Feasibility of Miscanthus as alternative bedding for dairy cows

Transcription

1 Veterinrni Medicin,, 1 (3): Originl Pper doi: 1.171/9-VETMED Fesibility of Miscnthus s lterntive bedding for diry cows S. Vn Weyenberg, T. Ulens, K. De Reu, I. Zwertvegher, P. Demeyer, L. Pluym Institute for Agriculturl nd Fisheries Reserch (ILVO), Technology nd Food Science Unit, Merelbeke, Belgium ABSTRACT: Despite diry frmers wreness of the importnce of correct bedding for the helth nd comfort of their cows, they re often frugl with respect to these bedding mterils in order to reduce costs. In ddition, frmers re currently dependent on the vilbility nd price fluctutions of trditionl bedding mterils. For these resons, the scientific literture s well s the trde press point to n intensifying serch for ffordble lterntive bedding mterils. The im of this study ws to investigte whether Miscnthus, woody grss tht requires low input but genertes high yield, could replce strw in deep little diry cow cubicles. The cows cubicles were lined for two consecutive 1-dy periods with strw/chlk/wter bedding, followed by two consecutive 1-dy periods with ground-miscnthus/chlk bedding. No significnt differences were found in the following prmeters: loss/ wste of bedding mteril, bcteril growth in cubicles, cow skin lesions (except for crpus lesions), cow clenliness or cow comfort. Dust concentrtions mesured s PM 1 were higher when cubicles were filled with strwbsed bedding, but never exceeded workspce qulity sfety limits. Bsed on these results, one cn conclude tht Miscnthus hs potentil s vible lterntive to strw when used s bedding mteril. On-frm cultivtion of Miscnthus my increse diry frmer self-sufficiency nd could reduce bedding costs. Keywords: bedding mteril; cow comfort; bcteril lod List of bbrevitions ADF = cid detergent fibre, ADL = cid detergent lignin, CCI = cow comfort index, DM = dry mtter, NDF = neutrl detergent fibre, PM = prticulte mtter, SUI = stll use index Incresed evidence for the key role of diry cttle bedding in mintining nd promoting cow helth nd cow comfort hs rised the interest of both frmers nd reserchers in the type nd mount of bedding mteril used on diry frms. Bedding mteril cn be n importnt source of bcteril exposure, with potentil negtive impct on tet end colonistion, udder helth nd milk qulity (Hogn nd Smith 1997; Zdnowicz et l. ; Smpimon et l. ; Verbist et l. 11). In ddition, cows prefer dry nd soft bedding mteril tht gives them sufficient friction to lie down nd rise without slipping (Chplin et l. ; Fregonesi et l. 7). Incresed lying times nd better cow comfort re ssocited with higher feed intke (Metz 19), incresed rumintion ctivity, better clw helth (Vokey et l. 1), less stress (Munksgrd et l. 1999; Fisher et l. ) nd higher milk production (Clmri et l. 9; Cook nd Nordlund 9). Indeed, Bruijnis et l. (13) reveled tht better lying surfce (mttress nd bedding) ws one of the most cost-efficient mesures for optimising niml helth nd welfre. Fctors tht cn influence lying time include the dry mtter content, wter holding cpcity nd prticle size of the bedding mteril, but lso the presence of sufficient mount of bedding. Tucker et l. (9) found tht cows spent three more minutes lying down per ech dditionl kilogrm of shvings (Tucker et l. 9). The mount of bedding vilble depends on bed- 11

2 Originl Pper Veterinrni Medicin,, 1 (3): doi: 1.171/9-VETMED ding mngement (mount provided nd frequency of removl), s well s the cpcity of the bedding mteril to remin in the cubicles. Both qulity nd mount of bedding lso influence cow clenliness nd the presence of lesions. Lesions t crpus nd hock re generlly ssocited with hrd bedding mterils (Wery nd Tszkun ; Wechsler et l. ; Vokey et l. 1). Lesions cn develop into more severe injuries due to continuous pressure nd friction relted to the lying surfce (Schulze et l. 7). In ddition, bedding is known to hve gret impct on dust concentrtion in diry brns (Smdi et l. 1). Strw nd swdust, the most commonly used types of bedding, cn be very dusty (Breum et l. 1999). A lyer of chlk powder (lime) on strw or swdust, which is used to bsorb moisture, lso dds to the dust concentrtion (Smdi et l. 1). Dust, especilly the concentrtion of the dust frction up to 1 µm (PM 1 ), cn hve n dverse effect on niml helth nd productivity nd cn lso negtively impct worker helth (Tki et l. 199; Dolejs et l. ; Cmbr-Lopez et l. 1). Despite diry frmers growing wreness of the importnce of good qulity bedding mteril, they do not lwys pply this knowledge in prctice becuse of the cost ssocited with good bedding. They either use lower mounts of bedding mteril or choose lower qulity mterils. Cost is therefore one reson to serch for lterntives; nother is the restricted nd fluctuting vilbility of trditionl bedding mteril (minly swdust). In this study we explore the possibility of using Miscnthus s bedding mteril for diry cows. Miscnthus is C perennil woody grss tht cn be cultivted on-frm over period of 1 yers (Jezowski ). It hs low nutritionl requirements nd high yields in wide rnge of soils (Heo et l. 1). The nnul dry production rte is on verge t/h (Scurlock 1999), but cn rise to 3 t/h depending on the conditions (Dnltos et l. 7). Its high biomss potentil nd other chrcteristics hve led reserchers to study the potentil of Miscnthus s bio-energy crop. Its low input, high yield nd the fct tht it cn be cultivted on swmpy lnd with low vlue for crop production or livestock lso mke it worth studying s n lterntive to strw for use s bedding mteril. On-frm cultivtion of Miscnthus my led to greter self-sufficiency of frmers nd could men tht they will be less dependent on the vilbility nd price fluctutions of trditionl bedding mterils. The im of this pilot study ws to explore the potentil of Miscnthus gignteus s lterntive bedding in deep litter cubicles for diry cttle. MATERIAL AND METHODS Experimentl setup. Fifty lctting Holstein diry cows were housed in stble with 9 deep litter cubicles. The cubicles, which were rrnged in two rows, mesured m nd hd open fronts. The cows were subjected to longitudinl study on well-mnged, privtely-owned Belgin diry frm from October 1 to November 1. They were milked by milking robot nd forced cow trffic (feed to milk) ws used. Cubicles were normlly filled with combintion of kg strw (.9 m 3 ), kg wter nd kg chlk powder (lime). Ech dy, dirt nd feces were mnully removed from the cubicles nd the remining bedding ws eqully distributed over the cubicle by rking. After 1 dys, the bedding in the cubicles ws refreshed. To test the effect of replcing strw by Miscnthus, n experiment ws set up in which smples nd mesurements were crried out in two consecutive periods of 1 dys with cubicles filled with the trditionl combintion of strw/chlk/wter s bedding mteril. Afterwrds, strw ws replced by kg (3.1 m 3 ) Miscnthus for two consecutive periods of 1 dys. After hrvest for use in the bio-energy industry, Miscnthus is typiclly chopped. But for use s bedding mteril, the Miscnthus must be ground due to the presence of lrge woody pieces. Before use s bedding, bse smple ws tken nd nlysed for dry mtter (DM) (EN 13), neutrl detergent fibre (NDF), cid detergent fibre (ADF), cid detergent lignin (ADL) (dopted from Vn Soest et l. 1991) nd prticle size. Prticle size ws determined by pplying g of mteril (strw or Miscnthus) on shker with six sieves (dimeters 31., 1,,, nd 1 mm) during min. Bse smple nlyses reveled reltively low DM content of the ground Miscnthus. For this reson, no dditionl wter ws dded when using Miscnthus s bedding. Insted, combintion of kg ground Miscnthus (3.1 m 3 ) nd kg lime ws pplied to 9 cubicles for two consecutive periods of 1 dys. Mesurements of bedding. Five cubicles (.%) distributed t equl distnces throughout the stble were selected. On Dy (immeditely fter dding fresh bedding), Dy 9 nd Dy 1 (before 1

3 Veterinrni Medicin,, 1 (3): Originl Pper doi: 1.171/9-VETMED Figure 1. Digrm of the grid used to mesure bedding depth of selected cubicles eqully distributed throughout the stble on dy nd dy 1. The grid mesured cm nd ws plced on the cubicle prtition. Mesuring points re indicted s blck circles refreshing the bedding mteril), nine equl size subsmples were tken from stndrdised plces from the bck /3 of the cubicles nd mixed s one smple. Mixed smples from ech of the five selected cubicles were cooled (1 C) nd on the sme dy trnsported to the lb for nlysis of DM content (EN 13) nd microbiologicl nlyses. Microbiologicl nlyses were performed on the dy of smpling. The totl count of erobic bcteri, Enterobctericee, Escherichi coli (E. coli) nd Enterococcus spp. ws determined on Plte Count Agr (PCA, Bio-Rd 3-7, Hercules, CA, USA), Violet Red Bile Glucose Agr (VRBGA, Bio- Rd 3-), Rpid E. coli (Bio-Rd 3 99), nd in Slnetz nd Brtley medium (S&B, Oxoid CM377, Bsingstroke, Hmpshire, UK), respectively. The incubtion time/temperture conditions were 7 ± 3 h t 3 ± 1 C for totl erobic bcteri, ± h t 37 ± 1 C for Enterobctericee, 1 ± 3 h t ± 1 C for E. coli nd ± 3 h t 37 ± 1 C for Enterococcus spp. Another five cubicles, lso eqully distributed throughout the stble, were selected for the purpose of estimting the bedding wste over 1 dys. A grid (1.1 m width nd 1. m length) ws used to mesure the distnce between nine fixed points nd the bedding surfce (Figure 1). The grid ws plced on the cubicle prtition nd the distnce ws mesured on Dy (immeditely fter dding new bedding) nd Dy 1 (before bedding refreshment) using lser distnce meter (Leic Disto TM A, leic Geosystems AG, Heerbrugg, Switzerlnd). For ech of the five cubicles, the men depth of bedding ws clculted s the men depth mesured t the nine loctions. Mesurements on diry cttle. From the cows, cows were rndomly selected (%) nd mrked. On Dy (new bedding mteril),, 9, 1 (before refreshing the sme bedding mteril), 1 (Dy of the second period of 1 dys), 3 (Dy 9 of the second period of 1 dys) nd (Dy 1 of the second period of 1 dys), the cows were scored for the presence of skin lesions nd clenliness. Scoring ws done by four trined observers. Skin lesions were scored on -point scle (1 = no swelling, no lesions; = hirless ptches; 3 = swelling; = dmged skin in form of scb; = dmged skin in form of wound, dpted from Regul et l. ()) nd ssessed t plces on the cow s body: crpl joint, trsl joint (medil nd lterl), knee, tuber ischium nd tuber coxe. All loctions were scored for the left nd right side, nd the highest score of both sides ws registered s the finl score. Becuse of non-normlity, results re reported s medin (minimum mximum). For further sttisticl nlyses, the -point scle ws reduced to binry scle with = cows with score 1 nd 1 = cows with score > 1. Clenliness of the udder (both front nd bck side), lower legs nd flnk/upper legs were scored on -point scle: = no dirt;. = minor splshing, between % of the totl surfce covered with dirt; 1 = % of the totl surfce covered with dirt; 1. = between nd 7% of the surfce covered with dirt; = totl surfce covered with dirt. Similrly s for the scoring of skin lesions, left nd right legs nd both flnks were scored nd the highest score of both sides ws registered s the finl score. Becuse of non-normlity, results re reported s medin (minimum mximum). For further sttisticl nlyses, the -point scle ws reduced to binry scle: = cows with score or.; 1 = cows with score 1 for udder nd flnk. For the lower legs, the binry scle consisted of for cows with score < 1. nd 1 for cows with score 1.. Two indices of cow comfort. Cow comfort ws determined using two indices: the cow comfort index (CCI) nd the stll use index (SUI). To clculte these two indices, five types of behviour were registered s defined in Vn Gstelen et l. (11): (1) ctive in lley (i.e. cow in the lley nd engged in ctivity), () witing in the witing room to be milked by the robot or being milked, (3) stnding idle in the lley, () stnding with t lest one foot in cubicle, () lying in the cubicle. At three time points per dy (:3, 13: nd 17:) the number of cows expressing ech type of behviour ws counted. These time points were chosen bsed on roughge feeding time in the stble (9: nd 1:) nd 13

4 Originl Pper Veterinrni Medicin,, 1 (3): doi: 1.171/9-VETMED on literture dt reporting n verge number of cows lying down t these time points (Munksgrd et l. 11). CCI nd SUI were clculted s the proportion of cows touching cubicle tht re ctully lying down (CCI) nd s the proportion of cows not involved in feeding or milking tht re lying down (SUI) (Cook et l. ). Behviourl observtions were mde t Dy,, 9 nd 1 (just before refreshing the bedding mteril). Dust, temperture nd reltive humidity. To mesure dust, temperture nd reltive humidity, Grimm 1.19 erosol spectrometer (Grimm Aerosol Technik GmbH & Co. KG, Ainring, Germny) ws used. It ws mounted in wetherproof housing nd equipped with temperture nd reltive humidity sensor. This direct-reding portble PM monitor opertes on the principle of scttered light bems. It counts the number of prticles in 3 size frctions. The counts re converted into PM mss density. The instrument provided continuous dt output with mesurement intervls of one minute for ech of the 3 chnnels which mesure prticle sizes rnging from. μm to 3 μm. The spectrometer ws protected from the cows by being plced in n iron housing which obstructed the cows entry into the cubicle contining the spectrometer s well s the djcent cubicle. The spectrometer ws put in the first cubicle next to the selection fence. For further sttisticl nlyses, men dily PM 1, PM. nd PM 1 concentrtions nd men dily temperture nd reltive humidity were lso mesured. Sttisticl nlyses. Prior to sttisticl nlysis, ll dt were checked for unlikely vlues (% of the dust concentrtion mesured t minute level ws excluded for tht reson). To pproximte normlity, logrithmic trnsformtion of ll bcteril counts ws used. To determine the effect of type of bedding, dys fter dding fresh bedding (time) nd their interction (independent vribles), liner mixed regression models were performed with dy s rndom effect to correct for repeted mesurements within one cubicle (in cse of bcteril counts or depth mesurements s dependent vribles) or within one mesuring period (in cse of CCI, SUI, PM 1, PM., PM 1 s dependent vribles). The covrince between the repeted mesurements ws modelled using the AR (1) structure. Next, logistic regression models were built with the binry lesion or clenliness scores s dependent vribles. Agin, dy ws dded s rndom effect to correct Tble 1. Anlysis of the dry mtter content nd the fibre frction of chopped strw nd chopped for repeted mesurement from one cow. The fit of the finl models ws evluted by exmintion of the norml probbility plots of the residuls. All sttisticl nlyses were performed using SAS 9.3 (SAS Institute Inc, NC, USA). Sttisticl significnce ws ssumed when P <.. RESULTS Bse smples Chopped strw Chopped Miscnthus DM (%) NDF (% on DM bsis) ADF (% on DM bsis).. ADL (% on DM bsis). 11. DM = dry mtter, NDF = neutrl detergent fibre, ADF = cid detergent fibre, ADL = cid detergent lignin The nlysis of chopped strw nd chopped Miscnthus pointed to higher fibre concentrtion nd lower minerl content in Miscnthus thn in strw (Tble 1). Dry mtter content ws high nd comprble between the two bedding types (91.3% nd 91.9% for chopped strw nd chopped Miscnthus, respectively). Grinding Miscnthus to mke it usble s bedding resulted in lower DM content (9%). As expected, higher proportion of Dry mtter (%) b Dys fter refreshing bedding mteril Figure. Men dry mtter content ± stndrd devition (%) in five mrked cubicles filled with either combintion of strw/chlk/wter (blck), or combintion of Miscnthus/chlk/wter (grey), mesured on Dy (immeditely fter bedding ws refreshed) nd Dy 9 nd 1 fter bedding ws refreshed. Significnt differences in time re indicted with different letters 1

5 Veterinrni Medicin,, 1 (3): Originl Pper doi: 1.171/9-VETMED Tble. Prticle size distribution of chopped strw nd ground < 31. mm < 1 mm < mm < mm < mm < 1 mm Chopped strw (%) Ground Miscnthus (%) smll prticles were present in ground Miscnthus compred to chopped strw (Tble ). Mesurements of bedding DM content significntly incresed between dy (fresh bedding mteril) nd Dy 9 fter refreshing of the bedding mteril (Figure ). No significnt increse ws seen fterwrds for both types of bedding. Dry mtter content did not significntly differ between the two bedding tretments. Bcteril counts were generlly low t Dy nd significntly incresed with the time tht the bedding ws used, except for Enterobctericee (Figure 3). For Enterobctericee, Enterococcus spp. nd totl erobic bcteri, no significnt difference between strw nd Miscnthus ws found (P >.). A significnt interction time bedding type ws found for E. coli (P <.1). E. coli concentrtions in unused Miscnthus (Dy ) were significntly higher thn in unused strw (Dy ). The loss of bedding mteril over the course of 1 dys, mesured s the men depth between nine reference points nd the bedding surfce, ws comprble between the two bedding types (P >.) (Figure ). Mesurements on diry cttle Due to two cows drying off, only 1 cows remined in the cohort group to monitor skin lesions nd clenliness throughout the study. Medin (minimum mximum) skin lesions scores for strw nd Miscnthus, respectively, were 1 (1 3) nd Log 1 Enetrobctericee (cfu/g) (A) Dys fter refreshing bedding mteril Log 1 E. coli (cfu/g) (B) b c Dys fter refreshing bedding mteril Log 1 Enetrococcus (cfu/g) (C) b Dys fter refreshing bedding mteril Log 1 Totl erobic count (cfu/g) 1 1 (D) b b b b Dys fter refreshing bedding mteril Figure 3. Men bcteril counts ± stndrd devition (log1 cgf/g) in five mrked cubicles filled with either combintion of strw/chlk/wter (blck), or combintion of Miscnthus/chlk/wter (grey), mesured on Dy (immeditely fter bedding ws refreshed) nd Dy 9 nd 1 fter bedding ws refreshed. Significnt differences in time bedding (for E. coli figure 3B) or in time (for Enterobctericee, Enterococcus nd totl erobic bcteri) re indicted with different letters 1

6 Originl Pper Veterinrni Medicin,, 1 (3): doi: 1.171/9-VETMED Distnce between reference point nd bedding surfce (cm) Dy Dy 1 Figure. Men depth ± stndrd devition (cm) of the bedding of five mrked cubicles mesured s men perpendiculr distnce between nine reference points nd bedding surfce of either strw/chlk/wter (blck), or Miscnthus/chlk/wter (grey) 1 (1 3) for crpus, 1 (1 ) 1nd 1 (1 ) for medil trsus nd 1 (1 ) nd 1 (1 3) for lterl trsus, 1 (1 ) nd 1 (1 ) for knee, 1 (1 ) nd 1 (1 ) for tuber ischium, 1 (1 ) nd 1 (1 ) for tuber coxe. No significnt effects were found for time tretment (P >.), over time (P >.), nor between tretments (P >.) for trsl lesions, knee lesions, tuber ischium lesions nd tuber coxe lesions (Figure ). However, the risk of hving lesion with score > 1 t the crpl joint ws higher when cubicles were filled with strw compred to Miscnthus (odds rtio = 3., P =.). Medin (minimum mximum) clenliness scores for strw nd Miscnthus were, respectively, ( ) nd. ( 1.) for the front udder,. ( ) nd. ( 1.) for the bck udder,. ( ) nd. ( 1.) for the flnk/upper legs nd 1 ( ) nd ( ) for the lower legs. No significnt effects were found for time tretment (P >.), over time (P >.), nor between tretments (P >.) for clenliness of front udder, bck udder nd flnk/upper legs (Figure ). However, the odds of hving cow with > % of her lower legs covered with dirt ws.3 times higher when cubicles were filled with strw compred to Miscnthus (P =.). Two Indices of cow comfort CCI ws on verge ± % nd ± % nd SUI ws on verge 71 ± 1% nd ± 7% for strw nd Miscnthus, respectively. No significnt influence ws found between dys fter dding fresh bedding (P =.93 nd P =.37 for CCI nd SUI, respectively), Number of cows without crpus lesions 1 1 (A) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Number of cows without lesions t hock nd medil trsl joint 1 1 (B) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Number of cows without lesions t hock nd lterl trsl joint 1 1 (C) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Number of cows without lesions t knee 1 1 (D) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Number of cow with lesions t tuber ischium 1 1 (E) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Number of cows without lesions t tuber coxe 1 1 (F) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Figure. Number of cows without skin lesions t six different loctions from totl number of scored cows of. Cubicles were filled for dys with either combintion of strw/chlk/wter (grey), or combintion of Miscnthus/chlk/wter (white). On Dy 1 new bedding mteril ws dded 1

7 Veterinrni Medicin,, 1 (3): Originl Pper doi: 1.171/9-VETMED 1 (A) 1 (B) Number of cows with less thn % of the front udder covered with dirt Number of cows with less thn % of the bck udder covered with dirt Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Number of cows with less thn % of the flnk/upper legs covered with dirt (C) Number of cows with less thn % of the lower legs covered with dirt (D) Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Dy Dy Dy 9 Dy 1 Dy 1 Dy 3 Dy Figure. Number of cows in which less thn % of the totl surfce of the udder (front nd bck side) nd flnk/ upper legs ws covered with dirt (A, B nd C) or less thn % of the totl surfce of the lower legs ws covered with dirt (D). Cubicles were filled for dys with either combintion of strw/chlk/wter (grey), or combintion of Miscnthus/chlk/wter (white). On Dy 1 new bedding mteril ws dded nor between the two bedding tretments (P =. nd P =. for CCI nd SUI, respectively). Despite the reltively lrge vrition between mesurements (up to 1% for SUI), no effect of mesurement times (:3, 1:3 nd 17:) ws found (P =.9 nd P =.99 for CCI nd SUI, respectively). Dust, temperture nd reltive humidity PM 1 concentrtion (µg/m 3 ) Dys fter refreshing bedding mteril Figure 7. Men dily PM 1 concentrtion ± stndrd devition (µg/m³) in cubicles filled with either combintion of strw/chlk/wter (blck), or combintion of Miscnthus/chlk/wter (grey) Figure 7 illustrtes the men dily PM 1 concentrtions mesured during the 1 dys fter refreshing the bedding mteril. The PM 1 concentrtion ws significntly influenced by the interction time bedding type (β =.9; P =.1). When cubicles were filled with combintion of strw/ chlk/wter, the PM 1 concentrtions were higher. These concentrtions incresed with incresed use of the bedding mteril. In ddition, PM 1 concentrtions lso tended to be positively ssocited with the stble temperture (β =.91; P =.). Although they were not significnt, the PM 1 concentrtions were lso slightly higher on the dy the cubicles were filled with fresh bedding (β = 3.9; P =.1). In contrst to PM 1, no significnt effects of bedding type, dy fter bedding refreshing, or stble temperture were found for PM. or PM 1 (P >.). Over the two 1-dy periods, men PM. concentr- 17

8 Originl Pper Veterinrni Medicin,, 1 (3): doi: 1.171/9-VETMED tions were 1. ± 1. µg/m 3 nd. ± 3.33 µg/m 3 nd men PM 1 concentrtions were 1.7 ± 1.3 µg/m 3 nd 1. ±. µg/m 3 for strw nd Miscnthus, respectively. Both concentrtions were negtively ssocited with reltive humidity (β =.9; P =.1 nd β =.; P =. for PM. nd PM 1, respectively). DISCUSSION Anlyticl differences of DM content nd fibre frctions between strw nd Miscnthus were s expected, considering the plnt versus wood origin nd comprble with results found in the literture (Qin et l. 1; Ververis et l. ). Even fter being chopped, Miscnthus still contins high proportion of reltively thick nd woody stlks, which mkes chopped Miscnthus unsuitble for ppliction s bedding mteril. The lignin content of Miscnthus ws lmost twice s high s tht of strw. Grinding Miscnthus mkes it pproprite s bedding nd seems to decrese the DM content (91.9% chopped versus 7.7% ground Miscnthus). After grinding, Miscnthus ws stored outside covered with plstic for four weeks. Incresed wter content ws probbly cused by this storge under the following environmentl conditions: men temperture of 11.7 C nd men reltive humidity of 9%. Huismn nd Kortleve (199) reported n incresed wter content of Miscnthus due to storge under higher reltive humidity. The smller prticle size of ground Miscnthus results in n incresed surfce re. Enlrged wter holding cpcity s result of incresed surfce re my fvour bcteril growth (Rendos et l. 197). On the other hnd, s ll woody plnts, Miscnthus contins fewer free nutrients, such s simple sugrs, pentoses nd mino cids nd more cellulose nd lignin thn strw, ll of which might impede bcteril growth (Rendos et l. 197). No differences in bcteril content between the two combintions of bedding mterils were found in this study, except for E. coli on Dy. Higher E. coli concentrtions were found t the strt in Miscnthus compred to strw. However, these differences disppered the longer the bedding mteril ws used. Totl erobic bcteri nd E. coli counts during use (Dy 9 nd 1) were lower thn those found by Kudi et l. (9). In their study verge totl erobic bcteri nd E. coli on different bedding mterils (snd, swdust nd strw) for diry cttle were 1 11 cfu/g nd 1 cfu/g, respectively. The dily mnul removl of dirt nd feces from the cubicles in this study performed on well-mnged herd might hve prevented the high bcteril growth in both types of bedding reported by Kudi et l. (9) nd Dodd et l. (19). No difference ws found between the men depth between nine reference points nd the bedding surfce; thus, one cn conclude tht the loss of the two bedding mterils during 1-dy period ws similr. After 1 dys of use without dding extr bedding mteril, the mount of bedding ws drmticlly reduced. However, the remining mount ws still cceptble nd did not result in n increse in skin lesions t the crpus or the trsus, for exmple, nor in reduction in the clenliness of the cows over time. The likelihood of hving skin lesions ws generlly low in this study. Also, the number of cows with skin lesions nd the severity of the lesions (mximum skin lesion score) ws low, gin demonstrting tht this ws well-mnged herd. No differences were found between the two tretments, except for crpus lesions. Although it is questionble whether the durtion of the tril (four weeks) ws long enough to properly evlute the incidence of skin lesions, the probbility of crpus lesions ws higher when strw ws used insted of Miscnthus. Besides the mount of bedding mteril, which ws comprble between strw nd Miscnthus, lso the physicl properties of the surfce (softness, therml comfort, coefficient of friction, compressibility, etc.) cn lso contribute to the development of skin lesions t crpus nd hock (Tucker et l. 9; Tucker nd Wery ). Although none of these physicl chrcteristics were mesured during this tril, subjective observtions during dily rking of the cubicles reveled tht Miscnthus ws much looser in consistency. Together with urine, the strw/chlk/wter combintion tended to form clots tht did not degrde, wheres dry strw normlly degrdes during use. More effort ws therefore needed to loosen strw clots nd to eqully distribute the mteril over the cubicle by rking. The formtion of hrd clots might be responsible for the increse in crpus lesions. However, these subjective results need to be confirmed nd the prcticl relevnce of this smll difference in crpus lesions is debtble. The probbility of high clenliness score (mening soiled cow s indicted by more thn % 1

9 Veterinrni Medicin,, 1 (3): Originl Pper doi: 1.171/9-VETMED dirt for udder nd flnk nd more thn % dirt for lower legs) ws low in this study. No differences were found between the use of strw nd Miscnthus except for on the lower legs. Fifteen cows with more thn % dirt coverge on their lower legs were found in strw-filled cubicles just before new strw ws dded to the cubicle (Dy 1). The odds for dirty cow were 3.3 when strw ws used compred to Miscnthus. It should be noted, however, tht these higher odds were more likely to hve been cused by dirty wlking re s result of defective clening robot on Dy 13 nd 1 during the tril with strw thn by the type of bedding mteril. Significnt differences were no longer present if those two dys were excluded from the nlysis. The two indices of cow comfort were high: for strw nd Miscnthus, CCI ws on verge % nd %, respectively nd SUI ws on verge 71% nd %, respectively. A vlue greter thn % for CCI is proposed s relistic gol (Cook et l. ). Considerble vrition in both CCI nd SUI ws found between nd within dys, which indictes tht one mesurement my not reflect the verge of the indices over time, s previously noted by Cook et l. () nd vn Gstelen et l. (11). Nevertheless, no significnt effect of mesurement time (:3, 1:3 nd 17:) on CCI nd SUI ws found. CCI for both strw nd Miscnthus ws comprble or slightly higher thn results found in the literture for mttresses (7 7%), snd (%) nd box compost (1%), wheres the SUI ws comprble or slightly lower ( 7% for mttresses, 7 7% for snd nd % for box compost) (Cook et l. ; vn Gstelen et l. 11). In ddition, both indices did not significntly differ between strw nd Miscnthus. Although lying time nd durtion of lying ws not recorded in our study, sufficient nd comprble cow comfort when using both types of bedding mteril cn be ssumed. No significnt differences in CCI nd SUI were found in time, indicting n immedite cceptnce of Miscnthus by the cows without prior fmilirity t the strt of the experiment nd sufficient mount of soft bedding surfce remining fter 1 dys of use. Cow comfort is lso known to be influenced by the dryness of the bedding mteril. According to Fregonesi et l. (7) lying time is reduced by more thn 11 h dy when DM content decreses from.% to.%. More relevnt DM contents for prctice (9. 3.7%) were tested by Reich et l. (1), reveling more moderte decreses in lying time (mximum 1.1 h/dy). Moreover, it ws concluded tht the decrese in lying time ws only relevnt below % DM nd tht cows my not hve strong preference for drier bedding when options between nd 9% re vilble (Reich et l. 1). In this study, DM content t the strt ws 71.% nd 7.3% for strw nd Miscnthus combintions, respectively. A significnt increse towrds 9.% nd 93% ws seen within 1 dys. Although lying time ws not mesured, we did not expect either negtive influences of this high DM content or differences in lying time between the two types of bedding mteril cused by DM content. In generl, men dily PM 1 concentrtions were low ( 9 µg/m 3 ) nd comprble with dt found for diry brns ( 13 µg/m 3 (Dolejs et l. ); 37 µg/m 3 (Schmidt et l. ); 1 17 µg/m 3 (Joo et l. 13); µg/m³ (Tki et l. 199)). PM 1 concentrtions remined fr below the workplce ir qulity sfety limits of 3 nd 1 µg/m 3, s prescribed by the Belgin government (Royl Decree of 3 June 11). Both fresh bedding combintions were quite wet (DM content of the combintion 71.% nd 7.3% for strw nd Miscnthus, respectively). Wet bedding mteril cn cpture dust (Tki et l. 199). However, during the use of the bedding, the mteril becomes more nd more dry (DM content of 9.% nd 93.% for the combintion strw/chlk/wter nd Miscnthus/ chlk on Dy 1, respectively) due to ventiltion nd wter evportion. But PM 1 concentrtions only incresed towrds the end of the mesuring period when the combintion strw/chlk/wter ws used. No such increse ws seen in the two 1-dy periods when the cubicles were filled with Miscnthus. The degrdtion of strw during use nd formtion of smll prticles might be responsible for the incresed relese of dust prticles. The plnt versus wood origin of the bedding mteril might lso contribute to the greter dust concentrtion in strw-bsed versus Miscnthus-bsed bedding (Smdi et l. 1). Dust generted by bedding mterils my lso be influenced by niml ctivity or humn hndling (Tki et l. 199), explining the tendency of PM 1 to be higher on dy due to the clening of cubicles nd the refreshing of bedding mteril. Besides bedding, environmentl fctors such s reltive humidity nd temperture cn ffect dust 19

10 Originl Pper Veterinrni Medicin,, 1 (3): doi: 1.171/9-VETMED concentrtions. At high reltive humidity, dust prticles will contin condensed wter nd ggregte, which might explin the negtive reltionship between PM 1 nd PM. nd the reltive humidity (Tki et l. 199). In contrst, higher tempertures potentilly result in more dispersed nd drier prticles nd my therefore contribute to the increse in PM 1 concentrtions (Joo et l. 13). To our knowledge, this is the first study tht investigtes the potentil use of Miscnthus gignteus s bedding mteril in deep litter cubicles for diry cttle. As no significnt differences could be found in bcteril growth, the cpcity of the mteril to remin in the cubicles, skin lesions (except for crpus lesions) nd clenliness score of the cows, nd two cow indices for cow comfort (CCI nd SUI), it cn be concluded tht strw cn indeed be replced by Miscnthus. In ddition, dust concentrtions were significntly lower fter Miscnthus ws used for 1 dys s bedding. The dust concentrtions never exceeded workspce ir qulity sfety limits when either strw or Miscnthus were used. Consequently, diry frms my increse their self-sufficiency nd become less dependent on the vilbility nd price fluctutions of the trditionl bedding mterils by cultivting Miscnthus onfrm. Totl cultivtion costs of Miscnthus vry from 7 euro/t to 7 euro/t, depending on the length of the production period (1 to 1 yers), the verge yield (1 t/h to t/h), dry mtter percentge, hrvest method nd post-hrvest costs, herbicide ppliction nd locl production circumstnces (Bullrd 1999; Bullrd 1; DEFRA 1; Khnn et l.; Styles et l. ; Muylle nd Snuwert 1). These clculted costs from literture re much cheper thn the prices for strw (from 9 euro/t up to 1 euro/t). However, the extr costs of the grinding process of Miscnthus re not included in the totl production costs nd must be tken into ccount. Future studies might include totl cost/benefit clcultion. Acknowledgements The uthors would like to thnk Lurens Vn Dooren (KAHO Sint-Lieven) for the field work support, the technicl stff of ILVO for the nlysis nd Mirim Levenson for lnguge correction. Also, thnks to Peter Coucke for providing Miscnthus nd LV Den Hmer for the use of stbles nd cows. REFERENCES Breum NO, Nielsen BH, Lyngbye M, Midtgrd U (1999): Dustiness of chopped strw s ffected by lignosulfonte s dust suppressnt. Annles of Agriculture nd Environmentl Medicine, Bruijnis MR, Hogeveen H, Stssen EN (13): Mesures to improve diry cow foot helth: consequences for frmer income nd diry cow welfre. Animl 7, Bullrd M (1999): Miscnthus gronomy (for fuel nd industril uses). Technicl Report NF3. Accessed from: reserch.htm. Bullrd M (1): Economics of Miscnthus production. In: Jones MB, Wlsh M (eds.): Miscnthus for Energy nd Fibre, London Jmes nd Jmes (Science Publicers). Clmri L, Clegri F, Stefnini L (9): Effect of different free stll surfces on behviourl, productive nd metbolic prmeters in diry cows. Applied Animl Behviour Science 1, Cmbr-Lopez M, Arnink AJA, Zho Y, Clvet S, Torres AG (1): Airborne prticulte mtter from livestock production systems: A review of n ir pollution problem. Environmentl Pollution 1, Chplin SJ, Tierney G, Stockwell C, Loque DM, Kelly M (): An evlution of mttresses nd mts in two diry units. Applied Animl Behviour Science, 7. Cook NB, Nordlund KV (9): The influence of the environment on diry cow behvior, clw helth nd herd lmeness dynmics. Veterinry Journl 179, Cook NB, Bennett TB, Nordlund KV (): Monitoring indices of cow comfort in free-stll-housed diry herds. Journl of Diry Science, Dnltos NG, Archontoulis SV, Mitsios I (7): Potentil growth nd biomss production of Miscnthus gignteus s ffected by plnt density nd N-fertiliztion in centrl Greece. Biomss nd Bioenergy 31, 1 1. DEFRA (1): A review of the potentil of gint grsses for U.K. griculture. Reserch nd Develoment Finl Project Report NF19. Accessed from gov.uk/reserch/project_dt. Dodd FH, Higgs TM, Brmley AJ (19): Cubicle mngement nd coliform mstitis. Veterinry Record 11, 3. Dolejs J, Mst O, Toufr O (): Elimintion of dust production from stbles for diry cows. Czech Journl of Animl Science 1, Fisher AD, Verkerk GA, Morrow CJ, Mtthews LR (): The effects of feed restriction nd lying deprivtion on pituitry-drenl xis regultion in lcting cows. Livestock Science 73, 3. 13

11 Veterinrni Medicin,, 1 (3): Originl Pper doi: 1.171/9-VETMED Fregonesi JA, Veir DM, von Keyserlingk MAG, Wery DM (7): Effects of bedding qulity on lying behvior of diry cows. Journl of Diry Science 9, 7. Heo HS, Prk HJ, Yim J, Sohn JM, Prk J, Kim S, Ruy C, Jeon J, Prk Y (1): Influence of opertion vribles on pst pyrolysis of Miscnthus sinensis vr. purpurscens. Bioresource Technology 11, Hogn JS, Smith KL (1997): Bcteri counts in swdust bedding. Journl of Diry Science (), 1 1. Huismn W, Kortleve W (199): Mechniztion of crop estblishment, hrvest, nd post-hrvest conservtion of Miscnthus sinensis Gignteus. Industril Crops nd Production, Jezowski S (): Yield trits of six clones of Miscnthus in the first 3 yers following plnting in Polnd. Industril Crops nd Production 7,. Joo HS, Ndegw PM, Heber AJ, Ni JQ, Bogn BW, Rmirez- Dorronsoro JC, Cortus EL (13): Prticulte mtter dynmics in nturlly ventilted freestll diry brns. Atmospheric Environment 9, Khnn M, Dhungn B, Clifton-Brown J (): Costs of producing Miscnthus nd switchgrss for bioenergy in Ilinois. Biomss nd Bioenergy 3, 93. Kudi AC, Bry MP, Nib AT (9): Mstitis cusing problems within the diry cttle environment. Interntionl Journl of Biology 1, Metz JHM (19): The rection of cows to short-term deprivtion of lying. Applied Animl Behviour Science 13, Munksgrd L, Ingvrtsen KL, Pedersen LJ, Nielsen VK (1999): Deprivtion of lying down effects behvior nd pituitry-drenl xis responses in young bulls. Act Agriculture Scndinvic Section A-Animl Science 9, Munksgrd L, Rushen J, de Pssille AM, Krohn CC (11): Forced versus free trffic in n utomted milking system. Livestock Science 13,. Muylle H, Snuwert E (1): Energy crops in Flnders:. Het from Miscnthus (in Germn). Proceedings Open Energy, Mrch, 1, Flnders, Belgium. Qin J, Yng Y, Jing J, Yi Z, Xio L, Ai X, Chen Z (1): Comprison of lignocellolose composition in four mjor species of Miscnthus. Africn Journl of Biotechnology 11, Regul G, Dnuser J, Spycher B, Wechsler B (): Helth nd welfre of diry cows in different husbndry systems in Switzerlnd. Preventive Veterinry Medicine, 7. Reich LJ, Wery DM, Veir DM, von Keyserlingk MAG (1): Effects of swdust bedding dry mtter on lying behvior of diry cows: A dose-dependent response. Journl of Diry Science 93, Rendos JJ, Eberhrt RJ, Kesler EM (197): Microbil popultions of tet ends of diry cows, nd bedding mterils. Journl of Diry Science, Smdi S, vn Eerdenburg FJCM, Jmshidifrd AR, Otten GP, Droppert M, Heederik DJJ, Wouters IM (1): The influence of bedding mterils on bio-erosol exposure in diry brns. Journl of Exposure Science nd Environmentl Epidemiology, Smpimon OC, Sol J, Kock PA (): An outbrek of Klebsiell pneumonie mstitis (in Germn). Tijdschrift voor Diergeneeskunde 131,. Schmidt DR, Jcobsen LD, Jnni KA (1): Continuous monitoring of mmoni, hydrogen sulfide nd dust emissions from swine, diry nd poultry brns. ASAE Pper No., ASAE, St. Joseph, USA. Schulze WH, Gygx L, Myer C, Wechsler B (7): Leg lesions nd clenliness of finishing bulls kept in housing systems with different lying re surfces. Veterinry Journl 17, 77. Scurlock JMO (ed.) (1999): Miscnthus: review of Europen experience with novel energy crop. Ok Ridge Ntionl Lbortory, Ok Ridge Styles D, Thorne F, Jones MB (): Energy crops in Irelnd: An economic comprison of willow nd Miscnthus production with conventionl frming systems. Biomss nd Bioenergy 3, 7 1. Tki H, Pedersen S, Johnsen JO, Metz JHM, Groot Koerkmp PWG, Uenk GH, Phillips VR, Hoden MR, Sneth RW, Short JL, White RP, Hrtung J, Seedorf J, Schroder M, Linkert KH, Wthes CM (199): Concentrtions nd emissions of irborne dust in livestock buildings in northern Europe. Journl of Agriculturl Engineering Reserch 7, Tucker CB, Wery DM (): Bedding on geotextile mttresses: How much is needed to improve cow comfort? Journl of Diry Science 7, 9 9. Tucker CB, Wery DM, von Keyserlingk MAG, Beuchemin KA (9): Cow comfort in tie-stlls: Incresed depth of shvings or strw bedding increses lying time. Journl of Diry Science 9, 9. vn Gstelen S, Westerln B, Houwers DJ, vn Eerdenburg FJCM (11): A study on cow comfort nd risk for lmeness nd mstitis in reltion to different types of bedding mterils. Journl of Diry Science 9, 7. Vn Soest PJ, Robertson JB, Lewis BA (1991): Methods for dietry fiber, neutrl detergent fiber, nd nonstrch polyscchrides in reltion to niml nutrition. Journl of Diry Science 7, Verbist B, Piessens V, Vn Nuffel A, De Vuyst L, Heyndrickx M, Hermn L, Vn Coillie E, De Vliegher S (11): Sources other thn unused swdust cn introduce Klebsiell pneu- 131

12 Originl Pper Veterinrni Medicin,, 1 (3): doi: 1.171/9-VETMED monie into diry herds. Journl of Diry Science 9, Ververis C, Georghiou K, Christodoulkis N, Snts P, Snts R (): Fiber dimensions, lignin nd cellulose content of vrious plnt mterils nd their suitbility for pper production. Industril Crops nd Production 19,. Vokey FJ, Gurd CL, Erb HN, Glton DM (1): Effects of lley nd stll surfces on indices of clw nd leg helth in diry cttle housed in free-stll brn. Journl of Diry Science, 99. Wery DM, Tszkun I (): Hock lesions nd free-stll design. Journl of Diry Science 3, Wechsler B, Schub J, Friedli K, Huser R (): Behviour nd leg injuries in diry cows kept in cubicle systems with strw bedding or soft lying mts. Applied Animl Behviour Science 9, Zdnowicz M, Shelford JA, Tucker CB, Wery DM, von Keyserlingk MAG (): Bcteril popultions on tet ends of diry cows housed in free stlls nd bedded with either snd or swdust. Journl of Diry Science 7, Received: 1 9 Accepted fter corrections: 1 1 Corresponding Author: Stephnie Vn Weyenberg, Institute for Agriculturl nd Fisheries Reserch (ILVO), Technology nd Food Science Unit, Burg. Vn Gnsbergheln 11, 9 Merelbeke, Belgium E-mil: stephnie.vnweyenberg@ilvo.vlnderen.be 13