Results and Discussion

Principal Component Analysis (PCA)

Fig. 10. Biplot which visualizes the results of the PCA.

Fig. 11. Biplot which visualizes the results of the PCA.

PCA was used on the dataset to reduce dimensionality, identify patterns, and capture the main sources of variation in the concentrations of TEs across nests (sites), sampling dates, and depths (Fig 10-11).
The length and direction of the arrows indicate the contribution of each variable to the Principal Components.
Variables that are close together are highly correlated, while those that point in opposite directions are negatively correlated (Fig 10).
Elements that load similarly on a component might originate from the same source or interact similarly with environmental factors, which can be valuable for tracing sources or understanding geochemical behavior.
The biplot suggests that the TEs that are clustered together (Fig 10) may fall into distinct groupings based on their chemical behavior, sources, or environmental associations.

The Pb, As, Mo, Sb, and Cd are clustered together (Fig 10 & 11). These elements are associated with a distinct source, such as anthropogenic pollution or heavy metal contamination, as Pb, As, and Cd are often associated with industrial activities or pollutants (Appelo & Postma, 2013). This could imply that samples aligning with this arrow direction have higher concentrations of these potentially toxic elements.
Ag and Cs arrows pointing in separate, individual directions suggest that these TEs do not correlate strongly with any of the other elements, meaning they might have distinct sources or environmental behaviors. For example, Cs could be associated with specific soil types or mineral phases, while Ag might be influenced by different factors altogether, possibly due to isolated contamination or low overall concentrations.
Li and Ni forming a separate group indicates that they are correlated with each other but are distinct from the other two main clusters. This grouping might represent elements with unique geochemical behavior or sources different from both the natural cluster (Th, La, etc.) and the likely anthropogenic cluster (Pb, As, etc.) (Hooda, 2010). These elements might be less mobile or more retained within certain soil or sediment layers, which could explain why they do not align with the directions of the other clusters.

Correlation Analysis

Correlation of TE concentrations with Sampling Date:

Cd, As, Pb, and Be have strong positive correlations with date. This means that the concentrations of these TEs tend to increase from Jul to Aug 2023.
Sb, V, La, U, and Ga also show moderate positive correlations with sampling date, suggesting a general trend of increasing concentrations over time for these elements, though not as strong as Cd or As.
TEs like Ag, Mo, Se, and Th show weak to moderate positive correlation, indicating little to no consistent pattern with sampling date.

Correlation of TE concentrations with Depth:

Sb, As, Pb, and Cd have strong positive correlations with Depth. This suggests that the concentrations of these elements tend to increase with depth, meaning they may be more abundant in deeper layers of the sampled material.
V, Mo, La , U , Ga, and Se show moderate positive correlations with Depth. This indicates a tendency for these elements to increase with depth (Carrillo-González et al., 2006), though the relationship is not as strong as with Sb or As.
Ag and Th have very weak correlations with Depth, suggesting that the concentrations of these elements do not change significantly with depth.
Li and Ni have weak to moderate negative correlations with Depth, implying that these elements may be more concentrated in the upper layers, decreasing as depth increases.

Descriptive Analysis of Data

Establishing low contamination of samples in the lysimeter

Fig 12. Concentration of trace elements found in the leachate of the polyethylene lysimeter.

The concentrations of TEs in the leachates of the UHMWPE were low. Zinc was the only trace element that raised concern, but still was very low (2.64 ppb).

Fig 12. The soil solutions sampled using the lysimeters show a wide array of trace elements with a wide range of concentrations.

Fig 13. Ratios of measured concentrations of the samples relative to the blank (deionized water).

Key Takeaways

Principal Component Analysis (PCA) and correlation analysis of the concentrations identified significant patterns and associations between certain elements and soil depth and time.
This suggests that trace elements with similar chemical properties such as Pb, As, and Cd tend to display distinct mobility patterns compared to trace elements such as Li and Ni which have different behavior in the soil.
The polyethylene lysimeters demonstrated low blank values for trace elements, indicating minimal release and adsorption of trace elements. The concentration of dissolved TEs in soil solutions were much higher compared to the blank (deionized water) values, with ratios ranging from 17 times for Molybdenum to 2,900 times for Lanthanum.
This implies that the metal-free lysimeter is a reliable tool for collecting soil solutions with a wide range of concentrations for trace elements of interest, and without significant contamination, enabling accurate measurement of TE concentrations.

References

Appelo, C. A. J., & Postma, Dieke. (2013). Geochemistry, groundwater and pollution. A.A. Balkema Publishers.

Carrillo-González, R., Šimůnek, J., Sauvé, S., & Adriano, D. (2006). Mechanisms and Pathways of Trace Element Mobility in Soils. In Advances in Agronomy (Vol. 91, pp. 111–178). https://doi.org/10.1016/S0065-2113(06)91003-7

Hooda, P. S. (2010). Trace elements in soils, 1st ed. John Wiley & Sons Ltd, West Sussex, United Kingdom

Kabata-Pendias, A., 2010. Trace Elements in Soils and Plants (4th ed.). CRC Press, Boca Raton, Florida, USA.

Shotyk, W. (2022). Environmental significance of trace elements in the Athabasca Bituminous Sands: facts and misconceptions. In Environmental Science: Processes and Impacts (Vol. 24, Issue 9, pp. 1279–1302). Royal Society of Chemistry. https://doi.org/10.1039/d2em00049k

Page updated

Report abuse