The Fate and Transport of Viruses through Surface Water Constructed Wetlands

Abstract

Coliphage removal efficiency and the effects of wetland hydrology on virus transport were determined for constructed wetlands in San Jacinto, CA. Mathematical models were used to further characterize virus transport. MS2, an F‐specific RNA (FRNA) coliphage was used as a model for human enteric viral behavior. Two wetland types were studied, a one‐phase cell and three‐phase cell. These wetlands received unchlorinated secondary effluent at a constant rate. The mean residence time in the wetlands was 9 ± 3 d as determined using bromide as a conservative tracer. Assuming 100% porosity, a plug flow model predicts this mean residence time within the experimental standard deviation (8 d). This suggests that a negligible volume was occupied by vegetation and settled solids. The convection‐dispersion equation adequately simulated the residence time distribution of the conservative tracer. MS2 removal in the wetlands was experimentally determined to be 97 ± 3%. There was no distinction between the two wetland types in terms of removal efficiency. The average coliphage decay rate was calculated to be 0.44 per day. However, the error involved with using the first order decay rate was high, 83 ± 12%. Therefore, first order decay does not adequately describe removal processes within the wetland. Most virus removal occurred within the first 3 m (k = 4.0 ± 1.8 d−1) with a removal efficiency of 85.3 ± 0.6%. The remainder had a decay rate of 0.20 ± 0.17 d−1 with a removal efficiency of 56 ± 33%.

Transport of Coliphage PRD1 in a Surface Flow Constructed Wetland

“…Compared with bromide, PRD1 concentrations showed a considerable range of variability in the experimental data. These discrepancies between PRD1 experimental and predicted concentrations, mostly at site 2 ( r 2 = 0.68), are similar to other wetland studies and are likely a result of deficient mixing of the virus in the wetland ( Chendorain et al, 1998 ; Gersberg et al, 1987 ; Sheuerman et al, 1989 ). At site 3, an r 2 of 0.94 was estimated by comparing the best CDE simulation with PRD1 experimental concentrations.…”

“…Other studies have found inactivation rates in secondary effluent ranging from 0.01 to 0.18 d ‐1 at 10 to 25°C. In contrast to PRD1, a decay rate of 0.35 d ‐1 and 93 to 98% removal for MS2 have been reported in surface flow wetlands with a detention time of 7.65 days ( Chendorain et al, 1998 ).…”

Prediction of velocity profiles and longitudinal dispersion in salt marsh vegetation

“…7 indicates that stem-shear dispersion can be neglected at distances of Ͼ80 cm, based on our measured velocity profile and Eq. 12, we estimate that the normalized depth-shear dis- Chendorain et al (1998) obtained a value of K x /Uh ϭ 50 in a study of longitudinal dispersion in a 46-cm-deep, 70-mlong emergent Scirpus californicus and Scirpus acutus wetland; Scirpus reeds have a similar, though not identical, morphology to Spartina species.…”

Influence of hummocks and emergent vegetation on hydraulic performance in a surface flow wastewater treatment wetland

“…Fractional recovery of the bromide tracer was determined by integrating the concentration‐flow time series at the wetland outlet after correcting for the background bromide concentration ( C background ) and tracer loss due to groundwater seepage, where m is mass of tracer added (mg), t is time (s), and C seep is the tracer concentration associated with groundwater seepage (mg L −1 ), as given by the average of the inlet ( C in ) and outlet ( C out ) concentrations, Fractional recoveries less than 1.0 (Table 1) indicate a loss of bromide that is attributable to experimental uncertainties in wetland tracer tests [ Chendorain et al , 1998; Lin et al , 2003]. Outlet tracer concentration data were therefore scaled to account for bromide loss [ Atkinson and Davis , 2000], where C scaled is the scaled tracer concentration (mg L −1 ) used in the subsequent moments analysis and solute transport modeling.…”