Heavy Metal Index
Ask the IndexAboutIndex Pro

Analysis of Heavy Metal Pollution in Soil along the Shuimo River by the Grey Relational Method and Factor Analysis

Cite this pageVersion history

Overview

Zhang and colleagues measured heavy metals in riparian soil along the Shuimo River in Urumqi, Xinjiang, China, then used grey relational analysis and factor analysis to classify soil quality and infer source groups. This is primary environmental-context evidence, not food, ingredient, or consumer-product occurrence evidence. The paper does not sample crops, finished foods, cosmetics, supplements, drinking water, or branded products.

The routeable finding is that the upper and middle river reaches were classified as Grade III / slight pollution under the authors’ soil-quality scheme, while downstream zones were Grade II / good soil quality. The six evaluation metals were total arsenic, lead, zinc, copper, nickel, and total chromium. Cadmium and total mercury were screened but excluded from the modeled evaluation because the measured values were described as low or not detectable.

Key numbers

  • Sampling frame: 42 soil sampling points; 122 soil samples after some hard-texture locations could provide only one or two of the planned three layers.
  • Planned profile depths: 0-20 cm, 20-40 cm, and 40-60 cm.
  • River-reach aggregation: nine averaged zones, S1-S3 upstream surface/middle/lower layers, S4-S6 midstream surface/middle/lower layers, and S7-S9 downstream surface/middle/lower layers.
  • Evaluation analytes: lead (Pb), Zn, total arsenic (tAs), Cu, nickel (Ni), and chromium (Cr). cadmium (Cd) and total mercury (tHg) were not carried into the grey-correlation/factor models because they were low or not detectable.
  • Exact worked concentration vector for S1, from the paper’s grey-correlation example, in mg/kg soil: Pb 22.92, Zn 79.42, tAs 14.74, Cu 40.63, Ni 50.95, Cr 54.36.
  • Figure 2 pattern: Zn and Cr reached their maxima at S5; Ni, Cu, Pb, and tAs reached their maxima at S3. Ni reached its minimum at S7; Zn, Cr, and Cu reached their minima at S8; Pb and tAs reached their minima at S9.
  • Grey relational classification: S1-S6 were Grade III / slight pollution; S7-S9 were Grade II / good soil quality.
  • Maximum relevance coefficients in Table 2: S1 0.936, S2 0.917, S3 0.905, S4 0.897, S5 0.848, S6 0.855, S7 0.867, S8 0.914, S9 0.910.
  • Factor-analysis diagnostics: KMO = 0.754; Bartlett significance reported as 0 < 0.05; two main factors explained 96.325% of the variance.
  • Rotated factor contribution rates: Factor 1 about 59.493%; Factor 2 about 36.831%.
  • Source interpretation: Factor 1 grouped Zn, Cu, Cr, Pb, and Ni and was interpreted as industrial pollution; Factor 2 grouped tAs with shared Pb/Ni contribution and was interpreted as human-household/agrochemical source context.

Methods (brief)

Soils were collected along the Shuimo River at 2 km intervals, with additional opposite-bank sampling, for 42 sampling points total. Samples were air dried, cleaned of plant/debris material, crushed without metal contact, sieved through 10 mesh and then 100 mesh nylon sieves, mixed by quartering, and bagged. Land-use settings included industrial areas, residential areas, woodland, agricultural land, vegetable land, roadside bare ground, artificial grassland, mountainous areas, and water outfalls.

All samples were screened with a portable X-ray fluorescence spectrometer. Thirty percent of the samples were also tested by national-standard methods after HF/HNO3/HClO4 digestion. Total arsenic was determined by atomic fluorescence spectroscopy, and Zn, Cu, Ni, and Cr were determined by flame atomic absorption spectrophotometry. The authors state that rapid-test results agreed with the national-standard method at more than 95% accuracy, so the XRF dataset was used for statistical analysis.

The study averaged the 122 samples into nine reach/layer zones before grey relational classification and factor analysis. Soil quality thresholds were drawn from GB15618-2018, GB36600-2018, and Xinjiang local soil background values.

Implications

Certification: This source should not enter any HMTc product or ingredient benchmark pool. It can support environmental source-context notes for Chinese river-adjacent agricultural and industrial landscapes, especially where riparian soils may influence food-chain risk through irrigation, vegetable plots, or downstream sediment transfer.

Courses: Useful for showing how environmental soil papers can carry real metal measurements while still requiring careful evidence routing. The study is also a teaching example for grey relational analysis, factor analysis, and source-apportionment language.

App: Context only. The data can inform regional-risk flags or source-attribution explanations, but the study does not measure edible tissue and does not justify product-level concentration estimates.

Microbiome: Not addressed.

Update history

No substantive edit history is available in this build. The full commit record is available in git.

Referenced by