Spatial and temporal dynamics of larval fish communities in relation to environmental variability in the NW Mediterranean

PhD p og am in Ma ine Sciences
Spa ial and empo al dynamics o la al ish
communi ies in ela ion o en i onmen al
a iabili y in he NW Medi e anean
Doc o al hesis by:
Vanesa Ma ía Raya Rod igál a ez
Thesis ad iso :
Ana Ma ía Saba és F eijo
Ins i u de Ciències del Ma , CSIC
Ba celona, May 2025
This hesis was conduc ed in he Ins i u de Ciències del Ma (ICM – CSIC) and was made
possible hanks o he p ojec s CACO (REN 2002-01339/MAR), FISHJELLY (MAR-CTM2010-
18874) and WINFISH (CTM2015-68543-R). This wo k acknowledges he Se e o Ochoa Cen e
o Excellence acc edi a ion (CEX2019- 000928-S).
A mi amilia, po su apoyo incondicional.
A odas las pe sonas que he encon ado en el camino,
po que cada paso me ha lle ado a donde es oy y a que es a esis sea posible.
i
Table o con en s
Abs ac iii
Keywo ds i
P e ace. P e acio
Acknowledgemen s. Ag adecimien os ix
Lis o ables xiii
Lis o igu es xi
1. In oduc ion 1
1.1. In oduc ion
3
1.1.1. Gene al ea u es o he NW Medi e anean
3
1.1.2. The Medi e anean Sea as a biodi e si y ho spo 6
1.1.3. The Medi e anean Sea as a clima e change ho spo 6
1.1.4. The ele ance o ich hyoplank on s udies 8
1.1.5. La al ish assemblages 10
1.1.6. E olu ion o ish la ae s udies in he Medi e anean Sea 11
1.1.7. La al ish communi ies and en i onmen al condi ions in he NW Medi e anean 13
1.2. Objec i es
15
1.3. Gene al me hods 17
1.3.1. Field sampling and oceanog aphic da a 17
1.3.2. Fish la ae iden i ica ion 19
1.3.3. S a is ical analysis o ish la ae da a 20
2. Resul s 21
Chap e 1. Va iabili y o la al ish assemblages unde con as ing win e en i onmen al
condi ions in he NW Medi e anean 23
Abs ac 25
2.1.1. In oduc ion 26
2.1.2. Ma e ials and me hods 28
2.1.3. Resul s 32
2.1.4. Discussion 43
Re e ences 49
Chap e 2. Recu ence o he spa ial s uc u e o summe la al ish assemblages linked o
hyd odynamics in he NW Medi e anean 61
Abs ac 63
2.2.1. In oduc ion 64
2.2.2. Ma e ials and me hods 65
2.2.3. Resul s 68
2.2.4. Discussion 81
Re e ences 85
Chap e 3.
Long- e m a iabili y o la al ish communi y in he NW Medi e anean in
summe 95
Abs ac 97
2.3.1. In oduc ion 98
2.3.2. Ma e ials and me hods 99
2.3.3. Resul s 101
2.3.4. Discussion 113
Re e ences 118

ii
Chap e 4. The Box-Balance Model: A new ool o assess ish la al su i al, applied o
ield da a on wo small pelagic ish 131
Abs ac 133
2.4.1. In oduc ion 134
2.4.2. Ma e ials and me hods 136
2.4.3. Resul s 141
2.4.4. Discussion 150
2.4.5. Conclusions 155
Re e ences 155
Appendix. Applying he Box-Balance Model (BBM) o ield da a 163
3. Gene al discussion 165
3.1. Seasonali y in la al ish communi ies 168
3.1.1. Di e si y o la al ish communi ies 168
3.1.2. In luence o en i onmen al a iables on la al ish communi ies 169
3.2. La al ish communi ies in he con ex o clima e change 173
3.2.1. Long- e m changes in summe la al ish communi ies 174
3.3. Compa a i e analysis o la al su i al 177
3.4. Fu u e di ec ions 179
4. Conclusions 181
5. Gene al bibliog aphy 185
6. Annex. Published wo k 209
iii
Abs ac
The ea ly de elopmen al s ages o ish, eggs and la ae, ound in he plank onic en i onmen a e subjec
o a high mo ali y. Thus, he s udy o la al su i al has been a key issue in ishe ies science since he
ea ly 20 h cen u y. Spa ial pa e ns in he la al ish communi ies a e in luenced by a complex a ay o
en i onmen al p ocesses ha in e ac s wi h ish biology a di e en empo al and spa ial scales. These
p ocesses include hose o la ge scale, such as clima e pa e ns and seasonal and in e annual
en i onmen al a iabili y, which de e mine adul s’ dis ibu ion and hei spawning s a egies. A local and
sho ime scale, la al ish communi ies a e shaped by he hyd odynamics ha in luence ish la al
dispe sal and e en ion, and by biologic ac o s, such as ood concen a ion and p eda ion, ha ul ima ely
de e mine hei su i al.
This hesis cha ac e ises he s uc u e o he la al ish communi y in summe and win e in he Ca alan
coas (NW Medi e anean), an a ea wi h a wide a ay o en i onmen al condi ions and high
hyd odynamic ac i i y. The aim is o unde s and i s spa ial and in e annual a iabili y in esponse o
changes in en i onmen al condi ions, including oceanog aphic a iables and hyd odynamic p ocesses.
Wi hin he con ex o clima e change, he hesis desc ibes long- e m changes in he s uc u e o he
summe la al ish communi ies and aims o unde s and he in e ac ions be ween la ae o es ablished
species and species ha a e expanding no hwa ds in he a ea.
To in es iga e he in luence o win e en i onmen al condi ions on he s uc u e o ish la al
communi ies, wo win e s, 2017 and 2018, wi h con as ing en i onmen al condi ions we e compa ed.
2017 was mild, while 2018 was mo e se e e, wi h in ense e ical mixing and deep-wa e o ma ion and
cascading e en s ha enhanced shel -slope wa e exchanges. Di e ences in he s uc u e o la al ish
communi y we e ound in ela ion o shel -slope wa e exchange p ocesses.
A high spa ial he e ogenei y in la al ish communi ies was de ec ed in he summe s o 2003, 2004 and
2012, ela ed o en i onmen al ac o s, such as he con inen al shel s uc u e, la i udinal di e ence in
su ace empe a u e, p ima y p oduc i i y, and s a i ica ion le el. Hyd odynamic s uc u es such as
ins abili ies o he No he n Cu en and an icyclonic eddies, also played an impo an ole in he
con igu a ion o hese communi ies.
In summe , o e h ee decades, 1980s, 2000s and 2010s, an inc ease in sea wa e empe a u e and a
dec ease in chlo ophyll we e de ec ed. Changes in he composi ion and abundance o he la al ish
communi y we e also de ec ed. These we e mainly due o he p esence o wa m wa e species in he a ea
o he i s ime, o o hei inc ease in abundance, in he 2000s in ela ion o he no hwa d expansion o
he adul s' ange. O he species showed a decline in abundance o e ime, p obably due o he dec ease in
chlo ophyll.
i
This wo k quan i a i ely compa ed he su i al chances o la ae o E. enc asicolus (an es ablished
species) and S. au i a (a species expanding no hwa ds). To his aim, a new me hod, he Box-Balance
Model, was de eloped o e alua e he ole o hyd odynamic s uc u es in hei mo ali y. The model
e ealed ha despi e he wa ming end would con ibu e o he expansion o S. au i a, i has no ye
de eloped an adap a ion s a egy as success ul as ha o E. enc asicolus, a well-es ablished species in he
a ea.
Keywo ds
Biological oceanog aphy, la al ish dis ibu ion, shel -slope exchanges, deep-wa e o ma ion,
en i onmen al condi ions, hyd odynamic s uc u es, ish spawning habi s, in e annual a iabili y,
la al ish g ow h, la al ish su i al.
P e ace. P e acio
Desde el pun o de is a cien í ico, el Ma Medi e áneo es un sis ema único, po su si uación en e
con inen es, po su ci culación e mohalina a pequeña escala y po albe ga ecosis emas complejos, de
g an biodi e sidad y con un g an núme o de especies endémicas. Es además uno de los ma es donde
los impac os an ópicos son más nume osos y lle an ac uando desde iempos emo os. Los
ecosis emas del Ma Medi e áneo son muy ulne ables, y los peces suponen un elemen o impo an e
de odos es os ecosis emas. Además, los es adios iniciales de la ida de los peces son las e apas que
es án some idas a una mayo p esión. Po su al a ulne abilidad, las la as de peces suponen una pieza
única pa a a e igua cómo a ec an los cambios inducidos po las ac i idades humanas a los
ecosis emas ma inos. Es a esis p e ende con ibui a la comp ensión del papel que desempeñan es as
piezas cla e en el sis ema del Ma Medi e áneo.
Pe o es a esis no su ge solamen e con la in ención de con ibui al conocimien o cien í ico. Del amo
al ma nace es a esis, po que el Medi e áneo es mucho más que una inmensa masa de agua en e
con inen es y mucho más que un ecosis ema único. Nace ambién la ce eza de que la conse ación de
los océanos y en pa icula , del ma Medi e áneo es necesa ia. Y pa a la conse ación es necesa io
p ime o el conocimien o. Po eso es a esis pa a mí signi ica mucho más que un í ulo académico.
xii

xiii
Lis o ables
Table 2.1.1.
Mean abundance and s anda d de ia ion (No. la ae 10 m
-
2
), ela i e
abundance (RA, in %) and equency o occu ence (FO, in %) o all iden i ied la al ish
axa in 2017 and 2018. The axa a e lis ed acco ding o adul habi a (shel and oceanic). 35
Table 2.1.2.
Summa y o he simila i y pe cen age ou ine (SIMPER): P- alue ob ained in
he pai wise compa isons, he i e species ha mos con ibu e o he clus e di e ences
(o de ed om highes o lowes con ibu ion), and he cumula i e pe cen age con ibu ion
o hese di e ences. 38
Table 2.1.3.
Summa y o he canonical co espondence analysis (CCA) esul s. The
indi idual a iables a e o de ed by he pe cen age o a iance explained (Exp. Va .), he
signi icance o each a iable (P) oge he wi h i s es s a is ic (F- alue), and he in e -se
co ela ions be ween he en i onmen al a iables and he i s wo axes o he CCA. 40
Table 2.1.4.
Weigh ed mean dep h (WMD, in m) and s anda d de ia ion, du ing day and
nigh , o he mos abundan la al ish axa, in 2017 and 2018. 44
Table 2.2.1.
Mean abundance and s anda d de ia ion (No. la ae 10 m
-
2
), ela i e
abundance (RA, in %) and equency o occu ence (FO, in %) o all iden i ied la al ish
axa in July 2003, 2004 and 2012. 71 - 72
Table 2.2.2.
Summa y o he canonical co espondence analysis (CCA) esul s. Indi idual
a iables a e o de ed by he pe cen age o a iance explained (Exp. Va .), in ela ion o
he o al and cons ained a iance. The signi icance o each a iable (P) oge he wi h i s
es s a is ic (F- alue), and he in e se co ela ions o en i onmen al a iables wi h he
h ee signi ican axes o he CCA a e p o ided. 76
Table 2.3.1.
Mean abundance and s anda d de ia ion (No. la ae 10m
-
2
) o all iden i ied
la al ish axa in he 1980s, 2000s and 2010s su eys. 103 -105
Table 2.3.2.
Summa y o he edundancy analysis (RDA) esul s. The indi idual a iables
a e o de ed by he pe cen age o a iance explained (Exp. Va .), he signi icance o each
a iable (P) oge he wi h i s es s a is ic (F- alue), and he in e -se co ela ions be ween
he en i onmen al a iables and he i s wo axes o he RDA. 108
Table
2.4.
1.
Summa y o he eg ession analyses o : 1) he co ec ion o he ish la ae
SL a e sh inkage by p ese a ion; 2) he age-leng h ela ionship; and 3) he ca chabili y
ac o (q) o day/nigh samples compa isons (see ex o mo e de ails). SE(a) and SE(b)
a e he s anda d e o o he coe icien s and SE is he s anda d e o o he es ima e. 141
Table
2.4.
2.
Summa y o he Wilcoxon ank-sum es o he di e ences be ween day and
nigh abundances a leng h o Eng aulis enc asicolus and Sa dinella au i a la ae. 142
Table
2.4.
3.
Linea eg ession analyses o la al log-abundances a age o he mo ali y
es ima ion o Sa dinella au i a and Eng aulis enc asicolus by su ey, speci ying he age
anges (S, in days) used. SE is he s anda d e o o he es ima e. Compa isons among
su eys a e shown by he ANCOVA o bo h species. 144
Table
2.4.
4
. Linea eg ession o co ec ed la al log-abundances a age o he mo ali y
es ima ions, as applied o he BBM o S. au i a and E. enc asicolus, speci ying he age
anges used o each su ey (S, in days). SE is he s anda d e o o he es ima e. 145
xi
Lis o igu es
Fig.
1.
1.
Map o he Medi e anean Sea showing he ci cula ion o he A lan ic Wa e and a eas
o deep wa e o ma ion. 3
Fig.
1.
2.
Spa ial ends (°C pe decade) o e 1981–2020, and o al change mul i-model mean
p ojec ed changes om 1995–2014 o 2081–2100 (F om IPCC, 2021). 6
Fig.
1.
3.
P ojec ed wa e cycle changes. Long- e m (2081–2100) p ojec ed annual mean
changes (%) ela i e o p esen -day (1995–2014) o (a) p ecipi a ion and (b) o al uno .
(F om IPCC, 2021). 7
Fig. 1.4.
Diag am o he di e en ac o s ha a ec he o ma ion, main enance and dis up ion
o la al ish assemblages, ac ing a di e en empo al and spa ial scales. Adap ed om Doyle
(1993). 10
Fig. 1.5.
Pee e iewed SCI publica ions be ween 1970 and 2024 con aining he wo ds
ich hyoplank on, ‘ ish la *’ o ‘la al ish’ and ‘la al ish assemblage*/communi *’ (LFA:
La al ish assemblages) wo ldwide and in he Medi e anean. The o dina e axis on he le
indica es he numbe o la al ish publica ions and he o dina e axis on he igh he numbe o
publica ions o la al ish assemblages. Sou ce: Scopus. 12
Fig.1.6.
Sampling a ea co e ed by he di e en su eys conduc ed in he Ca alan coas . The
a ea ma ked wi h he blue s ipe pa e n co esponds o he win e su eys, he yellow a ea o
he summe su eys conduc ed in 2000s and 2010 and he salmon a ea o he s udy pe o med
o e he h ee decades. 17
Fig. 1.7.
Ana omical, mo phological and some specialised cha ac e s o pos - lexion la ae
used o he la al iden i ica ion. 19
Fig. 2.
1.
1.
S udy a ea in he NW Medi e anean, sou h o he Gul o Lion. C osses indica e
sampling s a ions (CTD and Bongo ne ) du ing 2017 (o ange) and 2018 (da k blue) su eys.
T iangles indica e s a ions whe e only CTD cas s we e pe o med. O ange ci cles co espond o
MOCNESS ne samplings. S a s co espond o dep h-s a i ied sampling a ixed s a ions in
2017 (o ange) and 2018 (blue) su eys. Open ellipses (g een, Coas , Shel , Slope and Palamós
canyon) indica e he e ical p o iles o oceanog aphic a iables shown in igu e 4. The
isoba hs shown a e 50, 200, 1000 and 2000 m. 28
Fig. 2.1.2.
Spa ial dis ibu ion o su ace (10 m) empe a u e (SST, a and b), salini y (SSS, c
and d), chlo ophyll-a (SSChla, e and ) and dynamic heigh ela i e o 500 m (DH) o e laid on
densi y (, g and h) in win e 2017 (le panels) and 2018 ( igh panels). A ows indica e he
low di ec ion and in usions o he No he n Cu en and he ex usions o coas al wa e s. The
isoba hs shown a e 50, 200, 1000 and 2000 m. 31
Fig 2.1.3. Ve ical p o iles o densi y (

), mean (ci cle) and s anda d de ia ion (ho izon al
lines) o CTD s a ions sampled in 2017 and 2018. 32
Fig. 2.1.4. Ve ical p o iles o empe a u e ( ed), salini y (blue), densi y (

, black), and
chlo ophyll-a (g een), o ou selec ed s a ions o 2017 and 2018, a di e en loca ions: coas
(a and b), shel (c and d), slope (e and ) and he Palamós Canyon (g and h) ( o s a ion
loca ions see igu e 1). 34
Fig. 2.1.5.
Spa ial dis ibu ion o o al ish la ae (a and b), la ae o shel (c and d) and oceanic
(e and ) ish axa, acco ding o he classi ica ion in Table 1, o e laid on ba hyme y (isoba hs
shown: 50, 200, 1000 and 2000 m). 36
x
Fig. 2.1.6.
(a) Wa d’s hie a chical clus e ing o s a ions in win e 2017 and 2018, and hea map
showing he abundance o he ish la ae (colou g ada ion om whi e o da k ed ep esen s
om absence o he highes abundance), he six clus e s a e indica ed by colou ed boxes
labelled wi h numbe s. (b) Spa ial dis ibu ion o sampling s a ions showing he g oups
iden i ied in he clus e analysis. (c) Non-me ic mul idimensional scaling (nMDS) o dina ion
plo . Symbols o he same colo indica e clus e s o s a ions wi h simila species composi ion.
Ci cles co espond o he s a ions sampled in 2017 and iangles o he s a ions sampled in 2018.
39
Fig. 2.1.7.
O dina ion plo o he canonical co espondence analysis (CCA) showing he
ela ionships be ween he ish la al axa, he sampling s a ions, and he explana o y
en i onmen al a iables ( ec o s) o he win e o 2017 and 2018. Symbols o he same colo
indica e clus e s o s a ions wi h simila species composi ion. Ci cles co espond o he s a ions
sampled in 2017 and iangles o he s a ions sampled in 2018. Abb e ia ions o he axa names
a e de ailed in Table 2.1.1. 40
Fig. 2.1.8.
Spa ial dis ibu ion o sco e alues o s a ions o he axis 1 (a and b) and axis 2 (c
and d) ha esul ed om he canonical co espondence analysis (CCA) o 2017 (a and c) and
2018 (b and d). Colou symbols deno e he s a ions belonging o he g oup iden i ied in he
clus e analysis. (a) and (b): Sco es o dina ion in ela ion o he i s o dina ion axis. (c) and
(d): Sco es o dina ion in ela ion o he second axis. Black line sepa a es posi i e and nega i e
alues. 41
Fig. 2.1.9.
Spa ial dis ibu ion o ish la ae o some ep esen a i e oceanic species o e laid on
sea su ace empe a u e (SST) and ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in
2017 and 2018. 42
Fig. 2.1.10.
Spa ial dis ibu ion o ish la ae o some ep esen a i e shel and uppe slope
species o e laid on sea su ace empe a u e (SST) and ba hyme y (isoba hs shown: 50, 200,
1000 and 2000 m) in 2017 and 2018. No e ha S. pilcha dus and A noglossus spp. a e
ep esen ed on a di e en size scale o M. pou assou and G. mac oph halmus. 43
Fig. 2.1.11.
Violin plo showing he e ical dis ibu ion o la ae o A c ozenus isso,
Myc ophum punc a um, Sa dina pilcha dus and Mic omessis ius pou assou, du ing day and
nigh , o 2017 and 2018. The wid h o he iolin ep esen s he p opo ion o la ae a each
dep h s a um. 45
Fig. 2.2.1.
S udy a ea o he Ca alan coas , in he NW Medi e anean. Blue ci cles indica e
sampling s a ions (CTD and Bongo ne ) du ing he July 2003 and 2004 su eys. C osses
indica e sampling s a ions (CTD and Bongo ne ) du ing he July 2012 su ey. 66
Fig. 2.2.2.
Spa ial dis ibu ion o su ace (5 m) empe a u e (SST; a, b and c), salini y (SSS; d, e
and ), geos ophic cu en ield (a ows) and dynamic heigh (DH) indic ed by he colou
g ading (g, h and i), chlo ophyll-a (SSChla; j, k and l) and pynocline dep h (PD; black lines)
o e laid on he maximum densi y g adien (Gmax; m, n and o), in July 2003 (le panels), 2004
(cen al panels) and 2012 ( igh panels). Da k a ows indica e he loca ion o he Blanes and
Palamós canyons. 69
Fig. 2.2.3.
Spa ial dis ibu ion o abundance o la ae o shel (uppe panels, a, b and c) and
oceanic (cen al panels, d, e and ) ish axa, acco ding o he classi ica ion in Table 1, and
species ichness (bo om panels, g, h and i), o e laid on ba hyme y. Isoba hs shown: 50, 200
and 1000 m. 73
Fig. 2.2.4.
(a) Dend og am showing he classi ica ion o he s a ion assemblages using a Wa d’s
clus e ing me hod based on he B ay–Cu is simila i y ma ix o he abundance o la al ish o
he Ca alan coas in 2003, 2004 and 2012. The six signi ican clus e s a e iden i ied by he
colou squa es. The scale (heigh ) co esponds o he squa ed dis ances. (b) Spa ial dis ibu ion
o sampling s a ions showing he g oups iden i ied in he clus e analysis. Isoba hs shown: 50,
200 and 1000 m. 73
Fig. 2.2.5. Plo o he mean abundance (No. pe 10m
2
) o axa o each clus e . 74
x i
Fig 2.2.6.
Non-me ic mul idimensional scaling (nMDS) o dina ion plo . Symbols o he same
colou indica e clus e s o s a ions wi h simila species composi ion. Squa es co espond o he
s a ions sampled in 2003, ci cles o he s a ions sampled in 2004 and iangles o he s a ions
sampled in 2012. 75
Fig. 2.2.7.
Venn diag am o he a ia ion pa i ioning analysis showing he pe cen age o
a ia ion explained in he la al ish assemblages by each e m, spa ial, en i onmen al, and
in e -annual. 76
Fig. 2.2.8.
O dina ion plo s o he en i onmen al canonical co espondence analysis (CCA) (le
panels, a and c) and o he spa ial CCA ( igh panels, b and d). Uppe panels, a and b:
O dina ion o he sampling s a ions in he CCA plo s, showing he ela ionships be ween he
sampling s a ions and he explana o y en i onmen al a iables (blue ec o s), o July o 2003,
2004 and 2012. Symbols o he same colou indica e clus e s o s a ions wi h simila species
composi ion. Squa es co espond o he s a ions sampled in 2003, ci cles o he s a ions sampled
in 2004 and iangles o he s a ions sampled in 2012. Lowe panels, c and d: O dina ion o he
la al ish axa in he CCA plo s, showing he ela ionships be ween he la al ish axa and he
explana o y en i onmen al a iables (blue ec o s), o July o 2003, 2004 and 2012. The
abb e ia ions o la al ish names a e indica ed in Table 2.2.1. 77
Fig. 2.2.9.
Spa ial dis ibu ion o sco e alues o s a ions o he axis 1 (uppe panels, a, b and c)
and axis 2 (lowe panels, d, e and ) ha esul ed om he en i onmen al canonical
co espondence analysis (CCA) o 2003 (le panels, a and c), 2004 (cen al panels, b and e)
and 2012 ( igh panels, c and ). Symbols o he same colou deno e s a ions belonging o he
same g oup iden i ied in he clus e analysis. Isoba hs shown: 50, 200 and 1000 m. 78
Fig. 2.2.10.
Spa ial dis ibu ion o he i ed sco e alues o s a ions o he axis 1 (colou
g adien ) ha esul ed om he spa ial canonical co espondence analysis (CCA) on he
de ended la al ish ma ix o 2003 (le panel), 2004 (cen al panel) and 2012 ( igh panel).
Geos ophic cu en ield (a ows) and ba hyme y ha e been o e laid. Isoba hs shown: 50, 200
and 1000m. 78
Fig. 2.2.11.
Spa ial dis ibu ion o ish la ae o ep esen a i e shel and oceanic species in
2003, 2004 and 2012. Se anus hepa us o e laid on SSChla (a, b and c), Sa dinella au i a on
SST (d, e and ), Eng aulis enc asicolus (g, h and i), T achinus d aco (j, k and l), and
Ce a oscopelus made ensis (p, q and ) on he geos ophic cu en ield, and A gy opelecus
hemigymnus on ba hyme y (m, n and o). Isoba hs shown: 50, 200 and 1000 m. 80
Fig. 2.3.1.
S udy a ea o he Ca alan coas , in he NW Medi e anean. Open ci cles indica e
sampling s a ions (CTD and Bongo ne ) du ing 1983 su eys, blue ci cles in 2003 and 2004
su eys and c osses in 2011 and 2012 su eys. 100
Fig. 2.3.2.
Box plo s o he en i onmen al a iables o each su ey. (a) Sea su ace empe a u e,
(b) Sea su ace salini y and (c) Sea su ace chlo ophyll-a. In each box plo , he da kened line
indica es he median; boxes show he in e qua ile ange and whiske s indica e he expec ed
ex en o 99% o he da a o a Gaussian dis ibu ion. The le e s a he bo om indica e he
g oups iden i ied by he Tukey's pos hoc es . 102
Fig. 2.3.3.
Box plo s o (a) o al la al abundance (N), (b) species ichness (S, numbe o
species, (c) Shannon-Wiene di e si y index (H’, using a na u al loga i hm) and (d) Pielou’s
measu e o e enness (J’) o each su ey. In each box plo , he da kened line indica es he
median; boxes show he in e qua ile ange and whiske s indica e he expec ed ex en o 99% o
he da a o a Gaussian dis ibu ion. The le e s a he bo om indica e he g oups iden i ied by
he Tukey's pos hoc es . 106
Fig. 2.3.4.
Two-way hie a chical clus e ing o su eys and hea map showing he abundance o
ish la ae (colou g ada ion om whi e o da k ed ep esen s om absence o highes
abundance). G oups o species a e indica ed by colou ed boxes labelled wi h a cod o le e s
and numbe s, and su ey g oups a e indica ed by a di e en colou . 107
x ii
Fig. 2.3.5.
O dina ion plo o he edundancy analysis (RDA) showing he ela ionships
be ween he la al ish axa, he sampling s a ions, and he explana o y en i onmen al a iables
( ec o s) o he summe 1980s, 2000s and 2010s. Blue symbols co espond o he s a ions
sampled in he 1980s, ed symbols o hose in he 2000s and yellow symbols o hose in he
2010s. Abb e ia ions o he axa names a e de ailed in Table 2.3.1. 109
Fig. 2.3.6.
Mean abundance and spa ial dis ibu ion o Eng aulis enc asicolus la ae o e laid
on ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s,
2000s and 2010s. 110
Fig. 2.3.7.
Mean abundance and spa ial dis ibu ion o Sa dinella au i a la ae o e laid on
ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s
and 2010s. 110
Fig. 2.3.8.
Mean abundance and spa ial dis ibu ion o Poma omus sal a ix la ae o e laid on
ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s
and 2010s. 111
Fig. 2.3.9.
Mean abundance and spa ial dis ibu ion o Ca anx honchus la ae o e laid on
ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s
and 2010s. No e ha his species was absen in he 1980s. 111
Fig. 2.3.10.
Mean abundance and spa ial dis ibu ion o Co is julis la ae o e laid on
ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s
and 2010s. 112
Fig. 2.3.11.
Mean abundance and spa ial dis ibu ion o Thalassoma pa o la ae o e laid on
ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s
and 2010s. No e ha his species was absen in he 1980s. 112
Fig. 2.3.12.
Mean abundance and spa ial dis ibu ion o Se anus cab illa la ae o e laid on
ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s
and 2010s. 112
Fig. 2.3.13.
Mean abundance and spa ial dis ibu ion o Myc ophum punc a um la ae o e laid
on ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in June and July o he 1980s,
2000s and 2010s. 113
Fig 2.4.1.
Map o he s udy egion, R, in he NW Medi e anean, showing he sampling s a ions.
G ey a ows indica e he ypical pa h o he No he n Cu en along he con inen al slope. 136
Fig 2.4.2. Age-leng h ela ionships o E. enc asicolus and S.au i a la ae. 141
Fig 2.4.3.
Abundances a leng h o E. enc asicolus and S. au i a la ae (ba s, le axis),
showing day/nigh di e ences. Ra io day/nigh (ci cles) o he adjus men o he cu e o
ob ain he ca chabili y ac o (q, igh axis) (See ex o de ails). 143
Fig 2.4.4.
La al mo ali y a es o E. enc asicolus and S. au i a in each su ey, es ima ed
om log-abundances a age, wi hou applying he day-nigh co ec ion (uppe panels), and
applying he co ec ion (lowe panels) used in he BBM. 143
Fig 2.4.5.
Maps o he su ace ci cula ion o e laid on he balances (see me hods) es ima ed a
each sampling s a ion, showing he s uc u es o la al impo / e en ion and expo /dispe sion
a eas in each su ey, o E. enc asicolus (le ) and S. au i a ( igh ). Eddies obse ed in Saba és
e al (2013) a e iden i ied. The whi e a eas indica e he sampling s a ions whe e he balance
de e mina ion was no possible. 146

x iii
Fig 2.4.6
. Box-balances in a eas o ne la al impo : The no h impo a ea (July 2003 and July
2004; uppe ows) and he sou h impo a ea (June 2004; lowe ow) o E. enc asicolus and S.
au i a. The e e ence maps (le column) indica e hei espec i e posi ion in he egion, A is he
a ea analysed. N indica es he numbe o sampling s a ions pooled. Solid line in ba g aphs
indica es he la al abundance a age p edic ed by he mo ali y es ima e. The ba s indica e he
abundances a age: in whi e hose ha do no con ibu e o he mo ali y es ima es, in blue he
po ion ha exceed he p edic ed alue (gain), and in ed he abundance ha would be equi ed
o each he p edic ed alue (loss). 147
Fig 2.4.7
. Box-balances in one o he e en ion a eas, he eddy A3 in July 2004 (uppe panels),
i s neighbou hood (cen al panels) and he o al cen al a ea (eddy A3 and i s neighbou hood)
(lowe panels). Desc ip ions o panels as in igu e 6. 148
Fig 2.4.8
. Box-balances in one o he dipola s uc u es, he eddy C2 in June 2004. Desc ip ions
o panels as in igu e 6. 149
Fig 2.4.9
. Box-balances in he sou h a ea o ne la al expo in June 2004 (uppe panels) and
July 2004 (lowe panels). Desc ip ions o panels as in igu e 6. 149
Fig. 3.1.
Schema ic o some po en ial e ec s o clima e- ela ed and o he an h opogenically
induced changes on ish ea ly li e his o y. The complexi y and po en ial o in e ac ions o
e ec s a e e iden om he many a ows om he po en ial s esso s (da k ed boxes) sha ed by
each in luen ial ecosys em componen (i.e., p eda o s and p ey o ea ly li e s ages, da k g een
boxes) and ea ly li e s age p ocess (da k blue boxes). Da k ed lines indica e di ec e ec s on
ea ly li e s ages, while yellow lines a e e ec s on p eda o s and p ey o ea ly s ages ha can, in
u n, impac ea ly li e su i al. This diag am is no exhaus i e in he numbe o po en ial e ec s
o s esso s and he s esso s a e b oad (e.g., “habi a deg ada ion” could ange om chemical
pollu ion o sound pollu ion, pelagic en i onmen s o ben hic) and a e no mu ually exclusi e.
To educe complexi y, in e ac ions among ea ly li e s age p ocess boxes a e no included.
Adap ed om Llopiz e al (2014). 173
1. INTRODUCTION
1. In oduc ion
3
1.1. In oduc ion
1.1.1. Gene al ea u es o he NW Medi e anean
The Medi e anean Sea is a unique ma ine egion, cha ac e ized by being su ounded almos comple ely
by land be ween con inen s (Fig. 1.1). I is connec ed o he A lan ic Ocean h ough he na ow S ai o
Gib al a (wid h~13 km, sill dep h~300 m) and o he Black Sea h ough he Da danelles - Ma ma a Sea -
Bospho us sys em. The Medi e anean is also connec ed o he Red Sea h ough he Suez Canal, al hough
in an a i icial manne . The S ai o Sicily di ides he Medi e anean Sea in o wo deep basins, and
connec s he Wes e n Medi e anean wi h he Ionian Sea (Robinson and Golna aghi, 1994). The
Medi e anean is ela i ely small (co e s a o al a ea o 2.5 million km2) and is a deep basin (a e age
dep h o 1460 m, maximum 5267 m). Con inen al shel es a e ypically na ow and accoun o less han
20% o he o al sea su ace. The ne e apo a ion exceeds p ecipi a ion and i e inpu s unc ioning as a
concen a ion basin wi h high salini y (37.5 – 39.5). Th ough he S ai o Gib al a , he Medi e anean
impo s nu ien poo A lan ic wa e and expo s emine alised nu ien ich wa e , esul ing in i s
oligo ophy and low p ima y p oduc i i y (Tanhua e al., 2013).
Fig. 1.1. Map o he Medi e anean Sea showing he ci cula ion o he A lan ic Wa e and a eas o deep wa e o ma ion.
The gene al ci cula ion o he Medi e anean is he mohaline and cu en pa e ns a e cyclonic (Millo ,
1999) (Fig. 1.1). The in low o A lan ic wa e , o ela i ely low salini y (~36.2), h ough he S ai o
Gib al a , ollows wo main pa hs. One pa ollows he No h A ican coas showing in ense mesoscale
ac i i y and en e s he eas e n basin h ough he S ai o Sicily (Millo and Taupie -Le age, 2005). The
o he pa lows no hwa ds ollowing he no he n side o he Balea ic Islands o he wes coas o
Co sica (Ga cía e al., 1994). On he eas e n side o he Ligu ian Sea he A lan ic wa e low om he
1. In oduc ion
10
ac i i y (Pepin 1991). Also e y impo an a e physical p ocesses, such as cu en s, which de ine la al
anspo pa hways o la ae o a ou able o un a ou able a eas, and on s and eddies ha a ou he
concen a ion/ e en ion o ish la ae in a eas sui able o hei su i al (Bakun, 2006; Go oni, 2005).
1.1.5. La al ish assemblages
A la al ish assemblage, o la al ish communi y, is a g oup o species whose la ae a e collec ed in he
same a ea a he same ime, and should be conside ed a s a ic snapsho ha does no necessa ily imply o
equi e e olu iona y con e gence o biological in e ac ions (e.g., compe i ion, p eda ion). I is ansien
by de ini ion, and he ju enile and adul phases o hese species may no concu in ime o space (Mille
2002). Co-occu ence o indi idual species in an assemblage sugges s ha hey sha e common
equi emen s du ing hei ea ly li e. An al e a ion in he ecosys em ha nega i ely a ec s one species is
likely o nega i ely a ec all species in he assemblage. Thus, by s udying he assemblage as a whole, i is
possible o ha e a be e unde s anding o he p ocesses ha egula e la al su i al o sol e he
' ec ui men p oblem'. To do his, i is c ucial o know he p ocesses in ol ed in he o ma ion o la al
ish assemblages, hei main enance and hei dis up ion (Mille , 2002). These p ocesses include a
complex a ay o physical and biological ac o s ha in e ac s wi h he biology o ish popula ions a
di e en empo al and spa ial scales (Doyle e al., 1993) (Fig. 1.4).
Fig. 1.4. Diag am o he
di e en ac o s ha a ec
he o ma ion, main enance
and dis up ion o la al ish
assemblages, ac ing a
di e en empo al and
spa ial scales. Adap ed om
Doyle (1993).
The main ac o s in ol ed in he o ma ion o la al ish assemblages a e he dis ibu ion ange o adul
ish popula ions and hei ep oduc i e s a egies, which ha e e ol ed o adap o he dominan
oceanog aphic cha ac e is ics o he egion in which hey a e ound (Pa ish e al., 1981). Thus, he
speci ic composi ion, dis ibu ion, and abundance o ish la ae wi hin he complex and a iable ma ine
ecosys ems a e he esul o a p ocess o co-adap a ion o species' spawning s a egies (e.g., iming and
loca ion o spawning, du a ion and equency o spawning) (Doyle e al., 1993). As a esul o hese

1. In oduc ion
11
s a egies, o sp ing a e placed whe e la ae can ind he bes condi ions o success ul eeding, g ow h
and su i al (Cowen and Sponaugle, 2009; Cushing, 1969; Siddon e al., 2011). Annual o la ge -scale
a iabili y in en i onmen al ac o s, such as clima e change, can lead o a ia ions in he dis ibu ion
ange o adul ish and in hei ep oduc i e s a egies (Fig. 1.4). These a ia ions a e e lec ed in he
dis ibu ion pa e ns and abundance o hei la ae (Doyle e al. 1993). Thus, changes in he la al ish
communi y p o ide ea ly indica o s o a ia ions in ish communi y dynamics (Doyle e al., 2009) and in
oceanog aphic and clima ic condi ions (Au h e al., 2011; McCla chie e al., 2018). Spa ial mesoscale and
seasonal changes in en i onmen al condi ions may cause ish spawning o occu a a place and ime no
sui able o success ul eeding, g ow h and su i al o la ae (Cushing 1969; Llopiz e al 2014).
The e o e, he seasonali y o he spawning pe iod o ish also plays a ole in he o ma ion o la al ish
communi ies, especially in mid and high la i udes, whe e seasonal a iabili y o en i onmen al condi ions
is g ea e han in low la i udes (Mille 2002). Some physical p ocesses can gene a e mac oscale s uc u es
(e.g. wa e mass sepa a ion on s, la ge cu en s) and mesoscale s uc u es (e.g. con e gen on s,
eddies), which can concen a e eggs and la ae, hus con ibu ing o he o ma ion o he assemblages
(Bakun 2006; Go oni 2005).
Physical p ocesses can also con ibu e o he main enance o la al ish communi ies by anspo ing
la ae o a ou able a eas o e aining hem in mesoscale s uc u es, o lead o dis up ion by anspo ing
hem o un a ou able a eas (Mille 2002). The main biological ac o s ha con ibu e o he main enance
o la al ish communi ies o lead o hei dis up ion a e hose ha in luence la al g ow h and su i al,
such as eeding and p eda ion, as men ioned abo e. Compe i ion o ood occu s mainly among la ae in
ad anced s ages and is impo an when p eys a e a low concen a ion. In his case, compe i ion could
lead o a educ ion o ood o le els ha would esul in lowe su i al o la ae o all species in he
communi y (Mille 2002). To a oid compe i ion o ood and space in he ea ly s ages, ish species (o en
belonging o he same genus o amily) ha occupy he same a ea and equi e simila habi a s spawn in
di e en ime pe iods and, in mos cases, no spa ially o e lapping (Saba és e al., 2007a; Tsikli as e al.,
2010).
1.1.6. E olu ion o ish la ae s udies in he Medi e anean Sea
Since he 1970s, he e has been a s eady inc ease in he numbe o published a icles on ich hyoplank on
wo ldwide (Fig. 1.5). Since he mid-1990s, his ype o s udies has become inc easingly ele an , showing
ha his is an ac i e opic in science. In he Medi e anean, he numbe o published wo ks on ish la ae
has inc eased signi ican ly since he 2010s. Mos ich hyoplank on s udies ocus on a limi ed numbe o
species, mainly hose o comme cial in e es . Resea ch on la al ish assemblages s a ed in he mid-
1990s, and ep esen a small pe cen age (<5%) o he o e all numbe o ish la ae s udies. A signi ican
p opo ion o his wo k on la al assemblages has been ca ied ou in he Medi e anean.
1. In oduc ion
12
Fig. 1.5. Pee e iewed SCI
publica ions be ween 1970 and 2024
con aining he wo ds
ich hyoplank on, ‘ ish la *’ o
‘la al ish’ and ‘la al ish
assemblage*/communi *’ (LFA:
La al ish assemblages) wo ldwide
and in he Medi e anean. The
o dina e axis on he le indica es he
numbe o la al ish publica ions and
he o dina e axis on he igh he
numbe o publica ions o la al ish
assemblages. Sou ce: Scopus.
In he Medi e anean, ea ly esea ch on ish la ae was aimed o desc ibe he la al de elopmen o ish
species o iden i ica ion pu poses, as well as o p o ide in o ma ion on loca ions and pe iods o
spawning a eas (e.g. Demi , 1961; Ma ina o, 1991). Among hem, i is wo h men ioning he excellen
wo k o Lo Bianco e al. (1956), which compiled a high numbe o desc ip ions o he iden i ica ion o
ish la ae specimens o he Medi e anean ish species. This pionee ing s udy cons i u es he basis o
ich hyoplank on esea ch in he Medi e anean egion, and o his day, hese desc ip ions se e as
e e ence guide o he iden i ica ion o specimens om ield samples.
Subsequen ly, esea ch on undamen al aspec s o la al biology was ini ia ed in some species, mainly he
Eu opean ancho y (Eng aulis enc asicolus) and he Eu opean sa dine (Sa dina pilcha dus).
Imp o emen s in o oli h eading and analysis echniques made i possible o add ess la al g ow h (e.g.
Dulčić, 1995; Palome a e al., 1988; Regne , 1980) and o quan i y he la al mo ali y a es (Palome a
and Lleona , 1989; Regne , 1985).
A c ucial opic is he analysis o la al eeding o nume ous species, he small pelagics (Bo me e al.,
2013; Ca alán e al., 2010), unas (Mo o e e al., 2008a; U ia e e al., 2019), ed mulle (Saba és e al.,
2015), spa ids (Sánchez-Velasco and No bis, 1997) and mesopelagic ish (Con e as e al., 2015; Saba és
and Saiz, 2000). S udies ha e p o ided aluable in o ma ion on ma ine ood webs, and g ea ad ances
ha e been possible hanks o he in oduc ion o no el echniques such as s able iso ope (Mi -A guimbau
e al., 2021; Quin anilla e al., 2020), a y acids (Cos alago e al., 2011; Rossi e al., 2006) and molecula
analyses (U ia e e al., 2019; Yeb a e al., 2019). Resea ch on he ela ionship be ween la al g ow h and
condi ion and i s in luence on la al su i al began in he 2000s (Ca alán e al., 2006; Cos alago e al.,
2011; Ramí ez e al., 2004).
As in all scien i ic ields, esea ch on ish la ae has been linked o di e en echnological ad ances. The
use o mul iple ne s and o he sampling sys ems ha allow he collec ion o samples a di e en dep h
le els o he wa e column has allowed he s udy o he e ical dis ibu ion o ish la ae, along wi h ha
1. In oduc ion
13
o hei p ey. These s udies ha e shown ha he la ae o mos species a e loca ed nea he su ace abo e
he he mocline (e. g. G ana a e al., 2011; Oli a and Saba és, 1997; Saba és, 2004), while he la ae o
oceanic species ha e p e e ence o subsu ace laye s (Oli a e al., 2010). Some s udies ha e shown ha
ce ain species pe o m daily e ical mig a ions o eeding (Oli a e al., 2001; Saba és e al., 2008;
Soma akis and Nikolioudakis, 2007). Ad ances in oceanog aphic da a collec ion and emo e sensing
sys ems ha e allowed p og ess in he in e ac ion be ween en i onmen al a iables, e.g. empe a u e,
salini y and chlo ophyll-a, as well as hyd odynamic s uc u es in he spa ial and empo al dis ibu ion o
ish la ae. Imp o emen s in compu ing ha e made i possible o pe o m complex s a is ical analyses
e icien ly and implemen indi idual based models o s udy connec i i y be ween a eas (Ma iani e al.,
2010; Ospina-Al a ez e al., 2015) and habi a models o o ecas la al ish dis ibu ions o pelagic
species unde clima e change scena ios (Macías e al., 2014; Maynou e al., 2020a).
Despi e all hese ad ances, he e a e s ill gaps in ou knowledge abou he biology and ecology o ish
la ae. I is impo an o u he p og ess in opics such as he axonomy o ish la ae, and undamen al
aspec s o hei biology (e.g. die , g ow h a es, swimming abili y, e c.) since all hese aspec s a e c ucial
o hei su i al. O he impo an aspec s o conside a e la al p eda ion by nume ous axonomic g oups
(jelly ish, c eno o s, adul ish, e c.) as well as compe i ion o esou ces.
1.1.7. La al ish communi ies and en i onmen al condi ions in he NW Medi e anean
The la al ish communi y in he NW Medi e anean was i s desc ibed by Saba és (1990). In his wo k,
he spa io- empo al dis ibu ion o ish la ae was analysed and a ma ked seasonali y in he p esence o
la ae in he plank on was e idenced in ela ion o he spawning pe iod o he adul s. This seasonali y has
also been desc ibed in subsequen wo ks in he a ea (Alemany, 1997; Ál a ez e al., 2012; Oli a e al.,
2010) and in o he Medi e anean egions (Kou akis e al., 2004; Za ad e al., 2020). Mos
Medi e anean ish species spawn du ing la e sp ing and summe , unde condi ions o oligo ophy and
s a i ica ion (Tsikli as e al., 2010). Summe ich hyoplank on is cha ac e ised by high species di e si y
and comp ises an ex ensi e assemblage o ne i ic species (e.g. Spa idae, Lab idae, Mullidae, Se anidae,
Scomb idae), small and medium pelagics (e.g. Eng aulidae, Clupeidae, Ca angidae) and oceanic species
(e.g. Myc ophidae, Gonos oma idae). In win e , when he wa e column is well mixed, species om
ela i ely cold wa e ep oduce, such as Sa dina pilcha dus, and o he species belonging o he amilies
Ammody idae, Gadidae and Pleu onec idae (Saba és e al., 2007a).
In addi ion o he spawning pe iod, he de e mining ac o in he dis ibu ion o ish la ae, and he e o e
in ish la al assemblages, is he habi a o he adul s and he loca ion whe e hey spawn. In he NW
Medi e anen, as in o he geog aphic egions, dep h explains he dis ibu ion o ish la ae, usually
sepa a ing ne i ic om oceanic species (Isa i e al., 2008; Saba és, 1990; Soma akis e al., 2011). Saba és
and Oli a (1996) demons a ed ha he densi y on associa ed wi h he No he n Cu en limi ed he
dis ibu ion o la ae o coas al species owa ds he open sea, sepa a ing he la ae o ne i ic and
1. In oduc ion
14
mesopelagic species. Mesoscale s uc u es associa ed wi h he No he n Cu en , such as an icyclonic
eddies, ha e demons a ed o e ain and concen a e ancho y la ae ha a e anspo ed by he cu en
(Saba és e al., 2013, 2007b).
Mos o he s udies ca ied ou along he Ca alan coas dealing wi h he ole o en i onmen al ac o s (e.g.
ba hyme y, empe a u e and chlo ophyll-a) and hyd odynamic s uc u es (e.g. cu en s, an icyclonic
eddies) on he dis ibu ion, concen a ion and dispe sal o ish la ae ha e been ca ied ou on pelagic
species, such as Eng aulis enc asicolus and Sa dinella au i a (Saba és e al., 2013, 2009, 2004). The
s udies add essing he e ec o en i onmen al ac o s on he ish la al communi y as a whole a e sca ce
and es ic ed o a limi ed a ea (Oli a e al., 2010; Saba és and Oli a , 1996). In addi ion, hese s udies
ha e been conduc ed in he sp ing-summe pe iod and he in o ma ion a ailable on his subjec o he
win e pe iod is sca ce and es ic ed o a ew species (sa dine, blue whi ing and mesopelagic ish) (Oli a
e al., 2003; Saba és e al., 2024). In he global wa ming con ex expe ienced by he Medi e anean, cold-
wa e ish species ha ep oduce in win e would be ad e sely a ec ed (Llo e , e al. 2015). The Ca alan
coas , loca ed sou h o he Gul o Lion, is one o he coldes a eas in he Medi e anean, whe e deep
con ec ion akes place in win e , and cold-wa e ish species, he mos h ea ened by he clima e change,
a e pa icula ly abundan .
As men ioned abo e, a no hwa d expansion o ish and la ae o wa m wa e species has been
documen ed along he Ca alan coas in ela ion o sea wa ming (Raya and Saba és, 2015; Saba és e al.,
2012, 2006). On a global scale, s udies on he in luence o clima e a iabili y on ish la ae communi y
a e e y sca ce (e.g. Au h e al., 2017; Koslow e al., 2013; Nielsen e al., 2020) and in he Medi e anean,
al hough hey ha e been add essed a he le el o a pa icula species (Maynou e al., 2020b; Saba és e
al., 2015, 2006), hey ha e no been conduc ed in he en i e la al ish communi y. Some ques ions
emain unknown, such as whe he he coexis ence o la ae o indigenous species wi h hose o newly
a i ed species implies possible compe i ion o esou ces o space. The in luence o ising sea
empe a u e on he la ae o es ablished ish species in he a ea also emains o be in es iga ed.
This hesis add esses hese ques ions by analysing he spa ial and in e annual a iabili y o he la al ish
communi ies on he Ca alan coas (NW Medi e anean) and explo es i s long- e m a iabili y in a clima e
change con ex . This egion, wi h i s high hyd odynamic ac i i y and wide ange o en i onmen al
condi ions (Saiz e al., 2014; Sala , 1996) p o ides an ideal loca ion o analyse he in luence o
en i onmen al ac o s on he spa ial s uc u e o he la al ish communi y. To elucida e whe he he
coexis ence be ween la ae o well es ablished species (E. enc asicolus) and a species ha is expanding
no hwa ds in ela ion o sea wa ming (Sa dinella au i a), which occupy a simila ophic niche, can be
de imen al o he la ae o E.enc asicolus, he su i al o bo h species is analysed. To his end, a new
ool (Box-Balance Model) has been de eloped o quan i y he dispe sal and e en ion o la ae o bo h
species and o e alua e he ole o hyd odynamic s uc u es in hei su i al along he Ca alan coas .
1. In oduc ion
15
1.2. Objec i es
The p esen hesis cha ac e izes he s uc u e o he la al ish communi y, i.e. i s composi ion, abundance
and spa ial dis ibu ion o he species in summe and win e , in he NW Medi e anean. The objec i e is o
unde s and i s spa ial and in e annual a iabili y in esponse o changes in en i onmen al condi ions,
including oceanog aphic a iables and hyd odynamic p ocesses. Wi hin he cu en con ex o clima e
change in he Medi e anean, he hesis aims o desc ibe long- e m changes in he s uc u e o he summe
la al ish communi ies and o unde s and he in e ac ions be ween la ae o well es ablished species and
species ha a e expanding no hwa ds in he a ea.
The gene al aim o he hesis is di ided in o se e al speci ic objec i es, which a e add essed in he
di e en chap e s:
Chap e 1. Va iabili y o la al ish assemblages unde con as ing win e en i onmen al
condi ions in he NW Medi e anean
This chap e p o ides he i s cha ac e isa ion o la al ish communi ies in wo consecu i e win e s o
con as ing en i onmen al condi ions in he Ca alan coas . The Ca alan coas is loca ed sou h o he Gul
o Lion, one o he coldes a eas in he Medi e anean, whe e deep con ec ion akes place in win e , and
cold wa e ish species, he mos h ea ened by he clima e change, a e pa icula ly abundan . In his
con ex , he objec i es a e: (a) o desc ibe he species composi ion, abundance and di e si y o he ish
la al communi y in win e , (b) o analyse he in luence o hyd odynamic p ocesses on he s uc u e o he
la al ish assemblages, and (c) o asce ain he ex en o which in e annual shi s in he la al ish
assemblages a e indica o s o changes in en i onmen al condi ions.
Chap e 2. Recu ence o he spa ial s uc u e o summe la al ish assemblages linked o
hyd odynamics in he NW Medi e anean
The objec i e o his chap e is o assess he ole o di e en sou ces o a iabili y, including
en i onmen al, spa ial and in e annual, on he con o ma ion o la al ish assemblages in summe along
he Ca alan coas . In pa icula , he in luence o en i onmen al ac o s (e.g. ba hyme y, empe a u e and
p ima y p oduc i i y) and mesoscale hyd odynamic s uc u es (e.g. cu en s, an icyclonic eddies) on he
spa ial s uc u e o he la al ish communi y is in es iga ed
Chap e 3. Long- e m a iabili y o la al ish communi y in he NW Medi e anean in summe
This chap e in es iga es he main changes expe ienced by he summe la al ish communi y o e h ee
decades along he Ca alan coas . The objec i es a e (a) o iden i y he long- e m changes in he species
composi ion, abundance and dis ibu ion o he la al ish communi y om 1980s o 2010s, and (b) o
explo e he links be ween en i onmen al a iables and he la al ish communi y ends wi hin he cu en
con ex o clima e change in he Medi e anean

1. In oduc ion
16
Chap e 4. The Box-Balance Model: A new ool o assess ish la al su i al, applied o ield da a
on wo small pelagic ish
In he ligh o he no hwa d expansion and inc easing abundance o wa m wa e ish species in he NW
Medi e anean, he objec i e o his chap e is o quan i a i ely compa e he su i al o la ae o
Eng aulis enc asicolus (well es ablished species) and Sa dinella au i a (species ha is expanding
no hwa ds) and o iden i y sui able (o unsui able) a eas o hei de elopmen . To add ess hese issues,
a new ool, he Box-Balance Model, has been de eloped, which allows he quan i ica ion o la al
dispe sion and e en ion by mesoscale dynamics in combina ion wi h g ow h and mo ali y a es o he
wo species.
1. In oduc ion
17
1.3. Gene al me hods
1.3.1. Field sampling and oceanog aphic da a
Ich hyoplank on da a used in his hesis come om biological-oceanog aphic esea ch p og ams
conduc ed by he Ins i u e o Ma ine Sciences (CSIC) in Ba celona (Fig. 1.6): ARECES 1980s (June, July
and Sep embe 1983), CACO 2000s (July and Sep embe 2003; June and July 2004), FISHJELLY 2010s
(June 2011 and July 2012) and WINFISH (Feb ua y 2017 and 2018). In all summe su eys ca ied ou
o he Ca alan coas (NW Medi e anean) he same a ea was co e ed ( om 40.15˚N o 42˚N and om
0.5˚E o 3.5˚E), be ween he coas and he con inen al shel b eak, and he same sampling me hodology
was applied. The win e su eys we e conduc ed in he no he n pa o he Ca alan coas ( om 41.2˚N o
42.3˚N and om 2.6˚E o 3.8˚E), using he same sampling me hodology. In he summe su eys ca ied
ou in he 2000s and 2010s and in he win e su eys he sampling a ea was ex ended owa ds he open
sea. All su eys we e ca ied ou wi h he R/V Ga cia del Cid.
Fig.1.6. Sampling a ea co e ed by he di e en su eys conduc ed in he Ca alan coas . The a ea ma ked wi h he blue s ipe
pa e n co esponds o he win e su eys, he yellow a ea o he summe su eys conduc ed in 2000s and 2010 and he salmon
a ea o he s udy pe o med o e he h ee decades.
A each s a ion, e ical p o iles o he basic hyd og aphic a iables ( empe a u e and salini y) we e
ob ained wi h a conduc i i y- empe a u e-dep h (CTD, Neil B own, WOCE s anda d), equipped wi h a
Sea-Tech in si u luo ome e . To calib a e he luo ome e , wa e samples we e collec ed a di e en
le els o he wa e column a each s a ion o de e mine chlo ophyll-a concen a ion. Wa e samples we e
il e ed h ough Wha man GF/F il e s and ozen. In he labo a o y, he chlo ophyll-a concen a ion
(μg l-1) was de e mined luo ome ically ollowing he me hod o Yen sch and Menzel (1963).
Chlo ophyll-a was ex ac ed om il e s imme sed in 6 mL o 90% ace one (24 h a 4 ºC in da kness),
1. In oduc ion
18
and he ex ac was analysed wi h a Tu ne Designs luo ome e calib a ed wi h pu e chlo ophyll-a. The
e ical p o iles ob ained by he CTD we e in e pola ed o 1-m dep h in e als. Con ou maps o
en i onmen al a iables (mainly empe a u e, salini y, densi y and chlo ophyll-a) we e d awn using
SURFER 13 (Golden So wa e, 2015).
Fish la ae we e collec ed a each s a ion du ing day and nigh , using a Bongo ne wi h a mou h opening
o 60 cm o diame e and a mesh size o 300 μm. The ne was owed a 2 kno s, obliquely, om a
maximum dep h ha depends on he su ey, o he su ace. The olume o il e ed wa e by he ne s was
es ima ed by means o a lowme e placed in he cen e o each ne mou h. In addi ion, in he win e
su eys dep h-s a i ied samplings we e pe o med using a MOCNESS ne o 1 m2 mou h opening and
300 μm mesh size. Zooplank on samples we e ixed in 5% o maldehyde bu e ed wi h sodium
e abo a e.
1.3.2. Fish la ae iden i ica ion
In he labo a o y, all ish la ae we e so ed om he p ese ed samples unde a s e eomic oscope. La ae
we e coun ed and iden i ied o he lowes axonomic le el possible by he same g oup o expe s, enabling
di ec compa ison be ween all he su eys o he same spawning season. The iden i ica ion o la ae
canno be done by dicho omous keys due o he impo an changes ha la ae unde go du ing hei
de elopmen . The assignmen o a specimen o a pa icula species is based on he desc ip ions o la al
de elopmen exis ing in he li e a u e. Fo his hesis, nume ous guides and o he publica ions we e used
o he iden i ica ion o ish la ae. Some o hem, speci ic o he Medi e anean, such as Lo Bianco e al.
(1956), Saba és (1988), Alemany (1997) and Demi (1961), and o he wo ks such as Mose e al. (1984)
Oli a and Fo uño (1991), Aboussouan (1975, 1967), Fahay (2007), and Richa ds (2005).
Usually, he p ocess o la al iden i ica ion is ca ied ou in se e al s eps. The i s s ep is o iden i y
unknown la ae by o de o amily, based on shape and gene al appea ance, ollowed by me is ic alues,
pigmen , and mo phology (Mille and Kendall, 2009). Some cha ac e s a y wi h la al de elopmen ,
hus size and s age o de elopmen should be conside ed when compa ing he shape o an unknown la a
wi h illus a ions and desc ip ions o known specimens.
The cha ac e s used in he iden i ica ion o ish la ae can be g ouped in o ou ypes: (1) Mo phological
cha ac e s, such as shape and leng h o he body, head and gu and ela i e posi ion o he ins. These
cha ac e s we e used o di e en ia ion be ween amilies and be ween gene a. (2) Me is ic cha ac e s,
such as numbe o spines and in ays o he numbe o myome es. Fin elemen s de elop g adually du ing
la al de elopmen , hus s a e o de elopmen mus be conside ed. The numbe o myome es co espond
o he numbe o e eb ae in he adul s age. This cha ac e was especially use ul o dis inguish some
e y simila species, such as Sa dina pilcha dus and Sp a us sp a us. (3) Pigmen a ion pa e n as
axonomic cha ac e on la ae is limi ed o melanopho es, since in o malin- ixed la ae, melanopho es
1. In oduc ion
19
a e he only pigmen s ha emain a e ixa ion. Melanopho es a e cha ac e is ic o each species and we e
used o di e en ia e be ween species o he same amily o genus, o example T achu us achu us om
T achu us medi e aneus o Se anus hepa us om Se anus cab illa. (4) Specialised la al cha ac e s
a e hose ha de elop exclusi ely du ing he la al s age and los in he ju enile s age (Fig. 1.7). These
include spines and c anial a mou head spines, used o he dis inc ion, o example o Ca anx honchus
om T. medi e aneus la ae. O he specialised cha ac e s, such as ba bel, cha ac e is ic o Mullus spp.,
and peduncula ed eyes o some myc ophid species we e also use ul in he dis inc ion o di e en species.
Fig. 1.7. Ana omical,
mo phological and some
specialised cha ac e s o pos -
lexion la ae used o he la al
iden i ica ion (F om Mille and
Kendall, 2009)
1.3.3. S a is ical analysis o ish la ae da a
The numbe o la ae o each species o each sampling s a ion was s anda dized o numbe pe 10 m2 o
he sea su ace and ln(x + 1) ans o med o educe he weigh o he dominan species. These ans o med
la al abundances we e used o pe o m s a is ical analysis and o d aw maps o ob ain he geog aphical
dis ibu ions o he la ae o he di e en species using SURFER 13 (Golden So wa e, 2015).
Mul i a ia e s a is ical analyses we e pe o med using he R s a is ical so wa e (R Co e Team, 2023) o
s udy he la al ish communi y. The aim was o iden i y pa e ns in he dis ibu ion o species by
iden i ying g oups o objec s (sampling s a ions o su eys) wi h simila cha ac e is ics in e ms o
speci ic composi ion and abundance o la ae o each species. Fo his pu pose, clus e ing analyses, such
as he Wa d's (1963) minimum a iance me hod and o dina ion echniques, in his case non-me ic
mul idimensional scaling (nMDS), we e used. To in es iga e he in luence o en i onmen al a iables on
he speci ic composi ion, abundance and dis ibu ion o ish la ae canonical o dina ion me hods such as
canonical co espondence analysis (CCA) and edundancy analysis (RDA) we e used.
2.1. Va iabili y unde con as ing en i onmen al condi ions
26
2.1.1. In oduc ion
Spa ial pa e ns in he la al ish assemblages (LFA), i. e. occu ence, dis ibu ion, and abundance o ish
la ae, a e in luenced by a complex a ay o en i onmen al p ocesses ha in e ac s wi h he biology o
ish a di e en empo al and spa ial scales (Doyle e al., 1993). These p ocesses include hose o la ge
scale, as clima e pa e ns (Di Pane e al., 2020; Guan e al., 2015), as well as en i onmen al a iabili y a
seasonal and in e annual ime scales, ha de e mine adul s dis ibu ion and hei spawning s a egies
(Au h, 2008; Doyle e al., 2009). A a local and sho ime scale, LFA will be shaped by he
hyd odynamics (e.g. ci cula ion, on s, ins abili ies) ha in luence ish la al dispe sal and e en ion
(Mose and Smi h, 1993; Su he s e al., 2023) and by biologic ac o s, such as ood concen a ion and
p eda ion (Houde, 2008; Mille and Kendall, 2009), ha ul ima ely de e mine ish ec ui men (Cushing,
1990; Houde, 2008). F om a global app oach, changes in ich hyoplank on assemblages p o ide ea ly
indica o s o shi s in ish communi y dynamics (Doyle e al., 2009) and oceanog aphic and clima ic
condi ions (Au h e al., 2011; Ciannelli e al., 2022; McCla chie e al., 2018).
The Medi e anean is a ela i ely small, semi-enclosed sea be ween con inen s. As a esul , i is e y
sensi i e, and esponds apidly o a mosphe ic o cing and an h opogenic impac s, as compa ed o oceanic
ime scales (Gio gi, 2006). In his con ex , i can be conside ed as an excellen si e o s udy he e ec s o
clima e a iabili y on ma ine communi ies. The basin is in he empe a e zone o he No he n
Hemisphe e wi hin a ela i ely na ow la i udinal ange (30–45ºN). This in ol es a ma ked seasonal
cycle, wi h he al e na ion o s a i ied and mixing pe iods ha esul s in s ong seasonali y in p ima y
p oduc ion, wi h i s annual maximum in la e win e /ea ly sp ing and minimum in la e summe (Ribe a
d’Alcalà e al., 2004). Recu en la e win e /ea ly sp ing phy oplank on blooms a e only egula ly
obse ed in he no hwes e n egion, and in e mi en ly in a ew o he a eas (D’O enzio and Ribe a
D’Alcalà, 2009). The seasonal bloom in he NW Medi e anean is igge ed by deep con ec ion episodes
ha occu o sho e in he Gul o Lion in win e . These episodes a e d i en by e apo a ion caused by
s ong, cold and d y no he ly winds (MEDOC GROUP, 1970; Scho e al., 1996). Occasionally, in ense
cooling and e apo a ion ex end o e he shel wa e s, which become dense han hose o sho e, and
cascade down-slope o he dep h whe e he same densi y is ound (Fieux, 1974; Sala and Fon , 1987). All
hese win e con ec ion episodes, esponsible o he deep-wa e o ma ion, a e he majo con ibu o s o
o e all p ima y p oduc ion in he whole wes e n Medi e anean. In e annual a iabili y in p ima y
p oduc ion is highly dependen on he ex ension, in ensi y and du a ion o deep-wa e o ma ion episodes,
which inc ease in colde and d ie yea s (La asa e al., 2022; Ma y and Chia é ini, 2010). O e he
shel es and uppe slope, whe e he bo om is shallowe , con ec ion can sus ain a high p oduc i i y du ing
win e (Sala e al., 2002).
The s udy a ea, o he Ca alan coas , is loca ed in he NW Medi e anean, sou h o he Gul o Lion, one
o he coldes a eas in he Medi e anean (Theocha is and Geo gopoulos, 1993). The con inen al shel is
gene ally na ow ( ypically less han 20 km wide), u owed by a se ies o deep canyons whose heads a e

2.1. Va iabili y unde con as ing en i onmen al condi ions
27
no a om he coas line. The oceanog aphic dynamics is la gely con olled by he No he n Cu en
(NC), a slope cu en associa ed o a shel -slope densi y on , de ec able down o abou 400 m dep h,
which sepa a es he ela i ely low salini y shel wa e s om he mo e saline open sea wa e s (Fon e al.,
1988). The NC displays high mesoscale a iabili y ha causes oscilla ions, meande ing and eddy
gene a ion (Rubio e al., 2005). The NC may in e ac wi h he canyons, s ongly modi ying he local
ci cula ion and gene a ing shel -slope exchanges (Flexas e al., 2008) ha play a key ole in he
dis ibu ion and abundance o plank onic o ganisms (Gue e o e al., 2016; Saba és e al., 2004).
In he NW Medi e anean, he p esence o ish la ae in he plank on shows a ma ked seasonali y ela ed
o he spawning cycles o he adul s (Palome a and Oli a , 1996; Saba és e al., 2007). Mos ish species
spawn du ing sp ing and summe (Tsikli as e al., 2010), when oligo ophy p e ails in he uppe wa e
column. In win e , a pe iod o high p ima y p oduc i i y and maximum phy oplank on biomass (Mo án
and Es ada, 2005; Ribe a d’Alcalà e al., 2004), ew ish species ep oduce (Oli a e al., 2003b). These
a e species cha ac e is ic o ela i ely cold wa e s (o No h A lan ic biogeog aphic a ini y), which a e o
impo an comme cial in e es , including he Eu opean sa dine, Sa dina pilcha dus, he A lan ic
macke el, Scombe scomb us, he Eu opean seabass, Dicen a chus lab ax and a ious species o
Gadi o mes (hake, Me luccius me luccius and blue whi ing, Mic omessis ius pou assou) and
Pleu onec i o mes, such as he ou -spo meg im, Lepido hombus boscii (Saba és e al., 2007).
In he Ca alan coas , mos ich hyoplank on s udies, conduc ed in he sp ing-summe mon hs, showed ha
LFA basically esponded o he ba hyme y, including he subma ine canyons, he p esence o con inen al
ou lows, and he ci cula ion pa e ns, such as he an icyclonic eddies and he shel -slope densi y on
(Ál a ez e al., 2015; Saba és and Oli a , 1996). Howe e , he in o ma ion a ailable on his subjec o
he win e pe iod is sca ce and es ic ed o a ew species (sa dine, blue whi ing and mesopelagic ish)
(Oli a e al., 2003b; Saba és e al., 2024).
The impac o clima e change on he s uc u e o ich hyoplank on assemblages has been s udied
wo ldwide (Au h e al., 2018; Hsieh e al., 2009). In he global wa ming con ex , in he Medi e anean
(Sala e al., 2019; Va gas-Yáñez e al., 2017), cold-wa e ish species could be ad e sely a ec ed (Llo e
e al., 2015), and i has been epo ed ha su i al o ea ly li e s ages o hese species is lowe in mild
win e s han in se e e win e s (Mi -A guimbau e al., 2022; Moyano e al., 2023).
The p esen s udy p o ides he i s cha ac e isa ion o LFA in wo consecu i e win e s o con as ing
en i onmen al condi ions in he NW Medi e anean Sea. Ou s udy a ea was he no he n Ca alan coas ,
sou h o he Gul o Lion, one o he coldes a eas in he Medi e anean, whe e deep con ec ion akes
place in win e , and cold wa e ish species, he mos h ea ened by he clima e change, a e pa icula ly
abundan . In his con ex , he objec i es a e: (a) o desc ibe he species composi ion, abundance and
di e si y o he ish la al communi y in win e , (b) o analyse he in luence o hyd odynamic p ocesses
2.1. Va iabili y unde con as ing en i onmen al condi ions
28
on he s uc u e o he LFA, and (c) o asce ain he ex en o which in e annual shi s in he LFA a e
indica o s o changes in en i onmen al condi ions.
2.1.2. Ma e ials and me hods
Sample collec ion and da a p ocessing
Two oceanog aphic su eys we e conduc ed in wo consecu i e win e s ( om 18 Feb ua y o 3 Ma ch
2017 and om 14 o 26 Feb ua y 2018) in he Ca alan Sea (NW Medi e anean Sea; Fig. 2.1.1) on boa d
he R/V Ga cía del Cid. Sampling s a ions, 33 in 2017 and 29 in 2018, we e loca ed app oxima ely 7−14
km apa , and placed on ansec s pe pendicula o he sho eline, om nea he coas o he slope (Fig.
2.1.1). A each s a ion, e ical p o iles o basic hyd og aphic a iables ( empe a u e, salini y and
luo escence) we e acqui ed wi h a conduc i i y- empe a u e-dep h (CTD) p o ile equipped wi h a
luo ome e . In o de o calib a e he luo ome e , wa e samples o chlo ophyll-a de e mina ion we e
collec ed a selec ed s a ions, e enly dis ibu ed o e he a ea, wi h a ose e sys em a di e en dep h
le els down o 70 m, h oughou day and nigh . To ge geos ophic ci cula ion pa e ns, as well as
e ically in eg a ed in o ma ion on he dynamic s uc u es, dynamic heigh s (Asch and Checkley, 2013)
we e calcula ed wi h a e e ence dep h le el o 500 m. Whe e s a ion dep h was lowe , dynamic heigh s
we e ex apola ed using he con inui y equa ion applied o he deepes le el o h ee-s a ion clus e s (c .
Hidaka, 1940).
Fig. 2.1.1. S udy a ea in he NW Medi e anean, sou h o he
Gul o Lion. C osses indica e sampling s a ions (CTD and
Bongo ne ) du ing 2017 (o ange) and 2018 (da k blue)
su eys. T iangles indica e s a ions whe e only CTD cas s
we e pe o med. O ange ci cles co espond o MOCNESS ne
samplings. S a s co espond o dep h-s a i ied sampling a
ixed s a ions in 2017 (o ange) and 2018 (blue) su eys. Open
ellipses (g een, Coas , Shel , Slope and Palamós canyon)
indica e he e ical p o iles o oceanog aphic a iables shown
in igu e 4. The isoba hs shown a e 50, 200, 1000 and 2000 m.
2.1. Va iabili y unde con as ing en i onmen al condi ions
29
Chlo ophyll-a concen a ion (µg l–1) was de e mined luo ome ically (Yen sch and Menzel, 1963). Wa e
samples o 100 o 200 ml we e il e ed h ough Wha man GF/F il e s. Chlo ophyll-a was ex ac ed om
il e s imme sed in 6 ml o 90% ace one (24 h a 4°C in da kness). The ex ac was analysed using a
Tu ne Designs luo ome e calib a ed wi h pu e chlo ophyll-a. The ela ionship be ween chlo ophyll-a
concen a ion and luo escence ob ained on each c uise was used o con e he con inuous CTD
luo escence egis e o chlo ophyll-a concen a ion.
Fish la ae we e collec ed a each s a ion using a Bongo ne wi h a mou h opening o 60 cm diame e and
a mesh size o 300 μm. The ne was owed obliquely om a maximum dep h o 500 m o he su ace, a a
essel speed o 2 kno s. In he 2017 su ey, some o he s a ions o e he slope we e sampled using a
MOCNESS ne o 1 m2 mou h opening and 300 μm mesh size o pe o m dep h-s a i ied zooplank on
samplings (Fig. 2.1.1). In addi ion, o analyse he day-nigh e ical dis ibu ion o la ae, dep h-s a i ied
samples we e ob ained a se e al ixed s a ions, one in 2017 and h ee in 2018, du ing 24 - 48 h, a oiding
sunse and sun ise hou s (see Fig. 2.1.1). As he ne enabled sampling 8 dep h s a a, hese we e de ined
acco ding o he bo om dep h. The MOCNESS ne was deployed o collec he samples obliquely,
mo ing om deep o shallow laye s, a a ship speed o 2 - 2.5 kno s. The olume o il e ed wa e by he
di e en ne s was es ima ed by means o calib a ed lowme e s. All zooplank on samples we e
immedia ely ixed in 5% o maldehyde bu e ed wi h sodium e abo a e. In he labo a o y, all ish la ae
we e so ed and iden i ied o he lowes possible axonomic le el based on exis ing desc ip i e li e a u e
(Lo Bianco e al., 1956). The numbe o la ae collec ed wi h he Bongo ne was s anda dized o numbe
pe 10 m2 o he sea su ace. The la ae collec ed by he MOCNESS ne a each dep h s a um we e
s anda dized o 1000 m3 o il e ed wa e .
S a is ical analyses and spa ial pa e ns o species
Fo each species and su ey, he mean abundance alues, equency o occu ence (FO, pe cen age o
s a ions whe e a axon occu ed) and he ela i e abundance (RA, pe cen age con ibu ion o a axon o
he o al abundance o indi iduals) we e calcula ed. To de ec signi ican di e ences in en i onmen al
a iables (SST, SSS, SSSig, SSChla and DH) be ween yea s, - es s we e pe o med. The h eshold o
s a is ical signi icance o all analyses was se a α = 0.05.
To analyse he spa ial and empo al pa e ns in LFA, mul i a ia e analyses we e pe o med using he
“clus e ” (Maechle e al., 2022) and “ egan” (Oksanen e al., 2022) packages in he s a is ical so wa e R
. 4.2.1 (R Co e Team, 2022). Da a o ish la al abundance we e ln(x + 1) ans o med o educe he
weigh o he dominan species, and axa wi h FO < 5% we e no included in he analyses. To assess he
exis ence o g oups o s a ions wi h simila axa composi ion and abundance, a non-me ic
mul idimensional scaling (nMDS) (Cla ke, 1993), and a clus e analysis we e pe o med based on he
B ay-Cu is dissimila i y index (B ay and Cu is, 1957). The dend og am was cons uc ed using he
Wa d’s hie a chical agglome a ion me hod (Wa d, 1963). The g oups o s a ions we e plo ed on a map o
2.1. Va iabili y unde con as ing en i onmen al condi ions
30
he sampled egion o iden i y geog aphical pa e ns i hey exis . Pe mu a ional mul i a ia e analysis o
a iance (Adonis es , 999 pe mu a ions) (Ande son, 2001) and pai wise es s we e used o es o
signi ican di e ences in ish la al abundance and composi ion be ween yea s and be ween clus e s. To
quan i y he con ibu ion o he species o he dissimila i y be ween yea s and clus e s, simila i y
pe cen age ou ine (SIMPER) was pe o med (Cla ke, 1993).
To s udy he in luence o en i onmen al condi ions on he ish la al communi y canonical
co espondence analysis (CCA) was used. The en i onmen al da a ma ix included: he oceanog aphic
a iables a he su ace (10 m), empe a u e (SST), salini y (SSS), densi y (SSSig) and chlo ophyll-a
(SSChla); mean chlo ophyll-a in he uppe 50 m o he wa e column (MChla_50); dynamic heigh (DH);
bo om dep h (log- ans o med; Dep h); dis ance om a sampling s a ion o he coas (Dis _Coas ). This
las a iable was conside ed o compensa e he high bo om dep hs eached close o he coas in he
subma ine canyons. DH was included because i can be used as an indica o ha in eg a es he e ec s o
mul iple oceanic p ocesses (Asch and Checkley, 2013). I is a p oxy o he in eg a ed empe a u e and
salini y o he uppe wa e column. Thus, a eas wi h esh, wa m seawa e co espond o highe dynamic
heigh s han a eas wi h sal y, cool seawa e , which ha e lowe dynamic heigh s. Mo eo e , DH can be
used o iden i y hyd odynamic s uc u es, such as eddies and meande s (Asch and Checkley, 2013). As
he DH alues in he Medi e anean a e nega i e, hey we e ans o med in o posi i e alues in he
analysis o con enience by adding a cons an . All hese ac o s in luence he spawning habi a o ishes.
Collinea i y be ween pai s o a iables was assessed using Pea son’s co ela ion coe icien s (cu -o
alue |0.6|). As a esul , SSSig and MChla_50 we e disca ded. The CCA analysis was pe o med using
he o wa d selec ion me hod o ank he a iables in hei o de o impo ance and selec only hose ha
con ibu ed signi ican ly o explaining he a iance.
All maps we e d awn using SURFER 13 (Golden So wa e, 2015). Con ou maps o en i onmen al
a iables we e ob ained using he minimum cu a u e g idding me hod. Con ou maps o he CCA sco es
we e ob ained using he k iging in e pola ion me hod, in oducing he a iog am i ed by he leas
squa es c i e ion p e iously calcula ed.
To isualise he a ia ion in la al abundance wi h dep h, iolin plo s we e pe o med using a Gaussian
densi y unc ion. Weigh ed mean dep h (WMD) o each species was calcula ed as he mean dep h
sampled by each ne (MOCNESS da a, s anda dised o 1000m3) weigh ed by he p opo ion o la ae
caugh in he ne (Hawes e al., 2020).
2.1. Va iabili y unde con as ing en i onmen al condi ions
31
Fig. 2.1.2. Spa ial
dis ibu ion o su ace
(10 m) empe a u e
(SST, a and b), salini y
(SSS, c and d),
chlo ophyll-a (SSChla,
e and ) and dynamic
heigh ela i e o 500 m
(DH) o e laid on
densi y (, g and h) in
win e 2017 (le
panels) and 2018 ( igh
panels). A ows
indica e he low
di ec ion and in usions
o he No he n Cu en
and he ex usions o
coas al wa e s. The
isoba hs shown a e 50,
200, 1000 and 2000 m.

2.1. Va iabili y unde con as ing en i onmen al condi ions
32
2.1.3. Resul s
Hyd og aphic condi ions
The mos s iking ea u e ha appea ed when compa ing SST and, in pa icula , SSS dis ibu ions
be ween yea s was he p esence o a wa e mass o low empe a u e (< 12.8 ºC) and salini y (< 37.5)
along he coas in win e 2017. This band o coas al wa e s co e ed a signi ican pa o he con inen al
shel (Fig. 2.1.2 a and c). SST and SSS inc eased owa ds he open sea, wi h alues ~13.6 ºC and ~38.4,
espec i ely, o sho e. In 2018, SST and SSS alues showed a smoo he ansi ion be ween coas al and
o sho e wa e s (Fig. 2.1.2 b and d). SST was sligh ly lowe han in 2017, bo h nea he coas , ~12.6 ºC,
and o sho e (~13.4 ºC; Fig. 2.1.2 a and b). The ange o SSS was much lowe han in 2017, wi h ~38.1
nea he coas and ~38.5 o sho e (Fig. 2.1.2 c and d). O e all, SST was no s a is ically di e en be ween
yea s (p = 0.9802), bu signi ican di e ences we e ound in SSS (p < 0.05).
SSSig ollowed he same pa e n as SSS, wi h he highes alues o e he slope (Fig. 2.1.2 e and ). In
2017, a ma ked g adien ela ed o he low salini y coas al wa e mass was e iden o e he shel (Fig.
2.1.2 e). As obse ed in SSS, signi ican di e ences in SSSig we e de ec ed be ween yea s (p < 0.05).
The dynamic heigh o e laid on densi y clea ly showed he p esence o he No he n Cu en , associa ed
wi h he shel -slope densi y on , lowing sou hwes wa d along he con inen al slope wi h signi ican ly
highe DH alues in 2017 han in 2018 (p < 0.05). In bo h yea s, an an icyclonic eddy was de ec ed o e
he Palamós Canyon, mo e ma ked in 2018 (Fig. 2.1.2 e and ). Two in usions o o sho e wa e s o he
coas we e de ec ed, one in he no h o he a ea (~ 42.2°N) and he o he ollowing he sou he n side o
Fig 2.1.3.
Ve ical p o iles o densi y
(), mean (ci cle) and s anda d
de ia ion (ho izon al lines) o CTD
s a ions sampled in 2017 and 2017
2.1. Va iabili y unde con as ing en i onmen al condi ions
33
he an icyclonic eddy, o e he Palamós canyon. These in usions we e also e iden in he SST and SSS
dis ibu ions (Fig. 2.1.2 a, b, c and d), al hough in 2017 hey we e masked by he p esence o he low
salini y coas al wa e s. In he Blanes canyon in 2017, an in usion o o sho e wa e s owa ds he coas
was de ec ed (Fig. 2.1.2 e), while in 2018 he e was an o sho e sp eading o coas al wa e s owa ds he
sou heas h ough he canyon, as also shown by he low empe a u e and salini y alues (Fig. 2.1.2 b and
d).
High chlo ophyll-a concen a ions we e associa ed o he low salini y su ace laye in 2017, al hough
ela i ely high su ace chlo ophyll-a pa ches (> 1 µg l-1) we e also ound beyond he low salini y coas al
band (Fig. 2.1.2 g). In 2018, su ace chlo ophyll-a was, in gene al sligh ly lowe han in 2017, excep in
he icini y o he Blanes canyon whe e high alues (nea ly 2 µg l-1) ex ended o sho e, in ela ion o he
o sho e sp eading o coas al wa e s (Fig. 2.1.2 h). No signi ican di e ences we e obse ed in
chlo ophyll-a be ween yea s ( - es , p = 0.4742).
In gene al, he wa e column p esen ed sligh s a i ica ion in 2017, due o he coas al wa e mass o low
salini y, while i was mo e homogeneous in 2018 (Fig. 2.1.3). The e ical p o iles o hyd og aphic
pa ame e s o he di e en s a ions ep esen a i e o each zone (coas , shel , slope and he Palamós
canyon; see igu e 2.1.1 o s a ion loca ions) showed clea di e ences be ween bo h yea s (Fig. 2.1.4). In
2017, he coas al wa e mass o low empe a u e and salini y was de ec ed a su ace o 20 m dep h,
wi h a clea pycnocline, while on he es o he shel he e ical densi y g adien was weake (Fig. 2.1.4
a and c). O e he slope, he empe a u e p o ile showed a laye o lowe alues be ween 70 and 250 m,
sugges ing subduc ion o shel and uppe slope wa e s (Fig. 2.1.4 e). In he Palamós Canyon, he
signa u e o coas al wa e s sinking was de ec ed be ween 100 and 350 m, as shown by he lowe
empe a u e and salini y a his dep h in e al (Fig. 2.1.4 g). In 2018 he wa e column was mo e
homogeneous e e ywhe e (coas , shel and slope) and he empe a u e, salini y and densi y p o iles
showed a sligh inc ease wi h dep h (Fig. 2.1.4 b, d and ). The an icyclonic eddy iden i ied in he
ho izon al dis ibu ions o e he Palamós Canyon could be obse ed down o a dep h o 100 m wi h a
clea pycnocline (Fig. 2.1.4 h). Below his dep h, aces o sinking o cascading e en s we e de ec ed a
400 m and below 1000 m, as shown by he empe a u e and salini y in e sions, much deepe han in he
p e ious yea (Fig. 2.1.4 h). The e ical dis ibu ion o chlo ophyll-a p esen ed di e ences be ween
yea s in ela ion o he s a i ica ion o he uppe laye , al hough wi hin each yea he pa e ns we e
simila ac oss zones. In 2017 high chlo ophyll-a alues we e de ec ed in su ace wa e s down o 40 m
dep h, mainly in he coas , while in 2018, he chlo ophyll-a dis ibu ion was mo e homogeneous down o
~150 m (Fig. 2.1.4).
2.1. Va iabili y unde con as ing en i onmen al condi ions
34
Fig. 2.1.4. Ve ical p o iles o empe a u e ( ed), salini y (blue), densi y (, black), and chlo ophyll-a (g een), o ou selec ed
s a ions o 2017 and 2018, a di e en loca ions: coas (a and b), shel (c and d), slope (e and ) and he Palamós Canyon (g and
h) ( o s a ion loca ions see igu e 2.1.1).
2.1. Va iabili y unde con as ing en i onmen al condi ions
35
Table 2.1.1. Mean abundance and s anda d de ia ion (No. la ae 10m-2), ela i e abundance (RA, in %) and equency o
occu ence (FO, in %) o all iden i ied la al ish axa in 2017 and 2018. The axa a e lis ed acco ding o adul habi a (shel and
oceanic).
2.1. Va iabili y unde con as ing en i onmen al condi ions
42
Fig. 2.1.9. Spa ial dis ibu ion o ish la ae o some ep esen a i e oceanic species o e laid on sea su ace empe a u e (SST)
and ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in 2017 and 2018.
Fish la al e ical dis ibu ion
Mos o he axa showed a deepe la al dis ibu ion in 2018 han in 2017 (Table 2.1.4). La ae o oceanic
species, inhabi ing o e he slope and subma ine canyons, showed a deep and ela i ely wide e ical
dis ibu ion in he wa e column, being shallowe du ing he day han a nigh (Table 2.1.4, Fig 2.1.11 a
and b). Fo ins ance, la ae o A. isso we e loca ed in he i s ~120 m du ing he day and deepe a nigh ,
be ween 40 m and 350 m (Fig. 2.1.11 a). The e ical dis ibu ion o M. punc a um la ae ex ended in o
he i s 100 m o he wa e column, excep du ing he nigh in 2018, which we e ound down o ~250 m
(Fig. 2.1.11 b). The e ical dis ibu ion o la ae o shel species was shallowe han ha o oceanic
species and deepe in 2018 han in 2017 (Table 2.1.4). They we e close o he su ace du ing he day han
a nigh . Fo ins ance, he e ical dis ibu ion o S. pilcha dus la ae ex ended o he i s 100 m, being
pa icula ly abundan in he uppe 40 m du ing he day. A nigh in 2018, hei dis ibu ion was deepe
eaching ~250 m dep h (Fig. 2.1.11 c). As obse ed o he shel species, he e ical dis ibu ion o la ae
o species ha inhabi s he shel b eak and uppe slope, e.g. M. pou assou, showed a ela i ely supe icial
dis ibu ion du ing he day and deepe a nigh (Table 2.1.4). La ae o hese species we e loca ed in he
i s 100 m du ing he day, wi h he highes abundances a ~50 m. A nigh hey showed a wide e ical
dis ibu ion, ex ending o ~400 m in 2017 (Fig. 2.1.11 d).

2.1. Va iabili y unde con as ing en i onmen al condi ions
43
Fig. 2.1.10. Spa ial dis ibu ion o ish la ae o some ep esen a i e shel and uppe slope species o e laid on sea su ace
empe a u e (SST) and ba hyme y (isoba hs shown: 50, 200, 1000 and 2000 m) in 2017 and 2018. No e ha S. pilcha dus and
A noglossus spp. a e ep esen ed on a di e en size scale o M. pou assou and G. mac oph halmus.
2.1.4. Discussion
En i onmen al se ing
The main oceanog aphic condi ions ound du ing he s udy we e hose expec ed o he season, wi h a
coas al-o sho e g adien o SST and SSS and he p esence o he No he n Cu en associa ed wi h he
shel -slope densi y on along he slope (Fon e al., 1988). Howe e , in 2017 he coas al zone was
occupied by cold and low-saline wa e s a he su ace, due o he occu ence o hea y ain episodes a ew
days be o e he c uise (Saba és e al., 2024). Al hough he sea su ace empe a u es de ec ed in bo h
c uises we e qui e simila , win e 2017 was mild, wi h a sho pe iod o cold empe a u es and limi ed
e ical mixing (Mi -A guimbau e al., 2022). The s a i ica ion de ec ed in he coas al a ea was due o he
low salini y su ace wa e s (Fig. 2.1.4 a), bu also o sho e, whe e subduc ion o coas al wa e s in o mid-
dep h laye s was obse ed, as e idenced by hei lowe empe a u e and salini y alues (Fig 2.1.4 e and
g). The s a i ica ion and mild win e condi ions ha yea would es ic e ical mo ions and hinde deep
con ec ion (Ma gi ie e al., 2020; Pa as-Be ocal e al., 2022). In con as , win e 2018 was mo e
se e e, empe a u es eached lowe alues and pe sis ed o a longe pe iod han in 2017 (Mi -A guimbau
e al., 2022). These condi ions would a ou he mixing o he wa e column and dense wa e o ma ion
2.1. Va iabili y unde con as ing en i onmen al condi ions
44
wi h subduc ion and cascading e en s down he slope o deepe laye s, as shown by he empe a u e and
salini y in e sions de ec ed a ~400 m and below 1000 m in he Palamós Canyon (Fig. 2.1.4 h). In he
Gul o Lion, he in ense con ec ion episodes epo ed ha yea eached a leas 1900 m (Fou ie e al.,
2022; Ma gi ie e al., 2020). Such p ocesses o deep-wa e o ma ion, con ec ion and cascading, a e
pa icula ly impo an , as hey a e esponsible o shel -slope wa e exchanges in he egion (I ano e al.,
2004). Thus, he condi ions ound in 2018 would enhance he shel -slope exchanges while he
hyd odynamic condi ions o 2017 would es ic hese p ocesses. The dis inc si ua ions obse ed be ween
he wo yea s would also be e lec ed in he di e ences in DH alues, highe in 2017 han in 2018, as DH
e ically in eg a es he e ec s o mul iple oceanic p ocesses (Asch and Checkley, 2013).
Table 2.1.4. Weigh ed mean dep h (WMD, in m) and s anda d de ia ion, du ing day and nigh , o he mos abundan la al ish
axa, in 2017 and 2018.
2.1. Va iabili y unde con as ing en i onmen al condi ions
45
Fig. 2.1.11. Violin plo showing he e ical
dis ibu ion o la ae o A c ozenus isso,
Myc ophum punc a um, Sa dina pilcha dus and
Mic omessis ius pou assou, du ing day and nigh ,
o 2017 and 2018. The wid h o he iolin
ep esen s he p opo ion o la ae a each dep h
s a um.
En i onmen al in luence on he s uc u e o la al ish assemblages
In bo h yea s, he alues o he species ichness we e compa able o hose usually ound du ing he
mixing pe iod in he Medi e anean (Kou akis e al., 2004; Oli a e al., 2010, 2014; Za ad e al., 2020).
Howe e , hese alues we e lowe han hose ound in sp ing-summe (e.g. Ál a ez e al., 2015; Saba és
e al., 2007), since mos Medi e anean ish species ep oduce du ing he sp ing–summe pe iod (Tsikli as
e al., 2010). The p edominance o clupeid la ae, S. pilcha dus and S. sp a us, ound in he wo s udied
yea s (Table 2.1.1) is a common ea u e o win e ich hyoplank on in he Medi e anean (Kou akis e al.,
2004; Oli a e al., 2003a; Saba és e al., 2004) and in o he a eas a ound he wo ld (Di Pane e al., 2020;
Moyano and He nández-León, 2009; Rod iguez, 2008). These species, and la ae o shel species in
gene al, we e mo e abundan in 2017 (~70% o he o al la al abundance) in he coas al band o low
salini y and high chlo ophyll-a wa e s.
In 2018, oceanic ish la ae we e mo e dispe sed h oughou he s udy a ea in ela ion o he shel -slope
exchange p ocesses o ha yea . The alues o ela i e abundances o oceanic species we e simila o
hose epo ed o many LFA du ing pe iods o e ical mixing in he wes e n Medi e anean (Oli a e
2.1. Va iabili y unde con as ing en i onmen al condi ions
46
al., 2014), as well as in o he geog aphical a eas (Moyano and He nández-León, 2009; Oso io-Zúñiga e
al., 2018; Sassa and Konishi, 2015). The high la al concen a ions o bo h shel and oceanic species
de ec ed in he Blanes Canyon would suppo he ole o subma ine canyons as a condui o la ae om
deep-wa e spawning si es o coas al g ounds, highligh ing hei c ucial ole in shaping LFA (Mo dy e
al., 2019).
Clus e analysis sepa a ed he LFA in o wo b anches (Fig. 2.1.6 a), wi h a g ada ion om la ae o shel
o oceanic ish species. In 2017, he coas al and o sho e assemblages we e well delimi ed, wi h he
coas al g oup 6, associa ed o he low salini y and high SSChla wa e s, bounded by he ma ked densi y
g adien ha would e ain hese la ae in he p oduc i e coas al wa e s. The assemblage o oceanic
species was loca ed o sho e o his densi y g adien and clea ly sepa a ed om he cos al g oup. This
oceanic assemblage appea ed on he inne pa o he shel -slope on and he associa ed No he n
Cu en . The p esence o la ae o oceanic ish species in he inne pa o he shel -slope on has been
obse ed in p e ious occasions in he a ea and has been ela ed wi h he posi ion o he on a om he
coas (Masó e al., 1998; Saba és and Oli a , 1996), in ou s udy o e he 1000 m isoba h. In 2018, he
assemblages we e no so geog aphically delimi ed. In addi ion, he absence o a s ic ly coas al g oup and
he occu ence o ansi ional assemblages (3, 4 and 5) would e idence he ole o he shel -slope wa e
exchange p ocesses in he anspo o ish la ae ac oss he shel his yea .
The CCA esul s showed he ela ionship o oceanic clus e s (1 and 2) wi h high SST alues,
cha ac e is ic o o sho e wa e s. This ela ionship is con i med by he geog aphical ep esen a ion o he
sco es and he spa ial dis ibu ion o he oceanic ish axa (Fig. 2.1.8). The ela ionship o clus e 1
(mainly in 2017) and clus e 2 (2018) wi h high and low DH alues, espec i ely, e lec s he dis inc
hyd odynamic condi ions ound each yea , wi h impo an shel -slope wa e exchange p ocesses in 2018.
In ha yea , he high abundances o la ae o oceanic species and hei wide dis ibu ion all o e he a ea,
including he shel s a ions, e ealed hei anspo om o sho e a eas owa ds he shel (Fig. 2.1.5),
whe e e ical mixing can sus ain a high p oduc i i y (Sala e al., 2002). I should be no ed ha la ae o
oceanic species wi h a deepe e ical dis ibu ion, such as A. isso and A. hemigymnus, did no unde go
his coas al anspo as was obse ed o M. muelle i and M. punc a um (Fig. 2.1.9). These di e ences in
he anspo o oceanic ish la ae in ela ion o hei e ical dis ibu ion dep h we e p e iously epo ed
(Oli a e al., 2010; Va gas-Yáñez and Saba és, 2007). The cascading and e ical mixing p ocesses
de ec ed in 2018 would dispe se la ae in he wa e column, which would esul in a deepe e ical
dis ibu ion o la ae han in 2017, while a ou ing an homogeneous e ical dis ibu ion o chlo ophyll-a
in he i s 150 m o he wa e column (Fig. 2.1.4 and h). These s ong e ical mixing p ocesses ha e
been epo ed o lead o high p ima y p oduc i i y and subsequen zooplank on abundance (Es ada e al.,
2014; Vand omme e al., 2011). In addi ion o he shel -slope exchanges induced by hese hyd odynamic
condi ions in 2018, he in usions o he No he n Cu en de ec ed in bo h yea s would also a ou he
anspo o hese la ae owa ds he coas , as p e iously epo ed in he a ea (Masó e al., 1998; Saba és
2.1. Va iabili y unde con as ing en i onmen al condi ions
47
e al., 2018). The p esence o la ae o oceanic species o e he shel , associa ed wi h cu en ins abili ies,
is a common p ocess in di e en geog aphical egions (Cu i a e al., 2018; Go oni and Spach, 1999;
Sassa and Konishi, 2015).
Conce ning la ae o shel species, his s udy e idences he associa ion o he coas al g oup, 6, wi h he
band o low SSS, SST and high SSChla alues ound in 2017. This s a i ied p oduc i e a ea would o e
a ou able condi ions o eeding and de elopmen o la ae o hese species. These esul s ag ee wi h he
obse a ions o Ca alán e al., (2006), who ound ha in he mixed season, he dis ibu ion o shel ish
la ae was associa ed wi h he high concen a ions o zooplank onic o ganisms. In ha yea , he shallow
e ical dis ibu ion o hese species, such as S. pilcha dus, would be ela ed o he e ical densi y
g adien ha would es ic hei diel e ical mig a ion o e en hei passi e nigh sinking (Da is e al.,
1990; Oli a e al., 2001). In 2018, he ho izon al dis ibu ion o shel species was wide ac oss he shel
and slope, as hei o sho e dis ibu ion was limi ed by he loca ion o he shel -slope on . This sugges s,
as obse ed in he a ea, ha he on would ac as a ba ie p e en ing he dispe sal o la ae, and o he
plank onic o ganisms, owa ds he open sea (Gue e o e al., 2016; Saba és and Oli a , 1996; Saiz e al.,
2014). In his yea , he e ical dis ibu ion o hese species, such as S. pilcha dus (Fig. 2.1.11), was
deepe han ha ound in o he s udies (John, 1985; Oli a e al., 2001; Saba és, 2004), sugges ing ha he
homogenei y o he wa e column would acili a e he e ical mo emen s o la ae. Mo eo e , he
cascading e en s de ec ed in 2018 may o e whelm he dep h- ela ed beha iou al pa e ns o la ae,
causing hem no o ollow hei usual e ical dis ibu ion pa e ns as obse ed in o he a eas (G ay and
Miskiewicz, 2000).
Among he species ha appea ed in he cen al pa o he CCA, wi hou an ob ious ela ionship wi h any
en i onmen al a iable, i is wo h men ioning M. pou assou and G. mac oph halmus whose adul s
inhabi he shel b eak and he uppe slope (Ba os-Ga cía e al., 2018; Mi -A guimbau e al., 2022).
La ae o M. pou assou showed a wide dis ibu ion o e he shel and slope in bo h yea s, being limi ed
o sho e by he shel -slope on . They showed a wide e ical dis ibu ion in he wa e column, which
ag ees wi h p e ious s udies conduc ed in o he a eas (Ådlands ik e al., 2001; Hillg ube and
Kloppmann, 2000). Small la ae o his species would be loca ed in deep wa e s, whe e he adul s dwell,
and la ae g adually ise owa ds he su ace eaching he illumina ed and mo e p oduc i e su ace laye s
o eeding (Saba és e al., 2024). The wide ho izon al and e ical dis ibu ion o his species would
allow i o ake ad an age o he high su ace p oduc i i y in he coas al band in 2017, as well as he
ela i ely high chlo ophyll-a concen a ions, associa ed wi h he e ical mixing, de ec ed down o abou
150 – 200 m dep h ou side he shel . La ae o G. mac oph halmus we e ound on he oceanic side o he
coas al band in 2017, bu we e absen in hese p oduc i e wa e s, sugges ing ha he su ace densi y
g adien would p e en hei dis ibu ion owa ds he coas . In addi ion, he shallowe e ical dis ibu ion
o his species wi h espec o M. pou assou, pa icula ly du ing he day, would u he limi hei eaching

2.1. Va iabili y unde con as ing en i onmen al condi ions
48
he coas . In 2018, he sca e ed dis ibu ion o la ae o e he whole a ea could be ela ed o he high
hyd odynamics ha yea .
In 2018, he absence o a s ic ly coas al g oup and he occu ence o ansi ional assemblages (3, 4 and
5), would e idence he ole o he shel -slope wa e exchange p ocesses in he anspo o ish la ae
ac oss he shel in ha yea . These p ocesses we e mo e e iden o e he canyons, whe e all g oups we e
de ec ed (Fig. 2.1.6 b). P e ious s udies al eady epo ed ha canyons in e ac wi h he No he n Cu en
and enhance he shel -slope wa e exchanges, a ou ing he anspo o ish la ae and o he plank onic
o ganisms ac oss he shel and inc easing p ima y and seconda y p oduc i i y (Al a ez e al., 1996;
Flexas e al., 2008; Masó e al., 1998). Bo h he high abundance o shel species in he ansi ion g oup 5
(Fig. 2.1.6 a), ela ed o high CCA sco es o axis 1, and he spa ial dis ibu ion o sco es a om he coas
o e he Blanes canyon in 2018 (Fig. 2.1.8 b), sugges ha he ex usion o coas al wa e s mixed wi h
o sho e wa e s, esul ing in a e y p oduc i e a ea (Fig. 2.1.2 h). The Palamós Canyon would be a
pa icula case, whe e a clea deep cascading was de ec ed ha yea , which would channel wa e and
pa icles om coas al a eas o he deep ocean (Canals e al., 2013; Romano e al., 2017). The cascading
e en s would enhance he anspo o oceanic la ae owa ds he coas a he su ace as e lec ed in he
p esence o clus e s 1, 2 and 3 o e he canyon and in he dis ibu ion o nega i e alues o axis 1 sco es
e y close o he coas (Fig. 2.1.8 b). The esponsible o his su ace anspo would be he eddy de ec ed
in he canyon (Saba és e al., 2024, 2013), as occu s in o he a eas (Cu i a e al., 2016).
The di e ences in he s uc u e o he LFA in he wo win e s s udied a e closely linked o he
en i onmen al condi ions obse ed each yea , mainly o he anspo p ocesses associa ed wi h he deep-
wa e o ma ion episodes. Thus, LFA a e good indica o s o he a iabili y o he en i onmen al
p ocesses.
La al ish assemblages unde a clima e change scena io
In he no hwes e n Medi e anean, clima e ends show an inc ease in ai empe a u e along wi h a
dec ease in p ecipi a ion and in he equency o no he ly winds (e.g. Alpe e al., 2002; Jo dà e al.,
2017; Vicen e-Se ano and Rod íguez-Camino, 2017). A sea, an inc easing end in empe a u e and
salini y has been de ec ed in su ace and in e media e wa e s (e.g. Piñei o e al., 2019; Sala e al., 2019;
Va gas-Yáñez e al., 2017). This scena io in ol es an in ensi ica ion o he wa e column s a i ica ion
and, consequen ly, a dec ease in he nu ien injec ion in o he pho ic laye and he p ima y p oduc ion
(Cal o e al., 2011). In he open sea, as s a i ica ion in he uppe laye becomes mo e pe sis en ,
a ou able condi ions o he deep con ec ion become less equen (G ignon e al., 2010) and since
win e 2014 deep-wa e o ma ion episodes did no each he deepes le els o he basin (Ma gi ie e al.,
2020; Sala e al., 2019). In addi ion, long- e m p edic ions poin o a dec ease in he occu ence o deep
con ec ion e en s in u u e clima e scena ios (Josey and Sch oede , 2023; Macias e al., 2018). In his
con ex , he p esen s udy p o ides an ou line o he e ec s ha hese ends may ha e on ish la ae by
2.1. Va iabili y unde con as ing en i onmen al condi ions
49
compa ing wo en i onmen ally con as ing yea s. 2017 could be conside ed as an example o a u u e
mild win e , in which deep con ec ion and he consequen shel -slope exchange p ocesses we e inhibi ed,
while win e 2018 was close o a cu en “ ypical” win e , wi h qui e in ense e ical mixing and dense
wa e cascading ha enhanced he anspo o la ae o oceanic species owa ds he coas . In addi ion,
he deep con ec ion and mixing p ocess esul s in high p ima y p oduc ion and zooplank on abundance
(Es ada e al., 2014; Vand omme e al., 2011) ha would enhance he su i al o ish la ae (Mi -
A guimbau e al., 2022).
In win e 2017, s a i ied condi ions and limi ed e ical mixing would be expec ed o dec ease p ima y
p oduc ion in he su ace laye s, sugges ing ha en i onmen al condi ions we e no so sui able o la al
de elopmen . The high abundance o la ae o shel ish species (i.e S. pilcha dus and A noglossus sp.),
and M. pou assou, ound in he coas al band would appa en ly con adic he abo e s a emen s. Howe e ,
i mus be conside ed ha hese coas al wa e s, associa ed o hea y ain episodes, we e highly p oduc i e.
In any case, in he Medi e anean, such coas al e ilisa ion mechanisms a e subjec o g ea a iabili y
and a e o sho du a ion (1 – 3 weeks), which would no be ime enough o main ain zooplank on
popula ions (Salgado-He nanz e al., 2019). Mo eo e , long e m declining ends in he su ace
chlo ophyll concen a ion ha e been epo ed in coas al a eas in he Medi e anean (Gómez-Jakobsen e
al., 2022). This would imply ha in successi e win e s he la ae o hese species would ace
p og essi ely less a ou able condi ions o hei de elopmen and su i al.
Unde a clima e change scena io, i is expec ed ha he dec ease in he e ilisa ion p ocesses in he open
ocean, such as deep-wa e o ma ion (by con ec ion o cascading) and he consequen mixing p ocess,
would lead o inc easingly less a ou able condi ions o he de elopmen and su i al o la al ish
popula ions. The p oduc i i y o coas al wa e s is subjec o high a iabili y and is highly in luenced by
local and inc easingly i egula p ocesses, such as con inen al wa e inpu s, s o ms and winds. La al ish
popula ions in hese a eas would he e o e be s ongly condi ioned by he a iabili y o hese p ocesses,
which would esul in a high a iable and unp edic able ec ui men a e o ish popula ions.
Re e ences
Ådlands ik, B., Coombs, S., Sundby, S., Temple, G., 2001. Buoyancy and e ical dis ibu ion o eggs
and la ae o blue whi ing (Mic omesis ius pou assou): Obse a ions and modelling. Fish. Res. 50,
59–72. h ps://doi.o g/10.1016/S0165-7836(00)00242-3
Alpe , P., Ben-Gai, T., Baha ad, A., Benjamini, Y., Yeku ieli, D., Colacino, M., Dioda o, L., Ramis, C.,
Homa , V., Rome o, R., Michaelides, S., Manes, A., 2002. The pa adoxical inc ease o
Medi e anean ex eme daily ain all in spi e o dec ease in o al alues. Geophys. Res. Le . 29, 31-
1-31–4. h ps://doi.o g/10.1029/2001GL013554
2.1. Va iabili y unde con as ing en i onmen al condi ions
50
Al a ez, A., Tin o é, J., Saba és, A., 1996. Flow modi ica ion and shel -slope exchange induced by a
subma ine canyon o he no heas Spanish coas . J. Geophys. Res. 101, 12043–12055.
Ál a ez, I., Rod íguez, J.M., Ca alán, I.A., Hidalgo, M., Ál a ez-Be as egui, D., Balbín, R., Apa icio-
González, A., Alemany, F., 2015. La al ish assemblage s uc u e in he su ace laye o he
no hwes e n Medi e anean unde con as ing oceanog aphic scena ios. J. Plank on Res. 37, 834–
850. h ps://doi.o g/10.1093/plank / b 055Ande son, M.J., 2001. A new me hod o non-pa ame ic
mul i a ia e analysis o a iance. Aus al Ecol. 26, 32–46.
Asch, R.G., Checkley, D.M., 2013. Dynamic heigh : A key a iable o iden i ying he spawning habi a
o small pelagic ishes. Deep. Res. Pa I Oceanog . Res. Pap. 71, 79–91.
h ps://doi.o g/10.1016/j.ds .2012.08.006
Au h, T.D., 2008. Dis ibu ion and communi y s uc u e o ich hyoplank on om he no he n and cen al
Cali o nia Cu en in May 2004-06. Fish. Oceanog . 17, 316–331. h ps://doi.o g/10.1111/j.1365-
2419.2008.00481.x
Au h, T.D., B odeu , R.D., Soulen, H.L., Ciannelli, L., Pe e son, W.T., 2011. The esponse o ish la ae
o decadal changes in en i onmen al o cing ac o s o he O egon coas . Fish. Oceanog . 20, 314–
328. h ps://doi.o g/10.1111/j.1365-2419.2011.00586.x
Au h, T.D., Daly, E.A., B odeu , R.D., Fishe , J.L., 2018. Phenological and dis ibu ional shi s in
ich hyoplank on associa ed wi h ecen wa ming in he no heas Paci ic Ocean. Glob. Chang. Biol.
24, 259–272. h ps://doi.o g/10.1111/gcb.13872
Ba os-Ga cía, D., Bañón, R., A on e, J.C., Fe nández-Pe al a, L., Ga cía, R., Iglésias, S.P., Sellos, D.Y.,
Ba ei os, J.P., Comesaña, Á.S., De Ca los, A., 2018. New insigh s in o he sys ema ics o No h
A lan ic Gaid opsa us (Gadi o mes, Gadidae): lagging synonymies and hidden di e si y. Ma . Biol.
Res. 14, 17–29. h ps://doi.o g/10.1080/17451000.2017.1367403
B ay, J.R., Cu is, J.T., 1957. An O dina ion o he Upland Fo es Communi ies o Sou he n Wisconsin.
Ecol. Monog . 27, 325–349.
Cal o, E., Simó, R., Coma, R., Ribes, M., Pascual, J., Saba és, A., Gili, J.M., Peleje o, C., 2011. E ec s
o clima e change on Medi e anean ma ine ecosys ems: The case o he Ca alan Sea. Clim. Res. 50,
1–29. h ps://doi.o g/10.3354/c 01040
Canals, M., Company, J.B., Ma ín, D., Sànchez-Vidal, A., Ramí ez-Llod à, E., 2013. In eg a ed s udy o
Medi e anean deep canyons: No el esul s and u u e challenges. P og. Oceanog . 118, 1–27.
h ps://doi.o g/10.1016/j.pocean.2013.09.004
Ca alán, I.A., Oli a , M.P., Palome a, I., Be dale , E., 2006. Link be ween en i onmen al anomalies,
g ow h and condi ion o pilcha d Sa dina pilcha dus la ae in he no hwes e n Medi e anean. Ma .
Ecol. P og. Se . 307, 219–231. h ps://doi.o g/10.3354/meps307219
2.1. Va iabili y unde con as ing en i onmen al condi ions
51
Ciannelli, L., Canna acciuolo, A., Kons andinidis, P., Mi asole, A., Wong-Ala, J.A.T.K., Gue a, M.T.,
D’Amb a, I., Riginella, E., Cianelli, D., 2022. Ich hyoplank on assemblages and physical
cha ac e is ics o wo subma ine canyons in he sou h cen al Ty henian Sea. Fish. Oceanog . 1–17.
h ps://doi.o g/10.1111/ og.12596
Cla ke, K.R., 1993. Non‐pa ame ic mul i a ia e analyses o changes in communi y s uc u e. Aus . J.
Ecol. 18, 117–143. h ps://doi.o g/10.1111/j.1442-9993.1993. b00438.x
Cushing, D.H., 1990. Plank on p oduc ion and yea -class s eng h in ish popula ions: An upda e o he
ma ch/misma ch hypo hesis. Ad . Ma . Biol. 26, 249–293. h ps://doi.o g/10.1016/S0065-
2881(08)60202-3
Cu i a, A., Quinci, E.M., Pa i, B., Bonomo, S., Bonanno, A., Musco, M., To i, M., Placen i, F.,
Basilone, G., Geno ese, S., A me i, G.M., Spanò, A., A culeo, M., Mazzola, A., Mazzola, S., 2016.
Di e en key oles o mesoscale oceanog aphic s uc u es and ocean ba hyme y in shaping la al
ish dis ibu ion pa e n: A case s udy in Sicilian wa e s in summe 2009. J. Sea Res. 115, 6–17.
h ps://doi.o g/10.1016/j.sea es.2016.04.005
Cu i a, A., To i, M., Za ad, R., Zgozi, S., Ja boui, O., Quinci, E.M., Hamza, M., Abdul a ah, E.,
Haddoud, D., El Tu ki, A., Ramadan, A., Missaoui, H., Mi sud, R., Bonomo, S., Mazzola, S., Pa i,
B., 2018. Linking su ace hyd odynamics o plank onic ecosys em: he case s udy o he
ich hyoplank onic assemblages in he Cen al Medi e anean Sea. Hyd obiologia 821, 191–214.
h ps://doi.o g/10.1007/s10750-017-3483-x
D’O enzio, F., Ribe a D’Alcalà, M., 2009. On he ophic egimes o he Medi e anean Sea: A sa elli e
analysis. Biogeosciences 6, 139–148. h ps://doi.o g/10.5194/bg-6-139-2009
Da is, T.L.O., Jenkins, G.P., Young, J.W., 1990. Diel pa e ns o e ical dis ibu ion in la ae o
sou he n blue in Thunnus maccoyii, and o he una in he Eas Indian Ocean. Ma . Ecol. P og. Se .
59, 63–74.
Di Pane, J., Koubbi, P., Gi aldo, C., Le eb e, V., Caboche, J., Ma chal, P., Loo s, C., 2020. Recen
changes in ich hyoplank onic assemblages o he eas e n English Channel. J. Sea Res. 157.
h ps://doi.o g/10.1016/j.sea es.2020.101848
Doyle, M.J., Mo se, W.W., Kendall, A.W., 1993. A compa ison o la al ish assemblages in he
empe a e zone o he no heas Paci ic and no hwes A lan ic Oceans. Bull. Ma . Sci. 53, 588–644.
Doyle, M.J., Picquelle, S.J., Mie , K.L., Spillane, M.C., Bond, N.A., 2009. La al ish abundance and
physical o cing in he Gul o Alaska, 1981-2003. P og. Oceanog . 80, 163–187.
h ps://doi.o g/10.1016/j.pocean.2009.03.002
Es ada, M., La asa, M., Emeliano , M., Gu ié ez-Rod íguez, A., Fe nández-Cas o, B., Ise n-Fon ane ,
J., Mou iño-Ca ballido, B., Sala , J., Vidal, M., 2014. Seasonal and mesoscale a iabili y o p ima y
2.1. Va iabili y unde con as ing en i onmen al condi ions
58
Salgado-He nanz, P.M., Racaul , M.F., Fon -Muñoz, J.S., Bas e e xea, G., 2019. T ends in
phy oplank on phenology in he Medi e anean Sea based on ocean-colou emo e sensing. Remo e
Sens. En i on. 221, 50–64. h ps://doi.o g/10.1016/j. se.2018.10.036
Sassa, C., Konishi, Y., 2015. La e win e la al ish assemblage in he sou he n Eas China Sea, wi h
emphasis on spa ial ela ions be ween mesopelagic and comme cial pelagic ish la ae. Con . Shel
Res. 108, 97–111. h ps://doi.o g/10.1016/j.cs .2015.08.021
Scho , F., Visbeck, M., Send, U., Fische , J., S amma, L., Desaubies, Y., 1996. Obse a ions o deep
con ec ion in he Gul o Lions, no he n Medi e anean, du ing he win e o 1991/92. J. Phys.
Oceanog . 26, 505–524. h ps://doi.o g/10.1175/1520-0485(1996)026<0505:OODCIT>2.0.CO;2
Su he s, I.M., Schae e , A., A che , M., Roughan, M., G i in, D.A., Chapman, C.C., Sloyan, B.M.,
E e e , J.D., 2023. F on al eddies p o ide an oceanog aphic iad o a o able la al ish habi a .
Limnol. Oceanog . 68, 1019–1036. h ps://doi.o g/10.1002/lno.12326
Theocha is, A., Geo gopoulos, D., 1993. Dense wa e o ma ion o e he Samo h aki and Limnos
Pla eaux in he no h Aegean Sea (Eas e n Medi e anean Sea). Con . Shel Res. 13, 919–939.
h ps://doi.o g/10.1016/0278-4343(93)90017-R
Tsikli as, A.C., An onopoulou, E., S e giou, K.I., 2010. Spawning pe iod o Medi e anean ma ine ishes.
Re . Fish Biol. Fish. 20, 499–538. h ps://doi.o g/10.1007/s11160-010-9158-6
Vand omme, P., S emmann, L., Be line, L., Gaspa ini, S., Mousseau, L., P ejge , F., Passa iume, O.,
Gua ini, J.M., Go sky, G., 2011. In e -annual luc ua ions o zooplank on communi ies in he Bay o
Ville anche-su -me om 1995 o 2005 (No he n Ligu ian Sea, F ance). Biogeosciences 8, 3143–
3158. h ps://doi.o g/10.5194/bg-8-3143-2011
Va gas-Yáñez, M., Ga cía-Ma ínez, M.C., Moya, F., Balbín, R., López-Ju ado, J.L., Se a, M., Zunino,
P., Pascual, J., Sala , J., 2017. Upda ing empe a u e and salini y mean alues and ends in he
Wes e n Medi e anean: The RADMED p ojec . P og. Oceanog . 157, 27–46.
h ps://doi.o g/10.1016/j.pocean.2017.09.004
Va gas-Yáñez, M., Saba és, A., 2007. Mesoscale high- equency a iabili y in he Albo an Sea and i s
in luence on ish la ae dis ibu ions. J. Ma . Sys . 68, 421–438.
h ps://doi.o g/10.1016/j.jma sys.2007.01.004
Vicen e-Se ano, S.M., Rod íguez-Camino, E., 2017. Obse ed a mosphe ic ends in he Ibe ian
Peninsula. Cli a Exch. 20–23.
Wa d, J.H., 1963. Hie a chical G ouping o Op imize an Objec i e Func ion. J. Am. S a . Assoc. 58, 236–
244. h ps://doi.o g/h ps://doi.o g/10.1080/01621459.1963.10500845

2.1. Va iabili y unde con as ing en i onmen al condi ions
59
Yen sch, C.S., Menzel, D.W., 1963. A me hod o he de e mina ion o phy oplank on chlo ophyll and
phaeophy in by luo escence. Deep. Res. Oceanog . Abs . 10, 221–231.
h ps://doi.o g/10.1016/0011-7471(63)90358-9
Za ad, R., Rod íguez, J.M., Alemany, F., Cha e , A., Ja boui, O., Missaoui, H., 2020. La al ish
communi y composi ion and dis ibu ion o he cen al-sou he n medi e anean unde summe and
win e condi ions. Ac a Ich hyol. Pisca . 50, 313–324. h ps://doi.o g/10.3750/AIEP/02957
Chap e 2
Recu ence o he spa ial s uc u e o summe la al ish
assemblages linked o hyd odynamics in he NW Medi e anean
2.2. Recu ence o he spa ial s uc u e in summe
63
Recu ence o he spa ial s uc u e o summe la al ish
assemblages linked o hyd odynamics in he NW Medi e anean
Vanesa Raya, Jo di Sala , Ana Saba és
(submi ed)
Abs ac
In his s udy we aim o in es iga e he con ibu ion o a ious sou ces o a iabili y, en i onmen al,
spa ial, and in e annual on he la al ish assemblages (LFA) in summe o he Ca alan coas , an a ea
wi h a wide a ay o en i onmen al condi ions and high hyd odynamic ac i i y. The s udy was based on
h ee ich hyoplank on su eys ca ied ou in July 2003, 2004 and 2012. The numbe o iden i ied axa,
belonging o shel and oceanic ish species, was qui e simila among c uises, a ound 80, wi h he small
pelagics Eng aulis enc asicolus and Sa dinella au i a being he dominan species. A high spa ial
he e ogenei y in LFA was de ec ed h oughou he s udied a ea. Mos o his he e ogenei y was ela ed o
he spa ial a iabili y o en i onmen al ac o s, such as he con inen al shel s uc u e, la i udinal
di e ence in su ace empe a u e, chlo ophyll-a concen a ion, and s a i ica ion le el. Hyd odynamic
s uc u es, mainly he he mal on ac oss he shel , ins abili ies o he No he n Cu en and an icyclonic
eddies, also played an impo an ole in he con igu a ion o he LFA. Al hough hese showed ce ain
in e annual a iabili y, he ecu ence o hyd odynamic s uc u es, oge he wi h he spawning habi s o
adul s, means ha he spa ial s uc u e o LFA is main ained in di e en yea s.

2.2. Recu ence o he spa ial s uc u e in summe
64
2.2.1. In oduc ion
La al ish assemblages (LFA), i. e. occu ence, dis ibu ion, and abundance o ish la ae wi hin he
complex and a iable ma ine ecosys ems, a e he ou come o co-adap a ion p ocess o species spawning
s a egies (e. g. spawning ime and loca ion, spawning du a ion and equency) (Doyle e al., 1993). These
s a egies ha e e ol ed o adap o speci ic ea u es o he en i onmen s ha ish inhabi , ensu ing ha he
o sp ing is placed whe e la ae can ind he bes condi ions ha allow success ul eeding, g ow h and
su i al (Cowen and Sponaugle, 2009; Peck e al., 2012). Spa ial pa e ns in LFA a e ul ima ely
de e mined by en i onmen al p ocesses ha ope a e a di e en empo al and spa ial scales (Doyle e al.,
1993). These p ocesses include hose o la ge scale, as clima e pa e ns (Di Pane e al., 2020; Koslow e
al., 2013), as well as en i onmen al a iabili y a seasonal and in e annual ime scales, which de e mine
adul dis ibu ion and hei spawning s a egies (Au h, 2008; Doyle e al., 2009). A a local and sho ime
scale, LFA will be de e mined by hyd odynamic s uc u es (e.g. cu en s, on s, eddies) ha in luence
ish la al dispe sal and e en ion (Mose and Smi h, 1993; Su he s e al., 2023) and by biologic ac o s,
such as ood concen a ion and p eda ion (Houde, 2008; Mille and Kendall, 2009), ha ul ima ely
de e mine ish la ae su i al (Cushing, 1990; Houde, 2008).
The Medi e anean ish auna is cha ac e ised by high di e si y (Bianchi and Mo i, 2000; Quigna d and
Tomasini, 2000). Species o sub opical o igin a e mainly ound in he eas e n basin and he sou he n
Medi e anean, whe e wa e empe a u e is highe han a e age (Theocha is and Geo gopoulos, 1993).
Cold- empe a e species inhabi he no he n a eas (Gul o Lion, Ligu ian Sea and no he n Ad ia ic:
Bianchi and Mo i, 2000) whe e he wa e is colde wi h a seasonal a ia ion a he su ace anging om
13 o 25ºC (Sala , 1996). In he Medi e anean, he p esence o ish la ae in he plank on shows a ma ked
seasonali y ela ed o he spawning cycles o he adul s (Saba és e al., 2007a). Mos ish species spawn
du ing sp ing and summe (Tsikli as e al., 2010), when oligo ophy p e ails in he uppe wa e column.
The highe empe a u es o he summe pe iod a ou as e la al g ow h, and he s abili y o he wa e
column could con ibu e o main ain ood pa ches (Saba és e al., 2007a). The summe pe iod is
cha ac e ized by a s a i ied wa e column, wi h a ma ked he mocline ha limi s e ical mixing.
Consequen ly, p ima y p oduc ion emains concen a ed a he deep chlo ophyll maximum, a hin laye a
he deepes le els o he pho ic zone (Es ada, 1985). Su ace p oduc i i y is es ic ed o some coas al
zones (A ienza e al., 2016) and o a eas unde he in luence o uno wa e s, mainly he Rhône and he
Eb o (Sala , 1996).
The s udy a ea, o he Ca alan coas , in he NW Medi e anean Sea, is cha ac e ized by a ma ked
la i udinal g adien o su ace empe a u e. The no he n sec o , which is mo e di ec ly in luenced by
s ong no he ly winds, is colde han cen al and sou he n pa s and a he mal on de elops a su ace,
pe pendicula o he coas , associa ed wi h he sou he n limi o he s ong no he lies (Saba és e al.,
2009). The con inen al shel is gene ally na ow, widening in he sou he nmos pa , in he icini y o he
2.2. Recu ence o he spa ial s uc u e in summe
65
Eb o i e del a, and in he no h be ween he wo majo subma ine canyons (Blanes and Palamós
canyons). The oceanog aphic dynamics is la gely con olled by he No he n Cu en (NC), which
ypically lows sou hwes wa ds along he con inen al slope a app oxima ely 30−50 cm s−1, a he su ace
(Millo , 1990; Sala , 1995). The NC is in geos ophic equilib ium wi h a shel -slope densi y on ,
de ec able down o abou 400 m dep h, and lea es eshe and coole wa e s on i s coas al side (Fon e al.,
1988; Millo , 1990). The NC displays high mesoscale a iabili y ha causes oscilla ions, meande ing and
eddy gene a ion (Rubio e al., 2005). These mesoscale s uc u es a e igge ed by he complex ba hyme y
(canyons, ab up changes in he di ec ion o he con inen al slope, widening o na owing shel es) ha
s ongly modi y he local ci cula ion and gene a e shel -slope exchanges (Flexas e al., 2008; Masó e al.,
1998).
Mos o he s udies ca ied ou along he Ca alan coas dealing wi h he ole o en i onmen al ac o s (e.g.
ba hyme y, empe a u e and chlo ophyll-a), as well as hyd odynamic s uc u es (e.g. cu en s,
an icyclonic eddies), on he dis ibu ion, concen a ion and dispe sal o ish la ae ha e been ca ied ou
on pelagic species, such as Eng aulis enc asicolus and Sa dinella au i a (Saba és e al., 2013, 2009, 2004).
The s udies add essing he e ec o en i onmen al ac o s on he ish la al communi y as a whole a e
sca ce and es ic ed o a limi ed a ea (Oli a e al., 2010; Saba és and Oli a , 1996). In he p esen s udy,
we in es iga e he species assemblages o he ish la al communi y along he Ca alan coas . This a ea,
wi h i s high mesoscale ac i i y and wide a ay o en i onmen al condi ions (Saiz e al., 2014; Sala ,
1996) p o ides an ideal loca ion o analyse he e ec o he en i onmen al ac o s on he spa ial s uc u e
o LFA. The s udy is based on h ee ich hyoplank on su eys ca ied ou in summe 2003, 2004 and 2012.
The aim was o assess he ole o di e en sou ces o a iabili y, including en i onmen al, spa ial, and
in e annual on he con o ma ion o LFA in summe along he Ca alan coas .
2.2.2. Ma e ials and me hods
Oceanog aphic da a and ich hyoplank on sampling
Th ee oceanog aphic su eys we e conduc ed along he Ca alan coas , NW Medi e anean, in July: 18–
25, 2003; 21–29, 2004 and 12–23, 2012 (Fig. 2.2.1). In 2003 and 2004, samplings we e conduc ed a 66
s a ions loca ed along ansec s pe pendicula o he sho eline, om nea he coas o he slope. In 2012,
he sampling was ex ended o sho e bu , due o he bad wea he condi ions, he wo no he n ansec s
could no be ca ied ou and a o al o 73 s a ions we e sampled. In all su eys, he s a ions we e placed
~14 km apa , and he dis ance be ween ansec s was ~18.5 km (Fig. 2.2.1).
A each s a ion, e ical p o iles o he basic hyd og aphic a iables ( empe a u e, salini y and
luo escence) we e ob ained wi h a Neil B own Ma k III-CTD (WOCE s anda d) equipped wi h a Sea-
Tech in si u luo ome e . The e ical p o iles we e in e pola ed o 1-m dep h in e als. The dynamic
heigh s we e calcula ed wi h a e e ence le el o 600 m and, whe e dep h was lowe , we e ex apola ed
2.2. Recu ence o he spa ial s uc u e in summe
66
using he con inui y equa ion applied o he deepes le el o he h ee-s a ion clus e s (c . Hidaka, 1940).
Geos ophic eloci y componen s a he s a ions we e de i ed om a s eam unc ion ob ained by
in e pola ing dynamic heigh s o e he whole a ea (S ewa , 2008). The maximum e ical densi y
g adien was calcula ed by cen ed di e ences a 1-m in e als, and he mean dep h o he pycnocline was
es ablished whe e he maximum o he e ical densi y g adien was obse ed. To calib a e he
luo ome e , wa e samples o chlo ophyll-a de e mina ion we e collec ed wi h a ose e sys em a h ee
dep hs down o 80 m, h oughou he day and nigh (see Saba és e al., 2009 o me hodological de ails).
Fig. 2.2.1. S udy a ea o he Ca alan coas , in he
NW Medi e anean. Blue ci cles indica e
sampling s a ions (CTD and Bongo ne ) du ing
he July 2003 and 2004 su eys. C osses indica e
sampling s a ions (CTD and Bongo ne ) du ing
he July 2012 su ey.
Fish la ae we e collec ed a each s a ion, du ing day and nigh , using a Bongo ne wi h a mou h opening
o 60 cm o diame e and a mesh size o 300 μm. The ne was owed a 2 kno s, obliquely om a
maximum dep h o 200 m o he su ace. The olume o il e ed wa e by he ne s was es ima ed by
means o a lowme e placed in he cen e o each ne mou h. Zooplank on samples we e ixed in 5%
o maldehyde bu e ed wi h sodium e abo a e. In he labo a o y, all ish la ae we e so ed om he
p ese ed samples. La ae we e iden i ied o he lowes axonomic le el possible.
Da a analysis
The numbe o la ae o each axon collec ed a each s a ion was s anda dized o numbe pe 10 m2 o he
sea su ace. Fo each species and su ey, he mean abundance alues, equency o occu ence (FO,
2.2. Recu ence o he spa ial s uc u e in summe
67
pe cen age o s a ions whe e a axon occu ed) and he ela i e abundance (RA, pe cen age con ibu ion
o a axon o he o al abundance o indi iduals) we e calcula ed.
To analyse he spa ial and in e annual pa e ns in LFA, mul i a ia e analyses we e pe o med using he
“clus e ” (Maechle e al., 2022) and “ egan” (Oksanen e al., 2022) packages in he so wa e R . 4.3.1
(R Co e Team, 2022). Da a o la al abundance we e ln(x + 1)- ans o med o educe he weigh ing o
he dominan species. Analyses we e pe o med conside ing all h ee su eys oge he , and only axa ha
we e p esen in a leas 5% o all he s a ions sampled (a o al o 59 axa) we e included. The h eshold o
s a is ical signi icance o all analyses was se a α = 0.05. To assess he exis ence o g oups o s a ions
wi h simila axa composi ion and abundance, a clus e analysis was ca ied ou using he B ay-Cu is
index as he dis ance measu e (B ay and Cu is, 1957) and he Wa d’s hie a chical agglome a ion me hod
(Wa d, 1963). G oups o s a ions we e plo ed on a map o he sampled egion o iden i y possible
geog aphical pa e ns. Pe mu a ional mul i a ia e analysis o a iance (Adonis es , 999 pe mu a ions;
Ande son, 2001) was used o es o signi ican di e ences in la al ish abundance and axa composi ion
be ween clus e s. The mean la al abundance o di e en axa in each clus e was calcula ed o isually
ep esen hese di e ences. In addi ion, a non-me ic mul i-dimensional scaling (nMDS), was pe o med
o p o ide a wo-dimensional ep esen a ion o he assemblage s uc u e.
The in luence o en i onmen , space and yea on he a iabili y in he composi ion and abundance o he
di e en axa in he LFA was analysed using he a ia ion pa i ioning me hod (Bo ca d e al., 1992;
Mood, 1971; Pe es-Ne o e al., 2006). This me hod conside s di e en se s o explana o y a iables
(en i onmen al, spa ial and in e annual) and es ima es he amoun o a ia ion in he ish la ae ma ix
ha can be a ibu ed exclusi ely o one o he o he se o explana o y a iables, a e emo ing he e ec
o he o he se s o a iables. I also allows o es ima e he amoun o a ia ion explained join ly by wo
explana o y da a se s. He ea e , we will use ‘ e m’ o e e o a da ase o explana o y a iables. The
a ia ion pa i ioning me hod was pe o med in wo s eps (Bo ca d e al., 2011; Legend e and Legend e,
2012). Fi s , canonical co espondence analyses (CCA) we e used o explain he a ia ion in he ish
la al communi y, one analysis o each e m. The second s ep was o pe o m pa ial canonical analyses
o de e mine he ela i e con ibu ion o each e m. The signi icance o he a ia ion explained by each
e m was assessed by a Mon e Ca lo pe mu a ion es (999 pe mu a ions). Finally, a ia ion pa i ioning
was pe o med.
The en i onmen al da a se used in he CCA o analyse he in luence o en i onmen al ac o s on he ish
la al communi y included: bo om dep h (log, Dep h); he su ace (5 m) empe a u e (SST), salini y
(SSS) and chlo ophyll-a (SSChla); maximum densi y g adien (Gmax); pycnocline dep h (PD);
geos ophic eloci y componen s; and su ace (10 m) dynamic heigh (DH). Gmax and PD we e
conside ed o analyse he e ec o he s eng h o he s a i ica ion. The geos ophic eloci y componen s
and DH we e included o analyse he in luence o hyd odynamic s uc u es (Asch and Checkley, 2013).
As he DH alues in he Medi e anean a e nega i e, hey we e ans o med in o posi i e alues in he
2.2. Recu ence o he spa ial s uc u e in summe
74
Fig. 2.2.5. Plo o he mean abundance (No. pe 10m2) o axa o each clus e .
Clus e analysis iden i ied six g oups o s a ions wi h signi ican di e ences be ween hem (p < 0.05; Fig.
2.2.4 a). The spa ial dis ibu ion o hese g oups showed a g ada ion in he ish la al communi y
composi ion, om shel o oceanic s a ions, mainly in 2003 and 2012 (Fig. 2.2.4 b and Fig. 2.2.5). G oup
A was he mos coas al assemblage in he h ee sampled yea s, and was mainly loca ed in he sou he n
hal o he a ea. I was cha ac e ised by he lowes numbe o axa, almos exclusi ely shel species, such
as E. enc asicolus, S. au i a, Spa idae and Gobiidae (Fig. 2.2.4 b and Fig. 2.2.5). G oups B, C and D we e
loca ed on he shel . G oup B in 2012 and 2004 and G oup C in 2003 and 2004 we e mainly ound in he

2.2. Recu ence o he spa ial s uc u e in summe
75
Eb o shel , and G oup D in he no he n hal in he h ee sampled yea s (Fig. 2.2.4 b). Clus e B showed a
ela i ely high abundance o some shel ben hic species, such as Se anus hepa us, T achinus d aco,
A noglossus spp. and Diplodus annula is, bu he oceanic species C. made ensis and C. b aue i, we e
also abundan (Fig. 2.2.5). The species composi ion o g oup C was simila o ha o g oup B. The
dis inc ion be ween hem was a highe abundance o la ae o some he mophilic species in he la e ,
such as S. au i a, Poma omus sal a ix and Ca anx honchus (Fig. 2.2.5). G oup D was cha ac e ised by
he p esence o bo h, shel (e. g. E. enc asicolus, S. au i a, S. hepa us, A noglossus spp. and Auxis ochei)
and oceanic axa (e.g. C. b aue i, Ce a oscopelus made ensis, Lampanyc us c ocodilus and
A gy opelecus hemigymmnus) (Fig. 2.2.5). G oups E and F we e ound o sho e in he h ee yea s.
Clus e E assembled s a ions loca ed on he shel b eak and on he slope, and clus e F he mos o sho e
s a ions (Fig. 2.2.4 b). These oceanic g oups exhibi ed he highes abundances o he oceanic axa, while
clus e E also had ele a ed abundances o he shel species (Fig. 2.2.5).
Fig 2.2.6. Non-me ic mul idimensional
scaling (nMDS) o dina ion plo . Symbols
o he same colou indica e clus e s o
s a ions wi h simila species composi ion.
Squa es co espond o he s a ions sampled
in 2003, ci cles o he s a ions sampled in
2004 and iangles o he s a ions sampled
in 2012.
The nMDS o dina ion analysis (2D s ess = 0.178) showed some seg ega ion be ween he six clus e s
(Fig. 2.2.6). The s a ions o he oceanic g oups (E, F) we e loca ed on he igh side o he plo , being
clus e E s a ions mainly loca ed on he nega i e side o he second axis and hose o he g oup F on i s
posi i e side. The o he ou g oups we e si ua ed on he le side o he plo . Clus e s A and B we e ound
on he posi i e side o he second axis and clus e s C and D on i s nega i e side (Fig. 2.2.6).
In luence o en i onmen on he spa ial and in e annual a iabili y o LFA
All e ms (en i onmen al, spa ial and in e annual) included in he a ia ion pa i ioning analysis we e
signi ican (p < 0.05) and explained 33.3% o he a ia ion in he la al ish communi y (Fig. 2.2.7). The
highes pe cen age o a ia ion was explained by he en i onmen al a iables (21.4%), and he
2.2. Recu ence o he spa ial s uc u e in summe
76
con ibu ion o spa ial and in e annual e ms was 14.4% and 7.6%, espec i ely. These alues include he
sha ed a ia ion be ween he h ee e ms (Fig. 2.2.7). Conside ing exclusi ely he e ec o each e m,
en i onmen explained 11.7%, space 7.1% and yea 4.9% o he o al a ia ion.
Fig. 2.2.7. Venn diag am o he a ia ion pa i ioning analysis showing
he pe cen age o a ia ion explained in he la al ish assemblages by
each e m, spa ial, en i onmen al, and in e -annual.
The en i onmen al CCA iden i ied se en signi ican a iables: Dep h, SST, SSS, SSChla, DH
(hyd odynamics), PD and Gmax (s a i ica ion) (Table 2.2.2). The i s o dina ion axis explained 60.03%
o he cons ained a iance and was nega i ely co ela ed wi h Dep h, and posi i ely wi h SSChla. The
second axis explained 16.94% o he cons ained a iance and was posi i ely co ela ed wi h SST and
nega i ely wi h PD. The spa ial CCA o he de ended ma ix o ish la ae was signi ican (p < 0.05).
The i s axis explained 46.3% and he second 20.7% o he cons ained a iance.
Table 2.2.2. Summa y o he canonical co espondence analysis (CCA) esul s. Indi idual a iables a e o de ed by he
pe cen age o a iance explained (Exp. Va .), in ela ion o he o al and cons ained a iance. The signi icance o each a iable
(P) oge he wi h i s es s a is ic (F- alue), and he in e se co ela ions o en i onmen al a iables wi h he h ee signi ican axes
o he CCA a e p o ided.
The o dina ion plo o he i s wo axes o en i onmen al CCA showed he seg ega ion o shel and
oceanic clus e s, as obse ed in he nMDS (Fig. 8 a). The shel g oups we e placed on he igh side o he
plo . G oup A was associa ed wi h he lowes Dep h and he highes SSChla and Gmax alues (Fig. 2.2.8
a). Clus e B was ela ed o high SST and Gmax and low PD and SSS. G oup C was associa ed wi h he
2.2. Recu ence o he spa ial s uc u e in summe
77
highes alues o SST, SSChla and Gmax and low alues o PD, SSS, and Dep h. Clus e D was
associa ed wi h high DH, SSChla and Gmax, and low Dep h alues. The oceanic g oups, E and F we e
loca ed on he le side o he plo , being clus e F associa ed wi h highe Dep h, SSS and PD and lowe
SSChla han clus e E (Fig. 2.2.8 a). The geog aphical ep esen a ion o he sco es o each s a ion on axis
1 om he en i onmen al CCA, showed hei ela ionship wi h Dep h, wi h posi i e alues in coas al
loca ions and nega i e o sho e (Fig. 2.2.9 a, b and c). The sco es on axis 2 exhibi ed he la i udinal
g adien obse ed o SST, wi h posi i e alues in he sou h and nega i e in he no h (Fig. 2.2.9 d, e and
). These esul s e idence he ela ionship be ween en i onmen al and spa ial e ms.
Fig. 2.2.8. O dina ion plo s o he en i onmen al canonical co espondence analysis (CCA) (le panels, a and c) and o he
spa ial CCA ( igh panels, b and d). Uppe panels, a and b: O dina ion o he sampling s a ions in he CCA plo s, showing he
ela ionships be ween he sampling s a ions and he explana o y en i onmen al a iables (blue ec o s), o July o 2003, 2004
and 2012. Symbols o he same colou indica e clus e s o s a ions wi h simila species composi ion. Squa es co espond o he
s a ions sampled in 2003, ci cles o he s a ions sampled in 2004 and iangles o he s a ions sampled in 2012. Lowe panels, c
and d: O dina ion o he la al ish axa in he CCA plo s, showing he ela ionships be ween he la al ish axa and he
explana o y en i onmen al a iables (blue ec o s), o July o 2003, 2004 and 2012. The abb e ia ions o la al ish names a e
indica ed in Table 2.2.1.
2.2. Recu ence o he spa ial s uc u e in summe
78
Fig. 2.2.9. Spa ial dis ibu ion o sco e alues o s a ions o he axis 1 (uppe panels, a, b and c) and axis 2 (lowe panels, d, e
and ) ha esul ed om he en i onmen al canonical co espondence analysis (CCA) o 2003 (le panels, a and c), 2004
(cen al panels, b and e) and 2012 ( igh panels, c and ). Symbols o he same colou deno e s a ions belonging o he same g oup
iden i ied in he clus e analysis. Isoba hs shown: 50, 200 and 1000 m.
Fig. 2.2.10. Spa ial dis ibu ion o he i ed sco e alues o s a ions o he axis 1 (colou g adien ) ha esul ed om he spa ial
canonical co espondence analysis (CCA) on he de ended la al ish ma ix o 2003 (le panel), 2004 (cen al panel) and 2012
( igh panel). Geos ophic cu en ield (a ows) and ba hyme y ha e been o e laid. Isoba hs shown: 50, 200 and 1000m.
2.2. Recu ence o he spa ial s uc u e in summe
79
The spa ial analysis iden i ied he di e en clus e s al hough hey we e no as clea ly seg ega ed as in he
en i onmen al analysis (Fig. 2.2.8 b). The geog aphical ep esen a ion o he i ed si e sco es o each
s a ion on axis 1 esul ing om he spa ial CCA showed simila spa ial s uc u es in he h ee s udied
yea s (Fig. 2.2.10). Posi i e alues we e obse ed in he sou he n shel and in he shel , be ween he
Palamós and Blanes canyons, a he no he n side o he an icyclonic eddies. Nega i e alues we e ound
o sho e in he no he nmos pa o he a ea, associa ed wi h he in usion o he NC, and in he cen al
pa ela ed o ins abili ies o he NC (Fig. 2.2.10).
The species o dina ion, bo h in he en i onmen al and in he spa ial CCA plo s, showed seg ega ion
be ween shel and oceanic species (Fig. 2.2.8 c and d). The shel species we e mainly loca ed on he
posi i e side o axis 1, associa ed wi h high alues o SSChla and Gmax in he en i onmen al CCA, and
low dep h and SSS. La ae o hese species, such as S. hepa us, we e loca ed all o e he shel , being
especially abundan in he sou he n a ea, ela ed o he highes concen a ions o SSChla and low salini y
wa e s om he Eb o i e uno (Fig. 2.2.8 c and Fig. 2.2.11 a, b and c). The he mophilic species e.g.
Sa dinella au i a, Ca anx honchus and Poma omus sal a ix we e associa ed wi h high SST and low PD
alues (Fig. 2.2.8 c). La ae o hese species, illus a ed by S. au i a, we e loca ed along he con inen al
shel , hei dis ibu ion owa ds he no h being limi ed by he posi ion o he he mal on (Fig. 2.2.11 d,
e and ). I should be no ed ha he abundance o hese species was pa icula ly low in 2012. La ae o E.
enc asicolus and T achinus d aco we e loca ed on he cen al pa o he en i onmen al CCA plo and
showed no clea ela ionship wi h any en i onmen al a iable (Fig. 2.2.8 c). E. enc asicolus, he mos
abundan species, was p esen in all assemblages. They we e loca ed along he en i e shel in he h ee
s udied yea s wi h highes abundances in he no he n hal o he a ea, in coincidence wi h he in usion o
he NC on he no he n bounda y and he p esence o an icyclonic eddies (Fig 2.2.11 g, h and i). La ae o
T. d aco, placed igh mos on he spa ial CCA plo , we e pa icula ly abundan in wo a eas: o e he
Eb o shel and in he shel be ween he Palamós and Blanes canyons, a he no he n side o he
an icyclonic eddies (Fig. 2.2.8 d and Fig. 11 j, k and l). In bo h CCAs, he oceanic species we e loca ed on
he nega i e side o he axis 1, associa ed wi h high Dep h and SSS and low SSChla and Gmax alues in
he en i onmen al analysis (Fig. 8 c and d). La ae o A. hemigymnus, placed le mos on he spa ial CCA
plo , we e loca ed o sho e, all along he slope, being especially abundan in 2012 (Fig. 2.2.11 m, n and
o). The maximum abundances o C. made ensis we e also loca ed o e he slope, al hough i s dis ibu ion
ex ended o he con inen al shel , pa icula ly in 2012 in he no h o he Eb o shel whe e an in usion o
he NC was de ec ed (Fig. 2.2.11 p, q and ).

2.2. Recu ence o he spa ial s uc u e in summe
80
Fig. 2.2.11.
Spa ial dis ibu ion o ish la ae o ep esen a i e shel and oceanic species in 2003, 2004 and 2012. Se anus hepa us
o e laid on SSChla (a, b and c), Sa dinella au i a on SST (d, e and ), Eng aulis enc asicolus (g, h and i), T achinus d aco (j, k
and l), and Ce a oscopelus made ensis (p, q and ) on he geos ophic cu en ield, and A gy opelecus hemigymnus on
ba hyme y (m, n and o). Isoba hs shown: 50, 200 and 1000 m.
2.2. Recu ence o he spa ial s uc u e in summe
81
2.2.4. Discussion
This s udy analyses he con ibu ion o a ious sou ces o a ia ion, including en i onmen al, spa ial, and
in e annual, on he LFA in summe o he Ca alan coas , an a ea wi h a wide a ay o en i onmen al
condi ions and high hyd odynamic ac i i y. Ou indings show ha en i onmen al condi ions we e he
main d i e o he obse ed a ia ion, wi h hei con ibu ion o spa ial a iabili y being highe han hei
in luence on in e annual a iabili y.
En i onmen al se ing
In he NW Medi e anean, he ci cula ion pa e n is domina ed by he No he n Cu en , wi h i s
meande ing beha iou and some in usions owa ds he coas . The meande ing beha iou o he NC leads
o some an icyclonic eddies on i s coas al side ha ha e been p e iously epo ed in he no h-wes e n
Medi e anean coas s (Flexas e al., 2002; Saba és e al., 2004). In he h ee s udied yea s, a ecu en
eddy was loca ed a nea ly he same posi ion, o e he Blanes canyon, ha would be a ou ed by he
in e ac ion be ween he NC and he canyon (Flexas e al., 2008; Rubio e al., 2005) (Fig. 2.2.2 g, h and i).
The p esence o cold wa e s in he no he n egion con ibu es o he o ma ion o he pe sis en he mal
on , no h o he Blanes canyon, in he h ee c uises. Howe e , in 2012, he he mal on was blu ed by
he s ong in usion o cold, low salini y wa e ad ec ed by he NC along he con inen al slope and
apped by he an icyclonic eddy (Fig. 2.2.2 c). These wa e s p obably come om he Rhône i e uno
in he Gul o Lion, as p e ious s udies ha e desc ibed ha hey can be ad ec ed by he NC sou hwa ds
along he con inen al slope, being de ec ed o he sou h o he Blanes canyon mainly in sp ing and
summe (Saba és e al., 2007b; Sala , 1996).
In July 2003 he su ace empe a u e was a ound 2°C highe han usual in he whole egion, eaching
ex emely high alues in he sou he n pa . These wa m wa e s we e he esul o he successi e hea
wa es ha a ec ed sou h-wes e n Eu ope du ing sp ing and summe o 2003 (Black e al., 2004;
Lu e bache e al., 2004). The anomalous wa ming was limi ed o he uppe le els o he wa e column,
con ibu ing o an inc ease in he s a i ica ion in ensi y (Spa nocchia e al., 2006), in acco dance wi h he
high Gmax and low PD alues obse ed in he sou he n pa o he s udy a ea his yea .
In he sou he n a ea, he mesoscale dynamics was in gene al domina ed by he Eb o i e uno and he
ela i ely wide con inen al shel . High SSChla alues we e associa ed wi h he low-salini y wa e pa ches
om he Eb o uno , as p e iously epo ed in ha a ea (A in e al., 2005; Saba és e al., 2008). The
change in he o ien a ion o he shel b eak in he no h o he Eb o shel a ec s he s abili y o he NC,
enhancing he in usion o slope wa e s in ha a ea (Sala e al., 2002). The ci cula ion pa e ns and he
SSS dis ibu ions sugges ha bo h p ocesses, in usions and i e uno , in e ac dynamically agains
each o he leading o a agmen ed s uc u e o small pa ches, clea ly e iden in 2003 and 2004 (Saba és
e al., 2013). Thus, he Eb o shel is usually cha ac e ized by high p oduc i i y and has been desc ibed as
2.2. Recu ence o he spa ial s uc u e in summe
82
an impo an spawning a ea o shel dwelling ish species (Palome a e al., 2007; Saba és e al., 2007a).
In 2012, a decline in he Eb o i e uno (Cozzi e al., 2018), would allow he in usion o slope wa e s
o each he coas . In his yea , a dec ease in he concen a ion o nu ien s om he uno o he Eb o and
Rhône i e s (Cozzi e al., 2018), p obably caused he educ ion in SSChla concen a ion de ec ed his
yea in he whole a ea.
LFA and in e annual changes
The high species ichness alues (79 axa) we e qui e consis en among yea s. These alues a e simila o
hose usually ound in summe in he Medi e anean Sea (e. g. Ál a ez e al., 2015; Saba és, 1990;
Soma akis e al., 2011a), and only lowe o hose epo ed o opical and sub opical a eas (e.g. Beckley
e al., 2018; Keane and Nei a, 2008; Meine e al., 2020). This species ichness is conside ably highe
han ha obse ed in win e and is ob iously ela ed o he spawning cycles o he adul s, since mos
Medi e anean ish species (e.g. Spa idae, Lab idae, Blennidae, Mullidae, Se anidae, Ca angidae)
ep oduce du ing he sp ing–summe pe iod (Tsikli as e al., 2010). The p edominance o clupei o mes,
Eng aulis enc asicolus and Sa dinella au i a (Table 2.2.1) is a common ea u e o Medi e anean LFA in
summe , being he o me he mos abundan species (Oli a e al., 2010; Tsikli as e al., 2009; Za ad e
al., 2013).
The highes abundances o la ae o shel species we e ound in he no he n hal o he s udy a ea,
sugges ing ha hyd odynamic s uc u es, as he NC, eddies and he empe a u e on , play a ole in
shaping hei spa ial dis ibu ion (Bakun, 2006; Golds ein e al., 2019; Saba és e al., 2007b). O e all,
la ae o oceanic species we e abundan o e he shel b eak, pa icula ly in 2012, when sampling was
ca ied ou u he o sho e. Howe e , his yea , la ae o oceanic species we e also de ec ed no h o he
Eb o shel and in he cen al egion, which would be ela ed o he in usion o slope wa e s owa ds he
coas (A ienza e al., 2016; Oli a e al., 2010).
The spa ial dis ibu ion o species ichness showed he highes alues in he no he n hal o he a ea, in
ag eemen wi h he LFA iden i ied. In his egion, whe e he shel is na ow and u owed by subma ine
canyons, he assemblages included a high numbe o species, bo h coas al and oceanic. In he wide Eb o
shel , he assemblages con ained a lowe numbe o species, excep a he shel b eak, due o he
coexis ence o la ae o shel and oceanic axa (Doyle e al., 1993; Isa i e al., 2008; Oli a e al., 2010).
While se e al s udies ha e shown high in e annual a iabili y in LFA in ela ion o en i onmen al ac o s
(e.g. Ál a ez e al., 2015; Au h e al., 2018; Hsieh e al., 2009), in he p esen s udy he a ia ion
pa i ioning me hod e idenced ha he in e annual di e ences in he LFA we e lowe han he spa ial
a ia ion. O his in e annual a iabili y, a small pa would be due o changes in en i onmen al ac o s,
such as SST and SSChla. Mos o his a iabili y would no be ela ed o any o he a iables conside ed
in his s udy, sugges ing ha LFA depend on o he ac o s, o example, hose ha may a ec la al
2.2. Recu ence o he spa ial s uc u e in summe
83
su i al, such as ood abundance (zooplank on) and he p esence o p eda o s (Houde, 2008; Peck e al.,
2012; Pepin e al., 2003). I may also e lec changes in ‘adul ela ed p ocesses’ (e.g, spawning ime,
du a ion and ecundi y; Isa i e al., 2008) ha may a ec he occu ence, dis ibu ion and abundance o
hei la ae (Doyle e al., 1993).
En i onmen al ac o s con olling LFA
The a ia ion pa i ioning me hod e ealed ha he en i onmen al ac o s play he mos impo an ole in
s uc u ing he LFA. Acco ding o he CCA esul s, dep h is he p ima y en i onmen al a iable shaping
he LFA, mainly e lec ing he spawning habi a o adul s, as widely epo ed in di e en sys ems
(Ál a ez e al., 2015; Richa dson e al., 1980; Soma akis e al., 2011b). The ela ionship be ween dep h
and LFA was e idenced in he coas al-o sho e o dina ion o assemblages, pa icula ly in he sou he n
egion, whe e he con inen al shel is wide . Some ew species we e ela ed o low dep hs, mainly hose
ound in he coas al assemblage (e. g. Buglossidium lu eum, Gobiidae) while mos o he shel species,
associa ed o in e media e dep hs, we e ound in he ansi ional clus e s, o e shel o he shel b eak.
La ae o oceanic species, such as A. hemigymnus, we e abundan o e he slope in he oceanic
assemblages, in ela ion o he highes dep hs. O he oceanic species, such as he myc ophid C.
made ensis, also showed high abundances o e he shel , belonging o he ansi ional and oceanic
assemblages. The p esence o C. made ensis la ae on he con inen al shel is common in o he pa s o
he Medi e anean and in he A lan ic (Isa i e al., 2008; John and Zelck, 1997). The la ae o his species
a e loca ed in he shallowe le els o he wa e column, unlike A. hemigymnus la ae which ha e a deepe
e ical dis ibu ion (Oli a e al., 2014), hus, hey a e mo e suscep ible o being anspo ed by
mesoscale s uc u es om he open sea (Saba és and Oli a , 1996; Va gas-Yáñez and Saba és, 2007).
SSS, a a iable ha inc eases owa ds he open sea, also played a ole in shaping he LFA. Coas al wa e s,
in luenced by i e uno showed lowe salini y han slope wa e s, sepa a ed by he shel -slope densi y
on (Fon e al., 1988). La ae o oceanic species, such as A. isso and he myc ophids (e. g.
Lampanyc us c ocodilus, C. made ensis) we e ela ed o high SSS alues, in ela ion o he habi a o he
adul s.
SST was also an impo an a iable in he s uc u e o LFA. I is known ha ish species exhibi op imal
empe a u e anges ha de e mine hei habi a ex ension and hei ep oduc i e pe iod (Au h e al., 2018;
Mu awski, 1993; Peck e al., 2012). In his con ex , a close ela ionship be ween he abundance and
dis ibu ion o la ae o he mophilic species (Sa dinella au i a, Poma omus sal a ix and Ca anx
honchus) and SST was e iden (Fig. 2.2.8 c). These species we e abundan in he sou he n hal o he
a ea, whe e empe a u e is wa me , mainly in 2003 and 2004, being absen in he no he nmos egion.
The spa ial dis ibu ion o S. au i a la ae, limi ed o he no h by he he mal on , is a good indica o o
his ela ionship wi h empe a u e. In he Medi e anean, a no hwa d expansion o he mophilic species
in ela ion o he empe a u e inc ease has been de ec ed in he wes e n (Llo e e al., 2015; Raya and
2.2. Recu ence o he spa ial s uc u e in summe
90
Masó, M., Saba és, A., Pila Oli a , M., 1998. Sho - e m physical and biological a iabili y in he shel -
slope egion o he NW Medi e anean du ing he sp ing ansi ion pe iod. Con . Shel Res. 18, 661–
675. h ps://doi.o g/10.1016/S0278-4343(98)00011-9
Maynou, F., Saba és, A., Sala , J., 2014. Clues om he ecen pas o assess ec ui men o
Medi e anean small pelagic ishes unde sea wa ming scena ios. Clim. Change 126, 175–188.
h ps://doi.o g/10.1007/s10584-014-1194-0
Meine , C.R., Clausen-Spa ks, K., Co nic, M., Su on, T.T., Rooke , J.R., 2020. Taxonomic Richness
and Di e si y o La al Fish Assemblages in he Oceanic Gul o Mexico: Links o Oceanog aphic
Condi ions. F on . Ma . Sci. 7. h ps://doi.o g/10.3389/ ma s.2020.00579
Mille , B.S., Kendall, A.W., 2009. Ea ly li e his o y o ma ine ishes. Uni e si y o Cali o nia P ess.
Millo , C., 1990. The Gul o Lions’ hyd odynamics. Con . Shel Res. 10, 885–894.
Mood, A.M., 1971. Pa i ioning Va iance in Mul iple Reg ession Analyses as a Tool Fo De eloping
Lea ning Models. Am. Educ. Res. J. 8, 191–202.
h ps://doi.o g/h ps://doi.o g/10.3102/00028312008002191
Mose , H.G., Smi h, P.E., 1993. La al ish assemblages o he Cali o nia Cu en egion and hei
ho izon al and e ical dis ibu ions ac oss a on . Bull. Ma . Sci. 53, 645–691.
Mu awski, S.A., 1993. Clima e Change and Ma ine Fish Dis ibu ions: Fo ecas ing om His o ical
Analogy. T ans. Am. Fish. Soc. 122, 647–658. h ps://doi.o g/h ps://doi.o g/10.1577/1548-
8659(1993)122<0647:CCAMFD>2.3.CO;2
Oksanen, J., Simpson, G.L., Blanche , F.G., 2022. Package “ egan”: Communi y Ecology Package. R
package e sion 2.6–2.
Oli a , M.P., Emeliano , M., Villa e, F., U ia e, I., Maynou, F., Ál a ez, I., Mo o e, E., 2010. The ole
o oceanog aphic condi ions and plank on a ailabili y in la al ish assemblages o he Ca alan
coas (NW Medi e anean). Fish. Oceanog . 19, 209–229. h ps://doi.o g/10.1111/j.1365-
2419.2010.00538.x
Oli a , M.P., Saba és, A., Alemany, F., Balbín, R., Fe nández de Puelles, M.L., To es, A.P., 2014. Diel-
dep h dis ibu ions o ish la ae o he Balea ic Islands (wes e n Medi e anean) unde wo
en i onmen al scena ios. J. Ma . Sys . 138, 127–138. h ps://doi.o g/10.1016/j.jma sys.2013.10.009
Ospina-Al a ez, A., Ca alán, I.A., Be nal, M., Roos, D., Palome a, I., 2015. F om egg p oduc ion o
ec ui s: Connec i i y and in e -annual a iabili y in he ec ui men pa e ns o Eu opean ancho y in
he no hwes e n Medi e anean. P og. Oceanog . 138, 431–447.
h ps://doi.o g/10.1016/j.pocean.2015.01.011

2.2. Recu ence o he spa ial s uc u e in summe
91
Palome a, I., Oli a , M.P., Sala , J., Saba és, A., Coll, M., Ga cía, A., Mo ales-Nin, B., 2007. Small
pelagic ish in he NW Medi e anean Sea: An ecological e iew. P og. Oceanog . 74, 377–396.
h ps://doi.o g/10.1016/j.pocean.2007.04.012
Peck, M.A., Huebe , K.B., Llopiz, J.K., 2012. In insic and Ex insic Fac o s D i ing Ma ch-Misma ch
Dynamics Du ing he Ea ly Li e His o y o Ma ine Fishes, 1s ed, Ad ances in Ecological Resea ch.
Else ie L d. h ps://doi.o g/10.1016/B978-0-12-398315-2.00003-X
Pepin, P., Dowe , J.F., Da idson, F.J.M., 2003. A spa ially explici s udy o p ey-p eda o in e ac ions in
la al ish: Assessing he in luence o ood and p eda o abundance on la al g ow h and su i al.
Fish. Oceanog . 12, 19–33. h ps://doi.o g/10.1046/j.1365-2419.2003.00221.x
Pe es-Ne o, P.R., Legend e, P., D ay, S., Bo ca d, D., 2006. Va ia ion pa i ioning o species da a
ma ices: Es ima ion and compa ison o ac ions. Ecology 87, 2614–2625.
h ps://doi.o g/10.1890/0012-9658(2006)87[2614:VPOSDM]2.0.CO;2
Quigna d, J.P., Tomasini, J.A., 2000. Medi e anean ish biodi e si y. Biol. Ma . Medi . 7, 1–66.
R Co e Team, 2022. R: A Language and En i onmen o S a is ical Compu ing.
Raya, V., Saba és, A., 2015. Di e si y and dis ibu ion o ea ly li e s ages o ca angid ishes in he
no hwes e n Medi e anean: Responses o en i onmen al d i e s. Fish. Oceanog . 24, 118–134.
h ps://doi.o g/10.1111/ og.12097
Raya, V., Sala , J., Saba és, A., 2020. The box-balance model: A new ool o assess ish la al su i al,
applied o ield da a on wo small pelagic ish. Ma . Ecol. P og. Se . 650, 289–308.
h ps://doi.o g/10.3354/meps13369
Richa dson, S.L., La oche, J.L., Richa dson, M.D., 1980. La al ish assemblages and associa ions in he
no h-eas Paci ic Ocean along he O egon coas , win e -sp ing 1972-1975. Es ua . Coas . Ma . Sci.
11, 671–699. h ps://doi.o g/10.1016/S0302-3524(80)80017-X
Rubio, A., A nau, P.A., Espino, M., Del Ma Flexas, M., Jo dà, G., Sala , J., Puigde àb egas, J., A cilla,
A.S., 2005. A ield s udy o he beha iou o an an icyclonic eddy on he Ca alan con inen al shel
(NW Medi e anean). P og. Oceanog . 66, 142–156. h ps://doi.o g/10.1016/j.pocean.2004.07.012
Saba és, A., 1990. Dis ibu ion pa e n o la al ish popula ions in he No hwes e n Medi e anean. Ma .
Ecol. P og. Se . 59, 75–82. h ps://doi.o g/10.3354/meps059075
Saba és, A., Ma ín, P., Raya, V., 2012. Changes in li e-his o y ai s in ela ion o clima e change:
blue ish (Poma omus sal a ix) in he no hwes e n Medi e anean. ICES J. Ma . Sci. 278, 1000–
1009. h ps://doi.o g/10.1093/icesjms/ ss053
Saba és, A., Masó, M., 1990. E ec o a shel -slope on on he spa ial dis ibu ion o mesopelagic ish
la ae in he wes e n Medi e anean. Deep Sea Res. Pa A, Oceanog . Res. Pap. 37, 1085–1098.
h ps://doi.o g/10.1016/0198-0149(90)90052-W
2.2. Recu ence o he spa ial s uc u e in summe
92
Saba és, A., Oli a , M.P., 1996. Va ia ion o la al ish dis ibu ions associa ed wi h a iabili y in he
loca ion o a shel -slope on . Ma . Ecol. P og. Se . 135, 11–20.
h ps://doi.o g/10.3354/meps135011
Saba és, A., Oli a , M.P., Sala , J., Palome a, I., Alemany, F., 2007a. Physical and biological p ocesses
con olling he dis ibu ion o ish la ae in he NW Medi e anean. P og. Oceanog . 74, 355–376.
h ps://doi.o g/10.1016/j.pocean.2007.04.017
Saba és, A., Sala , J., Masó, M., 2004. Spa ial he e ogenei y o ish la ae ac oss a meande ing cu en in
he no hwes e n Medi e anean. Deep. Res. Pa I Oceanog . Res. Pap. 51, 545–557.
h ps://doi.o g/10.1016/j.ds .2003.11.003
Saba és, A., Sala , J., Palome a, I., Emeliano , M., De Puelles, M.L.F., Oli a , M.P., 2007b. Ad ec ion o
ancho y (Eng aulis enc asicolus) la ae along he Ca alan con inen al slope (NW Medi e anean).
Fish. Oceanog . 16, 130–141. h ps://doi.o g/10.1111/j.1365-2419.2006.00416.x
Saba és, A., Sala , J., Raya, V., Emeliano , M., 2013. Role o mesoscale eddies in shaping he spa ial
dis ibu ion o he coexis ing Eng aulis enc asicolus and Sa dinella au i a la ae in he
no hwes e n Medi e anean. J. Ma . Sys . 111–112, 108–119.
h ps://doi.o g/10.1016/j.jma sys.2012.10.002
Saba és, A., Sala , J., Raya, V., Emeliano , M., Segu a-Nogue a, M., 2009. Spawning en i onmen al
condi ions o Sa dinella au i a a he no he n limi o i s dis ibu ion ange, he wes e n
Medi e anean. Ma . Ecol. P og. Se . 385, 227–236. h ps://doi.o g/10.3354/meps08058
Saba és, A., Za agoza, N., G au, C., Sala , J., 2008. Ve ical dis ibu ion o ea ly de elopmen al s ages in
wo coexis ing clupeoid species, Sa dinella au i a and Eng aulis enc asicolus. Ma . Ecol. P og. Se .
364, 169–180. h ps://doi.o g/10.3354/meps07461
Saiz, E., Saba és, A., Gili, J.M., 2014. The zooplank on, in: Go edo, S., Dubinsky, Z. (Eds.), The
Medi e anean Sea: I s His o y and P esen Challenges. pp. 1–678. h ps://doi.o g/10.1007/978-94-
007-6704-1
Sala , J., 1996. Re iew o hyd og aphic en i onmen al ac o s ha may in luence ancho y habi a s in
no hwes e n Medi e anean. Sci. Ma . 60, 21–32.
Sala , J., 1995. The in e ac ion be ween he Ca alan and Balea ic cu en s in he sou he n Ca alan Sea.
Ocean. Ac a 18, 227–234.
Sala , J., Ga cia, M.A., C uzado, A., Palanques, A., A ín, L., Gomis, D., Guillén, J., De León, A.,
Puigde àb egas, J., Sosped a, J., Velásquez, Z.R., 2002. Seasonal changes o wa e mass s uc u e
and shel slope exchanges a he Eb o shel (NW Medi e anean). Con . Shel Res. 22, 327–348.
h ps://doi.o g/10.1016/S0278-4343(01)00031-0
2.2. Recu ence o he spa ial s uc u e in summe
93
Soma akis, S., Isa i, S., Machias, A., 2011a. La al ish assemblages in coas al wa e s o cen al G eece:
e lec ions o opog aphic and oceanog aphic he e ogenei y. Sci. Ma . 75, 605–618.
h ps://doi.o g/10.3989/scima .2011.75n3605
Soma akis, S., Ram os, A., Palialexis, A., Vala anis, V.D., 2011b. Con as ing mul ispecies pa e ns in
la al ish p oduc ion ace in e -annual a iabili y in oceanog aphic condi ions o e he N.E.
Aegean Sea con inen al shel (Eas e n Medi e anean). Hyd obiologia 670, 275–287.
h ps://doi.o g/10.1007/s10750-011-0677-5
Spa nocchia, S., Schiano, M.E., Picco, P., Bozzano, R., Cappelle i, A., 2006. The anomalous wa ming o
summe 2003 in he su ace laye o he Cen al Ligu ian Sea (Wes e n Medi e anean). Ann.
Geophys. 24, 443–452. h ps://doi.o g/10.5194/angeo-24-443-2006
S ewa , R.H., 2008. In oduc ion o physical oceanog aphy. Texas A&M Uni e si y.
h ps://doi.o g/10.5860/choice.34-5103
Su he s, I.M., Schae e , A., A che , M., Roughan, M., G i in, D.A., Chapman, C.C., Sloyan, B.M.,
E e e , J.D., 2023. F on al eddies p o ide an oceanog aphic iad o a o able la al ish habi a .
Limnol. Oceanog . 68, 1019–1036. h ps://doi.o g/10.1002/lno.12326
Takasuka, A., Oozeki, Y., Kubo a, H., 2008. Mul i-species egime shi s e lec ed in spawning
empe a u e op ima o small pelagic ish in he wes e n No h Paci ic. Ma . Ecol. P og. Se . 360,
211–217. h ps://doi.o g/10.3354/meps07407
Theocha is, A., Geo gopoulos, D., 1993. Dense wa e o ma ion o e he Samo h aki and Limnos
Pla eaux in he no h Aegean Sea (Eas e n Medi e anean Sea). Con . Shel Res. 13, 919–939.
h ps://doi.o g/10.1016/0278-4343(93)90017-R
Tsikli as, A.C., 2008. Clima e- ela ed geog aphic shi and sudden popula ion inc ease o a small pelagic
ish (Sa dinella au i a) in he eas e n Medi e anean Sea. Ma . Biol. Res. 4, 477–481.
h ps://doi.o g/10.1080/17451000802291292
Tsikli as, A.C., An onopoulou, E., S e giou, K.I., 2010. Spawning pe iod o Medi e anean ma ine ishes.
Re . Fish Biol. Fish. 20, 499–538. h ps://doi.o g/10.1007/s11160-010-9158-6
Tsikli as, A.C., Kou akis, E.T., Sylaios, G.K., Kallianio is, A.A., 2009. Summe dis ibu ion o ish
la ae in no he n Aegean Sea. J. Ma . Biol. Assoc. Uni ed Kingdom 89, 1137–1146.
h ps://doi.o g/10.1017/S0025315409000356
Va gas-Yáñez, M., Saba és, A., 2007. Mesoscale high- equency a iabili y in he Albo an Sea and i s
in luence on ish la ae dis ibu ions. J. Ma . Sys . 68, 421–438.
h ps://doi.o g/10.1016/j.jma sys.2007.01.004
2.2. Recu ence o he spa ial s uc u e in summe
94
Wang, D., Yao, L., Yu, J., Chen, P., Hu, R., 2021. Response o En i onmen al Fac o s o Spawning
G ound in he Pea l Ri e Es ua y, China. J. Ma . Sci. Eng. 9, 763.
h ps://doi.o g/h ps://doi.o g/10.3390/jmse9070763 Abs ac :
Wa d, J.H., 1963. Hie a chical G ouping o Op imize an Objec i e Func ion. J. Am. S a . Assoc. 58, 236–
244. h ps://doi.o g/h ps://doi.o g/10.1080/01621459.1963.10500845
Za ad, R., Alemany, F., Rod iguez, J.M., Ja boui, O., Lopez-Ju ado, J.L., Balbin, R., 2013. In luence o
summe condi ions on he la al ish assemblage in he eas e n coas o Tunisia (Ionian Sea,
Sou he n Medi e anean). J. Sea Res. 76, 114–125. h ps://doi.o g/10.1016/j.sea es.2012.08.001
Chap e 3
Long- e m a iabili y o la al ish communi y in he NW
Medi e anean in summe

2.3. Long- e m a iabili y
97
Long- e m a iabili y o la al ish communi y in he NW
Medi e anean in summe
Vanesa Raya, Ana Saba és
Abs ac
This wo k in es iga es he main changes expe ienced by he summe la al ish communi y o e h ee
decades along he Ca alan coas , an a ea cha ac e ized by a wide a ay o en i onmen al condi ions. The
s udy was based on nine ich hyoplank on su eys ca ied ou in June, July and Sep embe in h ee
decades, 1980s, 2000s and 2010s, co e ing he same a ea and applying he same sampling me hodology.
Changes in he en i onmen al a iables we e de ec ed h oughou he s udy pe iod. An inc ease in sea
su ace empe a u e, pa icula ly ma ked in June, and a dec ease in su ace chlo ophyll-a associa ed wi h
a dec ease in uno o he Eb o and Rhône i e s was obse ed. Ma ked changes in he composi ion and
abundance o he la al ish communi y we e also de ec ed be ween June and July in he 1980s and he
ollowing decades. These changes we e mainly due o he p esence o he i s ime in he a ea o wa m
wa e species, e.g. Thalassoma pa o and Ca anx honchus, o o hei inc ease in abundance, e.g.
Sa dinella au i a and Poma omus sal a ix, in he 2000s in ela ion o he no hwa d expansion o he
adul s' ange. The p esence o la ae o wa m wa e species in he 2000s and 2010s con ibu ed o an
inc ease in speci ic ichness in he la al ish communi y compa ed o alues ob ained in he 1980s. O he
species showed a decline in abundance o e ime, p obably due o a dec ease in su ace chlo ophyll-a,
e.g. Eng aulis enc asicolus al hough o e exploi a ion is also an impo an ac o o conside . La ae o
o he species, such as mesopelagic ish, did no show changes in abundance o e he h ee decades. In a
u u e scena io o inc easing sea wa e empe a u e and ma ine hea wa es e en s, enhanced s a i ica ion,
and dec easing su ace p ima y p oduc ion, he changes ha may occu in he ish la ae communi y will
la gely depend on he adap i e esponses o indi idual species o he new en i onmen al condi ions.
2.3. Long- e m a iabili y
98
2.3.1. In oduc ion
The e ec s o global wa ming on he dynamics o ma ine ecosys ems a e widesp ead wo ldwide. High
empe a u es al e oceanog aphic condi ions which can in luence ish li e-his o y ai s such as
ep oduc ion (e.g. Pe i gas e al., 2013) and g ow h (e.g. Pö ne and Peck, 2010). I is well documen ed
ha in ela ion o inc easing sea wa e empe a u e, ish communi ies ha e unde gone polewa d shi s in
la i udinal dis ibu ion (Cos a e al., 2021; Di Pane e al., 2020; Walsh e al., 2015). Howe e , o a
species o colonise a new egion, i mus success ully ep oduce and i s o sp ing mus su i e (Ba es e
al., 2014; Saba és e al., 2006). The spawning s a egies o ma ine ishes (e. g. iming and loca ion,
du a ion and equency o spawning) a e adap ed o he speci ic ea u es o he habi a s, wi h he esul
ha o sp ing is placed whe e la ae can ind he bes condi ions ha allow hem o success ul eed,
g ow h and su i al (Cowen and Sponaugle, 2009; Cushing, 1969; Siddon e al., 2011). Changes in ish
dis ibu ion would esul in changes in spawning g ounds and in he imes and en i onmen s in which ish
eggs and la ae occu (Hsieh e al., 2009; Llopiz e al., 2014). The la al ish communi ies, i.e.
occu ence, dis ibu ion, and abundance o ish la ae a e in luenced by a complex a ay o en i onmen al
p ocesses ha in e ac s wi h he biology o ish a di e en empo al and spa ial scales (Doyle e al.,
1993). F om a global app oach, changes in he la al ish communi y p o ide ea ly indica o s o shi s in
ish communi y dynamics (Doyle e al., 2009) and oceanog aphic and clima ic condi ions (Au h e al.,
2011; Ciannelli e al., 2022; McCla chie e al., 2018).
The Medi e anean Sea is one o he mos exposed a eas wo ldwide o he nume ous impac s o clima e
change, including one o he highes wa ming a es, and inc easing equency o ex eme wea he e en s
(Da ma aki e al., 2019; Jo dà e al., 2012; Macias e al., 2018). The sea wa ming is known o a ec
p ima y p oduc ion a global and egional scales by inc easing s a i ica ion o he sea, educing he inpu
o nu ien s in o su ace wa e s om mixing (Ba ale e al., 2008; Macias e al., 2018). The Medi e anean
is also a biodi e si y ho spo (Bianchi and Mo i, 2000; Coll e al., 2010), h ea ened by many
an h opogenic impac s, such as o e ishing, habi a deg ada ion, and in asi e species (e.g. Lejeusne e al.,
2009). These h ea s in e ac syne gis ically and can se e ely al e he balance o highly ulne able
Medi e anean ecosys ems, he eby a ec ing ma ine biodi e si y and ish popula ions (Lejeusne e al.,
2009; S e giou e al., 2016).
The Medi e anean ish auna is cha ac e ised by an unusually high biodi e si y o a empe a e sea
(Bianchi and Mo i, 2000; Quigna d and Tomasini, 2000). Species o sub opical o igin a e mainly ound
in he eas e n basin and he sou he n Medi e anean, whe e wa e empe a u e is highe han a e age
(Theocha is and Geo gopoulos, 1993). Cold- empe a e species inhabi he no he n coldes a eas (Gul o
Lion, Ligu ian Sea and no he n Ad ia ic: Bianchi and Mo i, 2000). The dis ibu ion o ish species in he
Medi e anean has unde gone signi ican changes in ecen decades in ela ion o he inc ease in seawa e
empe a u e. Species cha ac e is ic o he wa m wa e s o he sou he nmos pa s o he Medi e anean a e
expanding hei ange owa ds he no he nmos and coldes a eas, a p ocess known as me idionaliza ion
2.3. Long- e m a iabili y
99
(e. g. Azzu o, 2008; Bianchi and Mo i, 2003; F ancou e al., 1994; Lejeusne e al., 2009). These
dis ibu ion changes do no occu exclusi ely among ish, bu ha e also been desc ibed in o he
axonomic g oups, such as algae, c us acea and cnida ia (Gue e o e al., 2018; Pa a icini e al., 2015).
In he NW Medi e anean, he p esence o ish la ae in he plank on shows a ma ked seasonali y ela ed
o he spawning cycles o he adul s (Saba és e al., 2007a). Mos ish species spawn du ing sp ing and
summe (Tsikli as e al., 2010), among hem ne i ic species (e.g. Spa idae, Lab idae, Mullidae,
Se anidae, Scomb idae), small and medium pelagics (e.g. Eng aulidae, Clupeidae, Ca angidae) and
oceanic species (e.g. Myc ophidae, Gonos oma idae). The summe ich hyoplank on communi y, hus, is
cha ac e ised by high speci ic di e si y in a pe iod when oligo ophy p e ails in he uppe wa e column.
The highe empe a u es o he summe pe iod a ou as e la al g ow h and hence a high p obabili y o
su i al (Ande son, 1988; Houde, 1987). The summe pe iod in he Medi e anean is cha ac e ized by a
s a i ied wa e column, wi h a ma ked he mocline ha limi s e ical mixing. Consequen ly, p ima y
p oduc ion emains concen a ed a he deep chlo ophyll maximum, a hin laye a he deepes le els o
he pho ic zone (Es ada, 1985). Su ace p oduc i i y is es ic ed o some coas al zones and o a eas
unde he in luence o uno wa e s, mainly he Rhône and he Eb o (Sala , 1996).
The s udy a ea, o he Ca alan coas , in he NW Medi e anean, is cha ac e ized by a ma ked la i udinal
g adien o su ace empe a u e. The no he n sec o , which is mo e di ec ly in luenced by s ong and cold
no he ly winds, is colde han cen al and sou he n pa s and a he mal on de elops a su ace,
pe pendicula o he coas , associa ed wi h he sou he n limi o he s ong no he lies (Saba és e al.,
2009). In he ligh o he no hwa d expansion and inc easing abundance o wa m wa e ish species in he
NW Medi e anean, he Ca alan coas , wi h a clea g adien o sea su ace empe a u e and wide a ay o
en i onmen al condi ions (Saiz e al., 2014; Sala , 1996), p o ides an ideal loca ion o analyse possible
changes in he ish la ae communi y in summe . The objec i es we e i) o iden i y he long- e m changes
in he species composi ion, abundance and dis ibu ion o he summe la al ish communi y om 1980s
o 2010s, and ii) o explo e he links be ween en i onmen al a iables and he la al ish communi y
ends wi hin he cu en con ex o clima e change in he Medi e anean.
2.3.2. Ma e ials and me hods
Da a sou ces
Ich hyoplank on da a came om biological-oceanog aphic esea ch p og ams conduc ed by he Ins i u e
o Ma ine Sciences (CSIC) in Ba celona (h p://www.icm.csic.es) in summe along he Ca alan coas
(NW Medi e anean) (Fig. 2.3.1): ARECES 1980s (June, July and Sep embe 1983), CACO 2000s (July
and Sep embe 2003; June and July 2004), and FISHJELLY 2010s (June 2011 and July 2012). All
su eys we e pe o med wi h he R/V Ga cia del Cid, he same a ea was co e ed ( om 40.15˚N o 42˚N
and om 0.5˚E o 3.5˚E), and he same sampling me hodology was applied. While 2–3 su eys pe
decade o e a hi y-yea pe iod may seem low, he sampling su eys encompass a wide mesoscale a ea,
2.3. Long- e m a iabili y
106
The analysis o he di e en me ics used o de ec possible changes in he la al ish communi y, showed
signi ican di e ences (p < 0.05) be ween he su eys. In gene al, he highes la al ish abundances (N)
we e ound in June and July in all h ee decades, wi h hose in Sep embe being sensibly lowe (Fig. 2.3.3
a). I is wo h no ing, howe e , he low abundances o la ae de ec ed in July 2012. The speci ic ichness
(S) showed he lowes alues in he 1980s, being signi ican ly di e en om la e yea s, and he highes
alues we e de ec ed in June 2004 (Fig. 2.3.3 b). Bo h, he Shannon-Wiene biodi e si y index (H') and
he Pielou's e enness index (J') we e signi ican ly lowe in June and July 1983 (Fig. 2.3.3 c and d).
Fig. 2.3.3. Box plo s o (a) o al la al abundance (N), (b) species ichness (S, numbe o species, (c) Shannon-Wiene di e si y
index (H’, using a na u al loga i hm) and (d) Pielou’s measu e o e enness (J’) o each su ey. In each box plo , he da kened
line indica es he median; boxes show he in e qua ile ange and whiske s indica e he expec ed ex en o 99% o he da a o a
Gaussian dis ibu ion. The le e s a he bo om indica e he g oups iden i ied by he Tukey's pos hoc es .
Tempo al a iabili y
The wo-way clus e analysis classi ied he su eys in o i e clus e s (p < 0.05) sepa a ed in o wo main
b anches, wi h Sep embe su eys in one b anch and hose o June and July in he o he (Fig. 2.3.4). The
analysis also di ided he species in o wo b anches, which in u n included ou majo g oups (A o D)
acco ding o hei occu ence and abundance. The uppe b anch (g oups A and B) comp ised axa ha
occu ed in ew su eys and we e p esen in ela i ely low abundance (Fig. 2.3.4). G oup A included
species, such as Ci ha us lingua ula, Pagellus aca ne and Sco paena po cus, which we e almos
exclusi ely ound in Sep embe mon hs (Fig 2.3.4, able 2.3.1). The species ha occu ed in di e en
mon hs bu wi h low abundance, i.e. he coas al ish Ch omis ch omis and An hias an hias and he
myc ophyd Lampanyc us pusillus, o med g oup B.1. G oup B.2 was o med by species sligh ly mo e
abundan , such as Oblada melanu a and Lab idae, ha we e ound exclusi ely, o we e pa icula ly

2.3. Long- e m a iabili y
107
abundan in June 83, and Diplodus sp. and he mesopelagic ish A c ozenus isso in July 2012 (Fig 2.3.4,
Fig. 2.3.4. Two-way hie a chical clus e ing o su eys and hea map showing he abundance o ish la ae (colou g ada ion
om whi e o da k ed ep esen s om absence o highes abundance). G oups o species a e indica ed by colou ed boxes
labelled wi h a cod o le e s and numbe s, and su ey g oups a e indica ed by a di e en colou .
2.3. Long- e m a iabili y
108
abundan in June 83, and Diplodus sp. and he mesopelagic ish A c ozenus isso in July 2012 (Fig 2.3.4,
able 2.3.1). The lowe b anch (g oups C and D) comp ised ela i ely abundan axa. Wa m wa e species
such as Ca anx honchus, Thalassoma pa o and Sphy aena sphy aena o med g oup C.1 and we e
mainly p esen in he clus e assembling June 2004 and 2011 and July 2003 and 2004, being absen in all
mon hs o 1983 (Fig. 2.3.4, able 2.3.1). G oup C.2.1 included species such as T achinus d aco,
Poma omus sal a ix and Symphu us nigescens, which we e p esen in all su eys, wi h he excep ion o
June 1983. G oup C.2.2 we e composed by species such as T achu us achu us, Se anus cab illa, Co is
julis, and he myc ophid ish Myc ophum punc a um (Fig. 2.3.4, able 2.3.1), which we e ound in June
and July o all decades, being p ac ically absen in Sep embe su eys. G oup D included he 12 mos
common and abundan species. G oup D.1 was o med by species such as Auxis ochei, T achu us
medi e aneus and he myc ophids Ce a oscopelus made ensis and Lampanyc us c ocodilus, which
showed low abundances in Sep embe 83 and 2003. Species such as he small pelagics, Eng aulis
enc asicolus and Sa dinella au i a, and he mesopelagic ish Cyclo hone b aue i o med g oup D.2 and
we e p esen and abundan in all su eys, al hough S. au i a la ae we e absen in June 1983.
Table 2.3.2. Summa y o he edundancy analysis (RDA) esul s. The indi idual a iables a e o de ed by he pe cen age o
a iance explained (Exp. Va .), he signi icance o each a iable (P) oge he wi h i s es s a is ic (F- alue), and he in e -se
co ela ions be ween he en i onmen al a iables and he i s wo axes o he RDA.
Rega ding he RDA in ela ion o en i onmen al condi ions, he i s o dina ion axis explained 46.9% o
he cons ained a iance and was nega i ely co ela ed wi h dep h (Table 2.3.2). The second axis
explained 30.83% o he cons ained a iance and was nega i ely co ela ed wi h empe a u e (Table
2.3.2). The o dina ion o he sampling s a ions in he RDA plo showed ha s a ions sampled in he 1980s
we e mainly loca ed in he uppe pa o he plo , associa ed wi h low empe a u e and high chlo ophyll-a
alues, and hose o he 2000s in he lowe pa , ela ed o high empe a u e and low chlo ophyll-a (Fig.
2.3.5). The s a ions sampled in he 2010s we e loca ed in he cen al pa o he plo . Species o dina ion
showed ha mos o hem we e placed in he cen al pa o he plo (Fig. 2.3.5). Howe e , some species
exhibi ed a ela ionship wi h speci ic en i onmen al a iables. Thus, he mesopelagic ish Cyclo hone
b aue i and he myc ophids (e.g. Lampanyc us c ocodilus, Myc ophum punc a um) we e loca ed in he
le pa o he plo , associa ed wi h high dep h alues. Ano he g oup o species, such as Oblada
melanu a and T achu us achu us placed in he uppe pa o he plo , showed a ela ionship wi h low
2.3. Long- e m a iabili y
109
empe a u e and high chlo ophyll-a alues. Wa m wa e species such as S. au i a, T achu us
medi e aneus and Poma omus sal a ix we e loca ed in he lowe pa o he plo , associa ed wi h highe
empe a u es.
Fig. 2.3.5. O dina ion plo o he
edundancy analysis (RDA) showing
he ela ionships be ween he la al
ish axa, he sampling s a ions, and
he explana o y en i onmen al
a iables ( ec o s) o he summe
1980s, 2000s and 2010s. Blue
symbols co espond o he s a ions
sampled in he 1980s, ed symbols o
hose in he 2000s and yellow
symbols o hose in he 2010s.
Abb e ia ions o he axa names a e
de ailed in Table 2.3.1.
Dis ibu ion ends ac oss decades
A ma ked decline in he abundance o E. enc asicolus la ae was de ec ed om he 1980s o he 2010s
(Fig. 2.3.6). Th oughou he whole s udied pe iod, la ae o his species we e ound all along he coas ,
wi h wo main cen es o abundance, a he no he n hal o he a ea and o he sou h, on he Eb o Ri e
shel . I is wo h no ing he high abundance o la ae de ec ed in he no he n pa in June and July o he
1980s.
S. au i a la ae we e absen in June 1983, hei abundance in July 1983 was ex emely low and hei
dis ibu ion was limi ed o he sou he n hal o he a ea (Fig. 2.3.7). The abundance o his species
inc eased ma kedly in he 2000s, i s dis ibu ion ex ended along he en i e shel , eaching he no he n
limi o he s udy a ea. In he 2010s howe e , i s abundance no ably declined, showing a e y coas al
dis ibu ion ex ending all along he s udy a ea.
The empo al dis ibu ion o P. sal a ix la ae o e he h ee decades was qui e simila o ha o S. au i a
la ae (Fig. 2.3.8). Tha is, la ae we e absen in June 1983, in July 1983 hei dis ibu ion was limi ed o
he Eb o shel . In he 2000s, hei dis ibu ion ex ended o he whole s udy a ea, while in he 2010s, hei
abundance dec eased no ably, showing a mo e coas al dis ibu ion.
2.3. Long- e m a iabili y
110
Fig. 2.3.6. Mean abundance and spa ial dis ibu ion o Eng aulis enc asicolus la ae o e laid on ba hyme y (isoba hs shown: 50,
200, 1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s.
Fig. 2.3.7. Mean abundance and spa ial dis ibu ion o Sa dinella au i a la ae o e laid on ba hyme y (isoba hs shown: 50, 200,
1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s.
Ca anx honchus la ae we e absen in he 1980s (Fig. 2.3.9). In 2000s, la ae we e ela i ely abundan ,
pa icula ly in June 2004, wi h he highes abundances mainly in he sou h, on he Eb o shel . They
showed a ela i ely coas al dis ibu ion ex ending close o he no he n limi o he s udy a ea. Thei
abundance declined in he 2010s, simila o he pa e n de ec ed in S. au i a and P. sal a ix.
Co is julis la ae we e ela i ely abundan in he 1980s, pa icula ly in July (Fig. 2.3.10). Thei
abundance declined sligh ly in he 2000s and inc eased again in he 2010s, in con as o he pa e n
shown by he species men ioned abo e. In all he s udied yea s, hei dis ibu ion ex ended along he
en i e coas .
2.3. Long- e m a iabili y
111
Fig. 2.3.8. Mean abundance and spa ial dis ibu ion o Poma omus sal a ix la ae o e laid on ba hyme y (isoba hs shown: 50,
200, 1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s.
Fig. 2.3.9. Mean abundance and spa ial dis ibu ion o Ca anx honchus la ae o e laid on ba hyme y (isoba hs shown: 50, 200,
1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s. No e ha his species was absen in he 1980s.
Thalassoma pa o la ae, like hose o C. honchus, we e absen in he 1980s (Fig. 2.3.11). In he
ollowing wo decades, hei abundance was e y simila , being always mo e common in June han in
July, showing a sca e ed dis ibu ion h oughou he s udy a ea.
The abundance o Se anus cab illa la ae showed simila abundances in all h ee decades, al hough
always highe in June han in July (Fig. 2.3.12). Thei dis ibu ion was widesp ead h oughou he a ea.
M. punc a um la ae did no show signi ican changes in abundance o dis ibu ion o e he h ee decades.
La ae we e loca ed a he shel b eak h oughou he s udy a ea (Fig. 2.3.13).

2.3. Long- e m a iabili y
112
Fig. 2.3.10. Mean abundance and spa ial dis ibu ion o Co is julis la ae o e laid on ba hyme y (isoba hs shown: 50, 200, 1000
and 2000 m) in June and July o he 1980s, 2000s and 2010s.
Fig. 2.3.11. Mean abundance and spa ial dis ibu ion o Thalassoma pa o la ae o e laid on ba hyme y (isoba hs shown: 50,
200, 1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s. No e ha his species was absen in he 1980s.
Fig. 2.3.12. Mean abundance and spa ial dis ibu ion o Se anus cab illa la ae o e laid on ba hyme y (isoba hs shown: 50,
200, 1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s.
2.3. Long- e m a iabili y
113
Fig. 2.3.13. Mean abundance and spa ial dis ibu ion o Myc ophum punc a um la ae o e laid on ba hyme y (isoba hs shown:
50, 200, 1000 and 2000 m) in June and July o he 1980s, 2000s and 2010s.
2.3.4. Discussion
In he p esen s udy, impo an changes in he ich hyoplank on communi y ha e been de ec ed o e he
h ee decades o he s udy pe iod, wi h a majo shi occu ing be ween he 1980s and he 2000s. While
hese changes may be he esul o he in e ac ion o mul iple ac o s, bo h abio ic and bio ic, empe a u e
has been iden i ied as he key d i e o hese changes, in luencing bo h adul ish popula ions and la al
s ages (e.g. Houde, 1989; Peck e al., 2012; Va y e al., 2023). The esul s o he di e en analyses
pe o med showed ha la ae o some wa m wa e ish species ha we e no p esen in he 1980s we e
de ec ed om he 2000s onwa ds. O he wa m wa e species inc eased in abundance and hei dis ibu ion
sp ead no hwa ds.
Changes in he alues o he en i onmen al a iables we e de ec ed along he h ee decades o he s udy
pe iod. The e was a empe a u e inc ease, pa icula ly ma ked o he June su eys. This is in acco dance
wi h he wa ming end documen ed in he Medi e anean (Skli is e al., 2012; Va gas-Yáñez e al., 2009),
wi h he mos signi ican empe a u e inc ease eco ded in he sp ing and summe mon hs (Sala e al.,
2019; Skli is e al., 2012; Yan and Tang, 2021). Ex emely high empe a u es we e de ec ed in July 2003
as a esul o he successi e hea wa es ha a ec ed sou h-wes e n Eu ope du ing sp ing and summe o
ha yea (Black e al., 2004; Lu e bache e al., 2004). Clima ic models p edic apid wa ming in he
Medi e anean along wi h a g ea e occu ence o ma ine hea wa e e en s (Da ma aki e al., 2019; Oli e
e al., 2019; Pas o and Khodaya , 2023; Smale e al., 2019).
Su ace salini y and chlo ophyll-a showed opposi e ends, inc easing and dec easing espec i ely, along
he h ee decades. The low salini y alues de ec ed in he 1980s would be ela ed o he high i e uno
wa e s, mainly he Eb o and he Rhône (Saba és, 1990) ha show a dec easing end o e ime (Cozzi e
al., 2018). In he NW Medi e anean, du ing he s a i ica ion pe iod, i e uno is he main mechanism
2.3. Long- e m a iabili y
114
ha can supply nu ien s o he su ace, and a good co espondence be ween low salini y and high
chlo ophyll-a a su ace is cha ac e is ic o his pe iod (Masó e al., 1998; Saba és e al., 2013, 2009). A
dec ease in chlo ophyll-a concen a ion has also been documen ed in ela ion o he dec ease in i e
uno wa e s, accompanied by a dec ease in nu ien s concen a ion (Colella e al., 2016; Cozzi e al.,
2018; Salgado-He nanz e al., 2022) and an in ensi ica ion o wa e column s a i ica ion due o sea
empe a u e inc ease (Coma e al., 2009; Du ieu de Mad on e al., 2011; Somo e al., 2006).
The clus e analysis showed ma ked changes in he composi ion and abundance o he la al ish
communi y be ween June and July in he 1980s and he ollowing decades. The RDA indica ed ha
empe a u e was he main en i onmen al a iable con ibu ing o hese changes. This analysis also
showed a clea ela ionship o wa m wa e species wi h empe a u e, which we e de ec ed o he i s
ime, o exhibi ed a ma ked inc ease in abundance in he 2000s, du ing he wa mes yea s o he s udied
pe iod. The p esence o la ae o wa m wa e species in he 2000s and 2010s, no de ec ed in he 1980s,
e.g. he ca angids, T achino us o a us, Ca anx honchus and he coas al ben hic ish, Thalassoma pa o, is
in ag eemen wi h he eco ds o sou he n ish species in he colde a eas o he NW Medi e anean om
he la e 1980s (Bianchi e al., 2019; F ancou e al., 1994; Lejeusne e al., 2009). Al hough sea wa ming
may be esponsible o hese dis ibu ion changes, species’ esponse o clima e change in ol es
in e ac ions wi h a ious bio ic and abio ic condi ions (Ba es e al., 2014) and he e o e o he ac o s, such
as he es ablishmen o ma ine ese es and he subsequen educ ion o p ohibi ion o ishing in such
a eas, may ha e a ou ed an inc ease in he popula ions o some he mophilic species (Bodilis e al.,
2003). In he case o ca angids, he no hwa d expansion o wa m wa e species (i.e. T achu us
medi e aneus, C. honchus, Ca anx c ysos, T. o a us and Se iola dume ili) could also be ela ed o he
inc ease o loa ing objec s and jelly ish in he Medi e anean, as hei ju enile s ages ha e been
equen ly ound associa ed wi h ish agg ega ing de ices, d i algae and jelly ish (Psomadakis e al.,
2011; Raya and Saba és, 2015; Til es e al., 2018). T. pa o is a coas al wa m wa e ish con ined be o e
1980s o he sou he n pa s o he Medi e anean coas s. In he 1980s, la ae o his species we e absen in
he s udy a ea, hei abundance inc easing in he ollowing decades. As commen ed abo e, his species
was de ec ed in he mid-1990s in he no he n sec o o he Medi e anean (Bianchi, 2007; Milazzo e al.,
2011). T. pa o coexis s wi h he sympa ic species C. julis ha ole a es coole wa e s and is ound along
he no he n coas o he Medi e anean (Milazzo e al., 2011). La ae o C. julis we e abundan in he
1980s and declined sligh ly in he 2000s. This decline would sugges a g ea e adap i e ad an age o T.
pa o, which ‘won’ he compe i ion o space o e C. julis (Milazzo e al., 2013). While sea wa ming can
modi y in e speci ic in e ac ions be ween simila ecological species (Milazzo e al., 2013), ou esul s
show ha la ae o bo h species coexis in space and ime while i emains o be con i med how he
in e ac ion be ween he wo species will e ol e in he nea u u e.
La ae o o he species, such as S. au i a and P. sal a ix, inc eased ma kedly in abundance be ween he
1980s and he 2000s, and hei dis ibu ion ex ended u he no h (whe e empe a u es a e coole ). This is
2.3. Long- e m a iabili y
115
in acco dance wi h he inc easing abundance and g adual no hwa d expansion o adul s o hese species
along wes e n Medi e anean coas s in ela ion o sea wa ming (Saba és e al., 2012, 2006). The
expansion, coloniza ion, and inc ease in abundance o na i e wa m wa e species in o he colde no h
Medi e anean egions is an expec ed ou come o me idionaliza ion (Azzu o e al., 2019; Llo e e al.,
2015). The p esence o la ae o wa m wa e species in he s udy a ea om he 2000s onwa ds indica es
ha hey ep oduce success ully and con i ms he es ablishmen o hese species in hei new dis ibu ion
anges (Saba és e al., 2006).
Howe e , i is impo an o highligh he low abundance o la ae o mos species de ec ed in July 2003
(Table 2.3.1), du ing he ma ine hea wa e. This sugges s ha la ae would be nega i ely a ec ed by
high empe a u es eached (~27ºC), p obably being abo e he physiological op imum o hei g ow h and
su i al (Maynou e al., 2014; Takasuka e al., 2008). Rising empe a u es, pa icula ly du ing hea
wa es, can ha e di e en impac s on ecosys ems and, consequen ly, in he s uc u e la al ish
assemblages (Nielsen, 2020). I is expec ed ha in he nea u u e, in a scena io o inc eased ma ine
hea wa es (Da ma aki e al., 2019; Oli e e al., 2019; Pas o and Khodaya , 2023; Smale e al., 2019)
ish la ae o mos species will no ind a ou able condi ions o hei g ow h and su i al.
La ae o some species, p esen in he 1980s, do no appea o ha e expe ienced majo changes in
abundance and dis ibu ion be ween he h ee decades (Table 2.3.1; Fig. 2.3.10 and Fig. 2.3.12). This is
he case o coas al ben hic species such as Se anus cab illa, Se anus hepa us, and Co is julis. These
species a e well es ablished in he a ea, ep oduce in sp ing-summe and a e no subjec o ishing
exploi a ion. The la ae o mesopelagic species, e.g. Myc ophum punc a um, Cyclo hone b aue i, and
A c ozenus isso also did no show hese changes be ween he h ee decades (Table 2.3.1; Fig. 2.3.13).
The oceanic habi a o hese species, as indica ed he RDA, would mean ha hey would be less
in luenced by he obse ed en i onmen al ends, inc easing empe a u e and dec easing i e uno ,
which a e mo e p onounced on he con inen al shel han in he open sea (Salgado-He nanz e al., 2022)
(Fig. 2.3.11).
The p esence o la ae o wa m-wa e species in he 2000s and 2010s con ibu ed o an inc ease o he
speci ic ichness (S) o he la al ish communi y compa ed o he ichness alues ob ained in 1980s (Fig.
2.3.3). In his con ex , some au ho s ha e al eady sugges ed ha he inc ease o wa m wa e species
would lead o a bio ic homogeniza ion o ish communi ies in he NW Medi e anean (Ben Rais Las am
and Mouillo , 2008). The Shannon-Wiene (H’) and Pielou’s (J) biodi e si y indices showed low alues
in he 1980s compa ed o he ollowing decades. These low biodi e si y alues would be ela ed o he
dominance o E. enc asicolus in he 1980s. The ma ked decline in he abundance o E. enc asicolus
la ae obse ed om he 1980s o he 2000s would be ela ed o he excessi e ishing p essu e on his
species (FAO, 2024). A dec easing end in ancho y popula ion has been epo ed in he NW
Medi e anean since he 1980s (FAO, 2023). Fishing exploi a ion esul s in unbalanced demog aphic
s uc u e o he small pelagic s ocks by p e e en ially emo ing la ge indi iduals (e.g. B osse e al.,
2.3. Long- e m a iabili y
122
Doyle, M.J., Picquelle, S.J., Mie , K.L., Spillane, M.C., Bond, N.A., 2009. La al ish abundance and
physical o cing in he Gul o Alaska, 1981-2003. P og. Oceanog . 80, 163–187.
h ps://doi.o g/10.1016/j.pocean.2009.03.002
Du ieu de Mad on, X., Guieu, C., Sempé é, R., Conan, P., Cossa, D., D’O enzio, F., Es ou nel, C.,
Gazeau, F., Rabouille, C., S emmann, L., Bonne , S., Diaz, F., Koubbi, P., Radako i ch, O., Babin,
M., Baklou i, M., Bancon-Mon igny, C., Bel iso, S., Bensoussan, N., Bonsang, B., Bouloubassi, I.,
B une , C., Cadiou, J.F., Ca lo i, F., Chami, M., Cha masson, S., Cha iè e, B., Dachs, J., Doxa an,
D., Du ay, J.C., Elbaz-Pouliche , F., Eléaume, M., Ey olles, F., Fe nandez, C., Fowle , S., F ancou ,
P., Gae ne , J.C., Galzin, R., Gaspa ini, S., Ghiglione, J.F., Gonzalez, J.L., Goye , C., Guidi, L.,
Guizien, K., Heimbü ge , L.E., Jacque , S.H.M., Je ey, W.H., Joux, F., Le Hi , P., Leblanc, K.,
Le è e, D., Lejeusne, C., Lemé, R., Loÿe-Pilo , M.D., Malle , M., Méjanelle, L., Mélin, F., Mellon,
C., Mé igo , B., Me le, P.L., Migon, C., Mille , W.L., Mo ie , L., Mos aji , B., Mousseau, L.,
Mou in, T., Pa a, J., Pé ez, T., Pe enko, A., Poggiale, J.C., P ieu , L., Pujo-Pay, M., Pulido-Villena,
Raimbaul , P., Rees, A.P., Ridame, C., Ron ani, J.F., Ruiz Pino, D., Sic e, M.A., Taillandie , V.,
Tambu ini, C., Tanaka, T., Taupie -Le age, I., Tede i, M., Tes o , P., Thébaul , H., Thou enin, B.,
Tou a ie , F., T onczynski, J., Ulses, C., Van Wambeke, F., Van epo e, V., Vaz, S., Ve ney, R.,
2011. Ma ine ecosys ems’ esponses o clima ic and an h opogenic o cings in he Medi e anean.
P og. Oceanog . 91, 97–166. h ps://doi.o g/10.1016/j.pocean.2011.02.003
Es ada, M., 1985. Deep Phy oplank on and Chlo ophyll Maxima in he Wes e n Medi e anean, in:
Medi e anean Ma ine Ecosys ems. Sp inge US, Bos on, MA, pp. 247–277.
h ps://doi.o g/10.1007/978-1-4899-2248-9_12
FAO, 2024. The S a e o Wo ld Fishe ies and Aquacul u e 2024 – Blue T ans o ma ion in ac ion, Na u e
and Resou ces. Rome. h ps://doi.o g/h ps://doi.o g/10.4060/cd0683en
FAO, 2023. The S a e o Medi e anean and Black Sea Fishe ies 2023 – Special edi ion. Gene al
Fishe ies Commission o he Medi e anean. Rome.
Fe e -Maza, D., Llo e , J., Muñoz, M., Faliex, E., Vila, S., Sasal, P., 2016. Links be ween pa asi ism,
ene gy ese es and ecundi y o Eu opean ancho y, Eng aulis enc asicolus, in he no hwes e n
Medi e anean Sea. Conse . Physiol. 4, 1–13. h ps://doi.o g/10.1093/conphys/co 069
Feuilloley, G., F omen in, J.M., Sa aux, C., I isson, J.O., Jalabe , L., S emmann, L., 2022. Tempo al
luc ua ions in zooplank on size, abundance, and axonomic composi ion since 1995 in he No h
Wes e n Medi e anean Sea. ICES J. Ma . Sci. 79, 882–900. h ps://doi.o g/10.1093/icesjms/ sab190
F ancou , P., Boudou esque, C.F., Ha melin, J.G., Ha melin-Vi ien, M.L., Quigna d, J.P., 1994. A e he
Medi e anean wa e s becoming wa me ? In o ma ion om biological indica o s. Ma . Pollu . Bull.
28, 523–526. h ps://doi.o g/10.1016/0025-326X(94)90071-X

2.3. Long- e m a iabili y
123
F iedenbe g, L.E., Bollens, S.M., Rollwagen-Bollens, G., 2012. Feeding dynamics o la al Paci ic
he ing (Clupea pallasi) on na u al p ey assemblages: The impo ance o p o is s. Fish. Oceanog .
21, 95–108. h ps://doi.o g/10.1111/j.1365-2419.2011.00611.x
Fukami, K., Wa anabe, A., Fuji a, S., Yamaoka, K., Nishijima, T., 1999. P eda ion on naked p o ozoan
mic ozooplank on by ish la ae. Ma . Ecol. P og. Se . 185, 285–291.
h ps://doi.o g/10.3354/meps185285
Ganias, K., Soma akis, S., Nunes, C., 2014. - P og ess in Clupei o m Sys ema ics. Biol. Ecol. Sa dines
Ancho ies 16–55. h ps://doi.o g/10.1201/b16682-6
Ga cía-Comas, C., S emmann, L., Ibanez, F., Be line, L., Mazzocchi, M.G., Gaspa ini, S., Piche al, M.,
Go sky, G., 2011. Zooplank on long- e m changes in he NW Medi e anean Sea: Decadal
pe iodici y o ced by win e hyd og aphic condi ions ela ed o la ge-scale a mosphe ic changes? J.
Ma . Sys . 87, 216–226. h ps://doi.o g/10.1016/j.jma sys.2011.04.003
Genne , M.J., Halliday, N.C., Simpson, S.D., Sou hwa d, A.J., Hawkins, S.J., Sims, D.W., 2010.
Tempe a u e-d i en phenological changes wi hin a ma ine la al ish assemblage. J. Plank on Res.
32, 699–708. h ps://doi.o g/10.1093/plank / bp082
Golden So wa e, 2015. SURFER o windows. Con ou ing and 3-D su ace mapping.
Gue e o, E., Gili, J.M., G inyó, J., Raya, V., Saba és, A., 2018. Long- e m changes in he plank onic
cnida ian communi y in a mesoscale a ea o he NW Medi e anean. PLoS One 13.
h ps://doi.o g/10.1371/jou nal.pone.0196431
Houde, E.D., 1989. Compa a i e g ow h, mo ali y, and ene ge ics o ma ine ish la ae: empe a u e and
implied la i udinal e ec s. Fish. Bull. 87, 471–495.
Houde, E.D., 1987. Fish Ea ly Li e Dynamics and Rec ui men Va iabili y. Am. Fish. Soc. Symp. 2: 17-
29. h ps://doi.o g/ci elle
Hsieh, C., Reiss, C.S., Hun e , J.R., Bedding on, J.R., May, R.M., Sugiha a, G., 2006. Fishing ele a es
a iabili y in he abundance o exploi ed species. Le e s 443, 859–862.
Hsieh, C.H., Kim, H.J., Wa son, W., Di Lo enzo, E., Sugiha a, G., 2009. Clima e-d i en changes in
abundance and dis ibu ion o la ae o oceanic ishes in he sou he n Cali o nia egion. Glob.
Chang. Biol. 15, 2137–2152. h ps://doi.o g/10.1111/j.1365-2486.2009.01875.x
Jo dà, G., Ma bà, N., Dua e, C.M., 2012. Medi e anean seag ass ulne able o egional clima e
wa ming. Na . Clim. Chang. 2, 821–824.
La igne, H., D’O enzio, F., Ribe a D’Alcalà, M., Claus e, H., Sauzède, R., Gacic, M., 2015. On he
e ical dis ibu ion o he chlo ophyll a concen a ion in he Medi e anean Sea: A basin-scale and
seasonal app oach. Biogeosciences 12, 5021–5039. h ps://doi.o g/10.5194/bg-12-5021-2015
2.3. Long- e m a iabili y
124
Lejeusne, C., Che aldonné, P., Pe gen -Ma ini, C., Boudou esque, C.F., Pé ez, T., 2009. Clima e change
e ec s on a minia u e ocean: he highly di e se, highly impac ed Medi e anean Sea. T ends Ecol.
E ol. 25, 250–260. h ps://doi.o g/10.1016/j. ee.2009.10.009
Lleona , J., Maynou, F., 2003. Fish s ock assessmen s in he Medi e anean: S a e o he a . Sci. Ma . 67,
37–49. h ps://doi.o g/10.3989/scima .2003.67s137
Llopiz, J.K., Cowen, R.K., Hau , M.J., Ji, R., Munday, P.L., Muhling, B.A., Peck, M.A., Richa dson,
D.E., Soga d, S., Sponaugle, S., 2014. Ea ly li e his o y and ishe ies oceanog aphy new ques ions in
a changing wo ld. Oceanog aphy 27, 26–41. h ps://doi.o g/10.5670/oceanog.2014.84
Llo e , J., Palome a, I., Sala , J., Sole, I., 2004. Impac o eshwa e inpu and wind on landings o
ancho y (Eng aulis enc asicolus) and sa dine (Sa dina pilcha dus) in shel wa e s su ounding he
Eb e (Eb o) Ri e del a (no h-wes e n Medi e anean). Fish. Oceanog . 13, 102–110.
h ps://doi.o g/10.1046/j.1365-2419.2003.00279.x
Llo e , J., Saba és, A., Muñoz, M., Demes e, M., Solé, I., Fon , T., Casade all, M., Ma ín, P., Gómez,
S., 2015. How a mul idisciplina y app oach in ol ing e hnoecology, biology and ishe ies can help
explain he spa io- empo al changes in ma ine ish abundance esul ing om clima e change. Glob.
Ecol. Biogeog . 24, 448–461. h ps://doi.o g/10.1111/geb.12276
Lu e bache , J., Die ich, D., Xoplaki, E., G osjean, M., Wanne , H., 2004. Eu opean Seasonal and
Annual Tempe a u e Va iabili y, T ends, and Ex emes since 1500. Science (80-. ). 303, 1499–1503.
h ps://doi.o g/10.1126/science.1093877
Macias, D., Ga cia-Go iz, E., S ips, A., 2018. Deep win e con ec ion and phy oplank on dynamics in
he NW Medi e anean Sea unde p esen clima e and u u e (ho izon 2030) scena ios. Sci. Rep. 8,
1–15. h ps://doi.o g/10.1038/s41598-018-24965-0
Maechle , M., Rousseeuw, P., S uy , A., Hube , M., S ude , M., Roudie , P., González, J., Kozlowski,
K., Schube , E., Mu phy, K., 2022. Package “clus e ”: Finding G oups in Da a: Clus e Analysis
Ex ended. R package e sion 2.1.4.
Masó, M., Saba és, A., Pila Oli a , M., 1998. Sho - e m physical and biological a iabili y in he shel -
slope egion o he NW Medi e anean du ing he sp ing ansi ion pe iod. Con . Shel Res. 18, 661–
675. h ps://doi.o g/10.1016/S0278-4343(98)00011-9
Maynou, F., Saba és, A., Sala , J., 2014. Clues om he ecen pas o assess ec ui men o
Medi e anean small pelagic ishes unde sea wa ming scena ios. Clim. Change 126, 175–188.
h ps://doi.o g/10.1007/s10584-014-1194-0
McCla chie, S., Gao, J., D enka d, E.J., Thompson, A.R., Wa son, W., Ciannelli, L., Bog ad, S.J.,
Tho son, J.T., 2018. In e annual and Secula Va iabili y o La ae o Mesopelagic and Fo age
2.3. Long- e m a iabili y
125
Fishes in he Sou he n Cali o nia Cu en Sys em. J. Geophys. Res. Ocean. 123, 6277–6295.
h ps://doi.o g/10.1029/2018JC014011
Milazzo, M., Mi o, S., Domenici, P., G is ina, M., 2013. Clima e change exace ba es in e speci ic
in e ac ions in sympa ic coas al ishes. J. Anim. Ecol. 82, 468–477. h ps://doi.o g/10.1111/j.365-
2656.2012.02034.x
Milazzo, M., Palme i, A., Falcón, J.M., Badalamen i, F., Ga cìa-Cha on, J.A., Sinopoli, M., Chemello,
R., B i o, A., 2011. Ve ical dis ibu ion o wo sympa ic lab id ishes in he Wes e n Medi e anean
and Eas e n A lan ic ocky sub idal: Local sho e opog aphy does ma e . Ma . Ecol. 32, 521–531.
h ps://doi.o g/10.1111/j.1439-0485.2011.00447.x
Mo o e, E., Oli a , M.P., Pankhu s , P.M., Villa e, F., U ia e, I., 2008. T ophic ecology o bulle una
Auxis ochei la ae and on ogeny o eeding- ela ed o gans. Ma . Ecol. P og. Se . 353, 243–254.
h ps://doi.o g/10.3354/meps07206
Nielsen, J.M., Roge s, L.A., B odeu , R.D., Thompson, A.R., Au h, T.D., Dea y, A.L., Du y-Ande son,
J.T., Galb ai h, M., Koslow, J.A., Pe y, R.I., 2020. Responses o ich hyoplank on assemblages o
he ecen ma ine hea wa e and p e ious clima e luc ua ions in se e al No heas Paci ic ma ine
ecosys ems. Glob. Chang. Biol. 506–520. h ps://doi.o g/10.1111/gcb.15415
Oksanen, J., Simpson, G.L., Blanche , F.G., 2022. Package “ egan”: Communi y Ecology Package. R
package e sion 2.6–2.
Oli a , M.P., Emeliano , M., Villa e, F., U ia e, I., Maynou, F., Ál a ez, I., Mo o e, E., 2010. The ole
o oceanog aphic condi ions and plank on a ailabili y in la al ish assemblages o he Ca alan
coas (NW Medi e anean). Fish. Oceanog . 19, 209–229. h ps://doi.o g/10.1111/j.1365-
2419.2010.00538.x
Oli a , M.P., Saba és, A., 1997. Ve ical dis ibu ion o ish la ae in he no h-wes Medi e anean Sea in
sp ing. Ma . Biol. 129, 289–300. h ps://doi.o g/10.1007/s002270050169
Oli e , E.C.J., Bu ows, M.T., Dona , M.G., Sen Gup a, A., Alexande , L. V., Pe kins-Ki kpa ick, S.E.,
Ben huysen, J.A., Hobday, A.J., Holb ook, N.J., Moo e, P.J., Thomsen, M.S., We nbe g, T., Smale,
D.A., 2019. P ojec ed Ma ine Hea wa es in he 21s Cen u y and he Po en ial o Ecological
Impac . F on . Ma . Sci. 6, 1–12. h ps://doi.o g/10.3389/ ma s.2019.00734
Palome a, I., Oli a , M.P., Sala , J., Saba és, A., Coll, M., Ga cía, A., Mo ales-Nin, B., 2007. Small
pelagic ish in he NW Medi e anean Sea: An ecological e iew. P og. Oceanog . 74, 377–396.
h ps://doi.o g/10.1016/j.pocean.2007.04.012
Pa a icini, V., Mangialajo, L., Mousseau, L., Pei ano, A., Mo i, C., Mon e alcone, M., F ancou , P.,
Kulbicki, M., Bianchi, C.N., 2015. Clima e change and wa m-wa e species a he no h-wes e n
bounda y o he Medi e anean Sea. Ma . Ecol. 36, 897–909. h ps://doi.o g/10.1111/maec.12277
2.3. Long- e m a iabili y
126
Pas o , F., Khodaya , S., 2023. Ma ine hea wa es: Cha ac e izing a majo clima e impac in he
Medi e anean. Sci. To al En i on. 861, 160621. h ps://doi.o g/10.1016/j.sci o en .2022.160621
Peck, M.A., Huebe , K.B., Llopiz, J.K., 2012. In insic and Ex insic Fac o s D i ing Ma ch-Misma ch
Dynamics Du ing he Ea ly Li e His o y o Ma ine Fishes, 1s ed, Ad ances in Ecological Resea ch.
Else ie L d. h ps://doi.o g/10.1016/B978-0-12-398315-2.00003-X
Peebles, E.B., Hall, J.R., Tolley, S.G., 1996. Egg p oduc ion by he bay ancho y Anchoa mi chilli in
ela ion o adul and la al p ey ields. Ma . Ecol. P og. Se . 131, 61–73.
h ps://doi.o g/10.3354/meps131061
Pennino, M.G., Coll, M., Albo-Puigse e , M., Fe nández-Co edo , E., S eenbeek, J., Gi áldez, A.,
González, M., Es eban, A., Bellido, J.M., 2020. Cu en and Fu u e In luence o En i onmen al
Fac o s on Small Pelagic Fish Dis ibu ions in he No hwes e n Medi e anean Sea. F on . Ma . Sci.
7, 1–20. h ps://doi.o g/10.3389/ ma s.2020.00622
Pepin, P., Dowe , J.F., 2007. Va iabili y in he ophic posi ion o la al ish in a coas al pelagic
ecosys em based on s able iso ope analysis. J. Plank on Res. 29, 727–737.
h ps://doi.o g/10.1093/plank / bm052
Pe i gas, P., Rijnsdo p, A.D., Dickey-Collas, M., Engelha d, G.H., Peck, M.A., Pinnega , J.K.,
D inkwa e , K., Hu e , M., Nash, R.D.M., 2013. Impac s o clima e change on he complex li e
cycles o ish. Fish. Oceanog . 22, 121–139. h ps://doi.o g/10.1111/ og.12010
Pö ne , H.O., Peck, M.A., 2010. Clima e change e ec s on ishes and ishe ies: Towa ds a cause-and-
e ec unde s anding. J. Fish Biol. 77, 1745–1779. h ps://doi.o g/10.1111/j.1095-8649.2010.02783.x
Psomadakis, P.N., Ben i egna, F., Gius ino, S., T a aglini, A., Vacchi, M., 2011. No hwa d sp ead o
opical a ini y ishes: Ca anx c ysos (Teleos ea: Ca angidae), a case s udy om he Medi e anean
Sea. I al. J. Zool. 78, 113–123. h ps://doi.o g/10.1080/11250001003680933
Quigna d, J.P., Tomasini, J.A., 2000. Medi e anean ish biodi e si y. Biol. Ma . Medi . 7, 1–66.
R Co e Team, 2023. R: A Language and En i onmen o S a is ical Compu ing.
Raya, V., Saba és, A., 2015. Di e si y and dis ibu ion o ea ly li e s ages o ca angid ishes in he
no hwes e n Medi e anean: Responses o en i onmen al d i e s. Fish. Oceanog . 24, 118–134.
h ps://doi.o g/10.1111/ og.12097
Rossi, S., Saba és, A., La asa, M., Reyes, E., 2006. Lipid bioma ke s and ophic linkages be ween
phy oplank on, zooplank on and ancho y (Eng aulis enc asicolus) la ae in he NW Medi e anean.
J. Plank on Res. 28, 551–562. h ps://doi.o g/10.1093/plank / bi140
Saba és, A., 1990. Changes in he he e ogenei y o mesoscale dis ibu ion pa e ns o la al ish
associa ed wi h a shallow coas al haline on . Es ua . Coas . Shel Sci. 30, 131–140.
h ps://doi.o g/10.1016/0272-7714(90)90059-Z
2.3. Long- e m a iabili y
127
Saba és, A., Ma ín, P., Llo e , J., Raya, V., 2006. Sea wa ming and ish dis ibu ion: The case o he
small pelagic ish, Sa dinella au i a, in he Twes e n Medi e anean. Glob. Chang. Biol. 12, 2209–
2219. h ps://doi.o g/10.1111/j.1365-2486.2006.01246.x
Saba és, A., Ma ín, P., Raya, V., 2012. Changes in li e-his o y ai s in ela ion o clima e change:
blue ish (Poma omus sal a ix) in he no hwes e n Medi e anean. ICES J. Ma . Sci. 278, 1000–
1009. h ps://doi.o g/10.1093/icesjms/ ss053
Saba és, A., Oli a , M.P., Sala , J., Palome a, I., Alemany, F., 2007a. Physical and biological p ocesses
con olling he dis ibu ion o ish la ae in he NW Medi e anean. P og. Oceanog . 74, 355–376.
h ps://doi.o g/10.1016/j.pocean.2007.04.017
Saba és, A., Sala , J., Palome a, I., Emeliano , M., De Puelles, M.L.F., Oli a , M.P., 2007b. Ad ec ion o
ancho y (Eng aulis enc asicolus) la ae along he Ca alan con inen al slope (NW Medi e anean).
Fish. Oceanog . 16, 130–141. h ps://doi.o g/10.1111/j.1365-2419.2006.00416.x
Saba és, A., Sala , J., Raya, V., Emeliano , M., 2013. Role o mesoscale eddies in shaping he spa ial
dis ibu ion o he coexis ing Eng aulis enc asicolus and Sa dinella au i a la ae in he
no hwes e n Medi e anean. J. Ma . Sys . 111–112, 108–119.
h ps://doi.o g/10.1016/j.jma sys.2012.10.002
Saba és, A., Sala , J., Raya, V., Emeliano , M., Segu a-Nogue a, M., 2009. Spawning en i onmen al
condi ions o Sa dinella au i a a he no he n limi o i s dis ibu ion ange, he wes e n
Medi e anean. Ma . Ecol. P og. Se . 385, 227–236. h ps://doi.o g/10.3354/meps08058
Saba és, A., Za agoza, N., G au, C., Sala , J., 2008. Ve ical dis ibu ion o ea ly de elopmen al s ages in
wo coexis ing clupeoid species, Sa dinella au i a and Eng aulis enc asicolus. Ma . Ecol. P og. Se .
364, 169–180. h ps://doi.o g/10.3354/meps07461
Saba és, A., Za agoza, N., Raya, V., 2015. Dis ibu ion and eeding dynamics o la al ed mulle (Mullus
ba ba us) in he NW Medi e anean: The impo an ole o cladoce a. J. Plank on Res. 37, 820–833.
h ps://doi.o g/10.1093/plank / b 040
Saiz, E., Saba és, A., Gili, J.M., 2014. The zooplank on, in: Go edo, S., Dubinsky, Z. (Eds.), The
Medi e anean Sea: I s His o y and P esen Challenges. pp. 1–678. h ps://doi.o g/10.1007/978-94-
007-6704-1
Sala , J., 1996. Re iew o hyd og aphic en i onmen al ac o s ha may in luence ancho y habi a s in
no hwes e n Medi e anean. Sci. Ma . 60, 21–32.
Sala , J., Ga cia, M.A., C uzado, A., Palanques, A., A ín, L., Gomis, D., Guillén, J., De León, A.,
Puigde àb egas, J., Sosped a, J., Velásquez, Z.R., 2002. Seasonal changes o wa e mass s uc u e
and shel slope exchanges a he Eb o shel (NW Medi e anean). Con . Shel Res. 22, 327–348.
h ps://doi.o g/10.1016/S0278-4343(01)00031-0

2.3. Long- e m a iabili y
128
Sala , J., Pascual, J., Flexas, M., Chin, T.M., Vazquez-Cue o, J., 2019. Fo y- i e yea s o
oceanog aphic and me eo ological obse a ions a a coas al s a ion in he NW Medi e anean: a
g ound u h o sa elli e obse a ions. Ocean Dyn. 69, 1067–1084. h ps://doi.o g/10.1007/s10236-
019-01285-z
Salgado-He nanz, P.M., Regaudie-De-Gioux, A., An oine, D., Bas e e xea, G., 2022. Pelagic p ima y
p oduc ion in he coas al Medi e anean Sea: Va iabili y, ends, and con ibu ion o basin-scale
budge s. Biogeosciences 19, 47–69. h ps://doi.o g/10.5194/bg-19-47-2022
Siddon, E.C., Du y-Ande son, J.T., Mue e , F.J., 2011. Communi y-le el esponse o ish la ae o
en i onmen al a iabili y in he sou heas e n Be ing Sea. Ma . Ecol. P og. Se . 426, 225–239.
h ps://doi.o g/10.3354/meps09009
Siokou-F angou, I., Ch is aki, U., Mazzocchi, M.G., Mon eso , M., Ribe a D’Alcala, M., Vaque, D.,
Zingone, A., 2010. Plank on in he open medi e anean Sea: A e iew. Biogeosciences 7, 1543–
1586. h ps://doi.o g/10.5194/bg-7-1543-2010
Skli is, N., So ianos, S., Gkanasos, A., Man zia ou, A., Ve a is, V., Axaopoulos, P., Lasca a os, A.,
2012. Decadal scale a iabili y o sea su ace empe a u e in he Medi e anean Sea in ela ion o
a mosphe ic a iabili y. Ocean Dyn. 62, 13–30. h ps://doi.o g/10.1007/s10236-011-0493-5
Smale, D.A., We nbe g, T., Oli e , E.C.J., Thomsen, M., Ha ey, B.P., S aub, S.C., Bu ows, M.T.,
Alexande , L. V., Ben huysen, J.A., Dona , M.G., Feng, M., Hobday, A.J., Holb ook, N.J., Pe kins-
Ki kpa ick, S.E., Scannell, H.A., Sen Gup a, A., Payne, B.L., Moo e, P.J., 2019. Ma ine hea wa es
h ea en global biodi e si y and he p o ision o ecosys em se ices. Na . Clim. Chang. 9, 306–312.
h ps://doi.o g/10.1038/s41558-019-0412-1
Somo , S., Se aul , F., Déqué, M., 2006. T ansien clima e change scena io simula ion o he
Medi e anean Sea o he wen y- i s cen u y using a high- esolu ion ocean ci cula ion model.
Clim. Dyn. 27, 851–879. h ps://doi.o g/10.1007/s00382-006-0167-z
S e giou, K.I., Soma akis, S., T ian a yllou, G., Tsia as, K.P., Giannoulaki, M., Pe ihakis, G., Machias,
A., Tsikli as, A.C., 2016. T ends in p oduc i i y and biomass yields in he Medi e anean Sea La ge
Ma ine Ecosys em du ing clima e change. En i on. De . 17, 57–74.
h ps://doi.o g/10.1016/j.en de .2015.09.001
Takasuka, A., Oozeki, Y., Kubo a, H., 2008. Mul i-species egime shi s e lec ed in spawning
empe a u e op ima o small pelagic ish in he wes e n No h Paci ic. Ma . Ecol. P og. Se . 360,
211–217. h ps://doi.o g/10.3354/meps07407
Theocha is, A., Geo gopoulos, D., 1993. Dense wa e o ma ion o e he Samo h aki and Limnos
Pla eaux in he no h Aegean Sea (Eas e n Medi e anean Sea). Con . Shel Res. 13, 919–939.
h ps://doi.o g/10.1016/0278-4343(93)90017-R
2.3. Long- e m a iabili y
129
Til es, U., Saba és, A., Blázquez, M., Raya, V., Fuen es, V.L., 2018. Associa ions be ween ish and
jelly ish in he NW Medi e anean. Ma . Biol. 165. h ps://doi.o g/10.1007/s00227-018-3381-4
Tsikli as, A.C., An onopoulou, E., S e giou, K.I., 2010. Spawning pe iod o Medi e anean ma ine ishes.
Re . Fish Biol. Fish. 20, 499–538. h ps://doi.o g/10.1007/s11160-010-9158-6
Vand omme, P., S emmann, L., Be line, L., Gaspa ini, S., Mousseau, L., P ejge , F., Passa iume, O.,
Gua ini, J.M., Go sky, G., 2011. In e -annual luc ua ions o zooplank on communi ies in he Bay o
Ville anche-su -me om 1995 o 2005 (No he n Ligu ian Sea, F ance). Biogeosciences 8, 3143–
3158. h ps://doi.o g/10.5194/bg-8-3143-2011
Va gas-Yáñez, M., Moya, F., Tel, E., Ga cía-Ma ínez, M.C., Gue be , E., Bou geon, M., 2009. Wa ming
and sal ing in he wes e n Medi e anean du ing he second hal o he 20 h cen u y: Inconsis encies,
unknowns and he e ec o da a p ocessing. Sci. Ma . 73, 7–28.
h ps://doi.o g/10.3989/scima .2009.73n1007
Va y, L., Roge s, L., Ha e, M., Howa d, R., Ciannelli, L., 2023. Tempe a u e-dependen spawning
beha iou and la al he mohaline associa ions o Be ing Sea g ound ish. ICES J. Ma . Sci. 80,
1881–1898. h ps://doi.o g/10.1093/icesjms/ sad113
Walsh, H.J., Richa dson, D.E., Ma ancik, K.E., Ha e, J.A., 2015. Long- e m changes in he dis ibu ions
o la al and adul ish in he no heas U.S. shel ecosys em. PLoS One 10, 1–31.
h ps://doi.o g/10.1371/jou nal.pone.0137382
Wa d, J.H., 1963. Hie a chical G ouping o Op imize an Objec i e Func ion. J. Am. S a . Assoc. 58, 236–
244. h ps://doi.o g/h ps://doi.o g/10.1080/01621459.1963.10500845
Wa nes, G.R., Bolke , B., Bonebakke , L., Gen leman, R., Liaw, W.H.A., Lumley, T., Maechle , M.,
Magnusson, A., Moelle , S., Schwa z, M., 2016. gplo s: a ious R p og amming ools o plo ing
da a. R package e sion 3.0. 1. The Comp ehensi e R A chi e Ne wo k. Whi ney,.
Welke , M.T., Pie ce, C.L., Wahl, D.H., 1994. G ow h and Su i al o La al Fishes: Roles o
Compe i ion and Zooplank on Abundance. T ans. Am. Fish. Soc. 123, 703–717.
h ps://doi.o g/10.1577/1548-8659(1994)123<0703:gasol >2.3.co;2
Yan, X., Tang, Y., 2021. Mul idecadal Va iabili y in Medi e anean Sea Su ace Tempe a u e and I s
Sou ces. Geophys. Res. Le . 48, 1–9. h ps://doi.o g/10.1029/2020GL091814
Chap e 4
The Box-Balance Model: A new ool o assess ish la al
su i al, applied o ield da a on wo small pelagic ish