Using models and connectivity analysis to the predict the current and future patterns of invasion by Silver-wattle (Acacia dealbata Link) in the Northwest Iberian Peninsula

Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e - wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula Alice Lucas Roxo Mes ado em Ecologia, Ambien e e Te i ó io Depa amen o de Biologia 2013 O ien ado P o esso Dou o João Hon ado P o esso Auxilia na Faculdade de Ciências da Uni e sidade do Po o e In es igado no Cen o de In es igação em Biodi e sidade e Recu sos Gené icos CIBIO -UP Faculdade de Ciências da Uni e sidade do Po o Coo ien ado Dou o a Joana Vicen e In es igado a no Cen o de In es igação em Biodi e sidade e Recu sos Gené icos CIBIO –UP Faculdade de Ciências da Uni e sidade do Po o FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 2 Acknowledgmen s Fi s o all, I would like o hank P o esso João Hon ado o he oppo uni y, he suppo and encou agemen he ga e me h oughou his yea . I also wan o hank Joana Vicen e, o all he a ailabili y, suppo , guidance and encou agemen p o ided his pas yea . To Rui Fe nandes, o all he help, a ailabili y, suppo , companionship and iendship h oughou his p ojec . To all he people om he esea ch g oup o Biodi e si y and Conse a ion Ecology, specially Angela Lomba, Paulo Al es, B uno Ma cos, João Gonçal es, An ónio Mon ei o, So ia Vaz and Ana Ri a Sil a, o he way hey ecei ed me and o all he assis ance hey p o ided. To F ede ico San a ém and Ve onica Ono e, o he la e 6y iendship, o being he bes iends you can ha e, and always being he e o me, o all he ollies and because wi hou hem my jou ney h ough his mas e would no be he same hing. To all my iends, especially Ismael Macaia, Po i io Cos a, Diogo Sil é io, Rica do Felicio, And é Gonçal es, Joana Rocha, Ma ga ida Hen ique, João Mou a, João San os, Sa a Mou inho and all Bio ocas o all he memo able momen s and all he laughs. To Miguel F iães, o all he suppo , help, dedica ion and company his pas yea , o being p esen in he wo s days, and o making my days be e . Finally, bu mos impo an ly o my amily, especially my pa en s, Ana and Paulo Sob al, and my b o he Tomas, o all hey ha e done o me, o all he suppo , dedica ion and ca e, because wi hou hem I wouldn’ be who I am oday and o be, abo e all, he bes iends you can possibly ha e. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 3 This ese ach was de eloped a FCUP wi hin p ojec “Biodi e sidad Vege al Amenazada Galicia-No e de Po ugal. Conoce , ges iona e implica ” (0479_BIODIV_GNP_1_E), unded by "2º Eixo p io i á io – Coope ação e Ges ão Conjun a e Meio Ambien e, Pa imónio e Ges ão de Riscos" o he "P og ama Ope acional de Coope ação T ans on ei iça Espanha – Po ugal (POCTEP) 2007- 2013". FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 4 Abs ac Biological in asions a e one o he majo p omo e s o biodi e si y loss wo ldwide. In asions by alien species a e hough enhanced by changes in clima e and dis u bance egimes, as well as by o he en i onmen al shi s. P ese ing na i e biodi e si y and ecosys ems om in asion by alien species equi es comp ehensi e s udies and measu es o an icipa e impac s, and o p o ec species and habi a s o high conse a ion alue. This can be achie ed using species dis ibu ion models (SDM) o p edic he dis ibu ion o in asi e alien species (IAS) in a egion o in e es , bo h o cu en condi ions and unde scena ios o u u e en i onmen al changes. In his esea ch, we modeled he dis ibu ion o Acacia dealba a Link (an agg essi e IAS) in he c oss-bo de con ex o Galicia and No h o Po ugal, in o de o p edic and explain he cu en dis ibu ion o Acacia dealba a and o o ecas possible impac s o u u e clima e change scena ios on ha dis ibu ion. We applied a combined p edic i e modeling (CPM) app oach, whe e we i ed species dis ibu ion models using subse s o p edic o s p e iously classi ied as ac ing a egional o local scales, ha we e spa ially combined o ob ain he inal p ojec ions. These combined models p edic a wide a ie y o po en ial species esponses, p o iding mo e in o ma i e p ojec ions o species dis ibu ions and dynamics han adi ional models. This app oach was complemen ed by connec i i y analyses o he dis ibu ion o he in ade , o e alua e cu en and u u e con lic s be ween his IAS and high conse a ion alue a eas. Model p ojec ions sugges ha he dis ibu ion o Acacia dealba a in he s udy a ea will inc ease in he u u e, i clima e changes scena ios a e con i med, and ha hese changes in en i onmen will inc ease he connec i i y o he species dis ibu ion in he a ea. The e o e, using CMPs and connec i i y analysis i is possible o suppo esou ce p io i iza ion o an icipa ion and moni o ing o IAS impac s. CMPs can suppo measu es o p e en in asions in his egion as well as c oss-bo de managemen in which bo h bo de s will wo k oge he and ocus hei e o s o p o ec and conse e speci ic impo an habi a s. Key wo ds: Acacia; Clima e change; P o ec ed a eas; Combined p edic i e modelling; Species dis ibu ion models. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 5 Resumo As in asões biológicas são um dos p incipais p omo o es da pe da de biodi e sidade a ní el global. Conside a-se que a in asão po pa e de espécies exó icas pode se es imulada po al e ações climá icas e dos egimes de pe u bação, bem como po ou as mudanças ambien ais. P ese a a biodi e sidade na i a e os ecossis emas das in asões po espécies exó icas eque es udos exaus i os e medidas pa a an ecipa os impac os das mesmas e p o ege as espécies e habi a s de ele ado alo de conse ação. A u ilização de modelos de dis ibuição de espécies (SDMs) pe mi e p e e a dis ibuição de espécies exó icas in aso as (EEI) numa ce a egião, an o pa a as condições a uais como pa a cená ios u u os de mudanças ambien ais. Nes e abalho, modelámos a dis ibuição de Acacia dealba a Link (uma in aso a ag essi a) num con ex o ans on ei iço (Galiza e No e de Po ugal), de o ma a explica a a ual dis ibuição da Acacia dealba a e a an ecipa os possí eis e ei os das al e ações climá icas nes a egião. Desen ol emos um modelo p edi i o combinado (MPC), em que os modelos de dis ibuição o am ajus ados usando subconjun os de a iá eis an e io men e classi icadas como impo an es à escala egional ou à escala local, endo os dois submodelos sido espacialmen e combinados de o ma a ob e as p ojeções inais. Es es modelos combinados o necem uma maio a iedade de po enciais espos as das espécies, p opo cionando p ojeções da dis ibuição e dinâmica de espécies mais in o ma i as que os modelos. Es a abo dagem oi complemen ada com análises de conec i idade da dis ibuições da espécie in aso a, pa a a alia os con li os a uais e u u os en e a sua dis ibuição e á eas de ele ado alo de conse ação. Os esul ados mos am que a dis ibuição de Acacia dealba a na á ea de es udo i á aumen a no u u o, se as p e isões de al e ações climá icas se con i ma em, e que es as al e ações no ambien e i ão aumen a a conec i idade da dis ibuição da espécie na á ea. Des a o ma, a a és da u ilização de MPCs e análises de conec i idade iden i icámos as á eas com maio p obabilidade de se em a e adas po es a in aso a, com pa icula des aque pa a a ede egional de á eas p o egidas, e é possí el apoia a p io ização de ecu sos pa a an ecipação e moni o ização dos impac os das EEIs. Es a abo dagem pode apoia medidas de p e enção de in asões nes a egião e pode ambém auxilia a ges ão ans on ei iça das in asões biológicas e de ou as ameaças a uais e u u as à in eg idade ecológica dos habi a s e à conse ação da sua biodi e sidade. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 6 Pala as – cha e: Acacia; Al e ações climá icas; Á eas p o egidas; Modelos p edi i os combinados; Modelos de dis ibuição de espécies. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 13 1. In oduc ion 1.1. Clima ic change and h ea s o biodi e si y The Ea h's biosphe e is acing an inc easing deg ada ion and loss o di e si y due o human ac i i ies and o he an h opogenic p essu e on he global en i onmen (Vi ousek e al., 1977). To unde s and hese changes i ’s impo an o be awa e o he concep o biodi e si y. Biodi e si y is he a iabili y among li ing o ganisms om all sou ces and he ecological complexes o which hey a e pa , which includes di e si y wi hin species, be ween species as well as o ecosys ems (Millenium Ecosys em Assessmen , 2005). Biodi e si y is undamen al o ecosys em s uc u e and unc ioning, and i suppo s he b oad spec um o goods and se ices ha humans de i e om na u al sys ems (S audinge e al., 2012). Human al e a ion o he global en i onmen has induced he six h ex inc ion e en in his o y o li e (Di zo and Ra en, 2003; Chapin e al., 2000) p omo ing changes in he global dis ibu ion o o ganisms (Chapin e al., 2000). Modi ica ions in species dis ibu ion can change bo h ecosys em p ocesses, and he esilience o ecosys ems o en i onmen al changes (S audinge e al., 2012, Chapin e al., 2000), wi h hough ul consequences in ecosys em se ices (Chapin e al., 2000). The majo causes o biodi e si y decline a e: land use changes, in asi e species, pollu ion, clima e change and o e exploi a ion esul ing om economic, sociopoli ical, cul u al, demog aphic, echnological, and o he indi ec d i e s (Millennium Ecosys em Assessmen , 2005). 1.2. Biological in asions as a d i e o biodi e si y change Biological in asions can be de ined as he in oduc ion and sp ead o exo ic o ganisms in o egions ou side o hei na i e ange (Méndez e al., 2011), and hey a e among he mos impo an human-d i en agen s o change in na u al en i onmen s wo ldwide (Na ščius e al., 2012). The a e o biological in asions is quickly inc easing las yea s (Essl e al., 2011) and in asi e species a e causing nega i e e ec s on bo h biodi e si y and human well-being (Essl e al., 2011). Impac s o in ade species include: modi ica ions in composi ion, s uc u e and unc ioning o he ecosys ems (Te e eai e al., 2013), habi a changes by elimina ion and/o educ ion o na i e species (Na ščius e al., 2012), bio ic homogeniza ion, loss o biodi e si y and loss o ecosys em se ices in many pa s o he wo ld (Cas o-Díez e al., 2011). Plan s in asions ha e become ecognized as a majo h ea o bo h na u al and human-exploi ed ecosys ems wo ldwide (Hobbs and Humph ies, 1995), as hey a ec FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 14 ecosys em s uc u e and unc ion (Lake and Leishman, 2004). Human ac i i ies helped (in en ionally and acciden ally) ha many plan species g ow ou side hei na i e anges, some o hese non-na i e plan s o e come se e al ba ie s, becoming in asi e (B adley e al. 2010). To a success ul in asion many di e en mechanisms a e in ol ed (Gulezian and Nybe g, 2010) and many di e en s ages/ecological p ocesses need o be o e come a di e en spa ial scales (B own e al., 2008; Richa dson e al., 2000). The i s ba ie co esponds o a majo geog aphical ba ie ha he plan o e comes h ough human help (Figu e 1, Ba ie A); a e , he species is called casual. The second ba ie co esponds o an en i onmen al il e , in ol ing bio ic and abio ic ac o s (Figu e 1; Ba ie B). Then, he species needs o o e come he ep oduc ion ba ie (Figu e 1; Ba ie C). A e o e coming hese h ee i s ba ie s, a axon is conside ed success ully na u alized. A species o popula ion can be called in asi e i o e comes he local/ egional dispe sal ba ie s (Figu e 1; Ba ie D), he en i onmen al ba ie (s) in human modi ied o alien-domina ed ege a ion (Figu e 1; Ba ie E), and he en i onmen al ba ie s in na u al o semi-na u al ege a ions (Figu e 1; Ba ie E; Richa dson e al., 2000). Mo eo e , he ex en o in asion by alien species is a unc ion o ecosys em-le el p ope ies (in asibili y), p opagule p essu e, cha ac e is ics o he in asi e species (in asi eness), and he p ope ies o he indi idual na i e species in he ecosys em (Wes phal e al., 2008). FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 15 Figu e 1 - S ages o in asion p ocess and hei ep esen a i e ba ie s ( om Richa dson e al., 2000). Alien in ade s a e o en “passenge s” p i ileged (Gae ne and Richa dson, 2011) by o he en i onmen al changes (Gae ne and Richa dson, 2011), such as ising empe a u e, inc eased a mosphe ic ca bon dioxide (CO2), ex eme e en s in p ecipi a ion, ni ogen (N) deposi ion, and dis u bances associa ed wi h changes in land use o land co e (B adley e al., 2009). Fo example, changes in clima ic condi ions migh di ec ly in luence he likelihood o alien species in a e i o y, and also a ec he na u aliza ion success (Wal he e al., 2009). Syne gic e ec s be ween alien in ade s and global changes di icul he indi idualiza ion be ween bo h he e ec s o alien species on he na i e ecosys ems, and he e ec s o he dis u bance ha lead o he ini ial plan in asion (Gae ne and Richa dson, 2011). 1.3. The speci ic case o in asi e acacias The in asi e alien plan s belonging o genus Acacia a e among he mos common in ade s in many pa s o he wo ld (González-Muñoz e al., 2012), a ec ing ecosys ems s uc u e and unc ioning wo ldwide (Lo enzo e al., 2010), and a e al eady p esen in sensi i e habi a s in Eu ope (coas al dunes, i e cou ses, na u al pa ks and biosphe e ese es; Lo enzo e al., 2010). The e o e in asions by Acacia genus Ba ie s: A B C DE F 1. Geog aphic 2. En i onmen al (local) 3. Rep oduc ion 4. Dispe sal 5. En i onmen al (Dis u bed habi a s) 6. En i onmen al (Na u al habi a s) Alien Na u alizedCasual In asi e FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 16 (wa les) ep esen s a h ea o na u al habi a s h ough compe i ion and eplacemen o na i e species, homogeniza ion o he communi y, and loss o na i e biodi e si y (Fuen es-Ramí ez e al., 2011). Some cha ac e is ics o his genus p omo e he h ea o na i e ecosys ems and species, pa icula ly he high coloniza ion capaci y, e.g. in places dis u bed by i e, ha es ing and o he ypes o human dis u bance (Fuen es- Ramí ez e al., 2011). Some key ai s a e p esen in his genus p omo ing he dominance in compe i i e in e ac ions wi h na i e species as: apid g ow h a es and abili y o ou -compe e na i e plan s; capaci y o accumula e high olumes o biomass; la ge, pe sis en seed banks, and he capaci y o ix ni ogen (Le Mai e e al., 2011). The e a e a leas eigh wa le species na u alized and po en ial in asi e plan s in he sou h o Eu ope (Lo enzo e al., 2010): A. dealba a, A. melanoxylon, A. e inodes, A. saligna, A. ka oo, A. pycnan ha, A. mea nsii, and A. longi olia. Th ee o he Acacia species: A. dealba a; A. melanoxylon, and A. longi olia a e he mos p oblema ic in ade s in F ance, Po ugal, Spain and I aly, pa icula ly in conse a ion a eas, being A. dealba a he mos widesp ead species (Lo enzo e al., 2010). 1.4. P o ec ed a eas Acco ding o In e na ional Union Conse a ion o Na u e (IUCN, 2008), a p o ec ed a ea is “a clea ly de ined geog aphical space, ecognized, and managed, h ough legal o o he e ec i e means, o achie e he long e m conse a ion o na u e wi h associa ed ecosys em se ices and cul u al alues” (h p://www.iucn.o g/). P o ec ed a eas comp ises mo e han 12.7% o he plane ’s land su ace (Geldmann e al., 2013) being an impo an ool and s a egy o main ain habi a in eg i y and species di e si y, ensu ing he main enance o na u al p ocesses ac oss landscapes and ecosys ems (h p://www.cbd.in /p o ec ed/o e iew/). The designa ion and main enance o p o ec ed a eas is he mos impo an and e icien conse a ion s a egy wo ldwide (Kleinbaue e al., 2010), none heless he impo ance o na u e ese es a ies depending he egions o he wo ld, ela ed o he deg ee o an h opic ans o ma ion o he e i o y (Pysek e al., 2002). Al hough p o ec ed a eas emain a c ucial key o global di e si y conse a ion s a egies, hey emain suscep ible o an h opogenic changes and consequences (Spea e al., 2013), especially o alien plan in asion. Alien in ade species a e e y p oblema ic o conse a ion alue a eas due o he po en ially h ea o he pe sis ence o na i e species (Uddin e al., 2013). The deg ee o a ese e in asion can be ela ed in some cases wi h he o al numbe o isi o s FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 17 (Pysek e al., 2002). Recen ly Pysek e al. (2002) shown ha na u e ese es all o e he wo ld a e in aded abou hal as o en as si es ou sides ese es. 1.5. Using models o achie e and an icipa e pa e ns o in asion Species dis ibu ion models (SDM) a e empi ical models ha ela e ield obse a ions o en i onmen al p edic o s a iables cen e ed on he s a is ical o heo e ical de i ed esponse su aces (Guisan and Thuille , 2005). The use o SDMs has g own in he las wo decades due o he ecognized applica ion alue in applied scien i ic ields as ecosys em managemen and biodi e si y conse a ion (Rod íguez-Reya e al., 2013). SDMs can also be applied in undamen al ecological ques ions as, such as he ecological impac s o clima e and land-use changes in biological in asions (Vicen e e al., 2011; Guisan and Thuille , 2005). Many scien i ic s udies in in asions applied SDMs as a ool, o example: o assess species in asion and p oli e a ion, o model species assemblages (biodi e si y, composi ion), o quan i y he en i onmen al niche o he species, o suppo app op ia e managemen plans o species eco e y (mapping sui able si es o species ein oduc ion), o suppo conse a ion planning and ese e selec ion (sugges ing su eyed si es o high po en ial o occu ence o a e species; Guisan and Thuille , 2005). Despi e many ecen applied pu poses o SDMs ela ed o clima e change and conse a ion planning, he use o hese ools in heo e ical ecology and e olu ion is e-eme ging (Guisan and Thuille , 2005). Acco ding o Nen wig e al. 2008, he main esea ch ields o biological in asions can be classi ied as: pa hways o biological in asions, ai s o a good in ade , pa e ns o in asion and in asibili y, ecological impac s o biological in asions, economy and socio-economy o biological in asions, and inally, p e en ion and managemen o biological in asions. One o he majo goals in biological in asion s udies is o unde s and which and how bio ic and abio ic ac o s cons ain he sp ead o in asi e species, de e mining cu en and u u e po en ial dis ibu ions (Wang and Jackson, 2011; Fa ashi e al., 2013). Resea ch e o s mus be di ec ed o ind ways o an icipa e in asions in o de o p o ec global biodi e si y om he e ec s o alien species (Vicen e e al., 2011; Jones, 2012). This goal can be pa ially achie ed de e mining he po en ial geog aphic ex en o in asions, equi ing p edic i e ools o assess he geog aphic dis ibu ion o in asi e species unde cu en and u u e en i onmen al condi ions (Vicen e e al., 2011; Fa ashi e al., 2013). These p edic ions can simpli y he ea ly de ec ion o in asi e species and maximize moni o ing e iciency (Jones, 2012). FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 18 Ongoing clima e and land-use changes a e o ecas ed o boos he in asion o alien species in se e al habi a s (Vicen e e al., 2011) being impo an o ha e a mul i-scale app oaches due o mos o he pa hways o species in oduc ions esul om human ac i i ies a se e al spa ial and empo al scales (Vicen e e al., 2011; Razgou e al., 2011). Al hough mos o he s udies using SDMs conside ed en i onmen al p edic o s a a single g ain size and ixed spa ial ex en , ac o s d i ing he dis ibu ion and abundance o o ganisms can ac a di e en scales (Vicen e e al., 2011; Guisan and Thuille , 2005). Fo example: opog aphy, geomo phology, human land-use o bio ic in e ac ions a e usually conside ed o ope a e a egional and local scales while clima e ope a e a he global scale (Vicen e e al., 2011; Eli h and Lea hwicK, 2009). The e a e se e al a ailable modelling ools (Thuille e al., 2009), being Biomod2 one o he newes and accu a e s a is ical package. Biomod2 is implemen ed in he R s a is ical so wa e (Geo ges and Thuille , 2013), allowing he calib a ion o ensemble o ecas ing species dis ibu ions, o e coming a ange o unce ain ies and assump ions in he me hodological models. This package allows he calib a ion o a model applying 10 di e en modelling echniques, and o o ecas he po en ial species dis ibu ion unde di e en en i onmen al condi ions (clima e and land use change scena ios; Thuille e al. 2009). Due o he complex na u e o species ange dynamics and he ela ed ac o s ac ing a mul iple spa ial scales, modelling species dis ibu ion is a challenging p ocedu e (Vicen e e al., 2011; Guisan and Thuille , 2005). 1.6. Objec i es In his wo k we applied a combined p edic i e modelling (CPM) amewo k o: (1) iden i y he a eas ha a e oday po en ially in aded by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula (Galicia and No h o Po ugal); and (2) o ecas he e ec s o clima e changes on he dis ibu ion o his species by p ojec ing he models o u u e condi ions. Ou a ionale was ha he applica ion o CMPs o p edic he cu en and u u e pa e ns o in asion can suppo measu es o p e en in asions in he s udy egion and suppo c oss-bo de managemen o p o ec impo an habi a s om in asion. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 19 2. Me hods 2.1. S udy a ea The s udy a ea is loca ed in he No hwes Ibe ian Peninsula (Fig. 2), co e ing he No h Po ugal and Galicia. This a ea p esen s a high plan biodi e si y, wi h mo e han 3000 species and subspecies, many o hem endemic o his egion (h p://www.biodi e sidade.eu/p /in o/p oxec o/). This egion is expe iencing a s ong en i onmen al dis u bance p omo ed by many ac o s, like in asi e species, c ea ing a se ious h ea o conse a ion o biodi e si y (h p://www.biodi e sidade.eu/p /in o/p oxec o/). Figu e 2 – Loca ion o he s udy a ea in Eu ope (a) and Ibe ian Peninsula (b) wi h digi al ele a ion model de ail (c). The a ea ex en is 35017 km², co e ing he ansi ion be ween he Medi e anean and A lan ic biogeog aphic egions (Rí as-Ma ínez e al., 2004). The s udy a ea is e y he e ogeneous, wi h ele a ion anging om he sea le el (wes ) o 2059 m in he eas e n moun ains, wi h he majo i e s, as Dou o, Tua, and Cenza, unning om eas o wes . The geology o he moun ains is o e whelmed by acid igneous and FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 20 me amo phic ocks ha o m acid soils (Roucoux e al., 2005). The s udy a ea p esen s a e y complex land co e (mo e han 11 classes o land co e , acco ding o he Eu opean En i onmen Agency) and a la ge ne wo k o p o ec ed si es: Pa que Nacional Peneda-Ge es, Pa que Na u al de Mon esinho and Pa que Na u al de Al ão in Po ugal and Se a Candán and Pena T e inca in Spain. Cu en ly his egion has a empe a e clima e (Roucoux e al., 2005) and ca ies he ull o ce o he wes e ly winds b inging cyclones om he A lan ic (Roucoux e al., 2005). The p ecipi a ion le els in he a ea a e he highes in he whole peninsula (Roucoux e al., 2005) (up o 3000mm pe yea in he eas e n moun ain summi s). The annual mean empe a u es anges om ca. 5°C o ca. 17°C, whe e he summe s a e cool (18-22ºC (Roucoux e al., 2005) and win e s a e mild and os - ee (10-13ºC a ound he coas ) (Roucoux e al., 2005). 2.2. Tes species and occu ence da a The es species is an alien in ade plan om he Leguminosae amily, Acacia dealba a Link (Fig. 3). This wa le species is o iginal om he sou heas o Aus alia and ound in a wide ange o di e en habi a s, om coas al o subalpine egions, and om high ain all o a id inland a eas, g owing in opical, sub opical and wa m empe a e egions (Lo enzo e al., 2009). Figu e 3 – Acacia dealba a Link. This species can be ound in a eas wi h o e 500 mm ain all, ypically a al i udes om 350-1000m abo e he sea le el (Lo enzo e al., 2009). In oduced in Eu ope in he 1820s (Ca ballei a and Reigosa, 1999), and plan ed as an o namen al plan in sou he n Eu ope, which o e ed a o able clima es o his de elopmen , wi h su icien sun exposu e an li le os s, becoming widely na u alized in his a ea (Lo enzo e al., FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 21 2009). In he Sou hwes Eu ope his species occu s in ipa ian zones, wa e cou ses and sunny edges o pinewoods, o on he sou h and wes acing slopes, whe e i o ms dense s ands ha choke in he na u al ege a ion and i o en in ades a ea unde in ensi e ag icul u e uses and u he away om he sea han o he Acacia species p esen in hese egions (Lo enzo e al, 2009). Acacia dealba a was in oduced in he Ibe ian Peninsula in he second hal o he 19 h cen u y (Sanz Elo za e al., 2004) and became a p oblema ic species in Po ugal and Spain, whe e i is h ea ening he na i e lo a and becoming a se ious en i onmen al p oblem. In addi ion o i s g ea colonizing capaci y i leads o a e y low co e ing o unde g ow h species caused by i s alellopa hic abili y (Vicen e e al, 2011). The occu ence da ase o Po ugal was collec ed om p e ious scien i ic wo ks (Vicen e e al., 2013) and Spain da a we e and p o ided by Di eccion Xe al de Conse acion da Na u eza da Xun a da Galicia. Sampling si es we e selec ed using a andom-s a i ied s a egy (Vicen e e al., 2013), employing clima e (mean annual empe a u e) and bed ock as s a i ying laye . Sampling uni s we e cells o 1km2 in which p esen s whe e collec ed be ween Ma ch and Ap il 2011. 2.3. Selec ion and classi ica ion o en i onmen al p edic o s We selec ed 53 p edic o s ha , by expe knowledge and acco ding o p e ious epo ing in scien i ic li e a u e, ope a e as de e minan s o ecology and dis ibu ion o he a ge species (see annex 1). To a oid co ela ion, we applied he Spea man ank co ela ion coe icien (non-pa ame ic es ), o selec he p edic o s wi h lowes co ela ion be ween each o he . We use he so wa e STATISTICA o selec he p edic o s and only hose wi h co ela ion lowe han 0.5 we e conside ed (Vicen e e al., 2011). The e o e we selec 17 en i onmen al p edic o s o calib a e he models (Table 1 desc ibes he p edic o s chosen). Clima e p edic o s we e ob ained using in o ma ion om he Wo lClim da abase in as e o ma wi h a spa ial esolu ion o 1km2 while he emaining en i onmen al p edic o s we e ob ained om de ailed en i onmen al maps. The en i onmen al p edic o s we e classi ied a p io i as “ egional” o “local” based on ecological heo y and hei spa ial scale o a ia ion using a s a is ical p edic o classi ica ion based on spa ial au oco ela ion (Vicen e e al., 2011).Using poin pa e n s a is ic (PPSA—spdep R package; a ailable a h p://c an. -p ojec .o g/web/ packages/spdep) we calcula ed Gea y’s c au oco ela ion measu e o all p edic o a iables o inc easing neighbou hood dis ances. A leas , we pe o med a hie a chical FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 22 clus e ing based on a Euclidean dis ance ma ix o all Gea y’s alues and he wo g oups in he inal classi ica ion ee we e e y well sepa a ed co esponding o p edic o s wi h local e sus egional in luence (Vicen e e al., 2011). FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 29 Figu e 5 – Possible combina ions o species dis ibu ions dynamics using combined models (Vicen e e al., 2011). The le e s A, B, C and D e e s o he di e en possible combina ions (A- egional and local sui abili y, B- only egional sui abili y. C- only local sui abili y and D- no sui abili y). 2.7. Cu en and u u e con lic s wi h p o ec ed a eas The models o he species we e used o o ecas he cu en and u u e species dis ibu ions and hei con lic wi h he p o ec ed a eas p esen in he s udy a ea, by spa ially o e lapping he po en ial dis ibu ion maps wi h a map o he p o ec ed a ea, co esponding o Na u a 2000 ne wo ks. Wi h his p ojec ion i 's possible o de e mine he po en ial h ea s o he p o ec ed a eas in he nea u u e, as well as he possible impac s ha may a ise om hem. By o e lapping he models o he species wi h a map o p o ec ed a eas (co esponding o Na u e Ne wo k 2000 – ICNF o Po ugal and o Espacios Na u ales P o egidos en España o Spain) he dis ibu ion maps o combined models allow a de ec ion o p esen and u u e in asion a eas. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 30 2.8. Connec i i y o he dis ibu ion o he in asi e species Fi s , we calcula ed he empo al changes in spa ial connec i i y o p edic ed sui able a eas o he es species ac oss he es a ea, using he connec i i y index de eloped by Randy e al., 2009 (Vicen e e al., 2013). We conside ed alue o species p esence: p obabili y o 1 o he esponse ype A, p obabili y o 0.5 o esponse ype B and C, and p obabili y o 0 o esponse ype D. The connec i i y index a ains a maximum alue o 1 when all he cells su ounding a ocal sui able cell a e also sui able (Vicen e e al., 2013). We quan i ied he changes in he spa ial ela ionship be ween p o ec ed a eas and he connec i i y o p edic ed a eas using he as e calcula o in A cGIS 9.3 (ESRI, 2010). Then, we analysed he end o connec i i y o he specie in space ( ull a ea and inside p o ec ed a eas) and ime (2020 and 2050). A las , a Spea man ank co ela ion es was used o iden i y signi ican ela ions be ween connec i i y and p o ec ion alue. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 31 3. Resul s 3.1. P edic ions and de e minan s o cu en dis ibu ions The po en ial dis ibu ion o he species Acacia dealba a unde cu en condi ions is illus a ed in Figu e 6. The combined p edic i e model was calib a ed using 17 en i onmen al a iables "a p io i" classi ied as " egional" and "local" (see Table 2). Figu e 6 - Spa ial p ojec ion o Acacia dealba a po en ial dis ibu ion using combined models, unde cu en condi ions (2000). . FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 32 Table 2 – En i onmen al p edic o s selec ed o calib a e he combined models. The pa ial egional model was calib a ed wi h 7 en i onmen al a iables, and an AUC alue o 0.849 was ob ained, while he pa ial local model was calib a ed wi h 10 en i onmen al a iables wi h a inal AUC alue o 0.887. F om he 7 egional en i onmen al p edic o s, he mos impo an o pa ial egional model calib a ion we e Minimum Tempe a u e o Coldes Mon h (MTCM, wi h a alue o 0.394), Annual P ecipi a ion (AP, wi h a alue o 0.282) and Tempe a u e Annual Range (TAR, wi h a alue o 0.238). Conside ing he pa ial local model, be ween he o al o 10 local p edic o s, he mos impo an we e Densi y o local oad ne wo k (DensRoad, wi h a alue o 0.392), pe cen age o a able land (pA L, wi h a alue o 0.173 )and pe cen age co e o mixed o es (pMixFo, wi h a alue o 0.197). Unde cu en clima e condi ions (2000), he po en ial pe cen age o p edic ed a ea o Acacia dealba a in a eas wi h egional and local sui abili y, ype A, was 18.3%. Fo a eas wi h only- egional sui abili y, ype B, he pe cen age o p edic ed a ea 27% while in he a eas wi h only local sui abili y, ype C, he pe cen age o p edic ed a ea was P edic o s Desc ip ion Scale o a ia ion AP Annual P ecipi a ion Regional Dis Ri Dis an e o main i e Regional GPP Mean g oss annual p ima y p oduc i i yMM Regional MTCM Min Tempe a u e o Coldes Mon Regional pCamb % o cambissoils (pe 1km2) Regional PS P ecipi a ion Seasonali y Regional TAR Tempe a u e Annual Range Regional DensRi Densi y o local hyd og aphic ne wo k Local DensRoad Densi y o local oad ne wo k Local Dis oad Dis ance o oads Local MPAR Mean pe ime e -a ea a io Local MSI Mean shape index Local PA L % co e o a able land (pe 1km2) Local PA S % co e o a i icial s ands (pe 1km2) Local pBiFo % co e o b oad-lea o es (pe 1km2) Local pCoFo % co e o coni e o es (pe 1km2) Local pMixFo % co e o mix o es (pe 1km2) Loca FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 33 14%. Finally, o a eas p edic ed o be unsui able o he species (unsui able local and egional condi ions), ype D, he pe cen age o p edic ed a ea was 40.7% (Table 3). Table 3 – P edic ed pe cen age o he s udy a ea occupied by each “occupancy ype” in he combined models, o he es species, unde cu en condi ions (2000). Occupancy ypes Pe cen age o p edic ed á ea Type A - egional and local sui abili y 18.3% Type B - only egional sui abili y 27.0% Type C - only local sui abili y 14.0% Type D - no sui abili y 40.7% 3.2. Fo ecas o u u e dis ibu ions The spa ial dis ibu ion o he es species unde u u e clima e condi ions o he yea 2020 is illus a ed in Figu e 7. The combined models o he A1 scena io (Fig. 7-b) and o he B2 scena io (Fig. 7-d) we e ob ained by o e lapping he spa ial p ojec ions o he local and he egional pa ial models. Figu e 7- Spa ial p ojec ions unde cu en condi ions (2000 - a) and u u e condi ions - 2020 A1 scena io (b) and Bb2 scena io (d) and 2050 A1 scena io (c) and B2 scena io (e). FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 34 Unde u u e condi ions (2020) conside ing A1 clima ic scena io, p edic ed a ea wi h local and egional sui abili y, ype A, was 22.3%.The pe cen age o a eas wi h only egional sui abili y, ype B, was 33.3%, and o a eas wi h only local sui abili y, ype C was 10%. Finally, unsui able egional and local a eas, ype D, we e p edic ed as 34.4%. Fo B2 clima ic scena io, ype A (bo h egional and local sui abili y) p edic ed a ea was 20%, ype B ( egional sui abili y) was 28.9%, o ype C (local sui abili y) was 12.3% and o ype D (bo h egional and local unsui abili y) was 38.8% (Table 4). Table 4 - P edic ed pe cen age o he s udy a ea occupied by each “occupancy ype” in he combined models, o he es species, unde u u e condi ions (2020), o he a1 and b2 scena ios. Pe cen age o p edic ed a ea Occupancy ypes A1 B1 Type A - egional and local sui abili y 22.3% 20.0% Type B - only egional sui abili y 33.3% 28.9% Type C - only local sui abili y 10.0% 12.3% Type D - no sui abili y 34.4% 38.8% A spa ial dis ibu ion o he species Acacia dealba a, unde u u e clima e condi ions o he yea 2050 is p esen ed in he igu e 7, whe e he combined model o he A1 scena io ( ig. 7- c) and o he B2 scena io( ig 7- e). In able 5 is p esen ed he es ima ed pe cen age o he s udy a ea occupied by each “occupancy ypes”. A eas p edic ed wi h bo h egional and local sui abili y (Type A) was 26,5% in A1 scena io and 22.8% in B2 scena io. The pe cen age o a ea p edic ed o be ype B (only egional sui abili y) was 46% in he a1 scena io and 36.4% in he b2 scena io, and he pe cen age o a eas p edic ed as ype C (only local sui abili y) was 5.8% in he a1 scena io and 9,4% in b2. A eas p edic ed as unsui able local and egional condi ions (Type D) was 21.7% and 31.4% in he a1 and b2 scena ios, espec i ely. Table 5 - P edic ed pe cen age o he s udy a ea occupied by each “occupancy ype” in he combined models, o he es species, unde u u e condi ions (2050), o he a1 and b2 scena ios. Pe cen age o p edic ed a ea Occupancy ypes A1 B2 Type A - egional and local sui abili y 26.5% 22.8% Type B - only egional sui abili y 46.0% 36.4% Type C - only local sui abili y 5.8% 9.4% Type D - no sui abili y 21.7% 31.4% FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 35 3.3. Range dynamics be ween yea s 2000 and 2050 Based on he po en ial dis ibu ion o he species p edic ed o he yea s o 2000, 2020 and 2050, we analysed he po en ial ange dynamics o he species Acacia dealba a be ween he ime pe iods: 2000-2020; 2000-2050; and 2020-2050. P edic ed changes in ange dynamics o he species Acacia dealba a be ween 2000-2020, 2000-2050 and 2020-2050 (unde clima e change scena ios) a e shown in Figu e 8. Fo he h ee ime pe iods, p edomina e dynamic ob ained was he esponse ype “no change” (see Table 6). In bo h scena ios, o he esponses ype “Imp o emen o he condi ions” and “Coloniza ion”, om he pe iods 2000-2020 o 2000-2050 an inc ease o he esponse was obse ed, and o he pe iod o 2020-2050 a dec ease was obse ed compa ed o he ime pe iod 2000-2050. Fo he esponse ype “De e io a ion o he condi ions”, o bo h scena ios, a dec ease end was obse ed om he pe iod o 2000-2020 o he pe iod o 2000-2050. F om ime pe iod 2000-2050 o 2020-2050 he ob ained alues we e he same. Finally, o he esponse ype “ex inc ion”, a dec easing o he alues was obse ed om he ime pe iod 2000-2020 o 2000-2050 and also om he ime pe iod o 2000-2050 o 2020-2050. Table 6 – Dynamic o en i onmen al sui abili y o he es species om combined modes, unde clima e change scena ios, o he pe iods 2000-2020, 2000-2050 and 2020-2050. Scena ios Dynamics No change Imp o emen De e io a ion Ex inc ion Coloniza ion A1B 2000-2020 77.9% 5.1% 1.0% 4.9% 11.1% 2000-2050 68.1% 8.6% 0.3% 2.0% 2.1% 2020-2050 80.0% 4.5% 0.3% 1.3% 13.9% B2 2000-2020 89.9% 2.7% 0.9% 4.3% 6.2% 2000-2050 79.1% 5.2% 0.6% 2.9% 12.2% 2020-2050 84.2% 3.4% 0.6% 2.2% 9.6% FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 36 Figu e 8 – Changes in he po en ial and dynamic o he es species be ween 2000-2020 ( o he A1 scena io – a- and he B2 scena io - d), 2000-2050 ( o he A1 scena io – b- and he B2 scena io - e) and 2020-2050 ( o he A1 scena io – c - and he B2 scena io - ). 3.4. Cu en and u u e con lic s wi h p o ec ed a eas In Table 7 we p esen he pe cen age o a ea o Na u a 2000 si es occupied by he di e en esponse ypes o he es species Acacia dealba a, o cu en (2000) and u u e (2020 and 2050) en i onmen al condi ions (Figu e 9).Fo he clima ic scena io A1, he pe cen age o p edic ed a ea wi h he esponse ypes A ( egional and local sui abili y o he species) wi hin Na u a 2000 si es inc eases in he yea 2020 and 2050. Conside ing p edic ed ype B ( egional sui abili y) and D (bo h unsui able egional and local) inside he Na u a 2000 si es, i ’s p edic ed no changes o he yea 2020 ollowed by an inc ease o p edic ed pe cen age o ype B ( egional sui abili y), and a dec ease in he ype D (bo h egional and local unsui abili y), o 2050. The pe cen age o occupancy o he ype C (local sui abili y) inside he Na u a 2000 si es dec eases bo h yea s. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 37 Fo he clima ic scena io B2, a dec ease o he pe cen age o occupancy o he species inside Na u a 2000 si es can be obse ed o he yea o 2020 o esponse ypes A ( egional and local sui abili y) and B ( egional sui abili y), ollowed by an inc ease in he yea 2050. An inc ease o he pe cen age is obse ed o esponse ypes C (local sui abili y) and D (bo h egional and local unsui abili y) o he yea 2020, while in 2050 he pe cen age dec eases. Table 7– Pe cen age o occupancy in Na u a 2000 o each occu ence ype ( egional and local sui abili y ( ype A), only egional sui abili y ( ype B), only local sui abili y ( ype C), and no sui abili y ( ype D), o he es species, o cu en (2000) and u u e (2000, 2050) condi ions. Pe cen age o p edic ed a ea Occupancy ype 2000 2020 a1 2020b2 2050a1 2050b2 Type A - egional and local sui abili y 9.0% 10.6% 8.6% 13.9% 11.0% Type B - only egional sui abili y 18.9% 18.9% 15.4% 34.5% 22.2% Type C - only local sui abili y 11.8% 10.2% 12.2% 6.9% 9.8% Type D - no sui abili y 60.3% 60.3% 63.8% 44.7% 57.0% Figu e 9- Spa ial p ojec ions o in asion wi hin Na u a 2000 si es (p o ec ed a eas o combined models, unde cu en (2000) and u u e condi ions (2020 and 2050). FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 38 P edomina e dynamic ype p edic ed by he combined models in each ime pe iod wi hin Na u a 2000, was “No change” (see Fig 10 and Table 8). Fo bo h clima ic scena ios, esponses ypes “de e io a ion o he condi ions” and “ex inc ion” dec eased om he ime pe iod 2000-2020 o 2000-2050 and om he ime pe iod 2000-2050 o 2020-2050. Conside ing he A1 clima ic scena io, he esponse ypes “Imp o emen o he condi ions” and “Coloniza ion”, om he ime pe iod 2000-2020 o 2000-2050, was inc easing and hen, om he ime pe iod 2000-2050 o 2020-2050, was dec eased. Conside ing he B2 clima ic scena io, om he ime pe iod 2000-2020 o 2000-2050 and 2000-2050 o 2020-2050, an inc easing was obse ed. Table 8 – Pe cen age o occupancy in Na u a 2000 o each dynamic ype (no change, imp o emen o he condi ions, de e io a ion o he condi ions, ex inc ion and coloniza ion), o he es species, o he pe iods 2000-2020, 2000-2050 and 2020-2050. Scena ios Dynamics No change Imp o emen De e io a ion Ex inc ion Coloniza ion A1B 2000-2020 82.2% 2.8% 1.2% 6.9% 6.9% 2000-2050 70.4% 5.5% 0.6% 3.9% 19.6% 2020-2050 79.3% 3.5% 0.1% 0.7% 16.4% B2 2000-2020 84.4% 1.3% 1.6% 8.1% 4.6% 2000-2050 84.3% 2.8% 0.8% 4.4% 7.7% 2020-2050 85.9% 2.9% 0.5% 2.0% 8.7% FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 45 4.4. CPMs, connec i i y o in ade dis ibu ions, and p o ec ed a eas Connec i i y analysis is use ul in ecology esea ch (Vicen e e al., 2013), and i elucida es pa e ns, he eby p o iding impo an insigh s in o he p ocesses ha d i e plan in asions in habi a s (Mino e al., 2009). Su aces wi h high connec i i y ep esen po en ial dispe sal co ido s and a eas wi h high connec i i y will be he ones whe e managemen may be less e ec i e, as local con ol measu es may be coun e ac ed by immig a ion and p opagule p essu e om neighbou ing in aded a eas (Vicen e e al., 2013). In his s udy, we used a connec i i y index based on he nea es neighbou s ha is conside ed a c ude connec i i y me ic, al hough simple o ob ain. Howe e , combined p edic i e models p ojec ions allowed o weigh he po en ial sui able a eas o he Acacia dealba a species by e alua ing and quan i ica ion o he a ailabili y o bo h egional condi ions and sui able local habi a . A eas wi h high connec i i y o he species will be he ones wi h less e ec i e managemen , as local con ol measu es may be coun e ac ed and p opagule p essu e om neighbo ing in aded a eas (Vicen e el al., 2013). Fo he es species, connec i i y was p edic ed o inc ease in he u u e. Al hough connec i i y o sui able a eas o he species inside p o ec ed a eas is lowe han ac oss he ull a eas (because p o ec ed a eas coincide wi h a eas o high ele a ion) he changes in en i onmen al condi ions p edic ed o he u u e may change he connec i i y o he species (Vicen e e al., 2013). In conclusion, he connec i i y analysis showed ha p o ec ed a eas will po en ially su e a high p essu e om Acacia dealba a unde u u e en i onmen al condi ions. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 46 5. Conclusions Using a combined p edic i e modeling app oach o o ecas he cu en and u u e dis ibu ion o Acacia dealba a in No hwes Ibe ian Peninsula, as well as assessing i s cu en and u u e con lic s wi h p o ec ed a eas, we conclude ha :  Models p edic ha he a ea wi h sui able condi ions o clima e ( egional) and habi a /landscape (local) will inc ease in he u u e, i he clima e change scena ios a e con i med;  The species cu en ly in ades na u e ese es in he s udy a ea, and models p edic ha he pe cen age o occupancy will inc ease in he u u e; and  Connec i i y analysis showed ha changes in u u e en i onmen may inc ease he connec i i y o Acacia dealba a dis ibu ion in he es a ea. F om an applied pe spec i e, ou esul s sugges ha :  Spa ial p edic ions o he models p o ide an impo an basis o he e icien con ol o his in asi e species in in aded a eas and o he ea ly de ec ion o new a eas o in asion;  CMP p ojec ions can suppo measu es o p e en in asions in his egion and p omo e c oss-bo de managemen o in asions; and  Connec i i y analyses a e impo an in in asion ecology, as hey can con ibu e o p io i ize managemen ac ions conside ing egional dispe sal co ido s and he p obabili y o ecoloniza ion a e local con ol measu es ha e been unde aken. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 47 Re e ences Ba ows, C.W., Mu phy-Ma iscal, M.L. (2012). Modelling impac s o clima e change on Joshua ees a hei sou he n bounda y: How scale impac s p edic ions. Biological Conse a ion 152, 29–36. Beaumon L.J., Gallaghe R.V., Downey P.O., (2009) Modelling he impac o Hie aciumspp. on p o ec ed a eas in Aus alia unde u u e clima es. Ecog aphy 32:757-764. Becke , T., Die z, H., Bille e , R.,Buschmann, H. (2005). Al i udinal dis ibu ion o alien plan species in he Swiss Alps. Pe spec i es in Plan Ecology, E olu ion and Sys ema ics 7, 173–183. B adley, B.A., Blumen hal, D.M., Wilco e, D.S., Ziska, L.H. (2010) . P edic ing plan in asions in an e a o global change. T ends in Ecology and E olu ion Vol.25 No.5, 311. B oennimann, O. and Guisan, A. (2008). P edic ing cu en and u u e biological in asions: bo h na i e and in aded anges ma e . Biol. Le . 4, 585–589. B own, K.A., Spec o , S, WU, W. (2008). Mul i-scale analysis o species in oduc ions: landscape and demog aphic models o imp o e managemen decisions abou non- na i e species. Jou nal o Applied Ecology, 45, 1639–1648. Ca ballei a, A., Reigosa, M.J. (1999). E ec s o na u al leacha es o Acacia dealba a Link in Galicia (NW Spain). Bo . Bull. Acad. Sin. 40, 87-92. Cas o – Díez, P., Gogoy, O., Saldanã, A., Richa dson, D.M. (2011). P edic ing in asi eness o Aus alian acacias on he basis o hei na i e clima ic a ini ies, li e his o y ai s and human use. Di e si y and Dis ibu ions 17, 934–945. Ca o d, J.A., Vesk, P.A., Whi e, M.D., Win le. B.A. (2011).Ho spo s o plan in asion p edic ed by p opagule p essu e and ecosys em cha ac e is ics. Di e si y and Dis ibu ions, (Di e si y Dis ib.) 17, 1099–1110. Chapin III, F.S., Za ale a, E.S., E ine , V.T., Naylo , R.N., Vi ousek, P.M., Reynolds, H.L.; Hoope , D.U., La o el, S., Sala, O.E., Hobbie, S.E., Mack, M.C., Díaz, S. (2000).Consequences o changing Biodi e si y. NATURE 405. C ossman, N.D., B yan, B.A., Cooke, D.A (2011).An in asi e plan and clima e change h ea index o weed isk managemen : In eg a ing habi a dis ibu ion pa e n and dispe sal p ocess. Ecological indica o s 11 1 8 3 – 1 9 8.Di zo, R., Ra en, P.H. (2003). Global S a e o Biodi e si y and Loss. Annu. Re . En i on. Resou . 28, 137–67. Eli h, J., Lea hwick, J.R. (2009). Species Dis ibu ion Models: Ecological Explana ion and P edic ion Ac oss Space and Time . Annu. Re . Ecol. E ol. Sys . 40, 677-697. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 48 Essl, F., Milasowszky, N., Thomas Di nböck, T. (2011). Plan in asions in empe a e o es s: Resis ance o epheme al phenomenon? Basic and Applied Ecology 12, 1–9. E angelis a, P.H., Kuma , S., S ohlg en, .j., Young, N.E. (2011). Assessing o es ulne abili y and he po en ial dis ibu ion o pine bee les unde cu en and u u e clima e scena ios in he In e io Wes o he US. Fo es Ecology and Managemen 262, 307–316. Fe nández, M., Hamil on, H., Al a ez, O., Guo, Q. (2012). Does adding mul i-scale clima ic a iabili y imp o e ou capaci y o explain niche ans e abili y in in asi e species? Ecological Modelling 246, 60– 67. Fische , D., Thomas, S.M., Niemi z, F., Reineking, B., Bie kuhnlein, C. (2011). P ojec ion o clima ic sui abili y o Aedes albopic us Skuse (Culicidae) in Eu ope unde clima e change condi ions. Global and Plane a y Change 78 , 54–64. Fuen es-Ramí ez, A., Paucha d, A., Ca ie es, L.A., Ga cía, R.A. (2011). Su i al and g ow h o Acacia dealba a s. na i e ees ac oss an in asion on in sou h-cen al Chile. Fo es Ecology and Managemen 261, 1003–1009. Galla do, B. and Ald idge, D.C. (2013). E alua ing he combined h ea o clima e change and biological in asions on endange ed species. Biological Conse a ion 160, 225–233. Gae ne , M., Richa dson, D.M., P i e , S.D.J. (2011). E ec s o Alien Plan s on Ecosys em S uc u e and Func ioning and Implica ions o Res o a ion: Insigh s om Th ee Deg aded Si es in Sou h A ican Fynbos. En i onmen al Managemen 48, 57– 69. Gaikwad, J., Wilson, P.D., Rangan han, S. (2011). Ecological niche modeling o cus oma y medicinal plan species used by Aus alian Abo igines o iden i y species- ich and cul u ally aluable a eas o conse a ion. Ecological Modelling 222, 3437– 3443. Gallien, L., Douze , R., P a e, S., Zimme mann, N.E., Thuille , W. (2012). In asi e species dis ibu ion models - how iola ing he equilib ium assump ion can c ea e new insigh s. Global Ecology and Biogeog aphy, (Global Ecol. Biogeog .) 21, 1126–1136. Geldmann, J., Ba nes; M., Coad, L., C aigie, I.D., Hockings, M., Bu gess, N.D. (2013). E ec i eness o e es ial p o ec ed a eas in educing habi a loss and popula ion declines. Biological Conse a ion 161, 230-238. Guisan, A., Thuille , W. (2005). P edic ing species dis ibu ion: o e ing mo e han simple habi a models. Ecology Le e s 8, 993–1009. Guisan, A., Zimme mann, N.E. (2000). P edic i e habi a dis ibu ion models in ecology. Ecological modelling 135, 147-186. FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 49 Gulezian, P.Z., Nybe g, D.W. (2010). Dis ibu ion o in asi e plan s in a spa ially s uc u ed u ban landscape. Landscape and U ban Planning 95, 161-168. González-Muñoz, N., Cos a-Teno io, M., Espiga es, T. (2012).In asion o alien Acacia dealba a on Spanish Que cus obu o es s: Impac on soils and ege a ion. Fo es Ecology and Managemen 269, 214–221. Fa ashi, A.; Kaboli, M. and Ka ami, M. (2013). P edic ing ange expansion o in asi e accoons in no he n I an using ENFA model a wo di e en scales. Ecological In o ma ics 15, 96–102. Hobbs, R.J., Humph ies, S.E. (1995). An In eg a ed App oach o he Ecology and Managemen o Plan In asions. Conse a ion Biology, Pages 761-770 Volume 9, No. 4. Jones, C.C. (2012). Challenges in p edic ing he u u e dis ibu ions o in asi e plan species. Fo es Ecology and Managemen 284, 69–77. Kleinbaue , I., Dullinge , S., Pe e seil, J., Essl, F. (2010). Clima e change migh d i e he in asi e ee Robinia pseudacacia in o na u e ese es and endange ed habi a s. Biological Conse a ion 143, 382–390. Lake, J.C., Leishman, M.R. (2004). In asion success o exo ic plan s in na u al ecosys ems: he ole o dis u bance, plan a ibu es and eedom om he bi o es. Biological Conse a ion Volume 117, Issue 2, Pages 215–226. Le Mai e, D.C., Gae ne , M., Ma chan e, E., Ens, E., Holmes, P.M., Paucha d, A., O’Fa ell, P.J., Roge s, A.M., Blancha d, R., Blignau , J., Richa dson, D.M. (2011). Impac s o in asi e Aus alian acacias: implica ions o managemen and es o a ion. Di e si y and Dis ibu ions, (Di e si y Dis ib) 17, 1015–1029. Liu, C., Whi e, M., Newell, G., G i ione (2013). Species dis ibu ion modelling o conse a ion planning in Vic o ia, Aus alia. Ecological Modelling 249, 68– 74. Lomba A., Pellissie , L., Randin , C., Vicen e, J. Mo ei a, F., Hon ado, J., Guisan, A. (2010). O e coming he a e species modelling pa adox: A no el hie a chical amewo k applied o an Ibe ian endemic plan . Biological Conse a ion 143, 2647– 2657 Lo enzo, P., González, L., Reigosa, M.J. (2010). The genus Acacia as in ade : he cha ac e is ic case o Acacia dealba a Link in Eu ope. Ann. Fo . Sci. 67, 101. Méndez, V., Llopis, I., Campos, D., Ho s hemkc, W. (2011). E ec o en i onmen al luc ua ions on in asion on s. Jou nal o Theo e ical Biology 281, 31–38. Millennium Ecosys em Assessmen (2005). Ecosys ems and human well-being: syn hesis epo . Island P ess, Washing on, DC. Online a h p://www.millenniumassessmen .o g/ FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 50 Mino , E.S., Tessel, S.M., Engelha d , K.A.M., Lookingbill, T.R. (2009). The ole o landscape connec i i y in assembling exo ic plan communi ies: A ne wo k analysis. Ecology. 90, 1802-1809. Moilanen, A., Nieminen, M., 2002. Simple connec i i y measu es in spa ial ecology. Ecology. 83, 1131–1145 Na ščius, A., Olenin, S., Zaiko, A., Minchin, D. (2012). Biological in asion impac assessmen sys em: F om idea o implemen a ion. Ecological In o ma ics 7, 46–51. Ohlemulle , R., Walke , S., Wilson, B. (2006). Local s egional ac o s as de e minan s o he in asibili y o indigenous o es agmen s by alien plan species. OIKOS 112: 493_/501. Pino, J., Fon , X., Ca bó, J., Ja é, M., Pallaès, L. (2005). La ge-scale co ela es o alien plan in asion in Ca alonia (NE o Spain). Biological Conse a ion 122, 339–350. Pysek, P., Ja osik, V., Kuce a, T. (2002). Pa e ns o in asion in empe a e na u e ese es. Biological Conse a ion 104, 13–24. Razgou , O., Hamme , J., Jones, G. (2011). Using mul i-scale modelling o p edic habi a sui abili y o species o conse a ion conce n: The g ey long-ea ed ba as a case s udy. Biological Conse a ion 144, 2922–2930. Richa dson, D.M., Pysek, P., Rejmanek. M., Ba bou , M.G., Pane a, F.D., C.J. Wes (2000). Na u aliza ion and in asion o alien plan s: concep s and de ini ions. Di e si y and Dis ibu ions 6, 93–107. Ri as-Ma ínez S., Penas, A., Díaz, T.E. (2004). Biogeog aphic Map o Eu ope. Ca og aphic Se ice, Uni e si y o Léon, Spain. Robinson, T. P., Van Klinken, R. K., Me e nich , G. (2010). Compa ison o al e na i e s a egies o in asi e species dis ibu ion modeling. Ecological Modelling 221, 2261– 2269. Rod íguez-Rey, M., Jiménez-Val e de, A., Ace edo, P. (2013). Species dis ibu ion models p edic ange expansion be e han chance bu no be e han a simple dispe sal model. Ecological Modelling 256, 1– 5. Roucoux, K.H., Ab eu, L., Shackle on, N.J., Tzedakis, P.C. (2005). The esponse o NW Ibe ian ege a ion o No h A lan ic clima e oscilla ions du ing he las 65ky . Fo es Ecology and Managemen 297, 84–93. Saume, I., Kowa ik. I. (2010). U ban i e s as dispe sal co ido s o p ima ily wind- dispe sed in asi e ee species. Landscape and U ban Planning 94, 244–249. Sanz-Elo za, M., Dana, E.D., Sob ino, E. (2004). A las de las plan as alóc onas in aso as de Espanã. Mad id: Di ección Gene al pa a la Biodi e sidad. S audinge , M.D., G imm, N.B., S aud , A., Ca e , L.S., Chapin III, F.S., Ka ei a, P., Ruckelshaus, M., S ein, B.A. (2012). Impac s o Clima e Change on Biodi e si y, FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 51 Ecosys ems, and Ecosys em Se ices: Technical Inpu o he 2013 Na ional Clima e Assessmen . Coope a i e Repo o he 2013 Na ional Clima e Assessmen . 296 p. Spea , P., Foxc o , L.C., Vezuidenhou , H., McGeoch, M.A. (2013). Human popula ion densi y explains alien species ichness in p o ec ed a eas. Biological Conse a ion 159, 137 - 147. Te e eai, F., Gae ne , M., Jacobs, S.M., Rica dson, D.M. (2013). Eucalyp us in asions in ipa ian o es s: E ec s on na i e ege a ion communi y di e si y s and s uc u e and composi ion. Fo es Ecology and Managemen 297, 84–93. Thuille , W., Richa dson, D.M., Pysek, P., Midgley, G.F., Hughes, G.O., Rouge, M. (2005).Niche-based modelling as a ool o p edic ing he isk o alien plan in asions a a global scale. Global Change Biology 11, 2234–2250. Thuille , W., La ou cade, B., Engle , R., A aujo, M.B. (2009). BIOMOD, a pla o m o ensemble o ecas ing o species dis ibu ions. Ecog aphy 21, 369-373. Thuille , W., Geo ges D., Engle R. (2013). Ensemble pla o m o species dis ibu ion modeling. T e howan,P.D., Robe son,M.P., McConnachie, A.J. (2011). Ecological niche modelling o an in asi e alien plan and i s po en ial biological con ol agen s. Sou h A ican Jou nal o Bo any 77, 137–146. Uddin, M.B., S einbaue , M.J., Jen sch, A., Mukul, S.A., Beie kuhnlein, C. (2013). Do en i onmen al a ibu es, dis u bances and p o ec ion egimes de e mine he dis ibu ion o exo ic plan species in Bangladesh o es ecosys em? Fo es Ecology and Managemen 303, 72–80. Vicen e, J., Al es, P., Randin, C., Guisan, A., Hon ado, J. (2010). Wha d i es in asibili y? A mul i model in e ence es and spa ial modelling o alien plan species ichness pa e ns in no he n Po ugal. Ecog aphy 33: 1081-1092. Vicen e, J., Randin, C.F., Gonçal es, J., Me ze . M.J., Lomba, A., Hon ado. J., Guisan, A. (2011). Whe e will con lic s be ween alien and a e species occu a e clima e and land-use change? A es wi h a no el combined modelling app oach. Biol In asions 13, 1209–1227. Vicen e, J.R., Fe nandes, R.F., Randin, C.F., B oennimann, O., Gonçal es, J., Ma cos, B., Pôças, I., Al es, P., Guisan, A., Hon ado, J.P (2013). Will clima e change d i e alien in asi e plan s in o a eas o high p o ec ion alue? An imp o ed model-based egional assessmen o p io i ize he managemen o in asions (in p ess) a Vicen e J., Pe ei a, H.M., Randin, C.F., Gonçal es, J., Lomba, A., Al es, P., Me zge , M., Ceza , J., Guisan, A. and Hon ado, J. (2013): A mul i-model dissimila i y amewo k FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 52 o e alua e he e ec s o en i onmen and dispe sal on alien plan in asions. Pe spec i es in Plan Ecology, E olu ion and Sys ema ics. (in p ess) b Vicen e, J., Randin, C.F., Po ie , J., Lomba, A., Gonçal es, J., B oennimann, O., Hon ado, J. and Guisan, A. (2013). A amewo k o assessing he scale o in luence o en i onmen al ac o s on ecological pa e ns. Me hods in Ecology and E olu ion (in e ision (?) c Vicen e, J.R., Pin o, A.T., A aujo, M.B., Ve bu g, P.H. Lomba, A., Randin, C.F., Guisan, A., Hon ado, J.P. (2013). Using Li e S a egies o Explo e he Vulne abili y o Ecosys em Se ices o In asion by Alien Plan s. Ecosys ems 16: 678–693 d Vi ousek, P.M., Mooney, H.A., Lubcenco, J., Melillo, J.M. (1977). Human domina ion o Ea h’s ecosys ems. Science 277, 494-499. Wal he , G., Roques, A., Hulme, P. E., Sykes, M. T., Pe Pysek, P., Kuhn, I., Zobel, M., Bache , S., Bo a-Duka , Z., Bugmann, H., Czucz, B., Daube , J., Hickle , T., Ja osik, V., Kenis, M., Klo z. S., Minchin, D., Moo a, M., Nen wig, W., O , J., Pano , V. E., Reineking, B., Robine , C., Semenchenko, V., Sola z, W., Thuille , W., Vila, M., Vohland, K., Se ele, J. (2009). Alien species in a wa me wo ld: isks and oppo uni ies. T ends in Ecology and E olu ion Vol.24 No.12. Wang, L., Jackson, D.A. (2011). Modeling he es ablishmen o in asi e species: habi a and bio ic in e ac ions in luencing he es ablishmen o By ho ephes longimanus. Biol In asions 13, 2499–2512. Wes phal, M.I., B owne, M., MacKinnon, K., Noble, I. (2008). The link be ween in e na ional ade and he global dis ibu ion o in asi e alien species. Biol In asions, 10:391–398. In e ne e e ences: CBD, Con en ion ON Biological Di e si y (1992), Websi e: www.cbd.in , assessed in: 24/04/13. IUCN, he In e na ional Union o Conse a ion o Na u e (2000), Websi e: www.iucn.o g, assessed in: 03/05/13. Biodi e sidade ameazada, Websi e: www.biodi e sidade.eu, assessed in: 21/02/13.Deli e ing Alien In asi e Species In en o ies o Eu ope, Websi e: h p://www.eu ope-aliens.o g/pd /Acacia_dealba a.pd , assessed in: 02/03!13 FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 53 Annex Annex 1: En i onmen al p edic o s selec o apply he Spea man ank co ela ion coe icien . Va iable ype P edic o s Desc ip ion Re e ences ANT Annual Mean Tempe a u e B oennimann and Guisan, 2008 Clima e AP Annual P ecipi a ion Ba ows and Mu phy- Maiscal, 2012 I Iso he mali y Cas o-Díez e al., 2011 MDR Mean Diu nal Range C ossman and Cooke, 2011 MTCM Min Tempe a u e o Coldes Mon h E angelis a e al., 2011 MTCQ Min Tempe a u e o Coldes Qau e Fé nandez e al., 2012 ; Fische e al., MTDQ Mean Tempe a u e o D ies Qua e Fische e al., 2011 MTWM Max Tempe a u e o Wa mes Mon h Gaikwad e al., 2011; Gallien e al., 2012 MTWaQ Mean Tempe a u e o Wa mes Qua e Godoy e al., 2008 MTWeQ Mean Tempe a u e o We es Qua e Ohlemulle e al., 2006; Pino e al., 2005 PCQ P ecipi a ion o Coldes Qua e T e howan e al., 2011; Thuille e al. 2005 PDM P ecipi a ion o D ies Mon h Vicen e e al., 2010; Vicen e e al., 2011 PDQ P ecipi a ion o D ies Qua e Vicen e e al., 2013a; Vicen e e al., 2013b PS P ecipi a ion Seasonali y Vicen e e al., 2013c; Vicen e e al., 2013d PWM P ecipi a ion o We es Mon h Vicen e e al., 2013ª PWaQ P ecipi a ion o Wa mes Qua e PWeQ P ecipi a ion o We es Qua e TAR Tempe a u e Annual Range TS Tempe a u e Seasonali y Land co e pA L % co e o a able land (pe 1 km²) (Landscape pA S % co e o a i icial s ands (pe 1 km²) Chy y e al., 2008 composi ion) pBiFo % co e o b oad-lea o es (pe 1 km²) h p://www.eu ope-aliens.o g pCoFo % co e o coni e o es (pe 1 km²) g/pd /Acacia_dealba a.pd pHAa % co e o he e ogeneous ag icul u al a eas (pe 1 km²) Ohlemulle e al., 2006 pMiFo % co e o mix o es (pe 1 km²) Pino e al., 2005 pOs % co e o open spacie wi h small o no ege a ion (pe 1 km²) Vicen e e al., 2013a pPas % co e o pas u es land (pe 1 km²) Vicen e e al., 2013d pPeC % co e o pe manen c ops (pe 1 km²) pSH % co e o sc ub and/o he baceous ege a ion associa ions (pe 1 km²) Pwa % co e o wa e (pe 1 km²) T anspo DensRail Densi y o local ail ne wo k Ca o d e al. 2011 and DensRi Densi y o local hyd og aphic ne wo k Vicen e e al., 2010; Vicen e e al., 2011 hyd og aphic DensRoad Densi y o local oad ne wo k Vicen e e al., 2013a; Vicen e e al., 2013b ne wo k Dis Rail Dis ance o ails Vicen e e al., 2013c; Vicen e e al., 2013d Dis Ri Dis ance o i e s Dis Road Dis ance o oads Soil pCamb % o cambissoils (pe 1 km²) FCUP Using models and connec i i y analysis o p edic he cu en and u u e pa e ns o in asion by Sil e -wa le (Acacia dealba a Link) in No hwes Ibe ian Peninsula 54 pFlu i % o lu issoils (pe 1 km²) pHis % o his ossoils (pe 1 km²) pLi % o li hossoils (pe 1 km²) pLu i % o lu issoils (pe 1 km²) Vicen e e al., 2013a pPodz % o podzossoils (pe 1 km²) Vicen e e al., 2013b pRank % o anke ssoils (pe 1 km²) pRego % o egossoils (pe 1 km²) Lanscape Ed Edge Densi y S uc u e GPP Mean g oss annual p ima y p oduc i i y Lomba e al., 2010 MPAR Mean pe ime e -a ea a io Vicen e e al., 2010 MPFD Mean pa ch ac al dimensions Vicen e e al., 2013a; Vicen e e al., 2013b MPS Mean pa ch size Vicen e e al., 2013c; Vicen e e al., 2013d MSI Mean shape index Vicen e e al., 2013ª NumP Numbe o pa ches PSSD Pa ch size s anda de ia ion TE To al Edge