Skip to main content
ARS Home » Midwest Area » St. Paul, Minnesota » Soil and Water Management Research » Research » Publications at this Location » Publication #306287
Title: Identifying sources of emerging organic contaminants in a mixed use watershed using principal components analysis

Author
item KARPUZCA, EKREM - University Of Minnesota
item FAIRBAIRN, DAVID - University Of Minnesota
item ARNOLD, WILLIAM - University Of Minnesota
item BARBER, BRIAN - University Of Minnesota
item KAUFENBERG, ELIZABETH - University Of Minnesota
item Koskinen, William
item NOVAK, PAIGE - University Of Minnesota
item Rice, Pamela
item SWACKHAMER, DEBORAH - University Of Minnesota

Submitted to: Environmental Science: Processes & Impacts
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/1/2014
Publication Date: 8/1/2014
Publication URL: http://handle.nal.usda.gov/10113/60491
Citation: Karpuzca, E., Fairbairn, D., Arnold, W., Barber, B., Kaufenberg, E., Koskinen, W.C., Novak, P., Rice, P.J., Swackhamer, D. 2014. Identifying sources of emerging organic contaminants in a mixed use watershed using principal components analysis. Environmental Science: Processes & Impacts. 16:2390-2399.

Interpretive Summary: Principal components analysis (PCA) was used to identify sources of emerging organic contaminants in the Zumbro River watershed in southeastern Minnesota. Two main principal components (PCs) were identified, which together explained more than 50% of the variance in the data. Principal Component 1 (PC1) was attributed to urban wastewater-derived sources, including municipal wastewater and residential septic tank effluents, while Principal Component 2 (PC2) was attributed to agricultural sources. The variances of the concentrations of cotinine, DEET and the prescription drugs carbamazepine, erythromycin and sulfamethoxazole were best explained by PC1, while the variances of the concentrations of the agricultural pesticides atrazine, metolachlor and acetochlor were best explained by PC2. Mixed use compounds carbaryl, iprodione and daidzein did not specifically group with either PC1 or PC2. Furthermore, despite the fact that caffeine and acetaminophen have been historically associated with human use, they could not be attributed to a single dominant land use category (e.g., urban/residential or agricultural). Contributions from septic systems did not clarify the source for these two compounds, suggesting that additional sources, such as runoff from biosolid-amended soils, may exist. With respect to contaminant source allocation, analysis of detection frequencies may not tell the whole story and should be used with caution as a supplementary tool to multivariate statistical techniques, especially at sites characterized by mixed land use. This study showed that principal components analysis is a powerful technique in identifying sources of emerging contaminants according to the structure of their concentration patterns across different sites and seasons. Our results indicated that septic systems from residential areas had an impact on surface water quality in the study area. Furthermore, acetaminophen and caffeine were not ideal markers for urban/residential contamination sources and may need to be reconsidered as such in other areas as well.

Technical Abstract: Principal components analysis (PCA) was used to identify sources of emerging organic contaminants in the Zumbro River watershed in southeastern Minnesota. Two main principal components (PCs) were identified, which together explained more than 50% of the variance in the data. Principal Component 1 (PC1) was attributed to urban wastewater-derived sources, including municipal wastewater and residential septic tank effluents, while Principal Component 2 (PC2) was attributed to agricultural sources. The variances of the concentrations of cotinine, DEET and the prescription drugs carbamazepine, erythromycin and sulfamethoxazole were best explained by PC1, while the variances of the concentrations of the agricultural pesticides atrazine, metolachlor and acetochlor were best explained by PC2. Mixed use compounds carbaryl, iprodione and daidzein did not specifically group with either PC1 or PC2. Furthermore, despite the fact that caffeine and acetaminophen have been historically associated with human use, they could not be attributed to a single dominant land use category (e.g., urban/residential or agricultural). Contributions from septic systems did not clarify the source for these two compounds, suggesting that additional sources, such as runoff from biosolid-amended soils, may exist. Based on these results, PCA may be a useful way to broadly categorize the sources of new and previously uncharacterized emerging contaminants or may help to clarify transport pathways in a given area. Acetaminophen and caffeine were not ideal markers for urban/residential contamination sources in the study area and may need to be reconsidered as such in other areas as well.