Using principal component analysis to explore multi-variable relationships

Fraino, Patricia E.

doi:10.1038/s43017-023-00414-z

Cited by 6 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Factor analysis with the principal component method (FA-PCA) serves as a dimensionality-reduction technique facilitating the exploration of relationships among multiple variables simultaneously. −, In this research, FA-PCA was carried out to identify four factors (Fs) based on eigenvalues (>1), collectively representing 92.65% of the data set. Prior to interpreting the results of FA-PCA, the Kaiser–Meyer–Olkin (KMO) and Bartlett’s sphericity tests were employed to assess the validity of the analysis. , The KMO result was 0.68, and Bartlett’s test yielded 466.37 (df = 120, p < 0.01), affirming the utility of FA-PCA for dimensionality reduction.…”

Section: Resultsmentioning

confidence: 99%

Nutrients and Rare Earth Elements in Surface Sediments of Hongze Lake (China) Using the DGT Technique: Spatial Distribution Pattern and Probabilistic Risk

Gu,

Ma,

Jordan

et al. 2024

ACS EST Water

View full text Add to dashboard Cite

The escalating environmental impact of rare earth elements (REEs) in freshwater ecosystems, exemplified by Hongze Lake's increasing pollution levels from sediment-bound nutrients and REEs due to human activities, necessitates a comprehensive study. Our objectives are to unravel the spatial distribution patterns of nutrients and REEs in Hongze Lake's surface sediments and assess their combined toxicity effects on aquatic biota, aiming to fill a critical gap in understanding the ecological and toxicological risks associated with these contaminants. This study employed the diffusive gradients in thin films (DGT) technique to analyze surface sediments in China's Hongze Lake for nutrients and REEs. Significant fluctuations were observed in the concentrations of phosphate (PO 4 −P), nitrate (NO 3 −N), and ammonium (NH 4 −N), with some REEs exceeding predicted no effect concentrations (PNEC) for aquatic organisms. Correlation analysis suggested complex interrelationships among these elements, indicating potential common pollution sources. Factor analysis condensed 16 variables into 4 factors, unveiling distinct spatial distribution patterns in the lake. Additionally, a probabilistic ecotoxicological risk assessment, integrating DGT-labile concentrations and toxicity data, revealed a 1.26% probability of surface sediments in Hongze Lake inducing a combined toxic effect on aquatic biota. These findings provide specific numerical insights into pollution sources and risks within the Hongze Lake ecosystem. The results offer valuable information for current and future lake management and conservation efforts.

show abstract