr): In current exploration, average chromium (Cr) extent was found 32.45 ± 6.95 mg/kg, whereas the highest Cr level was detected in S2 (41.50 mg/kg) and the lowest was perceived in S5 (23.91 mg/kg) ( Table 1 and Figure 2). Cr occurrence in sediments of the Old Brahmaputra River might be due to two factors: (i) natural: an elevated levels of Crbearing elements (Chromite: is a crystallized rock made up mostly of iron (II) oxide and chromium (III) oxide components) [53]; and (ii) artificial: manufacturing events for instance electric power plant, metal plating, garments, cement, paper and pulp, steel and alloy, combustions of fossil fuel (gas, oil, coal) which are squaring Cr grounded oxidants (i.e. chromate and dichromate) [14,56,87]. As a result, waste emitted by these industries would be most likely the source of high Cr levels in disturbed sediments [88]. Cr level in this study was compared to that in previous Bangladeshi and international studies ( Table 2). Cr content in sediments of the current exploration was found lower than the Korotoa, Shitalakhya, Halda, Pasur, Bhairab, Dhaleshwari, Sundarbans River, Turag, Buriganga, Paira, Jamuna, Feni, Bangshi, Bakkhali River in Bangladesh and higher than the Meghna, Old Brahmaputra (Narsingdi), Padma, Karnaphuli, Rupsa, Sangu, Gumani, Dakatia, Louhajang, Gomti River in Bangladesh (Table 2). On the other hand, Cr extents of the current assessment was collated to other country's river and the present study exposed that Cr concentration was lower than the Yellow (China), Louro (Spain) and Gorges (Australia) River and higher than Ganges River (India) ( Table 2). Current study's Cr content was collated to numerous sediment quality guidelines, and it was determined that the Cr level was judged to be lower than the TEL, SEL, PEL, ERL, ERM and TET values and higher than the respective LEL value [62] (Table 1).
Nickel (Ni): In present assessment, average amount of Ni in sediment was determined to be 82 ± 11.50 mg/kg. Highest Ni level was detected at S4 (78.43 mg/kg) and the lowest perceived at S5 (49.09 mg/kg) ( Table 1 and Figure 2). Actually, S4 sampling site was received huge amount of municipal waste from nearby Kewatkhali area. Higher amount of Ni was found might be due to the industrial activities (incineration of waste and sewage, combustion of fossil fuels), domestic waste (steel kitchen utensils, inexpensive jewelries, tobacco smoking), natural process (wind-blown dust, weathering of rocks and soils) etc. [45,88,112]. Present study's Ni amount was greater than the Halda, Old Brahmaputra (Narsingdi), Padma, Rupsa, Dhaleshwari, Sundarbans River, Paira, Sangu, Dharla, Feni, Bangshi and Louhajang River in Bangladesh and lower than the Korotoa, Meghna, Turag, Buriganga and Surma River in Bangladesh (Table 2). Moreover, Ni extent of current analysis was also collated with other countries river and the result found that higher amount of Ni than the Niger (Nigeria), Louro (Spain), and Gorges (Australia) River (Table 2). Current study's Ni content was also evaluated to many sediment quality guidelines, and it was discovered that Ni extents was greater than the LEL, TEL, SEL, PEL, ERL and ERM, and lower than the TET value [62] (Table 1).
Copper (Cu): Result of the analysis revealed that mean copper (Cu) content was observed 14.24 ± 4.76 mg/kg, and while the highest Cu content was recorded at S2 (22.09 mg/kg) and the lowest was detected at S5 (9.53 mg/kg) ( Table 1 and Figure 2). The higher level of Cu was found in the study area which may be the result of anthropogenic activities such as electric product manufacturing/discarding industry, metal plating, pesticides, fungicides, leather processing, domestic waste pigmented materials, discarded cans, packets, diaper etc. [112][113][114][115][116] and inherent process (weathering of ancient construction, metallic proportion in constituent materials and stones, process of dirt creation) [85,117,118]. The comparison study showed that copper concentration in sediments of the current study was higher than the Old Brahmaputra (Narsingdi), Padma, Dhaleshwari and Surma Rivers in Bangladesh and lower than the Korotoa, Shitalakhya, Halda, Rupsa, Sundarbans River, Turag, Buriganga, Paira, Jamuna, Sangu, Dharla, Dakatia, Bangshi, Khiru and Balu Rivers (Table 2). Again, present study's Cu content was collated to diverse country's river and the present study became that Cu concentration was lower than Yellow (China); Louro (Spain); Gorges (Australia) and higher than the Ganges (India) and Chenab (Pakistan) ( Table 2). Current enquiry's Cu extent also checked to several sediment standard quality guidelines and found lower Cu level than the LEL, TEL, SEL, PEL, ERL, ERM and TET values [62] (Table 1).
Cadmium (Cd): Mean abundance of cadmium (Cd) was obtain 3.81 ± 1.89 mg/kg and whereas the highest Cd extent was witnessed at S3 and S4 (4.92 mg/kg) and the lowest concentration of Cd was found at S5 (0.47 mg/kg) ( Table 1 and Figure 2). In sediments of the Old Brahmaputra River Cd quantity was advanced could be linked to both anthropogenic activities as well as farming operations (sendoff phosphate fertilisers and pesticides), textiles wastewater (dying, silk-screening), battery (discarded auto rickshaw's batteries), lead smelting, municipal waste (especially plastic bags and wastes) around Mymensingh city corporation closed to the river [78,95,112], and lithogenic phenomenon (sediment contain Cd naturally) [112]. Current study's Cd amount in sediments was evaluated to previous studies completed in Bangladesh and abroad, and it discovered that Cd extent was greater than the Korotoa, Shitalakhya, Halda, Meghna, Pasur, Bhairab, Old Brahmaputra (Narsingdi), Padma, Karnaphuli, Rupsa, Dhaleshwari, Sundarbans River, Turag, Buriganga, Paira, Surma, Gumani, Dakatia, Bangshi, Khiru, Balu, Bakkhali, Louhajang, Gomti River in Bangladesh (Table 2). On the other hand, current examines Cd contents were collated to other country's river and found upper than the Ganges, India; Chenab, Pakistan; Niger, Nigeria; Louro, Spain (Table 2). Current analysis's Cd extent was also evaluated to several sediment standard quality guidelines and lesser than the SEL and ERM values and upper than the LEL, TEL, PEL, ERL and TET values [62] (Table 1).
Arsenic (As): The mean arsenic (As) content was found 2.02 ± 0.78 mg/kg and the highest As content was detected at S2 (3.24 mg/kg) and whilst the lowest was witnessed at S1 (1.07 mg/kg) ( Table 1 and Figure 2). The human-related actions as well as farming events (rat/ ant poison, weed killers) [116], preservation of wood materials (chromate copper arsenate treatment of wood), chemical waste (manufacturing of alloys, electronic materials, smelling), mining activities (excavation of river bed/dredging) that alters the hydraulic regime as well as fresh water habitat [112,121,122]. Present study's As extents were evaluated with previous explorations directed in Bangladesh and abroad, and found lower than the Korotoa, Halda, Pasur, Bhairab, Karnaphuli, Rupsa, Sundarbans River, Buriganga, Paira, Sangu, Louhajang River in Bangladesh and higher than the Padma, Gumani, Feni, Bangshi, Gomti River in Bangladesh (Table 2). Further, current study's As contents were collated to diverse country's river and found lower than the Yellow (China), Niger (Nigeria), and Gorges (Australia) ( Table 2). Current inquiry's As contents were also collated to some sediment quality guidelines and found that As level was lower than the respective LEL, TEL, SEL, PEL and TET values [62] (Table 1).

Deciphering interrelationships and possible sources of heavy metals in sediments
Pearson's correlation (PC): Pearson's correlation analysis is a useful statistical approach for displaying the level of dependence between two parameters [123,124]. Pearson's' correlation (PC) demonstrated that many of the metals tested had strong association ( Table 3). The Cu-Cr, Pb-Cd, and As-Cu disclosed pointedly substantial positive associations (r = 0.761-0.915, at p < 0.01) within factors and they are 0.755, 0.726, 0.737, respectively, ( Table 3), suggesting that poisonous substances are the consequence of typical unnatural contamination sources in the study region [125]. These examined harmful elements were discovered in sediments as a consequence of the dispersion of manufacturing effluents, commercial waste, municipal waste and agricultural runoff. Comparable inferences were likewise conveyed by Proshad et al. [53] in the Rupsa River, Ali et al. [59] in the Pasur River, Islam et al. [40] in rivers of Sundarban reserve forest, Bangladesh. In contrast, Cd-Cr, Pb-Cr, and As-Cr exposed significant (p < 0.05) positive correlation (0.649, 0.608, and 0.641, respectively), implying that the incidence of Cr is likely to be commercial rather than fertiliser runoff. Nevertheless, there was a negative association between Ni-Cr, Pb-Ni, As-Ni and As-Pb and they are −0.159, −0.023, −0.108 and −0.173, respectively. On the other hand, weak correlation (r = 0.462) was observed among Cu-Ni, Cd-Cu, Pb-Cu and As-Cd, pointing out the industrial activities and municipal wastes might be as the possible sources of them.
Principle component analysis (PCA): The PCA was manipulated to successfully analyse the parameter scores determined in sediment (S1). Three significant components were recognised centred on eigenvalues bigger than one, accounting for 95.89% of the system variation. The variance of the first principal component (PC-1) was 50.903 percent, with an eigenvalue of 3.054, indicating significant positive loading in Cr (0.937), Cu (0.864), and moderate positive loading in As (0.580), as shown in S1. PC-1 might be identified as unnatural origins because to the strong positive loading of Cr and Cu (industrial doings and agricultural runoff). The PC-2 had a variance of 26.748% and an eigenvalue of 1.605, specifying high positive loading in Pb (0.482) and Cd (0.573), as well as moderate positive loading in Cr (0.937). The PC-2 was implying that Pb and Cd came from manmade sources. Industrial discharge and extensive use of fertilisers (phosphate) in agricultural land and their runoff in the river is proposed to root a noteworthy upsurge in Cd intensities [45]. Moreover, Cr was allocated to different components (PC-1 and PC-2). The PC-3 had a variance of 18.239% and an eigenvalue of 1.094, and high loading factors for Ni (0.764) were discovered. In the study area, the unnatural inputs of Ni revealed a mixing source of both paedogenic and manmade composition. However, it appears that manmade sources for instance manufacturing, corporate, household discharges, regional incineration, animal manure and pesticide treatments in the study region have a significant impact Figure 3.

Assessment of environmental and ecological risk of heavy metals
In this study, the geo-accumulation index (Igeo) was employed to estimate the pollution load of possibly harmful components in surface sediments. The Igeo values at five stations in Old Brahmaputra River were typically in the succeeding approach ( Figure 4): Cd (4.4403 ± 1.4787) > Ni (−0.2811 ± 0.2603) > Pb (−0.4463 ± 0.8502) > Cu (−1.6189 ± 0.4489) > As (−1.9192 ± 0.5676) > Cr (−2.1151 ± 0.3122) (Figure 3). The Igeo score for Ni, Pb, Cu, As, and Cr throughout entire locations was in the uncontaminated band, signifying that Ni, Pb, Cu, As, and Cr could have natural origination in the sediments, incorporating geomorphological influences [126]. Most of the Igeo levels for Cd have been determined to be heavily to extremely pollute. Cd levels at their peak may be owing to inputs from air emissions, defused battery effluents, and Cd-plated goods [95]. This is critically relevant because Cd amounts could harm the preponderance of benthic creatures.
The enrichment factor (EF) is a useful metric for estimating pollution levels in the sites [127]. To assess manmade consequences on toxic substances in sediments, the EF for every constituent was determined. An EF score of roughly one implies that a certain element is primarily composed of crustal elements [128,129]. In the present study, Cd had the largest EF average 46.3245 ± 22.9527, and dropped in the succeeding direction of Ni (1.3710 ± 0.2679)> Pb (1.3559 ± 0.6267)> Cu (0.5571 ± 0.1863)> As (0.4614 ± 0.1778)> Cr (0.3864 ± 0.0828) (Figure 4). Average EF scores for Cu, As, and Cr appertain to cluster 1 or no enhanced; for Ni and Pb was defined as a minor enhanced in cluster 2, possibly because of the presence of banging automobile garages, dense traffic, insect killer and fertiliser industry, scuffle supplies and motor vehicle workspaces [74]. On the contrary, EF scores for Cd fall in to very severe enhanced (cluster 6). The study's findings demonstrated that most of the sediments were significantly tainted with Cd. The biggest reasons for this are the potential consequences of manufacturing and battery-powered autos in the research regions [53,130].
The average CF scores of Cr, Ni, Cu, Cd, Pb, and As were 0.3527 ± 0.0756, 1.2515 ± 0.2446, 0.5086 ± 0.1701, 42.2889 ± 20.9532, 1.2378 ± 0.5721, 0.4213 ± 0.1623, correspondingly and declined in direction of Cd > Ni > Pb > Cu > As > Cr ( Table 4). The degree of Cd poisoning was quite significant throughout all sediment samples from the Old Brahmaputra River, although Cr toxicity was negligible. These might be the outcome of large manufacturing releases or metallurgical processing [20]. However, Ni, and Pb, in sediments revealed moderate contamination, whereas Cu and As showed the low contamination. The study indicated that CD scores for all metals across all test sites directed a cumulative contamination caused by the simultaneous presence of all hazardous elements inspected [130,131]. Average CD score was 46.0607 ± 21.6477, putting it in the very high (CD ≥ 24) contamination amplitudes ( Table 4). The average mCD merit for sampling spots SD1 (8.7743), SD2 (8.8194), SD3 (9.7614), and SD4 (9.7489) represented very high extent of pollution; however, SD5 (1.2799) showed evidently nil to very low extent of pollution (Table  4). Various unnatural factors, in combination with paedogenic variables, are solely accountable for greater mCD levels and spatial variability [132]. The PLI is an instrument for evaluating the extent of metal toxicity that uses a set of comparing experimental indices [133]. A PLI score of one (1.00) confirms the presence of only background concentrations of pollutants, but scores more than one reflect sediment poisoning [134]. The extent of PLI was 0.6559-1.6309 depending on the outcomes. The threshold value (PLI > 1.00) was surpassed at all monitoring sites, designating conceivable metal effluence in the sediment ( Table 4).
The aggregate of the toxic units, characterised as the proportion of the calculated quantity to probable effect levels (PELs) merit, could be used to quantify the prospective cytotoxic activity of contaminants in surface sediments [37,62,135]. The S2 shows the toxic unit (TU) and sum of toxic units (TUs) for heavy metals in surface sediments of the Old Brahmaputra River. Toxic units of toxic metals declined in the direction of Ni > Cd > Cr > Pb > As > Cu. Overall toxic unit merits of Cr, Ni, Cu, Cd, Pb, and As in sediments were 0.3606, 1.6339, 0.0723, 1.0874, 0.2312, and 0.1189, individually. Aggregate of hazardous elements in maximum samples collected was less than 4, and only S4 was larger than 4, signifying that metal toxicity to the sediment-dwelling creatures of the study river was moderate to severe [39,136,137].
For the advanced understanding of sediment contamination and accompanying biological danger, both the individual risk factors (E i r ) and cumulative potential ecological risk (PER) are chiefly indispensable [41]. Average E i r scores of toxic metals drop succeeding the direction: Cd > Ni > Pb > As > Cu > Cr. The E i r scores of Cr, Ni, Cu, Cd, Pb, and As were fluctuated from 0.52 to 0.90, 6.26 to 10.01, 1.70 to 3.94, 156.67 to 1640.00, 2.14 to 9.50, and 2.23 to 6.75, separately (Table 5). Cd was the most major contribution to biological danger in the deposits, attributed to the dominance of individual toxicants on the poisoning phase. Nonetheless, risk imputable to Ni, Pb, As, Cu, and Cr were low in all sites of Old Brahmaputra River, Mymensingh. Consequently, Cd's eco-toxicological harm was substantially greater than that of all other metals calculated, and Cd should be a worry for earthly   [72]. Aside from that, the production facility is likely to be a significant generator of Cd [138]. Further, companies and stores that manufacture phosphate fertilisers and pesticides have been discovered to be heavier producers of several toxic components, as well as Cd [139]. Several investigations have concluded that increased Cd levels are driven by human doings [72,139,140]. Additionally, the average PER score of the all-sampling points was reduced from 171.14 to 1661.96, with an average score of 1289.83, suggesting that the studied poisonous metals' overall biological threat threshold was quite considerable. However, result of the study demonstrated that almost all study sites across the study region constituted a substantial risk to the local habitat (Table 5). This paper looked at the Nemerow integrated pollution index (NIPI) of toxic components in sediments. Overall monitoring stations' NIPIs ranged from 3.802 to 39.267, with an average of 30.393, showing that overall total emissions varied substantially among them (Table 6). In general, such data indicated that unnatural events poisoned the sediments [141]. Furthermore, the study's findings suggested that the sediments were significantly contaminated, having greater NIPI scores (≥3). Sites with greater pollution intensities were primarily in zones dominated by industry and extremely crowded transportation [142]. Moreover, globalisation, as well as the dispersion of manufacturing regions, vehicle stopping and garages, and commercial areas, could be to blame for such changes [143]. To estimate sediment quality, multiple potential effect concentrations quality (mPECQs) was analysed using a conventional approach to evaluating the biological prevalence of different contaminants [144,145]. The typical mPECQs score in sediment was 0.4314 (mPECQs < 1), suggesting that the sediment was innocuous and that the occurrence of poisoning was minor (<25%). The TRI values were calculated as 17.86799, 17.52842, 21.19537, 24.1213, and 2.660964 in sediments of S1, S2, S3, S4, and S5, respectively (Table 6). In the sediments of S5, meanwhile TRI score is TRI ≤ 5, there is danger of poisoning. On the other hand, sediments of S1 and S2 were found considerable toxic risk because of their TRI value is 15<TRI ≤ 20. In the meantime, S3 and S4 were showed very high toxic risk due to their TRI value is TRI>20. Arsenic contributes the most to TRI, trailed with Cu and Pb, correspondingly [83] (Kukrer et al., 2019). The modified hazard quotient (mHQ) is a way of calculating and grading the amount of surpassing for each single element by comparing the amounts of specific PTEs to sediment integrity ideal limits. The mHQ values (S3) at five stations in Old Brahmaputra River were mostly in the following manner: Cr (8.11 ± 1.74), Ni (2.94 ± 0.57), Cu (0.28 ± 0.10), Cd (3.96 ± 1.96), Pb

Conclusions
The sediments of the Old Brahmaputra River include six trace metals that are now well characterised in terms of their elemental distribution and ecological importance. The contamination situation, elemental dispersion, origin attribution, and significant environmental threats aspects are summarised as elemental profusion of Cr, Ni, Cu, Cd, Pb, and As being reasonably advanced than the acclaimed levels, with Cd and Cr achieving the highest and lowest enhancements, respectively. Descriptive statistics (coefficient of variation: 19.54-49.55%) definitely demonstrate that observed constituents were distributed in a nonuniform manner. Similarly, IDW maps revealed that the abundance of assessed metals is significantly greater in places with high loadings of industrial operations, economic and urban waste, chemical waste, and rearming runoff. Farming actions and manufacturing acts, rather than natural progressions, governed the elements' enrichments. According to the research findings, the following recommendations could be implemented to address the Old Brahmaputra River's existing problems: identify illegal industries set up on the river's bank; mandate the installation of effluent treatment plants for all industries; strict legislation for dumping wastewater into the river; establish a new regulatory authority for the Old Brahmaputra River; and initiate intensive research for environmental quarrelling.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
Sincere appreciation to the University Grants Commission (UGC) of the People's Republic of Bangladesh and the Research Cell of the Mawlana Bhashani Science and Technology University for the financial support (RC_FN_21/22-28) to carry out the research works efficiently and successfully.

Notes on contributors
Dr. Md. Sirajul Islam, a renowned Professor of Environmental Sciences. His research interests revolve around aquatic resource management including water pollution and remediation, wastewater treatment, environmental impact assessment, biodiversity conservation and climate change. He has developed an interest in multidisciplinary research on complex issues relating to toxic metals and its effects on the micro-environment. He has more than hundred publications in reputed journals in home and abroad, and presented many articles in national and international workshops, seminars and conferences. Rahiatul Jannat, obtained her undergraduate degree from the Department of Environmental Sciences and Resource Management (ESRM), MBSTU, Bangladesh. Currently, she is doing her MS in Environmental Science at the same department. She is interested in water quality, wastewater treatment and monitoring.

Rifat
Md. Humayun Kabir, serving as an Assistant Professor at the Department of Environmental Sciences and Resource Management, MBSTU, Bangladesh. He is skilled in environmental monitoring and assessment with a special focus on pollution and ecological risk assessment. He also trying to explore biogeochemical processes and the fate and transport of emerging chemicals that impact human and environmental health.
Dr. Md. Saiful Islam, a professor in the field of Soil Science. His research interests' natural resource management including water pollution and remediation, wastewater treatment, environmental impact assessment, biodiversity conservation and climate change.