Is coccolithophore distribution in the Mediterranean Sea related to seawater carbonate chemistry?

Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ doi:10.5194/os-11-13-2015 © Au ho (s) 2015. CC A ibu ion 3.0 License. Is coccoli hopho e dis ibu ion in he Medi e anean Sea ela ed o seawa e ca bona e chemis y? A. O iedo1, P. Zi e i1,2,5, M. Ál a ez3, and T. Tanhua4 1Ins i u e o En i onmen al Science and Technology (ICTA), Uni e si a Au ònoma de Ba celona (UAB), 08193 Bella e a, Spain 2Ea h & Clima e Clus e , Depa men o Ea h Sciences, FALW, V ije Uni e si ei Ams e dam, FALW, HV1081 Ams e dam, he Ne he lands 3IEO – Ins i u o Español de Oceanog a ia, Apd. 130, A Co uña 15001, Spain 4GEOMAR Helmhol z-Zen um ü Ozean o schung Kiel, Ma ine Biogeochemis y, Düs e nb ooke Weg 20, 24105 Kiel, Ge many 5ICREA, Ins i ució Ca alana de Rece ca i Es udis A ança s, Ba celona, Spain Co espondence o: A. O iedo (angelama ia.o [email protected] ) Recei ed: 31 Decembe 2013 – Published in Ocean Sci. Discuss.: 20 Feb ua y 2014 Re ised: 24 June 2014 – Accep ed: 1 No embe 2014 – Published: 9 Janua y 2015 Abs ac . The Medi e anean Sea is conside ed a “ho spo ” o clima e change, being cha ac e ized by oligo ophic o ul a-oligo ophic wa e s and apidly inc easing seasu - ace empe a u e and changing ca bona e chemis y. Coccol- i hopho es a e conside ed a dominan phy oplank on g oup in hese wa e s. As ma ine calci ying o ganisms hey a e expec ed o espond o he ongoing changes in seawa e ca bona e chemis y. We p o ide he e a desc ip ion o he sp ing ime coccoli hopho e dis ibu ion in he Medi e anean Sea and ela e his o a b oad se o in si u-measu ed en- i onmen al a iables. Samples we e aken du ing he R/V Me eo (M84/3) oceanog aphic c uise in Ap il 2011, be- ween 0 and 100m wa e dep h om 28 s a ions. To al di- a om and silico lagella e cell concen a ions a e also p e- sen ed. Ou esul s highligh he impo ance o seawa e ca bona e chemis y, especially [CO2− 3] bu also [PO3− 4] in un a eling he dis ibu ion o he e ococcoli hopho es, he mos abundan coccoli hopho e li e phase. Holo- and he - e ococcoli hopho es espond di e en ly o en i onmen al ac o s. Fo ins ance, changes in he e ococcoli hopho e as- semblages we e bes linked o he combina ion o [CO2− 3], pH, and salini y (ρ=0.57), al hough salini y migh be no unc ionally ela ed o coccoli hopho e assemblage dis i- bu ion. Holococcoli hopho es, on he o he hand, showed highe abundances and species di e si y in oligo ophic a eas (bes i , ρ=0.32 o nu ien s), h i ing in nu ien -deple ed wa e s. Clus e ing o he e ococcoli hopho es e ealed h ee g oups o species sha ing mo e han 65% simila i ies. These clus e s could be assigned o he eas e n and wes e n basins and deepe laye s (below 50m), espec i ely. In addi ion, he species Gephy ocapsa oceanica,G. muelle ae, and Emilia- nia huxleyi mo pho ype B/C a e spa ially dis ibu ed oge he and ace he in lux o A lan ic wa e s in o he Medi e anean Sea. The esul s o he p esen wo k emphasize he impo - ance o conside ing holo- and he e ococcoli hopho es sep- a a ely when analyzing changes in species assemblages and di e si y. Ou indings sugges ha coccoli hopho es a e a main phy oplank on g oup in he en i e Medi e anean Sea and can domina e o e siliceous phy oplank on. They ha e li e s ages ha a e expec ed o espond di e en ly o he a i- abili y in seawa e ca bona e chemis y and nu ien concen- a ions. 1 In oduc ion Ma ine phy oplank on cons i u es abou 1–2% o he global biomass among p ima y p oduce s (Falkowski, 1994); how- e e , i con ibu es o ∼46% o he p ima y p oduc ion in a global scale (Field e al., 1998). Coccoli hopho es ep esen ∼10% o global phy oplank on biomass (Ty ell and Young, 2009). They play an impo an ole in biogeochemical cy- Published by Cope nicus Publica ions on behal o he Eu opean Geosciences Union. 14 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s cles, con ibu ing o bo h he o ganic and ino ganic ca bon pumps h ough pho osyn hesis and calci ica ion. The la e is he main p ocess con olling PIC:POC ( ain a io). Fo ins ance, in he eas e n Medi e anean Sea, hey a e he main con ibu o o he ino ganic ca bon pump (CaCO3p oduc ion and lux) h oughou he yea (Knappe sbusch, 1993; Zi e i e al., 2000; Malin e no e al., 2003). Mos s udies looking a coccoli hopho e assemblages and dis ibu ion ake in o accoun pa ame e s such as nu i- en s, pho osyn he ically ac i e adia ion (PAR), empe a u e, salini y, and oxygen (e.g., Young, 1994; Zi e i e al., 1995; Hagino e al., 2000; Takahashi and Okada, 2000; Haida and Thie s ein, 2001; Co és e al., 2001; Igna iades e al., 2009). Pa ame e s ela ed o he seawa e ca bona e sys em ha e only ecen ly been conside ed due o hei impo ance o calci ica ion and he ongoing and p ojec ed changes di ec ly ela ed o he apidly inc easing a mosphe ic pCO2. Ca - bona e chemis y pa ame e s ha e been sugges ed as d i e s o coccosphe e mo phology modi ica ion in ield samples (Beau o e al., 2008, 2011; Meie e al., 2014; T ian aphyl- lou e al., 2010), Emiliania huxleyi blooms (Me ico e al., 2006; Ty ell e al., 2008), and changes in coccoli hopho e assemblage composi ion (Cha alampopoulou e al., 2011). Al hough i is no clea why coccoli hopho es calci y, calci i- ca ion is an ene gy-consuming p ocess o coccoli hopho es (B and, 1994; Balch, 2004), main ained by na u al selec ion o e millions o yea s, ha changes he ca bona e chem- is y o hei su ounding media. I is he e o e plausible ha he a ailabili y o he necessa y esou ces o ca y- ing ou calci ica ion (i.e., HCO− 3and CO2− 3)should acili- a e coccoli hopho e’s g ow h in he ocean. In his con ex , i is impo an o unde s and how ma ine calci ying o gan- isms could espond o he apid accumula ion o a mosphe ic CO2and hei in e ac ion wi h he ocean’s ca bona e chem- is y (K oeke e al., 2013). The Medi e anean Sea p o ides an ideal g ound o ex- plo e he ac o s con olling coccoli hopho e dis ibu ion be- cause o he well-known la ge g adien in physicochemical pa ame e s. I has a nega i e esh-wa e balance, wi h e ap- o a ion exceeding p ecipi a ion. Su ace wa e empe a u e, salini y, o al alkalini y (TA), and [CO2− 3] inc ease owa ds he eas e n basin. The Medi e anean Sea is one o he mos nu ien -poo egions o he global ocean (Dugdale and Wilk- e son, 1988), wi h a ophic s a us anging om meso ophic in he no hwes o ex emely oligo ophic in he eas (K om e al., 1991; Be man e al., 1984; Be land e al., 1988; Yacobi e al., 1995; Psa a e al., 2000). The spa ial dis ibu ion o he phy oplank on communi y along an eas –wes ansec shows ha coccoli hopho es can domina e along he Medi e anean Sea, in he Le an ine, Ionian, and Ty henian basins (Igna i- ades e al., 2009). Ocean acidi ica ion, wa ming, and changes in nu ien a ailabili y a e expec ed o signi ican ly al e p i- ma y p oduc ion a es, as well as he o e all plank on com- muni y s uc u e. S udies on coccoli hopho es dis ibu ion in he Medi e anean Sea a e mos ly egional (Dimiza e al., 2008; Malin e no e al., 2003), losing pa o he abo emen- ioned g adien s. Addi ionally, ca bona e chemis y pa ame- e s we e no a ailable in olde s udies ocussed on he dis- ibu ion o coccoli hopho es in he Medi e anean Sea. Thus compa isons be ween he di e en basins a e sca ce and he in luence o ca bona e chemis y pa ame e s on ac ual coc- coli hopho e assemblages emains he e o e unce ain. The p esen wo k in es iga es he egional and e ical dis- ibu ion o li ing coccoli hopho es in he Medi e anean Sea wi h espec o in si u-measu ed en i onmen al pa ame e s and wi h ocus on hose o he ca bona e chemis y. I p o- ides a desc ip ion o he la e sp ing ime coccoli hopho e assemblage’s composi ion and dis ibu ion in he Medi e - anean Sea, wi h a basin esolu ion ha has no been assessed be o e, and along physical and chemical g adien s. 2 Ma e ial and me hods 2.1 Sampling A de ailed wa e sampling was conduc ed du ing he M84/3 c uise om 6 o 28 Ap il 2011 onboa d he R/V Me eo (Tan- hua e al., 2013a). He e we in es iga e a subse o 81 sam- ples om 28 s a ions collec ed be ween 0 and 100m wa e dep h. Samples we e aken using a ca ousel o 24 Niskin bo les in e aced wi h he SBE911 CTD sys em (SeaBi d) ha p o ided he hyd og aphical da a o in si u empe a- u e, salini y, and dissol ed oxygen o he seawa e samples. Samples collec ed a <1m wa e dep h we e ob ained by illing a 5L plas ic con aine wi h su ace wa e . Be ween 1.5 and 4.5L o wa e was gen ly il e ed on o ace a e cellu- lose memb anes (Millipo e, 0.45µm po e size, 47mm diam- e e ). Memb ane il e s we e o en-d ied a 40◦C o ∼12h and s o ed in sealed Pe i dishes. Figu e 1 shows he lo- ca ion o all sampled s a ions du ing he c uise ajec o y. Table 1 shows he s a ions coo dina es, da es, and sampled dep hs. The bo le da a can be ound a h p://cdiac.o nl.go / p/oceans/CLIVAR/Me _84_3_Med_Sea/. 2.2 Phy oplank on analyses A po ion o each il e was placed on aluminum s ubs and gold-coa ed using an EMITECH K550X spu e coa e . The quan i ica ion and iden i ica ion o he main phy oplank- on g oups and coccoli hopho e species we e pe o med by JEOL-JSM 6300 and ZEISS-EVO MA10 scanning elec- on mic oscopes (SEMs). T ansec s o 5–15mm on he il e , co esponding o an a e age o 2.3mL o seawa e , we e obse ed a 3000x and phy oplank on g oups quan i- ied as coccoli hopho es, dia oms, and silico lagella es (Dic- yocha spp.). A subse o i e andom samples was analyzed unde a ligh mic oscope o es ima ion o coccoli hopho e and dia om cell abundance a 1000x. This was done in o - de o check whe he coun s a highe magni ica ion we e biased owa ds smalle cells. This compa ison showed ha Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s 15 Table 1. Loca ion, da e, and wa e dep h om which he samples we e collec ed. S a ion Mon h/day/yea Longi ude La i ude Bo . dep h Sampled dep h (m) (◦E) (◦N) (m) 100 50 25 5 0 287 4/6/2011 25.60 37.67 829 X∗X X 288 4/7/2011 26.22 35.65 2293 X X 291 4/8/2011 33.00 34.07 2474 X X X X 292 4/9/2011 35.17 33.99 1681 X 293 4/9/2011 34.42 34.00 2034 X 294 4/10/2011 31.00 33.70 2437 X X X X X 296 4/11/2011 28.77 33.58 2934 X 297 4/11/2011 26.02 34.40 4210 X X X 298 4/12/2011 24.33 34.50 3287 X X 299 4/12/2011 22.50 35.00 3117 X 302 4/13/2011 20.35 35.07 2968 X X X X X 305 4/14/2011 17.25 35.60 4440 X X 306 4/15/2011 19.00 36.50 3445 X X X 307 4/15/2011 19.30 37.90 3305 X X 308 4/15/2011 19.00 38.50 3462 X X X X 309 4/16/2011 18.80 39.50 805 X X X 313 4/16/2011 18.00 41.25 1105 X X 316 4/19/2011 11.50 38.60 1665 X X 319 4/20/2011 11.30 40.30 2880 X X 320 4/20/2011 10.61 38.75 2490 X X 321 4/20/2011 9.40 38.25 1565 X X X X 324 4/21/2011 5.60 38.65 2845 X X X X X 329 4/23/2011 2.00 37.90 2730 X X X X X 331 4/23/2011 0.00 37.05 2704 X X X X X 332 4/24/2011 −1.40 36.50 2340 X X 334 4/24/2011 −4.40 36.10 1228 X X X X 337 4/25/2011 −5.36 36.00 935 X X X X X 338 4/25/2011 −5.75 35.95 337 X ∗75m Figu e 1. Sampling s a ions p esen ed in his s udy collec ed du ing he M84-3 esea ch c uise. A composi e image is supe imposed showing he su ace chlo ophyll concen a ion (mgm−3)a an app oxima e da e o he sampling pe iod in he di e en basins. The ansec shown in he ollowing igu es includes he s a ions labeled in black. www.ocean-sci.ne /11/13/2015/ Ocean Sci., 11, 13–32, 2015 16 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s ela i e abundances o dia oms we e simila in he wo me h- ods. They comp ised be ween 1.7 and 3.3% (SD=0.85) o cells coun ed ia op ical mic oscopy and be ween 0.8 and 4.1% (SD=1.6) o cells coun ed ia SEM. Coccoli hopho e species we e iden i ied and hei absolu e and ela i e abun- dances coun ed. In samples wi h e y ew coccosphe es a la ge il e po ion was obse ed in o de o quan i y a min- imum o 100 cells (a maximum numbe o 420 cells we e coun ed). Lowe (CL)and uppe (CU)con idence in e als a 95% signi icance we e es ima ed ollowing Bollmann e al. (2002). Fo a 100-cell coun hese we e CL=82 and CU=102, and o a 420-cell coun : CL=382 and CU=422. Cell densi ies (numbe o cells pe li e o seawa e ) we e calcula ed. Emiliania huxleyi was sub-classi ied in o mo - pho ypes acco ding o Young e al. (2003). Fo each sam- ple he Shannon–Wiene di e si y index (H’) was calcula ed o he e ococcoli hopho es and holococcoli hopho es. These wo g oups we e ea ed sepa a ely because hey ep esen wo di e en s ages o a coccoli hopho e’s li e cycle,and ax- onomy be ween he wo does no always accoun o i . 2.3 En i onmen al pa ame e s A de ailed p o ocol o all measu ed en i onmen al a iables can be ound in Tanhua e al. (2013a). In si u salini y, em- pe a u e, and oxygen da a we e measu ed wi h a CTD (de- sc ibed in Sec . 2.1). O e all da a accu acies we e 0.002◦C o empe a u e and 0.003 o salini y. Mac onu ien s (phos- pha e and ni a e and silica e concen a ions) we e measu ed onboa d wi h a QuAA o au oanalyze om SEAL Analy - ical. The ollowing p o ocols om SEAL Analy ical we e ollowed: ni a e (NO3)(me hod no. Q-068-05 e . 4), phos- pha e (PO3− 4)(me hod no. Q-031-04 e . 2), and silicium (Si) (me hod no. Q-066-05 e . 3). The nu ien analy ical e o was de e mined on 5–7 sample eplica es aken a selec ed s a ions. The e o o ni a e is 0.08µmolkg−1, o phos- pha e 0.007µmolkg−1, and o silica e 0.10µmolkg−1. The ca bona e sys em was cha ac e ized by measu ing dis- sol ed ino ganic ca bon (DIC), pH, and o al alkalini y (TA). DIC con en was measu ed coulome ically using a SOMMA (single-ope a o mul i-me abolic analyze ) sys em. The p e- cision o he analysis is ±0.6µmolkg−1and he accu acy is 2.5µmolkg−1. The pH was measu ed by means o double- wa eleng h spec opho ome y, and i is epo ed a 25◦C on he o al scale. The ep oducibili y o he pH measu emen s was 0.0012. TA was analyzed ollowing a double-end-poin po en iome ic echnique. The p ecision o he TA measu e- men s was 0.1µmolkg−1. Mo e de ails abou he DIC, TA, and pH measu emen s and quali y con ol a e p esen ed in Ál a ez e al. (2014). In si u condi ions o o he CO2- ela ed a iables we e calcula ed om in si u pH and TA o he i s 100m wa e column. Calcula ions we e pe o med using he p og am CO2Sys (Lewis and Wallace, 1998). Equilib ium cons an s o Meh bach e al. (1973) e i ed by Dickson and Mille o (1987) we e chosen (Ál a ez e al., 2014). A cha ac e iza ion o he uppe 100m en i onmen al pa- ame e s is shown in Fig. 2; he p o iles o he comple e wa- e column a a highe spa ial ho izon al esolu ion and a ull desc ip ion o he physicochemical se ing a e p esen ed in Tanhua e al. (2013b) and Ál a ez e al. (2014) in his Ocean Science special issue. 2.4 S a is ical analyses The E-PRIMER .5 package was used o he ollowing anal- yses: –(1) The BIOENV ou ine, which compu es a ank co - ela ion be ween he elemen s o simila i y ma ices o en i onmen al pa ame e s and biological da a, was un o de ec he combined changes in en i onmen al pa- ame e s and species dis ibu ion among s a ions. The ou ine examines all possible combina ions o en i on- men al a iables and gi es he “bes i ” (highe Rho– Spea man ank co ela ion) o en i onmen al a iables explaining changes in biological communi ies. This es was pe o med o all he e ococcoli hopho e and holo- coccoli hopho e species con ibu ing >2% o he o al assemblage o each g oup. Be o e unning he ou ine, we checked o mu ual co ela ion among en i onmen- al a iables and selec ed a subse o hem o his ou- ine. These we e salini y, empe a u e, oxygen, pH, pa - ial p essu e o ca bon dioxide (pCO2), and he con- cen a ions o bica bona e ion (HCO− 3), ca bona e ion (CO−2 3), NO3+NO2, and PO3− 4. –(2) Hie a chical clus e analyses by g oup a e age. These we e pe o med o he e ococcoli hopho e and holococcoli hopho e species. Emiliania huxleyi mo - pho ype A was emo ed o he da a se used o un he clus e analyses. This was done o emphasize ou esul s on o e all communi y composi ion and no on E. huxleyi , which la gely domina ed he assemblages in ou samples. When clus e s among species we e de- ec ed, pai -wise Spea man co ela ions we e pe o med using he so wa e SPSS 18 o assess he en i onmen al pa ame e s in luencing changes in each species’ abun- dance. Fo he analyses pe o med using E-PRIMER so wa e, he biological da a we e ans o med in loga i hmic scale log(1+x) o a oid o e emphasizing he dominan species. En i onmen al da a we e s anda dized (−mean·STD) o b ing da a in o a compa able scale. Simila i y ma ices we e c ea ed o biological and en i onmen al da a. Fo he bio- logical da a he B ay–Cu is simila i y coe icien was used o examine simila i y be ween each sample’s pai . Euclidean dis ances we e used o c ea e he en i onmen al da a ma ix. Pai -wise Spea man co ela ions we e pe o med on he basis o non- ans o med, non-s anda dized da a. Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s 17 Figu e 2. En i onmen al pa ame e s a sampling s a ions o su ace wa e s du ing he M84-3 c uise. The wes –eas ansec along he Medi e anean Sea does no ake in o accoun he s a ions in he Aegean, Ad ia ic, he no he nmos Ionian, o no he n Ty henian egions. I includes he s a ions labeled in black (wes –eas ansec ) in Fig. 1. 3 Regional se ings The Medi e anean Sea can be di ided in o wo basins wi h a gene al coun e clockwise ci cula ion: he wes e n and eas - e n basins. The S ai o Sicily p e en s deep-wa e exchange be ween he wo. The uppe laye ( om he su ace down o 150–200m, acco ding o he basin and he season) is oc- cupied by he A lan ic Wa e (AW) (Ribe a d’Alcalà e al., 2003). In he wes e n Medi e anean, AW lows along he eas e n pa o he Albo an Sea in an an icyclonic ashion, and mo es o he wes e n Albo an in a mo e a iable pa - e n (Robinson e al., 2001). Fu he eas , he AW is ans- po ed along he Alge ian slope un il he Sa dinia Channel. Pa o he low hen p oceeds o he eas e n basin along he Tunisian slope, while he o he pa eaches no he n Sicily and ci cula es along I aly. In he no h, he Ligu o–P o enco– Ca alan Cu en , is o m by AW su ounding Co sica. AW closes i s wes e n gy e sou heas o Spain, whe e i encoun- e s he newly en e ed AW (see Robinson e al., 2001, o mo e de ails). In he eas e n basin, a je o A lan ic Wa e en- e s h ough he S ai o Sicily, meande s h ough he in e io o he Ionian Sea and con inues o low h ough he cen al Le an ine all he way o he sho es o Is ael (Robinson e al., 2001). Con ec ion occu s when he wa e app oaches he Rhodos gy e. Ano he pa o he AW low mo es o he Io- nian and a pa o i goes u he in o he Ad ia ic Sea. Deep- wa e o ma ion also occu s in he Ad ia ic gy e (Pina di and Mase i, 2000). Du ing he M84/3 c uise, he su ace laye in he wes e n basin was illed by he in low o ela i ely low-salini y AW h ough he S ai o Gib al a . The salini y minimum, in he su ace laye o he s ai , shows he en ance o his wa e in o he Medi e anean. Once i eaches he Sicily Channel (close o s a ions 304/305), he salini y minimum was ound a ∼100m dep h. (Hainbuche e al., 2014). In e ms o ca - bona e chemis y, AW om he su ace wes e n Medi e - anean was cha ac e ized by TA below 2560µmolKg−1, DIC nea 2250µmolKg−1, and pH o 8 (Àl a ez e al., 2014). In he su ace laye , AW wi h salini y lowe han 38 was de- ec ed in all he Ty henian s a ions excep 319 in he cen- e o he basin, whe e salini y was highe . The e, Ál a ez e al. (2014) sugges ha he uppe ∼50dba was occu- pied by p obably AW a ec ed by e apo a ion o mixed wi h he sal y in e media e wa e en e ing h ough he S ai o Sicily. Du ing he M84/3 c uise, modi ied AW and Le an ine Su ace Wa e (LSW) domina ed he eas e n Medi e anean, down o app oxima ely 100m wa e dep h. The LSW, o med www.ocean-sci.ne /11/13/2015/ Ocean Sci., 11, 13–32, 2015 18 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s ! 39! Figu e 3. Daily mean chlo ophyll-a concen a ion om sa elli e da a (Volpe e al., 2012) along he ansec o he M84/3 c uise, plo ed o he mon h o Ap il 2011. In his igu e, gaps co espond o cloud co e ed obse a ions and he empo al-spa ial loca ion o he samplings du ing ou s udy is indica ed by black do s. The ansec is ep esen a i e o 18Km in la i ude, ( wo pixels a 9Km esolu ion). Figu e 3. Daily mean chlo ophyll aconcen a ion om sa elli e da a (Volpe e al., 2012) along he ansec o he M84/3 c uise, plo ed o he mon h o Ap il 2011. In his igu e, gaps co espond o cloud-co e ed obse a ions and he empo al–spa ial loca ion o he samplings du ing ou s udy is indica ed by black do s. The ansec is ep esen a i e o 18km in la i ude ( wo pixels a 9km esolu ion). by in ensi e hea ing and e apo a ion, has he la ges salin- i y and empe a u e o he en i e Medi e anean (>17◦C, >38.9), and has TA alues a ound 2610µmolKg−1, DIC a ound 2270µmolKg−1, and pH25T a ound 8.03 (Ál a ez e al., 2014). Le an ine In e media e Wa e (LIW) was de ec ed a ∼50–100m in he Le an ine Sea, along i s pa h om he no heas e n Le an ine Sea, and in he Rhodes cyclonic gy e (Hainbuche e al., 2014, and e e ences he ein). Close o he M84/3 ajec o y, wind-induced upwelling and downwelling zones a e ound in di e en seasons (see Bakun and Agos- ini, 2001): o ins ance, (1) downwelling along he sou h- e n coas al bounda y du ing win e , e e sing o coas al up- welling o Libya in sp ing and o Alge ia in summe and all; (2) s ong upwelling in he eas e n Aegean Sea h ough- ou he yea , becoming ema kably in ense in he summe and all; and (3) s ong summe and all upwelling in he eas e n Ionian Sea. Figu e 3 shows daily a ia ions in cloud co e age du - ing he mon h o Ap il ( om sa elli e obse a ions). I p o- ides an o e iew o he me eo ological si ua ion and pos- sible sho - e m a iabili y ha may mask any ela ionship wi h long- e m a e age oceanog aphic g adien s. App oxi- ma ely hal o he sa elli e obse a ions we e cloud co e ed, wi h a minimum in cloud co e age (∼40% o obse a ions) a abou 25◦E (sou h o C e e). This minimum co esponds o he only segmen o he ansec di ec ly in luenced by no he n wind egimes. Such app oxima ion can be com- pa ed wi h he me eo ological epo s du ing he M84/3 (da a no shown), which shows ha ∼58% o he sampling days we e cloudy (a he ime o eco ding), wi h he majo i y o sunny days om sou he n C e e owa ds he Ad ia ic. The e- o e, he sampled days we e ep esen a i e, in e ms o cloud co e age, o he condi ions du ing Ap il 2011. 4 Resul s 4.1 Main phy oplank on communi y O e all, phy oplank on cell densi y was highes in he wes - e n Medi e anean Sea, in he S ai o Gib al a and in he Albo an Sea. Fo coccoli hopho es, he pa e n was no one o g adual inc ease owa ds he wes bu a he o local- ized spo s o highe cell densi ies close o Gib al a and a con inuous p esence, a lowe cell densi ies, in he es o he Medi e anean. The sa elli e-de i ed chlo ophyll acon- cen a ion du ing Ap il 2011 (Fig. 3) shows a simila pa - e n. Coccoli hopho es we e he mos abundan phy oplank- on g oup du ing he sampling, ela i e o siliceous phy- oplank on. They we e p esen in g ea numbe s in all he main Medi e anean basins and accoun ed o 68 o 99% o coun ed phy oplank on. Dia oms, al hough p esen in all s udied basins, displayed low concen a ions in he eas - e n Medi e anean, inc easing owa ds he wes . They we e on a e age 6% (maximum 25%) o o al phy oplank on. Silico lagella es (Dic yocha spp.) accoun ed on a e age o Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s 19 ! 40! Figu e 4. Dis ibu ion o coccoli hopho es (uppe panel), dia oms (middle panel) and silico lagella es (lowe panel) in a wes o eas ansec along he Medi e anean Sea ( he ansec includes he s a ions labeled in black in Figu e 1). Figu e 4. Dis ibu ion o coccoli hopho es (uppe panel), dia oms (middle panel), and silico lagella es (lowe panel) in a wes –eas ansec along he Medi e anean Sea ( he ansec includes he s a- ions labeled in black in Fig. 1). 1% o phy oplank on (maximum 9%). Figu e 4 shows he dis ibu ion o hese di e en phy oplank on g oups in he Medi e anean ansec . A species p esen mos ly a low cell densi ies was he xan hophy e Me ingosphae a medi e - anea. 4.2 En i onmen al ac o s con olling coccoli hopho e communi y dis ibu ion A o al o 70 coccoli hopho e species in he e ococcol- i hopho e li e s age and 45 in he holococcoli hopho e s age we e eco ded (see Appendix A). The majo i y o he quan- i ied cells we e in he he e ococcoli hopho e phase. The species Emiliania huxleyi la gely domina ed he coccol- i hopho e coun s in all s a ions excep o s a ion 319 (a he cen e o he Ty henian Basin), whe e Co isphae a g acilis and Rabdosphae a cla ige a we e he dominan species. The esul s om he Spea man’s ank-co ela ion-based ou ine (BIOENV) sugges ha holococcoli hopho es we e p e e en ially dis ibu ed in low-nu ien , high-pH seawa e s. (ρ=0.328, Table 2). They we e almos absen a 100m. He - e ococcoli hopho e dis ibu ion was bes linked o a combi- na ion o [CO2− 3], pH, and salini y wi h ρ=0.566 as well as o [PO3− 4] (Table 3). 4.3 Species assemblages When clus e ing all he species wi hin he he e ococcol- i hopho e li e s age, h ee g oups o assemblages we e iden i- ied ha sha ed mo e han65%simila i ies. Figu e5p esen s he esul s o u he clus e ing o he species wi h high simi- la i ies. The dis ibu ion o hese species along he eas –wes ansec (s a ions wi h black labels in Fig. 1) ha includes he s a ions in he Le an ine, Ionian (excluding 307–309), Ty henian (excluding 319), Alge ian, Albo an, and Gib al- a egions shows ha he h ee g oups a e dis inc i ely dis- ibu ed in he Medi e anean Sea (Figs. 5, 6, and 7). The i s g oup comp ises species ha we e mo e abundan in he eas e n s a ions: U. enuis,D. ubi e a,P. andelii,S. pul- ch a,R. cla ige a, and S. p o udens (R. xyphos was e y close o his clus e , wi h simila i y >50%). P. andelii, how- e e , was pa chily dis ibu ed all along he ansec . The sec- ond g oup includes E. huxleyi Type B/C and he Gephy o- capsa species: G. e icsonii,G. muelle ae, and G. oceanica. These species we e almos exclusi e o he wes e n basin. A hi d g oup was o med by Flo isphae a p o unda and Gladioli hus labella us, being closely ela ed o A. obus a and H. ca e i ( he las wo wi h simila i ies be ween 40 and 60%). The species in he la e g oup we e almos es ic ed o dep hs below 50m, wi h highe abundances a 100m, and we e pa chily p esen om he Alge ian o he Le an ine basins. Clus e ing analysis o he holococcoli h phase did no e eal any pa e n in he species composi ion among he di e en samples. Single Spea man co ela ions o he species ha we e clus e ed oge he e eal ha hei dis ibu ion can be be e explained by seawa e ca bona e chemis y pa ame e s; o ins ance, species ha we e mos ly abundan a eas e n s a- ions h i ed in wa e s wi h highe [CO2− 3] and pH and in he su ace. Among hese species, U. enuis,R. cla ige a,S. pul- ch a, and S. p o udens we e nega i ely co ela ed o phos- pha e concen a ions and only D. ubi e a showed a high pos- i i eco ela ionwi h empe a u e. Finally,F.p o unda and G. labella us we e co ela ed o [NO3+NO2] and nega i ely wi h empe a u e. Table 4 shows hese esul s. 4.4 Species di e si y (H’) He e o- and holococcoli hopho e species di e si y index (H’) changed sligh ly in he W–E ansec . Al hough co ela ions be ween H’ and he longi ude o he sample si es (◦E) in he i s 50m wa e column we e a he weak, he end was opposi e o he wo li e s ages. Fo ins ance, he e ococ- coli hopho e di e si y ended o dec ease owa ds he eas (ρ= −0.419, p=0.000), while holococcoli hopho e species di e si y ended o inc ease W–E (ρ=0.310; p=0.005) (Fig. 9). H’ index a 100m was o en ze o o bo h www.ocean-sci.ne /11/13/2015/ Ocean Sci., 11, 13–32, 2015 20 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s Table 2. Rho alues (ρ) o he bes Spea man’s ank co ela ions o all possible combina ions be ween he en i onmen al pa ame e s explaining pa e ns in holococcoli hopho e assemblages. Only ρ>0.2 a e shown. Desc ip ion as in Table 1. On he le , he numbe o a iables aken in o accoun ; on he igh , desc ip ion o he a iables and he highes ρ alues o he ank co ela ions a a gi en numbe o a iables. In he i s ow, he ρ alue o each a iable is shown in dec easing o de o con ibu ion o explain changes in he biological da a. Only ρ>0.2 a e shown. Numbe o Va iables (ρ) a iables 1 NO− 3+NO− 2(0.275); PO3− 4(0.236); pH (0.214); pCO2(0.204); 2 NO− 3+NO− 2pH (0.327) 3 NO− 3+NO− 2PO3− 4, pH (0.328) 4 NO− 3+NO− 2pH, O2, Salini y (0.311) 5 NO− 3+NO− 2PO3− 4, pH, O2, salini y (0.311) 6 NO− 3+NO− 2PO3− 4, pH, pCO2, O2, salini y (0.299) 7 NO− 3+NO− 2PO3− 4, pH, pCO2, O2, salini y, Tempe a u e (0.283) 8 All (0.275) Table 3. Rho alues (ρ) o he bes Spea man’s ank co ela ions o all possible combina ions be ween he en i onmen al pa ame e s explaining pa e ns in he e ococcoli hopho e assemblages. Desc ip ion as in Table 2. Numbe o Va iables (ρ) a iables 1 CO2− 3(0.551); pH (0.498); pCO2(0.397); PO3− 4(0.358); NO− 3+NO− 2(0.328); salini y (0.310); O2(0.226); 2 CO2− 3salini y (0.563) 3 CO2− 3pH, salini y (0.566) 4 CO2− 3pH, PO3− 4, salini y (0.565) 5 CO2− 3pH, pCO2, PO3− 4, salini y (0.539) 6 CO2− 3pH, pCO2, PO3− 4, salini y, empe a u e (0.517) 7 CO2− 3pH, pCO2, PO3− 4, salini y, O2, empe a u e (0.507) 8 All (0.491) g oups, being on a e age 0.3 o holococcoli hopho es, which a e mos ly p esen a su ace, and 1.3 o he e ococ- coli hopho es. 5 Discussion 5.1 Main phy oplank on communi y Al hough picoplank on can seasonally domina e phy oplank- on assemblages in he Medi e anean Sea (Decemb ini e al., 2009; Yacobi e al., 1995), p e ious s udies ha e o en sug- ges ed ha coccoli hopho es a e one o he mos abundan phy oplank on g oups in his sea, in bo h he eas e n (e.g., Go sis-Sk e as e al., 1999; Malin e no e al., 2003; Igna- iades e al., 1995, 2009; Rabi i e al., 1994; Zi e i e al., 2000) and wes e n basins (e.g Ba low e al., 1997; Ba cena e al., 2004, Igna iades e al., 2009). Ou indings sugges ha coccoli hopho es a e a main phy oplank on g oup in he en i e Medi e anean Sea ha domina ed o e siliceous phy- oplank on o he pe iod unde s udy (Fig. 4). Silico lagel- la es we e almos absen a 100m and mo e abundan a su - ace wa e s o he Ty henian Sea, wi h cell densi ies up o 6.7×103cellsL−1. Dia oms we e p e e en ially dis ibu ed in he wes e n Medi e anean Sea, in wa e s wi h lowe [CO2− 3] pH and highe PO3− 4a a maximum cell densi y o 1.6×104cellsL−1(Fig. 4). Dia om cell densi ies in he Medi e anean anged om a ew indi iduals in he Ionian Sea su ace o ∼3.5×104in he Ty henian Sea (June 1999; Igna iades e al., 2009), om 0.8×103 o 2×104(July 2005, Decemb ini e al., 2009), om 0.2×103 o 5×103 (Sep embe –Oc obe 2004, Las e nas; e al., 2011), and om 0.7×103 o 2.6×103cellsL−1(Decembe 2005; Decem- b ini e al., 2009) in he Ty henian Sea. Wi hou aking in o accoun in e annual a iabili y, his sugges s a pa e n o en- hanced dia om abundances in he i s pa o he yea and a dec ease a e July. In he las decades, many a emp s o explain he empo- al and spa ial phy oplank on dis ibu ion based on ew key en i onmen al ac o s ha e been made. Some o he mos discussed a e he hypo hesis o S e d up (1953), s a ing ha changes in phy oplank on abundances a e ligh - and nu ien - con olled ea u es, and he so-called Ma gale ’s mandala (Ma gale , 1978), which p oposes ha phy oplank on succes- sion depends on nu ien concen a ion and u bulence. Re- ga ding he possible ole o ligh in he ela i e success o Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s 21 Table 4. Th ee highes Spea man’s co ela ion (ρ) esul s o (1) he dominan species Emiliania huxleyi, (2) each o he species belonging o he obse ed clus e s wi h he en i onmen al pa ame e s, (3) he holo- and he e o-li e phases o he same species, and (4) he e o- and holococcoli hopho e abundances and di e si y (H’) and abundances o siliceous phy oplank on g oups. P e e en ial dep h and dis ibu ion along he wes –eas ansec is desc ibed. Numbe o samples aken in o accoun o he analysis is also shown (N). Signi icance le el was 0.05. Species/g oup Va iable ( ,N) P e . dep h P e . dis ibu ion E. huxleyi Type A pH (0.468; 73) CO2(-0.448; 73), CO2− 3(0.444; 73), 0–100m W–E E. huxleyi Type B/C Salini y (−0.691; 81), CO2− 3(−0.685;73), pH (−0.616; 73) 0–100m W G. e icsonii Salini y (−0.805; 81), CO2− 3(−0690; 73), pH (-0.576; 73) Abo e 50m W G. oceanica CO2− 3(−0.833; 73), Salini y (−0.803; 81), pH (−0.731; 73) 0–100m W G. muelle ae CO2− 3(−0.835; 73), Salini y (−0.830; 81), pH (−0.707; 73) 0–100m W U. enuis CO2− 3(0.626; 73), pH (0.605; 73), PO3− 4(-0.551; 74) Abo e 50m E R. cla ige a CO2− 3(0.727; 73), pH (0.701; 73), PO3− 4(-0.546; 74) Abo e 50m E D. ubi e a pH (0.600; 73), Tempe a u e (0.597; 81), CO2− 3(0.541; 73) Abo e 50m E S. pulch a pH (0.551; 73), CO2− 3(0.528; 73), PO3− 4(-0.399; 74) 0–100m E S. p o udens pH (0.557; 73), CO2− 3(0.489; 73), PO3− 4(-0.450; 74) Abo e 50m E F. p o unda NO− 3+NO− 2(0.637; 81), Tempe a u e (−0.553; 81), pH (−0.468; 73) Below 50m W–E G. labella us NO− 3+NO− 2(0.583; 81), Tempe a u e (−0.522; 81), O2(−0.481; 73) Below 50m W–E S. pulch a HOL PO3− 4(−0.308; 74) CO2(−0.261; 73), O2(0.260; 73) Abo e 50m W–E C. medi e anea HCO− 3(-0.294; 73) – – Abo e 50m W–E C. medi e anea HOL O2(0.240; 73) – – Abo e 50m W–E H. ca e i CO2(0.341; 73), pH (−0.339; 73), – 0–100m W–E H. ca e i HOL NO− 3+NO− 2(−0.284; 81), – – Abo e 50m W–E He e o-abund. CO2(−0.278; 73), – – 0–100m W–E Holo-abund. Dep h (−0.427; 81), NO− 3+NO− 2(−0.391; 81), CO2(0.339; 73) Abo e 50m E He e o-di e si y (H’) CO2− 3(−0.486; 73), Salini y (−0.426; 81), pH (−0.413; 73) – – Holo-di e si y (H’) HCO− 3(−0.460; 74) CO2− 3(0.446; 73), pH (0.388; 73) – – Dia oms CO2− 3(−0.617; 73), pH (−0.602; 73), PO3− 4(0.534; 74) Below 25m W Silico lagella es Si (0.254; 76) – – 0–100m W–E coccoli hopho es o e siliceous phy oplank on, we eco ded global and UV adia ion a he su ace, and a he ime o sampling a some o he s a ions, bu lack ligh da a a e - e y dep h. No signi ican co ela ion was obse ed o any o he phy oplank on g oups wi h adia ion da a (p>0.05, N=21). Mos likely, a esponse o changes in adia ion will no be immedia e (specially because he a iable ha we eco ded – cell densi ies – ela es o cell di ision). In such a case he sho empo al ex ension o ou da a (by minu e du - ing he sampling) could mask a co ela ion. La ge a ia ions in ligh a e also obse ed on he e ical p o ile. Holococ- coli hopho es and dia oms co ela ed wi h dep h (nega i e and posi i e co ela ions, espec i ely), al hough o dia oms his co ela ion was a he weak (ρ=0.240; p=0.031). This could indica e ha he wo g oups a e mo e sensi i e o ligh condi ions han he e ococcoli hopho es, bu i migh also be an a i ac o he co-co ela ion be ween dep h and nu ien s. Addi ionally, ligh a enua ion can di e be ween s a ions, and he e o e his compa ison is only app oxima e. In he case o nu ien s, i is impo an o no ice ha he dominance o coccoli hopho es o e siliceous phy oplank- on was clea in all he basins, including he Gib al a s ai , whe e ni a e and phospha e we e a ailable and silica e con- cen a ions we e abo e he hal -sa u a ion cons an o di- a oms (Ks_Si: ca. 3.5µM (Me ico e al., 2006; Leblanc e al., 2003; Sa hou e al., 2005) o 0.8–2.3µM (Nelson e al., 1976)) and we could ha e expec ed he communi y o be domina ed by as -g owing phy oplank on. In he in e- io o he Medi e anean, [Si] was p obably oo low (o - en below 1.0µmolKg−1) o suppo la ge dia om popula- ions, excep o he deepe laye s (100m) o he Ionian and Le an ine basins, which anged om 0.8 o 1.4µmolKg−1. Egge and Aksnes (1992) obse ed ha , a Si alues lowe han 0.6µM, Emiliania huxleyi ou compe ed he o he wise dominan Skele onema cos a um. Fo he da a se he e p e- sen ed, nu ien a iabili y alone does no explain he domi- nance o coccoli hopho es du ing Ap il 2011, a leas no di- ec ly. A possible phospha e limi a ion o o he phy oplank- on g oups (no ice he high N/P a ios in bo h he eas e n and wes e n basins as well as in he S ai o Gib al a ) canno be uled ou o explaining coccoli hopho e dominance o e o he g oups. Tu bulence does no accoun o i ei he , since coccoli hopho es domina ed in egions whe e wa e densi y was homogeneous h oughou he i s 100m as well as in egions whe e isoclines can be dis inguished (e.g., see salin- i y and empe a u e p o iles om Fig. 2). O e all, he eason o he dominance o coccoli hopho es o e siliceous phy o- plank on emains unclea . www.ocean-sci.ne /11/13/2015/ Ocean Sci., 11, 13–32, 2015 28 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s Table A1. Con inued. [37] Co onosphae a binoda a (Kamp ne ) Gaa de [38] C. medi e anea (Lohmann) Gaa de [39] Michaelsa sia ad ia icus (Schille ) Man on, B eme e Oa es [40] M. elegans G an, emend. Man on, B eme e Oa es [41] Ophias e o mosus G an [42] Ophias e hyd oideus G an [43] Sy acosphae a amplio a Okada e McIn i e [44] S. an hos (Lohmann) Jo dan e Young [45] S. bannockii (Bo se i e Ca i) C os, Kleijne, Zel ne , Billa d e Young [46] S. bo ealis Okada e McIn i e [47] S. co olla Lecal [48] S. delica a C os, Kleijne, Zel ne , Billa d e Young [49] S.dila a a Jo dan, Kleijne e Heimdal [50] S. his ica Kamp ne [51] S. lamina Lecal-Schlaude [52] S. ma ginopo a a Knappe sbusch [53] S. molischii Schille [54] S. nana (Kamp ne ) Okada e McIn y e [55] S. nodosa Kamp ne [56] S. no oi ica Knappe sbusch [57] S. ossa (Lecal) Loeblich e Tappan [58] S. pi us Halldal e Ma kali [59] S. p olonga a G an, ex Lohmann [60] S. p o udens Okada e McIn y e [61] S. pulch a Lohmann [62] S. o ula Okada e McIn i e [63] Sy acosphae a sp. ype D, sensu Kleijne [64] S. umula is Sánchez- Suá ez [65] Sy acospahae a sp. Sub-Family Umbellosphae oideae Kleijne [66] Umbellosphae a enuis (Kamp ne ) Paasche [67] F. p o unda Okada e Honjo [68] Gladioli hus labella us (Halldal e Ma kali) Jo dan e G een [69] Polyc a e galapagensis Man on e Oa es [70] Ce a oli hus c is a us No is Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s 29 Table A2. Coccoli hopho e species lis . Holococcoli hopho es. [1] Acan hoica qua ospina HOL =sp. a Sphae ocalyp a o C os e al. (2000) [2] An hosphae a aga ia Kamp ne emend. Kleijne [3] A. la ou cadii (Lecal) Kleijne [4] A. pe ipe o a a Kleijne [5] An hosphae a sp. Type A o C os and Fo uño (2002) [6] An hosphae a sp. Type C o C os and Fo uño (2002) [7] Calcidiscus lep opo us HOL (Mu ay e Blackman) Loeblich e Tappan [8] C. lep opo us ssp. quad ipe o a us HOL (Kamp ne ) Geisen, Billa d, B oe se, C os, P obe e Young [9] Calicasphae a conca a Kleijne [10] Calyp oli hina di e gens (Halldal e Ma kali) Heimdal [11] C. di e gens a . ube osa (Heimdal) Jo dan, Kleijne e Heimdal [12] C. mul ipo a (Gaa de ) No is [13] Calyp oli hopho a papilli e a (Halldal) Heimdal [14] Calyp osphae a cialdii Bo se i e Ca i [15] C. den a a Kleijne [16] C. heimdalae No is [17] C. sphae oidea Schille [18] Coccoli hus pelagicus ssp. b aa udii HOL (Gaa de ) Geisen, Billa d, B oe se, C os, P obe e Young [19] Co isphae a g acilis Kamp ne [20] C. s igilis Gaa de [21] C. y heniensis Kleijne [22] Co isphae a sp. Type A o Kleijne (1991) [23] Co onosphae a medi e anea HOL g acillima- ype =Calyp oli hopho a g acillima (Kamp ne ) Heimdal [24] Co onosphae a medi e anea HOL hellenica- ype =Zygosphae a hellenica Kamp ne [25] Gliskoli hus ami aka enae No is, o hog. emend., Jo dan e G een [26] Helicosphae a ca e i HOL =Sy acoli hus ca illi e us (Kamp ne ) De land e [27] Helladosphae a co ni e a (Schille ) Kamp ne [28] Homozygosphae a a e husae (Kamp ne ) Kleijne [29] H. spinosa (Kamp ne ) De land e [30] H. ia cha Halldal e Ma kalii [31] Homozygosphae a e cellii Bo se i e Ca i [32] Po icalyp a gaa de ae (Bo se i e Ca i) Kleijne [33] Po i ec oli hus sp. 2 o C os and Fo uño (2002) [34] Sphae ocalyp a adenensis Kleijne [35] S. quad iden a a (Schille ) De land e [36] Sphae ocalyp a sp. 1 o C os and Fo uño (2002) [37] Sphae ocalyp a sp. 3 o C os and Fo uño (2002) [38] Sphae ocalyp a sp. 6 o C os and Fo uño (2002) [39] Sy acoli hus schille i (Kamp ne ) Loeblich e Tappan [40] Sy acoli hus sp. ype A o Kleijne (1991) [41] Sy acosphae a an hos HOL =Pe iphyllopho a mi abilis (Schille ) Kamp ne [42] Sy acosphae a bannockii HOL =Zygosphae a bannockii (Bo se i e Ca i) Heimdal [43] Sy acosphae a pulch a HOL oblonga- ype =Calyp osphae a oblonga Lohmann [44] Sy acosphae a pulch a HOL pi us- ype =Calyp osphae a pi us Kamp ne [45] Zygosphae a amoena Kamp ne www.ocean-sci.ne /11/13/2015/ Ocean Sci., 11, 13–32, 2015 30 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s Acknowledgemen s. We hank he c ew membe s and scien is s onboa d he R/V Me eo du ing he M84/3 campaign o making a mon h o sampling possible and a wa m and welcoming ime onboa d. Thanks o Vincen Taillandie o helping wi h he chlo ophyll da a. This wo k was pa ially unded by he Eu o- pean Communi y’s Se en h F amewo k P og amme unde g an ag eemen 265103 (p ojec MedSeA) and by p ojec HOTMIX (CTM2011-30010-C02). The au ho s also acknowledge he “Agen- cia de Ges ión de Ayudas Uni e si a ias y de In es igación” (Becas FI) o he Ca alunya Gene ali a . We hank Sebas ian Meie and he anonymous e iewe o hei commen s ha g ea ly en iched he manusc ip . Edi ed by: N. K ess Re e ences Ál a ez, M., Sanleón-Ba olomé, H., Tanhua, T., Min op, L., Luche a, A., Can oni, C., Sch oede , K., and Ci i a ese, G.: The CO2sys em in he Medi e anean Sea: a basin wide pe spec i e, Ocean Sci., 10, 69–92, doi:10.5194/os-10-69-2014, 2014. Bach, L. T., Mackinde , L. C., Schulz, K. G., Wheele , G., Sch oede , D. C., B ownlee, C. and Riebesell, U.: Dissec ing he impac o CO2and pH on he mechanisms o pho osyn hesis and calci ica ion in he coccoli hopho e Emiliania huxleyi, New Phy- ol., 199, 121–134, doi:10.1111/nph.12225, 2013. Balch, W. M.: Re-e alua ion o he physiological ecology o coc- coli hopho es, in: Coccoli hopho es – F om Molecula P ocesses o Global Impac , edi ed by: Thie s ein, H. and Young, J. R., Sp inge , Be lin, 165–190, 2004. Bakun, A. and Agos ini, V. N.: Seasonal pa e ns o wind-induced up- welling/downwelling in he Medi e anean Sea, Sciencia Ma- ina, 65, 243–257, 2001. Ba cena, M. A., Flo es, J. A., Sie o, F. J., Pé ez-Folgado, M., Fab- es, J., Cala a , A., and Canals, M.: Plank onic esponse o main oceanog aphic changes in he Albo an Sea (Wes e n Medi e - anean) as documen ed in sedimen aps and su ace sedimen s, Ma . Mic opaleon ol., 53, 423–445, 2004. Ba low, R. G., Man ou a, R. F. C., Cummings, D. G., and Fileman, T. W.: Pigmen chemo axonomic dis ibu ions o phy oplank on du ing summe in he wes e n Medi e anean, Deep-Sea Res. P . II, 44, 833–850, doi:10.1016/S0967-0645(96)00089-6, 1997. Beau o , L., Couapel, M., Buche , N., Claus e, H., and Goye , C.: Calci e p oduc ion by coccoli hopho es in he sou h eas Paci ic Ocean, Biogeosciences, 5, 1101–1117, doi:10.5194/bg-5-1101- 2008, 2008. Beau o , L., P obe , I., de Ga idel-Tho on, T., Bendi , E. M., Ruiz- Pino, D., Me zl, N., Goye , C., Buche , N., Coupel, P., G elaud, M., Ros , B., Rickaby, R. E. M., and de Va gas, C.: Sensi i i y o coccoli hopho es o ca bona e chemis y and ocean acidi ica ion, Na u e, 476, 80–83, 2011. Be land, B. R., Benzhi ski, A. G., Bu lako a, Z. P., Geo gie a, L. V., Izmes ie a, M. A., Kholodo , V. I., and Maes ini, S. Y.: Con- di ions hyd ologiques es i ales en Medi e anee, epa i ion du phy oplank on e de la ma ie e o ganique, Oceanol. Ac a, 9, 163– 177, 1988. Be man, T., Townsend, D. W., El Sayed, S. Z., T ees, G. C., and Azo , Y.: Op ical anspa ency, chlo ophyll and p ima y p o- duc i i y in he Eas e n Medi e anean nea he Is aeli Coas , Oceanol. Ac a, 7, 367–372, 1984. Boeckel, B., Baumann, K. H., Hen ich, R., and Kinkel, H.: Coc- coli h dis ibu ion pa e ns in Sou h A lan ic and Sou he n Ocean su ace sedimen s in ela ion o en i onmen al g adien s, Deep- Sea Res. P . I, 53, 1073–1099, doi:10.1016/j.ds .2005.11.006, 2006. B and, L. E.: The salini y ole ance o o y-six ma ine phy oplank- on isola es, Es ua . Coas . Shel S., 18, 543–556, 1984. B and, L. E.: Physiological ecology o ma ine coccoli hopho es, in: Coccoli hopho es, edi ed by: Win e , A. and Siesse , W. G., Cam- b idge Uni e si y P ess, UK, 39–49, 1994. B oe se, A. T. C., B umme , G.-J. A., and Van Hin e, J. E.: Coccol- i hopho e expo p oduc ion in esponse o monsoonal upwelling o Somalia (no hwes e n Indian Ocean), Deep-Sea Res. P . II, 47, 2179–2205, 2000. Cha alampopoulou, A., Poul on, A. J., Ty ell, T., and Lucas, M. I.: I adiance and pH a ec coccoli hopho e communi y composi- ion on a ansec be ween he No h Sea and he A c ic Ocean, Ma . Ecol.-P og. Se ., 431, 25–43, 2011. Co ès, M. Y., Bollman, J., and Thie s ein, H. R.: Coccoli hopho e ecology a he HOT s a ion Aloha, Hawaii, Deep-Sea Res. P . II, 48, 1957–1981, 2001. C os, L.: Plank ic coccoli hopho es o he NW Medi e anean. PhD Thesis, Uni e si a de Ba celona, Ba celona, 181 pp., 2002. C os, L. and Es ada, M.: Holo-he e ococcoli hopho e li e cycles: ecological implica ions, Ma . Ecol.-P og. Se ., 492, 57–68, 2013. C os, L. and Fo uno, J.-M.: A las o no hwes e n Medi e anean coccoli hopho es, Sci. Ma ., 66, 7–182, 2002. Decemb ini, F., Ca oppo, C., and Azza o, M.: Size s uc u e and p oduc ion o phy oplank- on communi y and ca bon pa hways channelling in he Sou he n Ty henian Sea (Wes e n Medi e - anean), Deep-Sea Res. P . II, 56, 687–699, 2009. Dickson, A. G. and Mille o, F. J.: A compa ison o he equilib ium cons an s o he dissocia ion o ca bonic acid in seawa e media, Deep-Sea Res., 34, 1733–1743, 1987. Dimiza, M. D., T ian aphyllou, M. V., and De mi zakis, M. D.: Ve - ical dis ibu ion and ecology o li ing coccoli hopho es in he ma ine ecosys ems o And os Island (Middle Aegean Sea) du - ing la e summe 2001, Hellenic J. Geosci., 43, 7–20, 2008. Dugdale, R. C. and Wilke son, F. P.: Nu ien sou ces and p ima y p oduc ion in he eas e n Medi e anean, Oceanol. Ac a, Special Issue, 9, 179–184, 1988. Egge, J. K. and Aksnes, D. L.: Silica e as egula ing nu ien in phy oplank on compe i ion, Ma . Ecol.-P og. Se ., 83, 281–289, 1992. Falkowski, P. G.: The ole o phy oplank on pho osyn hesis in global biogeochemical cycles, Pho osyn h. Res., 39, 235–258, 1994. Field, C. B., Beh en eld, M. J., Rande son, J. T., and Falkowski, P. G.: P ima y p oduc ion o he biosphe e: in eg a ing e es ial and oceanic componen s, Science, 281, 237–240, 1998. F ada, M., P obe , I., Allen, M. J., Wilson, W. H., and De Va gas, C.: The “Cheshi e Ca ” escape s a egy o he coccoli hopho e Emiliania huxleyi in esponse o i al in ec ion, P. Na l. Acad. Sci. USA, 105, 15944–15949, 2008. F ada, M. J., Bidle, K. D., P obe , I., and de Va gas, C.: In si u su - ey o li e cycle phases o he coccoli hopho e Emiliania huxleyi (Hap ophy a), En i on. Mic obiol., 14, 1558–1569, 2012. Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/ A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s 31 Go sis-Sk e as, O., Pagou, K., Mo ai ou-Apos olopoulou, M., and Igna iades, L.: Seasonal ho izon al and e ical a iabili y in p ima y p oduc ion and s anding s ocks o phy oplank on and zooplank on in he C e an Sea and he S ai s o he C e an A c (Ma ch 1994–Janua y 1995), P og. Oceanog ., 44, 625–649, 1999. Hagino, K., Okada, H., and Ma suoka, H: Spa ial dynamics o coc- coli hopho e assemblage in he Eua o ial Wes e n-Cen al Paci ic Ocean, Ma . Mic opaleon ol., 39, 53–72, 2000. Hagino, K., Okada, H., and Ma suoka, H.: Coccoli hopho e assem- blages and mo pho ypes o Emiliania huxleyi in he bounda y zone be ween he cold Oyashio and wa m Ku oshio cu en s o he coas o Japan, Ma . Mic opaleon ol., 55, 19–47, 2005. Haida , A. T. and Thie s ein, H. R.: Coccoli hopho e dynamics o Be muda (N. A lan ic), Deep- Sea Res. P . II, 40, 1925–1956, 2001. Hainbuche , D., Rubino, A., Ca din, V., Tanhua, T., Sch oede , K., and Bensi, M.: Hyd og aphic si ua ion du ing c uise M84/3 and P414 (sp ing 2011) in he Medi e anean Sea, Ocean Sci., 10, 669–682, doi:10.5194/os-10-669-2014, 2014. Igna iades, L., Geo gopoulos, D., and Ka ydis, M.: Desc ip ion o he phy oplank onic communi y o he oligo ophic wa e s o he SE Aegean Sea (Medi e anean), Ma . Ecol.-P. S. Z. N. I., 16, 13–26, 1995. Igna iades, L., Go sis-Sk e as, O., Pagou, K., and K asakopoulou, E.: Di e si ica ion o phy o- plank on communi y s uc u e and ela ed pa ame e s along a la ge-scale longi udinal eas -wes ansec o he Medi e anean Sea, J. Plank on Res., 31, 411–428, 2009. Kleijne, A.: Holococcoli hopho ids om he Indian Ocean, Red Sea, Medi e anean Sea and No h A lan ic Ocean, Ma . Mi- c opaleon ol., 17, 1–76, 1991. Kleijne, A.: Ex an Rhabdosphae aceae (coccoli hopho ids, class P ymnesiophyceae) om he Indian Ocean, Red Sea, Medi e - anean Sea and No h A lan ic Ocean, Sc ip aGeol., 100, 1–63, 1992. Kleijne, A., K oon, D., and Ze enboom, W.: Phy oplank on and o amini e al equencies in no he n Indian Ocean and Red Sea su ace wa e s, Ne h. J. Sea Res., 24, 531–539, 1989. Knappe sbusch, M.: Geog aphic dis ibu ion o li ing and Holocene coccoli hopho es in he Medi e anean Sea, Ma . Mi- c opaleon ol., 21, 219–247, 1993. K oeke , K. J., Ko das, R. L., C im, R., Hend iks, I. E., Ramajo, L., Singh, G. S., Dua e, C. M., and Ga uso, J. P.: Impac s o ocean acidi ica ion on ma ine o ganisms: quan i ying sensi i i ies and in e ac ion wi h wa ming, Glob. Change Biol., 19, 1884–1896, doi:10.1111/gcb.12179, 2013. K om, M. D., K ess, N., B enne , S., and Go don, L.: Phospho ous limi a ion o p ima y p oduc i i y in he Eas e n Medi e anean Sea, Limnol. Oceanog ., 36, 424–432, 1991. Lange , G. and Bode, M.: CO2media ion o ad e se e ec s o sea- wa e acidi ica ion in Calcidiscus lep opo us, Geochem. Geophy. Geosy., 12, Q05001, doi:10.1029/2010GC003393, 2011. Lange , G., Geisen, M., Baumann, K.-H., Kläs, J., Riebesell, U., Thoms, S., and Young, J. R.: Species-speci ic esponses o calci- ying algae o changing seawa e ca bona e chemis y, Geochem. Geophy. Geosy., 7, Q09006, doi:10.1029/2005GC00122, 2006. Lange , G., Neh ke, G., P obe , I., Ly, J., and Zi e i, P.: S ain-speci ic esponses o Emiliania huxleyi o changing seawa e ca bona e chemis y, Biogeosciences, 6, 2637–2646, doi:10.5194/bg-6-2637-2009, 2009. Las e nas, S., Tunin-Ley, A., Ibañez, F., Ande sen, V., Pizay, M.- D., and Lemée, R.: Sho - e m dynamics o mic oplank on abun- dance and di e si y in NW Medi e anean Sea du ing la e sum- me condi ions (DYNAPROC 2 c uise; 2004), Biogeosciences, 8, 743–761, doi:10.5194/bg-8-743-2011, 2011. Leblanc, K., Quéguine , B., Ga cia, N., Rimmelin, P. and Raim- baul , P.: Silicon cycle in he NW Medi e anean Sea: seasonal s udy o a coas al oligo ophic si e, Oceanol. Ac a, 26, 339–355, 2003. Lewis, E. and Wallace, D. W. R: P og am de eloped o CO2sys- em calcula ions ORNL/CDIAC-105. Ca bon Dioxide In o ma- ion Analysis Cen eOak Ridge Na ional Labo a o y, U.S. De- pa men o Ene gy, Oak Ridge, Tennessee, 1998. Mackinde , L., Wheele , G., Sch oede , D., Riebesell, U., and B ownlee, C.: Molecula mechanisms unde lying calci ica- ion in coccoli hopho es, Geomic obiol. J., 27, 585–595, doi:10.1080/01490451003703014, 2010. Malin e no, E., Zi e i, P., and Co selli, C.: Coccoli hopho id dis i- bu ion in he Ionian Sea and i s ela ionship o eas e n Medi e - anean ci cula ion du ing la e all o ea ly win e 1997, J. Geo- phys. Res., 108, 8115, doi:10.1029/2002JC001346, 2003. Ma gale , R.: Li e- o ms o phy oplank on as su i al al e na i es in an uns able en i onmen , Oceanol. Ac a, 1, 493–509, 1978. Ma inez, J. M., Sch oede , D. C., La sen, A., B a bak, G., and Wil- son, W. H.: Molecula dynamics o Emiliania huxleyi and cooc- cu ing i uses du ing wo sepa a e mesocosm s udies, Appl. En- i on. Mic ob., 73, 554–562, 2007. McIn y e, A. and Bé, A. W. H.: Mode n coccoli hopho idae o he A lan ic Ocean. 1. Placoli hs and cy oli hs, Deep-Sea Res., 14, 561–597, 1967. Meh bach, C., Culbe son, C. H., Hawley, J. E., and Py kowi cz, R. M.: Measu emen s o he appa en dissocia ion cons an s o ca bonic acid in seawa e a a mosphe ic p essu e, Limnol. Oceanog ., 18, 897–907, 1973. Me ico, A., Ty ell, T., and Cokaca , T.: Is he e any ela ionship be ween he phy oplank on seasonal dynamics and he ca bona e sys em?, J. Ma ine Sys ., 59, 120–142, 2006. Mohan, R., Me gulhao, L. P., Gup ha, M. V. S. Rajakuma , A., Thamban, M., Anilkuma , N., Sudhaka , M., and Ra ind a, R.: Ecology o coccoli hopho es in he Indian sec o o he Sou he n Ocean, Ma . Mic opaleon ol., 67, 30–45, 2008. Mol ino, B. and McIn y e, A.: P ecession o cing o nu icline dy- namics in he equa o ial A lan ic, Science, 249, 766–769, 1990. Nejs gaa d, J. C., Gisme ik, I., and Solbe g, P. T.: Feeding and ep oduc ion by Calanus in- ma chicus, and mic ozooplank on g azing du ing mesocosm blooms o dia oms and he coccol- i hopho e Emiliania huxleyi, Ma . Ecol.-P og. Se ., 147, 197– 217, 1997. Nelson, D. M., Goe ing, J. J., Kilham, S. S., and Guilla d, R. R. L.: Kine ics o silicic acid up ake and a es o silica dissolu ion in he ma ine dia om Thalassiosi a pseudonana, J. Phycol., 12, 246–252, 1976. Nishida, S.: A las o Paci ic nannoplank ons, News Osaka, Mi- c opaleon y, 3, 1–31, 1979. Okada, H. and McIn y e, A.: Mode n coccoli hopho es o he Pa- ci ic and No h A lan ic Oceans, Mic opaleon ology, 23, 1–55, 1977. www.ocean-sci.ne /11/13/2015/ Ocean Sci., 11, 13–32, 2015 32 A. O iedo e al.: Coccoli hopho e dis ibu ion along ca bona e chemis y g adien s Okada, H. and McIn y e, A.: Seasonal dis ibu ion o mode n coc- coli hopho es in he wes e n No h A lan ic Ocean, Ma . Biol., 54, 319–328, 1979. Okada, H. and Honjo, S.: The dis ibu ion o ocean coccol- i hopho ids in he Paci ic, Deep-Sea Res., 20, 335–374, 1973. O iedo, A. M., Lange , G., and Zi e i, P.: E ec o phospho us lim- i a ion on coccoli h mo phology and elemen a ios in Medi e - anean s ains o he coccoli hopho e Emiliania huxleyi, J. Exp. Ma . Biol. Ecol., 459, 105–113, 2014. Pina di, N. and Mase i, E.: Va iabili y o he la ge scale gene al ci cula ion o he Medi e anean Sea om obse a ions and mod- elling: a e iew, Palaeogeog . Palaeoecl., 158, 153–173, 2000. P obe , I. and Houdan, A.: The Labo a o y cul u e o coccol- i hopho es, in: Coccoli hopho es: F om Molecula P ocesses o Global Impac , edi ed by: Thie s ein, H. R. and Young, J. R., Sp inge , Ge many, 217–250, 2004. Psa a, S., Tselepides, A., and Igna iades L.: P ima y p oduc i i y in he oligo ophic C e an Sea (NE Medi e anean): seasonal and in e annual a iabili y, P og. Oceanog ., 46, 187–204, 2000. Rabi i, S., Bianchi, F., Bold in, A., Da Ros, L., Socal, G., and To i, C.: Pa icula e ma e and phy oplank on in he Ionian Sea, Oceanol. Ac a, 17, 297–307, 1994. Raha , E., He u , B., Le i, A., Mulholland, M. R., and Be man- F ank, I.: Sp ing ime con ibu ion o dini ogen ixa ion o p i- ma y p oduc ion ac oss he Medi e anean Sea, Ocean Sci., 9, 489–498, doi:10.5194/os-9-489-2013, 2013. Ribe a d’Alcalà, M., Ci i a ese, G., Con e sano, F., and La ezza, R.: Nu ien a ios and luxes hin an o e looked p ocesses in he Medi e anean sea, J. Geophys. Res., 108, 8106, doi:10.1029/2002JC001650, 2003. Robinson, A. R., Leslie, W. G., Theocha is, A., and Lasca a os, A. : Medi e anean Sea ci cula ion, in: Encyclopedia o ocean sci- ences, Academic P ess, London, doi:10.1006/ wos.2001.0376, 2001. Rohling, E. J., Abu-Zied, R., Cas o d, J. S. L., Hayes, A., and Hoogakke , B. A. A.: The ma ine en i onmen : p esen and pas , in: The Physical Geog aphy o he Medi e anean, edi ed by: Woodwa d, J. C., Ox o d Uni e si y P ess, Ox o d, ISBN: 978- 0-19-926803-0, 33–67, 2009. Sa hou, G., Timme mans, K. R., Blain, S., and T egue , P.: G ow h physiology and a e o dia oms in he ocean: a e iew, J. Sea Res., 53, 25–42, 2005. S e d up, H. U.: On condi ions o he e nal blooming o phy o- plank on, ICES J. Ma . Sci., 18, 287–295, 1953. Takahashi, K. and Okada, H.: En i onmen al con ol on he bio- geog aphy o mode n coccoli hopho es in he sou heas e n In- dian Ocean o sho e o Wes e n Aus alia, Ma . Mic opaleon ol., 39, 73–86, 2000. Tanhua, T., Hainbuche , D., Ca din, V., Ál a ez, M., Ci i a ese, G., McNichol, A. P., and Key, R. M.: Repea hyd og aphy in he Medi e anean Sea, da a om he Me eo c uise 84/3 in 2011, Ea h Sys . Sci. Da a, 5, 289–294, doi:10.5194/essd-5-289-2013, 2013a. Tanhua, T., Hainbuche , D., Sch oede , K., Ca din, V., Ál a ez, M., and Ci i a ese, G.: The Medi e anean Sea sys em: a e iew and an in oduc ion o he special issue, Ocean Sci., 9, 789–803, doi:10.5194/os-9-789-2013, 2013b. T ian aphyllou, M., Dimiza, M., K asakopoulou, E., Malin e no, E., Lianou, V., and Sou e mezoglou, E.: Seasonal a ia ion in Emiliania huxleyi coccoli h mo phology and calci ica ion in he Aegean Sea (Eas e n Medi e anean), Geobios., 12 h Mee ing o he In e na ional Nannoplank on Associa ion, Lyon 2008, 43, 99–110, 2010. Ty ell, T. and Young, J. R.: Coccoli hopho es, in: Encyclopedia o Ocean Sciences, 2nd Edn., edi ed by: S eele, J. H., Tu ekian, K. K., and Tho pe, S. A., Academic P ess, San Diego, US, 3568– 3576, 2009. Ty ell, T., Schneide , B., Cha alampopoulou, A., and Riebesell, U.: Coccoli hopho es and calci e sa u a ion s a e in he Bal ic and Black Seas, Biogeosciences, 5, 485–494, doi:10.5194/bg-5-485- 2008, 2008. Va di, A., Ha ama y, L., Van Mooy, B. A., F ed icks, H. F., Kim- mance, S. A., La sen, A., and Bidle, K. D.: Hos - i us dynamics and subcellula con ols o cell a e in a na u al coccoli hopho e popula ion, P. Na l. Acad. Sci. USA, 109, 19327–19332, 2012. Volpe, G., Colella, S., Fo ne is, V., T onconi, C., and San ole i, R.: The Medi e anean Ocean Colou Obse ing Sys em – sys em de elopmen and p oduc alida ion, Ocean Sci., 8, 869–883, 2012 Yacobi, Y., Zoha y, T., K ess, N., Hech , A., Roba s, R., Waise , M., Wood, A., and Li, W.: Chlo ophyll dis ibu ion h oughou he sou he n Medi e anean in ela ion o he physical s uc u e o he wa e mass, J. Ma . Sys ., 6, 179–190, 1995. Young, J. R.: Func ions o coccoli hs, in: Coccoli hopho es, edi ed by: Win e , A. and Siesse , W. G., Camb idge Uni e si y P ess, Camb idge, UK, 63–82, 1994. Young, J., Geisen, M., C os, L., Kleijne, A., Sp engel, C., P obe , I., and Øs e gaa d, J.: A guide o ex an coccoli hopho e axon- omy, J. Nannoplank on Res., 1, 1–125, 2003. Zi e i, P., Thunell, R. C., and Rio, D.: Seasonal changes in coc- coli hopho e densi ies in he Sou he n Cali o nia Bigh du ing 1991–1992, Deep-Sea Res. P . I, 42, 1881–1903, 1995. Zi e i, P., Ru en, A., de Lange, G., Thomson, J., and Co selli, C.: P esen -day coccoli h luxes eco ded in cen al eas e n Medi e anean sedimen aps and su ace sedimen s, Palaeo- geog . Palaeocl., 158, 175–195, 2000. Zi e i, P., Baumann, K. H., Böckel, B., Bollman, J., and Young, J. R.: P esen day coccoli hopho e biogeog aphy in he A - lan ic Ocean, in: Coccoli hopho es: F om Molecula P ocesses o Global Impac , edi ed by: Thie s ein, H. R. and Young, J. R., Sp inge Ve lag, Be lin Heidelbe g, 2004. Zi e i, P., Thoms, S., P obe , I., Geisen, M., and Lange , G.: A uni e sal ca bona e ion e ec on s able oxygen iso ope a ios in unicellula plank onic calci ying o ganisms, Biogeosciences, 9, 1025–1032, doi:10.5194/bg-9-1025-2012, 2012. Ocean Sci., 11, 13–32, 2015 www.ocean-sci.ne /11/13/2015/