3.1 Magnetic response to heavy metal pollution in urban topsoil in various regions of China
Magnetic particles exhibit a pronounced effect in concentrating heavy metals (Xiao, 2019). The correlation between mass susceptibility and frequency susceptibility serves as an effective indicator of heavy metal contamination in soils. Figure 1 illustrates this relationship for urban topsoil samples across various regions in China. Building on the findings of prior studies (Wang et al., 2000; Lu et al., 2016; Han et al., 2021), it is generally accepted that a mass susceptibility (χlf) exceeding 100 × 10− 8 m3 kg− 1 coupled with a frequency susceptibility (χfd%) below 3% is indicative of soil contamination. Furthermore, within contaminated zones, a higher mass susceptibility concomitant with a lower frequency susceptibility signifies a greater level of pollution.
Consequently, with the exception of loess regions, a majority of provinces and cities across the six main areas exhibit varying degrees of magnetic signatures, as depicted in Fig. 1. This variation suggests that urban topsoils throughout these regions has been contaminated by heavy metals to varying extents. Notably, cities with heavy industrial backgrounds in the Northeast demonstrate the most acute levels of topsoil contamination. Moreover, the mass magnetic susceptibility (χlf) values in Northeast China are notably the highest, peaking at 942.376 × 10− 8 m3 kg− 1, and demonstrating a significant mean value. Conversely, the percentage frequency susceptibility (χfd%) registers as the lowest, averaging at 2.062% (Table 3). The pollution levels in North China, Northwest China, Central China, and East China exhibit similar patterns. In addition, applying Wilding’s (1984) classification for the coefficient of variation, where a CV > 0.36 indicates high variation, 0.16 < CV < 0.36 signifies medium variation, and CV < 0.16 denotes low variation, it emerges that most regions in this study fall into the high variation category. This reveals pronounced spatial disparities in the magnetic properties of urban topsoil across different areas. Concomitantly, the magnitude of variability (CV) is influenced by the sample size; regions with a higher number of cities involved tend to demonstrate greater variability. Both the Northwest and East China regions manifest considerable variability, attributable to their inclusion of numerous provinces and cities.
Table 3
Descriptive statistics of χlf and χfd% in different regions of China.
Region | Northeast | North | Northwest | Central | East | Southwest |
Sample number | 5 | 10 | 33 | 7 | 57 | 2 |
χlf (×10− 8 m3 kg− 1) | Minimum | 196.88 | 76.45 | 41.10 | 123.00 | 30.37 | |
Maximum | 2624.00 | 891.00 | 1129.00 | 313.00 | 1328.00 | |
Mean | 942.38 | 227.45 | 176.88 | 204.35 | 189.15 | 440.00 |
SD | 973.82 | 255.33 | 227.14 | 67.36 | 215.94 | |
CV(%) | 1.03 | 1.12 | 1.28 | 0.33 | 1.14 | |
χfd% (%) | Minimum | 0.90 | 2.06 | 0.64 | 2.30 | 0.39 | |
Maximum | 2.60 | 6.04 | 7.10 | 5.80 | 14.74 | |
Mean | 2.06 | 4.20 | 2.97 | 3.85 | 3.55 | 7.73 |
SD | 0.67 | 1.19 | 1.80 | 1.23 | 2.95 | |
CV(%) | 0.33 | 0.28 | 0.61 | 0.32 | 0.83 | |
Upon analyzing the causative factors, it has been ascertained that the predominant contributor to heavy metal contamination in the urban topsoil of Northeast China is the localization of sampling sites within the Anshan and Fuxin iron and steel industrial zones in Liaoning Province (Xiao, 2019; Wang et al., 2021). Lu (2000) posits that the threshold value indicative of superparamagnetic particles’ presence in soil is χfd% = 5%. This investigation reveals that, barring the southwest region, χfd% values across other areas fall below 5%. This suggests that pedogenesis is the key driver of magnetic anomaly accumulation in the urban topsoil of Southwest China, as opposed to other regions where anthropogenic activities or geological disturbances are the primary causes (Lu et al., 2000). Consequently, an examination of the magnetic properties of urban topsoil across various Chinese locales demonstrates regional disparities in heavy metal pollution. Industrialized zones or areas with intensive human activity are more susceptible to exacerbated soil contamination by heavy metals.
3.2 Magnetic susceptibility characteristics of urban topsoil across varying depths
The analysis of magnetic susceptibility (MS) variation in urban topsoil at different depths (refer to Fig. 2) reveals an initial increase followed by a decrease in mass-specific MS with greater sampling depths. The peak values were recorded in the 0–5 cm and 0–10 cm layers, averaging 232.72 × 10− 8 m3 kg− 1 and 287.20 × 10− 8 m3 kg− 1, respectively. These findings suggest that the most severe heavy metal contamination is present within the top 5 to 10 cm of soil. The pronounced variability at the 0–10 cm depth, as indicated by the high coefficient of variation at 1.63, may be attributed to data disparities (Table 4).
Table 4
Descriptive statistics of χlf in different sampling depth.
Sampling depth (cm) | 0–2 | 0–5 | 0–10 | 0–15 | 0–20 | 0–30 |
Sample number | 18 | 20 | 38 | 9 | 32 | 11 |
χlf (×10− 8 m3 kg− 1) | Minimum | 41.1 | 61.0 | 38.0 | 24.7 | 27.4 | 24.6 |
Maximum | 930.9 | 1129.0 | 2624.0 | 181.7 | 891.0 | 523.6 |
Mean | 190.17 | 232.72 | 287.20 | 70.39 | 27.39 | 24.60 |
SD | 197.32 | 246.76 | 468.93 | 50.58 | 194.84 | 150.95 |
CV(%) | 1.04 | 1.06 | 1.63 | 0.72 | 1.09 | 1.19 |
Analysis of the frequency-dependent magnetic susceptibility across varying topsoil strata (Fig. 3) reveals a distinct pattern: susceptibility initially increases with depth, then subsequently declines. Notably, the shallowest layer (0–2 cm) exhibits the lowest susceptibility, averaging 1.95%, followed by the 0–5 cm layer at 2.62% (Table 5). These findings suggest a more pronounced heavy metal contamination within the 0–2 cm layer compared to deeper strata, with the 0–5 cm layer also showing significant levels of pollution.
Table 5
Descriptive statistics of χfd% in different sampling depth.
Sampling depth (cm) | 0–2 | 0–5 | 0–10 | 0–15 | 0–20 | 0–30 |
Sample number | 18 | 20 | 38 | 3 | 30 | 5 |
χfd% (%) | Minimum | 0.64 | 0.39 | 0.70 | 2.82 | 1.17 | 0.9 |
Maximum | 3.50 | 6.84 | 14.74 | 7.09 | 8.19 | 10.2 |
Mean | 1.95 | 2.62 | 4.06 | 5.27 | 3.79 | 4.68 |
SD | 0.65 | 1.72 | 3.01 | 2.20 | 1.89 | 4.37 |
CV(%) | 0.332 | 0.655 | 0.741 | 0.418 | 0.499 | 0.933 |
Magnetic susceptibility typically peaks within the top 0–6 cm of soil, attributable to the accumulation of magnetic particles (Chen and Liu, 2023). The exact depth of this peak, however, varies across studies. Pang (2015) posits that magnetic minerals in various soil types predominantly reside beyond 5 cm depth, with the highest χlf values recorded at 1 cm. Conversely, Yan (2011) asserts that magnetic minerals concentrate primarily between 0–2 cm and 2–5 cm, with a marked increase in χlf; the value at 2 cm is 1.4 times that at 5 cm. Beyond this depth, up to 20 cm, the χlf values plateau, reflecting the local baseline; the χfd% at 0–2 cm registers as the lowest, at 1% across the entire profile. These observations align with findings from contaminated soils (Ju et al., 2004; Shen et al., 2007). In a study of Xuzhou’s urban topsoil, Liu (2020) found χlf values of 132.3 × 10− 8·m3·kg− 1 and 99.6 × 10− 8·m3·kg− 1 at depths of 0–2 cm and 3–10 cm, respectively, with χfd% at 3.5% and 4.8% (P < 0.05). This suggests anthropogenic magnetic particles in the topsoil, with more significant heavy metal pollution at 0–2 cm, corroborating Yan’s (2011) research. Therefore, synthesizing the findings discussed herein with those of other researchers, it is apparent that soil magnetic characteristic peaks are discernible, to varying extents, within the 0–5 cm sampling depth.
3.3 Pollution characteristics of heavy metals in urban topsoil under different sampling depths
The investigation into the variations of heavy metal concentrations at different topsoil strata (as illustrated in Fig. 4 and Fig. 5) indicates a general trend where these concentrations initially rise with increasing depth and subsequently decline. It is noteworthy that the depth at which peak concentrations of various heavy metals are observed exhibits minor differences. Specifically, the concentrations of lead (Pb), zinc (Zn), chromium (Cr), and manganese (Mn) hit their highest at the 0–5 cm depth interval (refer to Fig. 4). Conversely, copper (Cu), nickel (Ni), and iron (Fe) manifest their peak concentrations within the 0–10 cm soil layer. Regrettably, an insufficiency in data precludes a comprehensive explanation of cadmium (Cd) behavior, as depicted in Fig. 5.
The results of the aforementioned analysis indicate that in the topsoil layer (0–5 cm depth), there is a higher tendency for enrichment of Pb, Zn, Cr, and Mn. In contrast, Cu, Ni, and Fe exhibit a greater propensity for accumulation within the 0–10 cm soil depth. These findings are consistent with those of prior studies, albeit with certain variances. For instance, Pang’s (2015) investigation into the stratification of heavy metals in the vicinity of Jiuquan Iron and Steel Company, located in Jiayuguan City, revealed that the distribution patterns of heavy metals mirrored the trends in soil magnetic susceptibility. Moreover, the concentrations of various heavy metals displayed a broadly similar pattern, peaking within the 0–5 cm depth range.
Yan (2011) observed that the concentration profiles of five heavy metals (Pb, Zn, Mn, Cu, Fe) in the soil proximal to a steel factory in Lanzhou exhibited analogous trends with depth. These profiles can be categorized into three distinct phases delineated by a 10 cm threshold. In the 0–2 cm layer, there was a precipitous increase in heavy metal concentrations, reaching a peak and demonstrating significant enrichment. Between 2–10 cm, concentrations diminished progressively with depth, indicating slow deposition and translocation. At 10–20 cm, the concentrations stabilized, reflecting the background levels for the soil in the area under study.
The observed discrepancies are attributable to the choice of research subjects. Pang (2015) and Yan (2011) focused their investigations on the surface soil in the vicinity of Jiuquan Iron and Steel Corporation situated in Jiayuguan, and a steel mill in Lanzhou, respectively, examining particular instances. In contrast, this study encompasses a broader scope, targeting major urban areas in China affected by heavy metal contamination, thereby reflecting a generalized condition.
3.4 Magnetic response to heavy metal pollution in urban topsoil at varying sampling depths
Previous research indicates a robust correlation between soil heavy metals and magnetic measurements, with the strength of this correlation largely hinging on the source of pollution (Xiao, 2019). The question arises: how does the depth at which samples are taken influence this relationship? According to the Spearman’s correlation analysis of magnetic indices and heavy metal concentrations at varying depths, the correlation with low-frequency magnetic susceptibility (χlf) is more pronounced than that with frequency-dependent susceptibility (χfd%) (Table 6). Notably, the correlation between χlf and the concentrations of heavy metals demonstrates the greatest strength within the 0–5 cm soil layer, achieving levels of high statistical significance. Furthermore, anthropogenic activities appear to exert an influence on the association between heavy metals and magnetic properties in the uppermost soil stratum (0–2 cm), which is evidenced by a more robust correlation within the 0–5 cm and 0–2 cm intervals (Yan, 2011).
Table 6
Spearman correlation analysis of magnetic parameters and heavy metal concentrations at various soil depths.
Magnetic parameters | Sampling depth (cm) | Heavy metals (mg/kg) | (g/kg) |
Pb | Zn | Cr | Mn | Cu | Ni | Cd | Fe |
χlf (×10− 8 m3 kg− 1) | 0–2 | 0.810* | 0.755** | 0.286 | 0.143 | 0.691* | -0.048 | - | 0.103 |
0–5 | 0.459 | 0.830** | 0.857** | 0.833* | 0.806** | 0.452 | 0.217 | 1.000** |
0–10 | 0.453 | 0.309 | 0.083 | 0.107 | 0.562* | 0.175 | 0.477 | 0.393 |
0–20 | 0.644** | 0.416 | -0.502* | -0.541* | 0.406 | -0.160 | 0.786* | -0.600 |
χfd% (%) | 0–2 | 0.550 | 0.420 | -0.067 | -0.750 | 0.608* | 0.262 | - | -0.564 |
0–5 | -0.021 | -0.478 | -0.311 | -0.371 | -0.273 | -0.383 | -0.360 | -0.600 |
0–10 | -0.353 | -0.440 | 0.244 | -0.145 | -0.409 | -0.473 | 0.039 | -0.221 |
0–20 | -0.132 | -0.371 | -0.510* | -0.679** | -0.076 | 0.150 | 0.643 | - |
Note: ** and * denote significance levels at 0.01 and 0.05, respectively. |