Sources, toxicity potential, and human health risk assessment of heavy metals-laden soil and dust of urban and suburban areas as affected by industrial and mining activities

Sources and levels of heavy metals (HMs) in soil and dust of urban and suburban areas in Riyadh (industrial city) and Mahad AD’Dahab (mining area) cities in Saudi Arabia were reported in this study. Additionally, the concentrations of HMs in different soil particle size fractions (> 250, 63–250 and < 63 µm) were reported. Pollution extent, and ecological and human health risks associated with collected soil and dust samples were explored. Contamination levels of HMs were higher in dust as compared to soil samples at all sites. The average integrated potential ecological risk in dust samples of urban area of Mahad AD’Dahab was 139, and thus characterized as a very-high-risk criterion. Enrichment factor (EF), correlation analyses, and principal component analysis showed that aluminum (Al), cobalt (Co), chromium (Cr), iron (Fe), manganese (Mn), nickel (Ni), titanium (Ti), and zinc (Zn) had mainly the lithogenic occurrence (EF < 2). However, Zn, copper (Cu), and lead (Pb) in Riyadh, and cadmium (Cd), Cu, Zn, and Pb in the Mahad AD’Dahab were affected by industrial and mining activities, respectively, that were of anthropogenic origins (EF > 2). The hazard index values of dust and soil (< 63 µm) samples in both urban and suburban areas in Mahad AD’Dahab were > 1, suggesting non-carcinogenic risk. Therefore, the dust and soil samples from the mined area of Mahad AD’Dahab had a higher pollution levels, as well as ecological and human health risks than those from Riyadh. Hence, the pollution of such residential environments with HMs (especially Cd, Cu, Zn, and Pb) needs to be monitored.

Interest in the characterization of urban soils and dust has increased greatly in the last two decades. Owing to the rapid growth of urban zones accompanied by increase in residential areas, streets, commercial, and industrial zones, increasing pollution levels have been recorded in urban environments [1][2][3] . Prompt industrialization has resulted in the contamination of terrestrial environments with various pollutants such as heavy metals (HMs). Various sources are responsible for very high concentrations of HMs in different environmental compartments of urban and suburban areas. The elevated levels of HMs in soil could be either of lithogenic or anthropogenic origin 4 . Anthropogenic activities including emissions from vehicles, industrial waste, atmospheric deposition of dust and aerosols, and incinerators have introduced HMs into environments at significant levels 5,6 . Thus, anthropogenic activities are the main reason for HMs contamination of soil as well as airborne dust 7 .
Dust storms are a natural phenomenon in desert ecosystems and their frequency have increased in some parts of the world since 1950s 8 . The dust can be contaminated with HMs from industrial and vehicle emissions 9  www.nature.com/scientificreports/ quality control, soil and dust samples were analyzed in four replicates. Q test was applied to exclude abnormal values at a confidence level of 95%.

Levels of HMs in soil and dust samples.
In this study, contamination levels of HMs in soil and dust were characterized using the pollution index (PI), the integrated pollution index (IPI), the geo-accumulation index (I geo ), and the potential ecological risk index (RI). The pollution index and the IPI are used to assess environmental quality 30 . The PI is calculated as the ratio of the metal concentration in the sample to the background content of the corresponding metal in the lithosphere (content of the earth's crust). The following classification of PI was used: PI ≤ 1, low level; 1 < PI ≤ 3, middle level; and PI > 3, high level. The IPI of all measured metals in samples was defined as the mean value of the PI of the metals. The classification of IPI was as follows: IPI ≤ 1, low level; 1 < IPI ≤ 2, middle level; and IPI > 2, high level 31 . The I geo method was used to calculate the metal pollution levels 32 . The I geo is computed via the following equation.
where C i is the measured concentration of the metal i and B i is the geochemical background value of the metal. In this study, B i was the background content of the metal i (background in shale) 33 . The 1.5 constant was introduced to minimize the variation of background values. The following classifications were carried out according to I geo : unpolluted, I geo ≤ 0; unpolluted to moderately polluted, 0 < I geo ≤ 1; moderately polluted, 1 < I geo ≤ 2; moderately to strongly polluted, 2 < I geo ≤ 3; strongly polluted, 3 < I geo ≤ 4; strongly to extremely polluted, 4 < I geo ≤ 5; and extremely polluted (5 < I geo ) 32 .
The potential ecological RI originally mentioned by Hakanson 34 was also calculated to assess the degree of HMs pollution in soil and dust samples using the following equations.
where RI is the sum of all six risk factors for HMs, E i is the monomial potential ecological risk factor, T i is the metal toxic factor (with the values for each metal in the order of Zn = 1 < Cr = 2 < Cu = Ni = Pb = 5 < Cd = 30) 35 , f i is the metal pollution factor, C i is the concentration of metals in dust, and Bi is a reference value for metals 34 . Different RI classifications of metal pollution are low ecological risk (RI ≤ 150), moderated ecological risk (150 ≤ RI < 300), considerable ecological risk (300 ≤ RI < 600) and high ecological risk (RI ≥ 600).
The EF is a convenient measure for assessing the degree of metal contamination and determining its probable natural and/or anthropogenic sources 36 . For normalization, a reference Fe concentration is used because of its natural abundance. The EF was calculated using the following equation 37 where C m is the content of the examined metal in the soil sample, C Fe(soil sample) is the content of the reference metal (Fe) in the soil sample, C m(earth crust) is the content of the examined metal in the earth's crust, and C Fe(earth crust) is the content of the reference metal (Fe) in the earth's crust. In general, EF values much higher than 2 are mainly considered to indicate anthropogenic sources, while values less than 2 predominantly an origin in background soil material. Moreover, the EF also assists in determining the degree of metal contamination. Five contamination categories are recognized based on this factor: EF < 2 indicates deficiency to minimal enrichment; EF = 2-5, moderate enrichment; EF = 5-20, significant enrichment; EF = 20-40, very high enrichment; and EF > 40, extremely high enrichment 38 . In addition, the lithogenic and anthropogenic HMs content was calculated using the following equation 39 .
where [M] lithogenic is the metal concentration of lithogenic origin in the sample (mg kg -1 ), [Fe] sample is the total content of Fe in the soil sample (mg kg -1 ), and ( [M] [Fe] )lithogenic is the ratio of metal concentration to iron concentration in the earth's crust. Moreover, the anthropogenic HMs content was calculated using the following equation.
where [M] anthropogenic is the anthropogenic HMs content and [M]total is the total content of HMs measured in soil samples.
The Distribution factor (DF) has been widely used to assess the distribution of HMs and environmental risks in the different particle size fractions 5,40 . The (DF) index was calculated by Eq. (8) 41 : Ei, Risk assessment. Exposure assessment. Risk assessment is a multi-step procedure of estimating the nature and probability of adverse human health effects that are caused by HMs in an environmental medium 32 . Risk assessment is based on the consideration of human exposure to soil or dust via three different pathways, namely oral intake (ingestion), inhalation, and intake through skin exposure (dermal intake). The average daily doses (ADDs) through ingestion and dermal contact for dust and soil (< 63 µm) were calculated according to the following equations: All the definitions of the parameters and values of the variables for human health risk assessments are presented in Supplementary Table S1.
Non-cancer risk assessment. The non-carcinogenic quotients of exposure to HMs in dust and soil (< 63 µm) were calculated. Non-cancer risks are expressed as a hazard quotient (HQ). The HQ is the quotient of the ADD divided by the reference dose of a specific HMs and for the exposure through each pathway. The HQ of each metal was determined by the following equation 42 : To assess the overall potential non-cancer risk, the hazard index (HI) was calculated, as follows 5 : The value of HI ≤ 1 indicates that there is no significant risk of non-carcinogenic effects. On the other hand, there is a chance that non-carcinogenic effects may occur when HI > 1, and the probability increases as the value of the HI increases 43 .
Cancer risk assessment. The incremental lifetime cancer risk (ILCR) for an individual is estimated by multiplying the slope factor (SF) with the ADD over a lifetime exposure, as determined by Eq. (13).
An ILCR value of < 1.0E−06 is considered small, whereas an ILCR of 1.0E−06 to 1.0E−04 is in the range of the acceptable limit and an ILCR of > 1.0E−04 is likely to be harmful to humans 44 . Quality assurance and quality control. Great care was taken to avoid any metal contamination during the process of sampling, digestion, and analyses from the beginning of the study. All equipment and containers were soaked in 10% HNO 3 for 24 h and then rinsed thoroughly in deionized water prior to use. Each soil and dust sample was replicated four times during digestion and analyses. Q test was used for the identification and rejection of outliers and was applied to exclude abnormal readings at a confidence level of 95%. Moreover, three standard reference soils (Till-1, Till-2, and Till-4) were employed for quality control in HMs analyses in soil (Supplementary Table S2). Standard solutions with known concentrations were simultaneously analyzed in the experiments after each set of 10 samples, to ensure the analytical performance of the ICP-OES apparatus. The imprecision of the method was computed as the relative standard deviation targeted at ≤ 5%. The detection limits of ICP-OES were < 0.1 µg L -1 for Cd and Fe and 1 µg L -1 for Al, Ti, Co, Cu, Cr, Pb, Mn, Ni, and Zn. Values of the studied metals that were below the detection limits of ICP-OES were rejected. The recovery % of HMs was calculated according to the following equation.
where C ex is the HMs concentration extracted by solution (mg kg -1 ) and C ref is the concentration of HMs in the reference soil "Till" (mg kg -1 ).
Supplementary Table S2. Recovery of HMs content in the three certified reference materials (Till 1, Till 2, and Till 4) digested using the EPA 3051 method.

Results and discussion
Distribution of heavy metal concentrations in the soil particle size fraction. Supplementary Table S3a,b show the minimum, maximum, and average concentrations of HMs in the different particle size fractions of urban and suburban soils. The HMs in the different fractions of urban and suburban soils were regularly distributed. Among the particle size fractions, the total content of most of the investigated HMs tended to increase as the size of the soil particles decreased. The highest content was generally pronounced for the particle    www.nature.com/scientificreports/ size fraction of < 0.63 μm. The highest amount of metals that accumulated in the fine particles of < 63 μm could be explained by their high reactivity and their affinity toward HMs 45 . According to some researchers, HMs are often accumulated in the fine fraction, such as clay particles that act as metal sorbents, which is mainly due to their high surface area and negative surface charge 40,45 . It was generally observed that the total HMs concentrations in the mined soils of Mahad AD'Dahab were higher than those in the industrial activity-impacted soils of Riyadh. In this context, urban soils are often contaminated with HMs owing to anthropogenic sources. Previously, it has been reported that mining activities can result in significant metal accumulation in environmental compartments 46 . Alike, several studies found that mining operations are significant sources of HMs contamination in soils and higher contents of HMs in mining-impacted soils result from long-term activities 47,48 . The Distribution factor (DF) has been widely used to assess the distribution of HMs and environmental risks in the different particle size fractions 5,40 . Supplementary Fig. S3a,b show the minimum, maximum, and average DFs for HMs in the different particle size fractions of urban and suburban soils. These obtained DF values indicated a greater metal accumulation in the finer fraction (< 63 µm) compared with that of larger size fraction (63-250 μm). Our data were in line with previous findings on the preferential partitioning of HMs to fine soil particle size fractions 5,40,49 . This can be explained by the larger surface area of the fine particles, which enhances the adsorption capacity of the fine fraction. Additionally, finer soil particles can have higher contents of secondary clay minerals, which are very strong sorbents for HMs 49,50 . On the contrary, the coarser fractions of sand and silt can have a higher content of the primary mineral quartz (e.g., SiO 2 ), thereby leading to lower sorption capacity. It could be concluded that the fine particle fractions, especially the clay fraction, accumulated higher concentrations of HMs than the coarse fractions, thereby causing potential harm to human health and the environment 50 .
Supplementary Table S4 shows the Pearson correlation between the HM concentration in dust and soil < 63 µm over all the investigated areas. The Pearson correlation showed that there was a significant correlation between the total concentration of most HMs in dust and soil < 63 µm. The Pearson correlation was 0. 55  Heavy metals content in soil and dust samples. Table 1 shows the minimum, maximum, and mean concentrations of total HMs in the bulk soil samples collected from the investigated sites in the urban and sub- In the Riyadh area, the highest total concentrations of most HMs were found at site 16 (with the exception of Cu, Cd, and Co). In the Mahad AD'Dahab area, the highest total concentrations of Cd, Cu, Pb, and Zn were detected at site 3, which is close to the mining area. Moreover, the highest concentrations of Al, Co, Cr, Fe, Ni, and Ti were recorded at site 2.
In Saudi Arabia, quality guidelines for soil HMs have not been established. Therefore, in the current study, the concentrations of HMs in soil samples were compared with other guidelines, including the common range in the earth's crust, the average concentrations in world soils, the average shale values, and the Dutch optimum and Act target values, as shown in Table 1. The concentrations of most investigated HMs were lower than their corresponding values of the common range in soil according to Ref. 51 . However, in urban Mahad AD'Dahab sites, the average concentrations of Cd, Pb, and Zn were higher than their corresponding values of the common range (0.06, 10, and 50 mg kg -1 , respectively). Moreover, the maximum and average concentrations of Cu in urban Mahad AD'Dahab sites were higher than the average concentration in world soils.
Supplementary Table S5 shows the average metal concentrations in dust samples in relation to season. Generally, the highest average values of most of the HMs were detected in the spring season in urban and suburban areas of Riyadh and suburban areas of Mahad AD'Dahab. However, in urban areas of Mahad AD'Dahab, the highest average values of most HMs were found in the winter season. The variation in the order of the highest metal levels between these two seasons (winter and spring) could be explained by changes in meteorological conditions. Table 2 shows the comparison of the obtained average metal concentrations with those reported for other countries. The comparison of HMs levels in the collected dust samples with those of other countries showed that the average content of HMs in the investigated sites in the current study were lower than those of most other cities (e.g., Riyadh (Saudi Arabia), Khamees-Mushait (Saudi Arabia), Jeddah (Saudi Arabia), Middle and South of Iraq, Kermanshah (Iran), Rafsanjan SE (Iran), Shangqing (China), Xi'an (China), Hong Kong (China), Oslo and Madrid). However, they were higher than those measured in Luanda (Angola) (especially for Al, Co, Cr, Ni, Ti, and Zn).
These results indicated that soil and dust samples were contaminated with Cd, Pb, Cu, and Zn. The highest levels of these HMs in top 3 cm soil layer could be due to the accumulation of HMs-laden dust particles. Jian et al. 52 has previously reported the contamination of topsoil potentially hazardous elements such HMs. It was suggested that industrial and agricultural activities, traffic emissions and natural sources were responsible for topsoil contamination 52 . Likewise, Wang et al. 53 collected samples from roadside dust and found that > 90% of dust contained higher levels of HMs such as Cd, Cu, Hg, Pb, As, and Zn, thus posing serious health risks to ecosystem and human health. Similar results were reported by Cai et al. 54 for soil samples, which reported the contamination of soil by Cd, Pb and Zn through agricultural practices and traffic activities. The IPI values in soil samples collected in the Riyadh and Mahad AD'Dahab areas were < 1, indicating a low level of pollution with HMs. However, based on the maximal values of IPI in samples collected from urban areas of Riyadh and Mahad AD'Dahab, the dust samples were moderately and highly polluted, respectively. Table 4 shows the values of the geo-accumulation index of the soil and dust samples collected at the Riyadh and Mahad AD'Dahab sites. The results indicate that all soil and dust samples were unpolluted, with the exception of the presence of Zn in the dust samples. In the urban Riyadh area, the Igeo values of Zn ranged from 0.94 to 3.53, suggesting moderate or strong pollution levels. However, the Igeo values of Zn in the dust samples collected at the two sites of the Riyadh suburban area ranged from 1.75 (S18) to 2.35 (S19), suggesting moderate Table 3. The calculated data of the PI and IPI for heavy metals of soil and dust samples.   51 also found that the Igeo value indicated moderate to heavy contamination with Pb and Zn, and heavy to extreme contamination with Cd in urban dust collected in Jeddah, Saudi Arabia. Table 5 shows the EF values corresponding to the soil and dust samples collected at the Riyadh and Mahad AD'Dahab sites. Among all metals, the highest EF values were recorded for Cd (6.66), Pb (4.68), and Zn (3.80) in soil samples collected at site 3 of the urban Mahad AD'Dahab area, indicating significant enrichment for Cd and moderated enrichment for Zn and Pb. Our findings suggest that site 3 of the urban Mahad AD'Dahab area is a site-specific source of pollution and that the mining operations affect the surrounding area of Mahad AD'Dahab more than does the industrial activity of the Riyadh area. It was previously reported that the soil in areas surrounding the Mahad AD'Dahab mine contains high levels of HMs, such as Cd, Cu, Zn, and Pb 30 . According to the obtained results, the overall urban and suburban dust samples collected in the Riyadh area during the seasons were characterized by deficiency to minimal enrichment for most investigated metals (especially for Al, Ti, Mn, Ni, Co, and Cr). However, for Cd, some urban sites were characterized by significant enrichment (sites 2 and/ or 3 in the summer, winter, and spring seasons) or very high enrichment (sites 2 and 3 in the autumn season). In addition, dust samples collected at site 17 in summer, autumn, and winter, as well as at site 15 in winter, were characterized by significant enrichment for Cu. Our results also showed that most investigated sites were characterized by moderate enrichments (EF = 2-5) for HMs. Among all metals, the urban and suburban sites exhibited highest enrichment for Zn, with significant, very high, or extreme enrichment.
The dust samples collected at the suburban mined area of Mahad AD'Dahab had minimal to deficiency or moderate enrichment for all investigated metals (with the exception of Zn, with an EF ranging from 12.0 to 38.1, which indicates very high enrichment). Based on EF values, it was generally observed that urban dust samples  www.nature.com/scientificreports/ were polluted with metals to a greater extent than were suburban dust samples (especially for Cd, Cu, Pb, and Zn). Previously, it was suggested that the EF index could be applied to the determination of the origin of HM contamination, with EF values < 2 indicating that the HM originates from natural sources and EF values > 2 indicating that the HM originates from anthropogenic activities. In this context, the EF values obtained for Al, Co, Cr, Mn, Ni, and Ti at all sites of Riyadh and Mahad AD'Dahab were < 2; therefore, these metals cannot be considered as a major contamination concern. However, the EF values obtained for Cd, Cu, Pb, and Zn were > 2 ( Fig. 1), indicating a major contamination concern and the anthropogenic effects of these metals. Among all sites, site 3 at the mined urban area of Mahad AD'Dahab showed the highest EF values for these metals (especially for Pb, Cu, Zn, and Cd), mainly because of its closeness to the gold mine. These results with variations in the calculated coefficients (EF and PI) suggested that anthropogenic activities contributed in the accumulation of HMs in dust and soil of the studied areas. Similar results were reported by Cai et al. 55 for the contamination of agricultural soils with Pb, Zn, Cd and Hg. The assessment of the ecological risk potential was also investigated in the current study regarding Cd, Pb, Cu, Cr, Ni, and Zn. Table 6 shows the single potential ecological risk (Ei) and the integrated potential ecological risk (RI). Among all investigated HMs, Cd (maximum Ei value, 274) in dust samples collected at urban sites in Mahad AD'Dahab can be categorized as a moderate ecological risk. Conversely, among all urban and suburban areas, the average integrated potential ecological risk (RI) in dust samples collected at urban areas of Mahad AD'Dahab was categorized as moderate risk. Figure 2 shows the lithogenic and anthropogenic contribution to the levels of each metal in dust samples. It was reported previously that a value of 50% can be considered as the minimum value for anthropogenic contribution. Therefore, the results obtained indicated that the soil samples had a higher significant anthropogenic contribution (> 50%) was observed for Pb in the urban and suburban area of Riyadh, and urban area of Mahad AD'Dahab. Additionally, the relative higher anthropogenic contributions for Zn (37%) at urban area of Riyadh and for both Cd and Zn (42%) at urban area of Mahad AD'Dahab. Meanwhile, the dust samples had a higher Table 5. The calculated data of the enrichment factor (EF) for heavy metals of soil and dust samples.

Soil
Al www.nature.com/scientificreports/ lithogenic contribution (< 50%) regarding Al, Ti, Mn, Ni, Co, and Cr. Conversely, a higher significant anthropogenic contribution (> 50%) was observed for Zn and Pb in the Riyadh area, and for Zn, Pb, Cu and Cd in the Mahad AD'Dahab area. Generally, it can be speculated that, in the studied areas, these metals are originated from anthropogenic sources, indicating the possibility of contamination of the investigated areas with these metals as a result of industrial activities and heavy traffic at Riyadh, or mining activities at Mahad AD'Dahab.
The correlations between HM concentrations in dust samples collected from the Riyadh and Mahad AD'Dahab areas are presented in Supplementary Table S6. Pearson coefficients showed a significant positive correlation among various metals in dust samples. Most of the investigated HMs (with the exception of Cd and Cu at the Riyadh area and Cd, Cu, and Pb at the mining area) were significantly correlated with Fe, Al, or Ti. Our results indicated that Al, Cr, Co, Mn, Ni, Ti, and Zn in dust were more likely to stem from the same sources. However, a significant positive correlation was observed between Cd, Cu, and Pb, suggesting that these HMs in dust samples of the mining area originated from the same sources.
The results of PCA showed that four components were extracted for HMs in the dust samples at the Riyadh area (affected by industrial activities), whereas two components were extracted for the mining area of Mahad AD'Dahab. In the Riyadh area, Factors 1, 2, 3, and 4 accounted for 57.12%, 15.75%, 11.21%, and 5.98% of the total variance, respectively (Supplementary Table S7). Factor 1 was positively loaded with Al, Cr, Co, Fe, Mn, Ni, and Ti. Factors 2, 3, and 4 were dominated by Cu, Cd, and Zn, respectively. These results suggest that Al, Cr, Co, Fe, Mn, Ni, and Ti may originate from a lithogenic source; in contrast, Cu, Cd, and Zn may stem from anthropogenic sources. Moreover, at the mining area of Mahad AD'Dahab, Factor 1 was dominated by Al, Co, Cr, Fe, Mn, Ni, and Ti, while factor 2 was dominated by Cd, Cu, and Pb. The 2D plot of the PCA suggested that the HMs in both areas came from different sources (Fig. 3). The results of the 2D plot of the PCA showed a high Table 6. Average the single potential ecological risk (Ei) and integrated potential ecological risk (IR) in soil and dust of urban and suburban areas in Riyadh and Mahad AD'Dahab.  A loading between Cd, Cu, and Pb was also observed at the mining area of Mahad AD'Dahab. In addition, there was a loading between Zn, Pb, and Cu at the industrial area of Riyadh. These three metals (Cd, Cu, and Pb) at the mining area, and Zn and Pb in particular at the industrial area of Riyadh, exhibited an EF > 2 in most dust samples, with significant correlation between them. Therefore, these metals may stem from another source (mainly anthropogenic input). A study conducted by Yang et al. 56 to trace the sources of, and assess the contamination with, HMs in dust samples in a typical mining city in China reported that Ni, Zn, Fe, and Cr were associated with the first factor, which led the authors to suggest that these four metals originated mainly from common sources. However, they found that Cd and Pb were associated with the third factor, suggesting anthropogenic sources 54 . In another study based on EF and multivariate analysis, Almasoud et al. 57 found that the sources of Zn, Cu, and Pb were primarily anthropogenic, whereas Co, Ni, Fe, and Mn in the soil stemmed from industrial activities. In the mining Mahad AD'Dahab area, Alsaleh et al. 45 reported that the soil samples were characterized by moderate to high pollution levels of Cd, Cu, Zn, and Pb, whereas the concentrations of Fe, Mn, Ni, Co, and Cr were at background levels.
Human health risk assessment of HMs in dust and soil in urban and suburban areas. Heavy metals can have a serious impact on human health. For urban areas, the direct risks of HMs in recreational, residential, traffic, and industrial areas generally originate from oral ingestion and dermal exposure 58,59 . In this study, the non-carcinogenic HQ, HI, and ILCR of each HM were computed. The human health risk indicators www.nature.com/scientificreports/ of dust and soil of < 63 µm for HMs were calculated for all seasons in urban and suburban areas in Riyadh and Mahad AD'Dahab.

Non-carcinogenic and carcinogenic hazards assessment of HMs in dust.
The HQ values in the urban and suburban areas indicated that the investigated HMs in Mahad AD'Dahab had higher values than those of Riyadh in the exposure pathways for all human categories (except for Cr, Ni, and Pb in suburban areas). In Riyadh and Mahad AD'Dahab, the HQ values of HMs for all dust samples were < 1. Therefore, there were no adverse health effects (Supplementary Tables S8 and S9). Likewise, the HI values in the urban and suburban areas of Riyadh and Mahad AD'Dahab were < 1, thereby suggesting that there was no non-carcinogenic risk (Supplementary Table S10). The HI values indicted that urban dust samples collected from Mahad AD'Dahab had a more serious impact on human health than those from Riyadh. Therefore, more attention should be paid to the HM contamination in dust samples of such residential environments. Li et al. 60 assessed the health risks of nine HMs (As, Cd, Co, Cr, Pb, Cu, Zn, Mn, and Ni) in road dust in the Bayan Obo Mining Region in Inner Mongolia, China. In their study, the main exposure route was ingestion, and they found that Mn, Cr, Pb, and As were the main contributors to non-cancer risks in both children and adults. Meanwhile, according to the cancer risk assessment, Cr was the main contributor.
The HI values in Riyadh and Mahad AD'Dahab were < 1 for pathway exposure for all human categories. Also, the HI values were in the following order: children up to 6 years > children from 6 to 12 y > adults (Supplementary Figs. S5 and S6). In Riyadh, the highest HI values were found in site 16 followed by sites 2 and 3. In Mahad AD'Dahab, the highest HI values were found in site 3 followed by sites 2 and 1. The ΣHI values of each urban and suburban area of Riyadh were < 1 (except for site 16 with a value > 1) (Supplementary Fig. S7). Therefore, it could be speculated that there was no significant risk of non-carcinogenic health effects in this area (except for site 16). Additionally, the ΣHI values in sites 2 and 3 were close to 1 at 9.26E−01 and 8.97E−01, respectively; therefore, non-carcinogenic risks could be possible.
Al-Rashdi et al. 40 assessed the health risks of HMs (Fe, Pb, Cu, Mn, Zn, Co, and Cd) in dust from the Dammam area. The HQ and HI values were found to be < 1. However, they suggested that the potential health risks for children and adults due to exposure to HMs cannot be ignored. Similarly, Shabbaj et al. 51 assessed the health risks of HMs (Fe, Mn, Zn, Pb, Cd, V, Co, Ni, As, Cr, and Cu) in dust from the Jeddah area. They found no potential health risks for the investigated HMs (except for As). However, in the current study, the HI values for all the human categories of each urban area of Mahad AD'Dahab were > 1 (except for site 1 = 9.28E−01); thus, the Mahad AD'Dahab sites posed non-carcinogenic risks. Based on the overall average the hazard index (HI), higher non-carcinogenic risks in urban and suburban in dust samples of Mahad AD'Dahab were observed than those of Riyadh ( Supplementary Fig. S8). In this context, Al Bakheet et al. 28 reported that high concentrations of Pb, Cd, and Hg were observed in the blood samples of residents of Mahad AD'Dahab compared with those of Riyadh, and thus affected genes.
The relationships between the characteristics of dust pollution indicators, ecological risk (RI), and human health risks were investigated. The results showed that there was a significant correlation between the ΣHI and dust pollution indicators (IPI, EF, and I geo ) and RI (Table 8). In Riyadh, the significant relationship (r) between the ΣHI and dust pollution indicators of IPI, average EF, and average I geo had values of 0.77, 0.65, and 0.98, respectively (Table 8). In Mahad AD'Dahab, the r values between ΣHI and dust pollution indicators of IPI and average I geo were 0.89 and 0.97, respectively. Moreover, significant relationship (r = 0.75) was observed between the ΣHI and RI in current study. Therefore, it could be concluded from the results that the characteristics of dust pollution indicators and RI could be used to predict human health risks in such areas.
The lifetime cancer risks of Cr and Pb in dust for children up to 6 y, children from 6 to 12 years, and adults for different exposure pathways were calculated and presented in Table 7. The USEPA recommends a threshold of 1.00E−06 61 . In urban and suburban areas of Riyadh and Mahad AD'Dahab, the ILCRing and ILCRder for Cr were between 1.00E−06 and 1.00E−04 in all categories for human exposure pathways, thereby indicating that the inhabitants around the study areas of Riyadh and Mahad AD'Dahab may have a high risk of lifetime cancer development. On the contrary, the ILCRing and ILCRder for Pb were < 1.00E−06 in all categories for human exposure pathways, thereby indicating a lower risk 61 .
Generally, the mean values of ILCR for Cr and Pb in urban areas of Mahad AD'Dahab were higher than those for Riyadh. On the contrary, the mean values of ILCR for Cr and Pb in suburban areas of Mahad AD'Dahab were lower than those for Riyadh in all human categories. Additionally, among the two metals (Cr and Pb), the ILCR values of Cr in Mahad AD'Dahab and those in Riyadh for all human risk categories were close to the recommended threshold of 1.00E−06 to 1.00E−04, thereby indicating a possible risk of lifetime cancer development in the inhabitants around the study sites 61 .

Non-carcinogenic and carcinogenic hazards assessment of HMs in soil (< 63 µm). The HQ and
HI values in the soil samples (< 63 µm) of the urban and suburban areas indicated that the investigated HM concentrations in Mahad AD'Dahab were higher than those in Riyadh in the exposure pathways for all human categories (with the exception of Cd and Cr).
In Riyadh and Mahad AD'Dahab, the HQ values of HMs for all soil samples were < 1, thereby suggesting no adverse health impacts (Supplementary Tables S11 and S12). Additionally, the HI values in the urban and suburban areas of Riyadh and Mahad AD'Dahab were < 1 for all human categories, thereby suggesting that there was no non-carcinogenic risk (Supplementary Table S13). The HI values indicated that urban soil samples collected from Mahad AD'Dahab had a more serious impact on human health than those in Riyadh. Therefore, more attention should paid to the HM contamination in soil samples of such residential environments. In Riyadh, the HQ and HI values in urban areas were lower than those in suburban areas (except for Cu and Zn). In Mahad AD'Dahab,  Supplementary Fig. S11), indicating no adverse non-carcinogenic health effects. However, the ΣHI values for all human categories in the urban and suburban sites in Mahad AD'Dahab were > 1; thus, the Mahad AD'Dahab sites posed non-carcinogenic risks (Supplementary Figs. S12 and S13). Alsaleh 62 assessed the health risks of HMs (Cd, Cu, Pb, and Zn) in the soil of the Riyadh and Mahad AD'Dahab areas. In their study, the ΣHI values for all human categories in Riyadh and Mahad AD'Dahab areas were 1.07 and 28.90, respectively, which were above the allowable safe dose limits for non-cancer health effects. Mahad AD'Dahab area had higher ΣHI values than those obtained from our study. The higher values of ΣHI obtained by Alsaleh 62 could be explained by soil samples that were collected from the nearby Mahad AD'Dahab mine.
The relationships between the characteristics of soil pollution indicators, RI, and human health risks were established. The relationships showed that there was a significant correlation between the ΣHI and soil pollution indicators (IPI, EF, and I geo ) and RI (Table 8). In Riyadh, the r values between ΣHI and the soil pollution indicators of IPI, average EF, and average I geo were 0.69, − 0.59, and 0.95, respectively. In Mahad AD'Dahab, the r values between ΣHI and the soil pollution indicators of IPI, average EF, and average I geo were 0.98, 0.99, and 1.00, respectively. In addition, there was a significant relationship (r = 0.98) between the ΣHI and RI. From the obtained results, it could be concluded that the characteristics of soil pollution indicators and RI could be used to predict human health risks in such areas 63 .
The lifetime cancer risks due to Cr and Pb in soil for all the pathways for children up to 6 years, children from 6 to 12 years, and adults were calculated in this study. Table 7 shows the average ILCR for Cr and Pb in soil in the urban and suburban areas of Riyadh and Mahad AD'Dahab. In the urban and suburban areas of Riyadh and Mahad AD'Dahab, the ILCR ing , ILCRder, and ILCR ing+der were between 1.00E−06 and 1.00E−04 in all human categories, thereby indicating that the inhabitants around the study areas of Riyadh and Mahad AD'Dahab may have a high risk of lifetime cancer development. On the contrary, the ILCRing, ILCRder, and ILCR ing+der for Pb were < 1.00E−06 in all human categories, thereby indicating lower risk 61 . Generally, the mean values of ILCR for Cr and Pb in urban and suburban areas of Mahad AD'Dahab were higher than those in Riyadh (except for Cr in suburban areas). Also, among the two metals (Cr and Pb), the ILCR values of Cr in Mahad AD'Dahab and Riyadh for all human categories were close to the recommended threshold of 1.00E−06 to 1.00E−04, thereby suggesting a possible risk to the inhabitants around the study sites 61 . Previously, it has been shown that ingestion www.nature.com/scientificreports/ was the most hazardous pathway for developing cancer via HMs intake, followed by dermal contact and inhalation. Moreover, it was observed that carcinogenic risks for children via ingestion and dermal contact were more than that of adults 64 .

Conclusions
Levels and sources of HMs present in dust and soil of urban and suburban areas in Riyadh and Mahad AD'Dahab cities were studied and associated ecological and human health risks were estimated in this study. Overall, the levels of studied HMs were higher in dust samples than that of soil samples collected from the studied areas. The enrichment factor and PCA analyses exhibited that Al, Co, Cr, Fe, Mn, Ni, Ti, and Zn were of lithogenic origin, whereas, Cu, Zn, and Pb in Riyadh and Cd, Cu, Pb, and Zn in the Mahad AD'Dahab were of anthropogenic activities. The HI values for dust and soil (< 63 µm) samples were as: children up to 6 years > children from 6 to 12 years > adults. The HI values for all human categories for dust in urban areas of Mahad AD'Dahab and soil samples (< 63 µm) in urban and suburban areas of Mahad AD'Dahab were > 1, indicating non-carcinogenic risk.
The average values of ILCR for dust and soil (< 63 µm) samples from both urban and suburban areas of Riyadh and Mahad AD'Dahab were close to the recommended threshold for Cr; however, were lower for Pb. Higher levels and ecological and human health risks of HMs in residential areas of Riyadh and Mahad AD'Dahab could be owing to industrial and mining activities in these cities. Therefore, ecological and human health risk assessment showed that soil and dust of Mahad AD'Dahab were more polluted as compared to Riyadh. It is therefore recommended to monitor HMs pollution in dust and soil particles of such residential environments for sustainable ecosystem and human health.