The influence of Atlantic climate variability on the long-term development of Mediterranean cold-water coral mounds (Alboran Sea, Melilla Mound Field)

Abstract. This study provides a detailed reconstruction of climatic events affecting a cold-water coral mound located within the East Melilla Coral Province (Southeast Alboran Sea) over the last 300 ky. Based on benthic foraminiferal assemblages, macrofaunal quantification, grain size analysis, sediment geochemistry, and foraminiferal stable isotope compositions, a reconstruction of environmental conditions prevailing in the region is proposed. The variations in planktonic and benthic δ18O values indicate that cold-water coral mound formation follows global climatic variability. Cold-water corals develop during both interglacial and glacial periods, although interglacial conditions would have allowed better proliferation. Environmental conditions during glacial periods, particularly during the Last Glacial Maximum, appear to better suit the ecological requirements of the erect cheilostome bryozoan Buskea dichotoma. Benthic foraminiferal assemblages suggest that high organic carbon flux characterized interglacial periods. Results from this study imply that increased influence of warm and moist Atlantic air masses during interglacial periods led to increased fluvial discharge, providing nutrients for cold-water corals. Important interglacial Atlantic Water mass inflow further promoted strong Alboran Gyres, and thus mixing between surface and intermediate water masses. Increased turbulence and nutrient supply would have hence provided suitable conditions for coral development. In contrast, benthic foraminiferal assemblages and grain size distributions suggest that the benthic environment received less organic matter during glacial periods, whilst bottom flow velocity was reduced in comparison to interglacial periods. During glacial periods, arid continental conditions combined to more stratified water masses caused a dwindling of coral communities in the southeastern Alboran Sea, although aeolian dust input may have allowed these to survive. In contrast to Northeast Atlantic counterparts, coral mound build-up in the southeastern Alboran Sea occurs during glacial as well as during interglacial periods and at very low aggradation rates (between 1 and 9 cm ky−1). We propose that Buskea dichotoma plays an important role in long-term mound formation at the East Melilla Coral Province, noticeably during glacial periods.


the uneven surface of CWC cores, such as the direct measurement of air or of CWC skeletons, a post treatment of the dataset was carried out. X-ray fluorescence values with Argon counts higher than 6000, representing the measurement of air and thereby more porous/cracked media not representative for changes in sediment composition, were removed from the final dataset. In this study, we use the Log 10 normalized ratios (Gregory et al., 2019) Ti/Al and Si/Al as proxies for aeolian input, whilst the Log 10 Zr/Al and Rb/Al are used as proxies for fluvial input. Indeed, 180 titanium enrichment is considered a typical indicator of increased Saharan dust influence (Frigola et al., 2008;Itambi et al., 2009;Rodrigo-Gámiz et al., 2011), as aeolian deposits tend to concentrate heavy minerals that are rich in elements such as titanium or zirconium (Balsam et al., 1995;Itambi et al., 2009, Rodrigo-Gamíz et al., 2011. Silicates make up an important part of Saharan material, whilst they are rare in Alboran sediments (Caquineau et al., 2002;Masqué et al., 2003). Thus, in the same way as for titanium, enrichment in silica can be used as a proxy for 185 increased aeolian input originating from the Sahara (Rodrigo-Gámiz et al., 2011;Feenstra, 2020). Since rubidium is common in aluminosilicate minerals contained in fluvial material, the Rb/Al ratio is used as an indicator of terrestrial run-off in the western Mediterranean (Calvert and Pedersen, 2007;Martinez-Ruiz et al., 2015;Feenstra, 2020).
Though zirconium is generally considered as a proxy for aeolian input for the same reasons as Titanium (Rodrigo-Gámiz et al., 2011), it has been shown that sediments originating from major Moroccan rivers are considerably 190 enriched in zirconium (Stanley et al., 1975). We hence use the Zr/Al and Rb/Al ratios as regional proxies for fluvial input.

Grain-size analysis and organic geochemistry
Grain-size of the siliciclastic fraction was analysed using the Malvern Mastersizer 3000 at the Department of Geology, Ghent University (Belgium). The core was sampled with a small spoon (1 cm 3 ) every 5 cm. Large clasts 195 (>1 cm), such as coral or bryozoan fragments, were removed prior to analysis. Samples were placed in 35 % H 2 O 2 to remove organic matter and boiled until the reaction ended. Following this first step, samples were boiled in 10 % HCl for 2 minutes to dissolve CaCO 3 . Prior to measurement, samples were placed in 2 % sodium polymetaphosphate and boiled to assure complete disaggregation. Any remaining particle larger than 2 mm was sieved out before measurement. Eighty seven size classes were measured (from 0.01 to 2000 µm). Each sample was measured three 200 times and results were then averaged. Mean grain-size of the siliciclastic fraction ̅̅̅̅ (Folk and Ward, 1957) was calculated on the entire dataset with the Rysgran package for R (Gilbert et al., 2015;R Core Team, 2018). The sortable silt mean size ̅̅̅ , as defined by McCave et al. (1995; i.e. the mean of the 10-63 µm grain size range), was also calculated following the same procedure. Furthermore, following McCave and Hall (2006), the percentage of sortable silt (SS%) in the total <63 µm fraction was calculated. This percentage, together with the sortable silt mean 205 size, was used as a proxy for bottom current velocity (McCave and Hall, 2006;Toucanne et al., 2012). It has to be mentioned that the use of ̅̅̅ as a proxy for bottom current velocity on cores recovered from CWC mounds may be biased (e.g. Eisele et al., 2011). Indeed, the baffling effect of coral framework can locally reduce bottom current velocity and favour the deposition of fine sediments (Huvenne et al., 2009;Titschack et al., 2009;Fentimen et al., 2020), thus leading to an underestimation of ̅̅̅ during periods with high CWC content. Because of this, only relative 210 increases in ̅̅̅ are considered in combination with results obtained from other proxies. MinC. The RockEval6 technique produces an Oxygen and Hydrogen index, respectively corresponding to the quantity of CO 2 relative to TOC and the quantity of pyrolyzable organic compounds relative to TOC (Fantasia et al., 2019). These two indices give an indication about the origin of the organic matter present in the samples (Van Krevelen, 1993).

Microfaunal and macrofaunal investigations 220
The core was sampled (sliced) every 10 cm for micropalaeontological analysis. Samples were weighed dry, washed through a 63 µm mesh sieve and dried at 30 °C. Each fraction was then dry sieved through a series of 63, 125 and 2000 µm mesh sieves and weighed. A target number of 300 benthic foraminifera were identified from the fraction larger than 125 µm for each sample. If the residue contained more than 300 specimens, it was split using a dry microsplitter. Relative abundances (percentages) of benthic species were calculated from the total benthic 225 foraminiferal assemblage. The benthic foraminiferal density was calculated by dividing the total number of foraminifera of a given sample by the sample fraction's weight. The diversity Shannon index (H') was computed using the PRIMER6 software (Clarke and Gorley, 2006).
Samples prepared for micropaleontological analysis were further used to identify bryozoan species/genera at the 230 Department of Biological, Geological and Environmental Sciences, University of Catania (Italy) on the 125 µm to 2 mm and >2 mm sized fractions. Key intervals with high bryozoan content, previously identified by CT imagery, were selected. Dominant scleractinian corals and main brachiopod and bivalve species were identified at the lowest taxonomic level possible on the 2 mm sized fraction at the Department of Geosciences, University of Fribourg (Switzerland). 235

Oxygen and Carbon stable isotope analysis
Stable oxygen and carbon isotope compositions were measured on 5 to 12 specimens of the planktonic foraminifera Globigerina bulloides and the benthic foraminifera Cibicides lobatulus from the size fraction 212-250 µm in order to prevent any ontogenic effect on the measurements (Schiebel and Hemleben, 2017). The specimens were first cleaned three times with distilled water in an ultrasonic bath for 2 seconds. The measurements were then made using a 240 Thermo Fisher Scientific GasBench II connected to a Thermo Finnigan Delta Plus XL isotope ratio mass spectrometer at the Stable Isotope Laboratory of the University of Lausanne (Switzerland) according to the method adapted from Spötl and Vennemann (2003

Radiometric dating
Radiocarbon dating was performed on benthic foraminifera from 3 samples from the upper first meter of core MD13-3462G at the Laboratory of Ion Beam Physics, ETH Zürich, Switzerland (Table 1). The epibenthic foraminifera species Discanomalina coronata, Cibicides lobatulus and Cibicides refulgens were picked in order to obtain between 4 and 10 mg of pure carbonate. The samples were first dissolved in phosphoric acid. The resulting extracted CO 2 was 250 then converted to graphite and measured by Accelerator Mass Spectrometry (AMS) technique using the MICADAS dedicated instrument (Synal et al., 2007). Results were corrected for 13 C and calibrated using the Marine13 calibration curve (Reimer et al., 2013) and the software OxCal v4.2.4 (Ramsey, 2017). A reservoir age of 390 ± 80 years was applied to all ages (Siani et al., 2000).

255
Uranium-series dating was carried out on 10 CWC fragments (D. pertusum and M. oculata) using a multicollector inductively coupled plasma source mass spectrometer MC-ICPMS (Thermo Fisher Scientific Neptune plus ) coupled with a dissolver (Aridus I) at the Institute of Environmental Physics, Heidelberg University (Table 2). In order to constrain the chronostratigraphy of the core, well-preserved coral fragments were selected at the upper and lower boundaries of coral-rich units. These were identified based on visual core descriptions and CT-analysis (macrofaunal 260 quantification; Fig. 3). Coral fragments were physically cleaned with a Dremel ® drill tool and by sand blasting, and further chemically cleaned using a weak acid leaching prior to measurements. The detailed sample protocol is described by Frank et al. (2004), while spectrometry and chemical U and Th extraction and purification followed Wefing et al. (2017). Uranium-series coral ages were used to calculate mound aggradation rates.

Chronostratigraphy
The chronostratigraphy of core MD13-3462G is based on the combination of the coral ages (U-series dating), the planktonic and benthic stable oxygen isotope records, and the foraminiferal radiocarbon ages for the top first meter of the core (Fig. 3). Coral ages have been widely used to define the chronology of cores recovered from coral mounds. This approach provides satisfying results although age reversals down core have to be taken into account 270 (e.g. Rüggeberg et al., 2007;Frank et al., 2009;Matos et al., 2017). Indeed, reefs are fragile structures and can collapse, topple and fragment through the action of bioerosion, strong bottom currents, and gravity-driven processes, resulting in transport and redeposition of coral fragments (Beuck et al., 2005;Dorschel et al., 2007;White et al., 2007). In contrast, constructing a continuous age model based on stable isotope records is generally considered untrustworthy for cores collected from coral mounds since sedimentation is intermittent (Dorschel et al., 2005). 275 However, coral ages at the upper and lower boundaries of coral build-up phases in core MD13-3462G (e.g. at 390 and 507 cm depth) correspond to changes in the stable oxygen isotope records (Fig. 3), which in turn match the changes between Marine Isotope Stages (MIS; Lisiecki and Raymo, 2005). As such, the stable oxygen isotope records can, in the case of core MD13-3462G and in conjunction with coral ages, indicate important stratigraphic https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License. boundaries (Fig. 3). This is particularly relevant during times when CWCs did not grow and hence cannot serve to 280 construct a timeframe.
The coral ages indicate that core MD13-3462G extends approximately from 300 ka BP (Marine Isotope Stage 9) to the Holocene (Fig. 3, Table 2). The stratigraphic boundaries from the base of the core to ca. 600 cm depth were defined based on the coral ages as planktonic stable oxygen isotope compositions show little variation. The 285 boundaries of MIS 8 are the most poorly defined (Fig. 3). Due to difficulties to define precisely the stratigraphy of this section of the core, it will not be considered in detail during this study. In contrast, the planktonic and benthic

Sediment characterization
The sediment in core MD13-3462G consists mostly of macrofaunal remains (essentially corals and bryozoans) surrounded by a clay-to silt-sized carbonate/siliciclastic matrix. No important variation in the matrix sediment is 295 observed throughout the core. Carbonate content varies from ca. 10 to 86 %, whilst average values generally range between 40 and 60 % (Fig. 4). Total organic carbon content in the sediment varies between 0.16 and 1.13 wt% (Fig.   4). The highest TOC value is measured during late MIS 3 (1.13 wt%), whilst the lowest is recorded during MIS 8 (0.16 wt%; Fig. 4). The most important shifts to higher TOC values are observed during MIS 5, MIS 3 and at the transition between MIS 2 and MIS 1 (Fig. 4). High TOC values correspond to interglacials, whilst low values 300 correspond to glacials (Fig. 4). The sediment samples are further characterized by high Oxygen index values (> 200 mg C0 2 /g TOC; Supplementary data), indicating that the organic matter is oxidized and of essentially terrestrial origin (Espitalié et al., 1985).
The mean grain size of the siliciclastic fraction ( ̅̅̅̅ ) varies between ca. 6 and 14 µm (Fig. 4), whilst ̅̅̅ varies 305 between ca. 19 and ca. 26 µm (Fig. 4). Trends in ̅̅̅ follow those of ̅̅̅̅ (Fig. 4). Overall, a decrease in ̅̅̅ and ̅̅̅̅ is associated to intervals marking the transitions from interglacial to glacial periods (Fig. 4). Conversely, an increasing trend is observed from ca. 550 to ca. 375 cm depth, corresponding to the passage from the later phases of MIS 6 to the end of MIS 5 (Fig. 4). This trend is mirrored in ̅̅̅̅ (Fig. 4). A sharp decrease in ̅̅̅ and ̅̅̅̅ marks the passage from MIS 3 to MIS 2 and the later phase of MIS 2 (Fig. 4). The percentage of sortable silt (SS%) increases with ̅̅̅ 310 (Fig. 5). As discussed by McCave and Hall (2006) and McCave et al. (2017), the straight-line relationship (slope of ca. 0.125 µm/% and an intercept at 0% of ca. 17.5 µm) between ̅̅̅ and SS% is indicative of a sorting process induced by bottom currents (Fig. 5).
High CWC content is observed during interglacial periods, whilst low content characterizes glacial periods (Fig. 3).
During MIS 3 coral content shows a more staggered distribution, with a range of values from less than 10 vol% to ca. 320 27 vol% (Fig. 3).
In total 23 genera of bryozoans were identified. Buskea dichotoma is by far the dominant bryozoan species (Fig. 6).
Bryozoan content varies in general between 10 and 20 vol% (Fig. 3). Very high content is, however, observed during 325 MIS 2, reaching near to 70 vol%. The fragments, although delicate and fragile, are well preserved, large sized and unworn (Fig. 6). Bryozoans are absent during most of MIS 5. This absence corresponds to the interval when coral content is the most important (Fig. 3). Conversely, the maximum abundance of bryozoans during MIS 2 correlates to a minimum in coral content (Fig. 3).

330
Brachiopods are mainly represented by the co-occurrence of the species Gryphus vitreus and Terebratulina retusa ( Fig. 6). These two brachiopods are regularly associated to the bivalve Bathyarca pectunculoides (Fig. 6). These three inverterbrates have been formerly reported from Mediterranean CWC environments. Gryphus vitreus and Terebratulina retusa are also recorded from Pleistocene CWC deposits from Rhodes, Greece (Bromley, 2005), whilst Bathyarca pectunculoides was found at the Santa Maria di Leuca CWC province (Mastrototaro et al., 2010;335 Negri and Corselli, 2016). Gryphus vitreus was also found associated to -white corals‖ between 235 and 255 m depth off the coast of the Hyères Islands, France (Emig and Arnaud, 1988). Although being fragile, the shells are well preserved (Fig. 6). The brachiopod/bivalves concentrate as layers; hence they demonstrate a non-continuous distribution ( Fig. 3 and 6). They reach their highest abundance during glacial periods, in particular at the end of MIS 3 (30 vol% at 80 cm). Brachiopods and bivalves are completely absent during the last two interglacial periods (Fig.  340 3).

Stable carbon isotopes 365
The range of δ 13 C values of the planktonic G. bulloides is between -2.2 ‰ at 12 cm and -0.5 ‰ at 292 cm, whereas that for the benthic C. lobatulus is between 0.9 ‰ at 872 cm and 1.8 ‰ at 362 cm (Fig. 4). The planktonic δ 13 C record has more variability compared to the benthic δ 13 C record (Fig. 4). During MIS 6, the benthic δ 13 C is relatively high (ca. 1.5 ‰), whilst the planktonic δ 13 C record fluctuates between -0.6 ‰ and -1.5 ‰. A decrease in the planktonic δ 13 C record (from -0.7 to -1.5 ‰) marks the middle of MIS 5. In contrast, the benthic δ 13 C remains stable 370 and low (ca. 1.2 ‰) throughout MIS 5 (Fig. 4). The passage from MIS 4 to MIS 3 is characterized by a shift from the low planktonic δ 13 C recorded during MIS 4 (-1.5 ‰) to higher planktonic δ 13 C (-0.5 ‰). Conversely, benthic δ 13 C values shift from high (1.8 ‰) to lower values (1.3 ‰). The passage from MIS 2 to MIS 1 is marked by a sharp decrease in planktonic and benthic δ 13 C (from -1.2 ‰ to -2.2 ‰ and from 1.8 ‰ to 1.0 ‰ respectively). The last two glacial intervals, in particular MIS 4, are marked by a stronger difference between benthic and planktonic δ 13 C 375 values (Fig. 4).

Elemental geochemistry
The Ti/Al and Si/Al ratios follow the same general trend. Variations in Ti/Al and Si/Al ratios are more marked during MIS 7 and MIS 3, in comparison with the more stable values recorded during other periods. Maximum average Ti/Al and Si/Al values are reached during glacials, whereas interglacials record the lowest values (Fig. 8). 380 The Zr/Al and Rb/Al ratios follow the same trend, whilst differing strongly from the Ti/Al and Si/Al records. The Zr/Al and Rb/Al ratios demonstrate overall low values throughout the core. However, higher Zr/Al and Rb/Al ratios are reached at the end of MIS 6, and during MIS 5 (ca. 400 cm) and MIS 3 (ca. 100 cm). In the same way as for https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License.
Ti/Al and Si/Al records, Zr/Al and Rb/Al ratios demonstrate an important variability during MIS 3, in comparison to other periods where the records are comparatively stable (Fig. 8). 385

Humid continental conditions, fluvial discharge and increased food availability
During interglacial periods, benthic foraminiferal assemblages are marked by high abundances of the infaunal Bulimina spp., U. mediterranea and B. spathulata. Several authors describe Bulimina spp. as characteristic for 390 eutrophic and dysoxic environments (Phleger and Soutar, 1973;Lutze and Coulbourn, 1984;Jorissen, 1987;Schmiedl et al., 2000). In the Mediterranean Sea, they are dominant in the vicinity of the Po river delta in the North Adriatic Sea and close to the Rhône River delta (Jorissen, 1987;Mojtahid et al., 2009). The shallow infaunal U. mediterranea and the opportunistic B. spathulata are known to demonstrate a positive correlation with organic matter flux (De Rijk et al., 2000;Schmiedl et al., 2000;Fontanier et al., 2002;Drinia and Dermitzakis, 2010). 395 Moreover, Bulimina spp. and U. mediterranea are reported to be able to feed on fresh but also more refractory organic matter (De Rijk et al., 2000;Koho et al., 2008;Dessandier et al., 2016). Based on these observations, the benthic foraminiferal assemblage during interglacials would support a high organic matter export to the seafloor. The overall higher TOC levels during interglacials confirm that the sediment during these periods was relatively enriched in organic matter in comparison to glacial periods (Fig. 4). High abundance of the shallow infaunal G. subglobosa 400 has been linked to the deposition of fresh phytodetritus on the seafloor after bloom events (Gooday, 1993;Fariduddin and Loubere, 1997;Suhr et al., 2003;Sun et al., 2006). It is typically found in high energy (e.g. steep flanks, ridges) and well-oxygenated environments (Mackensen et al., 1995;Milker et al., 2009), and is a common taxon of the Alboran Platform and of CWC environments (Margreth et al., 2009;Milker et al., 2009;Spezzaferri et al., 2014). Mackensen et al. (1995) noted that G. subglobosa dominated in areas of the South Atlantic Ocean where 405 the organic carbon flux did not exceed 1 g.cm -2 yr -1 . In contrast, in the Mediterranean Sea, B. marginata is restricted to sites with an organic carbon flux >2.5 g.cm -2 yr -1 , whilst B. aculeata is associated to a flux of 3 g.cm -2 yr -1 (De Rijk et al., 2000). The last two interglacials (MIS 7 and MIS 5) are marked by an increased abundance of G. subglobosa at early stages followed by a general decline. Buliminids follow a converse trend, particularly during MIS 5, with lower abundances at early stages (Fig. 7). This suggests that conditions during the later stages of interglacials became 410 increasingly eutrophic and in turn less oxygenated at the sediment/water interface, as the consumption of organic matter led to oxygen depletion. These more environmentally stressful conditions resulted in decreased foraminiferal diversity and a proliferation of opportunistic taxa (Fig. 7). Overall lower abundances of Miliolids, which are typically found in well-oxygenated environments (Murray, 2006), further confirm eutrophication coupled to lower oxygenation at the seafloor during interglacials, specifically towards the end of interglacials (Fig. 7). Schmiedl et al. (2010) link the high abundance of U. mediterranea in the Aegean Sea to humid climatic conditions and increased river runoff. This observation is in agreement with overall increased fluvial and reduced aeolian input during interglacial periods at BRI, as evidenced by the Al-normalized elemental ratios (Fig. 8). Increased fluvial input has been widely linked in the eastern Mediterranean to more humid continental conditions during interglacial 420 times in response to a northern shift of the African monsoon (e.g. Gasse, 2000;Gasse and Roberts, 2005;Osborne et al., 2008;Coulthard et al., 2013). In contrast, the Alboran Sea lies below the maximum Inter-Tropical Convergence Zone northward position and is sheltered by the Atlas Mountain chain (Rohling et al., 2002;Tuenter et al., 2003;Lavaysse et al., 2009) Moulouya River which takes its source in the High Atlas Mountains (Snousi, 2004;Emelyanov and Shimkus, 2012;Tekken and Kropp, 2012). The basin of the Moulouya River covers approximately 54,000 km 2 , hence representing the largest river basin in Northwest Africa (Emelyanov and Shimkus, 2012;Tekken and Kropp, 2012). We propose that the influence of warm and moist Atlantic air masses during interglacial periods led to warmer and more humid conditions over Northwest Africa and torrential rain fall. This would have led to a strengthening of the Moulouya 435 River's flow rate, hence triggering episodes of important terrestrial organic matter input at BRI. These events may have in turn caused eutrophication and oxygen depletion at the seafloor, compatible with the benthic foraminiferal assemblages. Dysoxic conditions during interglacial periods would have hampered coral proliferation, as suggested by the low mound aggradation rates (Fig. 9). However, dysoxic conditions may have been limited to the sediment, thus only affecting foraminiferal communities and not fully preventing colonial corals living above the sediment 440 surface to develop. Such vertical decoupling between sediment and pelagic ecosystems has previously been observed in modern Norwegian CWC reefs (Wehrmann et al., 2009). Overall, high food availability triggered by increased fluvial discharge appears to be a decisive parameter governing coral proliferation at BRI.

Enhanced surface and intermediate water mass mixing 445
During interglacial periods, the high sea level and the increased evaporation in the Mediterranean leads to a more important inflow of low salinity MAW through the Strait of Gibraltar (Sierro et al., 2005). Thus, surface waters in the Alboran Sea are, in comparison to glacial periods, warmer and less dense. This is also noticed in the planktonic δ 18 O record (Fig. 3). The enhanced MAW flow during interglacials triggers stronger Western and Eastern Alboran Gyres, resulting in better mixing and downwelling. Knowing that the Banc des Provençaux and BRI are situated at 450 relatively shallow water depths and in the path of the eastward circulating branch/Eastern Alboran Gyre (Lanoix, 1974;Viúdez and Tintoré, 1995;Fig. 10), and that mixing between surface and intermediate water masses is https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License.
documented to occur down to ca. 300 m water depth (Heburn and La Violette, 1990), it is conceivable that the corals living currently at 327 m depth were bathed by, or situated at the limit of mixing between surface and intermediate water masses during interglacial periods. Higher input of MAW into the Alboran Sea would lead to an increased 455 contribution of surface waters to intermediate water masses (ShW) and a deepening of the pycnocline. This would promote the formation of internal waves and increase turbulence at the seafloor of BRI, as suggested by the slightly higher ̅̅̅ values during interglacials ( Fig. 4 and 5), and would have favoured coral proliferation by increasing lateral nutrient supply (Fig. 10). The slight offset between planktonic and benthic δ 13 C records towards the end of MIS 7 and MIS 5 indicate that water masses were becoming more stratified towards the end of interglacials and that the 460 contribution of MAW to intermediate water masses was hence possibly decreasing. Maximum Bulimina spp.
abundance, minimum G. subglobosa abundance, and decreasing benthic foraminiferal diversity may suggest that reduced mixing, in concomitance with important fluvial discharge (section 6.1.1) led to oxygen depletion at the seafloor at the transition between interglacial and glacial periods. Severe oxygen depletion may explain the decline of corals at the transition from interglacial to glacial periods. 465

Variability of cold-water coral mound formation between interglacial periods
Highest coral content is reached during MIS 5 and corresponds to a maximum in Buliminid abundance. The Alnormalized elemental ratios suggest that aeolian input during MIS 5 was relatively stable, whilst fluvial input would have increased throughout (Fig. 8). These stable conditions would have favoured a long-lasting coral proliferation dominated by the scleractinian D. pertusum (Fig. 3). Marine Isotope Stage 9 and 7 are also dominated by D. 470 pertusum. Although MIS 7 is poorly constrained, Al-normalized elemental ratios would indicate that this time period was more unstable than the previous interglacial period (Fig. 8) Rigid erect branching bryozoans such as B. dichotoma are known to be fragile, and hence to prefer low energy environments, being unable to withstand strong bottom currents and turbulence (Scholz and Hillmer, 1995;Bjerager and Surlyk, 2007). Eutrophic environments dominated by infaunal benthic foraminifera (e.g. Bulimina spp.) are unfavourable for erect bryozoans, the high concentration of suspended food particles clogging up their feeding 500 apparatus (Holbourn et al., 2002). Low ̅̅̅ values and reduced TOC content in the sediment confirm that glacial periods were marked by weak bottom current velocities and organic matter flux ( Fig. 4 and 5). The presence of brachiopod/bivalve layers dominated by the brachiopod Gryphus vitreus also characterizes the glacial macrofauna ( Fig. 3). This species is found between 160 and 250 m depth along the Mediterranean continental margin and thrives in areas dominated by moderate bottom currents (Emig and Arnaud, 1988). Thus, this species' co-occurrence with 505 bryozoans confirms that variations in sea level stand, hydrodynamics and trophic conditions govern the change in macrofaunal dominance at BRI. Low organic matter flux during glacial periods has been related to predominantly arid conditions over North Africa, in association with a weak North African monsoon (Gasse, 2000;Sierro et al., 2005). Such arid conditions led to the complete or severe desiccation of major African lakes during the last glacial, such as Lake Victoria (Talbot and Livingstone, 1989;Johnson, 1996). High Ti/Al and Si/Al elemental ratios would 510 indicate that aeolian input prevailed during glacial periods at BRI (at the exception of MIS 3, section 6.4), hence confirming that continental conditions were arid at these times ( Fig. 8 and 9).
The reduced precipitation and retreat of vegetation would have led to a dwindling of fluvial discharge at BRI, as evidenced by generally low Zr/Al and Rb/Al elemental ratios (Fig. 8). Glacial benthic foraminiferal assemblages are 515 characterized by the dominance of large epibenthic suspension feeding foraminifera, such as C. lobatulus and D.
coronata, together with the infaunal C. laevigata (Fig. 7). This follows observations made by Stalder et al. (2018) who noticed increased abundances of Cibicides spp., D. coronata and C. laevigata during glacial periods at BRI. These species share a preference for high quality fresh marine organic matter (De Rijk et al., 2000;Milker et al., 2009, Stalder et al., 2018. minimum ̅̅̅̅ values, thus confirming this species' affinity for fine grained glacial material (Fig. 4). In the Arctic basins and Norwegian-Greenland Sea, the dominance of the epibenthic Cibicides wuellerstorfi (a relative of C. https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License. lobatulus) reflects a relative low flux of organic matter (Linke and Lutze; 1993). This species tolerates vertical flux 525 rates <2 g.cm -2 .yr -1 (Altenbach, 1989). The dominance of C. lobatulus, D. coronata, C. laevigata and Miliolids would thus indicate that the seafloor during glacial periods received less but higher quality organic matter and became more oxygenated in response to the stronger influence of intermediate and deep water masses (Fig. 10).
These observations suggest that more arid conditions during glacial periods led to a shift from a more fluvial to a more marine influenced environment (Fig. 10). We propose that weaker but comparatively fresher organic matter 530 input favoured the development of the bryozoan B. dichotoma. This assumption is supported by experimental observations demonstrating how erect bryozoans feed essentially on diatoms and that suspension feeding foraminifera use the same food sources (Winston, 1977;1981;Best and Thorpe, 1994;Goldstein, 1999). Lower nutrient input appears in contrast to have been detrimental for coral proliferation but would not have led to their complete disappearance ( Fig. 3 and 10). It can be hypothesized that there may exist a threshold in the quality and 535 quantity of organic matter determining which of D. pertusum or B. dichotoma dominates the benthic environment at BRI.

Increased stratification and deep water overturning
As highlighted previously, the dominant macrofauna and low ̅̅̅ values (Fig. 3, 4 and 5) during glacial intervals at BRI indicate weaker bottom currents. Wang et al. (2019) relate low off mound ̅̅̅̅ and high benthic foraminiferal 540 δ 13 C values at BRI during glacials to a dominant influence of MAW coinciding with a low sea level stand. However, whilst the benthic foraminiferal δ 13 C values from core MD13-3462G are indeed relatively high during glacial periods, the planktonic foraminiferal δ 13 C values do not follow the same trend (Fig. 4). The decoupling between the planktonic and benthic δ 13 C records during the two last glacial periods, noticeably during MIS 4, suggests that water mass stratification was greater than during interglacial periods and that the seafloor was not under the direct 545 influence of surface MAW. During glacial periods, the flow of MAW was reduced due to lower sea level and the reduced evaporation over the Mediterranean (Sierro et al., 2005). This would have reduced the contribution of MAW to ShW and weakened Western and Eastern Alboran Gyres, which would have in turn led to less mixing between surface and intermediate water masses, whilst conversely increasing stratification.

550
Modern observations show that recently formed dense waters do not necessarily reach the deep western Mediterranean but may, in contrast, be located at intermediate water depths, above 1500 m depth (Sparnocchia et al., 1995;Millot, 1999;Ercilla et al., 2016). Ercilla et al. (2016) further exposed that WMDW can be identified at depths shallower than 500 m depth along the Moroccan margin and that it contributes to the overlying ShW, whilst deep water overturning and ventilation peaked during MIS 2 (Cacho et al., 2006;Toucanne et al., 2012). Increased 555 oxygenation of the seafloor, as evidenced by the benthic foraminiferal assemblage (Fig. 7), may suggest that the contribution of well-ventilated deep and intermediate water masses at BRI was more important during glacials than during interglacials (Fig. 10). The physical shape of BRI possibly plays a role in the rise of deep waters during glacial periods. In addition, the heavier benthic C-isotope record and the abundance of fresh organic matter feeding foraminifera (C. lobatulus and D. coronata) during glacial periods could indicate that these waters were also 560 https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License. nutrient-rich. Although stratification between surface and intermediate water masses was greater during glacials, the stronger flow of well-ventilated WMDW at BRI would explain the higher oxygen availability at the seafloor. Overall during glacial periods, and in particular during the LGM, enhanced contribution of nutrient-rich and well-ventilated WDMW to overlying ShW, coupled to reduced fluvial input and turbulence, would have promoted bryozoan proliferation (Fig. 10). However, such environmental conditions would be detrimental for coral proliferation (Fig.  565 10).

Fluctuating environmental conditions during the last glacial period
The benthic and planktonic foraminifera δ 18 O and δ 13 C values indicate that environmental conditions were particularly unstable during the last glacial period, as suggested by previous studies (Cacho et al., 2000;Martrat et al., 2004;Pérez-Folgado et al., 2004;Cacho et al., 2006;Bout-Roumazeilles et al., 2007). The last glacial shows a 570 strong variability in macrofaunal and benthic foraminiferal assemblages. Maximum coral content is reached during MIS 3 (Fig. 3). This increased coral content is associated to higher numbers of G. subglobosa and C. laevigata, together with phases of higher Zr/Al and Rb/Al elemental ratios ( Fig. 7 and 8). These observations suggest that corals and the benthic foraminiferal community positively responded to short phases of increased surface productivity related to important continental runoff during MIS 3. This is supported by observations made by 575 Rogerson et al. (2018), who documented more humid conditions during MIS 3 in comparison to the more arid MIS 4 and 2. Humid conditions would hence have promoted coral proliferation through increased fluvial input at BRI, in the same way as during interglacial periods (section 6.1). Nevertheless, the dominance of G. subglobosa coupled to the absence of Bulimina spp. and U. mediterranea suggests that conditions were less eutrophic than during peak interglacial periods and that the organic matter reaching the seafloor may have been less degraded. 580 At BRI, high planktonic foraminiferal δ 18 O values during the last glacial are associated with increased Ti/Al and Si/Al elemental ratios (Fig. 8). There is evidence that during times of increased aridity, enhanced African winds blew north towards the Alboran Sea (Magri and Parra, 2002;Bout-Roumazeilles et al., 2007). During Heinrich Event 1, the existence of a steppe/semi-desertic flora around the Alboran borderlands points to cold and dry climatic 585 conditions (Combourieu Nebout et al. 2009). The association of high Ti/Al and Si/Al ratios with high planktonic foraminiferal δ 18 O values confirms that increased aridity on land coupled to strong winds were concomitant with lower sea surface temperatures at BRI (Fig. 8). The arid continental conditions during these particularly cold spells would have led to reduced continental runoff. This could in turn explain the overall dwindling of coral communities during these cold events (Fig. 3). Cacho et al. (1999) and Martrat et al. (2004) showed that sea surface temperature 590 minima matched higher planktonic G. bulloides δ 18 O values in the Alboran Sea during the last glacial. Moreover, these sea surface temperature minima are concurrent with North Atlantic Heinrich Events, i.e. the deposition of icerafted detritus from massive iceberg discharges during some of the colder stadials (Heinrich, 1988;Bond et al., 1992). Ice-rafted detritus layers were observed as far south as the Portuguese margin (Lebreiro et al., 1996;Bard et al., 2000;Schönfeld and Zahn, 2000), the Gulf of Cádiz (Llave et al., 2006;Toucanne et al., 2007) and the Moroccan 595 margin (Kudras and Thiede, 1970). Rapidly decreasing sea surface temperatures were also associated to North https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License. Atlantic Heinrich events of the Bermuda Rise (Sachs and Lehman, 1999) and in the Alboran Sea (Cacho et al., 1999). Moreover, based on palynological and mineralogical evidence, Bout-Roumazeilles et al. (2007) revealed an intensification of wind, dust erosion and transport toward the Alboran Sea in provenance of western Morocco during North Atlantic cold events. Based on these observations, we tentatively suggest that the dwindling of coral 600 communities during the last glacial period may also be linked to the inflow of Atlantic glacial freshwater during Atlantic cold events. More precise investigations are however needed to assert this relationship.

Interglacial-glacial transition phases
The western Mediterranean is marked by abrupt interglacial-glacial transitions (Martrat et al. 2004). Benthic foraminiferal assemblages and ̅̅̅ would confirm that the environment at BRI also experienced such abrupt 605 transitions. Indeed, the interglacial-glacial transitions are characterized by increased ̅̅̅ values and T. angulosa abundances ( Fig. 3 and 7). Trifarina angulosa is typical for current-swept areas and can withstand permanent winnowing (Mackensen et al., 1995;Schönfeld, 2002;Margreth et al., 2009). These results suggest that transition phases between interglacial and glacial periods were characterized by winnowing at the seafloor. In contrast, benthic foraminiferal assemblages and ̅̅̅ would indicate that transition phases from glacial to interglacial periods were not 610 marked by winnowing or erosional events. These observations differ from the ones drawn from Northeast Atlantic mounds, where winnowing and mass wasting are considered as precursor events for the re-initiation of coral proliferation during interglacials (Dorschel et al., 2005;Rüggeberg et al., 2007). Thus, the environmental mechanisms triggering the reset of coral proliferation at the onset of interglacials at BRI appear to be different from the Northeast Atlantic. The re-establishment of coral proliferation during the last two interglacials at BRI is 615 concomitant with an increase in Buliminid abundance. This increase in Buliminid abundance is coupled to higher Rb/Al values at the transition between MIS 6 and 5 (Fig. 7, 8). These observations confirm that the recovery of coral proliferation at BRI is tightly linked to an increase in river runoff, which in turn reflects more humid continental conditions. A similar process has been reported from the Viosca Knoll area, where the dispersal of terrestrial organic matter by the Mississippi River triggers an increase in primary productivity, providing nutrients for coral 620 communities (Mienis et al., 2012). As such, water mass rearrangements appear to be of secondary importance, whilst the rapid increase in fluvial discharge would be the primary factor triggering coral proliferation at BRI.

Coral proliferation and environmental forcing
In the Northeast and Northwest Atlantic, corals thrive during interglacial periods whilst their proliferation is halted 625 during glacial periods (Dorschel et al., 2005;Rüggeberg et al., 2007;Frank et al., 2009;2011;Matos et al., 2015;2017). Coral proliferation at BRI does not follow the same pattern. Indeed corals also develop during interglacial periods, but also to a lesser extent during glacial periods (Fig. 3). Coral proliferation in the Northeast Atlantic is controlled by the northward advance of subtropical waters and of MOW (Henry et al., 2014;Boavida et al., 2019), whereas corals at BRI are influenced by the interplay between inflowing MAW and outflowing LIW, ShW and 630 https://doi.org/10.5194/cp-2020-82 Preprint. Discussion started: 26 June 2020 c Author(s) 2020. CC BY 4.0 License.
WMDW (Stalder et al., 2015;Wang et al., 2019;this study). Environmental control on coral development in both regions shares similarities but also shows differences. The positive response of corals to increased bottom current velocity is important in both regions. This follows the general consensus that strong bottom currents are decisive for the development of corals (e.g. White et al., 2005;Mienis et al., 2007;Roberts et al., 2009). The topography of Brittlestar Ridge I may favour the formation of Taylor columns and the retention of organic matter, 635 such as observed in the Rockall Trough (Northeast Atlantic, White et al., 2007). However, benthic foraminiferal assemblages associated to phases of coral proliferation in the Northeast Atlantic (Rüggeberg et al., 2007) and in the Southeast Alboran Sea (this study) differ. Benthic foraminiferal assemblages associated to phases of sustained coral proliferation at Propeller Mound (Northeast Atlantic) are essentially characterized by large epibenthic foraminifera (C. lobatulus, Cibicides refulgens, D. coronata, and Planulina ariminensis) and the infaunal Trifarina bradyi 640 (Rüggeberg et al., 2007). In contrast, at BRI, higher abundances of C. lobatulus, D. coronata and T. angulosa are associated to glacial periods or transition phases between interglacial and glacial periods with low coral abundance, while small infaunal foraminifera dominate phases of coral proliferation (Fig. 7). These contrasting observations suggest differences in food supply and bottom current regimes. Corals in the Northeast Atlantic thrive on fresh marine-derived organic matter resulting from the North Atlantic blooms which are fuelled by upwelling (Dickinson 645 et al., 1980). In contrast, corals at BRI are likely supplied by plankton blooms triggered by the input of degraded fluvial organic matter during interglacial times, whilst aeolian dust input allows corals to survive during glacial times by triggering local moderate bloom events in the area. In this regard, coral mounds situated in the Southeast Alboran Sea show more similarities to mounds located in the Viosca Knoll area or in the Gulf of Cadiz (Wienberg et al., 2010;Mienis et al., 2012). The respective shallow location and proximity of BRI to the continent explains the higher 650 influence of continental runoff on coral communities than in the deeper Northeast Atlantic sites. It can hence be expected that corals at BRI show higher sensibility to shifting continental climatic conditions.
Moreover, mound aggradation rates during the Holocene for core MD13-3462G (Fig. 9)  Indeed, mound aggradation in the Porcupine Seabight is restricted to interglacial periods, whilst glacials are marked by winnowing and erosive events (Rüggeberg et al., 2007;Frank et al., 2011). Long-term coral mound formation at BRI took place during interglacial and glacial periods, though at much lower aggradation rates than in the Porcupine Seabight ( Fig. 9; Frank et al., 2011). Mound aggradation rates in core MD13-3462G are comparable to inactive or 670 abandoned reefs in the Porcupine Seabight, i.e. <5 cm.ky -1 (Frank et al., 2011), thus suggesting that CWCs did not thrive at the site of core MD13-3462G but rather developed under stressful, possibly dysoxic, environmental conditions. Average long-term mound aggradation rates at BRI show more similarities with mounds situated along the Mauritanian margin that developed during the last glacial (28-45 cm.ky -1 ) but also during the last interglacial period (16 cm.ky -1 ; Wienberg et al., 2018;Wienberg and Titschak, 2015). In contrast with Atlantic CWC mounds, 675 mounds from the East Melilla Coral Province show a high contribution of the erect cheleistome bryozoan B.
dichotoma. Based on mound aggradation rates and macrofaunal content, we propose that B. dichotoma communities favoured mound formation at BRI, noticeably during glacial periods, by capturing fine-grained sediments in a similar way as CWCs do ( Fig. 3 and 9). As such, mounds at BRI stand out and may be considered as mixed B. dichotoma/CWC mounds, rather than CWC mounds per se. 680

Conclusions
The multiproxy study of core MD13-3462G provides information on the long-term development of a cold-water coral mound at Brittlestar Ridge I. Three important points can be concluded: (1) Cold-water corals develop mainly during interglacial periods. Their growth is promoted by the combination of 685 increased fluvial input and enhanced influence of Alboran Gyres. Increased fluvial organic matter inputs are driven by the increased impact of warm and moist Atlantic air masses with intensified Western and Eastern Alboran Gyres that lead to more important turnover between surface and intermediate water masses. This phenomenon is promoted by enhanced Modified Atlantic Water inflow at the Strait of Gibraltar. The seafloor was possibly depleted in oxygen at the end of interglacial phases. These results demonstrate the paramount importance of enhanced fluvial input as a 690 trigger for cold-water growth in the Southeastern Alboran Sea.
(2) Glacial periods are unfavourable for cold-water corals; in contrast the bryozoan Buskea dichotoma is more suited to glacial environmental conditions. The retreat of corals during glacial periods is triggered by arid continental conditions. Aeolian dust was the main fertilizing influence and may have enabled corals to survive throughout glacial 700 periods.
(3) Average coral mound aggradation rates are particularly low, varying between 1 and 9 cm.ky -1 . Mound formation takes place during glacial periods as well as during interglacial periods. Low mound aggradation rates during interglacials and glacials suggest that corals did not thrive but rather developed under stressful environmental 705 conditions at Brittlestar Ridge I. The erect cheleistome bryozoan Buskea dichotoma plays an important role in the long-term mound formation at Brittlestar Ridge I, noticeably during glacial periods. Overall, mound development at Brittlestar Ridge I is controlled by alternating aeolian and fluvial inputs, in response to North Atlantic climate dynamics.

710
From a wider perspective, this study demonstrates how cold-water coral environments can benefit from both fluvial and aeolian terrestrial input, during respectively interglacial and glacial periods. These results underline how coldwater coral systems are capable of withstanding important environmental changes and to survive and adapt to different climatic conditions.       Table 2. Details of Uranium-series isotope measurements (U/Th) carried out on 10 coral fragments. a) replicate of IUP-8504; b) replicate of IUP-8505; c) replicate of IUP-8507. Brackets denote activity ratios. All errors are 2 of the mean analytical