DOI: http://dx.doi.org/10.7306/gq.1457
Eco-biostratigraphic ad vances in late Qua ter nary geo chron ol ogy and palaeoclimate:
the mar ginal Gulf of Mex ico an a logue
Assimina ANTONARAKOU1, *, George KONTAKIOTIS1, Aristomenis P. KARAGEORGIS2, Eva BESIOU1, Stergios ZARKOGIANNIS1, Hara DRINIA1, Gra ham P. MORTYN3, 4 and Efthymis TRIPSANAS5
1 Na tional and Kapodistrian Uni ver sity of Ath ens, Fac ulty of Ge ol ogy and Geoenvironment, School of Earth Sci ences, De - part ment of His tor i cal Ge ol ogy-Pa le on tol ogy, Panepistimiopolis, Zografou, 15784, Greece.
2 Hel lenic Cen tre for Ma rine Re search, In sti tute of Ocean og ra phy, 46.7 km Ath ens-Sounio Av e nue, 19013 Anavyssos, Greece
3 Universitat Autànoma de Bar ce lona (UAB), In sti tute of En vi ron men tal Sci ence and Tech nol ogy (ICTA), Edifici Z Carrer de les Columnes, Bellaterra 08193, Spain
4 Universitat Autànoma de Bar ce lona (UAB), Faculty of Geography and His tory, Montalegre, 6, 08001 Bar ce lona, Spain
5 Hel lenic Pe tro leum Ex plo ra tion and Pro duc tion of Hy dro car bons S.A., 4A Gavrias, Marousi, Spain
Antonarakou, A., Kontakiotis, G., Karageorgis, A.P., Besiou, E., Zarkogiannis, S., Drinia, H., Mortyn, G.P., Tripsanas, E., 2019. Eco-biostratigraphic ad vances in late Qua ter nary geo chron ol ogy and palaeoclimate: the mar ginal Gulf of Mex ico an a - logue. Geo log i cal Quar terly, 63 (1): 178–191, doi: 10.7306/gq.1457
As so ci ate ed i tor: Wojciech Granoszewski
This study com bines high-res o lu tion plank tonic foraminiferal eco-biostratigraphy and palaeoclimatic data from the high-sed i - men ta tion-rate core JPC-26 from the north west ern mar gin of the Gulf of Mex ico (GoM). The eco-biozones rec og nized (GOMPFE1-12) be ing cor re lated with pub lished Mg/Ca-based sea sur face tem per a tures. This up dated palaeoclimatic and strati graphic ref er ence re cord fa cil i tates cor re la tions with the Green land ice core events and their cli ma tic re la tion ships, and also pro vides a solid strati graphic frame work for cor re la tions with other palaeoclimatic and palaeoceanographic re cords in the circum-GOM/Ca rib bean re gion. This multidisciplinary ap proach un der lines the util ity of sup port ing con ven tional dat ing meth od ol o gies with dif fer ent con straints, and fur ther re veals a pow er ful tool for re li ably cor re lat ing ma rine re cords be tween com pa ra ble deep-sea mar ginal set tings and co eval se quences of this re gion.
Key words: in te grated stra tig ra phy, Late Glacial–Holocene tran si tion, plank tonic foraminiferal eco-bioevents, deep-sea sed i - men tary cor re la tions, cli mate vari abil ity, palaeoceanography.
INTRODUCTION
In the last half-cen tury, con sid er able in ter est has de vel oped re gard ing the role of the (sub)trop i cal oceans in cli mate change, and in par tic u lar, oce anic sub-bas ins and mar ginal seas that are of ten more re spon sive to palaeoceanographic and palaeo - climatic changes than to broader global ocean bas ins, be cause of their smaller size and par tial iso la tion (Ma rino et al., 2009;
Kontakiotis et al., 2013; Kontakiotis, 2016). As an ex am ple, the trop i cal West ern Hemi sphere Warm Pool (WHWP), which en - com passes the Ca rib bean Sea and the Gulf of Mex ico (GoM), is an im por tant heat and mois ture source for cli mate in the
North At lan tic re gion (Wang and Enfield, 2001), and there fore acts as a key reg u la tor of sub po lar North At lan tic ocean og ra phy and cli mate in Eu rope. In par tic u lar, the GoM is a key area for global thermohaline cir cu la tion (THC), global heat ex change and cli mate sys tem (Broecker, 1991), be cause its hy dro graphic char ac ter is tics (high sea sur face tem per a ture (SST) and sa lin - ity (SSS) val ues) partly con trol Gulf Stream phys i cal prop er ties.
Its small vol ume, com pared with ma jor ocean bas ins, causes changes in its cli ma tic forc ing to be re corded vir tu ally in stan ta - neously in palaeoceanographic proxy data such as sta ble iso - topes, geo chem i cal ra tios, and microfossil abun dances. More - over, the di rect in flu ence of North At lan tic sur face wa ter on the fau nal as sem blages gives a mea sure of the close hy dro log i cal re la tion ship be tween the North At lan tic and the GoM wa ter masses, and un der lines the prom i nent role of the GoM in the un der stand ing of global cli ma tic evo lu tion (Thirumalai et al., 2018).
Due to its lat i tu di nal po si tion, the semi-en closed char ac ter, the high sed i men ta tion rates and close at mo spheric and oce -
* Cor re spond ing au thor, e-mail: aantonar@geol.uoa.gr Re ceived: November 2, 2018; ac cepted: January 31, 2019; first pub lished on line: April 1, 2019
anic con nec tions with the North At lan tic re gion, the GoM is among the most stud ied mar ginal bas ins. In par tic u lar, the in ter - play be tween Mis sis sippi River dis charge, the open ocean wa - ter masses and at mo spheric cir cu la tion cre ates a com plex and dy namic sys tem re spon si ble for hy dro log i cal fluc tu a tions on both an nual and long-term timescales. To better un der stand the im pact of these pa ram e ters on the re gional cli mate, a num ber of stud ies based on var i ous ma rine prox ies has fo cused on the hydrologic evo lu tion of the ba sin dur ing the late Qua ter nary (Metcalfe et al., 2000; Aharon, 2003, 2006; Poore et al., 2003;
Flower et al., 2004; LoDico et al., 2006; Nürnberg et al., 2008;
Ziegler et al., 2008; Richey et al., 2011, 2012; Spear et al., 2011a; Ras mus sen and Thom sen, 2012; Limoges et al., 2014;
del Monte-Luna et al., 2015; Antonarakou et al., 2015; Shakun et al., 2016; Reynolds et al., 2018). How ever, the re con struc tion of the vari abil ity of late Qua ter nary cli mate needs de tailed eco-litho-chronostratigraphic con trol (Triantaphyllou et al., 2009; Siani et al., 2010; Drinia et al., 2016; Koutrouli et al., 2018). In this re spect, no ta ble con tri bu tions are re quired from micropaleontological, biogeochemical, and sedimentological stud ies. In par tic u lar, mor pho log i cal and compositional chan - ges in the plank tonic foraminiferal as sem blages re corded in ma rine sed i ments and/or the wa ter col umn pro vide de tailed in - for ma tion on the phys i cal and chem i cal prop er ties of the sur - round ing wa ter masses and their re la tion ships with cli ma tic vari abil ity (Renaud and Schmidt, 2003; Schmidt et al., 2004a, b;
Spear et al., 2011b; Kontakiotis et al., 2011, 2013, 2016a;
Thirumalai et al., 2014; Antonarakou et al., 2015; Jonkers et al., 2015; Kontakiotis, 2016; Arellano-Torres and Machain-Castillo, 2017; Reynolds et al., 2018; Wil son et al., 2018). For in stance, it has been sug gested that plank tonic foraminifera al ter their shape and size, which de ter mine their weight and their over all den sity in di rect re sponse to changes in wa ter den sity and vis - cos ity to re tain or mod ify their hy dro dy namic be hav iour ac cord - ingly (Caromel et al., 2014; Zarkogiannis et al., 2019a, b). Un - der stand ing the driv ers of these changes, along with the im pli - ca tions for changes in biogeochemistry and cli mate, is key to un der stand ing the past hydroclimate. They also al low the def i ni - tion of a de tailed se quence of re gional eco-bioevents for the late Qua ter nary (e.g., tem po rary ap pear ance or dis ap pear ance, or sig nif i cant rel a tive abun dance vari a tions of spe cific taxa; Wil - son, 2012), re flect ing cli ma tic os cil la tions, that rep re sent an im - por tant tool for ac cu rate sub di vi sion of the late Qua ter nary strati graphic re cord and fa cil i tate cor re la tions be tween sites in the dif fer ent sub-bas ins. The pres ence of dis tinct tem po ral Qua ter nary or ganic-rich zones (ORZ; Tripsanas et al., 2013) in GoM suc ces sions rep re sents an ad di tional, in de pend ent tool for dat ing and cor re lat ing ma rine sed i men tary ar chives. They pro vide isochronous marker ho ri zons that can be cor re lated via geo chem i cal fin ger print ing with well-dated melt wa ter flood ing events (Tripsanas et al., 2007; Meckler et al., 2008; Sionneau et al., 2008; Montero-Serrano et al., 2009; Sionneau et al., 2010;
Vetter et al., 2017).
In this work we re fine and ex tend pre vi ous biostratigraphic schemes for the area in ves ti gated (Ken nett and Huddlestun, 1972; Ken nett et al., 1985; Flower and Ken nett, 1990; Mar tin et al., 1990) and fur ther high light the strati graphic re la tion ship be - tween foraminiferal and cli ma tic events since the late gla cial pe - riod. We re port a high-res o lu tion plank tonic foraminiferal quan - ti ta tive dis tri bu tion pat tern, and fur ther cor re late the eco- bioevents rec og nized with Mg/Ca-based SST re sults (Antonarakou et al., 2015). As a re sult, this up dated and com - pre hen sive re con struc tion can be con sid ered as a use ful
palaeoclimatic and strati graphic ref er ence re cord, fa cil i tat ing cor re la tions with the Green land ice core events and their cli ma - tic re la tion ships, and also pro vid ing a solid strati graphic frame - work for cor re la tions with other palaeoclimatic and palaeo - ceanographic re cords in the circum-GoM/Ca rib bean re gion.
OCEANOGRAPHIC SETTING
The GoM is a semi-en closed oce anic ba sin on the south - east ern mar gin of North Amer ica that cov ers an area of 1,555,000 km2 (Fig. 1). It is con nected to the At lan tic Ocean by the Florida Strait (FS) and to the Ca rib bean Sea by the Yucatan Chan nel (YC; INEGI, 2014). Its hy dro graphic prop er ties re sult mainly from the in ter play be tween the open ocean wa ter masses, Mis sis sippi River dis charge, and at mo spheric cir cu la - tion. The sur face cir cu la tion in the GoM is known to be a key el e - ment in the North At lan tic and the At lan tic Me rid i o nal Over turn ing Cir cu la tion (AMOC), as it con veys heat and salt from low to high lat i tudes via the Gulf Stream; it is largely in flu enced by: (1) the en er getic Loop Cur rent (LC) and its as so ci ated eddy field (Ohlmann et al., 2001; Morey et al., 2003b; Liu et al., 2012;
Gopalakrishnan et al., 2013; Schmitz et al., 2013; Weisberg and Liu, 2017) in the east ern part, and (2) a less prom i nent west-flow - ing an ti cy clonic gyre that moves on to the west ern part of the ba - sin, vary ing ap pre cia bly in in ten sity with sea son and lo ca tion (Behringer et al., 1977; Müller-Karger et al., 1991; Zavala-Hi - dalgo et al., 2006; Putrasahan et al., 2017). The most prom i nent sur face cir cu la tion fea ture in the GoM, the LC, brings warm and salty trop i cal wa ters orig i nat ing from the Ca rib bean Sea through the YC into the GoM, loops north west, and ex its into the North At lan tic Ocean through the Florida Straits (Florida Cur rent – FC), and even tu ally feeds the Gulf Stream (GS; Fig. 1; Elliott, 1982;
Blumberg and Mellor, 1985; Hofmann and Worley, 1986;
Jochens and DiMarco, 2008; Auladell et al., 2010; Oey et al., 2013). The lat i tu di nal ex ten sion of the Ca rib bean wa ter in flow (Ca rib bean Cur rent, CC) into the ba sin is sea son ally mod u lated by changes in the po si tion of the Intertropical Con ver gence Zone (ITCZ; Poore et al., 2004, 2005; Nürnberg et al., 2008; Ziegler et al., 2008). North ward mi gra tion of the ITCZ dur ing the bo real sum mer causes the LC to prop a gate far ther north, in flu enc ing the hy dro log i cal prop er ties of the en tire ba sin (tem per a ture, sa - lin ity, wind di rec tion), and fi nally cre ates drier con di tions dur ing the bo real win ter and rainy con di tions dur ing the bo real sum mer (Saha, 2010). Apart from the dis tinct role of the LC, flu vial fresh - wa ter dis charge (mainly the Mis sis sippi River sys tem) di rectly af - fected by the pre vail ing pre cip i ta tion pat terns and cli ma tic con di - tions (Montero-Serrano et al., 2010, 2011) also in flu ences the GoM sur face hydrography, by mod i fy ing the SSS, and by im - pact ing to some ex tent the thermocline cir cu la tion via the GS (Aharon, 2003; Morey et al., 2003a; Flower et al., 2004). More - over, the large sea sonal load ings of fresh wa ter, in or ganic and or - ganic mat ter, and nu tri ents main tain the north ern GoM as an ac - tive eco sys tem with dy namic phys i cal and biogeochemical pro - cesses, such as in ter ac tions of eddy cy clone-an ti cy clone pairs with the wa ter col umn, bot tom to pog ra phy, in creased off shore pro duc tiv ity, and advection of shelf ma te rial into the cen tral Gulf (Biggs and Müller-Karger, 1994; Toner et al., 2003; Wawrik et al., 2003; Corbett et al., 2004; Luo et al., 2016; Chen and Hu, 2017).
In the mod ern cli mate sys tem, GoM mean an nual SST is
~25.85°C with large sea sonal am pli tude (~8.5°C), whereas sa - lin ity var ies from about 34.4 to 36.3 psu (Antonov et al., 2010;
Locarnini et al., 2010).
MATERIAL AND METHODS
CORE SELECTION
Core JPC-26 (1995 m b.s.l.) was re cov ered from the north - west ern con ti nen tal slope of the GoM (26°22.44’N, 92°01.62’W) at a depth of 1995 m (Fig. 1), dur ing the R/V Knorr 1998 cruise, us ing the Woods Hole Ocean o graphic In sti tu tion jumbo pis ton corer. The core is lo cated off shore of the for mer mouth of the Mis sis sippi River within the Bryant Can yon area, which is rep re sented by a net work of intraslope bas ins sep a - rated from each other by sills/pla teaus (Tripsanas et al., 2007).
The up per 3 m of the core ana lysed rep re sents an un in ter - rupted last gla cial to Ho lo cene in ter val, in clud ing the Last Gla - cial Max i mum (LGM) event, and the last gla cial ter mi na tion pe - riod (T1). This core is also char ac ter ized by a Ho lo cene thick - ness of 1.5 m that can of fer a time res o lu tion on the or der of a few hun dred years per sam ple step. The Late Qua ter nary pro - vides the best time in ter val for this re search be cause sea level his tory is well con strained tem po rally; more over, cen ten nial to mil len nial scale chronostratigraphic res o lu tion is highly achiev - able through plank tonic bio-ecozonation, while palaeoclimate his tory is best con strained through pre vi ously in te grated sedimento logical, micropalaeontological, strati graphic and palaeoclimatic stud ies (Poore et al., 2003; LoDico et al., 2006;
Montero- Serrano et al., 2009, 2010; Wil liams et al., 2010; Ras - mus sen and Thom sen, 2012; Tripsanas et al., 2013; Limoges et al., 2014; Antonarakou et al., 2015; Arellano-Torres and Machain-Castillo, 2017).
LITHO-CHRONOSTRATIGRAPHY OF MARINE SEDIMENT CORE JPC-26
The sed i men tary re cord of the up per most 3 m of JPC-26 was sam pled at 2–5 cm in ter vals, and con sists of three dis tinct lithostratigraphic units (I–III). The up per unit I (0–1.52 m) con - sists of red dish-brown to pale yel low hemipelagic mud, and the spo radic pres ence of man ga nese-rich, dark brown mud ho ri - zons (3 cm thick; in in ter vals 0.10–0.32 m and 0.45–0.60 m). In unit II (1.52–1.90 m), hemipelagic muds were in ter ca lated with three or ganic-rich zones (0.60–0.95% Corg) slightly (ORZ I) or in tensely (ORZ II, III) bioturbated (Tripsanas et al., 2007). Their downcore tran si tion to pale green ish-grey hemipelagic muds of unit III (1.90–3 m) is char ac ter ized by the pres ence of two or - ganic-rich (0.75–1.30% Corg) red dish-brown (RT) and dark green (GT) mud turbidites. The chronostratigraphy of JPC-26 is based on eight ac cel er a tor mass spec trom e try ra dio car bon (AMS 14C) dates, sup ple mented by ad di tional tie-points de rived from the de ter mi na tion of both the iso to pic and fau nal pro file (Antonarakou et al., 2015).
MICROPALAEONTOLOGICAL ANALYSES
For de tailed quan ti ta tive micropalaeontological anal y sis, we picked spec i mens of plank tonic foraminifera from about 15 g of dry ma rine sed i ment, taken ev ery 2–5 cm along the core. All the sed i ment sam ples were washed with run ning wa ter through a 63 µm sieve. Af ter wash ing and dry ing at 50°C, the sam ples were sieved and the >125 µm size frac tion was split with an Otto mi cro-split ter into aliquots, from which at least 300 spec i mens Fig. 1. Bathymetric map of the Gulf of Mex ico (GoM) show ing the core lo ca tion and the main pat terns
of the sur face cir cu la tion (white ar rows, solid cir cles)
The top-right in set shows the names of the main cur rents, as de scribed in the text
of plank tonic foraminifera were picked, iden ti fied and counted us ing morphospecies-spe cific and/or spe cial ized tax o nomic ref er ences (Ken nett and Srinivasan, 1983; Hemleben et al., 1989; Kontakiotis et al., 2017). The num ber of plank tonic foraminifera counted is sta tis ti cally re li able for both eco- biostratigraphic and palaeoclimatic re con struc tions (Pat ter son and Fishbein, 1989; Capotondi et al., 1999; Drinia and Antonarakou, 2003; Sprovieri et al., 2003; Sbaffi et al., 2004;
Budillon et al., 2009; Triantaphyllou et al., 2009; Siani et al., 2010; Drinia et al., 2016; Bonfardeci et al., 2018). All the spec i - mens iden ti fied showed very good lev els of pres er va tion, with no traces of dis so lu tion or dam age found in the tests. Raw data were trans formed into per cent ages of the to tal abun dance and rel a tive per cent age abun dance curves were plot ted, af ter ex - clu sion of rare spe cies (<3%; Globigerinoides conglobatus, Globigerinella calida, Globigerinoides tenellus, Hastigerina pelagica, Globorotalia ungulata) and group ing of spe cies with phylo gen etic af fin i ties and sim i lar eco log i cal char ac ter is tics (e.g., Globigerina bulloides and Globigerina falconensis were grouped in G. bulloides gr., Globigerinoides trilobus and Globigerinoides sacculifer in G. sacculifer gr., Globigerinella aequilateralis and Globigerinella calida in the G. aequilateralis gr., Globorotalia menardii and Globorotalia tumida in G. menardii gr., and Neogloboquadrina pachyderma and Neogloboquadrina dutertrei in neogloboquadrinids re spec - tively). The plank tonic foraminiferal rel a tive dis tri bu tions are also used as a first-or der es ti mate of SST vari a tions. An in dex
of the SST vari a tions was con structed based on the down-core vari a tion of plank tonic foraminiferal abun dances, and it was re - ferred to as the Plank tonic Palaeoclimatic Curve (PPC). The PPC was ob tained by the for mula (w – c) ´ 100/(w + c), where:
w rep re sents the warm-wa ter in di ca tors and c the cold wa ter in - di ca tors (Rohling et al., 1993; Kontakiotis, 2016). For palaeoclimatic in ter pre ta tion, the PPC re cord was fur ther com - pared with pre vi ously pub lished SST proxy data (d18O, Mg/Ca) from the same or nearby cores (Flower et al., 2004; Ziegler et al., 2008; Wil liams et al., 2010; Ras mus sen and Thom sen, 2012; Schmidt et al., 2012; Tripsanas et al., 2013; Antonarakou et al., 2015; Arellano-Torres and Machain-Castillo, 2017).
RESULTS
SEDIMENTATION RATES
Within the GoM, Bryant Can yon is ex cep tional in that rapid sed i men ta tion sus tained by high rates of terrigenous sed i ment de liv ery and sur face ma rine pro duc tiv ity makes this ba sin an ex cel lent re corder of palaeoenvironmental and palaeoclimatic con di tions. Sed i men ta tion in the ba sin is mainly hemipelagic and is strongly linked to the Mis sis sippi River de tri tal sup ply (Bouma et al., 1985). This type of de po si tion is due to the rise of sea level that led to the con fine ment of river-sourced sed i ments to the con ti nen tal shelf, and the prev a lence of hemipelagic sed i - Fig. 2. High-res o lu tion SEM pho to mi cro graphs of the plank tonic foraminiferal spe cies most in dic a tive for the pres ent
study in core JPC-26
A – Orbulina universa; B – Sphaeroidinella dehiscens; C – Globogerinella aequilateralis; D – Globigerina bulloides; E – Globigerinoides ruber sensu stricto; F – Globigerinoides ruber sensu lato; G – Globigerinoides sacculifer; H – Neogloboquadrina dutertrei; I – Pulleniatina obliquiloculata; J – Globorotalia inflata; K – Globorotalia crassaformis; L – Globorotalia scitula; M – Globorotalia menardii; N – Globorotalia tumida; O – Globorotalia truncatulinoides
men ta tion over the con ti nen tal slope of the north west ern GoM.
The sed i ment re cord here in ves ti gated spans the in ter val from the LGM and the sub se quent tran si tion (T1) into the late Ho lo - cene (i.e. 0–21 ka; Antonarakou et al., 2015) with an av er age sed i men ta tion rate (SR) of 13.5 cm/ky, which is in agree ment with prior stud ies (Leventer et al., 1983; Aharon, 2003; Poore et al., 2003; LoDico et al., 2006; Tripsanas et al., 2007). Sed i men - ta tion rates of each unit re veal a strong down-core in creas ing trend, with the higher SR oc cur ring dur ing the gla cial sec tion (unit III; 18.18 cm/ky) and the deglacial melt wa ter pulses (unit II;
13.29 cm/ky). Lower SR oc curred dur ing the Ho lo cene in ter val (unit I; 11.44 cm/ky) and is con sis tent with other ob ser va tions of re duced sed i men ta tion rates af ter 13 cal ka BP (Wil liams and Kohl, 1986; Montero-Ser rano et al., 2009).
PLANKTONIC FORAMINIFERAL FAUNAL DISTRIBUTION
Qual i ta tive anal y sis of the plank tonic foraminifera al lows us to iden tify 24 spe cies lumped into 15 groups: Globigerinoides ruber alba, Globigerinoides ruber rosea, Globigerinoides sacculifer gr., Globigerina bulloides gr., Globogerinella aequi - lateralis gr., Globorotalia scitula, Globorotalia menardii gr., Orbulina universa, neogloboquadrinids, Globorotalia cras sa - formis, Globorotalia inflata, Globorotalia truncatulinoides, Pulleniatina obliquiloculata, and Sphaeroidinella dehiscens (Fig. 2). The strati graphic dis tri bu tion of the most abun dant spe - cies and spe cies groups over the last 21 kyr BP is shown in Fig - ure 3. The over all changes in the plank tonic foraminiferal abun - dances from JPC-26 are con sis tent with pre vi ously doc u - mented changes in the GoM foraminiferal as sem blages re lated to the tran si tion from gla cial to inter gla cial con di tions (e.g., Flower and Ken nett, 1995; Poore et al., 2003).
Globigerinoides ruber, G. sacculifer gr., neogloboqua - drinids, O. universa, G. crassaformis, G. menardii gr., and G. truncatulinoides are the most abun dant spe cies. With the ex - cep tion of the rep re sen ta tives of G. menardii gr., that are vir tu - ally ab sent from the base of the suc ces sion up to 1.36 m, they have an al most con tin u ous dis tri bu tion through out the in ter val stud ied. Globigerinoides ruber was the dom i nant spe cies over the en tire du ra tion (33.2% on av er age). The white (G. ruber alba) and pink (G. ruber rosea) va ri et ies of this spe cies were counted as sep a rate cat e go ries, show ing that the white va ri ety is about 15 to 30% more abun dant than the pink va ri ety (Fig. 3).
The spe cies G. menardii gr., P. obliquiloculata, G. sacculifer gr., and G. truncatulinoides dis play an up ward in crease in abun - dance. A spo radic dis tri bu tion is re corded for G. bulloides and G. inflata, since they are sig nif i cant con tri bu tors only in the basal part of the core with an av er age con tri bu tion of ~10–25%.
The lat ter ex ists in the fau nal com po si tion up to 1.40 m and then com pletely dis ap pears, while the for mer, al though it does not dis ap pear from the se quence, re duces its pop u la tions to <5% in the up per half of the core. Globorotalia scitula pres ents a dis - con tin u ous dis tri bu tion and shows mi nor abun dance peaks and/or in ter vals of oc cur rence lim ited mainly to the mid dle part of the core. The dis tri bu tion pat tern of neogloboquadrinids is char ac ter ized by high per cent ages in the up per half of the core, with their max i mum per cent ages oc cur ring dur ing the de po si - tion of or ganic-rich sed i ments and through out the Ho lo cene.
PLANKTONIC PALAEOCLIMATIC CURVE
From the base of the core up to 2.31 m (19.0 ka), the low est val ues of the PPC (~20–40%) are in dic a tive of the gla cial pe - riod (Fig. 4). This in ter val is char ac ter ized by the dom i nance of cold wa ter in di ca tors, such as G. inflata, G. crassaformis,
G. bullo ides and neogloboquadrinids, whereas warm wa ter spe cies are re stricted in these fau nal as sem blages. The palaeoclimatic curve shows rel a tive high vari abil ity upcore from 2.31–1.52 m (19.0–13.0 ka), dis play ing a char ac ter is tic pat tern cor re spond ing to deglaciation. The warm (~50–60%) in ter val 2.31–1.90 m (19.0–15.8 ka) is fol lowed by a sig nif i cant cool ing, as re flected by the abrupt drop to gla cial val ues (~20%). The rises to wards more pos i tive val ues in the PPC cor re spond to pro nounced deglacial warm ing peaks (MWFs of Aharon, 2003).
The brief re bound to slightly lower val ues around 60% at 12.5 ka is re lated to the Green land Stadial 2 (GS-2)/Youn ger Dryas (YD) cool ing event, and there fore, de spite the lower res o - lu tion of this part of the core, we can sug gest that the tran si tion to the Ho lo cene was not char ac ter ized by a grad ual warm ing, but in stead was punc tu ated by the cold YD in ter val. The Ho lo - cene sec tion of the re cord (1.36–0 m) is marked by a rel a tively very warm in ter val with val ues close to or greater than 80%.
Par tic u larly in the mid dle Ho lo cene, the PPC val ues ap prox i - mate 95%, in dic a tive of the Ho lo cene Cli ma tic Op ti mum (HCO).
PLANKTONIC FORAMINIFERAL ECO-BIOZONATION AND COMPARISON WITH PREVIOUS ECO-BIOSTRATIGRAPHIC SCHEMES
IN THE PERI-CARIBBEAN REGION
The de tailed quan ti ta tive dis tri bu tion of plank tonic fora - minifera re sulted in the iden ti fi ca tion of 12 eco-biozones (GOMPFE 1–GOMPFE 12) span ning ap prox i mately the last 21 kyr BP (Fig. 3). Each eco-biozone is de fined by us ing a pro - gres sive num ber from the top to the base of the core and iden ti - fied by lo cal (re-) oc cur rences and/or tem po rary dis ap pear - ances of the main spe cific taxa in as so ci a tion with more or less im por tant fluc tu a tions of their abun dance, in dic a tive of mod i fi - ca tion in ma rine con di tions, such as ver ti cal mix ing, as well as changes in ma rine pro duc tiv ity and sur face wa ter tem per a ture, sa lin ity and/or cur rent flow.
– GOMPFE 12 (2.96–2.61 m; 21.13–20.41 ka): this is rec - og nized be tween the base of the suc ces sion stud ied and 2.60 m. In this in ter val the as sem blages are dom i nated by the cold-wa ter spe cies G. inflata and G. cras saformis gr. This in ter val is also marked by a sig nif i cant con tri bu - tion of O. universa to the plank tonic fauna. The base of this eco-biozone falls within the mid dle part of the biozone Y2 of Ken nett et al. (1985).
– GOMPFE 11 (2.61–2.36 m; 20.41–19.01 ka): this zone is bounded by two in creases in G. crassaformis gr.
A near-ab sence of G. inflata, an in crease of G. ruber (both va ri et ies), the high est abun dance of G. bulloides, a strong de crease of O. universa, and the pres ence of G. scitula in the lower part are the main fea tures of this in ter val. This eco-biozone cor re sponds to the up per part of the biozone Y2 of Ken nett et al. (1985).
– GOMPFE 10 (2.36–1.91 m; 19.01–15.89 ka): the lower bound ary of this eco-biozone is iden ti fied by the sud den de crease in G. ruber rosea, an in crease in G. sacculifer, and the last con sis tent oc cur rence of G. inflata (Ken nett et al., 1985, bioevent f8 of Antonarakou et al., 2015). It is mainly char ac ter ized by a strong in crease in abun dance (acme) of G. crassaformis gr., O. universa and G. ruber, and the oc cur rence of G. truncatulinoides, neoglo - boquadrinids and G. aequilateralis which reach abun - dance val ues around 10%. This eco-biozone co in cides with the up per most part of the biozone Y2 of Ken nett et al. (1985).
– GOMPFE 9 (1.91–1.79 m; 15.89–15.28 ka): this is bounded by two sharp de creases in G. ruber rosea, and
62-CPJ eroc ni segal bme ssa larefinimarof cino tknalp fo snoi t au tculf evi ta ti tnauQ .3 .giF ( 62-CPJ eroc ni dezi ng ocer stneveoib larefinimarof cino tknalp eht sez ilo bmys eulb ni ”f‘‘ elihw ,seir adnuob cihpargitartsoib-oce eht tne se rper senil la tno z iroh eulb ehT,.la te uokaranotnA 5102tne ca jda eht fo kro wemarfcihpargitartsonorhc gn idnop se rroc eht dna MoG eht ni deni fed senozoib-oce larefinimarof cino tknalp eht neew teb emehcs noi ta le rroc eht nwohs si thgireht ot ;) setidibruT neerG – TG ,setidibruT deR – TR ,senoZ hciR-cina grO – ZRO ;)noi ger nae bbi raC/MoG-mucric( sn isab
eroc 62-CPJ eht fo dro cercihpargitartsoib-oce detar ge tnI .4 .giF ,.g.e( yh pa rgi tarts tneve ETAMITNI eht fo )der ni(slaidatsretnI dna )eulb ni(slaidatS dna lneerG ot deta le rroc era senozoib-oce ;tfel eht no nwohs si stneveoib-oce dezi ng ocer 21 fo noi sse ccus ehT ;2PSIG – 2 tce jorP teehS ecI dna lneerG2102 ,.la te yelkcolBeht ot dna ))4791( .la te duregnaMera stneve ci ta milc ;nwohs osla era seir adnuob enozoib-oce rof denia tbo sega eht ;stneve ci ta milc eht yb dethgi lhgih reh trufrebur .G( cifi ceps-epytohprom eht yb dna ,)CPP( evruc citamilcoealap cino tknalp eht ,noitaicalged tsal eht gn irud snoi t au tculfecna dnuba evi t aler rebur .G ).l.s d81 dna ,O ( de iduts eroc eht fo sdro cer aC/gM5102 ,.la te uokaranotnAni sa noi ta nal pxe rehto ;)3 er ugiF
is fur ther char ac ter ized by strong turn over in abun dance be tween the white and pink va ri et ies of G. ruber. The as - sem blages are char ac ter ized by an in crease of G. bulloides, and the sig nif i cant pres ence of warm wa ter spe cies (e.g., G. sacculifer gr., O. universa, G. aequi - lateralis gr.), in dic a tive of the be gin ning of deglaciation.
This eco-biozone matches the low er most part of biozone Y1 of Ken nett et al. (1985).
– GOMPFE 8 (1.79–1.65 m; 15.28–13.77 ka): this in ter val is marked by a pro gres sive up wards de crease in G. crassaformis gr., the re-oc cur rence of G. scitula, the near-dis ap pear ance of G. inflata, and a con com i tant in - crease in neogloboquadrinids. In the up per part a sig nif i - cant peak of G. bulloides marks the top of this eco-biozone. This eco-biozone co in cides with the mid - dle part of the biozone Y1 of Ken nett et al. (1985).
– GOMPFE 7 (1.65–1.53 m; 13.77–13.09 ka): the com po - si tion of this plank tonic micro fauna is sim i lar to that of eco-biozone 8, but dif fers by the ab sence of G. bulloides and G. scitula, and the last oc cur rence of G. inflata. At the base of this eco-biozone, P. oblioquiloculata ap pears in the as sem blages. This eco-biozone can be re lated with the mid dle part of biozone Y1 of Ken nett et al.
(1985).
– GOMPFE 6 (1.53–1.36 m; 13.09–11.70 ka): this eco-biozone shows a microfaunal com po si tion dras ti - cally dif fer ent to that of the pre vi ous zone, sug gest ing that it cor re sponds to the YD. Re-oc cur rences of G. sci - tula and G. bulloides de fine its base. In the in ter me di ate in ter val, the as sem blages are char ac ter ized by an in - crease in G. crassaformis, a de crease in O. universa and the pres ence of G. truncatulinoides and neogloboquadrinids. This eco-biozone is equiv a lent to the top of the biozone Y1 of Ken nett et al. (1985).
– GOMPFE 5 (1.36–1.26 m; 11.70–10.20 ka): high abun - dances of G. sacculifer gr. and neogloboquadrinids, a grad ual de crease in G. ruber alba and the ab sence of G. scitula and G. inflata char ac ter ize eco-biozone 5, the base of which is marked by the first con sis tent ap pear - ance of G. menardii group ( bioevent f7 of Antonarakou et al., 2015). This eco-biozone cor re sponds to biozone Y1A of Ken nett et al. (1985).
– GOMPFE 4 (1.26–1.01 m; 10.20–6.83 ka): this eco-biozone is de fined by a re-in crease in G. ruber alba, G. sacculifer gr. and neogloboquadrinids, as well as the pres ence of G. scitula. The base is de fined by the peak of G. sacculifer gr. (Poore et al., 2003) and min i mum abun dance of G. ruber alba (bioevent f6 of Antonarakou et al., 2015), while the top is marked by the dis ap pear - ance of G. crassaformis, which is in dic a tive of the mid-Ho lo cene Z1/Z2 bound ary (Ken nett et al., 1985;
bioevent f5 of Antonarakou et al., 2015) in the GoM. This eco-biozone co in cides with biozone Z2 of Ken nett et al.
(1985).
– GOMPFE 3 (1.01–0.71 m; 6.83–5.16 ka): a great ex pan - sion of warm wa ter spe cies char ac ter izes eco- biozone 3. The base of this in ter val is marked by the first con sis - tent oc cur rence of S. dehiscens. In creas ing per cent ages of warm wa ter spe cies (G. ruber alba, G. sacculifer gr.;
bioevent f4 of Antonarakou et al., 2015) and a strong re - duc tion or ab sence of neoglo boquadrinids and G. scitula char ac ter ize this eco- biozone that rep re sents the lower part of biozone Z1 of Ken nett et al. (1985).
– GOMPFE 2 (0.71–0.36 m; 5.16–2.38 ka): this is mainly char ac ter ized by an abrupt in crease in G. menardii gr., while other warm wa ter spe cies such as G. ruber,
G. sacculifer and O. universa are well-rep re sented. The re-ap pear ance of G. crassaformis (bioevent f3 of Anto - narakou et al., 2015) and de creases in G. bulloides, G. sacculifer gr., and G. aequilateralis gr. are also in dic - a tive of this in ter val. This eco-biozone is con gru ent with the mid dle part of biozone Z1 of Ken nett et al. (1985).
– GOMPFE 1 (0.36–0 m; 2.38–0.51 ka): this is pres ent in the up per most part of the re cord. It is char ac ter ized by a strong in crease in P. oblioquiloculata, O. universa and G. menardii gr. Near to the top of this eco-biozone, the min i mum and max i mum abun dances of G. ruber and P. oblioquiloculata (bioevents f1–f2 of Antonarakou et al., 2015; Poore et al., 2003) are rec og nized. All these con com i tant bioevents to gether with the com pletely ab - sence of G. bulloides and S. dehiscens rep re sent a dis - tinc tive late Ho lo cene fea ture in the GoM. This eco-biozone cor re sponds to the up per most part of biozone Z1 of Ken nett et al. (1985).
DISCUSSION
ECO-BIOSTRATIGRAPHIC APPROACH IN SUBTROPICAL MARGINAL SETTINGS
The quan ti ta tive dis tri bu tion of plank tonic foraminifera from the LGM to the pres ent dis plays a suc ces sion of microfaunal as sem blages (biozones) char ac ter iz ing the pro gres sively chang ing sur face wa ter en vi ron ment. In such mar ginal bas ins, the suc ces sive fau nal zones can be in di cated as ecozones, be - cause the ma jor ity of the spe cies are sen si tive in di ca tors of past en vi ron men tal changes with am pli fied sig nals (Drinia and Antonarakou, 2003; Antonarakou et al., 2007, 2018, 2019;
Drinia et al., 2007, 2008, 2016; Kontakiotis, 2012, 2016;
Kontakiotis et al., 2013, 2016a, b, 2017; Fenton et al., 2016;
Karakitsios et al., 2017a, b). The eco log i cal di ver gence be - tween the spe cies (Hemleben et al., 1989; Rohling et al., 1993), even be tween dif fer ent morphotypes of the same spe cies (Antonarakou et al., 2015; Kontakiotis et al., 2017), de ter mined by dif fer ent con trol ling fac tors (e.g., depth hab i tats, growth op ti - mum/stressed con di tions, pro duc tiv ity, strat i fi ca tion, near-/off - shore-con di tions), in com bi na tion with their dif fer ent test mor - pho logies (e.g., re lated to shell shape and size, thick ness, in ner po ros ity and pore sur face dis tri bu tion; Morard et al., 2009;
Kontakiotis et al., 2017), con trol di rectly their dis tri bu tion pat tern and re flect the re gional cli mate vari abil ity (Marcott et al., 2013).
There fore, this study uses the con cept of “as sem blage eco-biozones” to re fer to the eco log i cal re sponse of plank tonic foraminifera to en vi ron men tal changes, rather than evo lu tion - ary changes. In this sense plank tonic foraminifera are im por tant palaeocological and palaeoclimatological in di ca tors and such sub di vi sion of their re cord pro vides sig nif i cant in sights into the hy dro log i cal dy nam ics of the ba sin stud ied. More over, the ap - plied eco-biostratigraphic scheme ef fec tively re flects the global cli ma tic os cil la tions dur ing the lat est Qua ter nary. The ap pli ca - bil ity of this ap proach has been tested in many sed i ment cores col lected in var i ous mar ginal bas ins (e.g., Aegean Sea, Ionian Sea, Adri atic Sea, Tyrrhenian Sea, Ca rib bean Sea). The con - struc tion of such a de tailed eco-biochronostratigraphic frame - work through the iden ti fi ca tion of these bio-ecoevents in other more or less dis tant cores per mits the re fine ment of pre vi ous Late Qua ter nary re gional schemes (Ericson and Wollin, 1968;
Ruddiman, 1971; Ken nett and Huddlestun, 1972; Ken nett et al., 1985; Flower and Ken nett, 1990; Mar tin et al., 1990; Wil son, 2012), makes a pre cise cor re la tion be tween them and fa cil i - tates palaeoclimatic and/or palaeoceanographic in ter pre ta tions
of the time in ter val stud ied. Any mi nor dis crep an cies be tween the pres ent and pre vi ous stud ies con cern the de gree of synchroneity be tween the bio-ecozone bound aries, be ing due to (1) the dif fer ent size frac tions in plank tonic foraminifera anal - y ses adopted by the au thors (>100, 125, 150 or 175 µm), (2) dat ing un cer tain ties of 14C ages prob a bly due to the ap pli ca tion of a con stant sea-sur face 14C res er voir age cor rec tion in pre vi - ous stud ies, es pe cially dur ing the late gla cial–inter gla cial tran si - tion (e.g., Broecker et al., 1988), and to (3) an ab sence or a weak ac cu racy of 14C dat ing per formed in some cases, on mixed ben thic foraminifera and on pteropods (e.g., Ken nett et al., 1985; Flower and Ken nett, 1990).
LATE PLEISTOCENE–HOLOCENE INTEGRATED CHRONOSTRATIGRAPHIC FRAMEWORK, PALAEOCEANOGRAPHIC EVENT STRATIGRAPHY AND PALAEOENVIRONMENTAL IMPLICATIONS
A plank tonic foraminiferal eco-biostratigraphy was com - bined with the palaeoclimatic JPC-26 re cords to ob tain a de - tailed late Pleis to cene–Ho lo cene in te grated strati graphic frame work for the GoM (Fig. 4). Short-term cli ma tic and fau nal fluc tu a tions rec og nized dur ing this time span can be com pared to those pro posed by the INTIMATE group (Björck et al., 1998;
Lowe et al., 2008; Blockley et al., 2012) in Green land ice cores.
In the low er most part of the suc ces sion stud ied, eco-bio - zones GOMPFE 12 and GOMPFE 11 cover the time in ter val be - tween 21.1 and 19.0 ka, which chro no log i cally cor re sponds to the GS-2b event of the late gla cial pe riod and in cludes the LGM.
Dur ing this in ter val, the plank tonic as sem blages show gen er ally low di ver sity and are dom i nated by spe cies in di cat ing cold and pro duc tive wa ter con di tions. The cold est in ter val is iden ti fied in all cli ma tic re cords (d18O, Mg/Ca-SSTs, PPC) at the top of eco-biozone GOMPFE 12, char ac ter ized by a sig nif i cant de - crease in the warm wa ter spe cies G. ruber rosea and by the high est abun dances of G. inflata and G. crassaformis, in dic a tive of rather cool and deep mixed lay ers. In this part, heavier d18O val ues and min i mum SSTs of about 19°C are also re corded and as so ci ated with the LGM (Antonarakou et al., 2015). Nev er the - less dur ing the LGM, GoM SSTs do not dis play con tin u ously low tem per a ture val ues, which is con sis tent with rel e vant cli mate re - cords from the trop i cal and sub trop i cal At lan tic (Bard et al., 2000;
Flower et al., 2004; Schmidt et al., 2004; Weldeab et al., 2006). A no tice able event, that could partly ex plain this slight SST rise, is the sud den in crease of G. ruber alba, as sug gested by clear pos - i tive fluc tu a tions (e.g., in the up per part of eco-biozone GOMPFE 11). Gen er ally, G. ruber fluc tu a tions have been con - sid ered good re cord ers of cli mate vari abil ity (Sprovieri et al., 2006a, 2012; Budillon et al., 2009; Antonarakou et al., 2015), with its rel a tive abun dance pos i tively co-vary ing with SST (Hagen and Keigwin, 2002; Sprovieri et al., 2006b) and its abun - dance os cil la tions strongly con trolled by global cli mate forc ing (Sprovieri et al., 2006b). De spite the pre dom i nance of cold spe - cies dur ing the gla cial pe riod, an in crease (27.8–51.2%) of G. ruber alba dur ing the LGM is a com mon fea ture of mar ginal re cords (circum-GoM/Ca rib bean re gion, Med i ter ra nean Sea), prob a bly due to the fact that dur ing the LGM SSTs do not dis play their low est val ues (Cacho et al., 2001; Sbaffi et al., 2004; Siani et al., 2010; Kontakiotis, 2016).
Eco-biozone GOMPFE 10 cor re sponds to the GS-2a cli ma - tic event and in cludes the first step of the last deglaciation, dated in the sub trop i cal At lan tic to be tween 19.0 and 12.6 ka (Aharon, 2003, 2006; Carlson et al., 2008; Sionneau et al., 2010; Schmidt and Lynch-Stieglitz, 2011; Antonarakou et al., 2015). A slight de crease in the plank tonic foraminiferal di ver sity char ac ter izes the as sem blage, dom i nated by both warm and cold wa ter spe cies (G. crassaformis, neogloboquadrinids,
G. bullo ides, G. ruber and O. universa). Mg/Ca-based es ti - mates show a rel a tively sta ble cli mate pe riod with SSTs around 22°C. How ever, in the mid dle part of this eco-biozone (~17.0 ka), which has been as cribed to the Hein rich 1 event (Antonarakou et al., 2015), an in ten si fi ca tion of cold con di tions is sug gested by the oc cur rence of the heavi est d18O val ues in the whole re cord. Eco-biozone GOMPFE 9 cor re sponds to the up per part of GS2a. The strong neg a tive ex cur sion in PPC and Mg/Ca-SST re cords at the base of this in ter val may be ten ta - tively re lated to the Old est Dryas (OD) event, the age of which in the GoM is doc u mented be tween 16.0 and 14.7 ka (Lea et al., 2003; Wil liams et al., 2010).
The GOMPFE9/8 bound ary is close to the bound ary be - tween the GS-2a and GI-1 events. In this in ter val, a ten dency to ward warmer con di tions is in di cated by an up wards de creas - ing trend of cold wa ter spe cies (e.g., G. crassaformis) abun - dances and by a rel a tive warm ing trend in ferred from sig nif i cant d18O de ple tion in the iso to pic re cord and con com i tant in creased tem per a tures in the Mg/Ca-based re cord. Eco-biozones 8 and 7 cor re spond to the interstadial BÝlling/AllerÝd (B/A) sensu Mangerud et al. (1974) and the Green land GI-1 ep i sode (Lowe et al., 2008) dated to be tween 14.8 and 13.0 ka. This event falls within the or ganic-rich de pos its ORZ II–III, where a strong change in sed i men ta tion rate and in plank tonic biodiversity is re corded. These fea tures pre vent the iden ti fi ca tion of high-res - o lu tion cli mate os cil la tions. The ob served cli ma tic os cil la tions dur ing the GI-1 event show three warm/hu mid – GI-1e (BÝlling ) and GI-1a,c (AllerÝd) – events that were in ter rupted by two cold/dry – GI-1d (Older Dryas ) and GI-1b (Intra AllerÝd Cold Pe riod; IACP) events on a cen ten nial scale. All these short-term os cil la tions are con sid ered to be con tem po ra ne ous with GI-1e to GI-1a of the ice-core re cord (Fig. 4), sug gest ing a cli ma tic link be tween the study area and the high-lat i tude ar eas.
Over all, an im por tant cli ma tic warm ing (5–7°C based on Mg/Ca-SSTs in GI-1e; Antonarakou et al., 2015) and the pres - ence of warm and sea son ally strat i fied wa ter masses are in - ferred from the struc ture of the as sem blages of these eco-biozones, char ac ter ized by the pres ence of G. truncatu - linoides and neogloboquadrinids. The d18O re cord, on the whole, shows a ten dency to lighter val ues, sug gest ing a cli ma tic ame lio ra tion. How ever, the up wards deglaciation trend is of ten in ter rupted by brief cool ing events, ex pressed by pos i tive iso - tope ex cur sions and/or SST neg a tive peaks. Such an abrupt Mg/Ca-SST cold spell (~3°C), con com i tant with a slight de - crease in PPC and d18O re cords, char ac ter izes the up per most part of eco-biozone 8, at trib uted to the stadial GI-1d event dated in our re cord to ~13.8 ka. A de crease in G. ruber and O. universa ac com pa nied by an in crease of neogloboqua - drinids de fines the plank tonic foraminiferal as sem blage. A no - tice able event is the tem po rary ab sence of G. scitula in this time in ter val, which re cords the zonal bound ary be tween GI-1d and GI-1c (Minisini et al., 2007; Rouis-Zargouni et al., 2010; Siani et al., 2010).
The fol low ing eco-biozone GOMPFE 7 cov ers the time in - ter val be tween 13.7 and 13.0 ka and is at trib uted to the interstadial GI-1c event as in di cated by SST re con struc tion.
This in ter pre ta tion fits well with the plank tonic foraminiferal con - tent, char ac ter ized by a pro gres sive in crease in warm wa ter taxa (G. ruber, G. sacculifer gr., O. universa). The be gin ning of this event is punc tu ated by the re-oc cur rence of O. universa in - di cat ing in com bi na tion with G. ruber the es tab lish ment of a sea sonal thermocline (Tedesco and Thunell, 2003; Tedesco et al., 2007; Wejnert et al., 2010). The d18O sig nal in di cates a strong de ple tion at the base (GI-1c), and a weak up wards en - rich ment trend to cooler sea sur face wa ter con di tions, which is at trib uted to the IACP/GI-1b event (Al ley et al., 1993). Fi nally, at
the top of this in ter val, a sig nif i cant warm ing of about 3.5°C (GI-1a) is marked in the Mg/Ca-SST re cord. How ever, this event is not re corded in the PPC and d18O re cords, most likely due to the lower res o lu tion of the sed i men tary re cord and/or bioturbation ef fects in this part of the suc ces sion.
Fol low ing this warm in ter val, both Mg/Ca and d18O re cords show a cool ing trend of 3.5°C cen tered on 12.9–12.7 ka, prob a - bly rep re sent ing the strong cli ma tic cool ing linked to the GS-1 event in the Green land ice core re cords and the GS-1 pe riod, dated in core JPC-26 be tween 13.0 and 12.5 ka. The fea tures of the as sem blages (sig nif i cant in crease in cold wa ter eutrophic spe cies as so ci ated with a de crease in warm wa ter taxa) in di - cate a gla cial pe riod of gen er ally high sur face wa ter fer til ity that char ac ter izes this eco-biozone (GOMPFE 6).
The Ho lo cene sec tion of the re cord cor re sponds to eco-biozones GOMPFE 1–5, and is marked by higher di ver sity in plank tonic as sem blages com pared to the pre vi ous eco- biozones, and a rel a tively warm in ter val [neg a tive d18O val ues, el e vated Mg/Ca and PPC val ues; (–0.71)–(–3.42‰), ~24–30°C and >70% re spec tively], but with pro nounced warm-cold os cil la - tions with well-de fined min ima (e.g., at 6.1, 5.6, and 1.1 ka). The lat est Ho lo cene de lim its eco-biozone 5, which spans a short time in ter val (11.7–10.2 ka) in the JPC-26 re cord. SSTs reached Ho - lo cene val ues with a mean am pli tude change ob served dur ing the YD/Ho lo cene tran si tion of about 2°C. How ever, the cli ma tic ame lio ra tion is more ev i dent in the fol low ing eco-biozone (GOMPFE 4), cor re spond ing to the HCO in ter val, where lighter d18O and higher PPC val ues are re corded. The lat ter also in - cludes eco-biozone GOMPFE 3, and is the re sult of a sub stan tial in crease in warm wa ter taxa (G. ruber, O. universa, G. sacculifer gr., and G. aequilateralis gr.) and the pro gres sive re duc tion (into GOMPFE 4) or ab sence (into GOMPFE 3) of G. crassaformis and G. scitula. The oc cur rence of a short cool ing event in the lower part of eco-biozone 3 (~6.1 ka) is sug gested by sig nif i cant d18O de ple tion and re duc tion in the Globigerinoides ruber per - cent ages, cou pled with an in crease in abun dances of Globo - rotalia truncatulinoides as well as the re-oc cur rence of Globigerina bulloides. All these fea tures in the plank tonic as sem - blages rep re sent an ep i sode of im proved ven ti la tion in this ba sin at test ing to deep win ter con vec tion and ver ti cal mix ing in the wa - ter col umn (Schmuker and Schiebel, 2002; Arellano-Torres and Machain-Castillo, 2017). This may cor re late with the global-scale rapid cli mate change (RCC) at 6.0–5.0 ka BP re ported in the Mid dle Ho lo cene by Mayewski et al. (2004). A sub se quent SST rise be tween 5.1–2.4 ka de lin eates ecobiozone GOMPFE 2. The PPC sig nal is close enough to its max i mum val ues, and the spe - cies G. ruber, G. sacculifer, G. menardii gr. and neogloboqua - drinids are the main con stit u ents of the plank tonic fauna. The near-ab sence of G. crassaformis, G. inflata and G. scitula may in di cate that re duced ox y gen a tion in the deeper wa ter col umn was det ri men tal to that spe cies, sug gest ing sta ble strat i fi ca tion and poor deep-wa ter ven ti la tion in the GoM. Within this in ter val
and pro gres sively through the fol low ing eco-biozone, the d18O re cord shows the on set up wards of strong en rich ment and de - ple tion phases cor re spond ing to a long pe riod of marked sea - son al ity through out the late Ho lo cene, pos si bly as so ci ated with rel a tive changes in Mis sis sippi River run off. Fi nally, eco-biozone GOMPFE 1 pres ents the typ i cal mod ern foraminiferal as so ci a - tion, dom i nated by tem per ate and sub trop i cal taxa (Dowsett et al., 2003; Richey et al., 2007; Poore et al., 2013; Arellano- Torres and Machain-Castillo, 2017).
CONCLUSIONS
The multidisciplinary ap proach used here re sults in a de - tailed chronostratigraphic frame work for the last 21 ka for the GoM (also valid for the peri-Ca rib bean re gion), based on the study of as so ci a tions of plank tonic foraminifera from the deep-sea and the high sed i men ta tion rate core JPC-26. Twelve eco-biozones (GOMPFE 1–12) were iden ti fied by sharp changes in plank tonic foraminiferal taxa and com pared to the strati graphic event scheme pro posed by the INTIMATE group in or der to better iden tify the re la tion ships be tween past cli ma tic changes and the re sponse of microfaunal as sem blages in the sub trop i cal At lan tic. Most of these downcore fau nal vari a tions in re la tion to SST os cil la tions rep re sent mil len nial to cen ten nial cli ma tic vari abil ity, which are re corded in Green land ice cores over the last deglaciation. Among them, we rec og nized the gla - cial GS-2a to GS-2b events, the interstadial B-A/GI-1a-d events, the YD/GS-1 and the Ho lo cene warm in ter val. More - over, the rec og ni tion of a strati graphic re la tion ship be tween plank tonic foraminiferal bioevents and or ganic-rich sed i ments en hances the pos si bil ity for re li ably cor re lat ing ma rine re cords in such mar ginal set tings, of fer ing a better com pre hen sion of the palaeoceanographic his tory of these bas ins. These find ings rep re sent a re fine ment and ex ten sion of pre vi ous eco-bio - stratigraphic re cords in the sub trop i cal At lan tic, fa cil i tat ing re - gional and wide spread cli ma tic and microfaunal cor re la tions over the last 21 ky. A con clu sive cor re la tion of this work with oth ers well known from close or more dis tal ar eas in di cated the ro bust ness of these re sults, with par tic u lar em pha sis on multi-dis ci plin ary strat e gies and multiproxy stud ies per formed on late Qua ter nary cores with high sed i men ta tion rates, which can pro vide a pow er ful tool for mon i tor ing the palaeoclimatic evo lu tion of the sub trop i cal At lan tic in re la tion to the global cli - ma tic sys tem over the last gla cial cy cle.
Ac knowl edge ments. Con struc tive com ments by two anon y mous re view ers were crit i cal to im prov ing this manu - script, and Dr. W. Granoszewski (Co-Ed i tor) is also thanked for his ed i to rial han dling.
REFERENCES
Aharon, P., 2003. Melt wa ter flood ing events in the Gulf of Mex ico re - vis ited: im pli ca tions for rapid cli mate changes dur ing the last deglaciation. Paleoceanography, 18: 1079 doi:10.1029/ 2002PA 000840.
Aharon, P., 2006. En train ment of meltwaters in hyperpycnal flows dur ing deglaciation superfloods in the Gulf of Mex ico. Earth and Plan e tary Sci ence Let ters, 241: 260–270.
Al ley, R.B., Meese, D.A., Shuman, C.A., Gow, A.J., Tay lor, K.C., Grootes, P.M., White, J.W.C., Ram, M., Waddington, E.D., Mayewski, P.A., Zielinski, G.A., 1993. Abrupt in crease in Green land snow ac cu mu la tion at the end of the Youn ger Dryas event. Na ture, 362: 527.
Antonarakou, A., Drinia, H., Tsaparas, N., Dermitzakis, M.D., 2007. Micropaleontological pa ram e ters as prox ies of Late Mio -
cene sur face wa ter prop er ties and paleoclimate in Gavdos Is - land, East ern Med i ter ra nean. Geodiversitas, 29: 379–399.
Antonarakou, A., Kontakiotis, G., Mortyn, P.G., Drinia, H., Sprovieri, M., Besiou, E., Tripsanas, E., 2015. Bi otic and geo - chem i cal (d18O, d13C, Mg/Ca, Ba/Ca) re sponses of Globigerinoides ruber morphotypes to up per wa ter col umn vari - a tions dur ing the last deglaciation, Gulf of Mex ico. Geochimica et Cosmochimica Acta, 170: 69–93.
Antonarakou, A., Kontakiotis, G., Zarkogiannis, S., Mortyn, P.G., Drinia, H., Koskeridou, E., Anastasakis, G., 2018.
Plank tonic foraminiferal ab nor mal i ties in coastal and open ma - rine east ern Med i ter ra nean en vi ron ments: a nat u ral stress mon - i tor ing ap proach in re cent and early Ho lo cene ma rine sys tems.
Jour nal of Ma rine Sys tems, 181: 63–78.
Antonarakou, A., Kontakiotis, G., Vasilatos, C., Besiou, E., Zarkogiannis, S., Drinia, H., Mortyn, P.G., Tsaparas, N., Makri, P., Karakitsios, V., 2019. Eval u at ing the ef fect of ma rine diagenesis on Late Mio cene pre-evaporitic sed i men tary suc - ces sions of east ern Med i ter ra nean Sea. IOP Con fer ence Se - ries: Earth and En vi ron men tal Sci ences, 221: 012051 doi:
10.1088/1755- 1315/221/1/012051.
Antonov, J.I., Seidov, D., Boyer, T.P., Locarnini, R.A., Mishonov, A.V., Gar cia, H.E., Baranova, O.K., Zweng, M.M., John son, D.R., 2010. World Ocean At las 2009, Vol ume 2: Sa lin ity. In:
NOAA At las NESDIS 69 (ed. S. Levitus). Wash ing ton, D.C., U.S. Gov. Print ing Of fice.
Arellano-Torres, E., Machain-Castillo, M.L., 2017. Late Pleis to - cene–Ho lo cene vari abil ity in the south ern Gulf of Mex ico sur - face wa ters based on plank tonic foraminiferal as sem blages.
Ma rine Micropaleontology, 131: 44–58.
Auladell, M., Pelegrí, J.L., García-Olivares, A., Kir wan, A.D., Lipphardt, B.L., Martín, J.M., Pascual, A., Sangr´, P., Zweng, M., 2010. Mod el ling the early evo lu tion of a Loop Cur rent ring.
Jour nal of Ma rine Sys tems, 80: 160–171.
Bard, E., Rostek, F., Turon, J.-L., Gendreau, S., 2000. Hy dro log i - cal Im pact of Hein rich Events in the Sub trop i cal North east At - lan tic. Sci ence, 289: 1321–1324.
Behringer, D.W., Molinari, R.L., Fes ta, J.F., 1977. The vari abil ity of an ti cy clonic cur rent pat terns in the Gulf of Mex ico. Jour nal of Geo phys i cal Re search, 82: 5469–5476.
Biggs, D.C., Müller-Karger, F.E., 1994. Ship and sat el lite ob ser va - tions of chlo ro phyll stocks in in ter act ing cy clone-an ti cy clone eddy pairs in the west ern Gulf of Mex ico. Jour nal of Geo phys i cal Re search: Oceans, 99: 7371–7384.
Björck, S., Walker, M.J.C., Cwynar, L.C., Johnsen, S., Ka - ren-Luise, K., Lowe, J.J., Wohlfarth, B., 1998. An event stra - tig ra phy for the Last Ter mi na tion in the North At lan tic re gion based on the Green land ice-core re cord: a pro posal by the INTIMATE group. Jour nal of Qua ter nary Sci ence, 13: 283–292.
Blockley, S.P.E., Lane, C.S., Hardiman, M., Ras mus sen, S.O., Seierstad, I.K., Steffensen, J.P., Svens son, A., Lotter, A.F., Turney, C.S.M., Bronk Ramsey, C., 2012. Synchronisation of palaeoenvironmental re cords over the last 60,000 years, and an ex tended INTIMATE event stra tig ra phy to 48,000 b2k. Qua ter - nary Sci ence Re views, 36: 2–10.
Blumberg, A.F., Mellor, G.L., 1985. A sim u la tion of the cir cu la tion in the Gulf of Mex ico. Is rael Jour nal of Earth Sci ences, 34:
122–144.
Bonfardeci, A., Caruso, A., Bartolini, A., Bassinot, F., Blanc-Valleron, M.-M., 2018. Dis tri bu tion and ecol ogy of the Globigerinoides ruber–Globigerinoides elongatus morphotypes in the Azores re gion dur ing the late Pleistocene–Holocene.
Palaeo ge ogra phy, Palaeoclimatology, Palaeo ec ol ogy, 491:
92–111.
Bouma, A.H., Stelting, C.E., Coleman, J.M., 1985. Mis sis sippi Fan, Gulf of Mex ico. In: Sub ma rine Fans and Re lated Turbidite Sys tems(eds. A.H. Bouma, W.R. Normark and N.E. Barnes):
143–150. Springer New York, New York.
Broecker, W.S., 1991. The great ocean con veyor. Ocean og ra phy, 4: 79–89.
Broecker, W.S., Andree, M., Bonani, G., Wolfli, W., Klas, M., Mix, A., Oeschger, H., 1988. Com par i son be tween ra dio car bon
ages ob tained on co ex ist ing plank tonic foraminifera. Paleo - ceanography, 3: 647–657.
Budillon, F., Lirer, F., Iorio, M., MacrÒ, P., Sagnotti, L., Vallefuoco, M., Ferraro, L., Garziglia, S., Innangi, S., Sahabi, M., Tonielli, R., 2009. In te grated strati graphic re con struc tion for the last 80 kyr in a deep sec tor of the Sar dinia Chan nel (West ern Med i ter ra - nean). Deep Sea Re search Part II: Top i cal Stud ies in Ocean og - ra phy, 56: 725–737.
Cacho, I., Grimalt, J.O., Ca nals, M., Sbaffi, L., Shackle ton, N.J., Schönfeld, J., Zahn, R., 2001. Vari abil ity of the west ern Med i - ter ra nean Sea sur face tem per a ture dur ing the last 25,000 years and its con nec tion with the North ern Hemi sphere cli ma tic changes. Paleoceanography, 16: 40–52.
Capotondi, L., Maria Borsetti, A., Morigi, C., 1999. Foraminiferal ecozones, a high res o lu tion proxy for the late Qua ter nary biochronology in the cen tral Med i ter ra nean Sea. Ma rine Ge ol - ogy, 153: 253–274.
Carlson, A.E., Oppo, D.W., Came, R.E., LeGrande, A.N., Keigwin, L.D., Curry, W.B., 2008. Sub trop i cal At lan tic sa lin ity vari abil ity and At lan tic me rid i o nal cir cu la tion dur ing the last deglaciation. Ge ol ogy, 36: 991–994.
Caromel, A.G.M., Schmidt, D.N., Phillips, J.C., Rayfield, E.J., 2014. Hy dro dy namic con straints on the evo lu tion and ecol ogy of planktic foraminifera. Ma rine Micropaleontology, 106: 69–78.
Chen, S., Hu, C., 2017. Es ti mat ing sea sur face sa lin ity in the north - ern Gulf of Mex ico from sat el lite ocean color mea sure ments.
Re mote Sens ing of En vi ron ment, 201: 115–132.
Corbett, D.R., McKee, B., Duncan, D., 2004. An eval u a tion of mo - bile mud dy nam ics in the Mis sis sippi River deltaic re gion. Ma - rine Ge ol ogy, 209: 91–112.
Del Monte-Luna, P., Villalobos, H., Arreguín-Sánchez, F., 2015.
Vari abil ity of sea sur face tem per a ture in the south west ern Gulf of Mex ico. Con ti nen tal Shelf Re search, 102(C): 73–79.
Dowsett, H.J., Verardo, S., Poore, R.Z., 2003. Gulf of Mex ico plank tonic foraminifer trans fer func tion GOM2: pre lim i nary re - port. U.S. Geo log i cal Sur vey Open-File Re port OF 03-61.
Drinia, H., Antonarakou, A., 2003. Late Mio cene paleoclimatic vari a tions in the East ern Med i ter ra nean: an ecostratigraphical ap proach. Neues Jahrbuch für geologie und Paläontologie, 10:
603–616.
Drinia, H., Antonarakou, A., Tsaparas, N., Kontakiotis, G., 2007.
Palaeoenvironmental con di tions pre ced ing the Messinian Sa lin - ity Cri sis: A case study from Gavdos Is land. Geobios, 40:
251–265.
Drinia, H., Antonarakou, A., Kontakiotis, G., 2008. On the oc cur - rence of Early Plio cene ma rine de pos its in Ierapetra ba sin, E.
Crete. Bul le tin of Geosciences, 83: 63–78.
Drinia, H., Antonarakou, A., Tsourou, T., Kontakiotis, G., Psychogiou, M., Anastasakis, G., 2016. Foraminifera eco-biostratigraphy of the south ern Evoikos outer shelf, cen tral Aegean Sea, dur ing MIS 5 to pres ent. Con ti nen tal Shelf Re - search, 126: 36–49.
Elliott, B.A., 1982. An ti cy clonic Rings in the Gulf of Mex ico. Jour nal of Phys i cal Ocean og ra phy, 12: 1292–1309.
Ericson, D.B., Wollin, G., 1968. Pleis to cene Cli mates and Chro nol - ogy in Deep-Sea Sed i ments. Sci ence, 162: 1227–1234.
Fenton, I.S., Pearson, P.N., Dunkley Jones, T., Purvis, A., 2016.
En vi ron men tal Pre dic tors of Di ver sity in Re cent Plank tonic Foraminifera as Re corded in Ma rine Sed i ments. PLOS ONE, 11: e0165522.
Flower, B.P., Ken nett, J.P., 1990. The Youn ger Dryas Cool Ep i sode in the Gulf of Mex ico. Paleoceanography, 5: 949–961.
Flower, B.P., Ken nett, J.P., 1995. Bi otic re sponses to tem per a ture and sa lin ity changes dur ing the last deglaciation, Gulf of Mex ico.
In: Com mis sion on Geosciences, Ef fects of Past Global Changes on Life: 209–220. Na tional Acad emy Press, Wash ing - ton, D.C.
Flower, B.P., Hastings, D.W., Hill, H.W., Quinn, T.M., 2004. Phas - ing of deglacial warm ing and Laurentide Ice Sheet melt wa ter in the Gulf of Mex ico. Ge ol ogy, 32: 597–600.
Gopalakrishnan, G., Cornuelle, B.D., Hoteit, I., Rudnick, D.L., Owens, W.B., 2013. State es ti mates and fore casts of the loop
cur rent in the Gulf of Mex ico us ing the MITgcm and its adjoint.
Jour nal of Geo phys i cal Re search: Oceans, 118: 3292–3314.
Hagen, S., Keigwin, L.D., 2002. Sea-sur face tem per a ture vari abil - ity and deep wa ter re or gani sa tion in the sub trop i cal North At lan - tic dur ing Iso tope Stage 2–4. Ma rine Ge ol ogy, 189: 145–162.
Hemleben, C., Spindler, M., An der son, O.R., 1989. Mod ern Plank - tonic Foraminifera. Springer, New York.
Hofmann, E.E., Worley, S.J., 1986. An in ves ti ga tion of the cir cu la - tion of the Gulf of Mex ico. Jour nal of Geo phys i cal Re search:
Oceans, 91: 14221–14236.
INEGI, 2014. Anuario Estadístico y Geográfico por Entidad Federativa (in Spain). Instituto Nacional de Estadística y Geografía, Mex ico.
Jochens, A.E., DiMarco, S.F., 2008. Phys i cal ocean o graphic con - di tions in the deep water Gulf of Mex ico in sum mer 2000–2002.
Deep Sea Re search Part II: Top i cal Stud ies in Ocean og ra phy, 55: 2541–2554.
Jonkers, L., Reynolds, E., Richey, J., Hall, I.R., 2015. Lu nar pe ri - od ic ity in the shell flux of plank tonic foraminifera in the Gulf of Mex ico. Biogeosciences, 12: 3061–3070.
Karakitsios, V., Roveri, M., Lugli, S., Manzi, V., Gennari, G., Antonarakou, A., Triantaphyllou, M., Agiadi, K., Kontakiotis, G., Kafousia, N., de Rafelis, M., 2017a. A re cord of the Messinian sa lin ity cri sis in the east ern Ionian tec toni cally ac tive do main (Greece, east ern Med i ter ra nean). Ba sin Re search, 29:
203–233.
Karakitsios, V., Cornée, J.-J., Tsourou, T., Moissette, P., Konta - kiotis, G., Agiadi, K., Manoutsoglou, E., Triantaphyllou, M., Koskeridou, E., Drinia, H., Roussos, D., 2017b. Messinian sa - lin ity cri sis re cord un der strong fresh wa ter in put in mar ginal, in - ter me di ate, and deep en vi ron ments: the case of the North Aegean. Palaeo ge ogra phy Palaeoclimatology Palaeo ec ol ogy, 485: 316–335.
Ken nett, J.P., Huddlestun, P., 1972. Late Pleis to cene paleoclima - to logy, foraminiferal biostratigraphy and tephrochronology, west ern Gulf of Mex ico. Qua ter nary Re search, 2: 38–69.
Ken nett, J.P., Srinivasan, M.S., 1983. Neo gene Plank tonic Fora - minifera: a Phylo gen etic At las. Hutch in son Ross Pub lish ing Com pany, New York.
Ken nett, J.P., Elmstrom, K., Penrose, N., 1985. The last deglaciation in orca ba sin, gulf of Mex ico: high-res o lu tion plank - tonic foraminiferal changes. Palaeo ge ogra phy, Palaeoclima to - logy, Palaeo ec ol ogy, 50: 189–216.
Kontakiotis, G., 2012. Palaeoceanographic and palaeoclimatic study of East ern Med i ter ra nean dur ing Late Qua ter nary, based on plank tonic foraminiferal as sem blages (in Greek with Eng lish ex tended ab stract). Na tional and Kapodistrian Uni ver sity of Ath - ens, Ath ens, Greece.
Kontakiotis, G., 2016. Late Qua ter nary paleoenvironmental re con - struc tion and paleoclimatic im pli ca tions of the Aegean Sea (east ern Med i ter ra nean) based on paleoceanographic in dexes and sta ble iso topes. Qua ter nary In ter na tional, 401: 28–42.
Kontakiotis, G., Mortyn, P.G., Antonarakou, A., Martínez-Botí, M.., Triantaphyllou, M.V., 2011. Field-based val i da tion of a diagenetic ef fect on G. ruber Mg/Ca paleothermometry: core top re sults from the Aegean Sea (east ern Med i ter ra nean). Geo - chem is try Geo phys ics Geosystems, 12: Q09004 doi:
10.1029/2011GC003692
Kontakiotis, G., Antonarakou, A., Zachariasse, W.J., 2013. Late Qua ter nary palaeoenvironmental changes in the Aegean Sea:
in ter re la tions and in ter ac tions be tween north and south Aegean Sea. Bul le tin of the Geo log i cal So ci ety of Greece, 47: 167–177.
Kontakiotis, G., Karakitsios, V., Mortyn, P.G., Antonarakou, A., Drinia, H., Anastasakis, G., Agiadi, K., Kafousia, N., De Rafelis, M., 2016a. New in sights into the early Plio cene hy dro - graphic dy nam ics and their re la tion ship to the cli ma tic evo lu tion of the Med i ter ra nean Sea. Palaeo ge ogra phy, Palaeoclimato - logy, Palaeo ec ol ogy, 459: 348–364.
Kontakiotis, G., Mortyn, P.G., Antonarakou, A., Drinia, H., 2016b.
As sess ing the re li abil ity of foraminiferal Mg/Ca ther mom e try by com par ing field-sam ples and cul ture ex per i ments: a re view.
Geo log i cal Quar terly, 60 (3): 547–560.
Kontakiotis, G., Antonarakou, A., Mortyn, P.G., Drinia, H., Anastasakis, G., Zarkogiannis, S., Möbius, J., 2017. Mor pho - log i cal rec og ni tion of Globigerinoides ruber morphotypes and their sus cep ti bil ity to diagenetic al ter ation in the east ern Med i - ter ra nean Sea. Jour nal of Ma rine Sys tems, 174: 12–24.
Koutrouli, A., Anastasakis, G., Kontakiotis, G., Ballengee, S., Kuehn, S., Pe-Piper, G., Piper, D.J.W., 2018. The early to mid-Ho lo cene ma rine tephrostratigraphic re cord in the Nisyros- Yali-Kos vol ca nic cen ter, SE Aegean Sea. Jour nal of Vol ca nol - ogy and Geo ther mal Re search, 366: 96–111.
Lea, D.W., Pak, D.K., Pe ter son, L.C., Hughen, K.A., 2003. Syn - chroneity of Trop i cal and High-Lat i tude At lan tic Tem per a tures over the Last Gla cial Ter mi na tion. Sci ence, 301: 1361–1364.
Leventer, A., Wil liams, D.F., Ken nett, J.P., 1983. Re la tion ships be - tween an oxia, gla cial melt wa ter and microfossil pres er va tion in the Orca Ba sin, Gulf of Mex ico. Ma rine Ge ol ogy, 53: 23–40.
Limoges, A., de Ver nal, A., Van Nieuwenhove, N., 2014.
Long-term hy dro log i cal changes in the north east ern Gulf of Mex ico (ODP-625B) dur ing the Ho lo cene and late Pleis to cene in ferred from or ganic-walled dinoflagellate cysts. Palaeo ge - ogra phy, Palaeoclimatology, Palaeo ec ol ogy, 414: 178–191.
Liu, Y., Lee, S.-K., Muhling, B.A., Lamkin, J.T., Enfield, D.B., 2012. Sig nif i cant re duc tion of the Loop Cur rent in the 21st cen - tury and its im pact on the Gulf of Mex ico, Jour nal of Geo phys i cal Re search, 117, C05039-8, doi: 10.1029/2011JC007555 Locarnini, R.A., Mishonov, A.V., Antonov, J.I., Boyer, T.P., Gar -
cia, H.E., Baranova, O.K., Zweng, M.M., John son, D.R., 2010.
World Ocean At las 2009, Vol ume 1: Tem per a ture. In: NOAA At - las NESDIS 68 (ed. S. Levitus). U.S. Gov ern ment Print ing Of - fice, Wash ing ton, D.C.
LoDico, J.M., Flower, B.P., Quinn, T.M., 2006. Subcen - tennial-scale cli ma tic and hydrologic vari abil ity in the Gulf of Mex ico dur ing the early Ho lo cene. Paleoceanography, 21:
PA3015, doi: 10.1029/2005PA001243
Lowe, J.J., Ras mus sen, S.O., Björck, S., Hoek, W.Z., Steffensen, J.P., Walker, M.J.C., Yu, Z.C., 2008.
Synchronisation of palaeoenvironmental events in the North At - lan tic re gion dur ing the Last Ter mi na tion: a re vised pro to col rec - om mended by the INTIMATE group. Qua ter nary Sci ence Re - views, 27: 6–17.
Luo, H., Bracco, A., Cardona, Y., McWilliams, J.C., 2016.
Submesoscale cir cu la tion in the north ern Gulf of Mex ico: Sur - face pro cesses and the im pact of the fresh wa ter river in put.
Ocean Mod el ling, 101: 68–82.
Mangerud, J.A.N., Andersen, S.T., Berglund, B.E., Donner, J.J., 1974. Qua ter nary stra tig ra phy of Norden, a pro posal for ter mi - nol ogy and clas si fi ca tion. Boreas, 3: 109–126.
Marcott, S.A., Shakun, J.D., Clark, P.U., Mix, A.C., 2013. A Re con - struc tion of Re gional and Global Tem per a ture for the Past 11,300 Years. Sci ence, 339(6124): 1198–1201.
Ma rino, G., Rohling, E.J., Sangiorgi, F., Hayes, A., Casford, J.L., Lotter, A.F., Kuèera, M., Brinkhuis, H., 2009. Early and mid dle Ho lo cene in the Aegean Sea: in ter play be tween high and low lat i tude cli mate vari abil ity. Qua ter nary Sci ence Re views, 28:
3246–3262.
Mar tin, R.E., John son, G.W., Neff, E.D., Krantz, D.W., 1990. Qua - ter nary plank tonic foraminiferal as sem blage zones of the north - east Gulf of Mex ico, Co lom bia Ba sin (Ca rib bean Sea), and trop - i cal At lan tic Ocean: Graphic cor re la tion of microfossil and ox y - gen iso tope da tums. Paleoceanography, 5: 531–555.
Mayewski, P.A., Rohling, E.E., Curt Stager, J., Karlén, W., Maasch, K.A., Da vid Meeker, L., Meyerson, E.A., Gasse, F., van Kreveld, S., Holmgren, K., Lee-Thorp, J., Rosqvist, G., Rack, F., Staubwasser, M., Schnei der, R.R., Steig, E.J., 2004.
Ho lo cene cli mate vari abil ity. Qua ter nary Re search, 62:
243–255.
Meckler, A.N., Schu bert, C.J., Hochuli, P.A., Plessen, B., Birgel, D., Flower, B.P., Hinrichs, K.U., Haug, G.H., 2008. Gla cial to Ho lo cene terrigenous or ganic mat ter in put to sed i ments from Orca Ba sin, Gulf of Mex ico – a com bined op ti cal and biomarker ap proach. Earth and Plan e tary Sci ence Let ters, 272: 251–263.
