Strategies to control odours in livestock facilities: a critical review

In oduc ion Odou gene a ion is one o he mos ele an ai quali y issues o con ined li es ock ope a ions. Al hough odou exposu e has been adi ionally conside ed only a nuisance p oblem, i is now accep ed ha i can also impai heal h h ough di ec i i a ion o psychopa ho- logic mechanisms (Shus e man, 1999; Schi man & Williams, 2005). Fo his eason, in ecen yea s in en- si e esea ch has been conduc ed o assess and con ol odou s emi ed om li es ock acili ies. The e is also a endency o es ablish s ic e egula ions o imp o e ai quali y in he su oundings o li es ock acili ies (Nicell, 2009). Odou nuisances, howe e , cons i u e a conce n o e y complex na u e, in compa ison wi h o he ai qua- li y issues. Assessing odou s is condi ioned by he subjec i e cha ac e o nuisances and he in insic a- iabili y o odou emission and dispe sion p ocesses. Thus, measu ing odou s, de eloping odou dispe sion models and e alua ing aba emen s a egies ha e been adi ionally challenging asks o a me s and esea - che s. Much esea ch has been conduc ed so a in ela- ion o odou measu emen s and he applica ion o odou dispe sion models (Yu e al., 2010). Also, a wide num- be o odou aba emen echniques ha e been cha ac e- ized (e.g. Powe s, 1999). Howe e , ou cu en know- ledge seems o be s ill limi ed o de ini i ely sol e some odou p oblems o li es ock acili ies. The objec i e o his s udy is o c i ically e iew he s a e o knowledge on odou mi iga ion echniques in li es ock housing. Fi s ly, he complex na u e o li- es ock odou s is b ie ly e iewed. Then, we analyze he mos ele an odou con ol echnologies, hei S a egies o con ol odou s in li es ock acili ies: a c i ical e iew Y. Ubeda1, P. A. Lopez-Jimenez2, J. Nicolas3and S. Cal e 1* 1 Ins i u e o Animal Science and Technology. Uni e si a Poli ècnica de Valencia. Camino de Ve a, s/n. 46022 Valencia, Spain 2 Depa men o Hyd aulic and En i onmen al Enginee ing. Uni e si a Poli ècnica de Valencia. Camino de Ve a, s/n. 46022 Valencia, Spain 3 En i onmen al Moni o ing. Uni e si y o Liège. A da. Longwy, 185. B 6700 A lon, Belgium Abs ac Odou s gene a ed in li es ock buildings cons i u e one o he mos ele an ai quali y issues o in ensi e li es ock p oduc ion. Reducing nuisance episodes ela ed o odou exposu e is he e o e essen ial o a sus ainable li es ock p oduc ion. In his s udy, he s a e-o - he-a on odou mi iga ion echniques in li es ock housing is c i ically e iewed. Scien i ic ad ances in he las decade a e e ised and esea ch needs a e also iden i ied. The complex na u e o li es ock odou s is i s ly e iewed and examined. Then, he mos ele an odou con ol s a egies a e analyzed in e ms o p esen knowledge and u u e needs. The s a egies conside ed a e: nu i ional s a egies, manu e addi i es, building design, ai il a ion, manu e co e s, manu e ea men sys ems and windb eaks. Finally, u u e esea ch needs and p io i ies when es ablishing mi iga ion echniques a e iden i ied. Despi e impo an ecen ad ances, he e a e s ill some challenges o scien is s, p oduce s and egula o s, pa icula ly ela ed o ield e alua ion o odou s. The e o e, o con ol li es ock odou s e ec i ely, using s anda dized ield assessmen echniques will be equi ed. Also, in es iga ing measu emen and model e o s may be use ul o be e unde s and he limi a ions o he cu en me hods, as well as o iden i y esea ch p io i ies. Addi ional key wo ds: odou emission; odou dispe sion; odou mi iga ion; nuisance; ol ac ome y. * Co esponding au ho : [email p o ec ed].es Recei ed: 11-03-13. Accep ed: 08-11-13. This wo k has one Supplemen a y Table ha does no appea in he p in ed a icle bu ha accompanies he pape online. Abb e ia ions used: CP (c ude p o ein); EF (emission ac o ); FC ( e men able ca bohyd a e); FIDOL ( equency-in ensi y-du a- ion-o ensi eness-loca ion); VFA ( ola ile a y acid). Ins i u o Nacional de In es igación y Tecnología Ag a ia y Alimen a ia (INIA) Spanish Jou nal o Ag icul u al Resea ch 2013 11(4): 1004-1015 A ailable online a www.inia.es/sja ISSN: 1695-971-X h p://dx.doi.o g/10.5424/sja /2013114-4180 eISSN: 2171-9292 s eng hs and weakness and hei e iciency on odou po en ial educ ion. Finally, u u e esea ch needs and p io i ies a e p oposed o es ablish e ec i e mi iga ion s a egies o li es ock odou s. A complex nuisance p oblem The complex na u e o li es ock odou s An odou can be de ined as a human sensa ion ha occu s when ai bo ne chemical subs ances, called odo an s, s imula e senso y ecep o s in he nasal ca i y (Schi man e al., 2001). Li es ock odou s a e composed o a complex mix u e o compounds, which a ies be ween animal ypes. As an example, app oxi- ma ely 330 di e en odo ous compounds ha e been iden i ied in swine p oduc ion (Schi man e al., 2001), whe eas 110 compounds we e ound in dai y acili ies (Filipy e al., 2006). Li es ock odou s a e mainly gene a ed by he mi- c obial decomposi ion o o ganic ma e con ained in he diges i e ac o animals and in hei manu e, unde anae obic condi ions. Al hough odou s can also be o igina ed om o he sou ces (e.g. om he skin o he eed), s udies conduc ed in his sense sugges ha animals hemsel es cons i u e a mino sou ce o odou i compa ed wi h manu e (Ve does & Ogink, 1997; Kai e al., 2006). Impo an sou ces o li es ock odou s a e ela ed o animal houses, manu e s o age acili ies and ag icul u al applica ion o manu e, bu he ela i e odou load o each sou ce is no well de e mined so a . Odou o ma ion, concen a ion and emissions om li es ock acili ies depend on many ac o s, which a e summa ized in Fig. 1. Odou gene a ion is in insically de- e mined by he animals hemsel es, hei ood and ma- nagemen sys em o hei manu e. I may also be a ec- ed by daily and seasonal a ia ions, depending on cli- ma e and animal ac i i y. Finally, he amoun o odou emi ed o he a mosphe e is de e mined by he housing en ila ion a e and he exhaus odou concen a ion. A e iew o odou emission ac o s is shown in Suppl. Table 1 [pd online]. Al hough a majo pa o s udies on odou emissions ha e ocused on he measu- emen o odou emission a es, uni s used in he li e a- u e a e no always homogeneous. This he e ogenei y and he complexi y o odou measu emen s con ibu e o make compa isons among s udies mo e di icul (Gay e al., 2003). Odou nuisances The p ocesses by which odou s a e o med, eleased o ai , dispe sed in he a mosphe e, pe cei ed by indi iduals and cause nuisance in a popula ion a e complex no ully unde s ood. Van Ha e eld (2001) de ined some essen ial wo ds o a sound scien i ic S a egies o con ol odou s in li es ock acili ies: a c i ical e iew 1005 Figu e 1. O e iew o sou ces and ele an ac o s a ec ing odou concen a ion and emission om li es ock houses. VFA: ola- ile a y acid. Sou ce: Adap ed om Le e al. (2005). Pa icula e ma e Tempe a u e Humidi y Ai eloci y Ai exchange a e Feed and wa e in ake Feed composi ion: – P o ein con en – Aminoacid composi ion – Non-s a ch polysaccha ides Species Age Numbe Geno ype Heal h Beha iou Remo al equency Emi ing a ea Manu e collec ion sys em Physicochemical composi ion Bedding ma e ial Ven ila ion ai low Odou concen a ion Odou composi ion: VFA, indoles, phenoles, ska oles, sulphu -con aining compounds Feed Animal Manu e Odou o ma ion Hyd olisis and e men a ion unde anae obic condi ions En i onmen al condi ions Emission discussion. So, “annoyance” can be de ined as a se o complex human eac ions esul ing om an immedia e exposu e o an ambien s esso (e.g. an odou ). On he con a y, “nuisance” is caused by epea ed e en s o annoyance o e an ex ended pe iod o ime, which leads o modi ied o al e ed beha iou and can ha e a de imen al e ec on well-being and heal h. The e o e, he annoyance po en ial is an a ibu e o a speci- ic odou o mix u e o odo an s, whe eas nuisance po- en ial is a much b oade concep ha desc ibes he p o- babili y o nuisance occu ence in a pa icula loca ion. Human esponses o odou s, howe e , a e also highly a iable, and mus be unde s ood as a combina ion o i e in e ela ed componen s, known as a whole as FIDOL (Nicell, 2009). The FIDOL componen s e e o he F equency (how o en), In ensi y (how s ong), Du a ion (how long), O ensi eness (how unpleasan ), and he Loca ion (sensi i i y o neighbou s wi h ega ds o li es ock ope a ion). Since hese compo- nen s a e di icul o moni o , some coun ies (e.g. Ge many, The Ne he lands, Swi ze land, Aus ia, Bel- gium and Canada) ha e de eloped land use planning guidelines based on minimum sepa a ion dis ances be ween li es ock uni s and esiden ial a eas. This se back dis ance is calcula ed conside ing ou main ac o s, acco ding o he ollowing gene al equa ion (Nicolas e al., 2008): [1] whe e he se back dis ance (D) is calcula ed as a unc- ion o he ollowing ac o s: he dispe sion condi ions ( D), ecep o cha ac e is ics ( R), animal species ( A) and echnical ac o s such as aba emen s a egies ( T). The equa ion also conside s he a m size (N) and a i ing ac o o eal a m condi ions (α). Odou measu emen Odou measu emen echniques a ise om he un- damen al p emise ha odou s mus i s be measu ed objec i ely and ep oducibly measu ed be o e hey can be e ec i ely subjec ed o egula ion and be o e he e ec i eness o odou con ol echnologies can be assessed (Nicell, 2009). Despi e ecen ad ances, ob- aining p ecise odou measu emen s emains an elusi e a ge o esea che s and egula o s. The only s anda dized app oach which is in e na- ionally accep ed in odou measu emen is he dynamic ol ac ome y (CEN, 2003). This echnique de e mines odou concen a ions using he dilu ion- o- h eshold p inciple, bu s ill poses some challenges o measu- emen s in ield condi ions (Smi h e al., 2007). Odou concen a ions a e di ec ly ela ed o odou in ensi y, bu his ela ionship is no s aigh o wa d because in ensi y e e s o an indi idual’s pe cep ion o odou s eng h (Nicell, 2009). Mo e speci ic in o ma ion on he FIDOL componen s ela ed wi h odou nuisances can be ob ained only wi h al e na i e me hods, which a p esen a e no s anda dized. I would be he e o e con enien o no malize ield assessmen echniques a an in e na ional le el. In his sense, he Sni ing Team Me hod o plume measu emen s such as hose es ablished by he Ge man VDI 3940 could be mo e ealis ic o e alua e he impac o li es ock odou sou ces, which a e by na u e a iable and discon i- nuous in ime (Nicolas e al., 2008). Ob aining objec i e odou indica o s seems c ucial o iden i y e ec i e mi iga ion s a egies. To his aim, he ela ionship be ween odou pe cep ion and che- mical compounds has been subjec o in ense esea ch. This is a complex ask because analy ical echniques may ail o cha ac e ize he syne gic e ec s among odo an s. In addi ion, pa icula e ma e could play an impo an ole in odou pe cep ion (Bo che , 2001; Nicell, 2009). Con a ily, some gases such as NH3and SH2a e odou cons i uen s, bu hey canno be easily used as odou indica o s, since hei odou de ec ion is ela i ely high (Smee s e al., 2007; Blanes-Vidal e al., 2009a) and hei emission p ocess may di e om he elease pa hways o mos odo an s (Le e al., 2005; Blanes-Vidal e al., 2009a). Recen s udies ha e achie ed high ela ionships (R2~ 90%) be ween ol ac ome ic measu es and analy ic analyses using gas ch oma o- g aphy – mass spec ome y (Blanes-Vidal e al., 2009a; Zhang e al., 2010). Howe e , odo an s may di e conside ably among li es ock species, and he e- o e using a gene alized analy ical measu emen me hod o li es ock odou s seems a he momen no easible. Odou dispe sion modelling Fig. 2 summa izes he main ac o s which de e mine odou dispe sion om li es ock acili ies. Wea he condi ions (mainly wind speed and di ec ion, and a mosphe ic s abili y) cons i u e he main ac o de e - mining odou dispe sion. Fu he mo e, o he ac o s such as ege a ion and opog aphy also in luence in a ele an way. D=α D RN A T 1006 Y. Ubeda e al. / Span J Ag ic Res (2013) 11(4): 1004-1015 Acco ding o Smi h (1993), he dispe sion o li e- s ock odou s di e s om o he a mosphe ic pollu an s because o impo an pa icula i ies o odou sou ces, he na u e o odou s and he ecep o cha ac e is ics. In con as o indus ial sou ces, li es ock odou sou - ces usually co e a la ge a ea a o nea g ound le el and he plume ise is no mally i ele an . Fu he mo e, in li es ock buildings odou s a e emi ed om ope- nings o ans in walls and oo s, o ming complica ed s uc u es which a e di icul o model (Yu e al., 2010). Model alida ion is also complica ed because o he challenges o odou measu emen s, as well as he spa- ial and empo al a iabili y in emission a es. Finally, ecep o s may be ela i ely close o he sou ce o emissions, and in his case con en ional dispe sion models a e no eliable. Fo all hese easons, speci ic odou dispe sion models a e necessa y. In a ecen e iew on li es ock odou dispe sion modelling, he majo esea ch gaps o esea ch we e examined (Yu e al., 2010). Apa om he pa icula- i ies o odou dispe sion indica ed abo e, hese au ho s ques ioned he adi ional use o Pasquill- Gi o d coe icien s in he Gaussian dispe sion models. A mosphe ic s abili y has a la ge in luence on he dispe sion esul s, bu i seems oo simplis ic o desc ibe i using he Pasquill’s s abili y classi ica ion scheme. Finally, Yu e al. (2010) iden i ied ha he main d awback o mos models is he long- ime in eg a ion pe iods, no mally om 10 o 60 minu es. Using hese ime in eg a ion pe iods, sho - ime concen a ion luc ua ions a e usually igno ed by dispe sion models. Howe e , sho - ime luc ua ions may in luence conside ably he nuisance po en ial (Nicell, 2009) and he e o e should be conside ed when p edic ing he odou plume (Sykes & Gab uk, 1997; Mussio e al., 2001). In ecen yea s, ad ances in compu a ion ha e allowed he de elopmen and use o nume ical dispe sion mo- dels (Lin e al., 2009; Hong e al., 2011). Howe e , a majo challenge o odou modelling emains o be he S a egies o con ol odou s in li es ock acili ies: a c i ical e iew 1007 Figu e 2. Elemen s a ec ing odou dispe sion p ocess in li es ock con ining ope a ions. The odou plume unde h ee hypo he i- cal si ua ions (A, B and C) illus a es he a iabili y o odou exposu e as a unc ion o wea he condi ions, opog aphy, ege a ion and loca ion. Wea he condi ions Dis ance Si ua ion A Si ua ion B Si ua ion C O ien a ion F equency in ensi y du a ion Loca ion Topog aphy Vege a ion co e p ope alida ion be ween modelled da a and eal nuisance pe cei ed by he popula ion. Aba emen and con ol s a egies in li es ock housing O e iew o odou aba emen s a egies A wide a ie y o odou p e en ion and con ol s a- egies ha e been s udied in ecen yea s. Fo e e y odou con ol s a egy, i is necessa y o conside he cos s and he desi ed mi iga ion le el, as well as he applica- bili y o he o e all a m managemen (Powe s, 1999; Ullman e al., 2004). Also, i mus be accep ed ha only a pa ial educ ion o odou s can be achie ed (Nahm, 2003). Th ee main ways a e possible o minimize odou impac om li es ock houses: educing odou o ma- ion, using end o pipe echniques and enhancing dis- pe sion. Odou o ma ion can be educed by means o die a y manipula ion, addi i es, low-emission li es ock housing and manu e managemen sys ems. Howe e , he e ec i eness o hese s a egies may di e o di e en animal species. End-o -pipe echniques, such as ai sc ubbing, and manu e co e s ha e p o ed o be e y e ec i e o educe odou emission. I is also possible o p omo e he dispe sion o odou s in he a m su oundings o educe he nuisance po en ial a sensi i e a eas. Die a y manipula ion Despi e being ela i ely inexpensi e, die a y s a egies in li es ock p oduc ion may achie e impo - an en i onmen al bene i s. I has been demons a ed ha educing c ude p o ein (CP) can e ec i ely educe he p opo ion o exc e ed ni ogen, which is associa ed o lowe ammonia emissions, no only a he housing le el, bu also du ing manu e s o age and land appli- ca ion o manu e (Hayes e al., 2004; Le e al., 2007). Howe e , he cause and e ec ela ionship be ween eed cha ac e is ics and he exc e ion o odo ous com- pounds is no so e iden . Thus, he e is no ag eemen abou he e ec o low-p o ein die s on odou compound p o ile. Some s udies such as hose conduc ed by Hobbs e al. (1996) and Ke e al. (2006), showed a dec ease o odo an concen a ions in g owing and inishing pigs when using low-CP die s wi h amino acid supple- men a ion. Feeding pigs wi h low c ude p o ein die s educed odou concen a ion om 30 o 80%, acco ding o Hayes e al. (2004) and Le e al. (2007). On he o he hand, some au ho s epo ed ha odou concen a ion and o ensi eness inc eased when low-CP die s supple- men ed wi h syn he ic amino acids we e used (O o e al., 2003; O’Connell e al., 2006). Finally, o he s udies could no ind di e ences in odou composi ion and magni ude when low-CP die s we e used (Su on e al., 1999; Cla k e al., 2005; Leek e al., 2007). Some s udies ha e e idenced he in luence o he en e ic mic obial communi y on he exc e ion o odo ous subs ances. O’Shea e al. (2011) ound ha chi osan inclusion in die s inhibi ed lac obacilli popula ions and his was associa ed o inc eased En e obac e- iaceae popula ions and odou emissions om manu e. This sugges s ha lac ic-acid bac e ia may play an impo an ole in mi iga ing manu e odou emissions. Mc Alpine e al. (2012) ound ha he inclusion o xilanase (an enzyme which deg ade be a-xylan in o xylose) e ec i ely educed manu e odou emissions. Howe e , he inclusion o a p o ease did no in luence odou emissions. Also, g ow h-p omo ing subs ances help o educe odou emissions om pig slu y in a 53- 56% (Nahm, 2002). Howe e , as indica ed by McC o y & Hobbs (2001) diges i e addi i es do no cons i u e an al e na i e un il a ho ough unde s anding o mi- c obial p ocesses in li es ock was es is achie ed and he e o e mo e esea ch is s ill needed on he e ec s and modes o ac ion o speci ic mic oo ganisms and enzymes. The use o e men able ca bohyd a es (FC) is e ec- i e o educe ammonia ola iliza ion, bu hey may also a ec he odou emission p o ile (Le e al., 2005). The inc ease o eed FC s imula es mic obial ac i i y in animal gu and in he manu e, shi ing he ni ogen balance om u ine o aeces (Su on e al., 1999; Cla k e al., 2005). I has been widely demons a ed ha FC inco po a ed in o pig die s inc eases he ola ile a y acid (VFA) con en in manu e (Cla k e al., 2005; Ke e al., 2006; O’Connell e al., 2006), bu i emains unclea whe he he odou p o ile is a ec ed. Whe eas Lynch e al. (2008) ob ained a 41% inc ease o odou emissions when suga bee pulp was added o pig die s, a non-signi ican e ec o FC was ound by Cla k e al. (2005) and Le e al. (2007). As acknowledged by Leek e al. (2007) and Le e al. (2008), he balance be ween die a y FC and CP plays an impo an ole on odou p oduc ion and emission. Howe e , his ela- ionship mus be u he explo ed. 1008 Y. Ubeda e al. / Span J Ag ic Res (2013) 11(4): 1004-1015 Manu e addi i es Manu e addi i es a e po en ially e ec i e ools o educe a mosphe ic emissions om li es ock p oduc- ion, pa icula ly odou s and ammonia. To be use ul, hese addi i es mus be sa e in he en i onmen , in- expensi e and easy o apply (Va el, 2002). These p ope ies a e app ecia ed by manu ac u e s and a - me s, bu in some cases he comme cial con iden iali y may in ol e a lack o s anda d, independen es s. McC o y & Hobbs (2001) e iewed he use o manu e addi i es ega ding ammonia and odou educ ion. They concluded ha only sho - e m odou con ol can be achie ed using masking, disin ec ing, and oxidizing agen s. F equen eapplica ion o addi i es is he e o e equi ed o main ain an accep able odou educ ion. Wheele e al. (2011) classi ied manu e amend- men s in o i e ca ego ies: mic obial, chemical, disin- ec an , masking and adso ben s. As indica ed by Mackie e al. (1998), li es ock was es cons i u e a dynamic en i onmen in which ae obic and anae obic mic obial eac ions occu . As a esul o hese eac- ions, VFAs a e o med, mos o which a e o ensi e odo an s. The e o e, manu e addi i es co esponding on he i s h ee ca ego ies abo e men ioned ocus on inhibi ing some o hese mic obial p ocesses. Acco - ding o Va el (2002) an i-mic obial plan -de i ed oils may play a ole in inhibi ing odou s. Howe e , Amon e al. (1997) ound no signi ican e ec o using a manu e addi i e based on an ex ac o Yucca shidige a, and simila esul s we e ob ained by Wheele e al. (2011) o o he addi i es. Also, masking agen s (essen ial oils) and adso ben s (zeoli es) ha e been epo ed o ha e limi ed odou educ ion po en ial (Amon e al., 1997; Wheele e al., 2011). Animal housing Animal housing and managemen sys ems in luen- ce he emission o odou s. Howe e , in con as o he cu en ly well de ined “low-ammonia emission” housing sys ems, i is di icul o dis inguish “low-odou emission” housing sys ems. The eason o his is he la ge a iabili y wi hin housing sys ems (Mol & Ogink, 2004). Gene ally, low-ammonia emission housing sys ems also educe odou o ma ion, bu his ela ion is no always s aigh o wa d. In pig p oduc ion, educing he su ace con ac be ween slu y and he ai has been demons a ed o be e ec i e o educe no only ammonia, bu also odou emissions by up o 50% (Ogink & Koe kamp, 2001). These sys ems include pa ial sla ed loo s, iangula - shaped gu e s and equen slu y emo al by lushing. I is gene ally accep ed ha odo ous compounds a e mainly p oduced when animal was es a e subjec ed o anae obic condi ions (Mackie e al., 1998). The e o e, educing mois u e con en o manu e is e ec i e o e- duce odou p oduc ion. This can be achie ed by using bedding ma e ials wi h a p ope managemen (O’Neill & Phillips, 1991). Con a ily o ammonia emissions, speci ic esea ch ega ding he e ec o bedding ma e- ial on odou gene a ion is sca ce, and con adic o y esul s ha e been ob ained o pig and cow p oduc ion (Ngwabie e al., 2010; Wang e al., 2011). Fo laying hens hose sys ems including equen emo al, d ying o mixing up wi h li e in ol ed ele an odou e- duc ions wi h espec o he con en ional, long e m manu e s o age (Ogink & Koe kamp, 2001). Finally, in b oile s, a p ope egula ion o d inke s and e ige- a ion ogge s is e ec i e o con ol li e mois u e. Es ablishing p ope en ila ion lows can also con i- bu e by elimina ing he excess o wa e and educe odou gene a ion (Ullman e al., 2004). Ai cleaning sys ems Ai cleaning sys ems such as bio il e s, sc ubbe s and bio ickling il e s ha e been de eloped in he las decades o emo e ai bo ne pollu an s om li es ock acili ies. In hese sys ems he exhaus ai is o ced o pass h ough one o mo e se s o we packing ma e ial in which ce ain pollu an s a e e ained and elimina ed. Despi e hei high in es men and unc ion cos s, hey a e e y e ec i e o emo e pollu an s and hei cha- ac e iza ion is ela i ely easy in compa ison wi h o he odou mi iga ion echniques. Howe e , hese sys ems canno be applied in na u ally en ila ed buildings. The exhaus ai o con ined li es ock buildings has no mally high concen a ions o ammonia, which is ola ilized om he b eakdown o u ea and o he ni- ogen compounds o u ine and aeces. To emo e he ammonia om he ai , chemical sc ubbe s ha e been designed, which use acid solu ion. Melse & Ogink (2005) epo ed an a e age ammonia emo al e icien- cy o 96%, and acco ding o Ogink & Aa nink (2007) hey can also emo e pa icula e ma e by up o 90%. Howe e , hese sys ems a e ela i ely ine icien o emo e odou s ( ypically abou 30%) because only S a egies o con ol odou s in li es ock acili ies: a c i ical e iew 1009 soluble compounds a e emo ed, which does no inclu- de many odo an s (Melse & Ogink, 2005). Recen s udies ha e sugges ed ha he addi ion o an o ganic sol en o he wa e phase could inc ease he a aila- bili y o odou componen s o he bac e ial popula ion, enhancing he biodeg ada ion o odou (Melse e al., 2009). Bio ickling il e s, despi e he lowe ammonia e iciency, a e on a e age mo e e ec i e o educe odou s han chemical sc ubbe s (Melse e al., 2009). Bio il a ion is he mos e ec i e end-o -pipe ech- nique o educe odou s. When he ai passes h ough a we o ganic packing ma e ial, odou s a e i s e ained in he ma e ial su ace and hen decomposed by mi- c oo ganisms (Re ah & Mo gan-Sagas ume, 2005). The epo ed e iciency lies wi hin he ange om 80% o 99%, depending on he packing ma e ial, i s mois u- e con en and he emp y bed e en ion ime (Chen & Ho , 2009). To ob ain maximum odou and ammonia educ ions, hese sys ems should be ins alled a e a chemical sc ubbe , hus emo ing NH3and odou s in wo sepa a e s ages (KTBL, 2006). Manu e s o age and ea men Manu e s o age and ea men is conside ed a signi- ican sou ce o odou emissions in li es ock acili ies. Howe e , he ele ance o his sou ce in compa ison wi h li es ock houses and land applica ion is no well quan i ied in li e a u e. Among he a ailable echniques o odou con ol in manu e s o age acili ies, he use o co e s o educe gaseous emissions has been widely s udied in ecen yea s (Vande Zaag e al., 2008). These au ho s classi y co e s acco ding o hei o igin. Co e s o na u al o i- gin include na u ally occu ing c us s, s aw and o he c op esidues, woodchips and sawdus , expanded clay, pe li e, ege able oils and ae a ion oam. Syn he ic co- e s include pe meable co e s (plas ic g anules, ubbe g anules, hyd ophobic powde ) and impe meable co e s such as plas ic ilms. Finally, composi e co e s combine he bes aspec s di e en ma e ials. The use o co e s is a ac i e because o hei low cos compa- ed o o he aba emen s a egies. Acco ding o he e ision conduc ed by Vande Zaag e al. (2008), a e age odou educ ion o di e en co e s anges be ween 40% and 90%. Se e al s udies (Hudson e al., 2008; Blanes-Vidal e al., 2009b) suppo he idea ha he main mechanism o odou educ ion is ha co e s ac as a physical ba ie obs uc ing he ee exchange o ola ile compounds om he unde lying liquid o he a mosphe e. Mo e- o e , hese s udies also indica e ha na u al co e s may also educe odou s as a esul o a bio il a ion ac ion o he co e i sel . Howe e , u he esea ch seems necessa y o o e - come wo main gaps o knowledge on he use o co e s. On he one hand, he c oss e ec o di e en co e ma e ials on g eenhouse gas emissions should be be e cha ac e ized. On he o he hand, he o e all pe o - mance o co e s should be es ed unde a wide a ie y o condi ions de i ed om di e en manu e ypes, clima e condi ions o co e ageing. The use o co e s may be o highes in e es in hose egions whe e la ge lagoons a e used (i.e. in ensi e pig a ming in Sou he n Eu ope and he USA). Main enance o hese co e s may be a c i ical aspec o hese co e s, acco ding o he es s made by Aguila e al. (2012) o pe meable and impe meable co e s. O he manu e ea men s, despi e no being speci i- cally designed o aba e odou s, may lead o odou emission educ ion i hey al e he anae obic con- di ions o elimina e he p ecu so s o odo ous com- pounds. Solid-liquid sepa a ion can educe odou emission, bo h as a s and-alone echnique (Zhang & Wes e man, 1997) o combined wi h anae obic diges- ion (Hansen e al., 2006). Howe e , i can only be conside ed a mi iga ion echnique i he sepa a ion p ocess can emo e he ines ac ion o solids (pa i- cles smalle han 75 µm) o which VFAs a e a ached (Ndegwa e al., 2002). The ae obic ea men oxidizes manu e o ganic ma e ials in o s able p oduc s by ae obic bac e ia (Wes e man & Zhang, 1997). This p ocess s abilizes he manu e and does no allow he accumula ion o VFAs and o he odo an s. When he was e is s abilized, i can be s o ed o a long pe iod wi hou emi ing ele an odou s. In his echnique, he main decision is o balance he need o a minimum ae a ion ime wi h he inc easing cos o long ae a ion imes (Wes e man & Zhang, 1997; Zhang e al., 2004). In con as o he ae obic ea men , anae obic diges- ion p omo es he p oduc ion o VFA as a subs a e o me hane and ca bon dioxide p oduc ion. Because he deg ada ion o VFAs is enhanced du ing his p ocess, he odou elease po en ial o he diges a e du ing subsequen s o age and land applica ion is lowe han he undiges ed ma e ials (Hansen e al., 2006). Consi- de ing he low odou emission po en ial o diges ed ma e ials, he managemen o undiges ed ma e ials 1010 Y. Ubeda e al. / Span J Ag ic Res (2013) 11(4): 1004-1015 (slu ies and co-subs a es) becomes c ucial o comple e an e ec i e odou mi iga ion s a egy in he a m. Windb eaks I has long been demons a ed ha windb eaks al e he ai u bulence (Segine , 1975), bu hei e ec i- eness o enhance odou dispe sion and educe nuisan- ces has no been s udied un il he las decade. Wind- b eaks and shel e bel s a ound li es ock uni s a e no mally ege a ion sys ems ha edi ec wind and educe wind speed. This e ec on wind modi ies en i onmen al condi ions upwind and downwind shel- e ed zones, and hus changes he dispe sion o odou s (Tyndall & Colle i, 2007). Windb eaks no only enhance dispe sion by p omo ing ai u bulence, hey also se e as a il a ion ba ie whe e pa icula e ma e and odou s a e pa ly e ained. A wind-b eak posi io- ned nea an odou sou ce can educe he downwind leng h o he odou dispe sion plume, as e idenced by ield measu emen s and model simula ions (Lin e al., 2006). They a e also applicable a easonable cos s (Tyndall & G ala, 2009). Quan i ying he odou mi iga ion po en ial o wind- b eaks is ex emely complica ed, and depends on shel- e bel cha ac e is ics and wea he condi ions. This has been s udied using nume ical models (Lin e al., 2007; 2009) o scaled wind unnel es s (Ikeguchi e al., 2003). Odou s a e be e dispe sed using dense and high windb eaks, loca ed nea he odou emi ing sou ce (Lin e al., 2007). The po en ial educ ion o odou dispe sion by wind- b eaks is e iden and, conside ing i s low implemen a- ion cos , his echnique can be ecommended o educe odou nuisances. Howe e , s anda dizing his echni- que in p ac ice is challenging. The e is s ill a lack o scien i ic knowledge measu e in p ac ice how he spe- ci ic en i onmen al condi ions a e a ec ing he wide ange o possible windb eak ypes and disposi ions. Resea ch needs and u u e p io i ies Recen esea ch on odou science in li es ock p o- duc ion has signi ican ly imp o ed ou knowledge on odou na u e, measu emen echniques, dispe sion modelling, and mi iga ion echniques. I is ema kable ha he ange o a ailable mi iga ion echniques is nowadays no signi ican ly di e en om hose p opo- sed a numbe o yea s ago (O’Neill & Phillips, 1991; Powe s, 1999). As i is now accep ed, hese au ho s also conside ed he ele ance o eeding s a egies, li e- s ock housing, ai il e ing, addi i es, manu e co e s, manu e managemen sys ems and windb eaks, among o he s. Many o hese aba emen echniques we e p i- ma ily de eloped o educe ammonia emissions. How- e e , esea ch so a has gi en us an insigh in o he undamen als and po en ials o hose echniques o aba e odou s. The combina ion o mo e accu a e mea- su emen me hods and compu e ad ances has p o- ided he oppo uni y o quan i y he po en ial odou mi iga ion o di e en s a egies and o assess he nuisance po en ial in a m su oundings using nume i- cal dispe sion models. Howe e , e o s a e equi ed o o e come some aspec s o which scien i ic know- ledge is s ill incomple e, such as odou measu emen s and e o analysis. Despi e he ad ances in odou science, odou mea- su emen s ill poses a challenge o esea che s. Simila o o he a mosphe ic pollu an s o ag icul u al o igin, odou emissions a e highly a iable and in many cases di icul o ace because o he di use na u e o hei sou ces. In con as o o he a mosphe ic pollu- an s, which can be analy ically de e mined wi h p eci- sion, he subjec i e componen o odou pe cep ion makes p og ess mo e di icul . The ela ionship among odo an subs ances, odou concen a ion and odou in ensi y is now be e unde s ood, bu measu emen e o s a e s ill oo la ge o de ec ela i ely small e ec s o aba emen echniques (Clan on e al., 1999; Boeke & Haas, 2007; Banhazi e al., 2009). I is also s ill necessa y o quan i y mo e p ecisely and a easonable cos s he o he componen s o nuisance po en ial, na- mely equency, du a ion and o ensi eness. I is easo- nable ha adop ing s anda dized ield ol ac ome y echniques would allow ob aining be e es ima ions o he odou impac in he su oundings o di use sou ces. In his sense, adop ing an in e na ionally s an- da dized ield assessmen me hod seems an essen ial equi emen o adop a common s a egy ( o example a he Eu opean Union le el) o aba e odou nuisances. Quan i ying e o s o odou es ima ions may be o highes in e es o scien is s and policy make s. To his aim, unce ain y analysis can be pe o med acco - ding o he concep s and p ocedu es de ined by he Guide o he exp ession o unce ain y in measu emen (ISO, 1995). Al hough o odou emissions such an analysis is a p esen missing, po en ial unce ain y sou ces o odou impac assessmen a e ou lined in S a egies o con ol odou s in li es ock acili ies: a c i ical e iew 1011 Fig. 3. As an example, Boeke & Haas (2007) ound ha lowe and uppe limi s o measu emen unce - ain y, exp essed as a pe cen age o he odou concen- a ion, we e 25% and 400% espec i ely. Ob iously, his le el o unce ain y makes di icul es ablishing compa isons among al e na i es o educe odou emissions a a m le el. Apa om measu ing odou concen a ions, e- sea che s ind o he wo impo an sou ces o unce - ain y when de e mining odou emissions om li e- s ock buildings. The i s sou ce is he in insic a iabili y o odou emissions as a unc ion o in e e- la ed ac o s such as ai empe a u e and manu e mana- gemen , among o he s. Because o his a iabili y i is complica ed o ob ain a ep esen a i e ai sample, which leads o impo an sampling e o s. The second sou ce o unce ain y is he measu emen o en ila ion a es. In mechanically en ila ed buildings his unce - ain y has been epo ed o be a bes 5-10% o he measu ed en ila ion, bu in na u ally en ila ed buil- dings measu emen e o s may be conside ably highe (Van Buggenhou e al., 2009). Ve y open, na u ally en ila ed li es ock houses a e cu en ly an in e es ing al e na i e agains in ensi ely con olled buildings, pa icula ly in ca le and pig buildings. Howe e , in hese sys ems measu ing en ila ion lows and odou emission a es is ex emely unce ain, and he e o e i is a challenge o esea che s o cha ac e ize hese sys- ems in e ms o odou p oduc ion. When de e mining odou nuisance po en ial using dispe sion modelling, he co esponding modelling e o s a e added o he a o emen ioned odou emission unce ain y. Howe e , he con ibu ion o each indi i- dual unce ain y sou ce o he inal e o o modelled nuisance po en ials is o he momen unknown. Quan- i ying he unce ain y o modelled odou nuisance po- en ials and de e mining he con ibu ions o each indi idual sou ce would con ibu e o es ablish esea ch p io i ies conduc ing o educe he mos ele an e o s. This would also undoub edly lead o a be e unde - s anding o he p oblem by esea che s, a me s and policy make s. Conclusions The ange o a ailable mi iga ion echniques is nowadays simila o hose p oposed a numbe o yea s ago. Howe e , ecen esea ch on li es ock odou s has signi ican ly imp o ed ou knowledge on odou na u e, measu emen echniques, dispe sion modelling, and mi iga ion echniques. Accep ed odou aba emen s a- egies a a m le el a e: eeding s a egies, li es ock housing, ai il e ing, diges i e and manu e addi i es, manu e co e s, manu e managemen sys ems and windb eaks. Al hough hese echniques a e known o yea s, esea ch so a has gi en us an insigh in o he undamen als and po en ials o hose echniques. Despi e his inc easing knowledge o odou aba emen s a egies, hei e ec i e implemen a ion a a m le el needs a de ini e commi men o a me s and egula- o s. Resea ch also indica es ha odou p oblems mus 1012 Y. Ubeda e al. / Span J Ag ic Res (2013) 11(4): 1004-1015 Figu e 3. O e iew o unce ain y sou ces in he de e mina ion o odou impac s. Odou sou ce E o concen a ion Ven ila ion a e Va iabili y and sampling Model e o s Me e eological a ia ion Dispe sion Odou emission Recep o s 䊱 䊱 䊱 䊱 䊱 䊱 䊱䊱 Odou impac