Heavy Metal Contamination in Forest Reserved Soils Crossed by Roads, its Ecological Risks, and their Effects on Tree Biomass Stocking Potential

Metal contaminants such as Cadmium, Chromium, Nickel, and Lead are released and deposited in Reserved Forest soils as a result of heavily travelled roads. Their pollution interrupts the biogeochemical cycle in the natural environment, affecting plant productivity. However, this pollution's source, their ecological risks, and its effects on tree biomass productivity have yet to be examined. In order to examine this, an ecological study was conducted in two Assam Reserved Forests that are crossed by the National Highway (NH-15). Several ecological risk indices were used to assess potential ecological risks. Metal impacts on tree biomass stocks were predicted using regression analysis and Pearson's coecient. The results showed that metal concentrations in soil samples collected near roads were much higher than those away from roads. However, the overall mean concentration was within the Indian guidelines. Indices of soil contaminations and pollutions ranged from mildly contaminated to highly contaminated and from low polluted to highly polluted soils. The Cd (88.97%), and Pb (52.61%) were revealed to be highly the main contaminating and sources of pollution and ecological danger in the surface soils. The strongest Pearson correlation coecients between Cd-Cr (94%), Cd-Ni (74%), Cd-Pb (97%), Cr-Ni (88%), Cr-Pb (90%), and Ni-Pb (76%) suggest that metals are very comparable. While the strong negative relationships between tree biomass stock and metals, implying that metals are vital factors affecting tree biomass productivity. Thus, conservationists, ecologists, and policymakers must devise effective mitigation strategies for vehicular emission and car discharge caused by trac passing through reserved forests. Cr-Ni (88%) Cr-Pb (90%) and Ni-Pb (76%) suggesting the common source. metals demonstrated a negative correlation with tree stocking potential: TB-Pb (-80%), TB-Ni (-79%), TB-Cr (-76%), and TB-Cd (-71%), showing that metals had a considerable impact on tree stocking and productivity potential. for pretreatment and analysis. The soil samples were air-dried in a dust-free environment for 2 weeks at room temperature (25 °C), (Singh et al. 2015; Ngaba and Mgelwa 2020), ground with a wooden pestle, and mortar to break soil lumps. The soil samples were sieved through a 2 mm and homogenized in the laboratory (Singh et al. 2015; Ngaba and Mgelwa 2020), then stored in zip-lock polyethylene bags in desiccators for further digestion and analysis. using multiple The signicant in using an independent t-test at a signicance level of 0.05. All statistical analyses performed using the SPSS Software package (ver. 20.0; SPSS, Chicago, IL). Index (Igeo) and pollution index (PI) values for the metal contaminants studied. In Bhomoraguri RF, the Igeo values for Cd, Cr, Ni, and Pb, Ni were 0.86, -4.44, 0.87, and 2.79, respectively. While at Balipara RF, Cd, Cr, Ni, and Pb values were -2.09, -3.82, -1.40, and 1.46, respectively. In all of the forests tested, the Igeo values of the assessed metal pollutant along highways varied signicantly, Bhomoraguri RF, with mean values of 5.31, 3.25, 13.21, and 0.09, respectively. In Balipara RF, however, there was a slight pollution for Cd (1.59) and Ni (1.28) as well as a substantial pollution for Pb (8.04). These ndings imply that Pb, Cd, and Ni contribute signicantly to environmental pollution. In Bhomoraguri RF, the results of the ecological risk index revealed that Cd was the most signicant contributor to ecological risk, with a mean concentration of 88.97, followed by Pb (Ei = 52.61), Ni (Ei = 14.08), and nally Cr (Ei = 0.14). Meanwhile, for Cd, Cr, Ni, and Pb, the ecological risk mean values in Balipara RF were 1.02, 0.02, 0.22, and 2.31, respectively. According to Hakanson's criteria of 1980 for ecological risk assessment (if 80 ≤ Ei < 160) and if 40 < Ei ≤ 80), Cd (88.97) and Pb (52.61) were found in the categories of considerable potential ecological risk and moderately potential ecological risk, respectively, in the analyzed soil samples in Bhomoraguri RF. Cr and Ni, on the other hand, were classied as having a low ecological risk. The aggregate mean value of the examined soils in Balipara RF, on the other hand, falls within the category of low ecological risk. Although the overall amounts of metals in Balipara RF were classied as low ecological risk, this result shows that Cd and Pb were possible polluting and ecological risk factors in the examined forests. The independent t-test found a statistically signicant difference between the two forests analyzed. Apart from road-related discharge and emission, the large variations in potential ecological risks reported in Bhomoraguri RF versus Balipara RF could be attributable to the ongoing bridge construction project and the regular outow of the Brahmaputra around the Forest Reserve.


Introduction
The discharge of metal pollutants from automotive lubricants, vehicular emissions, and frequent road repair in heavily and quickly-traveled National highways that pass through Reserved Forests (RFs), all contribute to the accumulation of metal pollutants in surface soils (Rai et al. 2014; Ngaba and Mgelwa, 2020). This can be a valuable indicator of pollution levels and impacts on the environment , which can signi cantly impact plant productivity, particularly tree biomass stocking potential (Siddiqui et al. 2020). Metal deposition in many components of the environment, including soils, water, wetlands, and air, is of global concern today (Shi et al. 2019; Adimalla 2020; Dutta et al. 2021). Metal pollutants, such as nickel (Ni), iron (Fe), manganese (Mn), zinc (Zn), copper (Cu), chromium (Cr), arsenic (As), mercury (Hg), lead (Pb), and cadmium (Cd), has become a serious global concern (Carvalho et al. 2020; Singh et al. 2020). Their contamination of many elements of the environment and ecosystem disrupts the natural biogeochemical cycle (Borah et al. 2018; Kumar et al. 2019; Guarda et al. 2020). They are considered the most dangerous because of their non-biodegradability, ecological risks, toxicity, biogeochemical recycling, extended biological half-lives, and persistence (Bhatia et al. 2015; Sharma et al. 2018; Kaur et al. 2020), and some are poisonous even at extremely low doses. They can affect plant growth and alter their productivity (Sharma 2017;Singh et al. 2020). However, extensive ecological studies to assess levels and ecological risks of metal contaminants, particularly in high-tra cked roadways, in connection to plant productivity are rare in the literature.
Highways that pass through protected forest ecosystems such as National parks (NP), Wildlife sanctuaries (WLS), Biosphere reserves (BRs), and Reserve forests (RFs) have been identi ed as signi cant polluters (Gowd et al. 2010). They are also the source of other metals in the environment (Sarma et al. 2017;Devi et al. 2019; Ngaba and Mgelwa, 2020). The principal sources of heavy metal pollution in urban forestry are road tra c density patterns, automotive parts and their exhaust discharges, boiler and emission, tire wear and tear, and brake lining wears (Rai et al. 2014; Kaur et al. 2020). Due to high levels of trace metals in materials and chemicals used, activities such as road or bridge renovation, construction, and maintenance along forest habitats lead to metal deposition in soil surfaces (Bhatia et al. 2015; Siddiqui et al. 2020; Dutta et al. 2021). This has become a severe worry in many developing countries, with an alarming increase (Hu et  On the other hand, road construction and maintenance are among the most common sources of metal deposition and natural landscape change (Borah et al. 2018). This alters the physical environment and causes edge effects that last well beyond the building period of the road (Trombulak and Frissell 2000). According to studies, contamination of soils and plant parts is higher near high-tra c roads and decreases within 20 meters or can occur up to 200 meters away from the roadsides (Trombulak and Frissell 2000). According to (Sarma et al. 2017), vehicle-induced metal pollution changes leaf epidermal characteristics and leaf pigment concentration, all of which impede plant photosynthesis. Plants exposed to metals for an extended period may have reduced leaf growth and CO 2 assimilation activity (Khanam et al. 2020). Similarly, (Singh et al. 2020) pointed out that metal concentrations in surface soils and plant tissue caused by vehicular movements harm photosynthesis, transpiration, and productivity. Dust is mobilized and distributed by heavily travelled roadways, and when it settles on plants, it hinders photosynthesis, respiration, and transpiration, causing bodily damage on plant structure (Trombulak and Frissell 2000;Devi et al. 2019). Metal contaminants in soil are causing increasing concern around the world. Their toxicity, as well as their introduction into the food chain and the hazards that come with it (Adhikari and Bhattacharyya 2015). Metals accumulating in the terrestrial ecosystem beyond speci ed limits can harm soil hydrology and biota ; Guarda et al. 2020). They can cause chemical interactions that have synergistic effects on plant productivity, and as a result, their toxicity poses a risk to humans via the food chain (Fajardo et al. 2020). They are sensitive markers of pollution and a predictor of mineralogy and soil fertility (Adhikari and Bhattacharyya 2015). Even in low quantities, some heavy metals, in particular, may be exceedingly harmful to the environment. These activities alter soils' physical and chemical qualities, resulting in changes in the soil's natural behavior. As a result, the quality of groundwater may worsen, and the development pattern, morphology, and metabolism of microorganisms and the ecosystem's nutrient recycling mechanisms (Dutta et al. 2021). Metal concentrations have a big impact on physical properties like soil density, water holding capacity, organic carbon content, cation exchange capacity, texture, pH, and electrical conductivity in every environment (Rai et al. 2014; Trombulak and Frissell 2000). The worst-case scenario is that these harmful substances penetrate the food chain, causing direct health effects on individuals ).
Metal pollutants come from a variety of sources, which can be classi ed as either natural (lithogenic inputs via geochemical and chemical processes) or anthropogenic (urbanization, industrialization, agricultural expansion, livestock, dumping), (Alloway 2013; Adhikari and Bhattacharyya 2015; Devi et al. 2019; Khanam et al. 2020). The soil is the principal sink for all of these sources, as it is the most major natural potential sink for a variety of essential and non-essential metals (Adimalla 2020;Ng et al. 2016). Essential metal pollutants are those that living organisms require in trace amounts to support their metabolic functions (e.g., nickel (Ni), iron (Fe), manganese (Mn), zinc (Zn), and copper (Cu), whereas non-essential metals (e.g., chromium (Cr), arsenic (As), mercury (Hg), lead (Pb), and cadmium (Cd)) are not required for living organisms' growth (Ng et al. 2016). These metals are expected to provide a direct and long-term threat to the ecosystem if they are present in quantities higher than naturally occurring values. As a result, metal has gotten a lot of attention worldwide, and quantifying the accessible amounts in various ecosystems is seen as an essential undertaking. Metals that have accumulated on the surface of soils have become a severe problem (Sharma et al. 2018). They go through metabolic, geochemical, and chemical processes, which naturally enrich the soil with varying degrees of toxicity (Adhikari and Bhattacharyya 2015). The ecosystem's metal concentrations may be tolerated by the ecosystem at low levels but can become detrimental at larger levels (Gowd et al. 2010).
The United States Environmental Protection Agency (USEPA) categorized metals according to their toxicity priority (USEPA 1999; Shi et al. 2019). Metals including arsenic (As), cadmium (Cd), chromium (Cr), nickel (Ni), and lead (Pb) are known to be highly poisonous when it comes to their potential effects on the environment and ecology (Protano et al. 2014). Several studies have been carried out to determine the levels of metal contaminants and the ecological threats they pose in India's various ecosystems and land-use regimes (Chetia et  In India, the contribution of heavily used highways passing through urban Reserved Forests on heavy metal emission, ecological risk, and in uence on tree biomass stocking potentials are seldom observed. Therefore, we conducted an experiment in two RFs of Tezpur town, namely Balipara and Bhomoraguri, in Assam, a northeastern Indian state, to: (i) delineate the level of four metal pollutants (Cd, Cr, Ni, and Pb) in surface soil; (ii) quantify their ecological risk levels using various indices; and (iii) predict the effects of metal pollutants on tree biomass productivity and stocking potential. To achieve these objectives, we compared responses at control sites (about 200 metres away from the roadside within the RFs) with little or no exposure to vehicular emissions.

Materials And Methods
Description and selection of the study sites The present study was conducted in two RFs of Tezpur town in Assam, namely Balipara and Bhomoraguri. The National Highway No. 15 (NH-15), which connects all of the state's major cities and is one of the busiest highways for motorized freight transit, runs right via the two RFs throughout the year. The highway is known for its high tra c density and regular movement of heavy-duty vehicles from all northeast India. The two RFs of Balipara and Bhomoraguri were chosen due to the presence of a National Highway passing through, and other activities such as an ongoing bridge construction project and frequent river over ow caused by the Brahmaputra River. The study locations are depicted in detail in Fig. 1.
The two RFs, are found near Tezpur town in the Sonitpur district of Assam. The district is one of the state's 33 administrative districts, located in the north of the central Brahmaputra valley, between 92° 16' E and 93° 43' E longitudes, and 26° 30' N to 27° N latitudes, with an average elevation of 70 to 75 meters above mean sea level (Nath et al. 2013). RFs cover around 17.57 % of the district, covering an area of 935.38 square kilometres (Baruah et al. 2007). The district's northern and southern limits are bounded by the Himalayan foothills and the Brahmaputra River (Nath et al. 2013). The district is located in a subtropical climatic zone with a monsoon climate. Summers are hot and humid, with temperatures ranging from 7 to 36°C on average (Saxena et al. 2014). The rainy season starts early in April, with an annual average plumage of between 170 and 220 cm (Nath et al. 2013), in uencing the region's climate (Baruah et al. 2007). The rain falls heavily, a blessing and a curse for the people (Baruah et al. 2007;Nath et al. 2013).

Sampling and sample collection
Soil and tree samples were obtained at two sampling sites (near the roadside and away from the roadside) in the RFs. To compare the answers, the sites away from the roadway was considered as a control site, its samples were taken at a distance of about 200 meters away from the roadside (adopted from Trombulak and Frissell 2000;Singh et al. 2020). The 200-meter distance was deemed su cient to be free or minimally exposed to automobile exhaust and other road tra c metal discharges (Trombulak and Frissell 2000). The same scenario was used to evaluate characteristics related to tree biomass stocking. For estimating biomass stocks, a total of 24 dominant tree species with nearly identical height and girth were randomly selected from each site. Reagents and chemicals quality Analytical grade chemicals and reagents with a high spectroscopic purity of 99.9% were utilised without further puri cation throughout the laboratory analysis. For reagent dilutions, standards, and sample preparation, double distilled water was employed. Calibration curves were created by diluting benchmark stocks to create standards. Blanks were run regularly to verify that the analysis was of high quality, and washings were supplied at regular intervals (Sharma et al. 2018). Each sample was analyzed three times, and the average values were reported (Adhikari and Bhattacharyya 2015).

Geo-accumulation index (Igeo)
The geo-accumulation index (Igeo) is a quantitative parameter approach for determining heavy metal contamination in soil. Muller rst launched it in 1969 (Adimalla 2020). It assesses the extent of metal pollution in soils (Ngaba and Mgelwa 2020) and allows us to compare current and preindustrial levels (Gowd et al. 2010). The index can also be used to identify lithogenic impacts as a source of metal pollution (Shi et al., 2019). Eq. 1 can be used to calculate the geo-accumulation index (Igeo).
where C n denotes the measured metal pollutant concentration in soil, and B n denotes the metal concentration in the geochemical background (i.e., unpolluted sample or in natural forest). The levels of metal accumulation used in this study were: (i) practically uncontaminated (if Igeo ≤ 0); (ii) uncontaminated to moderately contaminated (if 0 < Igeo < 1); (iii) moderately contaminated (if 1 < Igeo < 2); (iv) moderately to heavily contaminated (if 2 < Igeo PLI = n C r n 1 × C r n 2 × C r n 3 × C r n 4 ⋯ × C r n n Where C r is the contamination factor of an individual pollutant and n is the number of metal pollutants that have been analyzed. According to Tomlinson et al. (1980), the value of PLI was divided into seven categories: (i) slight contamination (if PI < 0.25); (ii) moderate contamination (if 0.25 ≤ PI < 0.5); (iii) severe contamination (if 0.5 ≤ PI < 1); (iv) slight pollution (if 1 ≤ PI < 4); (v) moderate pollution (if 4 ≤ PI < 8); (vi) severe pollution (if 8 ≤ PI < 16) and (vii) excessive pollution (if PI > 16).

Nemerow pollution Index (PI N )
The nemerow pollution index (PI N ) was developed to assess the total metal pollution level in surface soils (Hu et al., 2013). It can be calculated as: Eq. 5. It measures the overall contamination degree of soils induced by the overall status of all heavy metals detected.

Enrichment index (Cp)
Only if the mean concentration of a substance exceeds the preindustrial reference value for the substance under examination is it classi ed as contaminated or enriched (Kumar et al., 2018). The contamination factor (Cf) and degree of contamination (Cd) established by Hakanson (1980) were used to measure the enrichment index (Cp) in surface soil as a result of heavy metals. The contamination factor (Cf) is the proportion of each determined metal pollutant concentration in soils to the speci c heavy metal's geochemical background. The total of all contamination factors for all metals analyzed shows the degree of environmental contamination (Cd) (Hakanson 1980). Thus, Eqs. 6 and 7 can be used to calculate the potential Cf and Cd, respectively.
Where C o is the mean concentration of an individual metal (e.g., Cd, Pb, Cr, Pb) and C n is the substance's preindustrial reference value (or natural forest concentration of the individual metal. The values of contamination factor (Cf) were classi ed as: (i) low contamination factor (if C ≤ 1); (ii) moderate contamination factor (if 1 ≤ C < 3); (iii) considerable contamination factor (if 3 ≤ C < 6); and (iv) very high contamination factor (if C ≥ 6). While the degree of contamination (Cd) was divided into four categories: (i) low degree of contamination (if Cd < 8); (ii) moderate degree of contamination (if 8 ≤ Cd < 16); (iii) considerable degree of contamination (if 16 ≤ Cd < 32); and (iv) very high degree of contamination (if Cd ≥ 32), (Hakanson 1980).

Ecological risk index (RI)
The ecological risk index (RI), which comprises the aggregate of risk indicators, was used to assess the overall possible ecological risk (Hakanson 1980 Where: BGTbelowground biomass; TB-total biomass; TCtree carbon.

Statistical analysis
The levels of physicochemical characteristics and metal pollutants in surface soils were represented using descriptive statistics. For parametric tests, data were standardized. The Shapiro-Wilk and Levene tests, respectively, were used to verify normality and homogeneity. A one-way ANOVA was used to see if the values of soil characteristics, metals, and tree stocks differed signi cantly. To see if the differences between the means were statistically signi cant at p = 0.05, the Fisher's Least Signi cant Difference Test (LSD) was used. The correlations and patterns of similarity between metals, soil physico-chemical parameters, and tree biomass stocks were discovered using Pearson's correlation coe cient and hierarchical cluster analysis. The Ward's and Euclidean distance methods were used to make a dendrogram. Empirical models for predicting the effects of metal pollution on tree biomass stocks were developed using stepwise multiple regression analysis. The signi cant differences in metal pollution levels between roadside and away from roadsides (control) were detected using an independent t-test at a signi cance level of 0.05. All statistical analyses were performed using the SPSS Software package (ver. 20.0; SPSS, Chicago, IL).

Results And Discussion
Physico-chemical properties of soil Where E i denotes the risk factor, T i denotes the toxic-response factor for a speci c metal, and f i denotes the pollution index. [ ] ( ) Within sampling sites, the arithmetic means for a few physicochemical parameters (pH, electrical conductivity (EC), and soil organic carbon (SOC) as well as the available nutrients (nitrogen (N), phosphorus (P), and potassium (K) differed signi cantly (F (10, 610) = 1039.82, p < 0.000). Table 1 shows the descriptive statistics of the selected soil physicochemical parameters (minimum, maximum, arithmetic mean, and standard deviation (Mean ± SD)). The mean ± SD values from both sampling sites (control and along roadsides) in Bhomoraguri RF were: 6.140. 16  For Bhomoraguri and Balipara RF, the independent t-test revealed that the two sampling sites from each forest were statistically signi cant at t = -0.57, df = 85.29, p < 0.000; and t = 40.39, df = 94.68, p < 0.000, respectively. In samples taken from control sites, the pH of the soil was determined to be strong to moderately acidic (5.27 and 6.14), while it was neutral to basic (6. 27  Contrary to popular belief, available P levels were higher in soil samples taken from roadsides (12.90 and 18.19 mg kg -1 , respectively) than in the control sites (2.94 and 14.68 mg kg -1 ). The main source of worry about the decreased available P in the control site might be attributed to the high uptake and xation of P by the dominant forest tree species (Tectona grandis), which used and stored a large amount of P. Not only that, but the quick utilisation of iron and aluminium by huge trees and regenerating plants, as documented by (Shrestha and Ka e 2020), could have had a part in the reduction of available P in the samples from the control sites. Furthermore, high P availability in soil samples collected near roadsides could be attributed to vehicular discharge of P-related compounds.

The concentration of metal pollutants in surface soils
The present investigation found a signi cant difference in metal pollutant concentrations between the two sampling sites (control and along roadsides) at the 95 % level (t = -33.65, df = 58.37, p < 0.000). The metal pollutant evaluated in soil samples collected from two places within the study forests was positive ( The mean concentrations of metal contaminants in surface soil were in the sequence Pb>Ni>Cr>Cd in all of the forests tested. However, for all metals studied, the mean concentration was higher in Bhomoraguri RF than in Balipara RF. This could be owing to the ongoing large-scale bridge construction project along the Brahmaputra River, which runs through the forests. As a result, there are a variety of road diversions in the area, all of which have a serious concern for tree removal to maximize the number of vehicles passing the RF. The creation of these temporary road diversions results in a signi cant shift in road geometry and vehicle speed. Devi et al. (2019) said that road geometry has a direct impact on vehicle deceleration and acceleration, hence controlling metal pollutant load discharge and emissions from vehicular engines, as well as wear and tear.
Despite considerable differences in concentration levels among the examined metal pollutants (Cd, Cr, Ni, and Pb), the mean concentrations were determined to be below-allowed limits when compared to the Indian standards guideline for natural and roadside soils (Bhatia et al. 2015; Kaur et al. 2020). According to Adimalla (2020) and Kaur et al. (2020), the majority of the examined metals were even lower than the safe level of other international guidelines. For example, the mean Cd concentration along the roadside in Bhomoraguri and Balipara RF was 2.97 and 0.53 mg kg -1 , respectively, although the permitted Cd limits in soil for natural and roadside soils vary from 3 to 6 mg kg - it was di cult to establish accurate comparisons from the present study's ndings due to a lack of similar studies from the region. However, some metal pollution concentration studies from various land uses in the region were compared for justi cation (Table 3).  ranging from uncontaminated to moderately contaminated, and moderately contaminated to heavily contaminated levels. However, all soil samples collected from the control site (away from roads) exhibited uncontaminated levels (i.e., Igeo ≤ 0). Pb (2 < Igeo < 3) was found to have the highest environmental contamination in the surface soils by metal contaminants within sites, whereas Cr and Cd had the lowest levels. The high concentrations and abundance of Pb and Ni in surface soils around roads could be accredited to automotive emissions and automobile discharges due to periodic road repairs and construction. According to Tomlinson et al. (1980), the overall site characterization using the pollution load index (PLI) and the integrated pollution index (IPI) revealed that the examined sites (roadsides) in Bhomoraguri and Balipara RF were severely contaminated and highly polluted, respectively. With mean values of 1.43 and 0.69 for Bhomoraguri and Balipara RF, respectively ( Table 6) Tree biomass stocking potential A total of 24 dominant tree species were quanti ed for biomass (i.e., 12 trees from each forest) to assess the impacts of the examined metal pollutants (Cd, Cr, Ni, and Pb) on tree biomass stocking and production potential in the forests. From the control site and along roadsides of Bhomoraguri RF, totals of 111.05 Mg ha -1 with an average of 9.25 ± 1.09 stocking per tree (Mg tree -1 ) and 97.41 Mg ha -1 with an average of 8.12±

Evaluation of metals pollution and Ecological risk assessments
1.12 Mg tree -1 were reported, respectively (Table 7). These results equated to 203.77 Mg ha -1 and 168.02 Mg ha -1 CO 2 sequestration by tree species.
In Balipara RF, 98.76 Mg ha -1 was recorded, with an average of 8.23 ± 1.90 stockings per tree (Mg tree -1 ), and 89.82 Mg ha -1 with an average of 7.49± 1.85 Mg tree -1 . The CO 2 equivalent value was estimated to be 181.23 and 154.93 Mg ha -1 . Tectona grandis L.f. (19.18 Mg ha -1 ), Artocarpus integer (Thunb.) Merr (12.18 Mg ha -1 ), Dalbergia sissoo Roxb (11.87 Mg ha -1 ), and Ficus carica L (9.40 Mg ha -1 ) were the tree species that produced the most biomass and carbon stocks in Bhomoraguri RF. The dominant biomass stocks in Balipara RF, on the other hand, were Ficus carica L (25.20 Mg ha -1 ), Mimusops elengi L (17.09 Mg ha -1 ), and Stereospermum chelonoides DC (9.99 Mg ha -1 ). For Bhomoraguri and Balipara RF, the topmost biomass and carbon accumulator were found in Tectona grandis L.f. and Ficus carica L., respectively. Despite the slight difference in biomass stocking between the control site and the site along roads, an independent sample t-test revealed that the difference was not statistically signi cant; t = 0.476, df = 60, p = 0.636; and t = 0.298, df = 82, p = 0.766, for Bhomoraguri and Balipara RF, respectively. According to the study, metal pollutants may negatively affect tree biomass stocks and productive potential.
Correlation between physicochemical properties, metal pollutants, and tree biomass stocks A Pearson's correlation coe cient (r) and hierarchical cluster analysis (CA) were used to determine the degree of a relationship and similarity between metal pollutants, soil physicochemical parameters, and tree biomass stocks. Table 8 and Fig. 2 illustrate the results of the matrix for these variables. The soil metals were shown to be correlated with physicochemical parameters (pH, EC, and SOC), as well as available soil nutrients (N, P, and K) at p < 0.000. A strong positive association was discovered between pH and metals (Cd (95%), Cr (93%), Ni (80%), and Pb (98%)). In contrast, a weak positive correlation was observed between EC and metals (Cd (18%), Cr (12%), Ni (5%), and Pb (20%)). According to the ndings, soil pH has identi ed, each with its own set of features. Cr, Pb, Ni, Cd, and EC belong to the rst cluster. The pH, available N, SOC, and available P are found in the second cluster, while available K and tree biomass stock values are found in the third. Cluster one was divided into two sub-groups: Cr, Pb, and Ni in one, and EC and Cd in the other. Cluster two was likewise broken into two sub-groups: pH and available N in one, and SOC and available P in the other. The correlation coe cient was found to correspond with the groupings of soil parameters, metal pollutants, and values for tree biomass stocks. This connection grouping supports the idea that the metal contaminants under investigation have common sources of enrichment in surface soils, which could be substantially linked to automotive emissions in this situation.
Predicting the effects of metal pollutant concentrations on tree biomass stocking potential A stepwise multiple regression analysis was done to estimate the impacts of metal pollutants (Cd, Cr, Ni, and Pb) on tree stocking potential ( Table  9). The ndings revealed that these metals

Conclusion
The following ndings could be drawn from the present study: i. In the surface soils near the highways, there is a clear indication of metal pollution enrichment. Although the mean metal pollutant concentrations (Cd, Cr, Ni, and Pb) were within the permitted levels of the Indian standards guideline for natural and roadside soils, However, soil samples obtained near roadsides had higher concentrations of these metals (e.g., 137.83 mg kg -1 for Pb) than those collected in control locations (10.27 mg kg -1 for Pb). The ndings show that the National Highway (NH-15), which runs through the RFs of Bhomoraguri and Balipara, considerably impacted metal levels in the surface soils. As a result, the primary sources of these metals were automotive lubricants discharged owing to road repair and construction, break tears and vehicular emissions.
ii. Apart from the pollution caused by fast-moving tra c, the large ongoing project of bridge construction across the Brahmaputra River may have facilitated the enrichment of metal pollutants because the construction has resulted in the formation of several road diversions around it, all of which have a great concern for tree clearance to increase the number of vehicles traversing the forests. The creation of these temporary road diversions results in a signi cant shift in road geometry and vehicle speed. As a result, the vehicle's acceleration is affected, which increases metal pollutant load discharge and emissions from vehicular engines, as well as wear and tear. The adsorption and mobility of these meta pollutants in surface soils may increase soil toxicity for biological systems, lowering plant species production.
iii. The soils in this study were classi ed as uncontaminated, moderately contaminated, and heavily contaminated based on geo-accumulation index (Igeo) values; results on pollution index (PI) values for the investigated soils revealed three categories (i.e., low, medium and high polluted category  contaminants have similar sources of enrichment in surface soils. The four metal contaminants that were discovered to be strongly negatively linked with tree biomass were found to be. The correlation coe cient matrix was in the following order: TB-Pb (-80%), TB-Ni (-79%), TB-Cr (-76%), and TB-Cd (-71%). According to regression analysis models, metals (Cd, Cr, Ni, and Pb) have a strong negative association with tree biomass stocks. Metal pollution were found to be one of the most important elements impacting plant physiological activities and productivity, which could have an impact on tree biomass stocks. As a result, it is critical to create a database of their concentration in soils along roads that pass through protected forests (e.g., Biosphere Reserves (BR), National Parks (NP), Wildlife Sanctuaries (WLS), and Reserved Forests (RF), as well as their potential ecological risk and effects on forest-based ecosystem services.

Declaration of interests
The authors declare that they have no known competing nancial interests or personal ties that may have in uenced the work presented in this study.

Funding:
No fund was received Note: Min-minimum, Max-maximum, SD-standard deviation Table 2 The measured concentration of metal pollutants in soil (mg kg -1 ) samples in two different sampling sites.  Present study Table 4 geo-contamination index (Igeo) and pollution index (PI) level of metal pollutants in the two studies forests Note: S1-Bhomoraguri RF, S2-Balipara RF, SD-standard deviation  Table 7 Tree-wise biomass, carbon stocks and carbon dioxide equivalent potential (Mg tree -1 ha -1 ) of the two studied forests Note: S1-Bhomoraguri RF, S2-Balipara RF, TB-Tree biomass stock, TC-Tree carbon stock, TCO2 e-Total Carbon dioxide equivalent Table 8 Correlation between soil physicochemical properties, metal pollutants, and tree biomass stocking potential Note: pH-Soil pH, SOC-Soil organic carbon, N-available nitrogen, P-available phosphorus, K-available potassium, Cd-available Cadmium, Cravailable Chromium, Ni-available Nickel, Pb-available Lead, and TB-Tree biomass stocks. Table 9 Regression equations to predict the effects of soil metal pollutants on tree biomass