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


     

Published online 20 February 2008
Published in J Environ Qual 37:725-735 (2008)
DOI: 10.2134/jeq2007.0351
© 2008 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 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 Google Scholar
Google Scholar
Right arrow Articles by Larney, F. J.
Right arrow Articles by McAllister, T. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Larney, F. J.
Right arrow Articles by McAllister, T. A.
Agricola
Right arrow Articles by Larney, F. J.
Right arrow Articles by McAllister, T. A.
Related Collections
Right arrow Nutrient Cycling
Right arrow Nutrient Management
Right arrow Industrial Waste
Right arrow Irrigation

TECHNICAL REPORTS

Waste Management

Physical and Chemical Changes during Composting of Wood Chip–Bedded and Straw-Bedded Beef Cattle Feedlot Manure

Francis J. Larney*, Andrew F. Olson, Jim J. Miller, Paul R. DeMaere, Francis Zvomuya and Tim A. McAllister

Agriculture and Agri-Food Canada, Research Centre, 5403 1st Ave. S., Lethbridge, AB, Canada T1J 4B1. Lethbridge Research Centre contribution no. 38707023

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

Received for publication July 4, 2007.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
In the 1990s, restrictions on incineration encouraged the forest industry in western Canada to develop new uses for their wood residuals by product. One such use was as a replacement for cereal straw bedding in southern Alberta's beef cattle (Bos taurus) feedlot industry. However, use of carbon (C)-rich bedding, such as wood chips, had implications for subsequent composting of the feedlot manure, a practice that was being increasingly adopted. In a 3-yr study, we compared composting of wood chip–bedded manure (WBM) and barley (Hordeum vulgare L.) straw-bedded manure (SBM). There were no significant differences in temperature regimes of SBM and WBM, indicating similar rates of successful composting. Of 17 physical and chemical parameters, five showed significant (P < 0.10) differences due to bedding at the outset of composting (Day 0), and 11 showed significant differences at final sampling (Day 124). During composting (10 sampling times), seven parameters showed significant bedding effects, 16 showed significant time effects, and four showed a Bedding x Time interaction. Significantly lower (P < 0.10) losses of nitrogen (N) occurred with WBM (19%) compared with SBM (34%), which has positive implications for air quality and use as a soil amendment. Other advantages of WBM compost included significantly higher total C (333 vs. 210 kg Mg–1 for SBM) and inorganic N (1.3 vs. 1.0 kg Mg–1 for SBM) and significantly lower total phosphorus (4.5 vs. 5.3 kg Mg–1 for SBM). Our results showed that wood chip bedding should not be a problem for subsequent composting of the manure after pen cleaning. In combination with other benefits, our findings should encourage the adoption of wood chips over straw as a bedding choice for southern Alberta feedlots.

Abbreviations: BD, bulk density • DM, dry matter • EC, electrical conductivity • GHG, greenhouse gas • IN, inorganic nitrogen • KEP, Kelowna-extractable phosphorus • OM, organic matter • PEP, percent extractable phosphorus • PIN, percent inorganic nitrogen • SBM, straw-bedded manure • TC, total carbon • TN, total nitrogen • TP, total phosphorus • VL, volume loss • WBM, wood chip-bedded manure • WC, water content


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
BEDDING is an important input at beef cattle feedlots on the Canadian prairies. In winter it mitigates cold stress on animals by providing insulation on snow or frozen ground, and in early spring it reduces stress induced by wet, muddy conditions during snowmelt (Mader, 2003). Although wood residuals have been commonly used as animal bedding, particularly for dairy cows and poultry, an inexpensive and plentiful supply of cereal straw bedding has precluded their widespread use in southern Alberta's cattle feedlot industry.

Historically, wood residuals emanating from the forestry sector in western Canada were incinerated in beehive or silo burners. Wood combustion emits nitric oxide (NO), nitrous oxide (N2O) (Winter et al., 1999), dioxins (Lavric et al., 2004), and particulate matter, which can negatively affect air quality. Therefore, policies were implemented in the mid-1990s to phase out wood burners operated by the forest and lumber industry to mitigate potentially damaging emissions (British Columbia Ministry of Environment, Lands and Parks, 1995). In the wake of these regulatory changes, the industry actively sought alternate uses for wood residuals. Marketing initiatives by Alberta-based forest product companies, coupled with drought years and low straw yields in the late-1990s, increased interest in the use of wood residuals (hereafter referred to as wood chips) as an alternative feedlot bedding to straw. Around the same time, interest in composting feedlot manure was high (Larney et al., 2006a; 2006b), and questions arose about the suitability of wood chip–bedded manure (WBM) for composting, compared with the more traditional straw-bedded manure (SBM).

Although barley straw and wood chips have similar levels of cellulose (360–420 g kg–1) and hemicellulose (210–270 g kg–1), a higher content of C in the form of lignin (~280 g kg–1) makes wood chips less biodegradable than straw (~110 g kg–1 lignin). Lodgepole pine (Pinus contorta var. latifolia Engelm.) from Montana has a C/N ratio of 661:1 (Allison, 1965), whereas barley straw has a C/N ratio of approximately 100:1. The much wider C/N ratio and the presence of more recalcitrant lignin (Tuomela et al., 2000) lowers the bioavailability of nutrients in wood chips, which has implications for composting manure from feedlot pens bedded with wood chips rather than straw. The effect of lignin on the bioavailability of outer cell wall components is thought to be largely a physical restriction with lignin molecules reducing the surface area available for enzymatic penetration and activity (Haug, 1993).

Wood chips are considered more of a bulking agent than straw when used in compost feedstocks (Rynk, 1992). The presence of wood chips may increase convective airflow through windrows, possibly increasing the supply of oxygen and enhancing aerobic decomposition. McCartney and Chen (2001) found that wood chips showed superior bulking properties and led to more free air space compared with straw and recommended their use in very high (3.7 m) windrows of municipal solid waste compost. Suzuki et al. (2004) reported that co-composting of wood chips with poultry manure was more effective than with other co-materials (urea, nitrogenous lime, food waste, coal ash, or volcanic ash). Mixtures of compost and wood chips have been used as biofilter media for control of livestock odors mainly because of their superior air flow (Nicolai and Janni, 2001). With respect to end-use, composts containing wood by-products, especially tree bark, have been shown to release toxic compounds (natural fungicides) that can suppress plant pathogens (Hoitink and Fahy, 1986; Litterick et al., 2004).

The present study is part of a larger one comparing the use of wood chip and straw bedding materials in feedlots. In terms of animal welfare, McAllister et al. (1998) demonstrated that cattle were cleaner on wood chip bedding because the abrading action of wood chips reduced the amount of tag (hair soiled and matted by bedding/mud/manure) on cattle hides by 30%. Moreover, they reported a lower bedding frequency requirement with wood chips, which translated into a 30% reduction in labor associated with bedding feedlot animals. The effects of wood chip vs. straw bedding on the chemical and bacterial properties of pen-floor manure (Miller et al., 2003) and runoff (Miller et al., 2006) as well as hydrological response (Olson et al., 2006) have been compared, as have their effects on the fate of coliform bacteria (Larney et al., 2003) and Giardia and Cryptosporidium (Van Herk et al., 2004) during composting. Hao et al. (2004) evaluated greenhouse gas (GHG) emissions during composting of WBM and SBM, and Zvomuya et al. (2005) reported on co-composting of WBM and SBM with phosphogypsum. With regard to end-use, the effects of fresh vs. composted beef cattle manure containing wood chips or straw on barley yield, nutrient uptake, and soil nutrient status (Miller et al., 2004); soil salinity and sodicity (Miller et al., 2005); and soil physical properties (Miller et al., 2000) have been reported.

This component of the study compares changes in physical and chemical properties during composting and reports mass balance estimates for wood chip-bedded and straw-bedded feedlot manure in southern Alberta.


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results and Discussion
 Conclusions
 REFERENCES
 
The study involved composting trials set up in three successive years at a research feedlot at the Agriculture and Agri-Food Canada Research Centre, Lethbridge, AB (49°38' N, 112°48' W). Cattle arrived at the feedlot in November/December (1997, 1998, and 1999) and departed the following June/July (1998, 1999, and 2000). At least three feedlot pens were bedded with wood chips and three with unchopped barley straw each year. The wood bedding (Sunpine Forest Products, Sundre, AB) included wood chips, bark, post peelings, and sawdust derived from a 4:1 lodgepole pine/white spruce [Picea glauca (Moench) Voss] mix. Each feedlot pen (273 m2 area) was stocked with 15 steers for a stocking density of 18 m2 head–1, which is typical of local commercial feedlots.

During each of the three 6- to 7-mo periods that the pens were stocked, bedding was added using a tractor and front-end-loader. Bedding frequency was at the discretion of the feedlot manager and based on wetness of the bedding pack and amount of tag on cattle hides in accordance with practices at commercial feedlots. In the first 2 yr, all bedding was weighed and sampled for water content (WC) before addition to the pens, and at pen cleaning all material was removed, weighed, and sampled for WC. This enabled an estimation of bedding added, expressed as a percent (dry wt) of material removed.

Windrow Management
Within 1 mo of cattle vacating the feedlot in 1998, 1999, and 2000, manure was removed from the pens with a loader and a truck equipped with a load cell and deposited into windrows on a concrete pad in an open-sided roofed composting facility (i.e., exposed to ambient air temperatures but not precipitation). Two replicates of each bedding type (wood chips or straw) were established to give four individual compost windrows per year. At formation, windrows were 10.6 to 11.4 m in length, 2.5 m wide at the base, and 2 m high. The mean (n = 6, 2 windrows x 3 yr) wet mass of fresh manure at the outset of composting (Day 0) was 16.9 (±1.0 SE) Mg windrow–1 for WBM and 15.8 (±0.1 SE) Mg windrow–1 for SBM.

Windrows were turned with a tractor-pulled EarthSaver windrow turner (Fuel Harvesters Equipment Inc., Midland, TX) 16 times in 1998 and eight times in 1999 and 2000 (Table 1 ). Turning frequency was reduced in 1999 and 2000 in an effort to more closely follow schedules at local commercial feedlots. Mean July–September (months common to all three compost trials) air temperatures, monitored at an automated weather station ~0.5 km from the compost facility, were 18.8°C in 1998, 16.0°C in 1999, and 17.2°C in 2000, which compared with a 30-yr (1971–2000) mean of 16.0°C.


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

  		Table 1. Compost sampling schedules in 1998, 1999, and 2000 and mean sampling day calculation for the 3 yr.

 
Windrow Temperatures
Thermocouples and a data logger (Sciemetric, Nepean, ON) were installed as soon as compost windrows were formed. They were removed just before each turning and reinstalled immediately after turning. Thermocouples were attached to three-pronged stainless steel forks to give a 3 x 3 (nine-point) sampling grid with each fork: three thermocouples per prong (prongs spaced 46 cm apart) at three vertical locations (bottom, middle, and top). The bottom, middle, and top locations were approximately 0.3, 0.6, and 0.9 m, respectively, above ground level. Three forks were inserted vertically into each bedding treatment (with the center prong of each fork aligned with the apex of the windrow) to give a total of 27 temperature monitoring points per treatment. In 1998, the three forks were placed in one replicate of each bedding treatment: near the center of the windrow and ~2 m from each end. In 1999 and 2000, the three windrow positions were ~3 m from each end of one replicate of each bedding treatment and at the center of the other replicate of each treatment. Temperatures were logged every 20 min and averaged to give mean daily values.

Compost Sampling
The sampling protocol involved making a cut with a skid-steer loader bucket perpendicular to the long axis of each windrow. This exposed two vertical faces at each cut, and samples were taken from top, middle, and bottom locations of each vertical face. In 1998, three cuts were made per windrow, exposing six vertical faces and therefore 18 samples. In 1999, one cut was made per windrow, giving two vertical faces and six samples. In 2000, two cuts were made per windrow, resulting in 12 samples. Sampling was performed 10 times during each composting trial. Table 1 outlines the sampling days and dates for each of the 3 yr and the arrival at an estimate of a mean sampling date for the 3 yr.

Physical Analyses
A subsample (~1.5 kg wet wt) was oven dried at 60°C for 5 d for WC determination (Hoskins et al., 2003). Dry bulk density (BD) of the manure/compost was estimated as the mass of dry matter (DM) (based on oven-dried subsample) divided by the volume (0.07 m3) of an aluminum bin. The bin was filled with material composited from the exposed windrow faces. Volume of the compost windrows was calculated from their length and circumference assuming a semi-circular shape (Larney et al., 2000). Circumference was estimated as the mean (n = 3) of the distances obtained by straddling the windrow with a measuring tape (i.e., from ground level to ground level at opposite sides of the windrow). Volume loss (VL) at each sampling time was expressed as a percent of the initial volume.

Chemical Analyses
Ammonium-N and NO3–N were determined on extracts from wet subsamples (10 g of manure/compost material shaken with 200 mL of 2 M KCl for 1 h) using an AutoAnalyzer II (Technicon Industrial Systems, Tarrytown, NY). The remaining analyses were performed on oven-dry material, which was coarse-ground to <2 mm and thoroughly mixed, except total C (TC) and total N (TN), which were measured on fine-ground (<0.15 mm) subsamples by combustion–gas chromatography using a CN analyzer (NA-1500; CE Instruments, Milan, Italy). Inorganic N (IN) was calculated as NH4–N + NO3–N, and percent inorganic N (PIN) was calculated as IN/TN x 100. Total P (TP) was determined by colorimetric analysis after wet digestion with H2SO4 and H2O2. Kelowna-extractable phosphorus (KEP) (Ashworth and Mrazek, 1995) was measured on 5 g of material extracted with 50 mL of solution (0.015 M NH4F, 1.0 M NH4OAc, and 0.5 M HOAc). Percent extractable P (PEP) was defined as KEP/TP x 100. pH and electrical conductivity (EC) were read on a meter calibrated with 0.1 M KCl using extracts obtained by shaking 30 g of material in 120 mL of deionized water for 1 h. Samples of wood chips and straw added as pen bedding in the first 2 yr were analyzed for TC, TN, TP, pH, and EC as outlined previously.

Mass balances (%, [initial – final]/[initial] x 100) for DM, water, total (DM + water), C, and N were estimated by assuming that (i) mass losses were wholly attributed to organic matter (OM) and (ii) the mass of ash (mineral matter) was conserved during the composting process (Bernal et al., 1998b; Petersen et al., 1998). Ash concentration (kg kg–1) was defined as 1 – OM (kg kg–1), where OM was estimated from TC concentration (Larney et al., 2005):

Formula 1[1]

Additionally, C and N mass balance estimates were derived for each of the nine sampling intervals (i.e., Days 0–7, 7–14, 14–21, etc.) to evaluate C and N loss rates over time.

Statistical Analyses
All statistical analyses were performed using SAS (SAS Institute Inc., 2006). Compost temperature data were analyzed using PROC MIXED with bedding as the fixed effect and year as a random effect (n = 3). All compost properties measured at 10 sampling times between Day 0 and Day 124 were analyzed using the MIXED procedure for repeated measures (Littell et al., 1996) with sampling time as the repeated measures factor. Bedding, sampling time, and bedding x sampling time interaction were fixed effects, and windrow (n = 2) and year (n = 3) were random effects in the mixed models. Models for mass balance estimates included bedding as a fixed effect and windrow and year as random effects. Appropriate covariance structures were selected according to Akaike's Information Criterion (Littell et al., 1996). The UNIVARIATE procedure indicated that BD, VL, TC, TN, IN, PIN, NH4–N, NO3–N, NH4–N/NO3–N ratio, and N/P ratio data did not conform to normal distributions, according to the Shapiro-Wilk statistic (P ≤ 0.05). These data were therefore loge(x)-transformed, with the exception of NH4–N/NO3–N ratio and N/P ratio data, which were loge(1 + x)-transformed; VL data, which were loge(100 + x)-transformed; and PIN data, which were arcsine of square root–transformed before analysis.

For compost parameters with significant (P < 0.10) time and/or Bedding x Time effects, polynomial orthogonal contrasts were used to determine if increased time of composting exerted a linear or quadratic effect. When the contrasts were significant, regression models were developed and evaluated on the corresponding treatment means. Mean comparisons were made using the DIFF option of the LSMEANS statement in SAS and using the Tukey-Kramer adjustment for multiple contrasts for all pairwise comparisons.


   Results and Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
Conclusions
 REFERENCES
 
Bedding Amounts and Properties
Over the 3 yr, bedding frequency averaged 11 times for wood chips and 20 times for straw during each 6- to 7-mo period that cattle were in the pens. This reflected the ability of wood chip bedding to allow water to infiltrate the bedding pack (McAllister et al., 1998), which resulted in drier surface conditions. However, because of the greater density of wood chips, a higher mass (mean 7.7 Mg pen–1, dry wt) was added compared with straw (mean 4.1 Mg pen–1, dry wt). Of the total mass of material removed from the straw-bedded pens at pen cleaning, straw added as bedding accounted for 172 g kg–1 (dry wt) in 1998 and 190 g kg–1 in 1999. Similarly, wood chips accounted for 220 g kg–1 in 1998 and 217 g kg–1 in 1999. The mean bedding concentrations were fairly similar at 180 g kg–1 for straw and 220 g kg–1 for wood chips, which is roughly a 4:1 manure/bedding ratio for both materials. The consistency in bedding rates between 1998 and 1999 indicates a similar rate in 2000 because the same bedding regime was used.

Total C levels were higher in wood chips (mean 506 kg Mg–1) than straw (mean 447 kg Mg–1) (Table 2 ). However, total N levels were lower in wood chips (mean 1.9 kg Mg–1) than straw (mean 5.7 kg Mg–1), and this gave rise to a much higher C/N ratio for wood chips (mean 330:1) than straw (mean 91:1). Total P values were higher in straw (mean 0.64 kg Mg–1) than wood chips (mean 0.28 kg Mg–1). The wood chips had a lower pH value (mean 5.4) compared with straw (mean 6.8). Straw EC levels were almost 10 times higher (mean 8.7 dS m–1) than wood chips (mean 0.9 dS m–1). The consistency of bedding properties between 1998 and 1999 suggested that materials in 2000 (analyses not conducted) were similar because the sources of straw and wood chips were unchanged.


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

  		Table 2. Properties of bedding materials in 1998 and 1999.

 
Windrow Temperatures
Temperature plays an important role in composting and is a good process indicator (Epstein, 1997; Rynk, 1992). The optimum range is 40 to 65°C, with microbial activity declining rapidly above 65°C and approaching inhibition above 73°C (Keener et al., 2000). Tuomela et al. (2000) reported that the optimum temperature for thermophilic fungi (40–50°C) corresponded with the optimum temperature for lignin degradation in compost.

Bedding type had no significant effect on overall compost temperature parameters. Over Days 0 to 97, a period common to all 3 yr, the mean absolute maximum temperature was 73.8°C for WBM vs. 73.3°C for SBM (P = 0.26). It required a mean of 36 d to attain maximum temperature for WBM, compared with 34 d for SBM (P = 0.49). Windrow temperatures were >55°C, the thermal kill threshold for pathogens in compost guidelines (CCME, 1996; USEPA, 1992), for 36 d with WBM vs. 39 d with SBM (P = 0.51). The guidelines state that temperatures >55°C should be maintained for at least 15 d, a criterion easily exceeded by both bedding types. The lack of differences in windrow temperatures corroborate the findings of Larney et al. (2003), who found no effect of bedding on the fate of coliform bacteria during composting. Additionally, Miller et al. (2003) found that the bedding effect was nonsignificant for four groups of bacteria (Escherichia coli, total coliforms, and total aerobic heterotrophs at 27 and 39°C) in pen floor samples from the same feedlot.

Dividing the composting duration into 10-d periods showed that all bedding comparisons for mean windrow temperature were nonsignificant (Fig. 1 ). During the period coming closest to significance (Days 21–30; P = 0.12), SBM was 1.4°C warmer than WBM (59.6 vs. 58.2°C). Numerically, the WBM treatment was only slightly warmer (by 0.3–0.5°C) than SBM during Days 51 to 60 and 61 to 70. In contrast to our results, Michel et al. (2004) found that straw-amended dairy manure attained lower compost temperatures than sawdust-amended manure and failed to meet the 55°C required for pathogen elimination. However, their manure/bedding ratio was closer to 1:1, compared with 4:1 in our study.


Figure 1
View larger version (41K):
[in this window]
[in a new window]

  		Fig. 1. Effect of bedding type on windrow temperatures for 10-d periods during composting (least square means, 1998–2000). Error bars represent SEM. P values for each comparison are presented above the bars.

 
Initial Manure Properties
Five (TC, TN, C/N ratio, PEP, and pH) of the 17 properties measured on fresh manure on Day 0 showed significant differences (P < 0.10) due to bedding (Table 3 ). Total C was significantly higher for WBM than SBM (394 vs. 302 kg Mg–1), and TN was significantly lower for WBM than SBM (14.3 vs. 18.3 kg Mg–1). This led to a significantly higher C/N ratio for WBM than SBM (28.2 vs. 16.7). The bedding effects on TC and C/N ratio agree with those for pen-floor manure taken from the same feedlot (Miller et al., 2003). The PEP was significantly higher for WBM (66.8% of TP in extractable form) than SBM (50.7%).


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

  		Table 3. Least square means of bedding treatments for compost properties at initial sampling on Day 0.

 
pH was significantly higher with SBM (8.4) than WBM (7.4) at the start of composting, whereas there was no significant difference (P = 0.25) in EC (13.4 dS m–1 for SBM vs. 10.3 dS m–1 for WBM) (Table 3). At a bedding concentration of 180 to 220 g kg–1 dry wt, wood chip bedding (pH 5.4) lowered the pH of fresh manure at pen clean-out compared with straw bedding. However, straw bedding (8.7 dS m–1), which was more saline than wood chips (0.9 dS m–1), did not affect the EC of fresh manure. Miller et al. (2003, 2006) also reported that the bedding effects on EC were nonsignificant for pen-floor and bedding pack manure from the same feedlot.

Properties during Composting
A total of 17 properties (3 physical and 14 chemical) were measured at each of the 10 sampling dates during the study (Tables 4 and 5 ). These parameters are grouped according to treatment effects.


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

  		Table 4. Effect of bedding and time of composting on physical properties and carbon and nitrogen properties (dry wt except water content) of compost.

 

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

  		Table 5. Effect of bedding and day of composting on phosphorus properties, pH, and electrical conductivity of compost.

 
Bedding Effect
Seven parameters (one physical [VL] and six chemical [TC, IN, PIN, C/N ratio, PEP, and pH]) showed significant bedding effects (Tables 4 and 5). However, two of these parameters (TC and pH) also showed significant Bedding x Time interactions. Volume loss was significantly higher (mean 17%) for SBM than WBM (mean 8.8%) (Table 4). The wood chips were a better bulking agent and hence reduced VL compared with straw bedding.

Averaged across all sampling times, IN, PIN, C/N ratio, and PEP were higher for WBM than SBM. Mean IN was 2.5 kg Mg–1 for the WBM treatment, compared with 2.0 kg Mg–1 for SBM (Table 4). Therefore, land application of WBM would result in higher rates of IN application, which is considered readily available for plant uptake. For PIN, WBM averaged 16.7%, compared with 11.7% for SBM over the composting period (Table 4).

The significantly higher average C/N ratio for WBM (25.5) compared with SBM (15.2) (Table 4) is reflective of the higher C/N ratio of wood chip bedding (330:1) compared with straw bedding (91:1). Even though the wood chip bedding accounted for only 220 g kg–1 of the total manure DM at pen cleaning, this proportion was high enough to influence the C/N ratio of subsequent compost. The PEP averaged 81.7% for WBM, compared with 70.1% for SBM (Table 5), which indicates higher mineralization of manure organic P in the presence of wood chips.

Time Effect
Composting time had a significant effect on all measured properties except EC (Tables 4 and 5). However, four of these parameters (TC, TN, N/P ratio, and pH) showed significant Bedding x Time interactions. Of the 12 remaining parameters, two (IN and PIN) showed significant linear responses, whereas the rest showed significant quadratic responses with time (Tables 4 and 5). Regression equations (based on the 10 sampling time means presented in Tables 4 or 5) had R2 values ranging from 0.76 for NO3–N to 0.99 for WC (Table 6 ).


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

  		Table 6. Relationships between time (x) and physical and chemical parameters (y) during composting (based on 10 sampling time means){dagger}.

 
Water content fell rapidly in early composting and then leveled off (i.e., the decline was 0.0057 kg kg–1 d–1 at Day 10 compared with 0.0004 kg kg–1 d–1 at Day 100). In contrast, BD increased rapidly during early composting (0.64% d–1 at Day 10) before leveling off (0.15% d–1 at Day 100). The trend in VL closely followed BD, being initially high (0.38% d–1 at Day 10) and falling during later composting (0.14% d–1 at Day 100).

For IN, the linear relationship with time predicted a decline of 0.9% d–1 over a 124-d composting period. For PIN, the decline was also linear at 0.2% d–1. The negative quadratic response between NH4–N concentration and time showed a slower decline during early composting, which became more rapid during late composting. For example, the rate of decline at Day 10 was 0.6% d–1 compared with 2.2% d–1 at Day 100. In contrast, the NO3–N response was one of an initial decline in concentration, reaching a minimum value on Day 34, followed by a slow increase at first (e.g., 0.6% d–1 at Day 40) with a more rapid increase toward the end of composting (e.g., 6.4% d–1 at Day 100). The contrasting behavior of higher NO3–N and lower NH4–N concentrations with compost time is reflective of nitrification (conversion of NH4–N to NO2–N and then NO3–N), which is a step in the degradation of OM during composting (Fauci et al., 1999; Peigné and Girardin, 2004).

The quadratic response of C/N ratio with time exhibited a rapid decline in the early stages of composting (0.084 U d–1 at Day 10). By Day 100, the daily rate of decline had more than halved to 0.039 U d–1. Larney and Olson (2006) also reported a significant quadratic relationship between C/N ratio and time (C/N ratio = 18.8 – 0.144 x time [d] + 0.0005 x [time]2; R2 = 0.94; P < 0.001) for composting of SBM. The current equation (y = 23.6 – 0.089x + 0.0002x2) (Table 6) had a higher a coefficient (intercept value), denoting the presence of C-rich wood chips. The b coefficient was lower, showing that the decline in C/N ratio with time in the present study (mean of SBM and WBM) was slower than that of SBM alone. For the NH4–N/NO3–N ratio, the daily rate of decline was slow initially (e.g., 0.5% d–1 on Day 10) but increased with time of composting (e.g., 6.5% d–1 at Day 100).

The response of TP with time showed an increase, more rapid at first, and then gradually reaching a peak at Day 93. Total P increases during composting because P is conserved in a system not subject to runoff, whereas the total mass of OM decreases (Felton et al., 2004). For KEP, the trend was similar to TP except that the initial increase with time was more rapid, as denoted by the higher b coefficient (0.036 vs. 0.024) (Table 6). The KEP peaked at Day 89. The PEP followed a trend similar to the other P parameters, with a rapid initial increase and a peak at Day 82. Composting effects on P availability are variable, with studies showing no change (Eghball and Power, 1999; McCoy et al., 1986) or lower availability in compost than fresh manure (Felton et al., 2004; Sharpley and Moyer, 2000). Adler and Sikora (2003) reported that composting did not change levels of water-extractable P or Mehlich-1 P in poultry manure, and, because TP concentration increased with composting, the PEP values declined with compost age for both extractants. With water extraction, they reported that PEP declined from 14 to 10% after 8 wk of poultry manure composting. A similar decline from 25 to 16% PEP was found with Mehlich-1 extraction. In contrast, PEP increased from 59 to 81% PEP (modified Kelowna-extractable) for feedlot manure in our study (Table 5).

Bedding x Time Interaction Effect
Four parameters (TC, TN, N/P ratio, and pH) showed significant Bedding x Time interaction effects, which meant their behavior with composting time differed depending on bedding type. For TC, SBM showed a significant quadratic relationship with time, whereas a slight but nonsignificant decrease in TC with time was evident for WBM (Fig. 2a ). For SBM, the most rapid decrease in TC concentration occurred in the first 7 d and continued until Day 91, by which time the compost had largely stabilized at 203 kg TC Mg–1.


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 2. Effect of bedding type and composting time on (a) total C, (b) total N, (c) N/P ratio, and (d) pH.

 
For TN, the interactions arose from the positive linear relationship with time for WBM and the lack of a relationship for SBM (Fig. 2b). For WBM, TN increased at a rate of 0.16% d–1. For the N/P ratio, there was no significant relationship with time for WBM (Fig. 2c). However, SBM showed a quadratic relationship whereby the greatest decrease in N/P ratio occurred in the early stages of composting before reaching a minimum value of 3.86 on Day 73.

For pH, the Bedding x Time interaction showed a cubic response for SBM and a nonsignificant relationship for WBM (Fig. 2d). For SBM, there was a decline from pH 8.1 to 7.2 in the first 36 d of composting, after which pH increased slowly to 8.0 on Day 105. This trend has been described in composting studies by Inbar et al. (1993) and Tuomela et al. (2000). Soluble and easily degradable C sources, such as monosaccharides, starch, and lipids, are used by microorganisms in the early stages of composting, and pH decreases as organic acids are released from these compounds during degradation. In the next stage, microorganisms start to degrade proteins, resulting in liberation of ammonium and an increase in pH. The downward tail of the cubic relationship from Day 105 to Day 124 (pH decrease from 8.0 to 7.7) may be related to the increased nitrification toward the end of composting, which is an acidifying process (Simandi et al., 2005).

Final Compost Properties
Day 124 represented the final sampling of the study and as such the end of thermophilic and mesophilic composting. Beyond this stage, compost would be considered ready for land application after a short curing period. Properties on Day 124 would be expected to give an indication of the behavior of the two composts when added to soil. Eleven of the 17 properties measured on Day 124 showed significant differences (P < 0.10) due to bedding (Table 7 ). This is much higher than on Day 0, where only five properties showed significant differences due to bedding (Table 3).


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

  		Table 7. Least square means of bedding treatments for compost properties at final sampling on Day 124.

 
The WBM was significantly wetter (319 vs. 261 kg kg–1) than SBM on Day 124. On a comparative wet weight basis, more water and less DM would be hauled with WBM than with SBM, leading to increased transportation costs per unit of DM. The significantly higher VL with SBM (41.8%) compared with 24.1% for WBM would also be advantageous from a haulage viewpoint because less volume of SBM would need to be transported. Because there was no difference in BD between the two beddings on Day 124 (both 0.29 Mg m3), fewer truckloads would be required to transport SBM to the field. The lower VL with WBM is likely due to the preservation of the bulking effect of wood chips during composting. Wood chips were visible in the compost on Day 124, whereas straw particles were subject to more microbial degradation. Our lower VL with WBM agrees with Michel et al. (2004), who reported lower VL values (33–79%) for sawdust-amended dairy manure compost than for straw-amended material (65–93%).

The WBM also had a higher TC concentration (333 kg Mg–1) than SBM (210 kg Mg–1) on Day 124, which means that the trend in the initial fresh manure was maintained. Therefore, on an equivalent dry weight basis, addition of WBM would increase soil C storage potential over SBM. Fortuna et al. (2003) reported that total soil C was increased by >14% relative to fertilized soils after 4 yr of compost (made from a 1:1 oak [Quercus rubra] leaves/dairy manure mix) addition. In the same study, compost application increased the size of the resistant C pool by 30% and the slow C pool by 10% after 6 yr.

The three N-related properties (IN, PIN, and NO3–N) were significantly higher for WBM than SBM. A higher PIN (8.4 vs. 5.3%) means that more of the total N was in the inorganic form for WBM, so that on an equivalent dry weight basis, 30% more IN (1.3 vs. 1.0 kg Mg–1) would be soil-applied compared with SBM. Because IN is readily available for plant uptake, WBM may have an advantage for a pre-seeding application to cropland. The higher NO3–N content of WBM (754 mg kg–1) compared with SBM (179 mg kg–1) was the main contributor to its higher IN and PIN values. The significant difference in C/N ratio at the outset (Table 3) was maintained at final sampling, where the C/N ratio was significantly higher for WBM (21.2) than SBM (12.4). The WBM also had a significantly lower NH4–N/NO3–N ratio (0.8) than SBM (4.5), indicating a more mature or stable compost (Bernal et al., 1998a). In companion field studies at Lethbridge, Miller et al. (2004) reported that N uptake and soil available N were significantly affected by interactions of bedding material with amendment type (fresh vs. compost), rate, or year, showing that bedding influenced certain components of the N cycle. However, barley DM yield and protein content were not affected by bedding material.

The WBM had a 15% significantly lower TP concentration than SBM (4.5 vs. 5.3 kg Mg–1) (Table 7). Therefore, use of WBM rather than SBM may be advantageous on soils of medium to high P loadings. However, even though WBM had less TP than SBM, there was no difference in KEP (4.0 vs. 3.9 kg Mg–1; P = 0.62) (Table 7) because a greater proportion of the TP in WBM was in extractable form. The significant bedding effect on PEP at Day 0 (WBM, 66.8%; SBM, 50.7%) (Table 3) was maintained during composting as WBM compost had PEP of 85.7% vs. 76.0% for SBM. The reason for increased extractability of P in the presence of wood chips is unclear. Miller et al. (2003) reported that a higher calcium content in straw than wood chips (1.0 vs. 0.4 g kg–1) may have caused increased phosphate precipitation, therefore reducing extractability in manure mixed with straw bedding.

As at initial sampling on Day 0, WBM had a significantly lower pH (7.1) than SBM (7.8) on Day 124 due to the influence of more acidic wood chips. Crop production on acid soils can be improved greatly by adjusting the pH to near neutrality. Although soil acidity is commonly corrected by liming, there is evidence that animal manure amendments can increase the pH of acid soils (Eghball, 1999; Whalen et al., 2002). The higher pH of SBM would be more effective in correcting the pH to near neutrality when applied to acid soils.

The trend of lower salinity with WBM (11.4 dS m–1 vs. 14.5 dS m–1 for SBM; P = 0.13) on Day 124 makes WBM a more attractive amendment for greenhouse potting media. Miller et al. (2005) found that soil Ca, Mg, K, Cl, EC, and K adsorption ratios were significantly higher with application of SBM than WBM materials at certain rates and years. They suggested that bedding material could be used as a management tool to control levels of certain salinity variables in soil. For example, high EC or K levels in soils where SBM was applied could potentially be lowered by converting to WBM.

Mass Balance Estimates
Two of the six mass balance estimates showed significant differences (P < 0.10) with bedding type (water and N), whereas four did not (DM, total mass, C, and P) (Fig. 3 ). Water losses were significantly higher (P = 0.02) with SBM (85.4% of initial) than WBM compost (77.5%), which corroborates the WC data of the materials on Day 124 (Table 7). It also agrees with Zvomuya et al. (2005), who reported that SBM compost lost more water (90.4%) than WBM compost (74.8%) during a composting study in southern Alberta.


Figure 3
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 3. Effect of bedding type on mass balance estimates of water, total (water + dry matter [DM]), DM, C, N, and P during composting.

 
Nitrogen losses were significantly higher (P = 0.10) from SBM (33.5%) than WBM (19.0%) (Fig. 3). This agrees with Zvomuya et al. (2005), who found that N losses from SBM were 19.9 vs. 4.5% for WBM in a co-composting experiment with phosphogypsum. Although overall N losses in that study were lower due to the presence of acidic phosphogypsum, the trend of higher losses with SBM was similar. Our results also agree with Michel et al. (2004), who reported lower N losses with sawdust-amended dairy manure composts (8–26%) than with straw-amended composts (15–43%). The higher N loss for SBM could be due to an initially higher TN content and lower C/N ratio (Table 3) or greater ammonia (NH3) losses due to a higher pH (Fig. 2d). Low C/N ratios (Al-Kanani et al., 1992; Eghball et al., 1997; Tiquia et al., 2002) or pH values >7 (Tiquia and Tam, 2000; Witter and Lopez-Real, 1988) are associated with higher N losses during composting. At Lethbridge, Hao et al. (2004) found that GHG emissions as N2O were unaffected by bedding type (SBM vs. WBM), which suggests that the difference in N losses between bedding types may arise from variation in NH3 volatilization.

Dry matter losses were similar for both bedding types (28.1% for WBM vs. 28.6% for SBM) (Fig. 3). The SBM had higher but nonsignificantly different total mass (64.7% vs. 57.9%; P = 0.15) and total C losses (49.3 vs. 38.6%; P = 0.16) than WBM. Hao et al. (2004) reported no difference between SBM and WBM in GHGs emitted as CO2 or CH4 at Lethbridge, which is in agreement with our C loss findings.

Phosphorus losses due to bedding type were not significantly different (Fig. 3), averaging 5.9%, which is lower than P losses from other manure composting studies: 12% for feedlot manure in Nebraska (Eghball et al., 1997), 12 to 21% for sawdust-amended and 1 to 38% for straw-amended dairy manure in Ohio (Michel et al., 2004), and 30 to 32% for solid pig manure in Iowa (Tiquia et al., 2002). Phosphorus losses during composting are generally attributed to runoff or leaching. Because our windrows were in a roofed facility, these processes did not occur.

Carbon and N mass losses followed first-order kinetic equations (Fig. 4 ) of the following form:

Formula 2[2]
where A is the maximum degradation value, k is the decomposition rate constant (d–1), and t is the composting time (d). First-order kinetic relationships have also been reported for OM decomposition during composting of sweet sorghum bagasse (Bernal et al., 1996) and co-composting of olive mill wastewater with agricultural by-products (Paredes et al., 2000, 2002).


Figure 4
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 4. Effect of bedding type on (a) carbon and (b) nitrogen losses during composting.

 
For C losses (Fig. 4a), the maximum degradation value was higher for SBM (51.4 ± 0.9% SE) than WBM (39.3 ± 1.6%). However, although WBM lost less C overall, the decomposition rate constants (k) were similar at 0.0504 ± 0.003 d–1 SE for SBM and 0.0514 ± 0.007 d–1 for WBM, showing that both bedding materials behaved similarly for loss of labile C. The half-life (t1/2) for labile C was calculated as:

Formula 3[3]
and showed similar values of 14 d for SBM and 13 d for WBM.

Nitrogen losses (Fig. 4b) showed similar responses with time, with the maximum degradation value being higher for SBM (37.6 ± 0.8% SE) than WBM (25.8 ± 1.0%). Although WBM lost less N overall, the decomposition rate constant (k) was higher (0.154 ± 0.04 d–1 SE) than for SBM (0.119 ± 0.01 d–1), indicating that readily decomposable N was lost faster from WBM. However, the t1/2 values of 5 d for WBM and 6 d for SBM were similar.


   Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
REFERENCES
 
Temperature regime is indicative of the success or failure of composting, and as such there was no deleterious effect from use of wood chip bedding, with a complete lack of significant differences in temperature regimes between SBM and WBM.

Although only 5 of 17 physical and chemical parameters showed differences due to bedding when measured in fresh manure at the outset of composting (Day 0), 11 of the 17 showed significant differences at the final compost sampling on Day 124. The more pronounced bedding effect in finished compost suggests that the influence of bedding is more of an issue when land applying finished compost vs. fresh manure. These differences were elicited with a 4:1 manure/bedding ratio. We would expect differences to become more pronounced if the ratio were reduced by the addition of greater amounts of bedding to feedlot pens.

Significantly lower losses of N occurred with composting of WBM compared with SBM, which has implications for air quality (NH3 emissions) and use of the product as a soil amendment. Other properties of WBM compost, perhaps making it more attractive than SBM compost as soil amendment, included significantly higher TC (if increased soil C storage is the aim, e.g., application to eroded or industrially disturbed lands) and IN levels (an advantage for a pre-seeding application to cropland) and significantly lower TP (advantageous on soils with medium to high P loadings).

Ethanol production has been promoted on the Canadian prairies as a means of diversifying agriculture and improving the environment (Freeze and Peters, 1999). Although the current ethanol industry is grain based, one future scenario is the production of ethanol from biomass sources such as cereal straw (Iogen Corporation, 2005). Therefore, the potential exists for diversion of cereal straw from livestock bedding to ethanol production. This may increase the price of straw, making the use of wood chips more economical.

Our work indicates that the incorporation of wood chip bedding into feedlot management regimes should not pose a problem for subsequent composting of the manure because most measured parameters reflected positively on WBM. Our composting results, in conjunction with other benefits from the use of wood chips gleaned from our larger study (e.g., cleaner cattle, less labor associated with bedding, no loss in crop production when land-applied), should encourage the adoption of wood chips over straw as a bedding choice for commercial feedlots.


   ACKNOWLEDGMENTS
 
We thank the Canada-Alberta Beef Industry Development Fund for partial financial support of this study. The technical help of Clarence Gilbertson, Bonnie Tovell, Wayne McKean, Alfredo Cárcamo, Mary-Lynn Muhly, and Wendi Smart is gratefully appreciated.


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




This article has been cited by other articles:


Home page
 Agron. J.Home page
J. J. Miller, B. W. Beasley, C. F. Drury, and B. J. Zebarth
Barley Yield and Nutrient Uptake for Soil Amended with Fresh and Composted Cattle Manure
Agron. J., September 1, 2009; 101(5): 1047 - 1059.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
W. Xu, T. Reuter, G. D. Inglis, F. J. Larney, T. W. Alexander, J. Guan, K. Stanford, Y. Xu, and T. A. McAllister
A Biosecure Composting System for Disposal of Cattle Carcasses and Manure Following Infectious Disease Outbreak
J. Environ. Qual., February 6, 2009; 38(2): 437 - 450.
[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 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 Google Scholar
Google Scholar
Right arrow Articles by Larney, F. J.
Right arrow Articles by McAllister, T. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Larney, F. J.
Right arrow Articles by McAllister, T. A.
Agricola
Right arrow Articles by Larney, F. J.
Right arrow Articles by McAllister, T. A.
Related Collections
Right arrow Nutrient Cycling
Right arrow Nutrient Management
Right arrow Industrial Waste
Right arrow Irrigation

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