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


     

Published online 2 February 2006
Published in J Environ Qual 35:505-515 (2006)
DOI: 10.2134/jeq2005.0205
© 2006 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
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 Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Miller, J. J.
Right arrow Articles by Selinger, L. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miller, J. J.
Right arrow Articles by Selinger, L. B.
Agricola
Right arrow Articles by Miller, J. J.
Right arrow Articles by Selinger, L. B.
Related Collections
Right arrow Water Quality
Right arrow Microorganisms
Right arrow Nutrients
Right arrow Runoff
Right arrow Animal Waste

TECHNICAL REPORTS

Surface Water Quality

Bedding and Within-Pen Location Effects on Feedlot Pen Runoff Quality Using a Rainfall Simulator

Jim J. Millera,*, Edith C. S. Olsonb, David S. Chanasykc, Bruce W. Beasleya, L. Jay Yankea, Francis J. Larneya, Tim A. McAllistera, Barry M. Olsond and L. Brent Selingere

a Agriculture and Agri-Food Canada, P.O. Box 3000, Lethbridge, AB, T1J 4B1 Canada
b Lethbridge Community College, 3000 College Drive South, Lethbridge, AB, T1K 1L6 Canada
c Department of Renewable Resources, General Services Building, University of Alberta, Edmonton, AB, T6G 2H1 Canada
d Alberta Agriculture Food and Rural Development, Lethbridge, AB, T1J 4V6 Canada
e Department of Biological Sciences, University of Lethbridge, Lethbridge, AB, T1K 3M4 Canada

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

Received for publication May 18, 2005.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soluble salts, nutrients, and pathogenic bacteria in feedlot-pen runoff have the potential to cause pollution of the environment. A 2-yr study (1998–1999) was conducted at a beef cattle (Bos taurus) feedlot in southern Alberta, Canada, to determine the effect of bedding material [barley (Hordeum vulgare L.) straw versus wood chips] and within-pen location on the chemical and bacterial properties of pen-floor runoff. Runoff was generated with a portable rainfall simulator and analyzed for chemical content (nitrogen [N], phosphorus [P], soluble salts, electrical conductivity [EC], sodium adsorption ratio [SAR], dissolved oxygen [DO], and pH) and populations of three groups of bacteria (Escherichia coli, total coliforms, total aerobic heterotrophs at 27°C) in 1998 and 1999. Bedding had a significant (P ≤ 0.05) effect on NH4–N concentration and load in 1999, SO4 load in 1998, SO4 concentration and load in 1999, and total coliforms in both years; where these three variables were higher in wood than straw pens. Location had a significant effect on EC and concentrations of total Kjeldahl nitrogen (TKN), Na, K, SO4, and Cl in 1998, and total coliforms in both years. These seven variables were higher at the bedding pack than pen floor location, indicating that bedding packs were major reservoirs of TKN, soluble salts, and total coliforms. Significantly higher dissolved reactive phosphorus (DRP), total P, and NH4–N concentrations and loads at the bedding pack location in wood pens in 1998, and a similar trend for TKN concentration in 1999, indicated that this bedding–location treatment was a greater source of nutrients to runoff than the other three bedding–location treatments. Bedding, location, and their interaction may therefore be a potential tool to manage nutrients, soluble salts, and bacteria in feedlot runoff.

Abbreviations: BP, bedding pack • DO, dissolved oxygen • DRP, dissolved reactive phosphorus • EC, electrical conductivity • PF, pen floor • SAR, sodium adsorption ratio • TKN, total Kjeldahl nitrogen


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
NUTRIENTS such as nitrogen (N) and phosphorus (P), soluble salts, and bacterial pathogens (e.g., Escherichia coli O157:H7) in runoff from beef feedlots have the potential to cause environmental pollution of water, soil, and air. Nitrogen and P may cause eutrophication of surface waters. High concentrations of soluble salts indicate soil and water salinization, and low electrical conductivity (EC) and high sodium adsorption ratio (SAR) may indicate soil sodicity. Bacterial pathogens in surface wa ters may impact the quality of drinking water for humans. The chemical properties of runoff derived from feedlots depend on factors such as soil type, climate, animal size, animal age and condition, animal species, housing, water consumption, climate, diet, feedlot surface, animal density, amount and type of bedding, and handling and storage of manure (Brady, 1974; Eghball and Power, 1994; Eck and Stewart, 1995). Two factors affecting the quality of surface runoff that have not received much attention are bedding and within-pen location effects. Bedding material may be a potential tool to manage the quality of feedlot runoff, and location may reveal the source of soluble salts, nutrients, and bacteria within feedlot pens.

In contrast to U.S. feedlots, most feedlots in Canada use bedding because of colder climatic conditions. Most feedlots in Alberta are bedded with barley straw, but increasing numbers of feedlots are using wood-chip bedding (Miller et al., 2003). Wood chips tend to have a higher carbon content, lower N and P values, greater acidity, and lower quantities of soluble salts than straw (Allison and Anderson, 1951). Miller et al. (2003) reported higher K and pH values in feedlot pen manure with straw than wood-chip bedding. They also found higher total carbon, carbon to nitrogen ratios, NH4–N, and available P in manure with wood-chip bedding than with straw bedding. Significant bedding effects on chemicals in feedlot manure suggest that runoff water quality from feedlot pens bedded with these two materials may also be different.

Wood, especially bark, contains phenols, organic acids, tars, tannins, ethyl alcohol, resins, and terpentine (Goldstein, 1982), which may be natural antibacterial inhibitors (Allison and Anderson, 1951). Kudva et al. (1998) found that small quantities of wood-chip bedding in cattle manure may have contributed to shorter survival times for Escherichia coli O157:H7. Nimenya et al. (2000) found that spruce sawdust inhibited urease-producing bacteria from converting urea to NH4 in dairy-cattle urine, and attributed the inhibitory effect to the tar contents of the wood. However, Miller et al. (2003) reported that populations of E. coli, total coliforms, and total aerobic heterotrophs were similar in beef feedlot pen manure with straw or wood-chip bedding, which suggested no antibacterial inhibition by the wood chips. We are unaware of any studies that have determined bacterial populations in surface runoff from beef feedlot pens bedded with straw versus wood chips.

Unpaved feedlot-pen surfaces in the Canadian prairies have a distinctive morphology with three major zones. The bedding mound zone is the highest elevation and consists mainly of bedding with lesser amounts of manure. The second zone is the bedding pack and consists mainly of equal parts bedding and manure, with lesser amounts of soil mixed in from the unpaved feedlot surface. Bedding packs or bedding mounds promote animal comfort and cleanliness by keeping cattle drier through better drainage, and by generation of heat by partial composting during the winter season (Winchell et al., 2000). The third zone furthest away from the bedding-pack mound is the pen floor, and consists mainly of soil with minor amounts of manure pats and bedding. In a typical commercial beef feedlot, the pen floor and bedding pack occupy the greatest area of a pen. Since the physical and chemical characteristics of the bedding pack and pen-floor zones are different, we might expect differences in the chemical and bacterial quality of runoff from these zones that might reveal greater sources of soluble salts, nutrients, and bacteria within the feedlot pen. Cattle behavior and distribution within feedlot pens may also affect the chemical and bacterial properties of feedlot runoff from different locations. We are unaware of any studies that have examined the effect of within-pen location on runoff quality. Miller et al. (2004) studied the quality of runoff from an entire beef feedlot that was bedded with straw and wood chips, but they did not measure runoff at different locations within each pen.

The objective of the study was to determine the influence of bedding type and within-pen location on the chemical and bacterial quality of runoff at a beef-cattle research feedlot in southern Alberta.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Design and Sampling Protocol
The 1.2-ha research feedlot at the Agriculture and Agri-Food Canada Research Centre at Lethbridge, Alberta, was used for the study. Each of the 32 pens in the feedlot measured 14 by 19.5 m. The feedlot capacity at the time of the study was approximately 500 head of beef cattle, with 15 cattle per pen for a stocking density of 18 m2 head–1, which is a stocking density comparable to that of commercial feedlots. Kennedy et al. (1999) reported a stocking density of 17 m2 head–1 for a 25 000-head feedlot in east-central Alberta. The cattle were steers weighing an average of approximately 300 kg each entering the pens and weighing about 580 kg leaving the pens. The cattle entered the feedlot in the fall and were removed in the spring of the following year (Table 1). Pen cleaning in 1998 and 1999 was conducted after the cattle were removed and the rainfall simulations were completed.


View this table:
[in this window]
[in a new window]
  		Table 1. Experimental treatment details, dates of cattle entering and leaving feedlot pens, and start and end dates of rainfall simulations.

 
Fresh bedding was added to the pens (Fig. 1) whenever the feedlot manager considered it warranted based on the amount of tag (mud, bedding, or manure) on the animals and the conditions of the bedding area. Generally, the same amount of straw or wood-chip bedding was generally added to each pen on the same date (Fig. 1). The material was added using a tractor with a front-end loader, which deposited the bedding on the bedding area of the pen, from where it was dispersed by the action of the cattle. This practice is commonly used in commercial feedlots. The straw bedding was unchopped barley straw. The wood-chip bedding was a mixture of sawdust and bark peelings derived mainly from 80% lodgepole pine (Pinus contorta var. latifolia Engelm.) and 20% white spruce [Picea glauca (Moench) Voss]. The chemical properties of these bedding materials have been reported previously (Miller et al., 2003). The bedding pack area occupied approximately 45% of the total pen area, the pen floor about 45%, and the bedding mound about 5% (visual observation). The bedding to manure ratio in 1998 and 1999 was measured as the mass of bedding added to each pen divided by the mass of manure (minus bedding) removed from each pen.


Figure 1
View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1. Dates of bedding addition (empty symbols) to feedlot pens in relation to rainfall simulations (half-filled symbols) in 1998 and 1999. Pens for bedding addition and rainfall simulations were the same in 1998 (a, b). Pens for bedding addition in 1999 were different than pens for rainfall simulations (c). Mean weight of bedding for three pens in 1999 was used. Weights of bedding applied to all pens in 1999 were similar.

 
The rainfall simulations were conducted in active (stocked) pens in the spring of 1998 and 1999 (Table 1), generally after all the bedding had been added (Fig. 1), and before pen cleaning. Rainfall simulations were conducted using the portable Guelph Rainfall Simulator II (Tossell et al., 1987). The rainfall simulator applied a rainfall intensity of 54 mm h–1 on a 1.0- by 1.0-m area of each site to generate runoff. This was achieved by using a 9.5-mm nozzle at a height of 1.0 m above the pen surface. For the Lethbridge region, a rainfall intensity of 54 mm h–1 for 20 min represents a return period storm of 5 yr. Details of the rainfall simulations are reported by Olson (2004). The dates when bedding was added to each feedlot pen, the mass of bedding added, and the dates of rainfall simulations are shown in Fig. 1. Bedding additions were measured for the pens where rainfall simulations were conducted in 1998. However, bedding additions were not available for the pens where rainfall simulations were conducted in 1999. Therefore, mean bedding amounts added to adjacent pens were used.

For all rainfall simulations, 19 consecutive runoff samples of approximately 700 mL each were collected after runoff commenced. The time to generate runoff ranged from 20 to 60 min, and the time to collect 19 samples ranged from 14 to 41 min. Every odd-numbered sample of the 19 samples was used for quantifying the chemical variables in 1998 and 1999, and statistical analyses were conducted on the mean value of these subsamples. The exceptions were total Kjeldahl nitrogen (TKN) and total P, where the chemical analysis was conducted on a composite of all 19 subsamples taken during runoff. Every even-numbered sample of the 19 samples was used for quantifying bacterial populations in 1998, and statistical analyses were conducted on the mean value of these subsamples. In 1999, bacterial analysis was conducted on a composite of all 19 subsamples taken during runoff. The change from subsamples in 1998 to composite samples in 1999 was initiated because time series analyses of the subsamples in 1998 showed bacteria numbers were relatively constant with time throughout the runoff event. Water samples were also taken from the rainfall simulator tank during each simulation and analyzed for chemical properties. Water samples were stored in plastic bottles, preserved with acid if required (American Public Health Association, 1995), and then stored at –20°C until analyzed (Olson, 2004).

Feedlot floor volumetric water content, manure pack gravimetric water content, manure pack depth, slope of rainfall simulator area (1-m distance), cumulative time of runoff, and total volume of runoff were measured as described by Olson (2004).

Chemical Analyses
Dissolved oxygen (DO), pH, EC, and temperature of water samples were measured at the feedlot immediately after collection using a portable water-quality meter and associated probes (MultiLine P4; Wissenschaftlich-Technische, Werkstätten, Germany). Ammonium N was analyzed using the automated phenate method (Technicon Industrial Systems, 1973) in 1998 and the automated salicylate method (Rhine et al., 1998) in 1999. Nitrate N was determined using the automated cadmium reduction method (Technicon Industrial Systems, 1972a) in 1998 and the automated hydrazine reduction method (Kempers and Luft, 1988) in 1999. Changes were made in the methods mainly because of safety concerns with the existing chemicals (i.e., phenol and cadmium). Quality control checks revealed good agreement between the former and latter methods. Dissolved reactive phosphorus (DRP) was determined using the automated ascorbic acid method (Technicon Industrial Systems, 1974b). Calcium and Mg were determined using atomic absorption spectroscopy while K and Na were determined by flame emission spectroscopy. Chloride was determined using the automated mercuric thiocyanate method (Technicon Industrial Systems, 1974a). Sulfate was determined using the automated barium chloride method (Technicon Industrial Systems, 1972b).

Total Kjeldahl nitrogen and total P were simultaneously determined on water samples digested with H2SO4 using USEPA Method no. 365.4 (USEPA, 1974). The method was modified by eliminating the HgO catalyst and increasing the duration and temperature of digestion. The digested extracts were then analyzed by the automated molybdenum blue colorimetric method for total P (Technicon Industrial Systems, 1986a) and the automated salicylate method for TKN (Technicon Industrial Systems, 1986b).

Bacterial Analyses
Runoff water samples were stored in capped, sterile urine sample cups in an ice-filled cooler and were transported to the laboratory within 1 to 2 h and plated within 4 h of collection. Samples were analyzed for total aerobic heterotrophs (TAH), E. coli, and total coliform bacteria by serially diluting the water samples to the appropriate numbers in sodium phosphate buffer (pH 6.5, 0.05 M). The dilutions were spread plated (100 µm) in duplicate onto Fluorocult LMX (Merck, Darmstadt, Germany) agar plates for enumeration of total coliform bacteria and E. coli. The TAH were similarly plated onto tryptic soy agar (TSA). The Fluorocult LMX plates were incubated aerobically at 37°C for 48 h. Escherichia coli was enumerated after 24 h as those colonies with the ability to hydrolyze 5-bromo-4-chloro-3-indolyl-D-galactopyranoside (X-GAL) and 4-methylumbelliferyl-D-glucuronide (MUG). Total coliform bacteria were enumerated after 48 h as those colonies hydrolyzing X-GAL. The TAH were incubated at 27°C and enumerated after 48 h. Selected colonies of presumptive total coliform bacteria and E. coli were isolated for confirmation of identity through membrane fatty acid composition (Paisley, 1996), cellular morphology, and biochemical characteristics (Smibert and Kreig, 1994). All 27 presumptive isolates were confirmed as E. coli. Of 115 presumptive coliform colonies, 102 isolates were confirmed as coliform but not E. coli. The other 13 isolates were confirmed as E. coli.

Data and Statistical Analyses
Mean values of chemical and bacterial properties were used in the statistical analysis. Data from 1998, when simulations were conducted at two locations on the bedding pack and pen floor of each pen, were averaged to give one mean for the bedding pack and one mean for the pen floor. Average runoff mass loads (kg ha–1) were calculated based on the mean concentration, total runoff volume, and area of runoff (1 m2).

The chemical and bacteria data were analyzed using SAS (SAS Institute, 1989) and a mixed model analyses (Littell et al., 1998). The main treatment effects of bedding (straw, wood chips) and location (BP = bedding pack, PF = pen floor) were assessed for the chemical and bacterial parameters. The data were analyzed separately for each year because different pens and number of sites within each pen were used in each year. Bacterial numbers were transformed to log10 values before statistical analyses. Treatment effects were considered significant at the 10% probability level. Step-wise regression was used to determine relationships between chemical and bacterial variables versus selected physical properties, with a 5% significance level required for entry of an independent variable into the model.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Effect of Environmental Variables on Runoff Water Quality
Since this experiment was conducted in a research feedlot where the experimental conditions were not as controlled as replicated research plots, and because the rainfall simulations were conducted over a one- to two-month period, a statistical analysis was conducted on certain environmental variables to determine if an analyses of co-variance would be required to analyze our runoff chemistry and bacteria data. We found no significant effects of bedding, location, or their interaction on feedlot floor and manure pack water content, feedlot floor and manure pack bulk density, manure depth, slope, surface roughness, and temperature of the runoff water. This indicated that these environmental variables were relatively constant throughout our experiment, and that an analysis of covariance was unwarranted.

Precipitation
Monthly precipitation in relation to the long-term mean is shown for the nine months when cattle were in the feedlot in 1997–1998 and for the seven months in 1998–1999 (Fig. 2). Monthly precipitation was 136 and 88% of the long-term mean for the first and second years of this study, respectively. Monthly precipitation during the months of the rainfall simulations were 178 and 97% of the long-term mean for the 1997–1998 and 1998–1999 periods, respectively.


Figure 2
View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2. Monthly precipitation at a feedlot in southern Alberta during 1997–1998 (a) and 1998–1999 (b) in relation to long-term average monthly precipitation (1971–2000).

 
Dissolved Reactive Phosphorus and Total Phosphorus
Bedding x location interaction had a significant effect on DRP and total P concentrations in 1998 (Table 2), where they were highest at the BP location in wood pens (Fig. 3a, 3b). There were no treatment effects on total P concentration in 1999. Similar to concentrations, DRP and total P loads in 1998 were significantly higher at the BP location in wood pens than the other three treatments (Table 3, Fig. 4a, 4b). Miller et al. (2003) reported similar total P values between straw versus wood-chip bedding, and between pen manure in straw and wood pens. They found that available P was six times higher for straw than wood bedding, but that it was 1.3 times higher for pen manure in wood than straw pens. Our P results indicated that the BP location in wood pens had the highest P pollution potential in wet years such as 1998.


View this table:
[in this window]
[in a new window]
  		Table 2. Bedding and within-pen location effects on selected chemical properties and concentrations in runoff from pens of a beef-cattle feedlot in southern Alberta (1998–1999).

 

Figure 3
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 3. Bedding x location effects on mean values (±1 SE) of dissolved reactive phosphorus, DRP (a), total P (b), and NH4–N (c) in 1998, and total Kjeldahl nitrogen (d) in runoff in 1999 from a feedlot in southern Alberta. Mean values with the same lowercase letter are not significantly (P > 0.05) different. Labels on the x axis are ST (straw), WD (wood), BP (bedding pack), and PF (pen floor).

 

View this table:
[in this window]
[in a new window]
  		Table 3. Bedding and within-pen location effects on mean mass loads of selected chemical properties in runoff from pens of a beef-cattle feedlot in southern Alberta (1998–1999).

 

Figure 4
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 4. Bedding x location effects on mean (±1 SE) mass loads of dissolved reactive phosphorus, DRP (a), total P (b), and NH4–N (c) in runoff from a feedlot in southern Alberta in 1998. Mean values with the same lowercase letter are not significantly (P > 0.05) different. Labels on the x axis are ST (straw), WD (wood chips), BP (bedding pack), and PF (pen floor).

 
Total Kjeldahl Nitrogen and Ammonium Nitrogen
Ammonium N concentration was significantly influenced by bedding x location in 1998 (Table 2), where it was highest at the BP location in wood pens (Fig. 3c). In 1999, NH4–N concentration was 2.3 times higher in wood than straw pens (Table 2). As reported by Miller et al. (2003), even though straw bedding had 44 times more NH4–N than wood bedding, pen manure at this same feedlot had higher NH4–N in wood than straw pens. They attributed this to the lower pH of manure with wood chips than with straw. We also found that the pH of runoff was lower in wood than straw pens in both years (Table 2). Average loads of TKN were 1.9 times higher at the BP than PF location in 1998, and loads were 2.0 times higher for wood than straw pens in 1999 (Table 3).

Location had a significant effect on TKN concentration in 1998 (Table 2), where it was two times higher at the BP than the PF location. There was a significant bedding x location effect on TKN concentration in 1999 (Table 2), where values were highest at the BP location in wood pens (Fig. 3d). Although total N was 25 times higher in straw than wood bedding, total N in pen manure at this same feedlot was similar (Miller et al., 2003). Overall, our N findings indicated that the BP location in wood pens had the highest NH4–N runoff pollution potential in wet years such as 1998. In drier years such as 1999, the BP location in wood pens had the highest TKN pollution potential, and wood pens had a higher NH4–N pollution potential than straw pens.

Dissolved Oxygen and pH
Location had a significant effect on DO concentration in runoff in both years (Table 2), where values were 1.3 times lower at the BP than PF location. Lower DO at the BP location is consistent with the high amount of manure and organic matter at this location. The decomposition of organic matter and microorganisms contributes to low DO levels in surface waters (Kadlec and Knight, 1996).

There was a significant bedding x location effect on pH (Table 2). Mean pH values in 1998 were higher for the BP location in straw pens (8.9) than the other three treatments (8.2–8.5). In 1999, mean pH at the BP location in straw pens (9.3) was significantly higher than the BP (8.5) and PF (8.8) locations in straw pens, but was similar to the PF location in wood pen (9.2). The pH of the bedding used at this feedlot was lower for wood (4.5) than straw (6.5), and the pH of the pen manure was also lower in wood than straw pens (Miller et al., 2003).

Salinity and Sodicity
Bedding had no effect on EC in either year (Table 2), and was consistent with no difference in EC of pen manure in straw and wood pens (Miller et al., 2003). Location had a significant effect on EC of runoff in 1998 (Table 2), where mean EC values were 1.9 times higher at the BP than PF location. The mean depth of the manure pack was greater at the BP location (1998 = 13.9 cm, 1999 = 17.6 cm) than the PF location (1998 = 5.3 cm, 1999 = 9.5 cm) (Olson, 2004); and the manure pack in this feedlot had a high EC between 7.8 and 8.9 dS m–1 (Miller et al., 2003). The SAR in 1998 was significantly affected by the interaction of bedding and location (Table 2). The SAR values were higher for the BP location in wood pens (9.7) than for the other three treatments (1.9–6.8), indicating that runoff from this bedding–location was the greatest source of sodicity.

Bedding had a significant effect on SO4 concentration in 1999 (Table 2), where it was 1.7 times higher in wood than straw pens. The mean depth of the manure pack was similar in straw (1998 = 9.1 cm, 1999 = 13.6 cm) and wood pens (1998 = 10.1 cm, 1999 = 12.9 cm) (Olson, 2004). Although SO4 concentration was higher in straw than wood bedding, mean values were similar in pen manure of straw and wood pens (Miller et al., 2003). Sodium, K, SO4, and Cl concentrations were significantly influenced by location in 1998 (Table 2), where values were 3.1 to 4.6 times higher for the BP than PF location. Calcium concentration was significantly influenced by bedding x location interaction in 1998 (Table 2), where mean values were significantly lower for BP location in wood pens (10.0 mg L–1) than for the other three treatments (14.5–19.0 mg L–1). There was also a significant bedding x location interaction on Mg concentration in both years (Table 2). In 1998, Mg was higher at the BP location in straw pens (22.0 mg L–1) than for the other three treatments (7.0–14.2 mg L–1); in 1999 it was highest at the PF location in straw pens (data not shown).

Similar results were found for average loads of soluble salts (Table 3). Mean SO4 loads were significantly higher in straw than wood pens in 1998, but the reverse trend was found in 1999. Loads of Na, K, SO4, and Cl in 1998 were significantly higher at the BP than PF location. Bedding x location interaction had a significant effect on loads of Ca in 1998, and Mg loads in both years (Table 3). Average Ca loads in 1998 were lower at the BP location in wood pens (1.5 kg ha–1) than for the other three treatments (2.2–2.9 kg ha–1). Magnesium loads in 1998 were highest at BP location in straw pens, and in 1999 they were highest at the PF location in straw pens (data not shown).

Bacteria
The main effects of bedding had a significant influence on total coliforms in both years (Table 4), where numbers were higher in wood than straw pens. The main effect of location had a significant effect on total coliforms in both years (Table 4), where values were higher at the BP than PF locations. There was a significant bedding x location effect on E. coli numbers in runoff in 1998 (Table 4). Numbers of E. coli in 1998 were higher at the BP (8.12 log CFU 100 mL–1; CFU = colony forming units) and PF (8.14 log CFU 100 mL–1) locations in wood pens than the BP (7.82 log CFU 100 mL–1) and PF (7.48 log CFU 100 mL–1) locations in straw pens. There were no treatment effects on E. coli in 1999, or total aerobic heterotrophs in either year (Table 4). This was consistent with the findings of Miller et al. (2003), who reported similar numbers of these three bacterial groups in straw and wood pen manure.


View this table:
[in this window]
[in a new window]
  		Table 4. Bedding and within-pen location effects on bacterial populations in runoff from pens of a beef-cattle feedlot in southern Alberta (1998–1999).

 
Our finding of significantly higher E. coli and total coliforms in runoff from wood chip than straw pens in 1998, and for total coliforms in 1999, was unexpected (Table 4). We expected lower bacterial numbers in runoff under wood than straw because of the potential antimicrobial properties of wood (Goldstein, 1982) and the greater concentration of total phenols in wood chip than straw bedding (Miller et al., 2003). Miller et al. (2003) reported no significant difference in populations of E. coli, total coliforms, or total aerobic heterotrophs at 27 and 39°C between pen manure with straw or wood chip bedding at this same feedlot. They suggested that no antibacterial effect by wood chips may have been due to the low absolute concentrations of total phenols produced by the bedding material, the high ratio of manure to bedding, and the high quantities of organic material. Larney et al. (2003) also reported that E. coli and total coliforms were not affected by bedding material during windrow composting of manure from this same feedlot.

No significant differences in bacterial numbers in pen manure from straw and wood-chip pens (Miller et al., 2003), but significantly higher bacterial numbers in runoff from wood chip than straw pens found in this study, suggest that the release rate of bacteria from pen manure to runoff may be different under these two bedding materials. The characteristics of the microbe (species, age, concentration), adsorbent (type, ionic form, particle size, concentration), and environment (pH, salt concentration, organic compounds, agitation, type of contact, temperature) are all factors that can affect sorption and desorption of microorganisms (Daniels, 1980). These factors could have contributed to greater release of bacteria into runoff under wood chips. Further research is required to determine exactly what factors control sorption and desorption of bacteria from bedding, feces, and pen manure (mixture of bedding, feces, and soil).

Treatment Effects on Runoff Quality
Treatment effects in this study were dependent on year, with more chemical and bacterial properties influenced by bedding and location in 1998 than 1999. This may be related to the wetter conditions, and more bedding used in 1998. Olson (2004) reported that the gravimetric water content of the manure pack within the pens was 65% higher in 1998 (0.38 kg kg–1) than in 1999 (0.23 kg kg–1), and this was consistent with higher precipitation in 1998 than 1999 (Fig. 2). In 1998, straw pens were bedded 20 times with 4.3 Mg of straw, and wood pens were bedded 14 times with 13.8 Mg of wood (McAllister et al., 1998). In 1999, straw pens were bedded 16 times with 3.8 Mg of straw, and wood pens were bedded seven times with 7.0 Mg of wood (A. Olson, personal communication, 2005). Therefore, 13% more straw bedding and 96% more wood bedding by weight was used in 1998 than 1999. The bedding to manure ratio in 1998 was 1:5 for straw and 1:4 for wood, and in 1999 it was 1:4 for straw and wood.

Bedding had a significant effect on NH4–N concentration and load in 1999, SO4 load in 1998, SO4 concentration and load in 1999, and total coliforms in both years. These three variables were higher in wood than straw pens. This was consistent with higher NH4–N and SO4 in wood than straw bedding (Miller et al., 2003). The exception was SO4 load in 1998, where values were higher in straw than wood pens.

Seven of the 14 chemical properties in 1998 and eight of the chemical properties in 1999 (Table 2) were not affected by bedding, and this may be related to the low ratio of bedding to manure. Similar findings were reported for chemical properties of pen manure at this same feedlot (Miller et al., 2003).

Location had a significant effect on EC and concentrations of TKN, Na, K, SO4, and Cl in 1998, and on total coliforms in both years. These seven variables were higher at the BP than PF location, indicating that bedding packs are major reservoirs of TKN, soluble salts, and total coliforms. The greater influence of location than bedding on the chemical properties was consistent with that for physical properties. The BP locations absorb more water and have a greater depth of manure, steeper slopes, and rougher surfaces than PF locations, whereas the PF locations have greater bulk densities than BP locations (Olson, 2004). Since visual observation indicated that manure dominated the BP location and soil dominated the PF location, and the manure pack has higher concentrations of N, P, and soluble salts than the soil, it seems likely that runoff from the BP location would have higher levels of nutrients and soluble salts. In addition, the A horizon is typically removed before a feedlot is constructed, so feedlot surface soil is low in nutrients because organic matter is removed.

More significant location effects on the variables in 1998 than 1999 may be related to the higher precipitation in 1998. Precipitation during the two months of rainfall simulations (June and July) was higher than the long-term mean in 1998 (Fig. 2), and this may have resulted in cattle spending more time on the BP location to keep dry. This would result in more manure pats deposited at the BP than PF location in 1998, and runoff may have transported more nutrients and soluble salts from these manure pats. In contrast, precipitation during the two months of rainfall simulations (May and June) in 1999 was slightly below the long-term mean (Fig. 2), and this may have resulted in a more uniform distribution of cattle between the BP and PF locations. In an average day, cattle spend about 2 h at the feedbunk, 1 h at the water trough, and the rest of the time near the bedding pack (T.A. McAllister, unpublished data, 2004). Bedding packs encourage animals to return to a specific site, and more frequent defecation at the bedding pack site can result in these areas becoming highly loaded with nutrients (Ziegler, 2005). Our results indicate that BP locations in feedlot pens are major sources of TKN, soluble salts, and total coliforms.

Significantly higher DRP, total P, and NH4–N concentrations and loads at the BP than PF location in wood pens in 1998, and a similar trend for TKN concentration in 1999 (Fig. 3 and 4), indicated that this bedding–location treatment was the greatest source of nutrients to runoff. It seems the combination of wood bedding and bedding pack contributes most P and N to runoff in feedlot pens, but we have no explanation for this. Bedding x location also affected E. coli numbers in the wet year (1998), where values were higher for both locations in wood pens than in straw pens. Significant bedding x location effects were found for pH in both years, SAR in 1998, Ca concentration and load in 1998, Mg concentration and load in both years, and total coliforms in both years. Different trends among the four bedding–location treatments were found for these five variables.

Relationship of Environmental Variables to Runoff Water Quality
Step-wise regression among chemical and bacterial properties versus physical properties revealed significant relationships for certain dependent variables (Table 5). Manure pack and feedlot floor water content had a significant effect on Ca concentration in runoff (Table 5). A positive relationship between Cl concentration in runoff and surface roughness and cumulative time of runoff (Table 5) indicated that a longer period of runoff and greater surface area of the manure pack and feedlot floor to the rainfall may have contributed to greater loss of these soluble non-adsorbed anions in runoff. Manure pack water content and cumulative time of runoff had a positive influence on DRP in runoff. The pH of runoff was significantly influenced by manure pack water content, feedlot floor water content, and the feedlot floor slope. There was a positive relationship between the EC of the runoff and the feedlot floor slope and manure water content, indicating that greater feedlot-floor slopes and higher manure water content may contribute to higher salinity in feedlot runoff. For bacteria, there was a negative relationship between total volume of runoff and log E. coli concentration, suggesting a dilution effect. The total volume of runoff and feedlot floor water content significantly influenced the total coliform populations in runoff, with runoff volume exerting a negative influence and feedlot floor water content exerting a positive influence. There was also a positive relationship between total aerobic heterotrophs and manure pack water content. This was consistent with the finding that microbial die-off in soils increased with decrease in water content (Reddy et al., 1981). The depth of the manure pack had no significant effect on any of the chemical or bacterial variables. Since the zone of rainfall–soil interaction that contributes to surface runoff is very shallow (2.0 cm) (Ahuja and Lehman, 1983), it is not surprising that there was a weak correlation between the depth of manure pack and chemical–bacterial properties. The proportion of manure, soil, and bedding material that occupies the surface area of the feedlot floor likely controls the quality of runoff rather than the depth of manure pack. Further research is required to more fully investigate the effect of these physical properties on the chemical and bacterial quality of feedlot runoff.


View this table:
[in this window]
[in a new window]
  		Table 5. Relationships (step-wise regressions) between various independent physical variables and selected dependent variables in runoff from pens of a beef-cattle feedlot in southern Alberta (1998–1999).

 

   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Bedding had a significant effect on NH4–N concentration and load in 1999, SO4 concentration and load in 1999, and total coliforms in both years; where these three variables were higher in wood than straw pens. Bedding may therefore be a potential tool to manage NH4–N, SO4, and total coliforms in runoff from feedlots. For example, converting from wood to straw bedding may reduce total coliforms in runoff, and this might lower the pollution potential of feedlot runoff. Since many feedlots irrigate adjacent cropland with catch basin effluent, bedding might also be a potential tool to manage total coliforms in effluent irrigation.

Location had a significant effect on EC and concentrations of TKN, Na, K, SO4, and Cl in 1998, and on total coliforms in both years. These seven variables were higher at the bedding pack than pen floor location, indicating that bedding packs are major reservoirs of TKN, soluble salts, and total coliforms. Although eliminating bedding mounds and associated bedding packs from feedlots may lower the pollution potential of runoff, animal health and comfort during cold winters would likely be affected. A possible compromise is frequent cleaning of manure from feedlot pens, which would minimize the depth of manure pack and time that the manure pack is present. This practice could minimize or eliminate the major reservoir of nutrients, soluble salts, and bacteria.

Significantly higher DRP, total P, and NH4–N concentrations and loads at the BP location in wood pens in 1998, and a similar trend for TKN concentration in 1999, indicated that this bedding–location treatment was the greatest source of nutrients to runoff.


   ACKNOWLEDGMENTS
 
Field assistance was provided by Ken Coles, Sean Robison, and Michael Verhage. Laboratory assistance was provided by Jim Braglin-Marsh, Clarence Gilbertson, Wayne McKean, Bonnie Tovell, Troy Bech, and Rhett Rasmussen. Ian Walker and Wendi Smart coordinated the bedding of the pens. Statistical advice was provided by Toby Entz.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
LRC Contribution no. (387)05009.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
F. J. Larney, A. F. Olson, J. J. Miller, P. R. DeMaere, F. Zvomuya, and T. A. McAllister
Physical and chemical changes during composting of wood chip-bedded and straw-bedded beef cattle feedlot manure.
J. Environ. Qual., March 1, 2008; 37(2): 725 - 735.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
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 Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Miller, J. J.
Right arrow Articles by Selinger, L. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miller, J. J.
Right arrow Articles by Selinger, L. B.
Agricola
Right arrow Articles by Miller, J. J.
Right arrow Articles by Selinger, L. B.
Related Collections
Right arrow Water Quality
Right arrow Microorganisms
Right arrow Nutrients
Right arrow Runoff
Right arrow Animal 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