| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a USDA-ARS, George E. Brown, Jr., Salinity Laboratory, 450 W. Big Springs Road, Riverside CA 92507-4617
b Department of Plant and Soil Sciences, University of Delaware, Newark, DE 19716
* Corresponding author (sbradford{at}ussl.ars.usda.gov)
Received for publication January 25, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
X174 (a representative somatic coliphage) and MS2 (a male-specific RNA coliphage) transport was controlled by inactivation induced by the solid phase. This conclusion was based on comparison of coliphage transport behavior at 5 and 20°C, mass balance information, and numerical modeling. Comparison of somatic coliphage transport data in the presence and absence of manure suspension revealed much higher effluent concentrations in the presence of manure. This difference was attributed to lower inactivation and higher detachment rates. The observed coliphage transport behavior suggests that survival of viruses may be extended in the presence of manure suspensions, and that transport studies conducted in the absence of manure suspension may not accurately characterize the transport potential of viruses in manure-contaminated environments. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Dissemination of animal viruses in the environment occurs as a result of application of animal manure and contaminated water to agricultural lands, and partitioning of manure components to flowing water. Viruses in turn can be transported with water and manure suspensions to surface water and through soils to groundwater. Knowledge of the processes that control the transport and fate of animal viruses is therefore needed to assess the risk and vulnerability of water resources to contamination, and to develop cost-effective treatment strategies to minimize human and animal exposure.
Considerable research has been devoted to the fate and transport of viruses in porous media (Schijven and Hassanizadeh, 2000; Jin and Flury, 2002). Attachment, detachment, and inactivation mechanisms have been identified as key processes that control the fate of viruses in the environment. Attachment and detachment depend on virus-virus, virus-solvent and virus-porous media interactions (Elimelech and O'Melia, 1990). Although viruses can have variably charged surfaces due to the presence of ionizable amino acids in their protein capsid (Gerba, 1984), most possess a net negative charge at a neutral pH (Jin and Flury, 2002). Under these circumstances, virus deposition is believed to be controlled by attachment onto positively charged clay edges and metal (iron, aluminum, and manganese) oxide surfaces (Farrah and Preston, 1993; Jin et al., 1997; Lukasik et al., 1999; Zhuang and Jin, 2003a). The presence of metal oxides has also been reported to enhance virus inactivation (Sagripanti et al., 1993; Pieper et al., 1997; Schijven et al., 1999; Chu et al., 2001; Ryan et al., 2002).
Several researchers have examined the influence of organic matter on virus transport (Powelson et al., 1991; Pieper et al., 1997; Zhuang and Jin, 2003b; Foppen et al., 2006). In these studies, humic or fulic acids were commonly used as surrogates for sewage or manure effluent. Some researchers reported that dissolved organic matter (DOM) enhanced microbe transport (Pieper et al., 1997; Johnson and Logan, 1996; Powelson and Mills, 2001). Blocking of favorable attachment sites by organic matter has typically been used to explain this enhanced transport (Pieper et al., 1997; Johnson and Logan, 1996; Zhuang and Jin, 2003b; Foppen et al., 2006). DOM has also been reported to sorb onto microbes and alter their electrophoretic mobility (Gerritson and Bradley, 1987). Increasing the negative charge of microbe surfaces diminishes its attachment onto negatively charged solid surfaces (Sharma et al., 1985). Conversely, other researchers have reported that organic matter inhibits virus transport due to hydrophobic interactions between the virus and grain surfaces that are coated with organic matter (Bales et al., 1993; Kinoshita et al., 1993).
Manure suspensions consist of a complex mixture of partially digested organic matter and microbial biomass, and therefore encompasses a wide range in particle sizes. Viruses represent only a small mass fraction of the total particles in suspensions. Straining of larger manure particles in down-gradient pores that are too small to allow particle passage would decrease the effective size of the pores or fill the smaller pore spaces completely. The potential implications of this manure deposition on virus transport are not yet known. Deposition-induced changes in the soil pore sizes could promote virus retention via straining, or induce changes in the pore-scale water flow field that would confine viruses to more conductive (larger and less reactive) regions of the pore space. Alternatively, adsorption of viruses onto mobile manure colloids could also potentially facilitate their transport potential (Jin et al., 2000; de Jonge et al., 2004).
This study examines the transport of indigenous somatic coliphage in dairy calf manure suspension and two indicator viruses (coliphage X174 and MS2) in the absence of manure suspension. Special attention was given to mechanisms of virus attachment, detachment, straining, and inactivation. Effluent concentration curves and the final spatial distributions of the coliphage were measured in column transport studies, whereas inactivation behavior was quantified by comparing transport results at different temperatures. Data analysis and interpretation was aided through mathematical modeling, mass balance considerations, and measurement of the particle size distributions in the effluent.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
X174 are commonly associated with fecal contamination and are typically found in high concentrations in lagoon water (Havelaar et al., 1986). Hence, MS2 and X174 are recommended indicators for male-specific and somatic coliphage, respectively, in water and waste water (USEPA, 2001). Coliphage MS2 is an icosahedral single-stranded RNA phage with a diameter of 26.0 to 26.6 nm (Van Duin, 1988) and has an isoelectric point of 3.9 (Zerda, 1982). The electrostatic (zeta) potential of MS2 is –17.7 ± 2.3 mV in 0.01 M NaCl solution at a neutral pH (You et al., 2003). The surface of MS2 is negatively charged under most natural environmental pHs (Redman et al., 1997), has many similar physical properties to enteroviruses (Havelaar, 1993), and is a reasonable surrogate for enterovirus transport (Schijven and Hassanizadeh, 2000). The survival of MS2 was reported to be similar to that of human pathogenic viruses (Schijven and Hassanizadeh, 2000). Coliphage X174 is a spherical, single-stranded DNA coliphage with a 27-nm diameter and an isoelectric point of 6.6 (Dowd et al., 1998). Considered to be a relatively conservative indicator for human virus transport, X174 is environmentally stable and has low hydrophobicity (Schijven and Hassanizadeh, 2000).
Transport experiments were conducted using somatic coliphage that were indigenous to dairy calf manure suspension, MS2, and X174. The concentrations of these coliphages in experimental solutions were determined by the soft-agar layer method (Adams, 1959). Approximately 2 mL of sample (influent, effluent, and soil solution) was collected for analysis. If the samples contained manure suspension, they were centrifuged at 10 000 rpm for 10 min (20°C). A serial dilution was made from 1 mL of the supernatant (manure suspension) or sample. One-half mL of log phase host bacterial strain was added to 3 mL of the soft trypticase soy agar (supplemented with nalidixic acid for indigenous somatic coliphage and X174, and ampicillin and streptomycin for MS2) kept at 46°C with a water bath. The tube was immediately removed from the water bath and 100 µL of serial diluted solution was added. The content of the tube was gently mixed and poured onto the trypticase soy agar plate, swirled and allowed to solidify. The host bacterial strain for MS2 was Escherichia coli (ATCC #700891), whereas E. coli (ATCC #700609) was used as host for indigenous somatic coliphage and X174. Plates were inverted and incubated at 37°C for 12 to 24 h. The number of plaque forming units (pfu) was then determined from plate counts.
Experimental solutions (deionized water) consisted of 0.001 M NaBr (influent suspension) or 0.001 M NaCl (resident and eluant solution) buffered to a pH of 6.73 using 5 x 10–5 M NaHCO3. The electrical conductivity of this solution was 0.14 dS m–1. In the absence of manure suspension, MS2 and X174 were added to the 0.001 M NaBr (influent) solution at concentrations of 2 x 105 and 1.7 x 105 pfu mL–1, respectively.
Unless specifically noted the transport experiments were conducted at approximately 20°C. To investigate coliphage transport under conditions of minimal inactivation, an additional experiment was run in a constant temperature cold room at 5°C. In this case, MS2 and X174 were added to the 0.001 M NaBr (influent) solution at concentrations of 6.8 x 104 and 3.9 x 104 pfu mL–1, respectively.
Holstein dairy calf manure was collected directly under the crates of 1- to 12-wk-old calves, thoroughly mixed with a stick, and stored at 4°C before use. The manure suspension was prepared by mixing a known mass of manure (wet weight) with the 0.001 M NaBr solution. This suspension was then filtered through a 103-µm stainless steel wire mesh. The concentrated suspension was then diluted to achieve a concentration of approximately 4.0 g l–1 (mass based on unfiltered weight). The pH and electrical conductivity of the filtered manure suspension were 8.8 and 0.38 dS m–1, respectively. Particle size distribution information for the influent and selected effluent manure samples were determined using a Horiba LA 930 laser scattering particle size analyzer (Horiba Instruments, Irvine, CA 92614). The initial concentration of indigenous somatic coliphage in the manure suspension was approximately 1.2 x 104 pfu mL–1, whereas no male-specific coliphage was detected.
Ottawa aquifer sand (U.S. Silica, Ottawa, IL) was used in the transport experiments. The Ottawa sands will be designated herein by the median grain size (d50) as: 710, 360, 240, and 150 µm. The coefficient of uniformity (Ui = d60 /d10; here x% of the sand was finer than dx) of the 710, 360, 240, and 150 µm sands was 1.21, 1.88, 3.06, and 2.25, respectively. Pore size distribution information for these Ottawa sands can be calculated from the capillary pressure-saturation curve presented by Bradford and Abriola (2001). Ottawa sands typically consisted of 99.8% SiO2 (quartz) and trace amounts of metal oxides, were spheroidal in shape, and had rough surfaces. The vast majority of the sands possessed a net negative charge at a neutral pH.
Many of the protocols for the column experiments were described in detail by Bradford et al. (2002); only a short summary is provided below. Borosilicate glass chromatography columns (Kimble/Kontes, Vineland, NJ) (15-cm long and 4.8 cm i.d.) equipped with a standard flangeless end fitting at the column bottom and an adjustable flow adaptor at the top were used in the experiments. The columns were wet-packed with the various porous media. The coliphage tracer suspension (with and without manure suspension) or eluant (0.001 M NaCl) solution was pumped upward through the vertically oriented columns at a steady rate for 250 min. Table 1 provides the porosity (
), column length, the duration of the tracer suspension pulse, and the average aqueous Darcy velocity (q) for the various column experiments. Effluent samples were collected in glass test tubes over the course of each column experiment using an autosampler, and the concentration of coliphage was measured using the procedures outlined above.
|
Inactivation experiments were conducted in 20-mL glass scintillation vials filled with 20 mL of the coliphage suspension. Vials were capped and slowly mixed using a Labquake orbital shaker (Barnstead/Thermolyne, Dubuque, IA) for 250 min. Duplicate samples (100 µl) of excess suspension were periodically collected from the vials and analyzed for coliphage concentration.
The HYDRUS-1D computer code (Simunek et al., 1998; Bradford et al., 2003) was used to simulate coliphage transport and deposition in the column experiments. Aspects of HYDRUS-1D that are relevant to coliphage transport in the saturated column experiments are briefly discussed below. To minimize the number of fitted model parameters, the values of the hydrodynamic dispersivity (
) that were used in the simulations were taken from bromide tracer data presented by Bradford et al. (2002) for the same sands and a similar velocity.
The aqueous phase coliphage mass balance equation is written as:
 | [1] |
w [-] is the volumetric water content, JT [Nc L–2 T–1] is the total coliphage flux (sum of the advective, dispersive, and diffusive fluxes), µw [T–1] is the coliphage inactivation rate in the aqueous phase, and Esw [Nc L–3 T–1] is the coliphage mass transfer terms between the aqueous and solid phases.
Two model formulations for Esw will be considered in this work. The most complex 2-site kinetic model is written as:
 | [2] |
b [M L–3; M denotes mass] is the soil bulk density, Satt [Nc M–1] is the solid phase concentration of attached coliphage, Sstr [Nc m–1] is the solid phase concentration of strained coliphage, katt [T–1] is the attachment coefficient, kdet [T–1] is the detachment coefficient, kstr [T–1] is the straining coefficient, 
str [-] is a dimensionless straining function, and µs [T–1] is the coliphage inactivation rate on the solid phase. Coliphage attachment, detachment, straining, and solid phase inactivation are modeled using the first, second, third, and fourth terms on the right hand side of Eq. [2], respectively.
The value of str in Eq. [2] is modeled as a function of distance and Sstr as:
 | [3] |
For simplicity, an alternative 1-site kinetic formation for Esw is also considered as:
 | [4] |
1 [-] is a dimensionless deposition function. In contrast to Eq. [2], deposition processes are lumped together in effective parameters in Eq. [4] and no attempt is made to separate attachment and straining mechanisms. To have the most flexibility to describe spatial distribution data, the value of 1 in Eq. [4] is modeled in an analogous manner to Eq. [3]. In this case, however, 1 is viewed as an empirical parameter that characterizes the observed spatial distribution and a physical interpretation for this distribution is not implicitly assumed. |
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
Interpretation of effluent and spatial distribution data of the somatic coliphage was facilitated by considering the fate of other manure components in the suspension. Figure 2
presents the cumulative size distribution of manure particles in the suspension initially and after 95 min (around 2.2 pore volumes) of passage through the various sands. More comprehensive information on manure suspension transport and deposition in these same sands was recently presented by Bradford et al. (2006). The initial manure suspension encompasses a wide range of particle sizes (<103 µm). Manure particles larger than around 13, 2, 1, and 1 µm were completely removed after passage through the 710, 360, 240, and 150 µm sands, respectively, due to mechanical filtration. This corresponded to ratios of manure particle to median grain size of 0.4 to 1.8%. Since the sands encompassed a wide range of pore sizes, it is logical to anticipate that straining and/or mechanical filtration processes can also play an important role in the deposition of smaller particles. Furthermore, particles retained by straining and/or mechanical filtration can decrease the effective pore size. Deposition of particles smaller than 5.4 µm could therefore induce straining of 27 nm coliphage (X174) according to the criterium (27 x 100/5400 = 0.5%) proposed by Bradford et al. (2003).
|
Figures 1a and 1b also present simulated somatic coliphage transport data using the 1-site kinetic deposition model. In this case, both attachment and straining processes were lumped together and no attempt was made to separate the mechanisms. Water and solid phase inactivation rate coefficients were assumed to be zero, because the recovered amounts of coliphage were quite high (86.5 to 95.8%). To improve the description of the spatial distribution, values of k1, ß, and S*max (= Smax/Nic; where Nic = Ci x 1 mL) were initially fitted to only the deposition data. These fitted values served as initial estimates in the simulations that also included effluent data. In this case, values of k1 and kdet were fitted to the effluent data. Table 2 provides a summary of model parameters (k1, ß, S*max, kdet, and ), as well as statistical parameters (coefficient of linear regression to effluent, re2, and spatial distribution data, rs2) to characterize the goodness of parameter fits. The simulated behavior shown in Fig. 1a and 1b and statistical parameters in Table 2 indicates that this model provided a reasonable description of both effluent and spatial distribution data.
To better deduce mechanisms of somatic coliphage deposition, the transport data was also simulated using the 2-site kinetic model. To separate the magnitudes of straining and attachment the spatial distribution data was divided into two regions. The first region (dimensionless depths of approximately 0 to 0.5) was initially characterized using fitted 1-site kinetic model parameters (k1, ß, and S*max). The remaining portion of the spatial distribution was then characterized by fitting a value of katt (irreversible attachment). With values of katt determined by this procedure, new values of kstr, ß, and Sstr*max were then fitted to the complete spatial distribution data. To include the influence of detachment, the value of katt was increased by a factor equal to kdet (Table 2). Table 3 provides a summary of model parameters (kstr, ß, Sstr*max, katt, kdet, and ), and statistical parameters for the goodness of model fit. The 2-site model simulations were very comparable to the 1-site modeling results presented in Fig. 1a and 1b, and were therefore not shown. The estimated percentage of somatic coliphage deposition due to straining for the 710, 360, 240, and 150 µm sand was 16, 38, 32, and 42%, respectively. Values of katt minus kdet (Table 3) increased with decreasing sand size, suggesting that attachment increased with increasing surface area of the sand.
|
X174, respectively, in 710-, 360-, 240-, and 150-µm Ottawa sands. The percentage of coliphage recovered in the effluent (Meff) is given in Table 1. Decreasing the sand size tended to produce lower effluent concentrations for MS2 (Meff ranged from 59.9 to 38.9%). The sand gradation may have also played a role in transport, since the more graded 240-µm sand exhibited slightly lower effluent concentrations (Meff = 38.9%) than the 150-µm sand (Meff = 44.3%). In contrast, X174 exhibited the opposite trend with respect to sand size and gradation than MS2 (Meff ranged from 46.6 to 67.2%).
|
X174 spatial distribution data in the various sands, respectively. The spatial distributions were highly dependent on the coliphage type and the sand size. In Fig. 4a most of the MS2 deposition occurred at the column inlet, with increasing retention occurring with decreasing sand size. In contrast, X174 exhibited more uniform deposition with depth.
|
X174. The total amounts of coliphage (Mtotal = Msand + Meff) that were recovered were also quite low, suggesting that the remaining coliphage was either irreversibly sorbed to the sands or inactivated. Losses in the column setup were quantified by running a blank column experiment (similar flow rate and volume) without any sand. Influent and final effluent concentrations after 250 min were very similar, suggesting that minimal coliphage losses were attributable to the experimental setup.
To better understand mechanisms of solid inactivation and/or irreversible sorption, additional column experiments were conducted for X174 and MS2 in 150-µm sand at an experimental temperature of 5°C. Significantly less inactivation was expected at 5 than 20°C (Schijven and Hassanizadeh, 2000; Jin and Flury, 2002). Figures 5a
and 5b present effluent concentration curves (Fig. 5a) and spatial distribution data (Fig. 5b) for X174 and MS2 in 150-µm sands at 5 and 20°C. Peak effluent concentration curves were much higher at 5 than at 20°C. Filtration theory predicts that the attachment coefficient will only slightly decrease (around 3.5% for a change in temperature from 20 to 5°C) with decreasing temperature (McCaulou et al., 1995). Hence, differences in transport behavior at 5 and 20°C were primarily due to inactivation. Comparison of Meff values at 20°C (Meff equaled 59% for X174 and 44% for MS2) and 5°C (Meff equaled 88% for X174 and 76% for MS2) in this sand suggests that 29% of the X174 and 32% of the MS2 were inactivated at 20°C in the 150-µm sands. The very low values of Msand (0.2 to 0.3%) in both 5 and 20°C systems also suggests that solid phase inactivation was an important elimination process. The presence of trace amounts of metal oxides and/or metal ions on sand surfaces has been reported to control solid phase inactivation (Sagripanti et al., 1993; Pieper et al., 1997; Schijven et al., 1999; Chu et al., 2001; Ryan et al., 2002).
|
|
X174 and MS2 in 150-µm sands at 5 and 20°C. The 1-site kinetic model was used to describe this transport data, and no attempt was made to separately deduce mechanisms of attachment and straining because of the confounding influence of inactivation. At 5°C we assumed that water phase inactivation was negligible (consistent with batch inactivation results discussed below) and values of k1, kdet, S*max, ß and µs were determined from the transport data. At 20°C values of k1, S*max, ß, and kdet were taken from the corresponding 5°C data (assuming that temperature had only a minor influence on deposition), and values of µs and µw were directly fitted to effluent and spatial distribution data, respectively. Table 4 summarizes the fitted model parameters, as well as statistical parameters for the goodness of the model fit to the data. The simulations provided a reasonable description of both effluent and spatial distribution data at 5 and 20°C.
Transport data for MS2 in the 710-, 360-, and 240-µm sands at 20°C were simulated using values of µs, µw, and kdet estimated from the 150-µm sand data, and by fitting values of k1, S*max, and ß to the transport data. A similar approach was taken for the 20°C data for X174 in these same sands, except a separate value of µw had to be fitted to the data from the 710- and 360-µm sands. Table 4 also summarizes these fitted model parameters. The values of re2 and rs2 in Table 4 indicate that this approach generally provided a reasonable description of the transport data. Values of rs2 for X174 data were sometimes low due to scatter in the measurement points.
The simulations discussed above provide insight on X174 and MS2 inactivation in the various sands. Table 4 implies that changes in the sand size had a minimal impact on µs for a given coliphage. The dependence on sand size was presumably already accounted for by the product of µs and S in Eq. [4], where S (deposition) implicitly included a dependence on the sand size. The value of µs in a given sand was much higher (77 to 88 times greater) for MS2 than X174, indicative of an increased sensitivity to solid phase inactivation. Table 4 also indicates that similar values of µs occurred for a particular coliphage at 5 and 20°C in 150-µm sand. This observation suggests that solid phase inactivation can still be important even at 5°C.
The relative concentrations of X174 and MS2 slowly decreased with increasing time in batch studies conducted at 20°C in the absence of sand. The value of µw was assumed to be first order and therefore determined as ln(C/Ci)/te; where te is the equilibration time. The batch value of µw was measured to be 0.03 h–1 for MS2 and 0.01 h–1 for X174. In general, X174 is more stable (survives longer) than MS2 (DeBorde et al., 1998). Measured batch values of µw were much smaller (0.03 h–1 compared to 0.86 h–1 for MS2; and 0.01 h–1 compared to 0.53 to 1.00 h–1 for X174) than those determined in the column studies (Table 4) and suggests that the solid phase impacted the measured value of µw in the column experiments. In further support of DeBorde's hypothesis, note in Table 4 that µw was similar in magnitude to µs for MS2 at 20°C. Schijven et al. (1999) also found that inactivation of deposited MS2 was similar to that of free MS2 in the water. In contrast, for X174 the value of µw was 53 to 100 times greater than µs, with higher values of µw occurring in the coarser textured sands. Water phase X174 was apparently more susceptible to inactivation than deposited X174, especially in the coarser textured sands. We hypothesize that dissolution of metal ions into the aqueous phase may have influenced the water phase inactivation rates for X174 (Sagripanti et al., 1993). The presence of different types of metals and surface oxide coatings in the various sands may have also contributed to differences in µw in the various sands.
Somatic Coliphage Transport in the Presence and Absence of Manure Suspension
The coliphage X174 is a recommended indicator for somatic coliphage in water and waste water (USEPA, 2001). In this section we compare and contrast transport behavior of X174 in the absence of manure suspension with indigenous somatic coliphage in manure suspension. Effluent and spatial distribution concentrations for the somatic coliphage were much higher in the presence (Fig. 1a and 1b) than in the absence of manure suspensions (Fig. 3b and 4b). For a given sand, the difference in Mtotal in the presence and absence of manure was 27.1 to 46.8% (Table 1), with greater differences occurring in coarser textured sands. Tables 2 and 4 indicate that the deposition coefficient was actually much higher in the presence of manure suspension than in the absence. Hence, all of these observations suggest that inactivation losses for somatic coliphage were much lower in the presence of manure suspension. This has important implications for virus fate in the field, and suggests that the viability of viruses may be enhanced in the presence of manure suspensions. Hence, transport studies conducted in the absence of manure suspension may not accurately characterize the transport potential of viruses in manure-contaminated environments.
The deposition coefficient (straining + attachment) was much higher in the presence of manure suspension than in the absence (Tables 2 and 4) due to differences in straining and attachment. Straining is expected to be much more significant in the presence of manure suspension than in the absence due to deposition of larger manure particles that can induce changes in the pore structure. Attachment may also be greater in the presence of manure suspension due to increases in the solution salinity (0.14 dS m–1 in the absence of manure suspension and 0.38 dS m–1 in the presence) (Zhuang and Jin, 2003b; Deshpande and Shonnard, 1999). Conversely, others researchers have postulated that attachment decreases in the presence of manure suspension due to blocking of favorable attachment sites (Pieper et al., 1997; Johnson and Logan, 1996; Powelson and Mills, 2001). We were unable to test this hypothesis due to the confounding influences of inactivation and straining.
After recovery of the bulk of the effluent concentration pulse, the coliphage breakthrough curves exhibited low concentration tailing (C/Ci values of around 10–3 to 10–4). Figure 6
presents a semilog plot of observed and simulated somatic coliphage effluent concentration curves for 150-µm sand in the presence and absence of manure suspension. One-site kinetic deposition model parameters are provided in Tables 2 and 4. The observed tailing behavior shown in Fig. 6 is fairly accurately described using a first-order detachment term. After around 3 PV, both effluent concentration curves exhibited low concentration tailing behavior, slowly decreasing with continued flushing with 0.001 M NaCl solution as a result of detachment. The relative concentration of somatic coliphage in the manure suspension was around 2.5 orders of magnitude higher than in the absence. Similar concentration tailing behavior was observed for somatic coliphage in the other sands. Higher effluent concentrations in the presence of manure suspension were previously (when PV were <3) attributed to lower inactivation rates. In the tailing region (>3 PV), higher effluent concentrations may also be due to higher detachment rates of somatic coliphage from manure particles and/or manure-induced changes in the sand surface properties. In the absence of manure suspension, both MS2 and X174 exhibited similar levels of low concentration tailing (C/Ci = 10–3 to 10–4) for the various sands.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Interpretation of effluent and spatial distribution data for the somatic coliphage was facilitated by considering the fate of other manure components in the suspension. The cumulative size distribution of manure components in the suspension initially and after passage through the packed columns was measured, and used to identify the mechanical filtration potential of these sands. Manure particles were completely removed by mechanical filtration when dp/d50 (dp = manure particle diameter) was greater than 0.4 to 1.8%. This suggests that particles retained by straining and/or mechanical filtration can decrease the effective pore size, and may induce straining of much smaller particles such as coliphage. Comparison of somatic coliphage concentrations in the manure suspension before and after centrifugation suggested that the vast majority of the somatic coliphage was not attached to the manure particles.
A 2-site kinetic deposition model was used to estimate the magnitudes of somatic coliphage attachment and straining in the various sands in the presence of the manure suspension. The magnitude of attachment was estimated from the spatial distribution data in the region that exhibited relatively uniform deposition with depth (dimensionless depths 0.2 to 1). With fixed values of the attachment coefficient, the straining parameters were then fitted to the spatial distribution data. This modeling approach provided a good description for both effluent and spatial distribution data, and indicated that straining accounted for 16 to 42% of the deposited somatic coliphage in the various sands. Magnitudes of both straining and attachment increased with decreasing sand size due to smaller pores and higher surface area, respectively.
In the absence of manure suspension, the transport of X174 and MS2 was controlled by solid and liquid phase inactivation. This conclusion was based on comparison of transport behavior for these coliphage at 5 and 20°C, mass balance information, and numerical modeling. At 5°C water phase inactivation was low and effluent concentration for X174 and MS2 were high in the finest sand. In this case, deposition coefficients and solid phase inactivation rates were fitted to the transport data, and the numerical simulations provided a reasonable description of the observed data. Fitted values of the solid phase inactivation rate suggested that this process can still be important at 5°C (especially for MS2), presumably due to interactions with trace amounts of metal oxides and/or metals on sand grain surfaces. Effluent concentration for X174 and MS2 in the finest sand at 20°C were much lower than at 5°C. Fitted values of the water phase inactivation rate were higher than those measured in the influent suspension, suggesting that the presence of the solid phase also influenced the water phase inactivation rate in these systems.
Comparison of somatic coliphage transport data in the presence (indigenous somatic coliphage) and absence (X174) of manure suspension revealed much higher effluent concentrations in the presence of manure. This observation was attributed to differences in inactivation. In the presence of manure suspension, much lower inactivation occurred, presumably due to sorption of organic components onto metals and/or metal oxide surfaces that otherwise would have induced inactivation (Foppen et al., 2006). After recovery of the bulk of the breakthrough curve, low concentration tailing was observed for both indigenous somatic coliphage and X174. In the presence of manure, the relative concentration for somatic coliphage in this tailing region was around 2.5 orders of magnitude higher in the presence than in the absence of manure. This may be due to lower inactivation rates and higher detachment rates in the presence of manure. The observed coliphage transport behavior has important implications for virus fate in the environment, and suggests that the viability of viruses may be enhanced in the presence of manure suspensions. Hence, transport studies conducted in the absence of manure suspension may not accurately characterize the transport potential of viruses in manure-contaminated environments.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. L. Pepper, H. Zerzghi, J. P. Brooks, and C. P. Gerba Sustainability of Land Application of Class B Biosolids J. Environ. Qual., September 2, 2008; 37(5_Supplement): S-58 - S-67. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Bradford, E. Segal, W. Zheng, Q. Wang, and S. R. Hutchins Reuse of Concentrated Animal Feeding Operation Wastewater on Agricultural Lands J. Environ. Qual., September 2, 2008; 37(5_Supplement): S-97 - S-115. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
Mtotal) for somatic coliphage in the presence (indigenous) and absence (
