| Autor: |
Marion JW; College of Public Health, Division of Environmental Health Sciences, The Ohio State University, 1841 Neil Avenue, Columbus, Ohio 43210-1351, United States., Lee J, Wilkins JR 3rd, Lemeshow S, Lee C, Waletzko EJ, Buckley TJ |
| Jazyk: |
angličtina |
| Zdroj: |
Environmental science & technology [Environ Sci Technol] 2012 Apr 17; Vol. 46 (8), pp. 4523-31. Date of Electronic Publication: 2012 Mar 21. |
| DOI: |
10.1021/es203962u |
| Abstrakt: |
Current approaches for assessing human health risks associated with cyanotoxins often rely on the quantification of microcystin. Significant limitations of current approaches are cost and time to obtain a result. To address these challenges, a numerical index for screening microcystin risks above the World Health Organization's (WHO) low-risk threshold for microcystin was developed for eutrophic Midwestern U.S. lakes based on water quality results from 182 beach water samples collected from seven Ohio lakes. In 48 (26.4%) samples we observed microcystin concentrations as measured by ELISA that exceeded the 4 μg/L microcystin threshold. A multivariable logistic regression model using practical real-time measures of in vivo phycocyanin (by fluorometry) and secchi depth was constructed to estimate the probability of a beach sample exceeding 4 μg/L microcystin. The final model achieved statistical significance (p = 0.030) as well as good calibration (as measured by the goodness-of-fit test comparing observed to expected counts within deciles of risk based on the model, p = 0.329) and discrimination (as indicated by the area under the receiver-operator-curve (0.795)). These results demonstrate two rapid and practical measures of recreational water quality are effective in identifying "at risk" lake conditions warranting additional management (e.g., advisory and/or advanced testing). |
| Databáze: |
MEDLINE |
| Externí odkaz: |
|
