Response distances of wild forest reindeer (Rangifer tarandus fennicus Lönnb.) and semi-domestic reindeer (R. t. tarandus L.) to direct provocation by a human on foot/snowshoes

Rangi e , 33, (1) 2013 32 (1), 2012 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no Response dis ances o wild o es eindee (Rangi e a andus ennicus Lönnb.) and semi-domes ic eindee (R. . a andus L.) o di ec p o oca ion by a hu- man on oo /snowshoes Mau i Nieminen Finnish Game and Fishe ies Resea ch Ins i u e, Reindee Resea ch S a ion, Toi oniemen ie 246, FIN-99910 Kaamanen (Co esponding au ho : [email p o ec ed]) Abs ac : The objec i e o he s udy was o examine esponse dis ances o wild o es eindee (Rangi e a andus ennicus Lönnb.) and semi-domes ic eindee (R. . a andus L.) in Finland and No way o di ec p o oca ion by a human on oo /snowshoes in 5 a eas and in 15 eindee he ding coope a i es du ing di e en seasons in 2010-12. The e we e no signi ican di e ences in mean he d size o in sigh , ale , ligh and closes esponse dis ances o wild o es eindee in he Kuhmo and Suomenselkä a eas. The encoun e dis ance in wild o es eindee was signi ican ly (P< 0.005) longe han in semi-domes ic eindee in Finland and in Finnma k, No way, and i inc eased wi h he g oup size. The sigh and he ale dis ances in wild o es eindee we e signi ican ly (P< 0.001) longe han in semi-domes ic eindee . In addi- ion, he ligh dis ance o wild o es eindee (mean 192 m) was signi ican ly (P< 0.001) and almos h ee imes longe han in semi-domes ic eindee in Finland (mean 68 m). The closes mean dis ance was in wild o es eindee 191m ( ange 100-320 m) bu only 44 m ( ange 2-110 m) in semi-domes ic eindee (P< 0.001). The sigh , ale , ligh and closes esponse dis ances we e sligh ly longe in No wegian han in Finnish semi-domes ic eindee . Howe e , hese dis ances we e signi ican ly (P<0.005) longe in Pohjois-Salla (no supplemen a y eeding) han in o he Finnish eindee he ding coope a i es and a he Kaamanen expe imen al s a ion. The mean ligh dis ance o eindee in Pohjois-Salla was 115 m bu only 65 m in o he coope a i es (P< 0.001). The closes dis ance o semi-domes ic eindee in Pohjois- Salla (mean 105 m) was mo e han 2.5 imes longe han in o he eindee he ding coope a i es (mean 40 m). The mean sigh , ale and ligh dis ances in wild o es eindee in au umn and win e we e signi ican ly longe (P<0.005) han in semi-domes ic eindee in Finland. Howe e , du ing summe hese dis ances in wild o es eindee he ds wi h young cal es we e signi ican ly longe (P<0.005). The mean he d size o Finnish semi-domes ic eindee was almos he same in di e en seasons, bu in wild o es eindee i was sligh ly bigge du ing win e and sp ing and smalle du ing summe and au umn, only 7-23 eindee . The mean encoun e and sigh dis ances in semi-domes ic eindee we e signi ican ly longe (P<0.005) in win e , bu he mean ale and ligh dis ances we e almos he same in win e and summe and sligh ly longe han du ing o he seasons. The esul s sugges ha he supplemen a y eeding p ac ice du ing win e may likely cause a educ ion in ligh dis ances in semi-domes ic eindee . Key wo ds: human dis u bance; esponse beha iou ; ligh dis ance; Rangi e a andus; wild o es eindee ; semi- domes ic eindee ; seasons; supplemen a y eeding Rangi e , 33, (1), 2013: 1-15 1 Rangi e , 33, (1) 2013 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no 32 (1), 2012 In oduc ion The e a e many ac i i ies in which people may nega i ely in luence he beha iou o wild Ran- gi e a andus, and also a ec hei mo emen and subsequen ange use (Wol e e al., 2000). Human ac i i ies and in as uc u e con i- bu e wi h noise o example om powe lines, gene a o s, windmills, and also om mo ing objec s like humans on oo , snowshoes and skis, snowmobiles, ou -wheele s, ca s, ai - c a s and helicop e s. While oads alone a e no likely pe cei ed as a h ea o eindee , in- c easing oads and a ic a e. Du ing he las 30-40 yea s ou doo eco ou ism, hiking, skiing and hun ing ha e been expanding and inc eas- ing ac i i ies in mo e emo e a eas, including moun ain habi a s o wild and semi-domes ic eindee (Helle & Sä kelä, 1993; Colman e al., 2001; Reime s e al., 2006). The e ha e also been many changes in eindee he ding and husband y p ac ices o e he las yea s, espe- cially in Finland (Nieminen, 2006). Acco ding o Baskin & Skogland (2001) eindee a e a an ea ly phase o domes ica ion, bu semi-domes ic eindee gene ally exhibi mo e elaxed igh and ligh beha iou com- pa ed o wild eindee (Reime s e al., 2000; 2006). The e o e, when compa ing beha iou be ween di e en eindee he ds, i is impo - an o know he o igin and his o y o he he ds in ques ion. While he ca ibou subspecies a e wild, he Fennoscandian und a eindee popu- la ion includes many domes ica ed he ds (Re- ime s & Colman, 2006). The semi-domes ic eindee he ds in No he n Finland, Sweden and No way, as well as he wild eindee he ds (wi h a mix o wild and domes ica ed o igin) in Sou he n No way, a e o iginally Eu asian wild und a eindee (Rangi e a andus a andus L.). DNA-analyses by Røed e al. (2008) suppo independen o igin o semi-domes ic eindee in Fennoscandia and Russia. The domes ic gene pools seem o mee only in eas e n Finland, mainly in Halla eindee he ding coope a i e. Howe e , he wild o es eindee (R. . ennicus Lönnb.) popula ion in eas e n Finland and he wild eindee popula ions in cen al No way ha e con ibu ed li le o no hing o he do- mes ic gene pool (Røed e al., 2008). Domes- ica ion is he i s s ep o selec ion, and i is a gued ha domes ica ion has mos ly esul ed in quan i a i e a he han quali a i e changes (Mignon-G as eau e al., 2005). Human ac i i ies a ec eindee /ca ibou h ough he senses o hea ing, sigh and smell. Acco ding o Flydal e al. (2001) he hea ing ca- paci y o eindee anges om 70 Hz o 38 kHz a a sound p essu e o 60 dB. I means ha al- mos all noises and ocaliza ions a e eadily pe - cei ed by eindee . Reindee as o he ungula es has appa en ly also e y good day and nigh i- sion. Reindee mos likely pe cei e colou s, bu no pa icula colou appea s o be dominan , and eindee p obably a e unable o dis inguish be ween ed and g een colou . I is mainly con- as s and mo emen s ha be ay human p es- ence. Because he eyes o eindee a e la e ally posi ioned, he combined isual ields o bo h eyes co e i ually 360o (Nieminen, 1994). I means ha eindee can also spo p eda o s and humans sneaking up om behind. Reindee ´s la e ally posi ioned eyes limi , howe e , he bin- ocula isual ield. Al hough eindee ´s sense o smell is no well documen ed, he capaci y o cap u e scen s e en unde un a ou able wind condi ions is well known by eindee /ca ibou hun e s, hike s and also eindee he de s. Some- imes smell alone can igge ligh o eindee wi hou inpu om o he senses. Reime s & Colman (2006) p edic ed ha eindee would also espond a g ea e dis ances o he di ec ly app oaching pe son when he wind ca ied he human scen o he eindee han when he eindee could no smell he human in ude . The s ongly ele a ed nasals o he wild o es eindee li ing only in o es ed egions indica e he e y keen sense o smell due o inc eased ol- ac o y mucous memb anes (Nieminen, 1980). 2 Rangi e , 33, (1) 2013 32 (1), 2012 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no I has long been ecognized ha lea ning plays an impo an ole in he manne and deg ee o which ungula es espond o humans (Geis , 1971), and he e a e usually h ee majo lea ned esponses which also a e alid o ein- dee : habi ua ion, a ac ion and a oidance. Domes ica ion, habi ua ion and sensi isa ion a e essen ial in shaping adap abili y o ein- dee and ca ibou (Reime s & Colman, 2006). Many impac s udies ocus on he beha iou al esponses o wildli e o humans, because hese a ibu es a e gene ally mo e amenable o s udy han o he o ms o esponse (Bejde e al., 2009). The cases o p esumed habi ua ion o sensi isa ion may ac ually ep esen di e ences in he ole ance le el o wildli e o an h opo- genic ac i i y. Fo example eindee show de- c eased ligh esponses in a eas wi h ela i e high amoun s o human ac i i ies (Colman e al., 2001; Reime s e al., 2009), indica ing he abili y o habi ua e o human ac i i ies. Habi - ua ion o humans would occu mo e eadily in ca ibou popula ions ha we e no hun ed and li ed in a eas lacking na u al p eda o s (Klein, 1980; Aas up, 2000). Habi ua ion and also o me expe iences wi h p eda o s signi ican ly in luence an ungula e´s pe cep ion o h ea . The popula ions wi h ew p eda o s lushed a g ea e dis ance han hose whe e p eda o s a e common. All p eda o s udies e iewed by S ankowich & Blums ein (2005) classi ied hu- mans as he p eda o and measu ed di e ences in ligh ini ia ion dis ance be ween ungula e popula ions ha di e ed wi h ega d o hu- man exposu e. Wild ungula e popula ions like S alba d eindee (R. . pla y hynchus V olik) exposed o a ela i ely high le el o human ac- i i ies ha e become habi ua ed o humans in a non- h ea ening con ex (Tyle , 1991). They a e likely o pe cei e less isk when app oached by humans han would animals in popula ions whe e encoun e wi h humans a e a e (Col- man e al., 2001; Lund, 2008). E en hough semi-domes ic eindee a e dis u bed by hu- man ac i i ies, hey can inc ease hei ole ance owa ds humans i insec ha assmen is se e e du ing summe (Ska in e al., 2004). Ru ing ac i i ies du ing au umn ob iously also a ec eindee beha iou mo e han he di ec ly ap- p oaching human obse e (Reime s e al., 2006). Fligh ini ia ion dis ance is he dis ance a which an animal begins o lee om an ap- p oaching p eda o o human. I is usually used in s udies, because i is easy o measu e and co - ela es wi h o he key aspec s o escape beha - iou e.g. ale dis ance (Blums ein e al., 2005). Acco ding o Vis nes & Nellemann (2008) ac- cu a e assessmen o impac s om human ac i - i y equi es, howe e , egional-scale and usually long- e m s udies. The objec i e o he p esen s udy was o examine esponse dis ances o wild o es eindee and semi-domes ic eindee in Finland and No way o di ec p o oca ion by humans on oo o on snowshoes. The obse a- ions we e collec ed du ing di e en seasons in a eas subjec ed o combina ions o high o low human ac i i y, supplemen a y win e eeding o eely-g azing in he o es s o on he ields. The objec i e was also o de e mine i ligh dis ances ha e changed o e he las yea s in e- sponse o inc eased human ac i i y and supple- men a y eeding o semi-domes ic eindee on na u al pas u es o in co als. Ma e ial and me hods S udy a eas and eindee he ds In o al, 55 di e en eindee he ds we e in- cluded in he p esen s udy: 17 wild o es eindee he ds (Rangi e a andus ennicus Lön- nb.) in he Suomenselkä a ea (9 in Pe ho, 1 in Kyyjä i and 7 in Alajä i municipali ies) and 3 in Kuhmo municipali y in he Kainuu a ea, 32 semi-domes ic eindee he ds (R. . a andus L.) in 15 eindee he ding coope a i es in Finland including a he d a Kaamanen expe imen al s a ion (150 eindee , enced a ea, 44 km2), and 3 Rangi e , 33, (1) 2013 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no 32 (1), 2012 3 he ds in No he n No way (2 in Kau okeino, Wes -Finnma k and 1 in Pykeija, Eas -Finn- ma k) (Fig.1). App oxima e he d size was 40 animals in wild o es eindee and 115 ani- mals in semi-domes ic eindee in Finland and No way. The he ds o wild o es eindee we e ound wi h he help o GPS-colla ed emales. In he s udied eindee he ding coope a i es in Finland o al numbe s o coun ed eindee a ied be ween 2200-8300 animals. Reindee he ds we e s udied du ing di e en seasons in he cen al a eas o he coope a i es. Acco d- ing o he eindee he de s hese eindee he ds ep esen a mix u e o animals owned by many he de s ( he animals ha e di e en ea ma ks), i.e. cons i u e o animals belonging o he main he d o he espec i e coope a i e. The wild o es eindee disappea ed om Finland o decades in he beginning o he 1900s. They came back o he Kuhmo a ea in he 1950s. Du ing 1979-80 wo o es eindee males and eigh emales we e ans e ed om Kuhmo o Salamajä i Na ional Pa k (Nie- minen & Lai inen, 1983) and since hen he popula ion has inc eased in he Suomenselkä a ea o o e 1100 indi iduals. Howe e , du - 4 Fig.1. S udy a eas o wild o es ein- dee in Kuhmo (64o 7´ N, 29o 31´ E) and Suomenselkä (Pe ho, Kyyjä i and Alajä i) (63o- 63o 13´ N, 24o- 24o 25´ E) and o semi-domes ic eindee in eindee he ding coope a- i es (1 = Pais un u i, 4 = Muddusjä - i, 7 = I alo, 8 = Hammas un u i, 9 = Salli aa a, 10 = Muo ka un u i, 14 = Ky ö, 18 = O aniemi, 23 = Pohjois- Salla, 24 = Salla, 25 = Hi asniemi, 28 = Poikajä i, 46 = Oi anki, 49 = Pudasjä i and 57 = Halla) (64o-70o N) and a he Kaamanen expe imen- al s a ion (69o 6´ N, 27o 11´ E) in Finland and in Kau okeino (69o N, 23o 2´ E) and Pykeija (69o 58´ N, 29o 38´ E) in Finnma k, No way. Rangi e , 33, (1) 2013 32 (1), 2012 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no ing he las yea s he numbe o wild eindee has dec eased due o p eda ion. In Kuhmo mu- nicipali y, nea he Russian bo de , he popula- ion o wild o es eindee has also dec eased, and he o al numbe o eindee is oday abou 900 (Kojola e al., 2009). No he n Finland di e s om he sou h mainly by he ype o o es , bu also by he ele a ed wa e shed a eas in cen al and eas e n Finland. Suomenselkä and Kainuu belong o he middle bo eal eg- e a ion zone (Ah i e al., 1968). The landscapes a e domina ed by mi es, and hey a e loca ed on wa e sheds. Bo h a eas a e di e se in al i- udes, moun ains, hills and alleys, ba en a eas wi h o es s, bogs and lakes. The landscapes a e domina ed by No wegian sp uce (Picea abies) and Sco s pine (Pinus syl es is) o es s wi h e icaceous hea he , lichen (Cladonia, Cladina spp.) and boggy a eas. The no he n eindee he ding coope a i es a e si ua ed in he no h bo eal ege a ion zone (Ah i e al., 1968). The e ain is domina ed by olling hills wi h di - e en aged o es s ands o mainly Sco s pine. Moun ain bi ch (Be ula pubescens cze epanowii) g ows in he slopes o he highes hills, and only he ops o he highes ells a e ba en. The co- ope a i es in he middle and sou he n pa s o eindee he ding a ea belong, like Kainuu, o he middle bo eal ege a ion zone (Ah i e al., 1968). Ag icul u e is common and he e a e many ields in he eindee he ding coope a- i es and also in wild o es eindee a eas. Be- cause he numbe s o wild o es eindee ha e dec eased as desc ibed abo e, no eindee was hun ed du ing he s udy pe iod in he Kuhmo and Suomenselkä a eas. In he eindee husband y a ea in Finland, and also in Finnma k, No way, lichens pas u es a e gene ally s ongly o e y s ongly wo n (lichen biomass < 100-300 kg d y weigh / ha) (Kumpula e al., 2009; Ma ila, 2006). As esul s o eindee g azing, lichen anges in moun ain a eas, la ge na ional pa ks, na u e e- se es and o he wilde ness a eas a e wo n in Finland (Nieminen, 2010). In he no he n coope a i es also he eindee summe pas u es a e wo n, and g azing has been he main eason o change ege a ion and cause e osion in some places. The cu en condi ion o win e pas- u es in combina ion wi h a con inuous s a e o change o he wo se show ha main aining he cu en numbe o eindee on hei na u al win e pas u e is no longe possible. Howe e , he body condi ion o eindee in Finland has usually been good e en du ing ha d win e s, due o he in ensi e supplemen a y eeding p ac ised in he coope a i es. To ally o e 40 million kg eed (calcula ed as d ied hay) a e used yea ly du ing win e s o eeding o semi- domes ic eindee , mainly on na u al pas u es o in co als in he middle o sou he n ein- dee he ding coope a i es (Nieminen, 2006). Supplemen a y win e eeding o eindee in Finnma k a ea, No way, is no common, bu in Finland only Pohjois-Salla eindee he ding coope a i e has he ded eindee on he na u al pas u es wi hou supplemen a y eeding du - ing win e s. The wild o es eindee a e eely- g azing in Suomenselkä and Kuhmo, and bo h win e and summe pas u es a e in a he good condi ion. In Kuhmo he amoun o lichen biomass (d y weigh ) has o e he las yea s been se en imes highe han in he nea by Halla eindee he ding coope a i e (Ma ila, 2004). The oad ne wo k in wild o es eindee a eas and also in eindee husband y a eas a e well de eloped, and e e y yea some wild o - es eindee and o e 4000 semi-domes ic ein- dee die by a ic. Also, some adul wild o es eindee and mo e han 4000 semi-domes ic eindee a e killed yea ly by big p eda o s in Finland (Nieminen, 2012). Da a collec ion In 2010-12, du ing he ou sampling pe iods o Sep embe -No embe (au umn and u - ing pe iod), Feb ua y-Ma ch (win e ), Ap il (sp ing) and July (summe ), a single and same 5 Rangi e , 33, (1) 2013 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no 32 (1), 2012 pe son ( he obse e ) on oo o on snowshoes (used in deep snow), d essed in da k clo h- ing, dis u bed wild o es and semi-domes ic eindee du ing dayligh hou s by di ec ly ap- p oaching hem. The obse e used binocula s and came a (Nikon D80) o documen beha - iou and places and measu ed la e (using 1 m s eps) esponse dis ances be ween he eindee and he obse e and he esul an dis ances by he eindee a e aking ligh . Upon loca- ion o a g oup (≥ 4 eindee ), 10 pa ame e s we e eco ded: 1) sample mon h, 2) g oup size (small: < 20 animals, medium: 20-49 animals, and la ge: >50 animals), 3) g oup composi ion ( emales and cal es, males), 4) dominan ac i i- y o he g oup when i s sigh ed (lying o g az- ing, u ing, mo ing), 5) wind di ec ion ela i e o he obse e (no wind, upwind o down- wind), 6) ege a ion ype (open ield, ma sh, o es ), 7) opog aphy o he su ounding a ea (le el, moun ain, lake ice and eeding place), 8) isibili y/wea he (sunny, cloudy, aining/ snowing), 9) snow dep h and 10) empe a u e. When a g oup o eindee was i s sigh ed, he obse e ook pic u es, and used he so called di ec app oach me hod: ad ancing di- ec ly owa ds he cen e o he g oup a a con- s an speed (abou 4 km/hou ) wi h < 10 sec- ond s ops, o ake pic u es and la e measu ed he ou addi ional esponse dis ances de ined below. When eindee a e i s dis u bed hey show signs o awa eness and igh by aising hei heads and ails, u ina ing and some imes jumping (deVos, 1960; Ho ejsi, 1981). The ini- ial ligh is o en ollowed by cu iosi y beha - iou : he dis u bed eindee ci cles a ound he in ude o ca ch he scen . All measu emen s we e made om he posi ion o he di ec ly app oaching obse e o he nea es eindee . The wildli e esponse dis ance e minology and me hodology ecommended by Taylo & Knigh (2003) wi h he modi ica ions ollow- ing Reime s e al. (2003; 2009) we e used in his s udy: 1) Encoun e dis ance (END): he dis ance used as he s a ing poin o he dis u bance. The eindee i s disco e s he p o oke by sigh o scen , indica ed by looking, s anding, u ning hei head o pausing om ea ing in a manne isible o he obse e . 2) Sigh dis ance (SD): he dis ance be ween he obse e and closes eindee when eindee in he g oup displayed an ale ed beha iou di- ec ed a he obse e . 3) Ale dis ance (AD): he dis ance a which he eindee g oup displayed an inc eased ale esponse by g ouping oge he o by indi iduals u ina ing wi h one hind leg ex ended ou wa d a an exagge a ed angle, while s a ing a he di- ec ly app oaching obse e (Fig. 2). 4) Fligh dis ance (FD): he dis ance om he di ec ly app oaching obse e o he g oup when he eindee ini ially ook ligh . 5) Closes dis ance (CD): he dis ance om he di ec ly app oaching obse e o he nea es animal i a g oup app oached he obse e im- media ely be o e inal wi hd awal. 6) Escape dis ance (ED): he sho es s aigh - line dis ance om whe e he eindee ook ligh in esponse o he obse e o whe e he eindee esumed g azing o bedded down. 7) Assessmen ime: he ime elapsed om ale o ligh ini ia ion es ima ed om measu ed dis ances and assuming a cons an obse e speed o abou 4 km/hou . S a is ical analyses Ini ially, da a o all a eas and bo h wild o es eindee and semi-domes ic we e pooled o analyse he e ec o each ac o on di e en dis ances sepa a ely. Co ela ion among de- penden a iables was es ed wi h o iginal (un- ans o med) da a using Spea man ank co e- la ion. The esponse a iables (END, SD, AD, FD and CD) we e i s ans o med in o hei na u al loga i hms p io o analysis. Sigh , ale , ligh and closes dis ances we e analysed wi h Mixed Models Analysis. Ma ginal F- es s we e 6 Rangi e , 33, (1) 2013 32 (1), 2012 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no used o he ull models con ained a ea, season, g oup size, wind di ec ion ela i e o he ob- se e , ege a ion ype and dominan ac i i y o eindee . To es o di e ences o p o oca ion me hods and o ela e independen a iables o eindee esponses, a mixed, s epwise analysis o ANOVA (analysis o a iance) was used. x2 - es was used o assess he ela i e seasonal e- quency in g oup-size classes. The s a is ical di - e ences esponse dis ances in di e en g oups we e es ed using - es . S a is ical es s we e ca ied ou by use o SPSS e . 7.0 o Win- dows. The da a we e examined o s a is ical signi icance a P<0.05. Resul s Du ing he ou sampling pe iods in 2010-12 o ally 55 independen eindee g oups we e encoun e ed and used in s a is ical analysis. A o al o 2216 eindee we e obse ed. O hese, 739 we e wild o es eindee and 1477 semi- domes ic eindee . Al hough 27 eindee g oups we e encoun- e ed du ing au umn, hese g oups ep esen ed only 22 % o he g oups in he la ge size class (> 50 animals) and 9.3 % o all o he eindee sampled. Du ing au umn, u ing season, mos eindee we e in ew la ge, mixed-sex g oups. To ally 14 g oups we e encoun e ed du ing win e , 50 % o he g oups was in he la ge size class and ep esen ed 20 % o all s udied eindee . Du ing sp ing and summe , 8 and 6 g oups, espec i ely, we e encoun e ed, and only 1-2 g oups we e in he la ge size class. Du ing summe , eindee we e dis ibu ed usu- ally in small- and medium-sized emale-cal Fig. 2. The ale dis ance (AD) o a wild o es eindee g oup in he Suomenselkä a ea, in Feb ua y 2010. The eindee g oup displayed an inc eased ale esponse by g ouping oge he o by indi iduals u ina ing wi h one hind leg ex ended ou wa d a an exagge a ed angle, while s a ing a he di ec ly app oaching obse e . Pho o Mau i Nieminen. 7 Rangi e , 33, (1) 2013 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no 32 (1), 2012 g oups. Mean he d size o wild o es eindee was 37, which was almos he same as o semi- domes ic eindee in Finland (mean 39). The g oups obse ed in No way (win e only) we e bigge (mean 178 animals). The encoun e (END), sigh (SD), ale (AD), ligh (FD) and closes dis ance (CD) we e posi i ely co ela ed be ween a eas, wi h a gene al dec ease in he co ela ion coe icien s om SD o CD (Table 1). FD inc eased wi h inc easing encoun e dis ance (END) (Table 2), as did also SD and AD, indica ing ha when he obse e app oached eindee om a he away hey esponded a longe dis anc- es. No signi ican di e ences, depending on whe he he app oach was made on oo o on snowshoes in win e , we e seen in any o he e- sponse dis ances. The e we e also no signi ican di e ences depending on a ea (Kuhmo and Suomenselkä) in mean he d size, END, SD, AD, FD o CD o wild o es eindee . The En- coun e dis ance (mean ± s anda d de ia ion) o wild o es eindee (332 ± 24 m) was, how- e e , signi ican ly (P< 0.005) longe han ha o semi-domes ic eindee in Finland and No way (226 ± 13 and 250 ± 6 m, espec i ely) (Table 3). END inc eased wi h g oup size in bo h wild o es eindee and semi-domes ic eindee . SD (253 ± 16 m) and AD (216 ± 13 m) in wild o es eindee we e signi ican ly (P< 0.001) longe han in Finnish semi-domes ic eindee (144 ± 7 and 94 ± 6 m, espec i ely). The FD in 8 Table 1. Spea man ank co ela ion o he i e dependen a iables encoun e (END), sigh (SD), ale AD), ligh (FD) and closes dis ance (CD). Encoun e dis ance Sigh dis ance Ale dis ance Fligh dis ance P P P P Sigh dis ance (SD) 0.843 0.001 Ale dis ance (AD) 0.736 0.001 0.895 0.001 Fligh dis ance (FD) 0.699 0.001 0.867 0.001 0.960 0.001 Closes dis ance (CD) 0.592 0.001 0.787 0.001 0.905 0.001 0.922 0.001 Table 2. Linea mixed-e ec s model o p edic ing ligh -ini ia ion dis ances (In ans o med) o g oups o wild o es eindee and semi-domes ic eindee dis u bed by anapp oaching obse e on oo /snowshoes in di e en a eas in Finland and No way in 2010-12. Va iable Es ima e SE d - alue P- alue Sigh dis ance, In e cep 3.748 1.229 26 3.05 <0.005 Season (win e s. au umn) -0.555 0.463 26 -1.19 0.246 (win e s sp ing) -0.307 0.604 26 -0.51 0.612 (win e s. summe ) -0.766 0.891 26 -0.96 0.348 G oup size (la ge s. small) 0.353 0.279 26 1.27 0.217 (la ge s.medium) 0.157 0.205 26 0.77 0.451 Wind (no wind s. upwind) 0.099 0.166 26 0.59 0.558 (no wind s. downwind) 0.001 0.209 26 0.01 0.995 Vege a ion ype (open s. o es ) -0.12 0.564 26 -0.21 0.833 (open s. ma sh) -0.199 0.522 26 -0.38 0.706 Ac i i y ( u ing s. eeding) 0.206 0.322 26 0.64 0.528 ( u ing s. lying) 0.008 0.231 26 0.03 0.974 ( u ing s. g azing) 0.003 0.205 26 0.02 0.987 Encoun e dis ace (ln m) 0.327 0.089 26 4.11 <0.005 Rangi e , 33, (1) 2013 32 (1), 2012 This jou nal is published unde he e ms o he C ea i e Commons A ibu ion 3.0 Unpo ed License Edi o in Chie : Bi gi a Åhman, Technical Edi o E a Wiklund and G aphic Design: Be il La sson, www. angi e .no wild o es eindee (192 ± 14 m) was almos h ee imes longe han in semi-domes ic ein- dee (68 ± 5 m) in Finland (P<0.001) (Fig. 3). The escape dis ance (ED) o eindee was pos- sible o measu e only in he moun ain a eas, and he mean ED o semi-domes ic eindee in Muo ka un u i eindee he ding coope a i e in Finland and also in Kau okeino, No way was 360 m. The mean dis ance be ween ale and ligh was 24 m in wild o es eindee and almos he same, 26 m in semi-domes ic eindee in Finland. Wi h an encoun e speed abou 4 km/ hou , hese dis ances sugges ha he e was a sepa a ion o 22-24 seconds, on a e age, om when he eindee g oups became ale un il hey ook ligh . The mean closes dis ance was 191 m ( ange 100-320 m) in wild o es eindee bu only 44 m ( ange 2-110 m) in semi-domes ic eindee in Finland (P<0.001). The mean SD, AD, FD and CD in No wegian semi-domes ic eindee he ds we e sligh ly, bu no signi ican ly longe han in Finnish semi- domes ic eindee . Howe e , he mean SD, AD, FD and CD o he semi-domes ic eindee in Pohjois-Salla (no supplemen a y eeding) we e signi ican ly (P<0.005) longe han in he o he Finnish semi-domes ic he ds (Table 4). The mean FD o he semi-domes ic eindee in Pohjois-Salla was 115 m, bu only 65 m in he o he he ds. The mean CD o semi-domes ic 9 Table 3. Obse ed esponse dis ances o wild o es eindee and semi-domes ic eindee in Finland and also o semi- domes ic eindee in Finnma k, No way, when p o oked by an obse e on oo /snowshoes in 2010-12 (da a pooled ac oss yea s). Wild o es eindee (Kuh, Suo) Semi-domes ic eindee (Fin) Semi-domes ic eindee (No ) mean ange n mean ange n mean ange n He d size 37 4-152 20 39 11-121 32 116 72-220 3 Encoun e (END) 332 120-600 20 226 100-400 32 250 232-255 3 Sigh (SD) 253 100-450 20 144 80-250 32 200 198-205 3 Ale (AD) 216 100-320 20 94 50-160 32 150 145-156 3 Fligh (FD) 192 100-320 20 68 30-120 32 135 120-150 3 Closes (CD) 191 100-320 15 44 2-110 26 120 100-140 2 Fig. 3. Response dis ances (mean ± SD) o wild o - es eindee (WFR) and o semi-domes ic eindee in Finland (SDR-Fin) and in No way (SDR-No ) in g oups dis u bed by a di ec ly app oaching hu- man on oo /snowshoes in 2010-12. Sample sizes a e: END (encoun e ), SD (sigh ), AD (ale ), and FD ( ligh dis ance), n = 20 o WFR, n = 32 o SDR-Fin and n = 3 o SDR-No , and CD (closes dis ance), n = 15 o WFR, n = 26 o SDR-Fin and n = 2 o SDR-No . eindee in Pohjois-Salla was 105 m, which was mo e han 2.5 imes longe han in he o he he ds (mean 40 m) (Table 4). The mean END, SD, AD and FD o wild o es eindee we e long in au umn and win- e , and signi ican ly longe (P<0.005) han in semi-domes ic eindee in Finland (Table 5, Fig. 4). Howe e , du ing summe hese dis- ances o wild o es eindee he ds wi h young cal es we e signi ican ly longe (P<0.005). The mean he d size o Finnish semi-domes ic eindee was almos he same du ing he ou