lable at ScienceDirect

Journal of Cleaner Production 86 (2015) 9e14

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier .com/locate/ jc lepro

Improving the nutrient content of sheep bedding compost by addingcattle manure

Monica Sarolli Silva de Mendonça Costa a, *, Taiana Cestonaro a,Luiz Antonio de Mendonça Costa b, Marcos Antonio Teofilo Rozatti a,Leocir Jos�e Carneiro a, Dercio Ceri Pereira a, Higor Eisten Francisconi Lorin a

a Research group on Water Resources and Environmental Sanitation, Agricultural Engineering Graduate Program, Western Paran�a State University,Cascavel, Brazilb CNPq/RHAE Scholarship, Rua Di Cavalcanti, 283 e Jardim Vit�oria, CEP 85813-280 Cascavel, Paran�a, Brazil

a r t i c l e i n f o

Article history:Received 1 May 2014Received in revised form31 July 2014Accepted 27 August 2014Available online 6 September 2014

Keywords:Rice huskOrganic matterCompostPrincipal componentsCluster

* Corresponding author. West Paran�a State UniverAgrícola, Center for Science and Technology, Rua UUniversit�ario, CEP 85819-110 Cascavel, Paran�a, Brazil.32203175; fax: þ55 45 32203262.

E-mail addresses: mssmc@ig.com.br, monica.costa@tcestonaro@hotmail.com (T. Cestonaro).

http://dx.doi.org/10.1016/j.jclepro.2014.08.0930959-6526/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Sheep bedding has favorable characteristics for composting, such as providing readily assimilable carbonand facilitating aeration, which are essential for the growth of composting microorganisms. However,when composted alone, it may have disadvantages related to the time of composting and to the fertilizervalue of the compost produced, due to its high lignocellulosic content. To investigate whether cattlemanure could improve composting of sheep bedding as well as the agronomic quality of the compostproduced, we mixed cattle manure with sheep bedding in varying proportions (0e100%) and evaluatedseveral parameters of the composting process and of the final compost, such as changes in temperature,macronutrients, and carbon content before and after the composting process. The addition of �50% cattlemanure to the mixture increased composting efficiency, resulting in a more stable compost with highernutrient content. The high content of rice husk in sheep bedding limited the addition of higher pro-portions of this component to the composting mixture. We identified an efficient way to improve thequality of compost generated during waste management of sheep manure and bedding.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Studies on sheep bedding have been conducted to evaluate theefficiency of the materials used in terms of production costs,development, animal well-being and behavior, as well as meatquality (Gonçalves et al., 2008; Wolf et al., 2010; Teixeira et al.,2012, 2013, 2014). Most results indicate that sheep bedding is ad-vantageous for these variables.

However, after removing the material used as bedding in orderto replace it, it must be stabilized in order to be reused as a source ofnutrients for plants and organic matter for the soil. Therefore, thecharacteristics of the material chosenwill directly affect the qualityof the final organic fertilizer. Borhan et al. (2014) consider it achallenge to find suitable materials for bedding that have high

sity (UNIOESTE), Engenharianiversit�aria, 2069 e JardimTel.: þ55 45 99416480, 55 45

unioeste.br (M.S.S.M. Costa),

moisture absorption, better cow comfort, high fertilizer value andthat promote improvements in the facility's internal environment.

In most cases, the material used for bedding is chosen based onvariables such as ease in obtaining (Avila et al., 2007; Wolf et al.,2010), water absorbency (Steef et al., 2009); emission of green-house gases (Garlipp et al., 2011), and water absorbency andemission of greenhouse gases (Borhan et al., 2014). The efficiency ofthe stabilization processes for this material and the agronomicvalue of the final product are most often overlooked.

Because it is a solid waste product, the sheep bedding can bestabilized through the composting process. The C:N ratio of thismaterial, and especially the quality of C, can affect the performanceof this process (Bernal et al., 2009). Rice husk is commonly used asbedding material, as it is a residue of the processing of cereal.However, its silica content makes it highly resistant to degradation.Leconte et al. (2009) observed that rice hulls were resistant tomicrobial attack, more than the sawdust itself. This was attributedto the large contact surface of the sawdust and to the waterretention aided by the waterproof rice husk (due to the presence ofsilica).

Thus, the use of only sheep bedding (sheep rice huskþwaste) inthe compost can cause unfavorable results, such as a long

M.S.S.M. Costa et al. / Journal of Cleaner Production 86 (2015) 9e1410

stabilization period and low fertilizer value of the final compost.Given these problems, the concept of co-composting with otherwaste products may be introduced to benefit the system. Co-composting involves mixing one or more types of materials in or-der to improve the characteristics of the main residue and facilitateits stabilization. This strategy is being evaluated to improve thecomposting of waste that is difficult to degrade; one can add wasteto decrease the C:N ratio (Zhang and He, 2006; Liu et al., 2011;Rashad et al., 2010) or straw to increase the C:N ratio of animalwaste (Petric et al., 2009; Das et al., 2011; Qian et al., 2014). Whencomposted without an easily degradable nitrogen source, rice huskpresents lower values of organic matter decomposition, a low AH/AF ratio, and lower cellulase activity. When mixed mostly withcomposted animal waste, the decomposition becomes more effi-cient (Rashad et al., 2010).

In the case of sheep bedding, mixing with cattle manure canimprove the composting process and the value of the final compost.Cattle manure presents good composting performance (including ahigh degradation rate of organic matter), as well as a high nutrientconcentration in the system, characteristics that are both envi-ronmentally and economically favorable (Orrico Junior et al., 2012).

Given that several composting parameters are usually moni-tored (in order to evaluate the performance and the final product)and that such parameters are interrelated, it is important to analyzedata using multivariate statistical techniques. These analysesconsider the dependency relationships between variables, studythe similarity between the sample elements, and allow simplifyingthe structure of data variability (Mingoti, 2005).

The aim of the current study was to combine sheepbeddingþ cattle manure while trying to establish an adequate ratiobetween the mixtures to maximize the efficiency of the procedureas well as the quality of the final compost.

2. Material and methods

2.1. Origin of residues: sheep bedding and cattle manure

The sheep bedding used in this study comes from a feedlotwhere the animals remain on the bed (rice husk) throughout thefattening cycle. The cattle manure come from a confinement sys-tem where the animal feed is composed of 60% forage from thewhole corn plant silage, and 40% concentrate, containing grainbran and mineral supplement. The characteristics of the sheepbedding and the cattle manure used in this study are shown inTable 1.

Table 1Characteristics of the sheep bedding and the cattle manure used as compostingsubstrates.

Parameter Substrate

Sheep bedding Cattle manure

pH 8.96 9.10EC (dS m�1) 12.76 10.23Carbon (%) 41.84 43.88TKN (%) 1.50 2.79Phosphorus (g kg�1) 5.67 7.89Potassium (g kg�1) 13.27 38.45Ash (%) 24.68 21.02Cellulose (%) 26.20 18.00Hemicellulose (%) 20.77 22.88HA/FA 0.84 1.23C/N 28 16

EC: Electrical conductivity; TKN: Total Kjedahl nitrogen; HA/FA: Humic acid/fulvicacid; C/N: Carbon/nitrogen.

2.2. Experimental set-up and procedures

The experiment was conducted at the Experimental Center ofAgricultural Engineering (NEEA) of the Western Paran�a State Uni-versity (UNIOESTE), Cascavel campus, Brazil, from April to July2012. Windrows were formed on a covered, concrete floor-composting pad, to protect compost from the weather. Ten wind-rows were formed, two from each experimental condition, basedon the following sheep bedding and cattle manure ratios: 0:100,25:75, 50:50, 75:25, 100:0, representing experimental conditionsT100, T75, T50, T25, and T0, respectively. Each windrow had an initialweight of 200 kg dry matter (DM) basis. Three days after windrowshad been formed (which was considered composting day 1), aninitial sample was collected from each windrow. Sub-samples werecollected at six different points within the swath. After being ho-mogenized, the sub-samples formed a single composite sample.After windrows were formed, manual turning was performed twicea week in the first month and weekly afterward to aerate andmoisten the material. Moisture content was recorded at eachturning by drying a small amount of windrowmaterial. The amountof water added to keep moisture content around 60% was calcu-lated individually for each windrow and depended on windrowweight at each turning. Composting was monitored until windrowtemperature was similar to the ambient temperature, when thematerial was considered stable.

2.3. Analytical methods

Windrow temperature was monitored daily at six points using amercury thermometer graduated to 250 �C, inserted to 0.2 m depthinto each windrow. Homogenized samples from each windrowwere analyzed at the beginning and end of composting. The sam-ples were pre-dried at 50 �C in a forced air oven up to constantmassto avoid losses, especially of nitrogen. After drying, the sampleswere ground in a grinder with a 20mesh sieve, and then used for alldeterminations (contents of nitrogen, phosphorus, potassium, ashand carbon and organic matter fractionation), which were analyzedin amoist sample. In all analyses, results were corrected to dry basis(105 �C).

Electrical conductivity (EC) and pH were determined in thesame solution, consisting of a suspension of sample in distilledwater (1:10 m/v). pH readings were done using a TECNAL TEC-3MPbench potentiometer and electrical conductivity was determinedusing an MS Tecnopon Special Equipment mCA 150 bench con-ductometer. Ash content was determined according to the Amer-ican Public Health Association (APHA et al., 2012).

Carbon (C) content was determined by ignition in a mufflefurnace at 550 �C for 12 h, according to Cunha-Queda et al. (2003).Organic matter content obtained after burning was divided by 1.8to determine carbon content. Organic matter fractionation and thequantification of carbon in humic acid (HA) and fulvic acid (FA)fractions used to calculate the HA/FA ratio were performed ac-cording to Benites et al. (2003).

Total Kjedahl nitrogen (TKN) was estimated using a Kjedahldistiller, according to Malavolta et al. (1997). The C/N ratio wascalculated from C and TKN estimates. Cellulose and hemicellulosecontents were obtained by determining neutral detergent fiber(NDF) and acid detergent fiber (ADF) in a Marconi MA-444/CIdigestion fiber apparatus using the sequential method describedby Campos et al. (2004). Phosphorus (P) and potassium (K) contentswere determined using the dry extraction procedure described inAlcarde (2009). Themethod involves heating the sample in amufflefurnace at 550 �C, thus allowing the destruction of organic matter.After firing, the residue is dissolved in HCl and transferred to avolumetric flask before analysis. Phosphorus readings were

M.S.S.M. Costa et al. / Journal of Cleaner Production 86 (2015) 9e14 11

performed using a Femto 700 Plus spectrophotometer, according toMalavolta et al. (1997), whereas potassium contentwas determinedby flame photometry using a Digimed DM-62 photometer.

Volume measurements were performed using a wooden box ofknown dimensions, where compost was placed. Windrow masswas monitored using a Filizola ID 1500 digital scale with 300 kgcapacity and 0.2 kg accuracy.

2.4. Experimental design and statistical analysis

We used a completely randomized design with five experi-mental conditions and two replicates, totaling 10 experimentalunits. The data were analyzed using principal component analysis(PCA) and cluster analysis.

We used PCA to investigate the relationships between the var-iables monitored and between the variables and experimentalconditions. The principal components (PCs) were extracted fromthe correlation matrix of the original variables to avoid any inter-ference from the measurement units of the variables studied. Weselected enough PCs to explain 80e90% of the variance of the data(Johnson and Wichern, 2007). The data used in the PCA corre-sponded to the variation of the parameters during composting, andwere calculated by measuring the difference between the final andinitial compost values, which were not transformed to percentagevalues.

We used cluster analysis to group the individuals (experimentalconditions). We calculated the matrix of Euclidean distances andused the average linkage hierarchical method to form the clusters.We standardized observations with mean 0 and variance 1. Ana-lyses were performed using the R statistical software (RDevelopment Core Team, 2012).

3. Results and discussion

3.1. Temperature

Composting time ranged from 82 days at T0 to 96 days at T100,T75, T50, and T25. Daily temperature variation during the compostingperiod is shown in Fig. 1.

Temperature was lower in windrows with higher amounts ofsheep bedding in the mixture during the entire composting period.The granulometric characteristic as well as the size of the particles

15

20

25

30

35

40

45

50

55

60

65

1 8 15 22 29 36 43 50 57 64 71 78 85 92

Tem

pera

ture

(°C

)

Composting day

T100T75T50T25T0

Fig. 1. Mean daily compost temperature variations during composting of sheepbedding mixed with cattle manure, in different proportions. Sheep bedding:cattlemanure proportions: T100 (0:100); T75 (25:75); T50 (50:50); T25 (75:25); T0 (100:0).

of the constituentmaterial in the bedding contributed to this result,as it promoted greater aeration and retained less heat in thewindrow. The thermophilic phase of mixtures with �50% cattlemanure (T50, T25, and T0) lasted approximately 25 days, whereasmixtures with smaller cattle manure increments resulted in shorter(10-day) thermophilic phases. Because the self-heating ability ofcompost highly correlates with the amount of biodegradable mat-ter (Guardia et al., 2010), this behavior reflects the lack of readilyavailable carbon and limited microbial activity of compostingmixtures with less than 50% cattle manure.

Leconte et al. (2009) found a longer thermophilic phase insawdust mixed with poultry manure (105 days) compared to ricehusk mixed with poultry manure (40e60 days), even though thetwo substrates in natura (sawdust and rice husk) exhibited similarC:N ratios and contents of nitrogen, cellulose, hemicellulose, andlignin. The authors argued that the lower microbial activity ofwindrows with rice husk (reflected in the shorter thermophilicphase) may be related to the lower water retention and specificsurface area for microbial attack, due to the wax coating and silicacontent of the rice husk. This argument provides a valid explanationfor the short thermophilic phases observed in our study for com-posting mixtures with greater than 50% rice husk.

3.2. Compost characteristics

The chemical characteristics and composition of the initialmaterial and the final compost for each experimental condition areshown in Table 2.

The highest proportions of sheep bedding in the mixture (T100,T75) resulted in an organic compost with lower TKN, P, and Kcontent and greater C:N ratio (Table 2). These results were ex-pected, because the sheep bedding mixture used in this studycontained high amounts of rice husk, which has low nutrientcontent. Because nitrogen is an important fertilizer, a higheramount of composted sheep bedding would be required to meet Nrequirements of cultures compared to the compost with cattlemanure.

In their review on composting of animal manures and chemicalcriteria for compost maturity assessment, Bernal et al. (2009)suggest that the C:N ratio values considered ideal for compostingrange between 25 and 35:1. However, they conclude that amongstthe controllable factors that influence manure composting, theselection of appropriate bulking agents plays an essential role incontrolling the decomposition rate and favoring N retentionwithinthe compost. Therefore, strategies such as adding a bulking agentwith degradable organic-C to enhance initial N immobilization andprocess control (moisture, temperature, aeration/turning and par-ticle size) have been shown to reduce ammonia volatilization andhence nitrogen losses, two major concerns in manure compostingfrom an environmental point of view.

The AH/AF ratio decreased with the reduction in the proportionof cattle manure in the mixture. This ratio represents the maturityof the compost, as it indicates an increase in polymerization rate ofthe material (Iglesias-Jim�enez et al., 2007). Thus, the higher theratio, the higher the stability of the compost. An increase inaliphatic compounds from degradation of the readily availableorganic matter such as polysaccharides is observed during the earlyphase of composting. These compounds are restructured into hu-mic acids during the composting process, indicating the loss ofaliphatic compounds that are replaced by aromatic compounds atthe end of composting (Amir et al., 2010). Kulikowska and Klimiuk(2011) observed that temperature was the main factor affectingpolymerization from fulvic to humic acids, as the degree of poly-merization increased more efficiently under thermophilic condi-tions. In agreement with these studies, we observed that longer

Table 2Characteristics of the initial material and final compost evaluated in composting sheep bedding mixed with cattle manure.

Treatment Parameter

TKNa Pb Kb Ca HA/FA C:N Asha Cell.a Hemi.a Massa Vol.c

T100 Initial 2.72 7.94 37.02 43.75 1.26 16 21.26 20.59 23.67 192.55 0.89Final 3.44 11.52 50.23 37.29 3.04 11 32.87 7.68 12.77 120.62 0.49

T75 Initial 2.39 6.98 30.41 43.59 1.12 18 21.54 23.96 21.84 196.39 0.96Final 3.07 10.54 40.25 37.41 2.58 12 32.66 11.58 13.07 129.78 0.59

T50 Initial 1.97 5.60 21.61 43.03 1.02 22 22.54 25.80 20.66 200.05 0.99Final 2.65 9.15 31.20 38.11 2.11 14 31.40 15.26 14.76 138.69 0.68

T25 Initial 1.55 4.73 14.65 42.74 0.95 28 23.07 27.24 19.51 191.28 1.00Final 2.23 8.01 18.62 39.26 1.57 18 29.34 18.16 16.51 142.99 0.76

T0 Initial 1.25 3.91 9.89 42.31 0.84 34 23.84 30.51 19.66 189.85 1.09Final 1.91 6.46 12.84 39.59 1.04 21 28.89 21.80 18.29 155.73 0.98

TKN, Total Kjedahl nitrogen; Cell., Cellulose; Hemi., Hemicellulose; Vol., Volume. HA/FA, ratio between concentrations of humic acid and fulvic acid.Sheep bedding:cattle manure proportions: T100 (0:100); T75 (25:75); T50 (50:50); T25 (75:25); T0 (100:0).

a concentration in %.b concentration in g kg�1.c expressed in m3.

M.S.S.M. Costa et al. / Journal of Cleaner Production 86 (2015) 9e1412

thermophilic phases were associated with higher degrees of poly-merization (i.e., higher HA/FA ratios) at the end of the compostingperiod.

The volume occupied by windrows with 100% sheep bedding(T0), even after the composting process, is twice that observed atT100. This finding highlights the difficulty of material degradation,mainly due to the presence of rice husk. Since there is little reductionin the volume of the windrows, the area required for composting isgreater when the sheep bedding is composted without mixtures.

3.3. Principal component analysis

The linear correlation coefficients between the variables areshown in Table 3.

The high number of correlations of moderate to strong magni-tude shows the interdependence of variables and the importance ofusing a statistical analysis such as PCA to address these relation-ships (Lattin et al., 2011). The first two PCs selected from the PCAalone explained 88.61% of the variance in the data.

Principal component 1 explained 79.22% of the variance and thevariables most responsible for this result were ash, HA/FA, K, and P(negatively correlated), and cellulose, hemicellulose, carbon, mass,and volume (positively correlated). PC1 can be characterized as anindex for comparing variables that increase in concentration andvariables that decrease in concentration in the composting process.

This index can be named organic/mineral fraction, as mostvariables that increased in concentration (ash, P, and K) are min-erals (except for HA/FA ratio), whereas variables that decreased inconcentration (cellulose, hemicellulose, carbon, mass, and volume)represented losses of the organic fraction. Thus, concentrations of

Table 3Linear correlation between the different variables in composting materials.

TKN P K C HA

TKN 1P 0.14 1K 0.26 0.69 1C �0.30 ¡0.76 ¡0.93 1HA/FA 0.33 0.71 0.95 ¡0.96 1Ash 0.30 0.75 0.93 ¡1.00 0Cell. �0.26 �0.54 �0.65 0.74 �0Hemi. �0.35 ¡0.71 ¡0.94 0.98 ¡0Mass �0.22 ¡0.71 ¡0.90 0.90 ¡0Vol. �0.28 ¡0.89 ¡0.89 0.96 ¡0

The bold values are correlations of moderate to strong magnitude.TKN, Total Kjedahl nitrogen; Cell: Cellulose; Hemi: Hemicellulose; Vol: Volume. HA/FA,

variables in the organic and in the mineral fractions change in thesame direction (negative and positive, respectively) during thecomposting process.

In contrast, while the values of variables in the organic fractiongroup become more negative, values of variables in the mineralgroup become more positive. The consumption of the organicfraction occurs mainly due to the requirement of C for the main-tenance of the microbial metabolism, reducing the mass of thecompost pile. Brito et al. (2012) observed a similar phenomenonduring cattle manure composting, in which nutrient contentcorrelated highly with a reduction in organic matter content.

Windrows corresponding to treatments T0 and T25 contributedpositively to PC1, whereas windrows corresponding to treatmentsT100 and T75 contributed negatively to PC1. In addition, treatmentT50 (windrows 5 and 6) exhibited a similar behavior to T100 and T75.Treatments T100 and T75 had the highest weights for the mineralgroup, as they were the treatments with the highest concentrationsof these variables, followed by T50. Conversely, T25 and T0 had thehighest weights for the organic fraction, as the two treatments hadthe lowest losses of this component.

The lower transformations of the organic fraction obtained us-ing low amounts of cattle manure in themixture (and consequentlylower nutrient content and stabilization of the process) are possiblymore related to the presence of rice husk in the compost than to thecharacteristics of sheep manure.

We observed that the use of rice husk as bedding has theadvantage of modifying the structure of manure pellets, which fa-cilitates microbial attack and, consequently, their degradation.However, rice husk e which is rich in cellulose, lignin, and silica e

makes the sheep bedding material itself difficult to decompose.

/FA Ash Cell. Hemi. Mass Vol.

.96 1

.66 ¡0.75 1

.98 ¡0.98 0.71 1

.96 ¡0.90 0.71 0.90 1

.93 ¡0.95 0.76 0.93 0.92 1

ratio between concentrations of humic acid and fulvic acid.

Tratamentos

Dis

tânc

iaE u

clid

iana

T0T25T50T75T100

5,03

3,36

1,68

0,00

Cluster 2Cluster 1

Fig. 3. Dendrogram constructed using average linkage hierarchical clustering andEuclidean distance of variables TKN, P, K, ash, HA/FA ratio, mass, volume, carbon,cellulose, and hemicellulose.

-1.0 -0.5 0.0 0.5 1.0 1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

0.0 0.5

PC1 Scores

PC

2 S

core

s

1

2

3

4

5

6

7

8

9

10

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

0.0

1.5TKN

1Phosphorus11PPhospho

PotassiumCarbon

uHA/FAPotassiPotassiHA/FAAsh Cellulose

Hemicellulose

CMass

nCelluloseCarbon

MassVolume

T100: 1 - 2T75: 3 - 4T50: 5 - 6T25: 7 - 8T0: 9 -10

T100

T75

T50

8T251T0

1.0

0.5

-0.5

-1.0

-1.5

Fig. 2. Biplot graph of the principal component analysis (PCA) for the variation ofparameters during composting of sheep bedding mixed with cattle manure in differentproportions. Sheep bedding:cattle manure proportions: T100 (0:100); T75 (25:75); T50(50:50); T25 (75:25); T0 (100:0). Arrows indicate the variables and numbers indicatethe treatments. The clustering of the treatments represented in the graph is based onPC1 only.

M.S.S.M. Costa et al. / Journal of Cleaner Production 86 (2015) 9e14 13

Among rice husk constituents, cellulose and lignin are found inhigh concentrations (59% and 20.9%, respectively), as is silica, whichrepresents 65% of the ash content in the husk (Leconte et al., 2009;Ma et al., 2012).

We observed that rice husk is weakened during compostingbecause of thermophilic temperatures, high density of microor-ganisms, and mechanical breakdown by turning. However, thereis little organic fraction available to microorganisms at highersheep bedding proportions, and when it ends, decompositiontends to occur slowly. Conversely, studies that used sheep manurein the composting process mixed with other residues withoutbedding reported high losses of organic matter (Canet et al.,2008). Orrico et al. (2007) reported satisfactory results inreducing windrow mass and pathogen density in goat manurecomposting.

In an attempt to improve the decomposition of rice husk, Kaosolet al. (2012) mixed this component with other residues, nitrogensources, or nitrogen in the form of urea. Serrami�a et al. (2010)observed that the addition of nitrogen as urea to compostingmixtures rich in lignocellulosic residues favored the degradation ofcellulose and hemicellulose, mainly by reducing composting timecompared to mixtures of lignocellulosic material with sheepmanure only. This reduction in composting time is due to the highmetabolic activity generated by the availability of large amounts ofnitrogen in the initial composting material. In our study, the use ofcattle manure was a source of nitrogen that promoted sheepbedding decomposition.

PC2 explained 9.39% of the variance of the data and correlatedwith TKN only. This variable did not correlate with any variablestudied and showed no defined behavior with the treatments. TKNcontent was similar in all treatments during the composting period,due to the reduction in carbon content, which has also beenmentioned by Kabor�e et al. (2010). Because organic matter degra-dation was lower in treatments with sheep bedding, a lower TKNcontent would be expected, as observed with P and K. Thus, the

lower microbial activity and consequently shorter thermophilicphase resulted in greater nitrogen conservation in treatments withhigher sheep bedding proportions. In general, TKN losses andcontent in composting are also related to the characteristics of theresidues used (Chen et al., 2010; VelascoeVelasco et al., 2011),frequency of turning, and environmental conditions (Parkinsonet al., 2004). Because in this study windrows were formed underthe same conditions, the initial composition of the mixture,regardless of sheep bedding proportion, did not affect TKN content.

3.4. Cluster analysis

A cluster analysis was performed between treatments (Fig. 3) tocomplement the principal component analysis.

The cluster analysis yielded results similar to those of PCAregarding the clustering of the treatments. The first clusteringshowed that treatments T100 and T75 were the most similar andhad the smallest Euclidean distance. In the next clustering step,T50 becomes part of the group that includes T25 and T0. At a similardistance, T25 and T0 are clustered. Then, the structure of the clusterremains stable for a long distance interval, resulting in the for-mation of two groups: the first formed by T100, T75, and T50, andthe second formed by T25 and T0. Thus, the addition of �50% cattlemanure in the mixture increased compost efficiency, whereas<25% cattle manure in the mixture reduced compost efficiency(Fig. 2).

4. Conclusions

The inclusion of �50% cattle manure mixed with sheep beddingresulted in a more stable organic compost, with higher nutrientcontent and HA/FA ratio. Thus, our study has identified an efficientway to improve the quality of compost generated during the wastemanagement of sheep bedding.

To facilitate the composting of sheep bedding without mixturesand the production of a higher quality compost, we suggestsubstituting the rice husk.

References

Alcarde, J.C., 2009. Manual de an�alise de fertilizantes. FEALQ, Piracicaba.Amir, S., Jouraiphy, A., Meddich, A., El Gharous, M., Winterton, P., Hafidi, M., 2010.

Structural study of humic acids during composting of activated sludge-greenwaste: elemental analysis, FTIR and 13C NMR. J. Hazard. Mater. 177, 524e529.

M.S.S.M. Costa et al. / Journal of Cleaner Production 86 (2015) 9e1414

APHA, AWWA, WEF, 2012. Standard Methods for the Examination of Water andWastewater, 22nd ed. American Public Health Association, American WaterWorks Association, Water Environment Federation, Washington.

Avila, V.S. de, Costa, C.A.F., Figueiredo, E.A.P. de, Rosa, P.S., Oliveira, U. de,Abreu, V.M.N., 2007. Materiais alternativos, em substituiç~ao �a maravalha comocama de frangos. Embrapa Suínos e Aves, Conc�ordia. Comunicado T�ecnico 465.

Benites, V.M., Madari, B., Machado, P.L. de A., 2003. Extraç~ao e fracionamentoquantitativo de substancias húmicas do solo: um procedimento simplificado debaixo custo. Embrapa Solos, Rio de Janeiro. Comunicado T�ecnico 16.

Bernal, M.P., Albuquerque, J.A., Moral, R., 2009. Composting of animal manures andchemical criteria for compost maturity. Bioresour. Technol. 100, 5444e5453.

Borhan, S., Rahman, S., Hammer, C., 2014. Water absorption capacity of flax and pinehorse beddings and gaseous concentrations in bedded stalls. J. Equine Vet. Sci.34, 611e618.

Brito, L.M.l, Mour~ao, I., Coutinho, J., Smith, S.R., 2012. Simple technologies for on-farm composting of cattle slurry solid fraction. Waste Manag. 32, 1332e1340.

Campos, F.P. de, Nussio, C.M.B., Nussio, L.G., 2004. M�etodos de An�alise de Alimentos.FEALQ, Piracicaba.

Canet, R., Pomares, F., Cabot, B., Chaves, C., Ferrer, E., Rib�o, M., Albiach, M.R., 2008.Composting olive mill pomace and other residues from rural southeasternSpain. Waste Manag. 28, 2585e2592.

Chen, Y., Huang, X., Han, Z., Huang, X., Hu, B., Shi, D., Wu, W., 2010. Effects of bamboocharcoal and bamboo vinegar on nitrogen conservation and heavy metalsimmobility during pig manure composting. Chemosphere 78, 1177e1181.

Cunha-Queda, A.C.F., Vallini, G., Bruno de Sousa, R.F.X., Almeida Duarte, E.C.N.F.,2003. Estudo da evoluç~ao de actividades enzim�aticas durante a compostagemde resíduos provenientes de mercados horto-frutícolas. An. do Inst. Super.Agron. 49, 193e208.

Das, M., Uppal, H.S., Singh, R., Beri, S., Mohan, K.S., Gupta, V.C., Adholeya, A., 2011.Co-composting of physic nut (Jatropha curcas) deoiled cake with rice straw anddifferent animal dung. Bioresour. Technol. 102, 6541e6546.

Garlipp, F., Hessel, E.F., van den Weghe, H.F.A., 2011. Characteristics of gas genera-tion (NH3, CH4, N2O, CO2, H2O) from horse manure added to different beddingmaterials used in deep litter bedding systems. J. Equine Vet. Sci. 31, 383e395.

Gonçalves, A.L., Lana, R. de P., Vieira, R.A.M., Henrique, D.S., Mancio, A.B.,Pereira, J.C., 2008. Avaliaç~ao de sistemas de produç~ao de caprinos leiteiros naRegi~ao Sudeste do Brasil. R. Bras. Zootec. 37, 366e376.

Guardia, A. de, Mallard, P., Teglia, C., Marin, A., Le Pape, C., Launay, M., Benoist, J.C.,Petiot, C., 2010. Comparison of five organic wastes regarding their behaviourduring composting: Part 1, biodegradability, stabilization kinetics and temper-ature rise. Waste Manag. 30, 402e414.

Iglesias-Jim�enez, E., Barral-Silva, M.T., Marhuenda-Egea, F.C., 2007. Indicadores de laestabilidad y madurez del compost. In: Moreno-Casco, J., Moral-Herrero, R.(Eds.), Compostaje. Mundi-Prensa, Madrid, pp. 243e284.

Johnson, R.A., Wichern, D.W., 2007. Applied Multivariate Statistical Analysis, sixthed. Prentice-Hall, New Jersey.

Kabor�e, T.W., Houot, S., Hien, E., Zombr�e, P., Hien, V., Masse, D., 2010. Effect of theraw materials and mixing ratio of composted wastes on the dynamic of organicmatter stabilization and nitrogen availability in composts of Sub-Saharan Africa.Bioresour. Technol. 101, 1002e1013.

Kaosol, T., Kiepukdee, S., Towatana, P., 2012. Influence of nitrogen containing wastesaddition on natural aerobic composting of rice straw. Am. J. Agric. Biol. Sci. 7,121e128.

Kulikowska, D., Klimiuk, E., 2011. Organic matter transformations and kineticsduring sewage sludge composting in a two-stage system. Bioresour. Technol.102, 10951e10958.

Lattin, J.M., Carroll, J.D., Green, P.E., 2011. An�alise de dados multivariados. CengageLearning, S~ao Paulo.

Leconte, M.C., Mazzarino, M.J., Satti, P., Iglesias, M.C., Laos, F., 2009. Co-compostingrice hulls and/or sawdust with poultry manure in NE Argentina. Waste Manag.29, 2446e2453.

Liu, D., Zhang, R., Wu, H., Xu, D., Tang, Z., Yu, G., Xu, Z., Shen, Q., 2011. Changes inbiochemical and microbiological parameters during the period of rapid com-posting of dairy manure with rice chaff. Bioresour. Technol. 102, 9040e9049.

Ma, X., Zhou, B., Gao, W., Qu, Y., Wang, L., Wang, Z., Zhu, Y., 2012. A recyclablemethod for production of pure silica from rice hull ash. Powder Technol. 217,497e501.

Malavolta, E., Vitti, G.C., Oliveira, S.A., 1997. Avaliaç~ao do estado nutricional dasplantas: princípios e aplicaç~oes, second ed. POTAFOS, Piracicaba.

Mingoti, S.A., 2005. An�alise de dados atrav�es de m�etodos de estatística multi-variada: uma abordagem aplicada. UFMG, Belo Horizonte.

Orrico, A.C.A., Lucas Júnior, J. de, Orrico Junior, M.A.P., 2007. Alteraç~oes físicas emicrobiol�ogicas durante a compostagem dos dejetos de cabras. Eng. Agríc. 27,764e772.

Orrico Junior, M.A.P., Orrico, A.C.A., Lucas Junior, J. de, Sampaio, A.A.M.,Fernandes, A.R.M., Oliveira, E. A. de, 2012. Compostagem dos dejetos da bovi-nocultura de corte: influencia do período, do gen�otipo e da dieta. R. Bras.Zootec. 41, 1301e1307.

Parkinson, R., Gibbs, P., Burchett, S., Misselbrook, T., 2004. Effect of turning regimeand seasonal weather conditions on nitrogen and phosphorus losses duringaerobic composting of cattle manure. Bioresour. Technol. 91, 171e178.

Petric, I., Sestan, A., Sestan, I., 2009. Influence of wheat straw addition on com-posting of poultry manure. Process Saf. Environ. Prot. 87, 206e212.

Qian, X., Shen, G., Wang, Z., Guo, C., Liu, Y., Lei, Z., Zhang, Z., 2014. Co-composting oflivestock manure with rice straw: characterization and establishment ofmaturity evaluation system. Waste Manag. 34, 530e535.

R Development Core Team, 2012. R: a Language and Environment for StatisticalComputing. R Foundation for Statistical Computing, Vienna.

Rashad, F.M., Saleh, W.D., Moselhy, M.A., 2010. Bioconversion of rice straw andcertain agro-industrial wastes to amendments for organic farming systems: 1.Composting, quality, stability and maturity indices. Bioresour. Technol. 101,5952e5960.

Serrami�a, N., S�anchez-Monedero, M.A., Fern�andez-Hern�andez, A., García-OrtizCivantos, C., Roig, A., 2010. Contribution of the lignocellulosic fraction of two-phase olive-mill wastes to the degradation and humification of the organicmatter during composting. Waste Manag. 30, 1939e1947.

Steef, J.J.; Lips, I~nguez de Heredia, G.M., Op den Kamp, R.G.M., van Dam, J.E.G., 2009.Water absorption characteristics of kenaf core to use as animal bedding ma-terial. Ind. Crops Prod. 29, 73e79.

Teixeira, D.L., Lama, M. de la, Villarroel, M., Garcia-Belenguer, S., Sa~nudo, C.,Maria, G.A., 2012. Effect of straw on lamb welfare, production performanceand meat quality during the finishing phase of fattening. Meat Sci. 92,829e836.

Teixeira, D.L., Lama, G.C.M. de la, Pascual-Alonso, M., Aguayo-Ulloa, L., Villarroel, M.,Maria, G.A., 2013. A note on lamb's choice for different types of bedding ma-terials. J. Vet. Behav. 8, 175e179.

Teixeira, D.L., Villarroel, M., María, G.A., 2014. Assessment of different organicbeddings materials for fattening lamb. Small Rumin. Res. 119, 22e27.

Velasco-Velasco, J., Parkinson, R., Kuri, V., 2011. Ammonia emissions during ver-micomposting of sheep manure. Bioresour. Technol. 102, 10959e10964.

Wolf, B.R., Molloy, H.R.B., Trayte, M.J., Rose, M.T., 2010. Behaviour of growing lambshoused on straw or woodchip bedding materials and their preference for floortype. Appl. Anim. Behav. Sci. 124, 45e50.

Zhang, Y., He, Y., 2006. Co-composting solid swine manure with pine sawdust asorganic substrate. Bioresour. Technol. 97, 2024e2031.