Estrogenicity patterns in the Swiss midland river Lützelmurg in relation to treated domestic sewage effluent discharges and hydrology

Abstract

Sewage treatment works (STW) discharge estrogenic effluent into rivers, which leads to variable estrogenicity of river water. Here, we characterize how the factors effluent and hydrology influence the estrogenicity of river water. We selected a river for which good hydrological data are available and collected water samples upstream and downstream from a STW discharge; effluent was sampled as well. Sampling took place during four 12-d periods, associated with the seasons, and always occurred in the morning. We also investigated the estrogenicity along the river, both by grab sampling and by passive sampling. Estrogens were analyzed by a recombinant yeast assay (YES); the estrogenicity of a sample was equated to the 17beta-estradiol standard of the YES (ng/L). Estrogenicity upstream from the STW was mostly close to the detection limit of the YES (maximum, 0.4 ng/L). Estrogenicity of effluent ranged between 0.2 and 7.7 ng/L; lower estrogenicity was associated with higher hydraulic retention times. Downstream from the STW, estrogenicity exceeded 1 ng/L on 25% of the days (maximum, 2.1 ng/L). Measured river water estrogenicity correlated positively and significantly with predicted estrogenicity based on effluent estrogenicity and effluent dilution factor. Grab samples taken along the river indicate that no significant sources of estrogens were upstream from the STW; downstream from the STW, the pattern of estrogenicity was highly variable. However, passive sampling showed that the estrogenicity of river water downstream from the STW decreased continuously with increasing distance from the STW, which is largely explained by dilution.

Estrogens in Swiss Rivers and Effluents – Sampling Matters

“…[22] Effluent EEQs were multiplied with the proportion of effluent in the river downstream from the discharge to calculate expected river EEQs. Measured and expected river EEQs matched rather well; [20] an important observation, as it provided strong support that the methods were appropriate, and that the observed variability in EEQs was not likely to be a method error, neither in Study I nor in Study II. As another link to Study I, it was observed that the average CV of the effluent EEQs was almost double that of the dilution factors (45% versus 25%).…”

“…In a general linearized model, HRT emerged as a significant determinant of effluent EEQ (P <0.01). [20] Whereas Johnson et al [23] showed the effect of HRT across sites, Study II showed that HRT clearly determines effluent EEQ at a single site.…”

“…Source [20] The aim of Study II was to investigate variability of EEQs in river water and also effluent in detail at a single site. An important criterion for the site selection was the availability of good data on river and effluent flow rates so that accurate dilution factors could be calculated.…”

“…In 2004, Petty and co-workers described a The association between HRT and EEQ was negative in all but one sampling period (small fi lled circles). [20] H y d r a u lic r e t e n t i on t i m e ( h ) novel sampler type that targets polar organic chemicals − the POCIS (polar organic chemical integrative sampler) − and successfully tested POCIS extracts in the YES. [29] For these reasons, POCIS were selected for a sampling campaign around five STWs.…”

“…This time also the effluent itself was sampled, so it would be possible to match EEQs from effluent with EEQs from river water. To do this, sampling locations had to be selected where good information is available on river flow rates (as in Study II [20] ). Second, the chemical analysis of samples was extended beyond the steroidal estrogens to include nonylphenol and bisphenol A, to allow for a more comprehensive comparison between YES and chemical analysis data.…”