Objective measures of spatial sffects in Spanish concert halls

ARCHIVES OF ACOUSTICS Vol. 37, No. 4, pp. 529–547 (2012) Copy igh c 2012 by PAN – IPPT DOI: 10.2478/ 10168-012-0063-y Objec i e Measu es o Spa ial E ec s in Spanish Conce Halls Sa a GIRÓN(1), Teófilo ZAMARRE˜ NO(1), Ped o BUSTAMANTE(2) (1) Depa amen o de F´ısica Aplicada II Uni e sidad de Se illa, E.T.S. A qui ec u a (IUACC) A . Reina Me cedes 2, 41012-Se illa, Spain; e-mail: {sgi on, eofilo}@us.es (2) Depa amen o de Cons ucciones A qui ec ónicas I, Uni e sidad de Se illa, E.T.S. A qui ec u a (IUACC) A . Reina Me cedes 2, 41012-Se illa, Spain; e-mail: bus aman[email p o ec ed] ( ecei ed Ap il 9, 2012; accep ed Oc obe 22, 2012) The p esen wo k consis s o a s a is ical s udy o he monau al (la e al- eflec ion ac ions and le el) and binau al acous ic pa ame e s (in e -au al c oss-co ela ion coefficien s) ha e alua e he amoun o ea ly and la e la e al acous ic ene gy encoun e ed in 9 pe o mance halls in Andalusia (sou he n Spain). Hall olumes ange be ween 6,163 m3and 34,594 m3and all enclosu es a e used o p esen a ions o symphonic conce s and o he music pe o mances. The majo i y o hese enues a e loca ed in p o incial capi als o he communi y and o en cons i u e he only p emises in he ci y whe e symphonic conce s can be held. The acous ic pa ame e s unde s udy he e we e de i ed om impulse esponses analyses using a sine-sweep signals which we e gene a ed and p ocessed by WinMLS 2004 so wa e in he oc a e- band equency cen ed om 125 o 4 kHz, and all pa ame e s we e spec ally a e aged acco ding o he ISO 3382-1 s anda d. A compa ison is p esen ed o monau al expe imen al esul s as a unc ion o sou ce- ecei e dis ance wi h he p edic ion o Ba on’s e ised heo y o conce halls, and he analyses o he acous ic pa ame e esul s a e ca ied ou in e ms o hei espec i e jus no iceable diffe ences: a he many mic ophone posi ions o he wo sou ce posi ions on s age, o he spa ial dis ibu ion o sea s in he audience zone ela i e o he cen al axis ( o le - and igh -hand sides) o he ooms, and o he p esence o he o ches a shell on s age. Resul s e eal ha he o ches a shell p opi ia es a pe cep ible dec emen in he alues o he ea ly la e al ene gy ac ion and an inc emen in he la e la e al le el a he audience sea s. In addi ion, a eg ession s udy e eals ha he wo kinds o measu es o la e ali y, monau al and binau al, a e co ela ed when he hall-a e age da a is conside ed, bu hey emain unco ela ed when all indi idual posi ions a e used. Likewise, he anges o a ia ion o he acous ic pa ame e s ound in hese halls a e na owe han hose specified in he ISO 3382-1. The pape concludes wi h a discussion on he ela ionships o hall-a e age da a o he fi e pa ame e s wi h eigh geome ic and acous ic a iables. Keywo ds: conce hall acous ics, di ec ional ene gy pa ame e s, binau al pa ame e s, spa ial imp es- sion, spa ial effec s. 1. In oduc ion A e he commencemen o he science o a chi ec- u al acous ics wi h W. C. Sabine’s con ibu ions in 1895 (Sabine, 1993), b eak h oughs and apid de el- opmen o elec oacous ics in he fi s hal o he wen- ie h cen u y p opi ia ed a g ea ad ance in gene al acous ics. Howe e , i is om he las hi d o he las cen u y ha he science o oom acous ics expe ienced a esu gence p opelled by he eme gen compu e and signal science. As a consequence o esea ch in ha new age, conside a ions o he han hose in he monau al empo al ac o s came in o play beyond he scope o e- e be a ion ime o de e mine op imal configu a ions o lis ening spaces (Be anek, 1962). In he 1960s, he fi s e e ences appea ed (Sch oede e al., 1966; Ma shall, 1967) ha conside ed he significance o spa ial effec s in he acous ics o conce halls. This pe cep ion e e s o how he acous ical imp ession in a oom diffe s om ha which lacks many delayed e- flec ions a i ing om all di ec ions. Soon a e , wo k by Ma shall (1968), and Ba on (1971) confi med ha his imp ession is p oduced by he ea ly la e al eflec ions, and his unde s anding has since had a sig- nifican impac on conce hall design (C eme , 1989; Be anek, 1992). As a ypical spa ial ac o o he sound field, Damaske and Ando (1972) defined he in e au al c oss-co ela ion coefficien , and in a efined 530 A chi es o Acous ics – Volume 37, Numbe 4, 2012 app oach, Ando (2007) de eloped a heo y o subjec- i e p e e ences in ela ion o empo al and spa ial ac- o s o he sound fields based on he modelling o he human audi o y-b ain sys em (Soe a e al., 2002). Rooms o music a e he mos isible and in e es - ing spaces in a chi ec u al acous ics whe e he h ee disciplines, o acous ics, a chi ec u e and music a e blended. Howe e , a ing he acous ic quali y o a hall o musical pe o mances is a non i ial and mul idi- mensional opic due o he lack o an unequi ocal c i- e ion o lis ene s’ music app ecia ion, in con as o ha o he one-dimensional comp ehension o speech (Peu z, 1971). Spa ial imp ession phenomenon is a gene al e m which co e s wo subjec i e pe cep ual ca ego ies: ap- pa en sou ce wid h (ASW), and lis ene en elopmen (LEV). ASW is he appa en audi o y wid h o he sound field c ea ed by a pe o ming en i y as pe cei ed by a lis ene in he audience a ea o a conce hall, and expe imen s ha e concluded ha i is mainly e- la ed o he ea ly la e al eflec ions eaching he lis- ene (Ba on, 1971). LEV is conside ed as he sub- jec i e imp ession by a lis ene o being en eloped by he sound field: a condi ion ha is, acco ding o ex- pe imen s, p ima ily ela ed o he la e la e al sound field (B adley e al., 2000). Two ypes o measu emen s o la e al eflec ions om he impulse esponse in a oom ha e eme ged: he monau al la e al ene gy ac ions (J) ela ed o he ene gy o ea ly o la e la e al eflec ions in he o me case, and he in e au al c oss-co ela ion coefficien (IACC) in he la e , which is a measu e o he diffe - ence in sound a he wo ea s and hence o la e ali y. To assess ASW, he ea ly la e al ene gy ac- ion JLF (Ba on, Ma shall, 1981), (o al e na- i ely, he ea ly la e al ene gy ac ion cosine JLFC (Kleine , 1989), is accep ed as a monau al di ec ion- ally influenced measu e. Likewise, he no malised in- e au al c oss-co ela ion unc ion IACF (Damaske, Ando, 1972) is a measu e o he dissimila i y be ween he a i als o he wa e a he wo ea s. F om his unc- ion he in e au al c oss-co ela ion coefficien s, IACC, a e hen gi en by: IACC 1, 2= max IACF 1, 2 o −1ms < τ < +1 ms. In which 1= 0 and 2= 80 ms (subsc ip E) a e cho- sen o he ea ly in e au al c oss-co ela ion coefficien IACCEwhich assesses he ASW pe cep ual a ibu e (Hidaka e al., 1995; Okano e al., 1998). In he same way, he la e la e al sound le el LJis p oposed as a s anda d (B adley, Soulod e, 1995a; 1995b) in he monau al impulse esponse measu emen in o de o p o ide a p ac ical indica o o he amoun o lis ene en elopmen in conce halls o sound a - i ing om all di ec ions (E jen e al., 2001), and he la e in e au al c oss-co ela ion coefficien IACCL (co ela ion calcula ed om 1= 80 ms o infini y (subsc ip L)) in he binau al measu emen , is p o- posed by Hidaka e al. (1995), and Okano e al., (1998). Al e na i ely, a u he pa ame e which in ol es a a io, he la e la e al ene gy ac ion JLLF, (Ba on, 2001) defined by he same a io as JLF bu o he la e in e al, o s udy lis ene in ol emen is also consid- e ed he e. Bo h monau al la e la e al pa ame e s a e ela ed o omnidi ec ional sound s eng h Gand cla - i y pa ame e C80 h ough: JLLF = 10(LJ−G)/10 1 + 10C80/10.(1) Due o he con amina ion o he esul s wi h back- g ound noise unde ce ain equencies, his pa ame e canno be ob ained di ec ly om WinMLS, and hence i s calcula ion in his wo k is as defined abo e. Acco ding o se e al expe imen s, he spec al con- en o la e al eflec ions ha con ibu e owa ds he ASW and LEV and la e al le els in he monau al pa- ame e s a e in he ou oc a e bands wi h cen e e- quencies 125, 250, 500, and 1000 Hz (Hidaka e al., 1995), espec i ely. This measu e a ou s he equency egion whe e he wa eleng hs a e mos ly longe han he acous ical dis ance be ween he le - and igh - hand sides o a head. In he case o he IACC pa- ame e , no widesp ead ag eemen ye exis s on how o calcula e a e ages o e a ious oc a e bands: in ac ISO 3382-1:2009(E) (2009) does no speci y his poin . Th oughou his wo k he spec al a e age o he IACC pa ame e is de e mined ollowing Hidaka’s ins uc ions (Hidaka e al., 1995; Okano e al., 1998), which a e based on p e ious expe imen s o Blaue and Lindemann (1986a, 1986b) and which concludes ha 500, 1 k, and 2 kHz a e he oc a e bands o equal and p incipal impo ance o he ASW a ibu e co - e ing he effec on spa ial imp ession ca ego ies o he en i e oc a e band equency ange (Okano e al., 1998). Hence o wa d, hese expe imen al measu es ob- ained by a i hme ically a e aging wi hin hese e- spec i e a o emen ioned equencies a e named JLFm, JLFCm,JLLFm, IACCEm, and IACCLm, espec i ely. Acco ding o in e na ional s anda ds (ISO, 2009) he equency-a e aged la e la e al sound ene gy le el, LJm mus be a e aged in ene gy. Al hough De V ies e al. (2001) sugges ed a de- composi ion o he measu es o supp ess in e e ence effec s o which he human ea is appa en ly insensi- i e, he e ised ISO s anda d (ISO, 2009) has p o- posed JLFm,JLFCm as objec i e measu emen s o he pe cep ual lis ene aspec ASW, a jus -no iceable di - e ence o 0.05, and a ypical ange o alues om 0.05 up o 0.35 ( equency-a e aged alues in single posi- ions in non-occupied conce and mul i-pu pose halls up o 25,000 m3). Fo LJm, i s jus -no iceable di - e ence emains unknown and o IACC acco ding o Cox e al. (1993) a jus -no iceable diffe ence o 0.075 is assumed. S. Gi ón, T. Zama e˜no, P. Bus aman e – Objec i e Measu es o Spa ial Effec s in Spanish Conce Halls 531 Wi hin his field, o he pieces o wo k (Mo imo o e al., 2008) ha e p oposed expe imen s o cla i y he essen ials o ASW and LEV om he poin o iew o audi o y beha iou by sugges ing ha he acous- ic componen s o he eflec ions o e and beyond he uppe limi o he p ecedence effec con ibu e owa ds ASW and LEV, espec i ely. Fu he mo e, he compo- nen s om behind he lis ene gene a e g ea e LEV (Mo imo o e al., 2001). Abdou and Guy (1996) also poin ed ou he necessi y o spa ial in o ma ion on he sound fields o he e alua ion o he acous ics o ooms and he diagnosis and co ec ion o he causes o acous ic de ec s. This pape aims o in e p e and discuss he esul s o monau al and binau al measu es o spa ial imp es- sion in 9 pe o mance spaces o sou he n Spain which a e used as conce halls and o o he musical pe o - mances. Analyses a e ca ied ou in e ms o sea posi- ions and hall-a e age alues wi h significan physical a iables o he acous ic field. The spa ial dis ibu ion o he a ious pa ame e s, he diffe ences ound due o he wo sou ce posi ions on he s age, and o he o ches a shell a e analysed in e ms o hei espec- i e jus -no iceable diffe ences (JND). These analyses could p o ide a s a ing poin o u he s udies o co ela ion be ween objec i e pa ame e s and spa ial imp ession. 2. Expe imen al me hod 2.1. Measu emen echnique The p ocedu es employed he e a e hose es ab- lished in he ISO 3382-1 s anda d (ISO, 2009), whe eby measu emen s we e ca ied ou in unoccupied ooms and on unoccupied s ages (no public, no musicians), sa e y cu ains we e always d awn open, and, in ce - ain cases wi h he o ches a shell configu a ion in he s age, chai s and music s ands we e p esen . Tempe a- u e and ela i e humidi y we e moni o ed du ing he measu emen s which e ealed a ange o a ia ion o he en i onmen al condi ions du ing measu emen s in all enclosu es o 16.8–25.2◦C o he empe a u e, and 38–60% o he ela i e humidi y, espec i ely, wi h he excep ion o Falla G and Thea e which was measu ed in win e , whose minimum empe a u e was 14.1◦C. The a ia ions be ween he en i onmen al condi ions o he halls ha ha e been measu ed wi h and wi h- ou an o ches a shell in he s ages ne e exceeded 1◦C in empe a u e and 8.5% in ela i e humidi y espec- i ely, excep o Falla G and Thea e whose measu e- men s we e ca ied ou in win e and sp ing wi h a diffe ence o 9◦C in empe a u e and 12% in ela i e humidi y, espec i ely. Monau al and binau al impulse esponses (IRs) we e measu ed o de e mine, among o he s, he ol- lowing pa ame e s o each equency band be ween 125 Hz o 4000 Hz and in all ecei e posi ions: he ea ly la e al ene gy ac ion (JLF )and ea ly la e al ene gy ac ion cosine (JLFC), he la e la e al ac ion (JLLF), he la e la e al le el (LJ), and he binau al in- e au al c oss-co ela ion coefficien s (IACC), o s udy he spa ial imp ession phenomena in he ooms. The IR has been ob ained a each ecep ion poin using sine sweep signals which we e gene a ed and analysed by WinMLS 2004 so wa e ia he VX Pocke 2 sound ca d om Digig am. The omnidi ec ional dodecahed al sou ce DO12 wi h i s In e M-1000 am- plifie , is placed a he mos usual poin o loca ion o he na u al sou ce (S) on he o ches a pla o m (none in he o ches a pi which was always co e ed wi h wooden panels o hea y cu ains) a a heigh o 1.50 m om he floo . Whe e e possible, measu e- men s in wo sou ce posi ions on he s age ha e been ca ied ou (see Table 1 and Fig. 1). In all cases he S1 posi ion coincides wi h he cen al sou ce posi ion on he s age; and hese wo sou ce loca ions coincide o he configu a ions wi h and wi hou an o ches a shell. Two mic ophones ha e been used: mul ipa e n config- u able Audio-Technica AT4050/CM5 (wi h an app o- p ia e signal-condi ioning Ea hwo ks LAB1 amplifie ) o measu e la e al ene gy ac ions and le els; and a o so ype HSU III simula o (Code 1323) om Head Acous ic (wi h OPUS amplifie ) o eco d binau al IR. They a e all loca ed a he app oxima e heigh o he head o a sea ed pe son, ∼1.20 m om he floo , in a p ede e mined numbe o sea s wi hin he a ious audience zones ( hese sea s a e coinciden o he mea- su emen s ca ied ou wi h and wi hou he o ches a shell on he s age in ce ain halls). 2.2. The halls esea ched In his sec ion, da a o he 9 conce halls unde esea ch is p esen ed: Co doba G and Thea e (CGT), loca ed in Co doba; Falla G and Thea e (FGT), lo- ca ed in Cadiz; Huel a G and Thea e (HGT), loca ed in Huel a; Isabel la Ca ólica Thea e (ICT), loca ed in G anada; Lope de Vega Thea e (LVT), loca ed in Se ille; Miguel de Ce an es Thea e (MCT), lo- ca ed in Malaga; Manuel de Falla Audi o ium (MFA), loca ed in G anada; Maes anza Thea e (MT), lo- ca ed in Se ille; and Villama a Thea e (VT), lo- ca ed in Je ez de la F on e a (Cadiz). They appea in alphabe ical o de in acco dance wi h he ac onyms assigned om hei names on he MIREM web si e (digi al scenic and musical enclosu e map o Spain), whe e u he in o ma ion on he enclosu es is a ail- able (MIREM, 2012). De ails on he s age shell o hese spaces and s age suppo pa ame e s ha e been pub- lished elsewhe e (Gi ón e al., 2010). In all hese buildings, se e al ypes o cul u al ac- i i ies, such as ope a, hea e, dancing, cinema and conce s, may be pe o med, excep in he Manuel de Falla Audi o ium whe e only conce s and eci als a e p esen ed. These a e pe o mance spaces wi h a fixed 532 A chi es o Acous ics – Volume 37, Numbe 4, 2012 public-s age ela ionship and a on al desc ip ion, ex- cep he Manuel de Falla Audi o ium which p esen s a bi- on al desc ip ion, (i consis s o wo ooms A and B, see Fig. 1). I is also pe inen o add ha all 7 p oscenium hea es we e e u bished by he e- gional go e nmen du ing he 1980s and 1990s, and ha acous ic measu es ha e been ca ied ou a e hese e u bishmen s. In Table 1, he yea o comple ion o inaugu a ion, a chi ec u al ype, he numbe o floo s, he numbe o poin s o ecep ion o he mic ophones, and he sea capaci y a e all summa ized in alphabe ical o - de , oge he wi h he mos ele an a chi ec u al and acous ical dimensions (León e al., 2007), whose cal- cula ions a e desc ibed below. Mean wid h W: o ho seshoe-shaped halls his is he maximum wid h on he g ound floo excluding boxes; i i is ec angula bu i egula , hen a mean alue is aken as in ICT; in he case o MFA, his geo- Fig. 1. G ound plan and longi udinal sec ion o 4 halls o he o al spaces analysed. The posi ions o he sou ce and he ecep ion poin s on each floo a e also shown. me ical ac o excludes boxes a bo h sides o oom A; and in MT his is i s ans e se diame e . Mean heigh H: calcula ed as he a e age o he heigh o he oom in he mou h o he s age, in he middle, and a he ea o he hall. Mean dep h D: he a e age o he ho izon al mea- su e ha exis s om he s age o he back o he oom on all floo s. (In he case o MFA, i is he leng h o oom A). Audience su ace SA: he a ea occupied by all sea s in he hall. To al olume V: hall and s age olume (in he wo configu a ions i i exis s wi h and wi hou he o ches- a shell). Re e be a ion ime T: he e e be a ion ime mea- su ed in he unoccupied halls and a e aged spa ially and spec ally in he 500 and 1000 Hz oc a e bands. Abso p ion A: he abso p ion ob ained using he e e be a ion ime in Sabine’s o mula. S. Gi ón, T. Zama e˜no, P. Bus aman e – Objec i e Measu es o Spa ial Effec s in Spanish Conce Halls 533 Table 1. Rele an da a o he conce halls s udied. Hall CGT FGT HGT ICT LVT MCT MFA MT VT Yea o comple ion 1873 1910 1923 1952 1929 1870 1978 1991 1928 TypologyaHP HP SP SP HP HP RA CA SP Numbe o sea s N946 1038 601 689 733 1058 A: 890 B: 413 1800 1200 Hall olume [m3] 6071 8114 4800 5035 5902 6594 A: 4274 B: 1536 20321 7988 Numbe o loo s 4 4 4 3 4 4 1 1 3 S age olume no shell [m3] 4631 5556 1963 1742 3363 4907 3421 14273 3703 S age olume wi h shell [m3] 872 1728 – – – 945 – 1850 574+405 To al olumeb[m3] no shell V′10702 13670 6763 6777 9265 13873 9231c34594 11691 To al olumeb[m3] wi h shell V6943 9842 – – – 9911 – 22171 8562 Hall mean wid h, W[m] 17.0 17.7 10.4 18.0 16.6 17.7 A+B: 19.7 37.0 22.8 Hall max dep h, D[m] 25.3 25.6 17.4 21.9 27.1 23.3 A+B: 24.4 36.3 24.9 Hall mean heigh , H[m] 12.4 14.9 16.7 17.3 18.9 19.9 A+B: 11.1 14.3 14.9 Sea su ace, SA[m2] 536 687 354 449 411 699 A+B: 786 1156 555 R. poin s no shell 22 15 16 12/12 14/14 15 22/22 – 18/18 wi h shell 17 15/15 – – – 15/15 – 15/15 17 T[s] no shell 1.17 1.89 1.41 1.26 1.44 1.26 2.33 – 1.85 wi h shell 1.19 1.86 – – – 1.14 – 2.51 1.70 A[m2]no shell 1473 1165 772 866 1036 1773 638 – 1017 wi h shell 939 852 – – – 1400 – 1422 811 aHP – Ho seshoe p oscenium loo plan. SP – Shoebox p oscenium loo plan. RA – Rec angula audi o ium loo plan. CA – Cylind ical audi o ium loo plan. bHall and s age olumes. cHalls A and B plus s age. To comple e his in o ma ion and o a oid excessi e leng h, Fig. 1 shows he plo ed g aphs o he g ound plans and longi udinal sec ion o only he CGT, MFA, MT, and VT, espec i ely. The plans include he poin s o ecep ion on each floo in he hea es wi h diffe en Fig. 2. Views om he s age o he in e io s o he halls; o MFA he iew is om oom B. symbols, and he wo posi ions o he sou ce on he s age. In addi ion, Fig. 2 shows a se o pho og aphs o he in e io s o each hall aken om hei s age, and om oom B in he case o MFA. 534 A chi es o Acous ics – Volume 37, Numbe 4, 2012 3. Analyses o expe imen al esul s and discussions This sec ion co e s, in i s a ious subsec ions, he analyses o monau al and binau al pa ame e s s udied in he 9 conce enues, bo h in e ms o hei spa- ial alues in he spaces as well as in e ms o he ela ionships be ween he diffe en ypes o pa ame- e s ha e alua e he same pe cep ual a ibu es o spa ial imp ession. The co ela ions be ween he hall- a e age pa ame e s wi h he geome ic and acous ic pa ame e s o he halls a e also ca ied ou . In all cases, he analyses p esen ed e e o he esul s o he spec- al a e ages o he pa ame e s as s a ed a he end o Sec. 1. These esul s can be used o u he s udies o co ela ion be ween objec i e pa ame e s and spa ial imp ession. As a s a ing poin , a compa ison by eg ession o he expe imen al alues o he wo monau al ea ly la - e al ac ions, JLFm and JLFCm, is ca ied ou . Re- g ession analysis shows ha he wo pa ame e s a e linea ly ela ed when he esul s a e compa ed o each hall and include he esul s o he wo s age configu a- ions, and when he esul s a e s udied in all indi idual posi ions in all he halls oge he . In he o me case he weakes co ela ion is, o FGT wi hou he o ches- a shell: JLFCm = 0.569 JLFm + 0.132, R2= 0.6144, P = 0.0005, (2) and he s onges linea co ela ion also o FGT, bu wi h he shell: JLFCm = 0.785 JLFm + 0.104, R2= 0.9560, P < 0.0001. (3) By conside ing he esul s in all indi idual posi ions o all halls (321 pai s o alues, o he wo configu a ions o he s age and he wo sou ce posi ions) he bes fi co esponds o a s aigh line: JLFCm = 0.837JLFm + 0.091, R2= 0.8554, P < 0.0001. (4) The good esul s o s a is ical analysis indica e ha hese wo pa ame e s a e e y simila and ha he expe imen al p ocedu e is co ec and can he e o e be used al e na ely o ei he di ec ional pa ame e o ea ly la e al ene gy. Simila esul s ha e been ob ained wi h diffe en equipmen in wo ship spaces o a com- mon ypology (Gi ón e al., 2008). Hence o wa d, he esul s o JLFm a e used since his pa ame e p esen s he bes acili ies o measu ing and has expe ienced he wides use in o he expe imen al wo k, he eby a- cili a ing compa ison. 3.1. Spa ial dis ibu ion o he la e al acous ic pa ame e s. E ec o he o ches a shell As a fi s s ep o his analysis, all he pa ame e s we e s udied as a unc ion o sou ce- ecei e dis ance. In gene al, he empi ical da a se appea s as a cloud o poin s and ails o indica e any p edic able beha iou in e ms o sou ce- ecei e dis ance. Fu he mo e, o he wo possible s age configu a ions, hese expe imen- al esul s we e also compa ed wi h he heo e ical al- ues de e mined by Ba on’s e ised heo y o sound p opaga ion in halls (Ba on, Lee, 1988), whe e e- e be a ion ime T, (measu ed in he unoccupied halls and a e aged spa ially and spec ally in he 500 and 1000 Hz oc a e bands), and he o al olume V, a e he only magni udes in ol ed in hei calcula ions. In o de o a oid excessi e leng h, he de ailed e- sul s o his s udy a e no shown. Howe e , i s main conclusions can be summa ized as: A) The e is con- side able sca e om he heo e ical p edic ions and he u hes de ia ions gene ally occu in he la e la - e al le el LJm ega dless o he ype o hall and o he sou ce- ecei e dis ance. B) The influence o he o ches al shell on he s age in he heo e ical p edic- ions is e y small. In he expe imen al esul s he di - e ences a e small, simila o he findings ob ained by B adley (1996), bu pe cep ible (in e ms o JND), as will be analysed la e . C) Ba on’s e ised heo y o sound p opaga ion in halls indica es ha la e la e al sound le els a e mo e sensi i e o changes in e e be a- ion ime and oom olume han ea ly la e al ac ions (Ba on, 2001), and hence he effec o adding an o - ches a shell may be he g ea es on he la e la e al a i ing sound le els a audience sea loca ions. In he absence o p edic able beha iou when s udying he esul s o hese fi e pa ame e s e sus dis- ance o he sou ce on he s age, i was decided o con- duc a s udy o hei spa ial dis ibu ion by ollowing a common me hodology o he fi e pa ame e s s udied. This consis s o e alua ing he pa ame e s acco ding o dis ances om he poin s o ecep ion o he cen al axis o he oom, x( o le - and igh -hand sides), and by no malizing hese dis ances wi h espec o hal he a e age wid h o each oom, w, hen hese ela i e dis- ances a e p esen ed as (x/w). This me hod o e alu- a ion is jus ified by he ac ha spa ial imp ession is ela ed o he p esence o ea ly and la e la e al eflec- ions, and ha he close he walls, he highe in ensi y should be. Fo he analysis o his comp ehensi e spa ial dis- ibu ion in all ooms, he alues o he acous ic pa- ame e s ela ed o spa ial imp ession a e g ouped in o disc e e in e als o he a o emen ioned ela i e dis- ances (x/w). The magni ude o hese in e als is se a 0.25, whe e posi ions x/w > 1exp ess he posi ions o he mic ophone e y close o he side walls, which in all cases co espond o posi ions in a eas o boxes and S. Gi ón, T. Zama e˜no, P. Bus aman e – Objec i e Measu es o Spa ial Effec s in Spanish Conce Halls 535 side e aces on he g ound o uppe floo s and he e- o e usually unde an o e hang, (e.g. in CGT, FGT, HGT, LVT, MCT and MFA). I should be bo ne in mind ha all he pa ame e s s udied a each posi ion ha e been a e aged in equency a he oc a e bands, wi h he p ocedu es es ablished by ISO 3382-1 (ISO, 2009), lis ed in Sec. 1. On he o he hand, in small ci ies, i is e y com- mon o adap pe o mance halls o he needs o o ches- al music by means o o ches a shells. Despi e hei widesp ead use, e y li le published in o ma ion exis s on hei acous ical p ope ies and hei influence in he acous ic field o he hall. This sec ion also analyses he influence o shells on he pa ame e s conside ed in he diffe en a eas. 3.1.1. Ea ly la e al ene gy ac ion JLFm In Fig. 3a, he ea ly la e al ene gy monau al pa- ame e JLFm is plo ed in o dina es and scaled a in- e als o 1 JND. In abscissa, he ela i e dis ance o he cen al axis o he oom is gi en o all he halls s udied wi hou he o ches a shell and wi h esul s o he wo posi ions o he sound sou ce (S1 and S2 on he s age, in he d awings o Fig. 1). The numbe o measu emen poin s in each disc e e posi ional zone is also specified a he op o he g aph. The ypical ange o his pa ame e (exp essed in Fig. 3a as ISO) acco ding o ISO 3382-1 (ISO, 2009), is (0.05, 0.35) o non-occupied conce halls and mul i-pu pose halls wi h a olume o less han 25,000 m3. Figu e 3a also indica es ano he na owe ange ound in hese pe o - mance spaces (0.10, 0.30) which means a cu o 1 JND in bo h he uppe and lowe limi s o he ange wi h espec o hose specified in ISO 3382-1. I can be ob- se ed ha his ange includes, in each case, he as majo i y o expe imen al esul s. The c i e ion used in hei specific de e mina ion acco ding o Fig. 3b, is discussed la e . The filled symbols linked by lines show he a e age alues o he JLFm pa ame e in each posi ional in e - al, and he deg ee o spa ial dispe sion in each in e - al is cha ac e ized by he s anda d de ia ion ( e ical ba s). The spa ial dispe sion ends o inc ease sligh ly o posi ions close o he side walls x/w >1 and he analysis o he esul s indica es ha , o x/w be ween 0.25 and 0.75, he a e age alues o he pa ame e a e e y cons an and only o he a ea nea he cen e o he halls, x/w < 0.25, does JLFm educe i s alues by 0.5 JND. As a complemen o Fig. 3a, and in o de o no mal- ize he expe imen al da a wi h espec o he spa ial dis ibu ion in he a eas conce ned, Fig. 3b exp esses in o dina es he pe cen age o he expe imen al da a in each in e al o 1 JND (n) o alues o JLFm in each posi ional a ea (x/w) in ela ion o he o al da a in he same a ea, named n∆(x/w)(see op o Fig. 3a). a) b) Fig. 3. a) Ea ly la e al ene gy ac ions in he ela i e dis- ance in e als o he cen al axis o he halls, o all halls, o he wo sou ce posi ions and o he wi hou shell on he s age. The mean alues and hei s anda d de ia ions o each in e al a e also shown. The ypical ange o alues in his wo k and in ISO 3382-1 a e p esen ed; b) ac ion o da a o alues o JLFm in s eps o 1 JND in he diffe en in e als o x/w ela i e o he o al numbe o da a in each in e al co esponding o he esul s shown in pa (a). I can be seen ha o each in e al he o al sum ails o each 100%, he eason being ha only hose alues o he acous ic pa ame e o which he da a lies mos ly abo e 10% in he whole ange a ia ion we e conside ed as significan in he wo figu es. Residual alues co espond o he ange o a ia ion o JLFm be- ween (0.00, 0.10) in he lowe ange and (0.30, 0.45) in he op ange. This de e mines he ypical ange o pa- ame e a ia ion in hese ooms. In his dis ibu ion, i can be obse ed how he pe cen age o he wo ma- jo anges (black and da k g ey ba s) inc emen s when app oaching he la e al walls. In o de o disce n he effec o he o ches al shell on he beha iou o he pa ame e , Fig. 4a shows a sim- ila s udy o JLFm bu conside ing only he ou pe - o mance halls (CGT, FGT, MCT, and VT) which 536 A chi es o Acous ics – Volume 37, Numbe 4, 2012 a) b) c) Fig. 4. a) Ea ly la e al ene gy ac ions in e ms o he di - e en in e als o ela i e dis ance o he cen al axis o 4 halls, a S1 sou ce posi ion and o he wo s age configu- a ions. The mean alues and hei s anda d de ia ions o each in e al and he ypical ange o alues in his wo k (wi h shell and no shell) and in ISO 3382-1, a e also shown; b) ac ion o da a o alues o JLFm in s eps o 1 JND in diffe en in e als o x/w ela i e o he o al amoun o da a in he same in e al, o 4 halls wi hou he o ches a shell; c) idem wi h he o ches a shell on he s age. ha e been s udied wi h he wo configu a ions o he s age and solely a he sou ce posi ion S1, o ensu e compa ible da a in he wo configu a ions. The s udy hus shows he possible effec when he amendmen o only one a iable is in ol ed ( he shell). (Al hough he amoun o da a should coincide, he e emains a sligh diffe ence due o echnical p oblems ha a ose in one space (CGT), see he op o Fig. 4a). Resul s confi m he end ob ained o he wi hou - shell configu a ions ha we e s udied o all halls in Fig. 3a, and he main enance o he ange es ablished (0.10, 0.30) o bo h s a is ical popula ions o da a ( he eby alida ing s a is ical esul s o he wi h-shell configu a ion), and also show ha shells lead o a sig- nifican dec ease in he a e age alues o he pa ame e JLFm in all audience zones. Specifically, in he cen e o he s alls, x/w ∈(0.00, 0.25), while be ween 0.50 and 0.75, he dec ease is abou 0.5 JND. Fo he zone o boxes and uppe balconies unde o e hangs, he o de o he dec ease is 1 JND. This sugges s, a leas ini- ially, ha shells ocus sound by dec easing he ini ial la e al eflec ions, especially in he a eas unde bal- conies. In addi ion, spa ial dispe sion p esen s simila be- ha iou emaining e y simila in he o he a eas and eaches i s highes alue, a x/w >1, he a ea nea he side walls. Resul s also confi m he na owe ange deduced be o e o he no-shell configu a ion in his pa ame e (0.10, 0.30), and ano he e en na owe ange in he uppe limi (0.10, 0.25) o he wi h-shell configu a- ion. The effec o he p esence o he o ches a shell on his ea ly la e al ene gy pa ame e in he audience sea s al hough small is audible, as seen in B adley’s wo k on a se o Ame ican halls (B adley, 1996). Complemen a y o hese s a emen s, Figs. 4b and 4c include he esul s o he pe cen age o da a in each posi ional in e al o his pa ame e , in s eps o 1 JND in each in e al, whe eby only hose esul s o he pa ame e o which he e a e significan alues abo e 10% in all a eas a e included. These esul s co - espond o he ange om 0.10 o 0.30 o he s age wi hou he o ches al shell and a e simila o he con- clusion d awn om Fig. 3b, and om 0.10 o 0.25 o he shell configu a ion. I is also wo h no ing ha he black ba does no appea i he shell is p esen . 3.1.2. La e la e al le el LJm A simila analysis was pe o med wi h he o he monau al pa ame e associa ed wi h he subjec i e pe cep ion, known as he en elopmen o sound o he subjec , LJm. Thus Fig. 5a shows he expe imen- al esul s ob ained a he wo sou ce posi ions in all halls wi hou he shell, and he a e aged alues o he pa ame e in he diffe en posi ional a eas. Mean al- ues a e e y simila in all posi ional zones, and lowe spa ial dispe sion is p esen a he cen e o he s alls and in he icini y o he side walls. Fu he mo e, i is wo h no ing ha he ange o a ia ion o he pa am- e e s udied in all he ooms can be se om −6dB o 0 dB which is a na owe ange o a ia ion om he ypical ange specified in ISO 3382-1 o conce halls and mul ipu pose ooms wi h a olume <25,000 m3. Fo he comple ion o his s udy, and in o de o no malize he dis ibu ion o da a, Fig. 5b p esen s he pe cen age o he amoun o da a, in s eps o 2 dB, ela i e o he o al amoun o da a in each posi ional in e al agains hese spa ial in e als. I is wo h no ing ha o he highes in e al (−2dB, 0 dB) he e is an inc emen o ecep ion poin s om he cen e o he halls o he la e al walls (black ba ), excep o he x/w > 1posi ions. Since he JND o his pa ame e emains unknown, he alue o 2 dB S. Gi ón, T. Zama e˜no, P. Bus aman e – Objec i e Measu es o Spa ial Effec s in Spanish Conce Halls 537 a) b) Fig. 5. a) A e aged la e la e al le el in he diffe en in e - als o no malized dis ance o he cen al axis o he halls: o all halls, he wo sou ce posi ions and wi hou o ches- a shell. The means and hei s anda d de ia ion alues a e also shown, oge he wi h he ypical ange o alues in his wo k and he ypical ange in ISO 3382-1; b) ac ion o da a o alues o LJm in s eps o 2 dB in he diffe en in e als o x/w ela i e o he o al amoun o da a in he same in e al, o all halls wi hou he o ches a shell on he s age. has been chosen due o he wide ange o he a ia ion: om −8.57 dB o +8.15 dB. In his ep esen a ion, he esul s o he alues o his pa ame e ha ail o p esen a significan amoun o da a (abo e 10%) in all posi ional zones a e igno ed. By es ic ing he s udy o he 4 ooms in which acous ic measu emen s we e aken in he wo config- u a ions, wi h and wi hou he o ches a shell on he s age, se o Figs. 6, he a e aged alues emain con- s an in all posi ional a eas and confi m he beha iou o he pa ame e alues in all ooms s udied in Fig. 5a, al hough sligh ly highe alues a e encoun e ed. Bo h Figs. 5a and 6a, main ain he uppe alue o he ange o a ia ion o he pa ame e in hese spaces (0 dB) de- spi e he significan numbe o poin s ha a e igno ed abo e 0 dB, because hey a e less han 10% in each 2 dB jump. Fu he mo e, hese esul s ela e mos ly o ecep o s o MFA ( his space has no o ches al shell) and a e he e o e no shown in Fig. 6a. In ac , he dis ibu ion o mean alues is app oxima ely −2dB when he shell is p esen and app oxima ely −4dB in i s absence (Fig. 6a). The gap be ween he wo con- figu a ions ends o dec ease o he a eas closes o he side walls (x/w > 0.75). Rega ding he spa ial dis- pe sion as assessed by he s anda d de ia ion, his is gene ally in he same o de o magni ude o he wo configu a ions o he s age, in all posi ional zones o he audience a ea. a) b) c) Fig. 6. a) La e la e al le el esul s in he diffe en in e als o no malized dis ance o he cen al axis o he halls o 4 halls, a S1 sou ce posi ion and o he wo s age config- u a ions. The mean alues and hei s anda d de ia ions o each in e al a e also shown, oge he wi h he ypi- cal ange o alues in his wo k and he ISO 3382-1 ange; b) ac ion o da a o alues o LJm in s eps o 2 dB in he diffe en in e als o x/w, ela i e o he o al amoun o da a in he same in e al, o 4 halls wi hou he o ches a shell; c) idem wi h he o ches a shell on he s age. As o he complemen a y s udy conduc ed h ough Figs. 6b and 6c, he da ase wi h alues o n/n∆xabo e he 10% limi sugges s sho ening he LJm ypical al- ues o be ween −8dB and 0 dB o he wi hou -shell configu a ion, in Fig. 5b he ange is na owe (−6dB, 0 dB), and be ween −6dB and +2 dB o he configu- a ion wi h he shell. The shell he e o e inc eases he 544 A chi es o Acous ics – Volume 37, Numbe 4, 2012 In all cases, he linea fi s s udied show e y weak co ela ions. Taking in o accoun he a o emen ioned assump ions, he only eliable co ela ions a e hose o (1-IACCEm)wi h A(wi h nega i e slope), (1- IACCLm)wi h T, whe eby he pa ame e inc eases when e e be a ion ime inc eases, and ha o LJm wi h V/N which also has a posi i e slope Eqs. (10)– (12). The alues and linea eg essions co esponding o he binau al pa ame e s a e plo ed in Figs. 11a and b, espec i ely. LJm =−0.763(V/N) + 5.159, R2= 0.4861, P = 0.0081,(10) (1 −IACCEm) = −0.0001A+ 0.668, R2= 0.5538, P = 0.0035,(11) (1 −IACCLm) = 0.044T+ 0.759, R2= 0.5828, P = 0.0024.(12) In he same way as he simple image model indi- ca es in conce halls and ope a houses, he mean hall ea ly la e al ene gy ac ions a e influenced by he hall wid h. O he pieces o wo k ha e ound his co ela- ion, o ins ance Gade (1989) concludes ha JLFm has no meaning ul connec ion o diffuse field heo y a) b) Fig. 11. a) Hall-a e age alues o (1-IACCEm) e sus A; b) (1-IACCLm) alues e sus T, in he wo configu a ions o he s ages and hei linea fi s. bu shows a highly significan ela ionship wi h hall wid h. Based on he da a om all 32 halls, he co e- la ion is R2= 0.43 and nega i e. Howe e , i he anal- ysis is es ic ed o 16 ec angula halls, i inc eases o R2= 0.67. In addi ion, he a io 2H/W which de e - mines he ela i e a i al ime o ea ly la e al and e - ical eflec ions, as Wes (1966) sugges ed o conce halls, seems o be a geome ical measu e in connec ion wi h ea ly la e al ac ion measu es. In he halls s ud- ied in his wo k, he wo geome ical magni udes W, and 2H/W p esen a e y poo ela ionship wi h he monau al ac ion JLFm and wi h he binau al mea- su e (1-IACCEm). Pe haps he low co ela ion in his wo k is due o he ac ha o ches a shells p o oke au- dible a ia ions in he acous ic pa ame e s, while he associa ed geome ical quan i ies emain unchanged. To sum up, unde he la ge olume o da a ana- lyzed in his wo k and he many a iables included i would be desi able o ollow B adley’s e y ecen sug- ges ions (B adley, 2011) in connec ion wi h ca ying ou expe imen al wo k o a mo e ocused na u e. 3.4. Summa y and discussion In his sec ion, a synopsis o he s udy is un- de aken and he esul s a e p esen ed o he objec- i e acous ic pa ame e s, (spec ally a e aged in acco - dance wi h he ISO 3382-1 s anda d (ISO, 2009)) ha quali y he spa ial imp ession in 9 pe o mance halls in sou he n Spain (Andalusia) in he a ious app oaches. The dependence o monau al and binau al pa am- e e s on sou ce- ecei e dis ance shows no p edic able beha iou and expe imen al esul s ha e been com- pa ed wi h he heo e ical calcula ions o Ba on’s e- ised heo y, which needs only he mean e e be a ion ime and o al hall olume o each hall (Subsec. 3.1, (Ba on, Lee, 1988)). The s udy o spa ial dis ibu ion o acous ic pa- ame e s in he ooms was join ly ca ied ou by scal- ing he esul s a in e als o 1 JND (0.05 o JLFm and JLLFm, 0.075 o (1-IACCm)measu es, and 2 dB o LJm, whose JND is no known) depending on he ela i e posi ions o he poin s o ecep ion o he cen- al axis o each oom ( o le and igh -hand sides), which we e no malized ela i e o hal hei espec i e mean wid h, (Subsec. 3.1). The o ches a shell com- monly p oduces a small bu pe cep ible effec on hese pa ame e s measu ed in he audience a ea, since o JLFm, Fig. 4a, he change may each a alue o 1 JND, especially in he a eas closes o he side walls. In ad- di ion, he ange o a ia ion o JLFm in he se o alues o all ooms is (0.10, 0.30) in he configu a ion wi hou shell (Figs. 3a and 4a), and (0.10, 0.25) wi h he o ches al shell, Fig. 4a, which sugges s ha he shell a ou s a edis ibu ion o he acous ic ene gy in he audience a ea in hese pe o mance spaces. Fu - he mo e, he a e age alues o JLFm in all posi ional S. Gi ón, T. Zama e˜no, P. Bus aman e – Objec i e Measu es o Spa ial Effec s in Spanish Conce Halls 545 in e als a e lowe when he shell is p esen on s age han when i is absen . Fo he la e la e al le el LJm, howe e , he shell seems o a ou an inc ease o he pa ame e , and in bo h configu a ions o he s age he ange o a ia ion is also much na owe han he yp- ical ange s a ed in ISO 3382-1 (ISO, 2009), (−8, 0) dB o he no-shell configu a ion, and (−6, 2) dB o he wi h shell, see se o Figs. 6. As o he binau al pa ame e s, (Table 2), bo h he ea ly la e al ene gy as well as he la e ene gy ex- hibi simila spa ial dispe sion and a e lowe han he monau al pa ame e s a he a ious posi ional in e - als. Fo (1-IACCEm), he ypical ange in hese spaces is (0.40, 0.70) o he wo configu a ions o he s age. Pa ame e (1-IACCLm)exhibi s he same beha iou in i s mean alues in each posi ional zone wi h a sligh peak in he a ea be ween he cen al axis and side walls o he oom (s alls zone). The s anda d de ia- ions show ha he spa ial dispe sions a e no wide, and a e sligh ly highe , gene ally, in he configu a ion wi hou shell. The ypical ange o his pa ame e in hese a eas is (0.75, 0.90) ega dless o whe he he e is an o ches a shell on s age. The effec on la e al ene gy pa ame e s measu ed in he audience zone due o he change o sou ce po- si ion on s age has been also s udied in e ms o he absolu e alue o he diffe ences in hei alues, scaled a in e als o 1 JND a each ecep ion poin o each oom. No p edic able beha iou o hese diffe ences in he pa ame e s associa ed wi h he a ia ion o sou ce- ecei e dis ance, no o he diffe en la e ali y o he sou ce sound posi ion has been de ec ed. In a s a is i- cal su ey o all ecep ion poin s, classified acco ding o how hey see he sou ce in all ooms oge he , mos o he sea s ha e a ia ions ha a e in he ange 1–2 JND o he ea ly la e al ene gy pa ame e s, whils o he la e la e al ene gy, mos poin s lie wi hin diffe - ences <1JND (Figs. 7 and Table 3). Conside ing he beha iou o hese diffe ences o each oom, JLFm p esen s e y homogeneous alues in MCT and LVT halls, as shown in Fig. 8a, which bo h ha e ho seshoe ypology. Gene ally o he la e la e al le el, LJm, Fig. 8b, a g ea e change is ound in he alues o hese diffe ences om place o place wi hin all he halls s udied. The esul s o he compa ison be ween alues o he wo ypes o measu es o la e al ene gy (monau al and binau al o he same empo al in e al) ob ained in he 9 halls in all he indi idual posi ions show ha he co ela ions a e e y weak and ha he ela ionships a e no s a is ically significan , as shown in he se o Figs. 9. Howe e , ela ionships imp o e when he da a on each hall is conside ed exclusi ely, especially in ce - ain halls, and when all hall-a e age da a is s udied o a compa ison be ween JLFm and (1-IACCEm). In he la e case, he inclusion in he s udy o he esul s o a se o 15 halls om Okano’s wo k (Okano e al., 1998) u he imp o es he co ela ion ound (Fig. 10a). Fo he la e la e al ene gy pa ame e s, no simila da a is a ailable in he li e a u e. Howe e , by aking in o ac- coun he alues o he 9 halls s udied in his wo k o he wo configu a ions o he s age, he ela ionship be- ween (1-IACCLm)and LJm pa ame e s becomes sim- ila o ha ob ained p e iously o he ea ly la e al ene gy pa ame e s as shown in Fig. 10b. Finally, he s udy in Subsec. 3.3.2 o he dependen- cies o hese hall-a e age pa ame e s wi h eigh acous- ic and geome ical pa ame e s, (namely, mean wid h W, mean heigh - o-wid h a io H/W, mean heigh - o- dep h a io H/D, mean dep h- o-wid h a io D/W , o al- olume- o-numbe -o -sea s a io V/N, audience- su ace- o-numbe -o -sea s a io SA/N, e e be a ion ime T, and abso p ion A), enables he conclusion ha he s a is ically significan ela ionships a e he ollowing linea eg essions be ween he a iables: (1-IACCEm)wi h A, Fig. 11a, wi h nega i e slope; (1-IACCLm)wi h T, Fig. 11b, wi h posi i e slope; and ha o LJm wi h V/N, also wi h nega i e slope. I can be obse ed ha , in all cases, he independen a iable is ela ed o he e e be a ion cha ac e is ic o he oom (i mus be aken in o accoun ha he majo abso ben su ace in hese spaces is he audi- ence zone, which is closely ela ed o he V/N a io). In his analysis, no o he geome ical o acous ic a i- ables ha e shown hei po en ial influence on he alues o he pa ame e s ela ed o spa ial imp ession. 4. Conclusions O he a ious aspec s s udied in his wo k on he monau al di ec ional and binau al pa ame e s (spec- ally a e aged in acco dance wi h he ISO 3382-1 s anda d) o e alua e he spa ial effec s o he sound fields in 9 pe o mance spaces in Andalusia (sou he n Spain), i is fi s wo h emphasising ha he depen- dence o he pa ame e s on sou ce- ecei e dis ance in each hall has e ealed ha la e al acous ic ene gy is no dis ibu ed acco ding o he heo e ical p edic ions o he diffuse e ised heo y by Ba on. As a esul , a common me hodology has been used o all a eas and pa ame e s unde s udy. In his me hodology he po- si ion o he ecei ing poin s a e exp essed in ela ion o hei dis ance om he symme y axis o he oom (le - and igh -hand sides) no malized wi h espec o hal he a e age wid h o he oom, and he acous- ic pa ame e s a e scaled a in e als o hei espec- i e JND. The s a is ical ea men is he same o he pa- ame e s, da a om all ooms wi h sou ce in S1 and S2 wi hou shell, and hen he 4 ooms in which hey a e measu ed wi h he wo configu a ions on he s age. The main enance o conclusions when ea ing a wide s a is ical ensemble wi h he second case (smalle ) gi es alidi y o he esul s and conclusions when shells 546 A chi es o Acous ics – Volume 37, Numbe 4, 2012 a e on he s ages. S a is ical esul s show ha he spa- ial dis ibu ion o ea ly la e al ene gy ac ions p o- ides highe alues o posi ions close o he side walls (mean alues o x/w > 1exceed hose o he cen al a ea o e 1 JND), whe eas hose pa ame e s associ- a ed wi h la e la e al ene gy and binau al pa ame e s p esen a mo e uni o m s a is ical dis ibu ion in he posi ional zones. The applied me hodology has allowed, o each pa- ame e s udied, he sys ema ic es ablishmen o hei cha ac e is ic anges o he dis ibu ion o hei al- ues o he wo shell configu a ions on he s age. These anges a e gene ally smalle han hose sugges ed in ISO 3382-1. S a is ical esul s ha e shown ha he o ches al shell p omo es a edis ibu ion o acous ic ene gy in o he audience a ea and p oduces a no iceable dec ease o he ea ly la e al ene gy ac ion: his dec ease is g ea e o posi ions close o he side walls. Howe e , o he la e la e al le el he p esence o he shell on s age leads o an inc ease in i s alues and a simila i y in magni ude in all posi ional in e als. This sugges s, a leas ini ially, ha shells ocus sound by dec eas- ing he ini ial la e al eflec ions, especially in he a eas unde balconies. This is eflec ed in he alues o he limi s o he anges men ioned in he p e ious pa a- g aph. The change in he wo posi ions o he sound sou ce on s age also influences hese pa ame e s; i s influence is g ea e on he ea ly la e al ene gy pa ame e s, and he diffe ences ound due o he change a e mo e ho- mogeneous o he ho seshoe ypology. Fo he la e la e al ene gy pa ame e s, he e is a lowe incidence o change in posi ion o he sound sou ce and esul s p esen mo e uni o mi y in audi o ium ypologies. The analysis has no shown ha he ela i e posi ion o he ecei e in on o he sou ce (le / igh ) has a signi - ican effec in any o he pa ame e s analyzed. The change in posi ion o he sou ce implies ha , o all he pa ame e s analysed, mo e han 80% o ecei e s o he en i e enclosu es expe ience a ia ions lowe han 2 JND. Fo he (1-IACCLm)pa ame e , his pe cen - age eaches 100% o ecei e s. Monau al and binau al pa ame e s desc ibe e y diffe en physical cha ac e is ics o he sound field, and hus co ela ions a e e y weak when s udying pai s o esul s a each posi ion o signal ecep ion. Exclusi ely when each hall is desc ibed by a single alue ( he spa- ial a e age) an accep able ela ionship be ween he pa ame e s is ound bo h o he ea ly and la e ime- in e al pa ame e s. Finally, in o de o d aw conclusions in ela ion o design a iables, monau al and binau al pa ame e s ha e been compa ed wi h he mean alues o geome i- cal and acous ical a iables o he pe o mance spaces. Reg essions show ha nei he he mean wid h, no he mean-heigh /mean-wid h a io is significan and ha in all cases, pa ame e s ela ed o e e be a ion a e he key (abso p ion, e e be a ion ime, and V/N a io). Acknowledgmen The au ho s wish o exp ess hei g a i ude o he s aff and managemen o each hall o hei coope a ion in allowing he measu emen s o be ca ied ou . This wo k has been sponso ed by FEDER unds and he Spanish Minis y o Science and Technology (MCYT) wi hin he p ojec s wi h e e ences BIA2003-09306- CO4-02, and BIA2010-20523. Re e ences 1. Abdou A., Guy R.W. (1996), Spa ial in o ma ion o sound fields o oom-acous ics e alua ion and diagno- sis, Jou nal o he Acous ical Socie y o Ame ica, 100, 3215–3226. 2. Ando Y. (2007), Handbook o Acous ics, Thomas Ross- ing, Edi o , Sp inge -Ve lag, New Yo k Chap e 10. 3. Ba on M. (1971), The subjec i e effec s o fi s e- flec ions in conce halls - he need o la e al eflec- ions, Jou nal o Sound and Vib a ion, 15, 475–494. 4. Ba on M. (2001), La e la e al ene gy ac ions and he en elopmen ques ion in conce halls, Applied Acous ics, 62, 185–202. 5. Ba on M., Lee L-J. (1988), Ene gy ela ions in con- ce audi o ia I, Jou nal o he Acous ical Socie y o Ame ica, 84, 618–628. 6. Ba on M., Ma shall A.H. (1981), Spa ial imp es- sion due o ea ly la e al eflec ions in conce halls. The de i a ion o a physical measu e, Jou nal o Sound and Vib a ion, 77, 211–232. 7. Be anek L.L. (1962), Music, Acous ics and A chi ec- u e, John Wiley & Sons, New Yo k. 8. Be anek L.L. (1992), Conce hall acous ics, The Jou nal o he Acous ical Socie y o Ame ica, 92, 1– 39. 9. Blaue J., Lindemann W. (1986a), Audi o y spa- ciousness: Some u he psychoacous ic analyses, The Jou nal o he Acous ical Socie y o Ame ica, 80, 533– 542. 10. Blaue J., Lindemann W. (1986b), Supplemen- a y psychoacous ical esul s on audi o y spaciousness, Acus ica, 59, 292–293. 11. B adley J.S. (1994), Compa ison o conce hall mea- su emen s o spa ial imp ession, Jou nal o he Acous- ical Socie y o Ame ica, 96, 3525–3535. 12. B adley J.S. (1996), Some effec s o o ches a shells, Jou nal o he Acous ical Socie y o Ame ica, 100, 889- 898. 13. B adley J.S. (2011), Re iew o objec i e oom acous- ics measu es and u u e needs, Applied Acous ics, 72, 713–720. S. Gi ón, T. Zama e˜no, P. Bus aman e – Objec i e Measu es o Spa ial Effec s in Spanish Conce Halls 547 14. B adley J.S., Reich R.D., No c oss S.G. (2000), On he combined effec s o ea ly- and la e-a i ing sound on spa ial imp ession in conce halls, Jou nal o he Acous ical Socie y o Ame ica, 108, 651–661. 15. B adley J.S., Soulod e G.A. (1995a), The in- fluence o la e-a i ing ene gy on spa ial imp ession, Jou nal o he Acous ical Socie y o Ame ica, 97, 2263– 2291. 16. B adley J.S., Soulod e G.A. (1995b), Objec i e measu es o lis ene en elopmen , Jou nal o he Acous ical Socie y o Ame ica, 98, 2590–2597. 17. Cox T.J., Da ies J., Lam Y.W. (1993), The sensi i - i y o lis ene s o ea ly sound field changes in audi o ia, Acus ica, 79, 27–41. 18. C eme L. (1989), Ea ly eflec ions in some mod- e n conce halls, Jou nal o he Acous ical Socie y o Ame ica, 85, 1213–1225. 19. Damaske P., Ando Y. (1972), In e au al c oss co e- la ion o mul ichannel loudspeake ep oduc ion, Acus- ica, 27, 232–238. 20. De V ies D., Hulsebos E.M., Baan J. (2001), Spa- ial fluc ua ions in measu es o spaciousness, Jou nal o he Acous ical Socie y o Ame ica, 110, 947–954. 21. E jen P., B adley J.S., No c oss S.G. (2001), The effec o la e eflec ions om abo e and behind on lis- ene en elopmen , Applied Acous ics, 62, 137–153. 22. Gade A.C. (1989), An acous ical su ey o ele en Eu- opean conce halls - a basis o discussion o halls in Denma k, The Acous ics Labo a o y, Tech. Uni . o Denma k, [Repo No.44]. 23. Gi ón S., Galindo M., Zama e˜ no T. (2008), Dis- ibu ion o la e al acous ic ene gy in Mudeja -Go hic chu ches, Jou nal o Sound and Vib a ion, 315, 1125– 1142. 24. Gi ón S., Zama e˜ no T., Galindo M. (2010), Ex- pe imen al s udy o suppo pa ame e s in audi o ium and p oscenium s ages, Ac a Acus ica uni ed wi h Acus ica, 96, 1026–1041. 25. Hidaka T., Be anek L.L., Okano T. (1995), In e - au al c oss-co ela ion, la e al ac ion, and low- and high- equency sound le els as measu es o acous ical quali y in conce halls, Jou nal o he Acous ical So- cie y o Ame ica, 98, 988–1007. 26. In e na ional O ganiza ion Fo S anda diza ion ISO 3382-1:2009(E), (2009), Acous ics-Measu emen o oom acous ic pa ame e s, pa 1: Pe o mance ooms, Gene a, Swi ze land. 27. Kleine M. (1989), A new way o measu ing he la - e al ene gy ac ions, Applied Acous ics, 27, 321–327. 28. León A.L., Send a J.J., Na a o J., Zama e- ˜ no T. (2007), Acous ics and es o a ion in hea es o Andalucia [in Spanish: Ac´us ica y Rehabili ación en Tea os de Andaluc´ıa], Sec e a iado de Publicaciones de la Uni e sidad de Se illa, Se illa. 29. Ma shall A.H. (1967), A no e on he impo ance o oom c oss-sec ion in conce halls, Jou nal o Sound and Vib a ion, 5, 100–112. 30. Ma shall A.H. (1968), Le els o eflec ion masking in conce halls, Jou nal o Sound and Vib a ion, 7, 116–118. 31. MIREM (Mapa In o ma izado de Recin os Esc´enicos y Musicales; Fundación Au o , Sociedad Gene al de Au o es y Edi o es) h p://www.a ene sgae.com- /mi e/index.h m [Digi al Scenic and Musical Enclosu e Map, Au ho Founda ion] (Accessed Sep embe 2012, cu en ly only in Spanish). 32. Mo imo o M., Iida K., Sakagami K. (2001), The ole o eflec ions om behind he lis ene in spa ial imp ession, Applied Acous ics, 62, 109–124. 33. Mo imo o M., Nakagawa, K. Iida K. (2008), The ela ion be ween spa ial imp ession and he law o he fi s wa e on , Applied Acous ics, 69, 132-140. 34. Okano T., Be anek L.L., Hidaka T. (1998), Re- la ions among in e au al c oss-co ela ion coefficien (IACCE), la e al ac ion (LFE), and appa en sou ce wid h (ASW) in conce halls, Jou nal o he Acous i- cal Socie y o Ame ica, 104, 255–265. 35. Peu z V.M.A. (1971), A icula ion loss o consonan s as a c i e ion o speech ansmission in a oom, Jou - nal o he Audio Enginee ing Socie y, 19, 915–919. 36. Sabine W.C. (1993), Collec ed Pape s on Acous ics. (Camb idge: Ha a d Uni e si y P ess), Rep in ed by Acous ical Socie y o Ame ica, New Yo k. 37. Sch oede M.R., A al B.S., Sessle G.M., Wes J.E. (1966), Acous ical measu emen s in he Philha monic Hall (New Yo k), Jou nal o he Acous- ical Socie y o Ame ica, 40, 434–440. 38. Soe a Y., Nakagawa S., Tonoike M., Ando Y. (2002), Magne oencephalog aphic esponses co e- sponding o indi idual subjec i e p e e ence o sound fields, Jou nal o Sound and Vib a ion, 258, 419–428. 39. Wes J.E. (1966), Possible subjec i e significance o he a io o heigh o wid h o conce halls, Jou nal o he Acous ical Socie y o Ame ica, 40, 1245.