Response of phytoplankton to enhanced atmospheric and riverine nutrient inputs in a coastal upwelling embayment

1 Response o phy oplank on o enhanced a mosphe ic and i e ine nu ien inpu s 1 in a coas al upwelling embaymen 2 Teixei a, I.G.1,3*; A bones, B.1; F oján, M.1; Nie o-Cid, M.1; Ál a ez-Salgado, X.A.1; 3 Cas o, C.G.1; Fe nández, E.2; Sob ino, C.2; Tei a, E.2; Figuei as, F.G.1 4 1 CSIC, Ins i u o de In es igacións Ma iñas. Edua do Cabello 6, 36208 Vigo. Spain. 5 2 Uni e sidade de Vigo, Campus Lagoas-Ma cosende, 36310 Vigo, Spain. 6 3 CESAM – Cen e o En i onmen al and Ma ine S udies & Depa men o Biology, 7 Uni e si y o A ei o, Campus de San iago, 3810‑193 A ei o, 8 Po ugal. 9 *Co esponding au ho : [email p o ec ed] 10 KEYWORDS: phy oplank on, nu ien s, a mosphe ic deposi ion, i e ine inpu s, 11 upwelling sys em, NW Spain, Ria de Vigo 12 13 ABSTRACT 14 O e he pas decades, as a consequence o human ac i i y, he e was an inc ease in 15 nu ien inpu s o he ocean and hey a e expec ed o enhance e en mo e in he u u e. 16 Coas al a eas, accoun ing o a signi ican p opo ion o ma ine p ima y p oduc i i y, 17 a e he mos ulne able zones o an h opogenic impac s. The esponse o phy oplank on 18 communi ies o an inc ease in o ganic and ino ganic nu ien s le els om na u al 19 alloch honous sou ces was assessed in mic ocosm expe imen s conduc ed in a coas al 20 sys em a ec ed by in e mi en upwelling e en s (Ría de Vigo, NW Ibe ia). Th ee 21 nu ien addi ion expe imen s we e pe o med in sp ing, summe and au umn, when 22 2 su ace wa e was supplemen ed wi h 5 and 10% o a mosphe ic and i e ine ma e . 23 Pico-, nano- and mic ophy oplank on abundances, chlo ophyll a concen a ion (Chl a) 24 and p ima y p oduc ion a es (PP) we e measu ed and compa ed wi h hose in he 25 con ol seawa e sample (wi hou addi ions) a e 48 h o incuba ion. Simul aneous 26 expe imen s wi h con olled addi ions o ino ganic and o ganic nu ien s we e also 27 pe o med in o de o desc ibe he limi ing nu ien o phy oplank on g ow h a each 28 expe imen . The composi ion o he ma e inpu s and he s uc u e o he phy oplank on 29 communi ies de e mined he ype o esponse obse ed. Phy oplank on esponses a ied 30 among seasons, being posi i ely co ela ed wi h dissol ed ino ganic ni ogen (DIN) 31 concen a ions. As expec ed, he phy oplank on esponses o ex e nal nu ien inpu s 32 we e s onge unde low nu ien le els (summe ) han when phy oplank on was al eady 33 g owing in nu ien eple e condi ions (sp ing). Null and nega i e esponses o he 34 na u al inpu s we e obse ed in au umn, which sugges s ha he oceanic phy oplank on 35 ad ec ed o his coas al sys em du ing downwelling e en s could be occasionally 36 inhibi ed by hese nu ien inpu s. In a u u e global change scena io, cha ac e ized by 37 enhanced nu ien inpu s om i e ine and a mosphe ic o igin, he esponse o 38 phy oplank on communi ies will s ongly depend on he concen a ion and chemical 39 composi ion o hese inpu s and on he s uc u e o phy oplank on communi ies able o 40 espond o hem. 41 42 1. In oduc ion 43 Phy oplank on o ganisms a e he main p ima y p oduce s in he sea being 44 esponsible o abou 50% o he wo ld p ima y p oduc ion (Field e al. 1998). Coas al 45 a eas, despi e ep esen ing only 7% o ocean’s su ace, accoun o 15-20% o global 46 3 ma ine p ima y p oduc i i y (Wollas 1993, 1998; La uelle e al. 2009). Enhanced 47 nu ien luxes om con inen al uno , he a mosphe e and he adjacen open ocean, 48 and he e icien nu ien ecycling based on a closed coupling be ween pelagic 49 p oduc ion and ben hic egene a ion a e he easons behind he high p oduc i i y o he 50 coas al zone (Walsh 1991; Wollas 1998). All hese biogeochemical p ocesses a e 51 specially in ensi ied in coas al upwelling egions because o he enhanced en y o 52 nu ien s om he adjacen ocean in esponse o in ense and pe sis en equa o wa d 53 winds (Walsh 1991; Wollas 1998). 54 Due o i s close p oximi y o human popula ions, coas al a eas a e also he mos 55 ulne able zones o an h opogenic o cing (Jickells 1998). Ino ganic and o ganic 56 nu ien s and pollu an s esul ing om human ac i i ies each coas al a eas h ough 57 a mosphe ic deposi ion, con inen al uno and g oundwa e e luen s (Jickells 1998; 58 Doney 2010; S a ham 2012). O e he pas decades, he e was an inc ease in hese 59 nu ien inpu s o he ocean and hey a e expec ed o expand e en mo e in he u u e 60 (Ande son e al. 2002; Galloway and Cowling 2002; Galloway e al. 2004; Duce e al. 61 2008). Al e a ions in he magni ude and composi ion o ma e inpu s in coas al a eas 62 may induce signi ican changes in phy oplank on communi ies inhabi ing hose sys ems 63 (Pae l 1997; Peie ls and Pae l 1997; Sei zinge and Sande s 1999; Spa ha is e al. 2007). 64 Ex eme cases o nu ien en ichmen may in ol e he de elopmen o massi e 65 phy oplank on blooms, including ha m ul algal species (Pae l 1997; Ande son e al. 66 2002). 67 The Ría de Vigo is a p oduc i e embaymen in he coas al upwelling sys em o 68 he No hwes Ibe ian Peninsula (F aga 1981). Seasonal pa e ns in phy oplank on 69 composi ion and p oduc i i y in his sys em a e closely ela ed o he hyd og aphic 70 a iabili y (Figuei as and Ríos 1993; Figuei as e al. 2002). Upwelling e en s, occu ing 71 4 usually be ween Ma ch and Oc obe , in oduce nu ien ich subsu ace wa e s in o he 72 pho ic laye and a e ollowed by high concen a ions o phy oplank on cells and 73 chlo ophyll. On he con a y, he win e pe iod is a o able o downwelling and i is 74 cha ac e ized by lowe biomass accoun ed o small-size phy oplank on cells (Figuei as 75 e al. 2002). Ri e discha ge and a mosphe ic deposi ion may also cons i u e signi ican 76 sou ces o nu ien inpu s o his coas al embaymen , al hough o a lesse ex en as 77 compa ed wi h upwelling episodes (Gago e al. 2005; Rod íguez and Macías 2006; 78 Alonso-Pé ez and Cas o 2014; Fe nández e al. in 2016). The high p oduc i i y 79 cha ac e is ic o his sys em sus ains an in ensi e mussel p oduc ion and signi ican 80 ca ches o se e al ish and shell ish species, which a e ele an o he economy o he 81 egion (Figuei as e al. 2002; F oján e al. 2014). Thus, unde s anding he esponse o 82 he o ganisms in he basis o he ood web o an e en ual inc ease in nu ien s inpu s in 83 he Ría de Vigo may be de e minan o an icipa e i s consequences o he whole 84 ecosys em. 85 The ou comes o an inc ease in nu ien s on coas al mic obial plank on in his 86 sys em was es ed in p e ious expe imen al s udies, which ocused on he e ec o 87 con olled ino ganic and o ganic nu ien addi ions (Ma ínez-Ga cía e al. 2010; Tei a 88 el al. 2011) and na u al addi ions o ainwa e (Tei a e al. 2013; Ma ínez-Ga cía e al. 89 2015). In he p esen s udy, we aimed a examining he esponse o phy oplank on 90 communi ies in he Ría de Vigo o ealis ic inc easing amoun s o dissol ed ma e 91 inpu s om bo h a mosphe ic and i e ine o igin. To calcula e he cu en a e age 92 i e ine and a mosphe ic inpu s o he Ría de Vigo we conside ed he a e age i e low 93 o he Ri e Oi abén-Ve dugo, 17 m3 s–1 (Gago e al. 2005) and he a e age 94 p ecipi a ion o he Ría de Vigo, 7.7 mm d–1. Conside ing he su ace a ea o he ía is 95 (174 km2), he mean su ace mixing laye (2 m) and he a e age lushing ime o his 96 5 laye (5 days), he su ace mixing laye o he ía mus con ain abou 2% o i e and 97 2% o ainwa e . The e o e, he addi ions o 5% and 10% es ed he e would se e o 98 ob ain he esponse o he Ría de Vigo o u u e global change scena ios in which 99 human ac i i ies inc ease he quan i y wi hou al e ing he quali y o i e ine and 100 a mosphe ic inpu s. Fu he mo e, concomi an expe imen s wi h con olled addi ions o 101 ino ganic and o ganic nu ien s we e also pe o med in o de o desc ibe he limi ing 102 nu ien o phy oplank on g ow h a each season. 103 104 2. Ma e ials and Me hods 105 2.1 Sampling 106 Seawa e o he expe imen s was sampled om he middle o he Ría de Vigo 107 (42º14.09’ N, 8º47.18’ W) du ing he p oduc i e season in sp ing (May) and summe 108 (July), and a e he ansi ion o he unp oduc i e pe iod in au umn (Oc obe ) 2013 109 (Fig. 1). Tempe a u e, salini y and in si u luo escence down o 25 m dep h we e 110 ob ained wi h a SBE 9/11 CTD p obe and a Sea ech luo ome e a ached o a ose e 111 sample . Sub-su ace seawa e (3-4 m) was collec ed in 12 L acid-clean Niskin bo les 112 and il e ed h ough a 200 µm po e size mesh o emo e la ge zooplank on in o a la ge 113 acid-clean ca boy o ans e i o he coas al s a ion (Fig. 1) o p epa e he addi ion 114 expe imen s. 115 116 2.2 Ekman anspo and uno 117 The Ekman anspo (Qx; m2 s−1) pe pendicula o he coas , a p oxy o he 118 occu ence and in ensi y o coas al upwelling, was calcula ed om wind di ec ion and 119 eloci y, eco ded a Cabo Sillei o buoy (Fig. 1) acco ding o Bakun (1973): 120 6 Ría de Mu os Ría de A ousa Ría de Pon e ed a A Cape Finis e e Ría de Vigo -10.0 -9.5 -9.0 -8.5 -8.0 Longi ude (E) 42.0 42.5 43.0 43.5 La i ude (N) Sillei o Buoy -8.9 -8.8 -8.7 Longi ude (E) 42.1 42.2 42.3 42.3 42.4 La i ude (N) 75m 20m 40m B Ría de Vigo + IIM - CSIC X Ci y Hall Coas al s a ion 𝑄𝑥= 𝜌𝑎∗𝐶∗ [𝑉]∗𝑉 𝑦 𝑓∗𝜌𝑤 (1) 121 whe e ρa is he ai densi y (1.22 kg m−3), C is an empi ical d ag coe icien (1.3 × 10−3, 122 dimensionless), [V] is he wind speed (m s−1) wi h componen Vy, ƒ is he Co iolis 123 pa ame e a his la i ude (9.95 × 10−5 s−1), and ρw is he densi y o seawa e (~1025 kg 124 m−3). The sign o Qx was changed o associa e posi i e alues wi h o sho e anspo 125 (upwelling) o su ace wa e s. 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 Fig. 1: (A) Map o he wes Galician coas wi h he loca ion o he Sillei o buoy (◆) and (B) 144 map o he Ría de Vigo wi h he loca ion o he sampling s a ion (●), he i e Oi abén-Ve dugo (▲) and 145 he me eo ological s a ions in he IIM (+) and in he Ci y Hall (X). The coas al s a ion whe e addi ion 146 expe imen s ook place is indica ed by an a ow. 147 148 7 Runo was calcula ed using da a om Ri e Oi abén-Ve dugo (Fig. 1), acco ding o he 149 me hod desc ibed in O e o e al. (2010). 150 151 2.3 P epa a ion o na u al wa e s concen a es o he addi ion expe imen s 152 Ri e and ainwa e samples, and <10 µm a mosphe ic pa icles we e collec ed 153 and p ocessed o ob ain concen a es o he i e ine and a mosphe ic ma e inpu s o he 154 Ría de Vigo. The objec i e was o educe he olume o he o iginal wa e samples 10-155 old main aining hei chemical composi ion, i.e. wi hou addi ion o loss o any 156 componen p esen in he na u al samples. In his way, i was possible o es he e ec 157 o an inc ease in na u al ma e using small wa e olumes and wi hou signi ican 158 changes in salini y. 159 The Ri e Oi abén-Ve dugo (Fig. 1) was sampled a week be o e he addi ion 160 expe imen s in Ap il, July and Oc obe 2013. The wa e samples we e collec ed 161 ups eam o he eshwa e -seawa e in e ace, o gua an ee ha he chemical 162 composi ion o he i e samples ep esen ed ha o he wa e ha mixes wi h he 163 seawa e o he Ría de Vigo. Fi e li e s o each sample we e g a i y il e ed h ough a 164 p e-washed (wi h 10 L o ul apu e wa e ) dual-s age (0.8 and 0.2 µm) il e ca idge 165 (Pall-Ac opak supo Memb ane). The il a e was hen concen a ed 10- old using 166 o a o y e apo a ion wi h a Buchi R215 e apo a o unde mild condi ions (ba h 167 empe a u e: 25 ºC, acuum: 13 mba , condense : ace one/CO2) o a oid b eakage o 168 any o ganic compound p esen in he o iginal wa e samples. Analysis o he 169 concen a ion o ino ganic (ammonium, ni i e, ni a e and phospha e) and o ganic 170 (dissol ed o ganic ca bon and ni ogen) subs a es con i med ha he samples we e 171 concen a ed quan i a i ely bu main aining hei o iginal composi ion. 172 8 A MTX ainwa e sample (model FAS005AB) and a high olume PM10 MCV 173 PM1025 sample (model CAV-A/MS) ins alled a he oo o he Ins i u o de 174 In es igaciones Ma inas (IIM-CSIC) (Fig. 1) collec ed samples o we and d y 175 deposi ion o he Ría de Vigo. The MTX sample was equipped wi h a humidi y senso 176 opening he sys em only when i was aining and allowing sampling jus he we 177 ac ion o he a mosphe ic deposi ion. Rainwa e was collec ed om ou weeks o one 178 week be o e he addi ion expe imen s. Samples we e aken daily and ozen 179 immedia ely a e collec ion. A week be o e he expe imen , he daily samples we e 180 hawed a ambien empe a u e, mixed in one olume (6 L), and quan i a i ely 181 concen a ed ollowing he same p ocedu e as o he i e ine samples. Rainwa e was 182 collec ed only o he expe imen s in sp ing (May) and au umn (Oc obe ) 2013 because 183 we deposi ion was e y sca ce he weeks be o e he summe expe imen in July (36 184 mm accumula ed om 11 June o 10 July; me eo ological s a ion o he Vigo ci y hall – 185 Fig. 1). The high olume sample was used o collec a mosphe ic pa icles (1−10 µm) 186 on p ecombus ed (450 ºC, 4 h) 140 mm GF/F il e s (48 h sampling). The pa icles we e 187 collec ed he week be o e each o he h ee addi ion expe imen s, ope a ing du ing 48 h 188 a a a e o 30 m3 h–1. Then, he wa e -soluble ac ion (WSF) o one eigh h o he il e 189 was ex ac ed in 400 mL o he co esponding ainwa e concen a e by mechanical 190 s i ing du ing 40 min. In he summe expe imen , he WSF was ex ac ed in milli-Q 191 wa e . These p opo ions (1/8 o he il e in 400 mL o wa e ) we e decided o ob ain 192 10- old he expec ed concen a ions based on p e ious in o ma ion abou he 193 composi ion o we and d y deposi ion o he Ría de Vigo (Tei a e al. 2013; Ma ínez-194 Ga cía e al. 2015). Final mixed ex ac s we e il e ed h ough p ecombus ed (450 ºC, 4 195 h) 47 mm diame e Wha man GF/F il e s in an acid-cleaned glass il a ion sys em, 196 unde low N2 low p essu e, o be chemically cha ac e ized and used in he expe imen s 197 9 NATURAL Ri e ine A mosphe ic 0% 5% 10% Con ol Ino ganic O ganic Mixed CONTROLLED as a mosphe ic concen a e. Simila ly o i e ine concen a es, quan i a i e 198 concen a ion was main ained excep o he silica e because he educ ion o he 199 ainwa e olume was ca ied ou in a glass o a y e apo a o and he a mosphe ic 200 pa icles we e collec ed on o a glass ibe il e . 201 202 2.4 Na u al and con olled addi ion expe imen s 203 Fo bo h na u al and con olled addi ion expe imen s, 4 L UV- anspa en Whi l-204 pak® bags we e gen ly illed unde dim ligh condi ions un il 2 L o capaci y. Fo he 205 na u al addi ion expe imen s, i e ine and a mosphe ic concen a es we e added o 206 subsu ace seawa e , collec ed as desc ibed in Sec ion 2.1, in p opo ions o 0% (1% 207 ul apu e wa e : 99% seawa e ), 5% (0.5% concen a e: 0.5% ul apu e wa e : 99% 208 seawa e ) and 10% (1% concen a e: 99% seawa e ) o he o iginal (p e ious o 209 concen a ion) i e ine and a mosphe ic ma e ials (Fig. 2). These p opo ions also 210 ensu ed ha he inal salini y o he samples was kep cons an independen ly o he 211 amoun o ex ac added. 212 213 214 215 216 217 218 219 220 Fig. 2: Schema ic ep esen a ion o he na u al and con olled addi ion ea men s pe o med a each 221 season (see de ails in sec ion 2.4). 222 223 16 The le els o signi icance o - es s, pe o med be ween each ea men and he con ol, a e: *p<0.05, 354 **p<0.01, ***p<0.001. 355 In summe , Chl a and PP a es esponded posi i ely in all ea men s, including 356 na u al and con olled addi ions (Fig. 4c, d). The e we e only wo excep ions: PP a es 357 did no espond o o ganic inpu s and he esponse o Chl a o he 10% i e ine inpu 358 was ma ginally signi ican (P=0.07). In gene al, he esponse a io in his expe imen 359 was highe han in sp ing, wi h he esponse a ios o he na u al inpu s being highe 360 han 1.5 and o he con olled ino ganic inpu s eaching 4.5 o Chl a and 7 o PP (Fig. 361 4a-d). In au umn, he e was a nega i e esponse in Chl a o he na u al inpu s (~0.85) 362 and o he o ganic con olled addi ions (~0.40) (Fig. 4e). On he con a y, ino ganic and 363 mixed con olled addi ions led o a sligh inc ease in Chl a concen a ions. PP a es 364 inc eased in esponse o he 10% i e ine inpu s and o he mixed ea men , and 365 dec eased in esponse o he o ganic addi ions (Fig. 4 ). 366 Table 3: Ou pu o mul i a ia e Gene al Linea Model analysis o Chl a, PP and he se e al 367 phy oplank on g oups’ esponse a ios as dependen a iables and Expe imen and T ea men as ixed 368 ac o s. Synecho: Synechococcus- ype cyanobac e ia, APF: au o ophic pico lagella es, ANF: au o ophic 369 nano lagella es. 370 Tes s o Be ween-Subjec s E ec s Sou ce F Sig. Expe imen Chl a 23.860 0.001 PP 34.076 0.000 Synecho 9.405 0.010 APF 20.515 0.001 ANF 24.452 0.001 Dia oms 1.333 0.323 T ea men Chl a 15.742 0.001 PP 24.922 0.000 Synecho 3.788 0.052 APF 57.782 0.000 ANF 10.840 0.003 Dia oms 0.946 0.519 Expe imen * T ea men Chl a 3.385 0.057 PP 10.331 0.002 Synecho 1.638 0.262 APF 11.154 0.002 ANF 3.579 0.050 Dia oms 1.073 0.484 17 Table 4: Bon e oni pos hoc mul iple compa isons om Gene al Linea Models analysis wi h he 371 esponse a ios o Chl a, PP and he se e al phy oplank on g oups as dependen a iables and he 372 expe imen and ea men as ixed ac o s. Synecho: Synechococcus- ype cyanobac e ia, APF: au o ophic 373 pico lagella es, ANF: au o ophic nano lagella es. Fo ea men , only signi ican esul s a e shown. 374 Expe imen Sig. Chl a Sp ing Summe 0.024 Au umn 0.011 Summe Au umn 0.001 PP Sp ing Summe 0.001 Au umn 0.052 Summe Au umn 0.000 Synecho Sp ing Summe 0.012 Au umn 0.136 Summe Au umn 0.510 APF Sp ing Summe 0.545 Au umn 0.001 Summe Au umn 0.012 ANF Sp ing Summe 0.182 Au umn 0.001 Summe Au umn 0.013 Dia oms Sp ing Summe 1.000 Au umn 0.682 Summe Au umn 0.499 T ea men Sig. Chl a Ino ganic Na u al <0.01 O ganic 0.003 PP Ino ganic All <0.01 Synecho A m 10% Mixed 0.037 APF Mixed All <0.01 o ganic All <0.01 ANF Ino ganic Na u al <0.05 O ganic 0.002 Mixed O ganic 0.017 375 376 377 378 18 S a is ical analyses (Table 3) demons a ed ha bo h expe imen and ea men 379 had a signi ican e ec in he esponse a ios o Chl a and PP. The in e ac ion be ween 380 hese wo ac o s was also signi ican o PP esponse a ios. Thus, Chl a and PP 381 esponse a ios we e signi ican ly di e en among he h ee expe imen s (p < 0.01), 382 excep o he PP esponse a io be ween sp ing and au umn (Table 4). Among 383 ea men s, signi ican di e ences we e ound o Chl a and PP esponse a ios be ween 384 na u al inpu s and he ino ganic ea men in he con olled addi ions. The esponse o 385 he ino ganic ea men was also signi ican ly di e en om he o ganic ea men o 386 Chl a and om he mixed and o ganic ea men s o PP. No signi ican di e ences 387 we e ound in Chl a and PP esponse a ios among he samples ea ed wi h bo h ypes 388 and le els o na u al inpu s: i e ine and a mosphe ic. 389 The di e en phy oplank on g oups also esponded di e en ly o he 390 expe imen s and ea men s (Fig. 5, Table 3 & 4). While Synechococcus showed a 391 signi ican inc ease in he o ganic and mixed ea men in sp ing (Fig. 5a), a signi ican 392 dec ease was obse ed in au umn in esponse o a mosphe ic and o ganic addi ions (Fig. 393 5g). In summe , he low abundances o e en absence o Synechococcus in some samples 394 did no allow in e ing eliable esponses o he ea men s. Among APF, posi i e 395 esponses we e obse ed o 10% a mosphe ic, ino ganic and mixed addi ions in sp ing 396 and summe , and nega i e in all expe imen s o o ganic addi ions (Fig. 5b, e, h). ANF 397 showed posi i e esponses in he con olled addi ions and in he 5% addi ion o i e ine 398 inpu s in sp ing (Fig. 5c), and in he ino ganic and mixed ea men s in summe (Fig. 399 5 ). No esponse was obse ed in he au umn expe imen o ANF, excep a dec ease in 400 he esponse o he o ganic addi ions (Fig. 5i). Dia oms showed a high posi i e esponse 401 o he ino ganic ea men in he summe expe imen (Fig. 5d) while a sligh posi i e 402 esponse was also obse ed o he mixed addi ion. In sp ing and au umn, no esponses 403 19 o nega i e e ec s we e obse ed on dia oms, especially o o ganic and mixed 404 addi ions (no shown). 405 406 407 408 409 410 411 412 413 414 415 Fig. 5: Response a ios ( ea men /con ol) o phy oplank on g oups showing signi ican esponses o 416 ea men s in sp ing (a-c), summe (d- ) and au umn (g-i). Synecho: Synechococcus- ype cyanobac e ia, 417 APF: au o ophic pico lagella es, ANF: au o ophic nano lagella es. Ho izon al line deno es he esponse 418 a io = 1. A m.: A mosphe ic inpu s; I: con olled ino ganic addi ions; O: con olled o ganic addi ions; M: 419 con olled mixed addi ions. The le els o signi icance o - es s, pe o med be ween each ea men and 420 he con ol, a e: *p<0.05, **p<0.01, ***p<0.001. In he expe imen s o July and Oc obe , only one bag o 421 each ea men was analyzed o nano and mic ophy oplank on a 48h, so we don’ ha e s a is ical 422 signi icance le els o ANF and dia oms a hese expe imen s. 423 424 I is no iceable ha APF abundances we e be e s imula ed by he combina ion 425 o ino ganic and o ganic compounds ( he mixed ea men ) when compa ed wi h he 426 addi ion o only ino ganic o only o ganic nu ien s (Fig. 5b, e, h). In ac , he o ganic 427 20 ea men seems o be in some way de imen al o APF, showing lowe abundances 428 han in he con ol. By con as , ANF and dia oms ypically g ew mo e wi h he addi ion 429 o only ino ganic nu ien s (Fig. 5c, d, ) han in he mixed ea men . Synechococcus 430 showed a simila pa e n o APF o ganisms, wi h highes g ow h unde he combina ion 431 o o ganic and ino ganic compounds (Fig. 5a, g). Howe e , in his case, he o ganic 432 ea men seemed o be ha m ul o hese o ganisms only in au umn (Fig. 5g). 433 Also o he di e en phy oplank on g oups’ esponse a ios, he expe imen and 434 he ea men we e s a is ically signi ican ac o s (Table 3). Fo Synechococcus, 435 signi ican di e ences we e ound be ween he sp ing and summe expe imen s, while 436 o he small lagella es (APF and ANF) he signi ican di e ences occu be ween he 437 au umn expe imen and he o he wo expe imen s (Table 4). No di e ences we e ound 438 be ween he esponses o he di e en na u al addi ion ea men s. Mos signi ican 439 di e ences in esponse a ios be ween ea men s we e ound be ween na u al and 440 con olled addi ion ea men s. 441 442 Table 5: Pea son co ela ion coe icien s o analyses pe o med be ween Chl a concen a ions, PP a es 443 and abundances o he se e al phy oplank on g oups a he end o he incuba ion, and ini ial ino ganic o 444 o ganic nu ien s le els (including in si u and added) a each expe imen . Synecho: Synechococcus- ype 445 cyanobac e ia, APF: au o ophic pico lagella es, ANF: au o ophic nano lagella es. 446 Chla PP Synecho APF ANF Dia oms SPRING DIN 0.975*** 0.930*** 0.563 0.729* 0.920*** -0.229 DOC 0.422 0.236 0.565 0.341 0.216 -0.497 DON 0.421 0.232 0.573 0.349 0.210 -0.520 SUMMER DIN 0.902*** 0.687* 0.506 0.271 0.922*** 0.710* DOC 0.117 -0.221 0.038 -0.216 0.099 -0.118 DON 0.120 -0.214 0.088 -0.254 0.117 -0.093 AUTUMN DIN 0.657* 0,413 0.642* 0.070 0.104 0.120 DOC -0.208 -0.290 0.281 -0.563 -0.686* -0.775** DON -0.151 -0.289 0.340 -0.621 -0.697* -0.738* 447 21 In o de o explain he di e en phy oplank on esponses a each expe imen , we 448 analyzed hem sepa a ely sea ching o co ela ions wi hin he se e al a iables. Chl a 449 concen a ion was posi i ely co ela ed wi h DIN in he h ee expe imen s (Table 5). 450 Simila ly, also se e al phy oplank on g oups showed posi i e co ela ions wi h DIN, 451 bu no o all expe imen s. DIN was posi i ely co ela ed wi h APF and ANF in sp ing, 452 wi h ANF and dia oms in summe and wi h Synechococcus in au umn. No signi ican 453 co ela ions we e ound be ween he se e al phy oplank on g oups and o ganic 454 compounds (bo h DOC and DON) in any o he h ee expe imen s, excep o ANF and 455 dia oms in au umn which showed a nega i e signi ican co ela ion wi h o ganic 456 compounds. 457 Some di e ences in phy oplank on communi y composi ion we e obse ed in 458 esponse o nu ien amendmen s (Fig. 6). In sp ing, he p opo ion o dia oms inc eased 459 wi h i e ine and a mosphe ic inpu s whe eas wi h con olled addi ions he ela i e 460 dia oms abundance was dep essed and ANF a o ed (Fig. 6a). In au umn an inc ease in 461 he p opo ion o Synechococcus abundance was obse ed in esponse o he o ganic 462 ea men (Fig. 6c). In his same ea men , i was obse ed a dec ease in he p opo ion 463 o APF abundance (Fig. 6c). 464 22 465 Fig. 6: Rela i e abundance o majo phy oplank on g oups a he end o he expe imen in sp ing (a), 466 summe (b) and au umn (c). Synecho: Synechococcus- ype cyanobac e ia, APF: au o ophic 467 pico lagella es, ANF: au o ophic nano lagella es. C: Con ol (no addi ions); I: con olled ino ganic 468 addi ions; O: con olled o ganic addi ions; M: con olled mixed addi ions. 469 470 471 % 0 20 40 60 80 100 % 0 20 40 60 80 100 % 0 20 40 60 80 100 Synecho APF ANF Dia oms Pigmen ed dino lagella es Sp ingSumme Au umn a b c Ri e ine A mosphe ic C5%10% C I O MC 5%10% Con olled 23 4. Discussion 472 Pe o ming he expe imen s in sp ing (May), summe (July) and au umn 473 (Oc obe ) allowed in es iga ing he di e en hyd og aphic condi ions and biological 474 communi ies ypical o his coas al embaymen (Figuei as and Ríos 1993; Figuei as e 475 al. 2002; A bones e al. 2008). Jus a e he i s in ense upwelling pe iod o he yea in 476 sp ing, high Chl a and nu ien concen a ions we e obse ed in su ace wa e s, wi h a 477 ypical communi y domina ed in biomass by chain o ming dia oms (Chae oce os spp.). 478 On he con a y, in summe , despi e he p e ious upwelling e en s, we ound ha he 479 su ace laye was poo in nu ien s and domina ed by small plank on o ganisms. This 480 ac was due o he sinking o he su ace phy oplank on communi ies (domina ed by 481 la ge dia oms de eloped in he p e ious weeks - da a no shown), esul ing om he 482 elaxa ion o upwelling a o able winds. In au umn, ypical oceanic communi ies we e 483 ound a he su ace as a esul o a p e ious s ong downwelling e en which pushed 484 con inen al shel wa e s in o he Ría. High abundance o Synechococcus deno es he 485 p esence o his o sho e wa e inside he Ría de Vigo (Rod íguez e al. 2006). 486 The s a is ically di e en esponse a ios o phy oplank on among he 487 expe imen s pe o med in he h ee di e en seasons indica e ha he ini ial condi ions 488 a each expe imen we e impo an o de e mine he magni ude and he ype o esponse 489 o phy oplank on communi ies o nu ien s inpu s. Di e ences in ini ial condi ions 490 among he expe imen s which can in luence hese phy oplank on esponses may include 491 dissimila i ies in en i onmen al a iables (i.e. nu ien s a ailabili y), in he composi ion 492 o he na u al ma e inpu s o in he s uc u e o phy oplank on communi ies (Ma ínez-493 Ga cía e al. 2015). The high posi i e co ela ions ound be ween Chl a (and se e al 494 phy oplank on g oups) and DIN concen a ions in he incuba ion bags sugges ha DIN 495 a ailabili y is an impo an ac o de e mining he esponse o phy oplank on 496 24 communi ies in his in es iga ion. In addi ion, he highes phy oplank on esponses o 497 he se e al ea men s we e obse ed in summe , when in si u DIN le els we e he 498 lowes . In his case, phy oplank on g ow h seemed o be se e ely limi ed by nu ien s 499 and bo h ypes o inpu s (na u al and con olled) p omo ed enhanced phy oplank on 500 g ow h. A simila si ua ion, al hough wi h a lowe magni ude, occu ed in sp ing. In 501 au umn, howe e , when in si u ino ganic nu ien le els we e he highes , phy oplank on 502 did no espond o e en dec eased in abundance. In his case, in e ac ions be ween 503 se e al ac o s besides DIN mus ha e occu ed as discussed la e on. Ni ogen has been 504 widely ecognized as he main limi ing nu ien in ma ine sys ems when ligh in ensi y 505 is su icien (Wollas 1998). In ac , se e al wo ks ha e also obse ed s imula ion in 506 phy oplank on communi ies due o inc eases in ino ganic ni ogen om a mosphe ic 507 deposi ion (Pae l e al. 1990; Cui e al. 2016) o in con olled addi ions (Piehle e al. 508 2004; Xu e al. 2014). 509 In addi ion o ino ganic nu ien s, i has been obse ed ha o ganic compounds 510 may also s imula e phy oplank on g ow h in coas al a eas (Peie ls and Pae l 1997; 511 Sei zinge and Sande s 1999). In ac , p io esul s in he Ría de Vigo showed ha 512 o ganic ni ogen concen a ion also con ibu ed o he a iabili y in he esponse o 513 p ima y p oduc ion and Chl a o ainwa e addi ions (Ma ínez-Ga cía e al. 2015). 514 Howe e , in ou wo k, no signi ican co ela ions we e ound be ween phy oplank on 515 abundance and o ganic compounds, excep a nega i e co ela ion wi h ANF and 516 dia oms in au umn. The ew signi ican posi i e (al hough low) inc eases in 517 phy oplank on abundance in esponse o he o ganic ea men (Synechococcus and ANF 518 in sp ing and Chl a in summe ) may be ela ed o he occu ence o mixo ophy 519 (including di ec consump ion o o ganic compounds) in bo h seasons (Flynn and Bu le 520 1986; Eile 2006) o o an inc ease in nu ien s emine aliza ion media ed by enhanced 521 25 bac e ial ac i i y in he o ganic ea men in summe (Tei a e al. 2016). Nu ien 522 emine aliza ion in sp ing is unlikely o be impo an , because he e we e high in si u 523 nu ien le els and bac e ia did no espond o he o ganic inpu s in his expe imen 524 (Tei a e al 2016). As he chemical composi ion o he ainwa e addi ions he e was e y 525 di e en om ha in Ma ínez-Ga cía e al. (2015), i should be expec ed ha di e en 526 phy oplank on esponses may occu depending on he composi ion o na u al ma e 527 inpu s in he Ría de Vigo. 528 Mo eo e , some phy oplank on g oups appea ed o be a o ed by he 529 combina ion o ino ganic and o ganic o ms. This is he case o APF in sp ing and 530 summe , which p esen ed a highe esponse o he mixed ea men han o he ino ganic 531 o o ganic addi ions alone (Fig. 5b, e). This esponse can be ela ed o he concomi an 532 use o o ganic o ms, namely amino acids, by hese small phy oplank on cells (Flynn 533 and Bu le 1986; He nández-Ruiz e al., 2018). The same end was obse ed o 534 Synechococcus in sp ing (Fig. 5a) likely ela ed o hei mixo ophic capabili ies (Eile 535 2006). O he explana ions may also include he need o ex e nal seconda y me aboli es 536 o phy oplank on g ow h (eg. B12 i amin) p oduced by he e o ophic bac e ia (C o e 537 al. 2005; P ie o e al. 2016) o he occu ence o bac e i o y by he small euka yo es 538 (Sande s and Gas 2012; Ha mann e al. 2013). In bo h cases, phy oplank on would be 539 bene i ing om he s imula ion o bac e ial g ow h by he combina ion o he o ganic 540 and ino ganic compounds in he mixed ea men (Tei a e al. 2016). On he o he hand, 541 he lowe esponse o ANF and dia oms o he mixed compa ed o he ino ganic 542 ea men could be explained by compe i ion o ino ganic nu ien s wi h he o ganisms 543 s imula ed by he mixed ea men , including APF and bac e ia (Join e al. 2002; 544 Ma ínez-Ga cía e al. 2010). 545 32 Join , I., Hen iksen, P., Fonnes, G.A., Bou ne, D., Things ad, T.F., Riemann, B., 2002. 684 Compe i ion o ino ganic nu ien s be ween phy oplank on and bac e ioplank on in 685 nu ien manipula ed mesocosms. Aqua . Mic ob. Ecol. 29, 145-159. 686 Ké ouel, K., Amino , A.1997. Fluo ime ic de e mina ion o ammonia in sea and 687 es ua ine wa e s by di ec segmen ed low analysis. Ma ine Chemis y 57, 265–275. 688 La uelle, G.G., Roubeix, V., S e a o e,A., B odhe , B., Ciu a,D., Conley, D. J., Du , 689 H. H., Ga nie , J., Lancelo , C., Le Thi Phuong, Q., Meunie , J.-D., Meybeck, M., 690 Michalopoulos, P., Mo iceau, B., Nı´ Longphui , S., Loucaides, S., Papush, L., P es i, 691 M., Ragueneau, O., Regnie , P., Saccone, L., Slomp, C. P., Spi e i, C., and Van 692 Cappellen, P., 2009. An h opogenic pe u ba ions o he silicon cycle a he global 693 scale: Key ole o he land-ocean ansi ion. Global Biogeochem. Cycles, 23, GB4031, 694 doi:10.1029/2008GB003267 695 Lessa d, E.J., Swi , E., 1986. Dino lagella es om he No h A lan ic classi ied as 696 pho o ophic o he e o ophic by epi luo escence mic oscopy. J. Plank on Res. 8, 1209-697 1215. 698 Ma ínez-Ga cía, S., A bones, B., Ga cía-Ma ín, E.E., Teixei a, I.G., Se e , P., 699 Fe nández, E., Figuei as, F.G., Tei a, E., Ál a ez-Salgado, X.A., 2015. Impac o 700 a mosphe ic deposi ion on he me abolism o coas al mic obial communi ies. Es ua . 701 Coas . Shel S. 153, 18-28. 702 Ma ínez-Ga cía, S., Fe nández, E., Ál a ez-Salgado, X.A., González, J., Lønbo g, C., 703 Ma añón, E., Mo án, X.A.G., Tei a, E., 2010. Di e en ial esponses o phy oplank on 704 and he e o ophic bac e ia o o ganic and ino ganic nu ien addi ions in coas al wa e s 705 o he NW Ibe ian Peninsula. Ma . Ecol. P og. Se . 416, 17-33. 706 33 Menden-Deue , S., Lessa d, E.J., 2000. Ca bon o olume ela ionships o 707 dino lagella es, dia oms, and o he p o is plank on. Limnol. Oceanog . 45, 569-579. 708 O e o, P., Ruiz-Villa eal, M., Peliz, A., Cabanas, J.M., 2010. Clima ology and 709 econs uc ion o uno ime se ies in no hwes Ibe ia: in luence in he shel buoyancy 710 budge o Ría de Vigo. Sci. Ma . 74, 247-266. 711 Pae l, H.W., 1997. Coas al eu ophica ion and ha m ul algal blooms: Impo ance o 712 a mosphe ic deposi ion and g oundwa e as “new” ni ogen and o he nu ien sou ces. 713 Limnol. Oceanog . 42, 1154-1165. 714 Pae l, H.W., Rudek, J., Mallin, M.A., 1990. S imula ion o phy oplank on p oduc ion in 715 coas al wa e s by na u al ain all inpu s: Nu i ional and ophic implica ions. Ma . Biol. 716 107, 247-254. 717 Peie ls, B.L., Pae l, H.W., 1997. Bioa ailabili y o a mosphe ic o ganic ni ogen 718 deposi ion o coas al phy oplank on Limnol. Oceanog . 42, 1819-l1823 719 Pé ez, F.F., Padín, X.A., Pazos, Y., Gilco o, M., Cabanas, M., Pa do, P.C., Do al, M.D., 720 Fa ina-Bus o, L., 2010. Plank on esponse o weakening o he Ibe ian coas al 721 upwelling. Glob. Change Biol. 16, 1258-1267. 722 Piehle , M.F., Twomey, L.J., Hall, N.S., Pae l, H.W., 2004. Impac s o ino ganic 723 nu ien en ichmen on phy oplank on communi y s uc u e and unc ion in Pamlico 724 Sound, NC, USA. Es ua . Coas . Shel S. 61, 197-209. 725 P ie o, A., Ba be -Lluch, E., He nández-Ruiz, M., Ma ínez-Ga cía, S., Fe nández, E., 726 Tei a, E., 2016. Assessing he ole o phy oplank on–bac e ioplank on coupling in he 727 esponse o mic obial plank on o nu ien addi ions. J. Plank on Res. 38, 55-63. 728 34 Rod íguez, F., Ga ido, J.L., C espo, B.G., A bones, B., Figuei as, F.G., 2006. Size-729 ac iona ed phy oplank on pigmen g oups in he NW Ibe ian upwelling sys em: impac 730 o he Ibe ian Polewa d Cu en . Ma . Ecol. P og. Se . 323, 59-73. 731 Rod íguez, L., Macías, F., 2006. Eu ophica ion ends in o es soils in Galicia (NW 732 Spain) caused by he a mosphe ic deposi ion o ni ogen compounds. Chemosphe e 63, 733 1598-1609. 734 Sande s, R.W., Gas , R.J., 2012. Bac e i o y by pho o ophic picoplank on and 735 nanoplank on in A c ic wa e s. FEMS Mic obiol. Ecol. 82, 242-253. 736 Sei zinge , S.P., Sande s, R.W., 1999. A mosphe ic inpu s o dissol ed o ganic ni ogen 737 s imula e es ua ine bac e ia and phy oplank on. Limnol. Oceanog . 44, 721-730. 738 Spa ha is, S., Tsi sis, G., Danielidis, D.B., Chi, T.D., Mouillo , D., 2007. E ec s o 739 pulsed nu ien inpu s on phy oplank on assemblage s uc u e and blooms in an enclosed 740 coas al a ea. Es ua . Coas . Shel S. 73, 807-815. 741 S a ham, P.J., 2012. Nu ien s in es ua ies — An o e iew and he po en ial impac s o 742 clima e change. Sci. To al En i on. 434, 213-227. 743 Tei a, E., He nández-Ruiz, M., Ba be -Lluch, E., Sob ino, C., Teixei a, I.G., Ál a ez-744 Salgado, X.A., Nie o-Cid, M., Ma ínez-Ga cía, S., Figuei as, F.G., Fe nández, E., 745 2016. Bac e ioplank on esponses o i e ine and a mosphe ic inpu s in a coas al 746 upwelling sys em (Ría de Vigo, NW Spain). Ma . Ecol. P og. Se . 542, 39-50. 747 Tei a, E., He nando-Mo ales, V., Ma ínez-Ga cía, S., Figuei as, F.G., A bones, B., 748 Ál a ez-Salgado, X.A., 2013. Response o bac e ial communi y s uc u e and unc ion 749 o expe imen al ainwa e addi ions in a coas al eu ophic embaymen . Es ua . Coas . 750 Shel S. 119, 44-53. 751 35 Tei a, E., Ma ínez-Ga cía, S., Ca ei a, C., Mo án, X.A.G., 2011. Changes in 752 bac e ioplank on and phy oplank on communi y composi ion in esponse o nu ien 753 addi ions in coas al wa e s o he NW Ibe ian Peninsula. Ma . Ecol. P og. Se . 426, 87-754 104. 755 Walsh, J.J., 1991. Impo ance o con inen al ma gins in he ma ine biogeochemical 756 cycling o ca bon and ni ogen. Na u e 350, 53-55. 757 Wollas R. In e ac ions o ca bon and ni ogen cycles in he coas al zone, In: Wollas R., 758 Mackenzie F.T. and Chou L. (Eds.), In e ac ions o C, N, P and S Biogeochemical 759 Cycles and Global Change, 1993, NATO ASI Se ies (Se ies I: Global En i onmen al 760 Change), ol 4. Sp inge , Be lin, Heidelbe g.) 761 Wollas , R., 1998. E alua ion and compa ison o he global ca bon cycle in he coas al 762 zone and in he open ocean, in: B ink, K.H., Robinson, A.R. (Eds.), The Sea. John 763 Wiley & Sons, New Yo k, pp. 213-252. 764 Xu, Z.-H., Guo, Z.-R., Xu, X., Huang, D.-J., Sun, X.-X., Jiang, L.-M., Yang, J.-F., 765 2014. The impac o nu ien en ichmen on he phy oplank on and bac e ioplank on 766 communi y du ing a mesocosm expe imen in Nan'ao o Daya Bay. Ma ine Biology 767 Resea ch 10, 374-382. 768     Fig.S1:Tempo ale olu iono  heupwellingindexand uno  alues h oughou 2013. Expe imen da esa ema kedwi hdiscon inuouslines.     Upwelling index (m3 s-1 km-1) -4000 -2000 0 2000 4000 Runo (m3 s-1) 0 50 100 150 200 250 300 350 Jan Feb Ma Ap May Jun Jul Aug Sep Oc No Dec Sp ing Summe Au umn