JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 May 2009
Published in J Environ Qual 38:1066-1075 (2009)
DOI: 10.2134/jeq2008.0085
© 2009 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Open Access Article
This Article
Free via Open Access
Right arrow OA Abstract
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow An erratum has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gottschall, N.
Right arrow Articles by Lapen, D. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gottschall, N.
Right arrow Articles by Lapen, D. R.
Agricola
Right arrow Articles by Gottschall, N.
Right arrow Articles by Lapen, D. R.
Related Collections
Right arrow Water Quality
Right arrow Municipal Wastes
Right arrow Vadose Zone Processes and Chemical Transport
Right arrow Municipal Waste

TECHNICAL REPORTS

Vadose Zone Processes and Chemical Transport

Nitrogen, Phosphorus, and Bacteria Tile and Groundwater Quality Following Direct Injection of Dewatered Municipal Biosolids into Soil

N. Gottschalla, M. Edwardsa, E. Toppb, P. Boltona, M. Paynec, W. E. Curnoed, B. Ball Coelhob and D. R. Lapena,*

a Agriculture and Agri-Food Canada, Ottawa, ON, Canada, K1A 0C6
b Agriculture and Agri-Food Canada, London, ON
c Ontario Ministry of Agriculture, Food, and Rural Affairs, Stratford, ON
d Univ. of Guelph-Kemptville, Kemptville, ON

* Corresponding author (david.lapen{at}agr.gc.ca).

Received for publication February 14, 2008.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Methods and Materials
 Results
 Discussion
 Conclusions
 REFERENCES
 
Application of municipal biosolids (sewage) to agricultural land is a common practice to improve soil physical quality and fertility. The chosen method of land application can have a strong impact on the extent of adjacent water contamination by nutrients and bacteria. Dewatered municipal biosolids (DMB) were applied to silt-clay loam experimental field plots in Ontario, Canada using two application methods: (i) surface spreading followed by shallow incorporation (SS) and (ii) a newly developed implement that directly injects DMB into the topsoil (DI). The objective of this study was to compare N, P, and bacteria quality of tile drainage and shallow groundwater associated with each land application technique. There were no significant differences (P > 0.05) in N, P, and bacteria tile mass loads among the application treatments for time periods <100 d postapplication, when the greatest peak loads and peak tile water concentrations were observed. Both land application treatments caused groundwater Escherichia coli contamination to at least 1.2 m depth below surface after the first postapplication rainfall event, and NO3–N contamination to at least 2.0 m depth below surface. The DI treatment did, however, have significantly (P < 0.05) higher tile mass loads of total Kjeldahl N (TKN), total phosphorus (TP), E. coli, Enterococci, and Clostridium perfringens relative to the SS treatment for time periods >100 d postapplication. Nevertheless, relative to tile effluent data collected <100 d postapplication (no application treatment differences), peak loads, and concentrations during this time were, overall, considerably lower for both treatments. This finding, along with no significant differences in N, P, and bacteria groundwater concentrations among the application treatments, and that the direct injection technique could potentially reduce vector attraction problems and odor, suggests that the direct injection technique should be considered a dewatered municipal biosolid land application option.

Abbreviations: DMB, dewatered municipal biosolids • BDL, below detection limit • C, control • DI, direct injection • DOY, day of year • LMB, liquid municipal biosolids • SS, surface spreading • TKN, total Kjeldahl nitrogen • TP, total phosphorus


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Methods and Materials
 Results
 Discussion
 Conclusions
 REFERENCES
 
LAND application of municipal biosolids is sanctioned in many countries and jurisdictions (USEPA, 1999; European Commission, 2001; Schut, 2005). The material can serve as a source of organic material as well as nutrients for crop production (Joshua et al., 1998; Mantovi et al., 2005). The solids content of biosolids applied to land can vary; being typically >20% wet weight for dewatered municipal biosolids (DMB), to <1% wet weight for liquid municipal biosolids (LMB) (Schut, 2005). Dewatered materials may be considered more economically viable for land application, relative to LMBs, since reduced water content will reduce the cost of transportation of the waste to field sites (Wakeman et al., 1976). However, dewatering can effectively concentrate contaminant constituents relative to their liquid or wastewater counterparts that have much lower total solids contents (Withers et al., 2001; Mantovi et al., 2005), and moreover, dewatered biosolids may enhance vector attraction (e.g., insects) (USEPA, 1995; Labud et al., 2003), and can be more odorous than slurry or liquid municipal biosolids (Zhang and Westerman, 1997; Janiec, 2006).

Guidelines for the land application of biosolids in Ontario, Canada attempt to ensure the protection of surface and groundwater resources by imposing distance restrictions and measures related to the time of application, site characteristics, and nutrient, bacteria, and metal contents of the biosolids (Ministry of the Environment and Energy and Ministry of Agriculture, Food, and Rural Affairs, 1996; Government of Ontario, 2002). Nevertheless, while advanced wastewater treatment processes significantly reduce pathogenic microorganisms in the waste material, some do remain recalcitrant to treatment (Chauret et al., 1999; Gantzer et al., 2001). Dewatering digested biosolids through centrifugation has been shown to increase the numbers of fecal bacteria, which may indicate potential regrowth or reactivation of pathogenic microorganisms (Higgins et al., 2007; Qi et al., 2007).

The mode of land application of nutrient amendments can impart a strong effect on the degree of pollution by nutrients, bacteria, and other contaminants in artificial subsurface drainage systems like tile drainage networks (Kay et al., 2004; Heathwaite et al., 2006; Ball-Coelho et al., 2007; Lapen et al., 2008a, 2008b). Dewatered biosolids are typically applied to land via a variety of surface spreading techniques (Evanylo, 1999; Brooks et al., 2005; Janiec, 2006); however, new dewatered biosolid topsoil injection technologies may serve to alleviate problems associated with, in particular, odor, vector attraction, and surface runoff potential (Janiec, 2006). But the impact of such a practice on groundwater and tile drainage remains unclear, and in terms of tile drainage specifically, regulations regarding biosolid land application are limited (National Research Council, 2002). While biosolid decomposition has been shown to be greater when the material is surface applied in contrast to when the material is incorporated in the soil (Terry et al., 1979), the nature of bacteria survival in injected biosolids, relative to biosolids that are surface applied is not well documented.

The objectives of this study were to: (1) compare groundwater concentrations and tile drain mass loads of selected nutrients (i.e., forms of N and P) and bacteria (i.e., E. coli., Enterococci, and C. perfringens) resulting from the land application of DMB via two contrasting DMB land application methods: (i) direct injection (DI) and (ii) surface spreading followed by incorporation (SS); and (2) compare bacteria concentrations of the land applied DMB applied via the two land application methods.


   Methods and Materials
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Methods and Materials
Results
 Discussion
 Conclusions
 REFERENCES
 
Site Description and History
The study was conducted in Winchester, ON, Canada (45°03' N, 75°21' W) from July to December 2006 on experimental field plots with silt-clay loam soils. Crops on this field have consisted historically of corn (Zea mays L.), soybean (Glycine max L.), and wheat (Triticum spp.) under conventional and no-tillage practices. The site was not cultivated in 2005 until the land application of liquid municipal biosolids in October of that year (Lapen et al., 2008a). After the 2005 LMB application, the field was left uncultivated until the July 2006 land application of DMB by SS and DI described in this paper. Two days after application, the plots were rolled over once with a roller/packer to prepare the soil surface for direct drilling of short-season soybean. Weather data collected at the site, and tile effluent monitoring details are given in Lapen et al. (2008a).

Experimental Setup
Eight plots each being 100 m length (length of tile drain contributing area) centered over tile lines (100 mm diam. plastic tiles for six plots and 100 mm diam. clay tiles for two plots) were used in this study (Lapen et al., 2008a). Tiles were approximately 0.8 m below the surface at their deepest point along the plots. Of the eight plots at the site, the six plastic-tiled plots were designated for application treatments. The remaining two ceramic-tiled plots were used as "controls" and were located on an adjacent field that was hydrologically isolated from the application treatment plots. There were three plot replications for each application treatment: these plots alternated spatially as per Lapen et al. (2008a). T1(DI), T3(DI), and T5(DI) refer to application treatment plots where DI was performed, whereas T2(SS), T4(SS), and T6(SS) were application treatment plots associated with the SS treatment. C1(DI) and C2(SS) were the control plot designations, and tillage and planting operations for these plots were conducted to mimic those performed on the respective application treatment plots, using the same equipment before it came in contact with biosolids.

Dewatered Municipal Biosolids and Land Application
The DMB used in this study were anaerobically digested and dewatered with a high speed centrifuge. Subsequent to centrifuging, the material was pumped into dry box containers, and those containers were trucked to the study site. The DMB was land applied at a rate of approximately 8 Mt dry weight ha–1 using two distinctly different land application methods. The amount of material that was applied over the entire experimental field (~1.05 ha) over 2 yr (LMB and DMB application), was ~4 Mt dry weight; a value under the maximum allowable amount per 5 yr of approximately 8 Mt dry weight (Ministry of the Environment and Energy and Ministry of Agriculture, Food, and Rural Affairs, 1996). The land application approaches consisted of: (i) surface spreading using a Hydra-spread II spreader (Hagedorn, Paisley, ON, Canada) with subsequent (within a few hours postapplication) DMB incorporation (~0.1-m depth) using a Kongskilde vibro-shank cultivator (Kongskilde, Exeter, ON) (SS application treatment) and (ii) Terratec Environmental Ltd.'s dewatered biosolids direct injection system (DI application treatment) (e.g., Janiec, 2006). The DI system consisted of a dry box container which uses a hydraulic ram push blade to move DMB to the rear of the container where a high pressure pump moves the DMB into a distribution box. The material is then distributed to a suite of sweep-type injectors (McLaughlin et al., 2006) fitted on a tool bar mounted on the back of the ram box. The injector tool bar consisted of five up front and four rear tines with a tine spacing of approximately 0.4 m (tool bar width approximately 4 m). The injector tine disturbance is approximately 0.3 m wide by 0.11 m (nominal depth of injection). The injected DMB bands are roughly 0.03 to 0.05 m in diameter leaving the drop tube, however, press plates fitted on the back of the injection sweeps flatten the material somewhat to augment soil contact area. Each application approach (DI and SS) consisted of two implement passes, making for a plot application area of approximately 8 m width x 100 m length centered over the tile.

Sampling and Statistical Analyses
Land application of DMB took place on Day of Year (DOY) 188 (7 July 2006). Systematic and hydrograph-based tile drainage monitoring took place from DOY 170 (20 June ) to DOY 321 (17 November), whereas groundwater monitoring took place from DOY 151 (31May ) to DOY 340 (6 December), every 2 wk. Tile water was sampled during rain events (tile hydrograph events) pre- and postapplication using ISCO 6712 (ISCO Inc., Lincoln, NE) automated water samplers, set to trigger when a rainfall depth of 5 mm h–1 (summer) or 7 mm h–1 (fall) was achieved. Samples were also collected throughout the season during tile baseflow conditions. Shallow groundwater was monitored via piezometers at depths of 1.2 and 2.0 m below the surface. To preserve sample integrity, bottles were transported in ice-packed coolers back to the laboratory, where samples were refrigerated until analysis was conducted. If samples could not be analyzed for N and P within 1 wk, they were frozen until they could be processed (maximum 2 wk). Bacterial analyses were conducted within 24 h of sample collection. The N and P analyses (NO3–N, NH4–N, total Kjeldahl N (TKN), PO4–P, and TP) were performed at the Laboratory Services Branch of the Ontario Ministry of the Environment, in Etobicoke, ON. All TKN and TP concentrations were determined by colorimetry, preceded by acid digestion, according to Ontario Ministry of Environment (MOE) method E3368. The other N and P species were determined colorimetrically according to MOE method E3366.

Escherichia coli and Enterococci were enumerated using the Colilert and Enterolert (IDEXX Laboratories, Inc., Westbrook, MA) systems, which are based on the most probable number (MPN) method. Dilutions were employed, where necessary, to determine MPN for each sample. Clostridium perfringens was determined by MPN using a modified m-CP medium (Armon and Payment, 1988). Clostridium perfringens, a spore former, is often an indicator of human-based fecal contamination (Payment and Franco, 1993). The detection limit for bacterial analyses was 10 CFU L–1.

Escherichia coli concentrations were measured in land applied DMB samples from the SS and DI treatment plots, whereas soil was sampled for E. coli in the control plots over the course of the study period. For DI plots, DMB was uncovered and sampled directly at three separate locations within T3(DI); whereas on T6(SS), grab samples of DMB on the soil surface to incorporation depth, were taken at five locations at specific times postapplication. In addition, three replicate soil samples (0–0.2 m depth) were taken from control plots three times throughout the study period (preapplication, DOY 257 and 325). The DMB and soil samples were processed by mixing 20 g (wet weight) of sampled material with 180 mL of distilled water, and E. coli was determined on the soil/biosolid suspensions at Accutest laboratories (Ottawa, ON, Canada), according to standard methods procedure 9221F, and an Ontario Ministry of Environment procedure for analysis for sediment bacteria (Ley, 1991; APHA, 2005). A field olfactometer (Nasal Ranger, St. Croix Sensory Inc., Lake Elmo, MN) was used to measure odor (Nasal Ranger value) from both DI and SS plots.

Both ANOVAs and t tests, using SYSTAT v. 10 (SPSS Inc., Chicago, IL), were used to compare tile mass loads (for N, P, and bacteria) associated with the land application treatments and controls. Tile mass load data were compared within data groupings defined in terms of "early" (<100 d postapplication) and "late" (>100 d postapplication) study period (groupings partitioned on the basis of a mid-study season dry spell that resulted in no tile flow), whereas groundwater N, P, and bacteria concentrations were statistically compared primarily via ANOVA on a seasonal (DOY 199–340) basis which included all samples after the first rain event postapplication. Data were log-transformed to better meet the statistical assumptions of normality and homogeneity of variance. A significance level of 0.05 was used for all statistical analyses.


   Results
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Methods and Materials
 Results
Discussion
 Conclusions
 REFERENCES
 
General Background Information
Hourly total rainfall, tile discharge for tile T1(DI), and average daily air temperature during the study period are presented in Fig. 1 . Total precipitation that occurred between DOY 188 (day of application) and DOY 321 was 403 mm. There was no tile flow that occurred between roughly DOY 220 (8 August) and DOY 289 (16 October). Figure 2 shows soil water content and temperature by soil depth during the study period. Soil water content ranged from 12 to 28% volume for 10 cm, and 16 to 29% volume for 20 cm depth in soil; temperature ranged from 4 to 30°C for 10 cm and 4 to 22°C for 20 cm depth for the study period. Key parameters for the DMB are presented in Table 1 , including moisture content, N, P, and initial bacteria concentrations. The bulk density of the DMB extruded from the drop tubes of the DI system averaged 0.29 g cm–3 with a standard deviation of 0.03 g cm–3. The odor (Nasal Ranger value) associated with SS was nearly five times greater than that for the DI technique at application; but SS values were lower and generally on par with DI 48 h after application (M. Janiec and Conestoga-Rovers and Assoc., Agricultural Services Division, Waterloo ON, unpublished data, 2006).


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 1. Average daily air temperature, hourly precipitation, and hourly tile flow for the 2006 study period. Tile flow is for T1(DI).

 

Figure 2
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 2. Soil water content (%) and temperature (°C) at 10 and 20 cm depths.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Selected parameters for dewatered municipal biosolid (DMB) used in this study. Method detection limits are 0.01%, 1 mg kg–1 and 10 CFU L–1 for respective parameters.

 
Tile Water Quality
Ranges of pre- and postapplication concentrations for N, P, and selected bacteria in tile discharge are presented in Table 2 . Maximum concentrations for data collected <100 d postapplication are considerably higher for all parameters measured (except for PO4–P in DI) relative to concentrations >100 d postapplication. And all maximum concentrations after application were above those preapplication. Except for TP, all treatment tile preapplication maximum observed concentrations, were higher than respective concentrations for the control plots. For all plots, including controls, both pre- and postapplication concentrations of E. coli, NO3–N, and TP exceeded water quality guidelines (Table 2) (Ontario Ministry of Environment, 1994; Canadian Council of Ministers of the Environment, 2007). However, these concentrations would likely have been reduced to some extent by plant nutrient uptake if the plots had been planted with a cover crop before land application.


View this table:
[in this window]
[in a new window]

  		Table 2. Minimum and maximum observed concentrations of selected N, P, and bacteria pooled by treatment for the 2006 study period,

 
Tile drainage mass loads before land application in the application treatment plots for N, P, (mg 15 min–1) and bacteria (CFU 15 min–1) ranged from: 0 to 2005 for NO3–N, 0 to 6.7 for NH4–N, 0.027 to 5.66 for PO4–P, 0.2 to 16.1 for TP, below detection limit (BDL) to 8 x 106 for E. coli, BDL to 4 x 107 for Enterococci, and BDL to 2.2 x 105 for C. perfringens (n = 22 samples). During <100 d postapplication, peak tile N, P, and bacterial mass loads, with the exception of Enterococci, were considerably higher than those associated with the preapplication conditions (representing both base flow and rainfall-induced preferential flow events) (Fig. 3 ).


Figure 3
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 3. Tile drain water quality parameters in (mg) or (CFU) per 15 min for samples collected over the early study period (<100 d postapplication).

 
Analysis by ANOVA showed a significant effect of treatment for NO3–N and C. perfringens on tile drain mass loads for data collected <100 d postapplication (P < 0.05) (Table 3 ). Post-hoc (Bonferroni) tests showed no significant differences (P > 0.05) between the SS and DI treatments, but both application treatments had significantly higher NO3–N and C. perfringens mass loads relative to those for the control treatments (P < 0.05). Peak mass loads for both application treatments observed during <100 d postapplication were approximately: 3070 for TKN, 739 for NH4–N, 22512 for NO3–N, 1699 for TP, 135 for PO4–P, 2.3 x 107 for E. coli and 2.4 x 106 for C. perfringens (Fig. 3).


View this table:
[in this window]
[in a new window]

  		Table 3. Summary of ANOVA (early study period) and t test (late study period) results for log-transformed N, P, and bacteria mass loads presented with means ± standard error for each treatment.

 
Late study period (>100 d postapplication) TKN, TP, E. coli, Enterococci, and C. perfringens tile mass loads from the DI plots were significantly higher (P < 0.05) relative to the SS plots (Table 3). Water pumping problems associated with electrical demand in the control plot tile access pits precluded collection of a similar number of samples relative to the SS and DI tiles during the >100 d postapplication study period; hence, statistical comparisons among control and application treatments for this late study period were not made. The greatest observed mass loads for SS and DI tiles for data collected >100 d postapplication were approximately (mg 15 min–1 for N and P, and CFU 15 min–1 for bacteria): 2322 for TKN, 261 for NH4–N, 6921 for NO3–N, 302 for TP, 150 for PO4–P, 1.3 x 106 for E. coli and 9.5 x 105 for C. perfringens (Fig. 4 ).


Figure 4
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 4. Tile drain water quality parameters in (mg) or (CFU) per 15 min for samples collected over the late study period (>100 d postapplication).

 
Shallow Groundwater Quality
Preapplication and before the first postapplication rain event, groundwater N and P concentrations at 1.2 m depth for all plots averaged (mg L–1): mean ± SE of 3.8 ± 2.8 for NO3–N, 0.10 ± 0.005 for NH4–N, 0.03 ± 0.017 for PO4–P, and 0.56 ± 0.21 for TP. For the same period, bacteria concentrations (CFU L–1) at 1.2 m depth ranged from: 104 ± 71 for E. coli, 18,800 ± 15,159 for Enterococci, and 1040 ± 268 for C. perfringens; and at 2.0 m depth below surface, groundwater concentrations for all plots averaged (mg L–1 and CFU L–1): mean ± SE of 2.33 ± 1.05 for NO3–N, 0.52 ± 0.27 for NH4–N, 0.02 ± 0.006 for PO4–P, 0.15 ± 0.08 for TP, 16 ± 8.5 for E. coli, 3400 ± 1778 for Enterococci, and 123 ± 70 for C. perfringens. For the application treatment plots, postapplication, there were general increases in NO3–N, NH4–N, and E. coli at 1.2 m depth, whereas PO4–P, TP, Enterococci, and C. perfringens concentrations were similar to preapplication values. At 2.0 m depth, postapplication, N, E. coli, Enterococci, and C. perfringens concentrations increased, while P was similar to preapplication values.

There were significant effects of treatment for NO3–N and C. perfringens (P < 0.05, ANOVA) postapplication at 1.2 m depth (Table 4 ). Bonferroni post-hoc tests showed both DI and SS treatments had significantly higher study period (DOY 199–340) NO3–N concentrations relative to control treatments (Fig. 5 ), while the SS treatment had significantly higher (P < 0.05) C. perfringens concentrations than the controls. There were no significant differences between DI and SS treatments (P > 0.05) for NO3–N and C. perfringens. Although no significant treatment effects were found for E. coli postapplication, an additional t test showed there were significantly higher application treatment E. coli concentrations post- vs. preapplication at 1.2 m depth (P < 0.05). Escherichia coli concentrations at 1.2 m declined considerably over the course of the study period, while C. perfringens showed concentration increases toward the end of the study period (Fig. 6 ).


View this table:
[in this window]
[in a new window]

  		Table 4. Summary of ANOVA results for groundwater log-transformed N, P, and bacteria concentrations presented with means ± standard error for each treatment.

 

Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 5. Groundwater NO3–N concentrations (mg L–1) at 1.2 and 2.0 m depth for treatment and control plots for the study period. Data were averaged by treatment for each sampling event. Error bars represent standard error.

 

Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 6. Groundwater Escherichia coli and Clostridium perfringens (CFU L–1) for study period at 1.2 m depth.

 
For postapplication 2.0 m depth piezometer data, there were significant treatment differences for NO3–N, NH4–N, TP, and Enterococci concentrations (P < 0.05, ANOVA) (Table 4). Bonferroni post-hoc tests showed that both DI and SS again had significantly higher NO3–N concentrations than controls, while controls had significantly higher TP concentrations than both DI and SS treatments, and greater NH4–N and Enterococci concentrations than the SS treatment (P < 0.05).

Escherichia coli Concentrations in Land Applied Dewatered Municipal Biosolids
Escherichia coli concentrations in DI treatment DMB were generally greater than SS plot DMB over the course of the study period, especially shortly after land application (Fig. 7 ). Despite considerable variability in concentrations for both treatments, DMB in DI treatment plots were found to have significantly higher E. coli than SS plots (P < 0.05, t test). Maximum observed concentrations were 2.5 x 106 and 2.9 x 105 CFU L–1 for DI and SS respectively, and minimum values were 1.3 x 105 and 1.0 x 104 CFU L–1 for DI and SS, respectively.


Figure 7
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 7. Land applied DMB Escherichia coli concentrations (CFU L–1) for the study period. Data were averaged by treatment for each sampling event. Error bars represent standard error.

 

   Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Methods and Materials
 Results
 Discussion
Conclusions
 REFERENCES
 
While there were no application treatment differences in N, P, and bacteria mass loads in tile effluent during the rainfall-induced flow events immediately following application (<100 d postapplication), when observed study peak mass loads and concentrations were generally greatest, there were significantly (P < 0.05) higher mass loads of TKN, TP, E. coli, Enterococci and C. perfringens from DI plots relative to SS plots during tile flow events that occurred in fall after a prolonged summer dry-spell (>100 d postapplication). Hence, there was a greater degree of longer term N, P, and bacteria persistence associated with the DI application relative to the SS application. Late study period (>100 d postapplication) application treatment differences in N mass loads may be due to greater N loss through volatilization from the SS treatment plots during the extended summer dry period. Significant N loss from surface-applied dewatered biosolids can occur rapidly through volatilization, and the higher air temperatures soon after application would likely have increased the rate of loss (Harmel et al., 1997; Robinson and Polglase, 2000; He et al., 2003). Surface application has also been shown to favor N mineralization, while subsurface application (or full incorporation) favors nitrification, due to the greater soil water content and lower temperatures below the soil surface (Sierra et al., 2001). Within 100 d postapplication, NH4–N tile mass loads were generally higher for SS plots, while NO3–N export was higher for DI plots, as shown in Fig. 2. Inferred increases in mineralization on SS plots, coupled with ammonia volatilization, could have considerably reduced organic N, which would explain the significantly lower TKN mass loads relative to DI later in the study season.

There was also generally higher late study period (>100 d postapplication) PO4–P and significantly higher TP tile mass loads from DI plots (P < 0.05) (Fig. 3). Anaerobically digested sewage sludges contain mainly inorganic forms of P, predominantly orthophosphate (O'Connor et al., 2004; Smith et al., 2006), which can be readily leached into soil solution during rain events after application (Vadas et al., 2007). Rain events soon after application would have leached soluble inorganic P from both treatments, but for flow events that occurred >100 d postapplication, there was still considerable soluble P in DI plot tile drain effluent, while concurrent soluble P output from the SS plots was relatively lower. Biosolid-amended soils also tend toward mineralization of organic P (Kashem et al., 2004) and, given the greater surface area to volume ratio of the DMB subjected to mechanical breakdown during application activities on SS plots, compared to the DMB injected in a more intact manner below the surface, mineralization of organically-bound P could have been more rapid in the SS plots, freeing up earlier PO4–P for crop growth and transport, and further depleting P stores in the SS dewatered biosolids by the late study period, lowering TP tile mass loads.

The DI treatment plots also showed greater persistence of E. coli and C. perfringens in tile water and in land applied DMB, compared to SS plots over the course of the study period. Bacterial survival in soil is highly dependent on soil water content and temperature, with generally greater survival occurring in soils with higher water content and lower temperatures (Reddy et al., 1981; Santamaria and Toranzos, 2003; Zaleski et al., 2005). The DMB applied to DI plots had been modestly compacted and was deposited in large continuous log-like segments to a greater depth in the topsoil than the DMB that was incorporated on SS plots. Aside from greater potential for the SS DMB to be exposed to "hostile" atmospheric elements than DI DMB (i.e., surface exposure as well as greater DMB surface area), there was generally greater soil water content and cooler temperatures at 20 cm depth in soil, relative to 10 cm depth (differences were most pronounced during <50 d postapplication) (Fig. 2). Although DI DMB E. coli was generally higher, relative to SS, throughout the season, both treatments had values approaching control plot soil E. coli concentrations near the end of the monitoring period (~DOY 290); an observation consistent with other studies showing decreases to near background levels for bacteria within approximately 3 mo after biosolid application (Lang et al., 2003; Estrada et al., 2004; Akhand et al., 2008).

Both SS and DI treatments caused bacterial contamination of groundwater to at least 1.2 m depth, with significantly higher E. coli post- vs. preapplication for both application treatments, and significantly higher C. perfringens in SS vs. control plots on a seasonal basis (P < 0.05). Both application treatments also caused N contamination to at least 2.0 m depth, with significantly higher DI and SS NO3–N concentrations compared to controls on a seasonal basis (P < 0.05). The greatest E. coli concentrations observed occurred soon after application and then rapidly decreased, while C. perfringens concentrations remained more consistent throughout the study period, and the greatest concentrations were observed on DOY 340. Clostridium perfringens, a spore former, has been shown to be more persistent in environmental water matrices than E. coli (Medema et al., 1997) and is likely a less conservative indicator of fecal-based contamination than E. coli, once the bacteria is significantly introduced into a system (i.e., land application of liquid municipal biosolids in fall 2005 causing groundwater loading of C. perfringens).


   Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Methods and Materials
 Results
 Discussion
 Conclusions
REFERENCES
 
To summarize, the peak observed tile mass loads of N, P, and bacteria for both land application treatments occurred soon after land application, but there were no significant differences in loads among application treatments at this time; and both land application treatments after the first postapplication rainfall event caused some contamination of shallow groundwater. Direct injection did cause, however, significantly greater N, P, and bacterial contamination of tile water than the SS method over the long term (>100 d postapplication). Nevertheless, peak mass loads during this "late" study period were relatively smaller compared to those observed during <100 d postapplication study period flow events, with reductions of peak observations of approximately 25% for TKN, 82% for TP, 94% for E. coli and 60% for C. perfringens. Additionally, tile mass loads for key soluble nutrients such as NO3–N and PO4–P, were not significantly different among both land application treatments. Further, while maximum observed TP and E. coli concentrations in tile water, for instance, exceeded the provincial water quality objectives of 0.03 mg L–1 and 1000 CFU L–1 (Ontario Ministry of Environment, 1994) respectively, this occurred with both DI and SS treatments, and most observed concentrations were below these limits >100 d postapplication. Within 40 d postapplication, E. coli concentrations in the direct injected DMB were higher than those of the DMB that was surface spread plus incorporated. But both methods had, for the most part, DMB E. coli concentration parity after this time.

Considering the overall results of this study in terms of water contamination and land-applied DMB E. coli persistence, and the important public health and aesthetic concerns of vector attraction and odor that injection of DMB can reduce (USEPA, 1995; Labud et al., 2003), the direct DMB injection method should be considered a DMB land application option, especially where DMB are applied near populated areas and/or where surface runoff potential is a significant concern.


   ACKNOWLEDGMENTS
 
This research was supported by Agriculture and Agri-Food Canada's Agricultural Policy Framework GAPs program, and Ontario Ministry of Agriculture and Food and Rural Affairs/Ontario Ministry of Environment's Nutrient Management Joint Research Program. Field assistance by Mr. D. Irving (Univ. of Guelph-Kemptville College) is greatly appreciated. A special thanks to Mr. M. Janiec for utilization of Terratec Environmental Ltd.'s dewatered biosolids direct injection system and in-kind logistical support. A. Scott provided excellent laboratory technical assistance.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Methods and Materials
 Results
 Discussion
 Conclusions
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Methods and Materials
 Results
 Discussion
 Conclusions
 REFERENCES
 





This Article
Free via Open Access
Right arrow OA Abstract
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow An erratum has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gottschall, N.
Right arrow Articles by Lapen, D. R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gottschall, N.
Right arrow Articles by Lapen, D. R.
Agricola
Right arrow Articles by Gottschall, N.
Right arrow Articles by Lapen, D. R.
Related Collections
Right arrow Water Quality
Right arrow Municipal Wastes
Right arrow Vadose Zone Processes and Chemical Transport
Right arrow Municipal Waste

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