Old World megadroughts and pluvials during the Common Era

CLIMATOLOGY 2015 © The Au ho s, some igh s ese ed; exclusi e licensee Ame ican Associa ion o he Ad ancemen o Science. Dis ibu ed unde a C ea i e Commons A ibu ion NonComme cial License 4.0 (CC BY-NC). 10.1126/sciad .1500561 Old Wo ld megad ough s and plu ials du ing he Common E a Edwa d R. Cook, 1 * Richa d Seage , 1 Yochanan Kushni , 1 Kei h R. B i a, 2 Ul Bün gen, 3 Da id F ank, 3 Paul J. K usic, 4 Willy Tegel, 5 Ge a d an de Sch ie , 6 Laia And eu-Hayles, 1 Mike Baillie, 7 Claudia Bai inge , 8 Niels Bleiche , 9 Niels Bonde, 8 Da id B own, 7 Ma co Ca e , 10 Richa d Coope , 2 Ka a ina Ču a , 11 Ch is oph Di ma , 12 Jan Espe , 13 Ca ol G iggs, 14 Bjö n Gunna son, 15 Bjö n Gün he , 16 Emilia Gu ie ez, 17 K is o Haneca, 18 Samuli Helama, 19 F anz He zig, 20 Ka l-Uwe Heussne , 21 Ju a Ho mann, 22 Pa el Janda, 23 Raymond Kon ic, 24 Nesibe Köse, 25 TomášKyncl, 26 Tom Le anič, 27 Hans Linde holm, 28 S u Manning, 14 Thomas M. Mel in, 2 Daniel Miles, 29 Bu kha d Neuwi h, 30 Ku Nicolussi, 31 Paola Nola, 32 Momchil Panayo o , 33 Ionel Popa, 34 And eas Ro he, 35 K is ina Se igen, 28 And ea Seim, 28 Helene S a a, 36 Mi osla S oboda, 23 Te je Thun, 36 Mau i Timonen, 19 Ramzi Touchan, 37 Volodymy T o siuk, 23 Vale ie T oue , 37 Felix Walde , 9 Tomasz Ważny, 37,38 Rob Wilson, 39 Ch is ian Zang 40 Clima e model p ojec ions sugges widesp ead d ying in he Medi e anean Basin and we ing in Fennoscandia in he coming decades la gely as a consequence o g eenhouse gas o cing o clima e. To place hese and o he “Old Wo ld”clima e p ojec ions in o his o ical pe spec i e based on mo e comple e es ima es o na u al hyd oclima ic a iabili y, we ha e de eloped he “Old Wo ld D ough A las”(OWDA), a se o yea - o-yea maps o ee- ing econs uc ed summe we ness and d yness o e Eu ope and he Medi e anean Basin du ing he Common E a. The OWDA ma ches his o ical accoun s o se e e d ough and we ness wi h a spa ial comple eness no p e iously a ailable. In addi ion, megad ough s econs uc ed o e no h-cen al Eu ope in he 11 h and mid-15 h cen u ies ein o ce o he e idence om No h Ame ica and Asia ha d ough s we e mo e se e e, ex ensi e, and p olonged o e No he n Hemisphe e land a eas be o e he 20 h cen u y, wi h an inadequa e unde s anding o hei causes. The OWDA p o ides new da a o de e mine he causes o Old Wo ld d ough and we ness and a ibu e pas clima e a iabili y o o ced and/o in e nal a iabili y. INTRODUCTION Ins umen al obse a ions show ha he Medi e anean egion has been d ying since he 1970s (1). Coupled clima e model simula ions om Phase 5 o he Coupled Model In e compa ison P ojec a chi e, pe - o med as pa o he In e go e nmen al Panel on Clima e Change Wo king G oup 1 Fi h Assessmen Repo (2), u he sugges ha he Medi e anean will unde go se e e and widesp ead d ying in he coming decades as a consequence o ising g eenhouse gases (GHGs) (3). How- e e , he ecen d ying end has been a iably a ibu ed o a mix u e o na u al clima e a iabili y and an h opogenically o ced change (4,5), wi h simila challenges in unde s anding bo h we ing and d ying ends in o he pa s o Eu ope and a he a ield (6,7). This unce - ain y a ises because he ins umen al clima e eco ds used in a ibu ion s udies a e ela i ely sho , a e likely o be con ounded by unspeci ied le els o GHG o cing, and a e unlikely o include he ull ange o na u al a iabili y. In addi ion, de e mining he comple e ange o pos- sible u u eclima es a es omclima emodels equi esknowingbo h he esponse o changes in an h opogenic o cing and he po en ial ange o na u al a iabili y ha willunde liei .Clima emodelsa ees- sen ial ools o diagnosing ongoing and u u e clima e change, making i necessa y o de e mine how well hey ep esen each o hese d i e s o clima e a iabili y. An ex ended eco d o na u al hyd oclima ic a iabili y om ee- ing econs uc ed d ough and we ness be o e he ins umen al e a is a c ucial es ima e o pas clima e a iabili y caused by bo h o ced a - iabili y and in e nal a iabili y, which is ideal o assessing he ue ange o hyd oclima e a iabili y in he p eindus ial pas and he de- g ee o which clima e models p ope ly ep esen i . The No h Ame ican D ough A las (NADA), eleased in 2004 (8), made clea ha d ough s o 1 Lamon -Dohe y Ea h Obse a o y o Columbia Uni e si y, Palisades, NY 10964, USA. 2 Clima ic Resea ch Uni , Uni e si y o Eas Anglia, No wich NR4 7TJ, UK. 3 Swiss Fede al Re- sea ch Ins i u e WSL, Bi mensdo 8903, Swi ze land. 4 Na a ino En i onmen al Obse a o y, Messinia 24001, G eece. 5 Ins i u e o Fo es G ow h (IWW), Uni e si y o F eibu g, F eibu g 79106, Ge many. 6 Royal Ne he lands Me eo ological Ins i u e (KNMI), De Bil 3730, The Ne he lands. 7 Paleoecology Cen e , Queens Uni e si y, Bel as BT7 1NN, No he n I eland. 8 Na ional Museum o Denma k, Copenhagen DK-1220, Denma k. 9 Compe ence Cen e o Unde wa e A chae- ology and Dend och onology, O ice o U banism, Ci y o Zü ich, Zü ich 8008, Swi ze land. 10 TeSAF Depa men , Uni e si à degli S udi di Pado a, Ag ipolis, Legna o I-35020, I aly. 11 Bio echnical Facul y, Uni e si y o Ljubljana, Ljubljana SI-1000, Slo enia. 12 En i onmen al Resea ch and Educa ion (UFB), Mis elbach 95511, Ge many. 13 Depa men o Geog aphy, Johannes Gu enbe g Uni e si y, Mainz 55099, Ge many. 14 Co nell T ee Ring Labo a o y, Co nell Uni e si y, I haca, NY 14853, USA. 15 Depa men o Physical Geog aphy, S ockholm Uni e si y, S ockholm SE-106, Sweden. 16 Technische Uni e si ä D esden, Tha and D-01737, Ge many. 17 Depa men o Ecology, Uni e si y o Ba celona, Ba celona 08028, Spain. 18 Flande s He i age Agency, B ussels 1210, Belgium. 19 Na u al Resou ces Ins i u e Finland, Ro aniemi FI-96301, Finland. 20 Ba a ian S a e Depa men o Cul u al He i age, Thie haup en 86672, Ge many. 21 Ge man A chaeological Ins i u e (DAI), Be lin 14195, Ge many. 22 Jah inglabo Ho mann, Nü ingen 72622, Ge many. 23 Depa men o Fo es Ecology, Czech Uni e si y o Li e Sciences, P ague 16521, Czech Republic. 24 Labo Dend on, Basel 4057, Swi ze land. 25 Facul y o Fo es y, Is anbul Uni e si y, Bahcekoy, Sa iye 34473, Is anbul, Tu key. 26 Mo a ian Dend o-Labo , B no 61600, Czech Republic. 27 Slo enian Fo es y Ins i u e, Ljubljana SI-1000, Slo enia. 28 Depa men o Ea h Sciences, Go henbu g Uni e si y, Go henbu g SE-405, Sweden. 29 Ox o d Dend och onology Labo a o y, Ox o d Uni e si y, Ox o d RG4 7TX, UK. 30 DeLaWi –T ee Ring Analyses, Windeck D-51570, Ge many. 31 Ins i u ü Geog aphie, Uni e si ä Innsb uck, Innsb uck A-6020, Aus ia. 32 Dipa imen o di Scienze della Te a e dell’Ambien e, Uni e si à degli S udi di Pa ia, Pa ia 27100, I aly. 33 Dend ology Depa men , Uni e si y o Fo es y, Sophia 1756, Bulga ia. 34 Fo es Resea ch and Managemen Ins i u e, Calea Buco inei, Campulung Moldo enesc 725100, Romania. 35 Facul y o Fo es y, Uni- e si y o Applied Sciences Weihens ephan-T iesdo , F eising 85354, Ge many. 36 NTNU Uni e si y Museum, No wegian Uni e si y o Science and Technology, T ondheim 7012, No way. 37 Labo a o y o T ee-Ring Resea ch, Uni e si y o A izona, Tucson, AZ 85721, USA. 38 Nicolaus Cope nicus Uni e si y, To un 87-100, Poland. 39 School o Geog aphy and Geosciences, Uni e si y o S . And ews, S . And ews KY16 9AL, Sco land. 40 Ecoclima ology, Technische Uni e si ä München, F eising 85354, Ge many. *Co esponding au ho . E-mail: [email p o ec ed] RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 1o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om a se e i y and a longe i y no seen in he 19 h and 20 h cen u ies oc- cu ed mo e equen ly in ea lie cen u ies, and his has spa ked a ange o e o s o de e mine whe he clima e models can simula e such e en s (9,10). He e, we p esen he Old Wo ld D ough A las (OWDA), a new ee ing–based ield econs uc ion o pas d ough s and plu ials o e Eu ope, No h A ica, and he Middle Eas spanning he Common E a (CE), which will acili a e simila ad ances in unde s anding and modeling hyd oclima e a iabili y in he Old Wo ld. RESULTS The OWDA is a se o yea -by-yea maps o econs uc ed summe season [June-July-Augus (JJA)] sel -calib a ing Palme D ough Se- e i y Index (scPDSI) (11) on a 5414-poin hal -deg ee longi ude- by-la i ude g id (Fig. 1). The JJA scPDSI e lec s sp ing-summe soil mois u e condi ions and is he same season econs uc ed in he NADA (8) and Monsoon Asia D ough A las (MADA) (12). The OWDA is based on he same eg ession-based clima e ield econs uc ion me hod used o p oduce he NADA and he MADA. These common p ope ies allow o di ec compa isons o he d ough a lases. The OWDA also p o ides a longe and mo e spa ially comple e econs uc- ion o hyd oclima ic a iabili y o e he Old Wo ld compa ed o p e- ious es ima es based on ins umen al, his o ical, and na u al eco ds (Supplemen a y Ma e ials). In addi ion o he ee- ing da a om li ing ees, we assembled an excep ional amoun o his o ical, a chaeological, and sub ossil ee- ing da a o ex end many ch onologies back a millen- nium o mo e in ime (Supplemen a y Ma e ials). This allows us o make in o med s a emen s abou he p ope ies o hyd oclima ic a i- abili y du ing medie al imes in he Old Wo ld. Dend oclima ic econs uc ions a e adi ionally alida ed h ough compa isons o ee ing–based clima e es ima es wi h 20 h-cen u y ins umen al da a ha ha e no been used o calib a ion (8,12). This was pe o med as well o he OWDA by wi hholding 1901–1928 in- s umen al scPDSI da a om he 1928–1978 calib a ion pe iod o al- ida ion es ing (Supplemen a y Ma e ials).Howe e , he as amoun o his o ical clima e in o ma ion om Eu ope ex ending back many cen- u ies (13) allows o a mo e ex ensi e compa isons o OWDA econ- s uc ions wi h eco ded d ough s and plu ials. We compa e he OWDA o se e al his o ically documen ed ex eme hyd oclima ic e en s (wi h addi ional compa isons shown in Supplemen a y Ma e ials), wi h em- phasis on hose e en s ha e lec he sp ing-summe mois u e condi- ions mos ele an o he OWDA. The g ea d ough o 1921: On he basis o an e alua ion o ins u- men al clima e da a no used o calib a e he OWDA, his d ough was desc ibed as “a yea o unp eceden edly small ain all”o e la ge pa s o he B i ish Isles, wi h he wo s de ici occu ing in sou h- eas e n England whe e London may ha e expe ienced i s d ies yea since 1774 (14). The OWDA map o 1921 (Fig. 2A) shows he mos ex eme d ough occu ing in sou he n England, in acco dance wi h he epo . I also shows se ious d ough ex ending o e mos o cen al Eu ope bu less d y condi ions in peninsula I aly, a esul consis en wi h o he epo s om Eu ope (15). The g ea d ough o 1893: E alua ed again using ins umen al clima e da a, “… he absence o ain was phenomenal”du ing he pe iod Ma ch o June, wi h a ain all de ici g adien om 30 o 50% o no mal o e sou he n England o 50 o 90% o no mal o e Sco land and I eland (16). The 1893 OWDA map (Fig. 2B) shows a simila no h-sou h g adien o ain all de ici o e he B i ish Isles and also shows he d ough ex- ending o e con inen al Eu ope, as alluded o in his epo . The I ish amine o 1740–1741: This e en has been a ibu ed o unusually low win e and sp ing empe a u es in 1740, esul ing in c op ailu es and subsequen amine (17). The OWDA is no well sui ed o de e mining empe a u e anomalies because i p ima ily e lec s wa m season hyd oclima e. Howe e , clima e ield econs uc ions o seasonal p ecipi a ion om documen a y and ea ly ins umen al da a (18)indi- ca e ha sp ing-summe ain all o e I eland in 1741 was well below no mal ela i e o he mode n a e age. D ough o e I eland may he e- o e ha e con ibu ed o he se e i y o he amine h ough i s nega i e impac on ood p oduc ion in 1741. The OWDA map o 1741 (Fig. 2C) indica es se e e d ough o e I eland ha also ex ended o e England and Wales, consis en wi h p e iously epo ed eco d ain all de ici s (19). The g ea d ough s o 1616 and 1540: D ough s o e Czech lands ha e been econs uc ed om documen a y eco ds since 1090 CE, wi h i e “ou s anding d ough e en s”(1540, 1590, 1616, 1718, and 1719) desc ibed (20). We highligh he 1616 and 1540 d ough s he e bu no e ha all i e ou s anding d ough e en s o e Czech lands a e well exp essed in he OWDA (Supplemen a y Ma e ials). The 1616 d ough began in he sp ing and con inued h oughou he summe wi h “g ea hea , d ied-up i e s”and a ma k on a “hunge s one”on he Elbe Ri e (20).Simila condi ionsalsoex endedin oSwi ze landandGe many. The OWDA map o 1616 (Fig. 2D) indica es se e e o ex eme d ough o e cen al and eas e n Eu ope, much he same as indica ed on Czech documen a y eco ds. The 1540 d ough has been desc ibed as a “wo s - case”e en in e ms o bo h p ecipi a ion de ici and excessi e wa m h 0˚ 0˚ 20˚ 20˚ 40˚ 40˚ 40˚ 40˚ 60˚ 60˚ 0 200 400 600 800 1000 1200 1400 1600 1800 Ch onology s a yea Fig. 1. Map o he JJA scPDSI a ge ield (small black g id poin s) and he 106 ch onology ee- ing ne wo k used o econs uc ion. The e a e 5414 hal -deg ee scPDSI g id poin s. The OWDA ee- ing ne wo k ( illed iangles shaded by s a yea ) illus a es he easonably uni o m co e age o ch onologies ac oss he domain, excep o Russia. RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 2o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om o e cen al Eu ope ela i e o d ough s ha ha e occu ed o e he las cen u y (21). The OWDA map o 1540 (Fig. 2E) shows he widesp ead occu ence o mode a e o ex eme d ough s in cen al Eu ope, consis en wi h epo s (18,19), bu no a wo s -case e en e en among he examples shown he e. The g ea Eu opean amine o 1315–1317: This is one o he mos amous his o ical ca as ophes in la e medie al Eu opean his o y, a amine caused by a mul iyea pe iod o excessi e we ness ha made ood p oduc ion nea ly impossible (22). This plu ial ac ually began in 1314, bu he ollowing yea was conside ed he mos ca as ophic yea , when “… he e was uni e sal ailu e o c ops in 1315 in mos i no all lands o Eu ope om he Py enees o Sla ic egions, om Sco land o I aly”(22). The 1315 OWDA map (Fig. 2F) co esponds well o his pa e n o excessi e we ness and c op ailu e and also shows he d ie condi ions in sou he n I aly, “… which escaped he g ea c isis ha aged no h o he Alps”(22). The OWDA now p o ides s ong e i- dence o he hyd oclima ic condi ions o e Eu ope ha we e espon- sible o his ca as ophe and shows i s yea ly p og ession om 1314 o 1317 in ull de ail (Supplemen a y Ma e ials). The spa ial and empo al de ails o he OWDA econs uc ions a e a ma ked imp o emen in ou knowledge o changing mois u e condi- ions ac oss he Old Wo ld du ing he Common E a (Supplemen a y Ma e ials). These ew illus a ions alone show ha he OWDA now p o ides a new basis o he s udy o p e- and pos indus ial hyd o- clima ic a iabili ies and hei possible causes and consequences on he Old Wo ld. The “Medie al Clima e Anomaly”(MCA) (23)and “Li le Ice Age”(LIA) (24) pe iods a e pa icula ly in e es ing, as well as con o e sial, because o unce ain y o e hei condi ions ela i e o he mode n pe iod and he deg ee o spa ial he e ogenei y in p eindus ial clima e (25).Whiledeba eabou heexac cha ac e is ics o empe a- u e a iabili y du ing he MCA and LIA con inues, hyd oclima ic con- di ions emain e en mo e poo ly quan i ied. The NADA made clea ha he MCA pe iod in No h Ame ica expe ienced se e e and p o- longed “megad ough s”(26,27), wi h addi ional paleoclima e e idence o ele a ed a idi y documen ed elsewhe e in he No he n Hemisphe e (NH) (28). Ac oss Eu ope, glacie s ad anced du ing he LIA, es i ying o a gene ally coole clima e (24), bu he hyd oclima e signa u es o he MCA and LIA in he Old Wo ld ha e been poo ly cons ained. He e, we use he OWDA o he i s de ailed compa ison o he spa ial pa e ns o MCA and LIA hyd oclima e ac oss he Old Wo ld o he pe iods 1000–1200 and 1550–1750 CE, which all wi hin he gene ally accep ed ime spans o he MCA and LIA (23,24). The speci ic pe iod chosen he e o he MCA is also one in which a megad ough was p e iously econs uc ed o ha e occu ed in sou he n Finland (29). The mode n pe iod is de ined as 1850–2012. Fo each epoch, composi e maps o e- cons uc ed scPDSI (Fig. 3A) a e shown o he ull OWDA domain (uppe maps) as well as o jus he s a is ically signi ican (P<0.01) egions o we ness and d yness (lowe maps). The MCA pe iod ana- lyzed he e is signi ican ly d ie o e a la ge po ion o con inen al no h-cen al Eu ope and sou he n Scandina ia han ei he he LIA 0˚ 16˚ 32˚ 36˚ 48˚ 60˚ 0˚ 16˚ 32˚ 0˚ 16˚ 32˚ 36˚ 48˚ 60˚ 1921 1893 1741 0˚ 16˚ 32˚ 36˚ 48˚ 60˚ 0˚ 16˚ 32˚ 0˚ 16˚ 32˚ 36˚ 48˚ 60˚ 1616 1540 1315 ABC DEF –6 –4 –2 0 2 4 6 –6 –4 –2 0 2 4 6 –6 –4 –2 0 2 4 6 Fig. 2. OWDA maps o known yea s o hyd oclima ic ex emes. (A o F) The maps a e p esen ed in e e se ch onological o de based on documen a y clima e eco ds: om he bes yea s eco ded by ins umen al clima e eco ds [1921 (A) and 1893 (B)] o he lesse known yea s [1741 (C), 1616 (D), 1540(E), and 1315 (F)]. See he ex o de ails and e e o Supplemen a y Ma e ials o mo e examples o his o ical d ough s om documen a y eco ds. RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 3o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om 0˚ 16˚ 32˚ 36˚ 48˚ 60˚ 0˚ 16˚ 32˚ 0˚ 16˚ 32˚ 0˚ 16˚ 32˚ 0˚ 16˚ 32˚ 0˚ 16˚ 32˚ 36˚ 48˚ 60˚ −1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 36˚ 48˚ 60˚ 36˚ 48˚ 60˚ 1000−1200 1550−1750 1850−2012 >99% CL >99% CL >99% CL MCA LIA MOD –4 –2 0 2 4 900 1100 1300 1500 1700 1900 scPDSI Yea MCA LIA MOD B A scPDSI scPDSI scPDSI Fig. 3. Compa ison o mean scPDSI ields in he OWDA du ing pe iods associa ed wi h he MCA, LIA, and mode n pe iod (MOD). (A)Themean ields we e calcula ed o e he ime in e als indica ed, and he a eas in hose ields wi h signi ican mean anomalies o we ness o d yness (p< 0.01, wo- ailed, co ec ed o lag −1 au oco ela ion) a e indica ed in he middle se o maps. The a ea o maximum d yness du ing he MCA pe iod is indica ed by he yellow ec angle in he lowe MCA map. (B) A e age o OWDA econs uc ions om wi hin his ec angle. I con i ms he d ie condi ions du ing he MCA pe iod and also shows he occu ence o an ex ao dina y megad ough in he mid-15 h cen u y. CL, con idence le el. RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 4o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om o mode n pe iod, conside ably adding o he p e ious epo o an MCA megad ough in sou he n Finland (29), and i now mo e com- ple ely de ines he spa ial pa e n andex en o d ynessdu ing ha ime. In con as , he Romania and Uk aine egions o eas e n Eu ope ha e mo e simila pa e ns o d yness, and no he n Fennoscandia and Russia ha e mo e simila pa e ns o we ness, in all h ee epochs. No ably, he o e all iming o MCA d yness in no h-cen al Eu ope is consis en wi h ha desc ibed o la ge a eas o No h Ame ica (26,27)(seela e discussion). A summa y o he his o y o d ough and we ness since 870 CE in heco e egiono OldWo ldMCAd ough (Fig.3A,yellow ec angle) is p esen ed in Fig. 3B. The o e all mean ± 1se o is −0.44 ± 0.04 scPDSI uni s om he expec ed mean o ze o o he 1928–1978 calib a ion pe iod, which e lec s he gene al endency o d ie condi ions in he p eindus ial pas . In con as , he mos ecen pe iod (1998–2012) has been anomalously we (+0.97 ± 0.24). I is necessa y o go back o 1721– 1739 o ind a we e pe iod o compa able du a ion (+1.55 ± 0.24). As a ela i e index o d ough , scPDSI has a high deg ee o spa ial compa a- bili y ac oss a b oad ange o p ecipi a ion clima ologies (30). This allows us o compa e his d ough o ano he econs uc ed medie al mega- d ough occu ing a a ound he same imeinwes e nNo hAme ica (26). The 1000–1200 CE megad ough o e no h-cen al Eu ope has a econs uc ed mean o −0.72 ± 0.10 scPDSI uni s. By compa ison, he wo s megad ough in he Cali o nia and Ne ada egions o he NADA (26) las ed om 832 o 1074 CE (−0.84 ± 0.09, calcula ed a e adjus ing he mean o he Cali o nia/Ne ada se ies o ma ch ha o he no h- cen al Eu ope se ies o e hei 870–2005 common in e al). Thus, in e ms o ela i e d yness as modeled by he scPDSI, his MCA mega- d ough in he OWDA is compa able o one o he mo e excep ional MCA megad ough s in he NADA. Besides he MCA, Fig. 3B also e eals he occu ence o a mid–15 h- cen u y megad ough in no h-cen al Eu ope. The mos in ense d ough phase las ed o 37 yea s om 1437 o 1473 CE (−1.84 ± 0.20), wi h only wo isola ed yea s o posi i e scPDSI. The iming o his megad ough is simila o ha o he wo s d ough econs uc ed o ha e occu ed o e he pas 1000 yea s in he sou heas e n Uni ed S a es (27). This sugges s he exis ence o some common hyd oclima e o cing ac oss he No h A lan ic, pe haps ela ed o A lan ic Ocean sea su ace empe a u e a ia ions and/o he No h A lan ic Oscilla ion (31,32). Finally, a hi d megad ough occu ed om 1779 o 1827 (−1.34 ± 0.16). This pe iod has a subpe iod o “majo long-du a ion d ough ”(33) om 1798 o 1808 (−1.89 ± 0.38) in England and Wales iden i ied om ea ly ins umen al NADA MADA OWDA La i ude A NHDA egions –4 –2 0 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2 4 Z-sco es B NHDA a e ages Yea s Longi ude 0˚ −150˚ −100˚ −50˚ 50˚ 100˚ 150˚0˚ 20˚ 40˚ 60˚ Fig. 4. The NHDA based on he OWDA, NADA, and MADA. (A) The empe a e la i ude egions o d ough a lases wi hin he dashed boxes a e em- phasized o pu poses o compa ison because no all d ough a lases ha e bo eal ( o example, MADA) o opical ( o example, OWDA) econs uc ions. (B) The o iginal annually esol ed d ough econs uc ions in each egion we e a e aged om 1000 o 1989 CE, ans o med in o s anda d no mal de- ia es (Zsco es), and low pass– il e ed o emphasize a iabili y ha was >30 yea s in du a ion. The low pass– il e ed a e age se ies we e eno malized o elimina e any di e en ial weigh ing by egion and a e aged o p oduce he NHDA eco ds (no eno malized) shown in black. RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 5o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om and his o ical clima e in o ma ion. I is also he d ies pe iod wi hin he longe epoch (1779–1827) o pe sis en ly d ie - han-a e age condi ions o e no h-cen al Eu ope. Mo e gene ally, Fig. 3B e eals he exis ence o la ge-ampli ude decadal o cen ennial hyd oclima e a iabili y o e Eu ope and shows ha , like No h Ame ica, megad ough s in he Old Wo ld we e no es ic ed o jus he MCA pe iod. In compa ison, hyd o- clima e a iabili y o e he 20 h cen u y, al hough la ge, does no ap- pea unp eceden ed in ampli ude o end. Isola ing signals o ecen GHG-induced hyd oclima e change om his complex eco d o na u al a iabili y will be challenging. DISCUSSION The OWDA g ea ly expands ou unde s anding o he spa io empo al his o y o d ough s and plu ials in he Old Wo ld. I con i ms he oc- cu ence o an MCA megad ough in no h-cen al Eu ope, i s no ed in sou he n Finland (29), which is simila in iming, du a ion, and el- a i e in ensi y o ha ound in he NADA o he wes e n Uni ed S a es (26). This inding signi ican ly adds o he ealiza ion ha NH d ough s we e mo e p olonged du ing he MCA compa ed o he 20 h cen u y, wi h li le unde s anding as o why. In addi ion, he example maps illus- a e i s conside able ele ance o s udies o hyd oclima e impac on Old Wo ldsocie iesandcul u eso e heCommonE a. The OWDA is an impo an new ad ancemen owa d achie ing a ull and consis en spa ial co e age o hyd oclima ic a iabili y ac oss he NH land a eas du ing he Common E a, one o he p ima y goals o Kau man and he PAGES 2k Conso ium (34). Figu e 4 shows p o- g ess owa d he comple ion o a “No he n Hemisphe e D ough A las” (NHDA) o he empe a e la i ude egions o he OWDA, NADA, and MADA (Fig. 4A). This gi es us an oppo uni y o compa e hyd oclima ic a iabili y be ween con inen s on a hemisphe ic basis back o he MCA o he i s ime. A i s compa ison, based on no malized a e ages o econs uc ed d ough s smoo hed o emphasize >30-yea ime scales o a iabili y (Fig. 4B), shows li le commonali y in he iming o we and d y epochs be ween d ough a las egions a he con inen al scale o e he pas 1000 yea s (a e age = 0.05). Howe e , he e a e se e al pe iods o dis- inc an iphasing be ween egions, especially be ween he NADA and he MADA, which none heless sugges s some dynamical links be ween hem. The NHDA a e age has also shown a long- e m posi i e end om d ie o we e condi ions o e he empe a e la i udes o he NH since he MCA. The addi ion o he OWDA o he NADA and MADA hus allows o mo e comple e in es iga ions o a mosphe e-ocean dynamical in luences on he hemisphe ic pa e ns o hyd oclima e a iabili y o e he Common E a. The use o mul iple clima e ield econ- s uc ions (spa ially and ully da a-independen ) ep esen ed by he OWDA, NADA, and MADA should be e cons ain he modes o cli- ma e a iabili y esponsible and g ea ly imp o e ou unde s anding o he causes o hyd oclima ic a iabili y a in e annual o cen ennial ime scales. Fu he mo e, comple ion o he OWDA, oge he wi h a ailable simula ions o he las millennium, now allows us o de e mine whe he s a e-o - he-a clima e models in he NH con ain ealis ic hyd oclima e a iabili y a in e annual o cen ennial ime scales [see Sme don e al. (35) o a No h Ame ican example]. This knowledge is essen ial o assessing whe he models can co ec ly ep esen he ange o u u e hyd oclima es, including we and d y ex emes, ha combine o ced change wi h con inued na u al a iabili y. MATERIALS AND METHODS The scPDSI da a used o econs uc ion (11) a e based on g idded Clima ic Resea ch Uni Time Se ies (CRU TS) mon hly empe a u e and p ecipi a ion da a (36) upda ed o e sion CRU TS 3.21 and co e - ing he pe iod 1901–2012. We examined he changing densi y o p e- cipi a ion s a ions a ailable o in e pola ion on o he CRU TS 3.21 p ecipi a ion g id be o e 1950 and ound i o be s able back in ime o e mos o he OWDA domain (Supplemen a y Ma e ials). The g ea es excep ionwas oundinTu keyand heMiddleEas ,whe eli lelocal p ecipi a ion s a ion da a we e a ailable be o e 1930 o in e pola ion o he hal -deg ee g id. This is no desi able o e alua ing econs uc ion skills in hose egions o e he 1901–1927 eg ession model alida ion pe iod. Howe e , i is unlikely o ad e sely a ec he ee- ing econs uc- ions hemsel es because he 1928–1978 calib a ion pe iod used o p oduce he econs uc ions ba ely o e laps wi h he pe iod o educed local p ecipi a ion da a co e age. The co ela ion decay leng h (CDL) o he scPDSI ield was also e alua ed and ound o be ~800 km, on a - e age, bu wi h conside able la i udinal a iabili y. This in o ma ion was use ul o objec i ely de ining he sea ch adius used o ind ee- ing ch onologies o econs uc ion o pas d ough s (see la e discussion and Supplemen a y Ma e ials o de ails). The ee- ing da a used o p oduce he OWDA came om he In- e na ional T ee-Ring Da a Bank and om con ibu ions by Eu opean dend och onologis s, including a la ge quan i y om he dend oa ch- aeology communi y. Inco po a ing a chaeological ee- ing da a in o he ee- ing ne wo k, a e upda ing wi h mode n ee- ing da a om app o- p ia e younge li ing ees (37), enabled he econs uc ions o be ex ended back a millennium o mo e o e mos o he OWDA domain. The esul ing ee- ing ch onologies we e de eloped o clima e econ- s uc ion using he newes “signal- ee”me hods o ee- ing s anda diza- ion (38,39), wi h emphasis on p ese ing long- e m a iabili y due o clima e. Because o conside able a ia ion in he segmen leng hs o he a chaeological ee- ing se ies in each o hose ee- ing da a se s and he associa ed di icul y in p ese ing long- e m clima e a iabili y because o “segmen leng h cu se”(40), a modi ied signal- ee egional cu e s an- da diza ion (SF-RCS) (39) me hod was de ised and used (see Supple- men a y Ma e ials o a de ailed desc ip ion o how RCS was pe o med). The clima e ield econs uc ionme hodused op oduce heOWDA is he poin -by-poin eg ession (PPR) me hod (41), wi h an ex ension o he p ocedu e p oducing ensembles o clima e econs uc ions (12,42). The OWDA econs uc ion p esen ed he e is he mean o eigh ensem- ble membe s, wi h each membe being based on he weigh ing o each ee- ing ch onology used in PPR by some powe o i s co ela ion wi h scPDSI (42). The ini ial sea ch adius used by PPR o loca e ee- ing ch onologies o econs uc ing scPDSI a each g id poin was se o 1000 km, a small enla gemen o e he es ima ed CDL (800 km) o he ins umen al scPDSI da a o accoun o some o he i egula spacings o he ee- ing ch onology ne wo k shown in Fig. 1. A alue-added ou come o PPR was he abili y o e alua e he clima e sensi i i y o he ee- ing ch onologies used o scPDSI econs uc ion. This e alua ion, pe o med h ough a simple co ela ion analysis, conclu- si ely demons a ed he o e all mois u e sensi i i y o he ee- ing ch onologies o e mos o he OWDA domain, a esul ha is highly consis en wi h an independen e alua ion o he clima e esponse o ee- ing ch onologies in Eu ope (43). A high ele a ions in cen al Eu- ope and a high no he n la i udes in Fennoscandia, he le el o mois- u e sensi i i y does diminish and is inc easingly eplaced by sensi i i y RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 6o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om o g owing-season empe a u es. This change in clima e sensi i i y has also been desc ibed by Babs e al.(43). The success ul econs uc ion o d ough s in Fennoscandia, using a mix u e o mois u e- and empe a u e- limi ed ee- ing eco ds (44), indica es ha his change in clima e sensi- i i y is no a se ious issue ( e e o Supplemen a y Ma e ials o de ails). SUPPLEMENTARY MATERIALS Supplemen a y ma e ial o his a icle is a ailable a h p://ad ances.sciencemag.o g/cgi/ con en / ull/1/10/e1500561/DC1 In oduc ion OWDA as a scien i ic ad ancemen o e p e ious wo k G idded mon hly scPDSI a ge ield OWDA ee- ing ne wo k Clima e sensi i i y o OWDA ee- ing ch onologies Valida ion o OWDA ee- ing clima e esponse Augmen ing ee- ing ch onologies wi h his o ical ee- ing da a S anda dizing OWDA ee- ing da a o clima e econs uc ion S anda dizing OWDA his o ical/mode n ee- ing da a Es ima ing low- o medium- equency a iance e en ion Poin -by-poin eg ession Compa isons wi h Pauling sp ing-summe p ecipi a ion econs uc ions Addi ional alida ion es s o he OWDA Re e ences Table S1. Lis o ee- ing ch onologies used o p oducing he OWDA. Fig. S1. Map o he OWDA domain showing 5414 hal -deg ee g id poin s o JJA scPDSI (small black do s) and he 106 annual ch onology ee- ing ne wo k ( ed and blue iangles). Fig. S2. Maps, by decade (up o 1950), o he changing densi ies o p ecipi a ion s a ions (solid ed do s) a ailable o in e pola ion on he hal -deg ee egula g id used o p oduce he CRU TS p ecipi a ion ield (h p://badc.ne c.ac.uk). Fig. S3. Compa isons o calib a ion pe iod (1928–1978) and alida ion pe iod (1901–1927) scPDSI a e ages and hei a iances. Fig. S4. S a is ical p ope ies o g idded summe scPDSI da a o e he 1928–1978 calib a ion pe iod and es s o no mali y using a simple and obus es o no mali y based on join use o skewness and ku osis (69). Fig. S5. CDL be ween 5414 g id poin s o summe scPDSI used o econs uc ion o e he OWDA domain. Fig. S6. Summa y maps o co ela ions be ween summe scPDSI and he ee- ing ne wo k o e he 5414 g id poin s o he OWDA domain calcula ed o he 1928–1978 calib a ion pe iod, using he PPR p og am in he same way ha i was used o p oduce he OWDA econs uc ions. Fig. S7. Example o a his o ical/mode n ee- ing ch onology om no heas e n F ance de eloped by he i e a i e p ocedu e desc ibed in he ex . Fig. S8. O e lay plo s o To ne ask powe spec a ( equencies om 0 o 0.1) o di e en de ending op ions be o e (RCS/SSD) and a e (SF-RCS/SSD) he applica ion o he signal- ee me hod o he da a using he same cu e- i ing op ions: Op 0—RCS de ending (designed o p ese e he mos low- o medium- equency a iance); Op 1—nega i e exponen ial/linea de ending (mono onic noninc easing, leas lexible SSD op ion); Op 2—cubic smoo hing spline de ending based on he median segmen leng h o he da a (mode a ely da a-adap i e, ixed in e media e lexibili y); and Op 3— he F iedman a iable span smoo he (locally adap i e, e y lexible). Fig. S9. Example o he wo-s age SF-RCS me hod applied o he his o ical/mode n Que cus species (QUSP) ee- ing da a o no heas e n F ance used as an example by Aue e al.(48). Fig. S10. Addi ional low- o medium- equency a iance e ained in he his o ical/mode n ee- ing ch onologies using he wo-s age SF-RCS p ocedu e. Fig. S11. Calib a ion and alida ion s a is ical maps o he eigh -membe ensemble-a e age OWDA econs uc ions. Fig. S12. Co ela ions o OWDA JJA scPDSI econs uc ions wi h Pauling sp ing-summe p ecipi a ion econs uc ions p ima ily econs uc ed om long ins umen al and his o ical clima e indices (18). Fig. S13. Compa ison o OWDA and Pauling maps o 1540 (“yea -long unp eceden ed Eu opean hea and d ough ”)(21). Fig. S14. Maps o excep ional d ough s in Czech lands (20). Fig. S15. Maps o he g ea Eu opean amine (22). Fig. S16. OWDA mean and median maps o nine no ewo hy 17 h-cen u y d ough s o e England and Wales: 1634, 1635, 1636, 1666, 1667, 1684, 1685, 1694, and 1695 (33). Fig. S17. OWDA mean and median maps o eigh no ewo hy O oman Empi e d ough s: 1570, 1591, 1592, 1594, 1595, 1607, 1608, and 1610 (107). Fig. S18. OWDA mean and medianmaps o 12 no ewo hy p e-1450 his o icald ough s in England and Wales: 1084, 1129, 1136, 1222, 1242, 1252, 1263, 1272, 1284, 1288, 1305, and 1385 (110). Re e ences (45–110) REFERENCES AND NOTES 1. C. M. Philand as, P. T. Nas os, J. Kapsomenakis, K. C. Dou is, G. Tselioudis, C. S. Ze e os, Long e m p ecipi a ion ends and a iabili y wi hin he Medi e anean egion. Na . Haza ds Ea h Sys . Sci. 11, 3235–3250 (2011). 2. B. Ki man, S. B. Powe , J. A. Adedoyin, G. J. Boe , R. Boja iu, I. Camilloni, F. J. Doblas-Reyes, A. M. Fio e, M. Kimo o, G. A. Meehl, M. P a he , A. Sa , C. Schä , R. Su on, G. J. an Oldenbo gh, G. Vecchi, H. J. Wan, Nea - e m clima e change: P ojec ions and p edic abili y, in Clima e Change 2013: The Physical Science Basis. Con ibu ion o Wo king G oup I o he Fi h Assessmen Repo o he In e go e nmen al Panel on Clima e Change, T. F. S ocke , D. Qin, G.-K. Pla ne , M. M. B. Tigno , S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, P.M. Midgley, Eds. (Cam- b idge Uni . P ess, Camb idge, UK, 2013). 3. R. Seage , H. Liu, N. Hende son, I. Simpson, C. Kelley, T. Shaw, Y. Kushni , M. Ting, Causes o inc easing a idi ica ion o he Medi e anean egion in esponse o ising g eenhouse gases. J. Clima e 27, 4655–4676 (2014). 4. C. Kelley, M. Ting, R. Seage , Y. Kushni , The ela i e con ibu ions o adia i e o cing and in e nal clima e a iabili y o he la e 20 h cen u y win e d ying o he Medi e anean egion. Clim. Dyn. 38, 2001–2015 (2011). 5. M. Hoe ling, J. Eischeid, J. Pe lwi z, X. Quan, T. Zhang, P. Pegion, On he inc eased equency o Medi e anean d ough . J. Clima e 25, 2146–2161 (2012). 6. B. O lowsky, S. I. Sene i a ne, Elusi e d ough : Unce ain y in obse ed ends and sho - and long- e m CMIP5 p ojec ions. Hyd ol. Ea h Sys . Sci. 17,1765–1781 (2013). 7. P. G e e, B. O lowsky, B. Muelle , J. She ield, M. Reichs ein, S. I. Sene i a ne, Global assess- men o ends in we ing and d ying o e land. Na . Geosci. 7,716–721 (2014). 8. E. R. Cook, C. A. Woodhouse, C. M. Eakin, D. M. Meko, D. W. S ahle, Long- e m a idi y changes in he wes e n Uni ed S a es. Science 306, 1015–1018 (2004). 9. S. Coa s, J. E. Sme don, B. I. Cook, R. Seage , A e simula ed megad ough s in he No h Ame ican Sou hwes o ced? J. Clima e 28, 124–142 (2015). 10. S. Coa s, B. I. Cook, J. E. Sme don, R. Seage , No h Ame ican pancon inen al d ough s in model simula ions o he las millennium. J. Clima e 28, 2025–2043 (2015). 11. G. an de Sch ie , J. Ba ichi ich, K. R. B i a, P. D. Jones, A scPDSI-based global da a se o d y and we spells o 1901–2009. J. Geophys. Res. 118, 4025–4048 (2013). 12. E. R. Cook, K. J. Anchukai is, B. M. Buckley, R. D. D’A igo, G. C. Jacoby, W. E. W igh , Asian monsoon ailu e and megad ough du ing he las millennium. Science 328,486–489 (2010). 13. R. B ázdil, C. P is e , H. Wanne , H. Von S o ch, J. Lu e bache , His o ical clima ology in Eu ope—The s a e o he a . Clim. Change 70, 363–430 (2005). 14. C. E. P. B ooks, J. Glasspoole, The d ough o 1921. Q. J. Roy. Me eo ol. Soc. 48, 139–168 (1922). 15. L. C. W. Bonacina, The Eu opean d ough o 1921. Na u e 112, 488–489 (1923). 16. F. J. B odie, The g ea d ough o 1893 and i s a endan me eo ological phenomena. Q. J. Roy. Me eo ol. Soc. 20,1–30 (1894). 17. S. Engle , F. Mauelshagen, J. We ne , J. Lu e bache , The I ish amine o 1740–1741: Fam- ine ulne abili y and “clima e mig a ion”.Clim. Pas 9, 1161–1179 (2013). 18. A. Pauling, J. Lu e bache , C. Cas y, H. Wanne , Fi e hund ed yea s o g idded high- esolu ion p ecipi a ion econs uc ions o e Eu ope and he connec ion o la ge-scale ci cula ion. Clim. Dyn. 26,387–405 (2006). 19. P. D. Jones, K. R. B i a, Unusual clima e in no hwes Eu ope du ing he pe iod 1730 o 1745 based on ins umen al and documen a y da a. Clim. Change 79,361–379 (2006). 20. R. B ázdil, P. Dob o olný, M. T nka, O. Ko yza, L. Řezníčko á, H. Valášek, P. Zah adníček, P. Š ěpánek, D ough s in he Czech Lands, 1090–2012 AD. Clim. Pas 9, 1985–2002 (2013). 21. O. We e , C. P is e , J. P. We ne , E. Zo i a, S. Wagne , S. I. Sene i a ne, J. He ge , U. G ünewald, J. Lu e bache , M.-J. Alco o ado, M. Ba iendos, U. Biebe , R. B ázdil, K. H. Bu meis e , C. Camenisch, A. Con ino, P. Dob o olný, R. Glase , I. Himmelsbach, A. Kiss, O. Ko yza, T. Labbé, D. Limanówka, L. Li zenbu ge , Ø. No dl, K. P ibyl, D. Re sö, D. Riemann, C. Roh , W. Sieg ied, J. Söde be g, J.-L. Sp ing, The yea -long unp eceden ed Eu opean hea and d ough o 1540—A wo s case. Clim. Change 125,349–363 (2014). 22. H. S. Lucas, The g ea Eu opean amine o 1315, 1316, and 1317. Speculum 5, 343–377 (1930). 23. H. F. Diaz, R. T igo, M. K. Hughes, M. E. Mann, E. Xoplaki, D. Ba ioped o, Spa ial and empo al cha ac e is ics o clima e in medie al imes e isi ed. Bull. Am. Me eo ol. Soc. 92, 1487–1500 (2011). 24. J. A. Ma hews, K. R. B i a, The ‘Li le Ice Age’: Re-e alua ion o an e ol ing concep . Geog . Ann. 87 A,17–36 (2005). 25. PAGES 2k Conso ium, Con inen al-scale empe a u e a iabili y du ing he pas wo millennia. Na . Geosci. 6, 339–346 (2013). 26. E. R. Cook, R. Seage , R. R. Heim J ., R. S. Vose, C. He weije , C. Woodhouse, Megad ough s in No h Ame ica: Placing IPCC p ojec ions o hyd oclima ic change in a long- e m paleoclima e con ex . J. Qua . Sci. 25,48–61 (2010). 27. B. I. Cook, J. E. Sme don, R. Seage , E. R. Cook, Pan-con inen al d ough s in No h Ame ica o e he las millennium. J. Clima e 27, 383–397 (2014). RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 7o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om 28. R. Seage , N. G aham, C. He weije , A. L. Go don, Y. Kushni , E. Cook, Bluep in s o medie al hyd oclima e. Qua . Sci. Re . 26,2322–2336 (2007). 29. S. Helama, J. Me iläinen, H. Tuomen i a, Mul icen ennial megad ough in no he n Eu ope coincided wi h a global El Niño–Sou he n Oscilla ion d ough pa e n du ing he Medie al Clima e Anomaly. Geology 37,175–178 (2009). 30. N. Wells, S. Godda d, M. J. Hayes, A sel -calib a ing Palme D ough Se e i y Index. J. Clima e 17, 2335–2351 (2004). 31. S. Feng, Q. Hu, R. J. Oglesby, In luence o A lan ic sea su ace empe a u es on pe sis en d ough in No h Ame ica. Clim. Dyn. 37, 569–586 (2011). 32. R. Oglesby, S. Feng, Q. Hu, C. Rowe, The ole o he A lan ic Mul idecadal Oscilla ion on medie al d ough in No h Ame ica: Syn hesizing esul s om p oxy da a and clima e models. Glob. Plane . Change 84–85,56–65 (2012). 33. G. A. Cole, T. J. Ma sh, The impac o clima e change on se e e d ough s. Majo d ough s in England and Wales om 1800 and e idence o impac , in Science Repo : SC040068/SR1 (En i onmen Agency, B is ol, UK, 2006). 34. D. S. Kau man, PAGES 2k Conso ium, A communi y-d i en amewo k o clima e econ- s uc ions. Eos 40, 361–362 (2014). 35. J. E. Sme don, B. I. Cook, E. R. Cook, R. Seage , B idging pas and u u e clima e ac oss paleoclima ic econs uc ions, obse a ions, and models: A hyd oclima e case s udy. J. Clima e 28,3212–3231 (2015). 36. I. Ha is, P. D. Jones, T. J. Osbo n, D. H. Lis e , Upda ed high- esolu ion g ids o mon hly clima ic obse a ions—The CRU TS3.10 da ase . In . J. Clima ol. 34, 623–642 (2014). 37. W. Tegel, J. Vanmoe ke ke, U. Bün gen, Upda ing his o ical ee- ing eco ds o clima e econs uc ion. Qua . Sci. Re . 29, 1957–1959 (2010). 38. T. M. Mel in, K. R. B i a, A “signal- ee”app oach o dend oclima ic s anda disa ion. Dend och onologia 26,71–86 (2008). 39. T. M. Mel in, K. R. B i a, CRUST: So wa e o he implemen a ion o Regional Ch onology S anda disa ion: Pa 1. Signal-F ee RCS. Dend och onologia 32,7–20 (2014). 40. E.R.Cook,K.R.B i a,D.M.Meko,D.A.G aybill,G.Funkhouse ,The‘segmen leng h cu se’in long ee- ing ch onology de elopmen o palaeoclima ic s udies. Holocene 5,229–237 (1995). 41. E. R. Cook, D. M. Meko, D. W. S ahle, M. K. Clea eland, D ough econs uc ions o he con inen al Uni ed S a es. J. Clima e 12, 1145–1162 (1999). 42. E. R. Cook, P. J. K usic, K. J. Anchukai is, B. M. Buckley, T. Naka suka, M. Sano, PAGES Asia2k Membe s, T ee- ing econs uc ed summe empe a u e anomalies o empe a e Eas Asia since 800 C.E. Clim. Dyn. 41, 2957–2972 (2013). 43. F. Babs , B. Poul e , V. T oue , K. Tan, B. Neuwi h, R. Wilson, M. Ca e , M. G abne , W. Tegal, T. Le anic, M. Panayo o , C. U bina i, O. Bou iaud, P. Ciais, D. F ank, Si e- and species-speci ic esponses o o es g ow h o clima e ac oss he Eu opean con inen . Glob. Ecol. Biogeog . 22, 706–717 (2013). 44. K. Se igen, J. Bjö klund, E. R. Cook, H. W. Linde holm, A ee- ing ield econs uc ion o Fennoscandian summe hyd oclima e a iabili y o he las millennium. Clim. Dyn. 44, 3141–3154 (2014). 45. W. C. Palme , Me eo ological D ough , Tech. Rep. Wea he Bu eau Resea ch Pape No. 45, US Depa men o Comme ce, Washing on DC (1965). 46. G. an de Sch ie , K. R. B i a, P. D. Jones, T. J. Osbo n, Summe mois u e a iabili y ac oss Eu ope. J. Clim. 19, 2818–2834 (2006). 47. K. R. B i a, P. D. Jones, M. Hulme, Summe mois u e a iabili y ac oss Eu ope, 1892–1991: An analysis based on he Palme D ough Se e i y Index. In . J. Clima ol. 14,475–506 (1994). 48. I. Aue , R. Böhm, A. Ju ko ic, W. Lipa, A. O lik, R. Po zmann, W. Schöne , M. Unge sböck, C. Ma ulla, K. B i a, P. Jones, D. E hymiadis, M. B une i, T. Nanni, M. Mauge i, L. Me calli, O. Mes e, J.-M. Moisselin, M. Bege , G. Mülle -Wes e meie , V. K e on, O. Bochnicek, P. S as ny, M. Lapin, S. Szalai, T. Szen im ey, T. Cegna , M. Dolina , M. Gajic-Capka, K. Zanino ic, Z. Majs o o ic, E. Nieplo a, HISTALP –his o ical ins umen al clima ological su ace ime se ies o he G ea e Alpine Region. In . J. Clima ol. 27,17–46 (2007). 49. K. R. B i a, G. an de Sch ie , P. D. Jones, We and d y summe s in Eu ope since 1750: E idence o inc easing d ough . In . J. Clima ol. 29, 1894–1905 (2009). 50. R. B ázdil, P. Dob o olný, J. Lu e bache , A. Mobe g, C. P is e , D. Wheele , E. Zo i a, Eu opean clima e o he pas 500 yea s: New challenges o his o ical clima ology. Clim. Change 101, 7–40 (2010). 51. R. Touchan, D. Meko, M. K. Hughes, A 396-yea econs uc ion o p ecipi a ion in Sou he n Jo dan. J. Ame . Wa e Res. Assoc. 35,45–55 (1999). 52. R. Touchan, G. Funkhouse , M. K. Hughes, N. E kan, S anda dized p ecipi a ion index e- cons uc ion om Tu kish ee- ing wid hs. Clim. Change 72, 339–353 (2005). 53. R. J. S. Wilson, B. H. Luckman, J. Espe , A 500-yea dend oclima ic econs uc ion o sp ing- summe p ecipi a ion om he lowe Ba a ian o es egion, Ge many. In . J. Clima ol. 25, 611–630 (2005). 54. J. Espe , D. F ank, U. Bün gen, A. Ve s ege, J. Lu e bache , E. Xoplaki, Long- e m d ough se e i y a ia ions in Mo occo. Geophys. Res. Le . 34, L17702 (2007). 55. C. B. G iggs, A. DeGae ano, P. Kuniholm, M. New on, A egional high- equency econ- s uc ion o May-June p ecipi a ion in he no h Aegean om oak ee- ings, AD 1089-1989. In . J. Clima ol. 27, 1075–1089 (2007). 56. K. Ču a , M. De Luis, D. Ecks ein, L. Kaj ež-Boga aj, Recons uc ing d y and we summe s in SE Slo enia om oak ee- ing se ies. In J. Biome eo ol. 52, 607–615 (2008). 57. R. Touchan, K. J. Anchukai is, D. M. Meko, S. A alah, C. Baisan, A. Aloui, Long e m con ex o ecen d ough in no hwes e n A ica. Geophys. Res. Le . 35, L13705 (2008). 58. U. Bün gen, V. T oue a, D. F anka, H. H. Leuschne c, D. F ied ichsd, J. Lu e bache e, J. Espe , T ee- ing indica o o Ge man summe d ough o e he las millennium. Qua . Sci. Re . 29, 1005–1016 (2010). 59. R. J. Coope , T. M. Mel in, I. Tye s, R. J. S. Wilson, K. R. B i a, A ee- ing econs uc ion o Eas Anglian (UK) hyd oclima e a iabili y o e he las millennium. Clim. Dyn. 40, 1019–1039 (2013). 60. R. Wilson, D. Miles, N. J. Loade , T. Mel in, I. Tye s, R. Coope , K. R. B i a, A millennial long Ma ch-July p ecipi a ion econs uc ion o sou he n-cen al England. Clim. Dyn. 40, 997–1017 (2013). 61. N. Köse, Ü. Akkemik, H. T. Güne , H. N. Dal es, H. D. G issino-Maye , M. S. Öze en, T. Kındap, An imp o ed econs uc ion o May–June p ecipi a ion using ee- ing da a om wes e n Tu key and i s links o olcanic e up ions. In . J. Biome eo ol. 57,691–701 (2013). 62. S. Klesse, M. Ziehme , G. Rousakis, V. T oue , D. F ank, Synop ic d i e s o 400 yea s o summe empe a u e and p ecipi a ion a iabili y on M . Olympus, G eece. Clim. Dyn. 45, 807–824 (2015). 63. A. Nicaul , S. Alleaume, S. B ewe , M. Ca e , P. Nola, J. Guio , Medi e anean d ough luc ua ion du ing he las 500 yea s based on ee- ing da a. Clim. Dyn. 31, 227–245 (2008). 64. R. Touchan, K. J. Anchukai is, D. M. Meko, M. Sabi , S. A alah, A. Aloui, Spa io empo al d ough a iabili y in no hwes e n A ica o e he las nine cen u ies. Clim. Dyn. 37,237–252 (2010). 65. G. an de Sch ie , P. D. Jones, K. R. B i a, The sensi i i y o he PDSI o he Tho n hwai e and Penman Mon ei h pa ame e iza ions o po en ial e apo anspi a ion. J. Geophys. Res. 116, D03106 (2011). 66. K. E. T enbe h, A. Dai, G. an de Sch ie , P. D. Jones, J. Ba ichi ich, K. R. B i a, J. She ield, Global wa ming and changes in d ough . Na . Clim. Chang. 4,17–22 (2014). 67. M. New, P. D. Jones, Rep esen ing wen ie h cen u y space– ime clima e a iabili y. Pa 2: De elopmen o 1901–96 mon hly g ids o e es ial su ace clima e. J. Clim. 13, 2217–2238 (2000). 68. L. H. Shoemake , Fixing he F es o equal a iances. Am. S a . 57, 105–114 (2003). 69. R. B. D’Agos ino, A. Belange b, R. B. D’Agos ino J ., A sugges ion o using powe ul and in o ma i e es s o no mali y. Am. S a . 44, 316–321 (1990). 70. M. New, M. Hulme, P. Jones, Rep esen ing wen ie h-cen u y space– ime clima e a iabili y. Pa II: De elopmen o 1901–96 mon hly g ids o e es ial su ace clima e. J. Clim. 13, 2217–2238 (2000). 71. T. D. Mi chell, P. D. Jones, An imp o ed me hod o cons uc ing a da abase o mon hly clima e obse a ions and associa ed high- esolu ion g ids. In . J. Clima ol. 25, 693–712 (2005). 72. R. J. S. Wilson, J. Espe b, B. H. Luckman, U ilising his o ical ee- ing da a o dend oclima ology: A case s udy om he Ba a ian o es , Ge many. Dend och onologia 21,53–68 (2004). 73. K. Haneca, K. Ču a b, H. Beeckmanc, Oaks, ee- ings and wooden cul u al he i age: A e iew o he main cha ac e is ics and applica ions o oak dend och onology in Eu ope. J. A chaeol. Sci. 36,1–11 (2009). 74. K. Nicolussi, M. Kau mann, G. Pa zel , J. Plich an de , A. Thu ne , Holocene ee-line a iabili y in he Kaune Valley, Cen al Eas e n Alps, indica ed by dend och onological analysis o li ing ees and sub ossil logs. Vege . His . A chaeobo . 14, 221–234 (2005). 75. U.Bün gen,W.Tegel,K.Nicolussi,M.McCo mick,D.F ank,V.T oue ,J.O.Kaplan,F.He zig, K.-U. Heussne , H. Wanne , J. Lu e bache , J. Espe , 2500 yea s o Eu opean clima e a iabili y and human suscep ibili y. Science 331,578–582 (2011). 76. A. M. Ga cía-Suá eza, C. J. Bu le , M. G. L. Baillie, Clima e signal in ee- ing ch onologies in a empe a e clima e: A mul i-species app oach. Dend och onologia 27, 183–198 (2009). 77. N. Pede son, A. R. Bell, E. R. Cook, U. Lall, N. De ineni, R. Seage , K. Eggles on, K. P. V anes, Is an epic plu ial masking he wa e insecu i y o he g ea e New Yo k Ci y egion? J. Clim. 26,1339–1354 (2013). 78. D. M. Meko, Applica ions o Box-Jenkins Me hods o Time Se ies Analysis o he Recons uc- ion o D ough om T ee Rings. Unpublished Ph.D. disse a ion, The Uni e si y o A izona (1981). 79. J. F anke, D. F ank, C. C. Raible, J. Espe , S. B önnimann, Spec al biases in ee- ing clima e p oxies. Na . Clim. Chang. 3,360–364 (2013). 80. S. S . Geo ge, T. R. Aul , The imp in o clima e wi hin No he n Hemisphe e ees. Qua . Sci. Re . 89,1–4 (2014). 81. D. A. F ied ichs, V. T oue , U. Bün gen, D. C. F ank, J. Espe , B. Neuwi h, J. Lö le , Species- speci ic clima e sensi i i y o ee g ow h in Cen al-Wes Ge many. T ees –S uc u e and Func ion 23, 729–739 (2009). 82. F. Babs , M. Ca e , B. Poul e , C. U bina i, B. Neuwi h, D. F ank, 500 yea s o egional o es g ow h a iabili y and links o clima ic ex eme e en s in Eu ope. En i on. Res. Le . 7, 045705 (2012). RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 8o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om 83. J. Espe , U. Bün gen, M. Timonen, D. C. F ank , Va iabili y and ex emes o no he n Scandina ian summe empe a u es o e he pas wo millennia. Glob. Plan. Change 88-89,1–9 (2012). 84. T. M. L. Wigley, K. R. B i a, P. D. Jones, On he a e age alue o co ela ed ime se ies, wi h applica ions in dend oclima ology and hyd ome eo ology. J.Clim.Appl.Me .23,201–213 (1984). 85. R. L. Holmes, Compu e -assis ed quali y con ol in ee ing da ing and measu emen s. T ee-Ring Bull. 43,69–78 (1983). 86. C. Neh bass-Ahles, F. Babs , S. Klesse, M. Nö zli, O. Bou iaud, R. Neukom, M. Dobbe in, D. F ank, The in luence o sampling design on ee- ing-based quan i ica ion o o es g ow h. Glob. Change Biol. 20, 2867–2885 (2014). 87. M. G. L. Baillie, J. R. Pilche , A simple c oss-da ing p og am o ee- ing esea ch. T ee-Ring Bull. 38,35–43 (1973). 88. T.M.L.Wigley,P.D.Jones,K.R.B i a,C oss-da ingme hodsindend och onology.J. A chaeol. Sci. 14,51–64 (1987). 89. D. F ank, J. Espe , E. R. Cook, Adjus men o p oxy numbe and cohe ence in a la ge-scale empe a u e econs uc ion. Geophys. Res. Le . 34, L16709 (2007). 90. H. C. F i s, T ee Rings and Clima e (Academic P ess, London, 1976), p. 567. 91.E.R.Cook,L.A.Kai iuks is,Eds.,Me hods o Dend och onology: Applica ions in he En i onmen al Sciences (Kluwe Academic Publishe s, Do d ech , 1990), p. 394. 92. E. R. Cook, K. Pe e s, The smoo hing spline: A new app oach o s anda dizing o es in- e io ee- ing wid h se ies o dend oclima ic s udies. T ee-Ring Bull. 41,45–53 (1981). 93. J. H. F iedman, “A a iable span sca e plo smoo he ”S an o d Uni e si y Technical Repo No. 5(1984). 94. K.R.B i a,P.D.Jones,T.S.Ba holin,D.Ecks ein,F.H.Schweing ube ,W.Ka lén,P.Ze e be g, M. E onen, Fennoscandian summe s om A.D. 500: Tempe a u e changes on sho and long imescales. Clim. Dyn.7, 111–119 (1992). 95. K. R. B i a, T. M. Mel in, A close look a egional cu e s anda diza ion o ee- ing e- co ds: Jus i ica ion o he need, a wa ning o some pi alls, and sugges ed imp o emen s in i s applica ion, in Dend oclima ology,De elopmen s in Paleoen i onmen al Resea ch 11, M. K. Hughes, T. W. Swe nam, H. F. Diaz, Eds. (Sp inge , Do d ech , 2011), pp. 113–145. 96. T. M. Mel in, K. R. B i a, CRUST: So wa e o he implemen a ion o Regional Ch onology S anda disa ion: Pa 2 Fu he RCS op ions and ecommenda ions. Dend och onologia 32, 343–356 (2014). 97. T. M. Mel in, K. R. B i a, K. Nicolussi, M. G abne , Time- a ying- esponse smoo hing. Dend och onologia 25,65–69 (2007). 98. U. Bün gen, W. Tegel, K.-U. Heussne , J. Ho mann, R. Kon ic, T. Kyncl, E. R. Cook, E ec s o sample size in dend oclima ology. Clim. Res. 53, 263–269 (2012). 99. T. M. Mel in, H. G udd, K. R. B i a, Po en ial bias in ‘upda ing’ ee- ing ch onologies using egional cu e s anda disa ion: Re-p ocessing 1500 yea s o To ne äsk densi y and ing-wid h da a. Holocene 23, 364–373 (2013). 100. P. M. Kelly, M. A. R. Mun o, M. K. Hughes, C. M. Goodess, Clima e and signa u e yea s in Wes Eu opean Oaks. Na u e 340,57–60 (1989). 101. M. K. Hughes, P. I. Kuniholm, J. K. Eischeid, G. G. Ga in-Woll, C. B. G iggs, C. La ini, Aegean ee- ing signa u e yea s explained. T ee Ring Res. 57,67–73 (2001). 102. D. Meko, Dend oclima ic econs uc ion wi h ime a ying p edic o subse s o ee indices. J. Clim. 10,687–696 (1997). 103. E. R. Cook, R. D. D’A igo, M. E. Mann, A well- e i ied, mul ip oxy econs uc ion o he win e No h A lan ic oscilla ion index since AD 1400. J. Clim. 15, 1754–1764 (2002). 104. D. M. Allen, The ela ionship be ween a iable selec ion and da a augmen a ion and a me hod o p edic ion. Technome ics 16, 125–127 (1974). 105. W. T. Bell, A. E. J. Ogil ie, Wea he compila ions as a sou ce o da a o he econs uc ion o Eu opean clima e du ing he medie al pe iod. Clim. Chang. 1, 331–348 (1978). 106. M. J. Ing am, D. J. Unde hill, T. M. L. Wigley, His o ical clima ology. Na u e 276,329–334 (1978). 107. S. A. Whi e, Clima e change and c isis in O omon Tu key and he Balkans, 1590-1710. In P oceedings o In e na ional Con e ence on Clima e Change in he Middle Eas : Pas , P esen and Fu u e,20–23 No embe 2006, Is anbul Technical Uni e si y, pp. 391–409. 108. R. Touchan, G. M. Ga in, D. M. Meko, G. Funkhouse , Nesa E kan, M. K. Hughes, B. S. Wallin, P elimina y econs uc ions o sp ing p ecipi a ion in sou hwes e n Tu key om ee- ing wid h. In . J. Clima ol. 23, 157–171 (2002). 109. R. Touchan, E. Xoplaki, G. Funkhouse , J. Lu e bache , M. K. Hughes, N. E kan, Ü. Akkemik, J. S ephan, Recons uc ions o sp ing/summe p ecipi a ion o he eas e n Medi e anean om ee- ing wid hs and i s connec ion o la ge-scale a mosphe ic ci cula ion. Clim. Dyn. 25,75–98 (2005). 110. C. E. B i on, A Me eo ological Ch onology o A.D. 1450, Geophysical Memoi s,no.70 (H. M. Me eo ological O ice, London, 1937), p. 177. Acknowledgmen s: Re iews o ea lie d a s o his pape by K. Anchukai is, M. Cane, B. Cook, J. Sme don, and A. P. Williams imp o ed i s quali y and a e g ea ly app ecia ed. Funding: We hank he Na ional Oceanic and A mosphe ic Adminis a ion Clima e Change Da a and De ec- ion P og am o suppo ing his p ojec (awa d NA10OAR4310123) and he Na ional Science Founda ion Ea h Sys em His o y P og am (awa ds 0075956 and ESH0317288), ATM GEO/ATM Pa- leoclima e P og am (awa d 0758486), and awa d 1103314 o addi ional long- e m suppo in de eloping his esea ch in he Medi e anean egion. Au ho con ibu ions: E.R.C. was ully e- sponsible o da a p ocessing, s a is ical analysis, and p oduc ion o heOWDA. G. .d.S. p o ided he PDSI a ge ield da a. R.S. and Y.K. helped guide he de elopmen o he OWDA and wo ked on clima e dynamics in e p e a ion. W.T. and U.B. spea headed he con ibu ion o ee- ing da a used in he OWDA. K.R.B., D.F.,and P.J.K. con ibu ed da a and c i ically e iewed ea lie d a s. The emain- ing coau ho s con ibu ed ee- ing da a o he e o . Compe ing in e es s: The au ho s decla e ha hey ha e no compe ing in e es s. Da a and ma e ials a ailabili y: All da a needed o e alua e he conclusions in he pape a e p esen in he pape and/o he Supplemen a y Ma e ials. The OWDA econs uc ions a e a chi ed wi h associa ed me ada a (www.ncdc.noaa. go /da a-access/paleoclima ology-da a/da ase s/clima e- econs uc ion) o ou ine public access and use.Asecondpublic access si ewill be se up a he In e na ional Resea ch Ins i u e o Clima e and Socie y/Lamon -Dohe y Ea h Obse a o y Clima e Da a Lib a y (h p://i idl.ldeo.columbia. edu). The OWDA was made possible h ough he ee- ing da a (mos o hem unpublished) p o ided by he dend och onological communi y, which is engaged in a ious aspec s o Old Wo ld ee- ing esea ch. The a ailabili y o dend oa chaeological da a om Eu ope enabled he OWDA o be ex ended back o e mos o he CommonE a, which would ha e been impossible o do o he wise, and he newly a ailable g idded scPDSI da a om he Royal Ne he lands Me e- o ological Ins i u e (h p://climexp.knmi.nl) p o ided he high-quali y ins umen al da a o calib a- ion. O he ee- ing da a came om he In e na ional T ee-Ring Da a Bank (www.ncdc.noaa. go /da a-access/paleoclima ology-da a/da ase s/ ee- ing), om A. Billamboz (Dend olab A - chaeological Se ice, Baden-Wü embe g, Ge many), and om he a chi ed ee- ing da a con ibu ed by K. Tye s and I. Tye s o he ADVANCE-10K p ojec (h p://hol.sagepub.com/con- en /12/6/639.abs ac ). Addi ional da a ela ed o his pape may be eques ed om he au ho s. Lamon -Dohe y Ea h Obse a o y con ibu ion numbe 7938. Submi ed 4 May 2015 Accep ed 4 Augus 2015 Published 6 No embe 2015 10.1126/sciad .1500561 Ci a ion: E. R. Cook, R. Seage , Y. Kushni , K. R. B i a, U. Bün gen, D. F ank, P. J. K usic, W. Tegel, G. an de Sch ie , L. And eu-Hayles, M. Baillie, C. Bai inge , N. Bleiche , N. Bonde, D. B own, M. Ca e , R. Coope , K. Ču a , C. Di ma , J. Espe , C. G iggs, B. Gunna son, B. Gün he , E. Gu ie ez, K. Haneca, S. Helama, F. He zig, K.-U. Heussne , J. Ho mann, P. Janda, R. Kon ic, N. Köse, T. Kyncl, T. Le anič, H. Linde holm, S. Manning, T. M. Mel in, D. Miles, B. Neuwi h, K. Nicolussi, P. Nola, M. Panayo o , I. Popa, A. Ro he, K. Se igen, A. Seim, H. S a a, M. S oboda, T. Thun, M. Timonen, R. Touchan, V. T o siuk, V. T oue , F. Walde , T. Ważny, R. Wilson, C. Zang, Old Wo ld megad ough s and plu ials du ing he Common E a. Sci. Ad . 1, e1500561 (2015). RESEARCH ARTICLE Cook e al. Sci. Ad . 2015;1:e1500561 6 No embe 2015 9o 9 on Sep embe 28, 2016h p://ad ances.sciencemag.o g/Downloaded om