RESEARCH ARTICLE | EARTH, ATMOSPHERIC, AND PLANETARY SCIENCESENVIRONMENTAL SCIENCES A Late Pleistocene coastal ecosystem in French Guiana was hyperdiverse relative to today Pierre-O livier Antoinea,1 , Linde N. Wieringaa , Sylvain Adneta, Orangel Aguilerab , Stéphanie C. Bodinc , Stephen Cairnsd , Carlos A. Conejeros- Vargase, Jean-J acques Cornéef , Žilvinas Ežerinskisg , Jan Fietzkeh , Natacha O. Gribenskii, Sandrine Grouardj , Austin Hendyk, Carina Hoornl , Renaud Joannes- Boyaum,n , Martin R. Langero , Javier Luquep , Laurent Marivauxa , Pierre Moissetteq , Kees Noorenl , Frédéric Quillévérér, Justina Šapolaitėg , Matteo Sciumbatal,s, Pierre G. Vallat , Nina H. Witteveenl , Alexandre Casanovau , Simon Clavierv , Philibert Bidgrainu, Marjorie Gallayw , Mathieu Rhonéw, and Arnauld Heuretf,u Edited by Nils Stenseth, Universitetet i Oslo, Oslo, Norway; received July 8, 2023; accepted February 15, 2024 Warmer temperatures and higher sea level than today characterized the Last Interglacial interval [Pleistocene, 128 to 116 thousand years ago (ka)]. This period is a remarka- Significance ble deep-t ime analog for temperature and sea-l evel conditions as projected for 2100 AD, yet there has been no evidence of fossil assemblages in the equatorial Atlantic. The Last Interglacial interval (128 Here, we report foraminifer, metazoan (mollusks, bony fish, bryozoans, decapods, and to 116 ka) is a remarkable sharks among others), and plant communities of coastal tropical marine and mangrove deep-t ime analog for temperature affinities, dating precisely from a ca. 130 to 115 ka time interval near the Equator, at and sea- level conditions as Kourou, in French Guiana. These communities include ca. 230 recent species, some projected for 2100, that had not being endangered today and/or first recorded as fossils. The hyperdiverse Kourou mollusk been documented in the assemblage suggests stronger affinities between Guianese and Caribbean coastal waters equatorial Atlantic thus far. Here, by the Last Interglacial than today, questioning the structuring role of the Amazon Plume on tropical Western Atlantic communities at the time. Grassland-d ominated pollen, we report hyperdiverse fossil phytoliths, and charcoals from younger deposits in the same sections attest to a marine communities of coastal marine retreat and dryer conditions during the onset of the last glacial (ca. 110 to 50 ka), with and mangrove affinities, dating a savanna- dominated landscape and episodes of fire. Charcoals from the last millennia back from this interval and suggest human presence in a mosaic of modern-l ike continental habitats. Our results unearthed at the Europe’s provide key information about the ecology and biogeography of pristine Pleistocene Spaceport in Kourou, French tropical coastal ecosystems, especially relevant regarding the—widely anthropogenic— Guiana. Mollusk assemblages ongoing global warming. suggest stronger ecological affinities between Guianas and the French Guiana | ancient ecosystems | past biodiversity | Last Interglacial | climate change Caribbean than today. Grassland- dominated pollen, phytoliths, and During the last 2 My, climatic oscillations induced environmental fluctuations that resulted charcoals from younger deposits in drastic changes in biotic distribution all over the globe (1–3). For example, global sea in the same sections attest to a level fluctuated up to 120 m between glacial and interglacial maxima, primarily forced by marine retreat and dryer orbital cycles (4, 5). Such fluctuations led to iterative emergences and drownings of conditions during the Last Glacial low- elevation coastal areas, with deep consequences on marine taxa (3, 6). The Last Period (100 to 50 ka). These Interglacial interval (LIG; Marine Isotope Stage [MIS] 5e; 128 to 116 ka) is characterized records provide key ecological and by up to 4 to 6 m higher sea-l evel conditions and overall climatic conditions 2 to 4 °C biogeographic information about warmer than today (7–10), making this period a deep-t ime analog for temperature and sea- level conditions as projected for 2100 AD (11). Despite the existence of numerous data Late Pleistocene tropical coastal available worldwide, very little is known about LIG low-l and biotic assemblages near the ecosystems prior to human Equator, their ecology, and biogeography, more especially in the Atlantic region (8, 10–14) influence. (Fig. 1A and Dataset S1). This gap notably prevents from characterizing the tropical Atlantic biotic communities in the penultimate Earth’s warm episode (11, 15–17). The Guianas comprise a vast territory (ca. 2.5 million km2) near the Equator in South America. Today, this region shelters high levels of taxonomic diversity in both terrestrial The authors declare no competing interest. and aquatic ecosystems (19–21), but we know almost nothing about its past biodiversity This article is a PNAS Direct Submission. (22, 23). Guianese coastal areas are covered by mangrove vegetation and tidally influenced Copyright © 2024 the Author(s). Published by PNAS. river banks while, more inland, herbaceous swamps and savannas are followed by marsh This article is distributed under Creative Commons Attribution- NonCommercial- NoDerivatives License 4.0 and evergreen woodlands (24). This dense vegetation further hampers access to potential (CC BY- NC- ND). fossil-y ielding outcrops. Guianese coasts are strongly impacted by a huge flux of surface Although PNAS asks authors to adhere to United Nations waters of Andean- Amazonian origin (25), termed the Amazon Plume (AP). This north- naming conventions for maps (https://www.un.org/geospatial/mapsgeo), our policy is to publish maps as westward flux strongly structures the composition of recent tropical Western Atlantic provided by the authors. biotic communities, without a clue about its role in the past (25–27). 1To whom correspondence may be addressed. Email: Here, we report a hyperdiverse marine LIG assemblage, separated by a hiatus from pierre- olivier.antoine@umontpellier.fr. younger continental fossil communities, in a section spanning the last ca. 130 ka. This This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas. succession was uncovered during the titanic earthworks that were undertaken in 2015 to 2311597121/-/ DCSupplemental. 2020 for the Ariane 6 ELA4 launching pad (here KOU- AR6) at Europe’s Spaceport, Published March 25, 2024. PNAS  2024  Vol. 121  No. 14  e2311597121 https://doi.org/10.1073/pnas.2311597121   1 of 11 Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. Fig. 1. Location map (A) of the Middle and Late Pleistocene marine mollusk localities, between 60°S and 60°N in the Atlantic Ocean (Dataset S1), as retrieved in the Paleobiology Database (18), and of the Pleistocene–Holocene KOU-A R6 sections and sampling sites (B and C), by the Ariane 6 launcher pad (KOU- AR6-0 1 to -0 6). The dashed line denotes the cross-s ections as seen in Fig. 4. Composite stratigraphic section (D) at Kourou, French Guiana, with sedimentological descriptions and multi-p roxy age constraints. Sampling levels are approximate (for detailed information on each section and sampling efforts, see SI Appendix, Fig. S11). ASL, above sea level; B.P., before present; C, clay; cal, calibrated; Cg, conglomerate; cont., continental; Sa, sand; Si, siltite. Kourou, French Guiana (FG; Fig. 1 B–D). Our research team the behavior of equatorial Atlantic biodiversity during the LIG; includes many specialists from the fields of paleontology and pale- iii) the deep history of the AP as a major structuring element of obotany to fully reconstruct these ecosystems by using as many Western Atlantic marine communities; and iv) the relative abun- different taxonomic groups as possible. As a result, these fossil dance of now critically endangered marine taxa, in the absence of communities comprise over 270 taxa of foraminifers, metazoans human footprint. (mollusks, bryozoans, decapods, ray-fi nned fish, and selachians among others), and plants, further providing a glimpse into the Results and Interpretation composition of equatorial coastal ecosystems (marine and terres- trial), during both the LIG and the Last Glacial Period (LGP, ca. Pleistocene–Holocene Sections at the Ariane 6 Launch Pad, 115 to 12 ka). Kourou, FG. Six trenches and ditches were excavated and investigated These ancient ecosystems predate human arrival in the Guianas for their paleontological content over a 0.5- km2 surface in 2019 [ca. 10 ka B.P. (28–30)]. They can contribute to testing: i) the (KOU-A R6- 01 to -0 5) and 2021 (KOU-A R6-0 6; Fig. 1C). Common regional effects of the global marine retreat related to the LGP; ii) and salient features in these geological sections enabled us to build 2 of 11   https://doi.org/10.1073/pnas.2311597121 pnas.org Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. an 8-m - thick composite section (SI Appendix, Figs. S1–S6 and S11). must be bracketed. Two in situ charcoal samples from successive The base of the section is formed by the top of a Paleoproterozoic continental fluvial channels at KOU-A R6-0 4 provided consistent granitoid (crystalline basement), which coincides with current sea 14C ages at 47,053 ± 572 and 43,091 ± 284 calibrated years B.P. level (−1/+1 m). The granitoid is overlain by around 0.5 m of pink [cal B.P.], dating the base and the lower part of Unit C, respectively saprolites (chemically weathered granitoids), followed by a 7-m - thick (SI Appendix, Table S6). This time span falls within MIS 3, just sedimentary deposit (Fig.  1D and SI  Appendix, Fig.  S11). The preceding the LGM within the LGP (9, 33). Drastic climatic latter deposit includes three successive units with an erosional base changes are recorded during this interval at high and mid-l atitudes, (Units A–C). Based on its fossil content (see below), the lowermost but climatic models do consider that no seasonal temperature shift part of this sedimentary ensemble (Unit A, around 4 m thick) is occurred between MIS 3 and the LGM at the Equator (34). The unambiguously of marine origin and referable to the Middle–Late Holocene marine Demerara clays are not recorded in the investigated Pleistocene Coswine Formation (Fm.), documented in coastal areas loci (31, 32). Finally, the two successive riverine channels situated of FG (31, 32). just below the surface at KOU-A R6-0 4 (top of Unit C) were dated In all trenches, the fossil-r ich ensemble starts by marine deposits based on charcoals and yielded 14C ages of 1,938 ± 120 and 804 ± (Unit A lying on previously emerged bedrock, transformed into 55 cal B.P. (Fig. 1D). At that time, human settlement and land use paleosols higher up and topped by an emersion surface, thereby are well- documented locally (35, 36). Accordingly, the time intervals documenting a transgressive/regressive sedimentary cycle). More documented in the KOU-A R6 sections would be ca. 130 to 115 specifically, Unit A consists of a basal gray oyster- rich conglom- ka (Unit A, marine), ca. 110 to ca. 50 ka (Unit B, continental) and erate, transgressive and overlain by gray silty clays (1.5 m). It is 47 to 1 ka (Unit C, continental). covered by a khaki conglomerate in trenches KOU-A R6-0 3 and - 05, with quartz pebbles and oxidized elements, laterally equivalent to The Biotic Communities at KOU-A R6 (Pleistocene to Holocene). khaki or ocher sands in other trenches. Both conglomerates yield All the fossil specimens from the sampled sections are referable calcareous, phosphatic and siliceous, or carbonized marine mac- to living species, i.e., no extinct taxon is documented thus far at rofossils, plus foraminifers and palynomorphs (SI Appendix). KOU-A R6. Above the khaki conglomerate/sand, a regressive sequence starts with 1.5 m of variegated silty paleosols (blueish and ocher or beige Unit A (ca. 130 to 115 ka), Mangrove to Shallow Marine Environ­ and yellow), yielding only siliceous and phosphatized marine fos- ment. Unit A (Fig. 4A) documents a very short high sea- level sils (brachiopods and fish; KOU-A R6-0 6) attesting to their marine interval spanning the LIG. Based on our chronological constraints, origin, with a subsequent weathering due to pedogenesis. These this is most likely MIS 5e, 128 to 116 ka (9). This sequence yielded variegated silts become reddish upward and turn into either an hyperdiverse assemblages comprising 229 distinct taxa belonging iron crust (at KOU- AR6- 03, - 04 and 05) or a yellow quartz-r ich to a wide array of phyla, including foraminifers, mollusks, ray- siltite, yielding only continental palynomorphs and phytoliths (at finned fish, bryozoans, decapods, and sharks among others (Fig. 2), KOU- AR6- 06) and topping Unit A. Above it, Unit B is charac- but also plants (charcoals, phytoliths, and pollen; Fig. 3) derived terized, in all trenches, by around 1.5 m of brownish-o range, gray from nearby coastal habitats. In general, mollusks and decapods or yellow silts of continental origin. At KOU- AR6-0 6, Unit B dominate over other groups, with perfectly preserved delicate begins with a 15- cm- thick dark microconglomeratic peat [around shells, pointing to low- energy habitats and preservation in situ. 2.5 m above sea level (PN- 15a- b pollen and phytolith samples); Foraminifer communities (SI Appendix, Table S7) are mainly Fig. 1D]. The top of this continental sequence is cut or weathered composed of hyaline-p erforate benthic taxa (Fig. 2 A–C), indicative and replaced by modern soils and human-d isturbed surfaces in for shallow intertidal mangrove and subtidal environments (11 most of the studied sections. Nevertheless, the KOU-A R6- 04 species), and one individual of planktonic foraminifer (Fig. 2D). section, culminating 2 to 4 m above all other investigated sections The smallest benthic species (Nonion subturgidum, Elphidium magel- (SI Appendix, Fig. S11), provides information on a third unit of lanicum, Cerebrina claricerviculata, and Fursenkoina sp.) usually live fluvial origin, here termed Unit C. This unit consists of 2.5-m - thick in low- oxygenated sediments, while other ones tolerate low- salinity beige-b rownish coarse sands intercalated with charcoal-r ich micro- conditions and potentially occur in mangrove habitats and estuaries conglomeratic lenses and channels. with variable salinity conditions (Ammonia). All other benthic foraminifers are comparatively shallow marine, subtidal taxa, usually Age Constraints on the KOU-A R6 Sections. Nineteen samples were occurring in nearshore shallow-w ater environments with algae or dated through independent proxies to estimate the ages of Units A seagrass vegetation. The open ocean influence was probably low. and C (SI Appendix, Tables S2–S6). From the base of Unit A (basal The foraminifers found in all trenches strongly recall the associations conglomerate), aragonitic Astrangia corals were dated by U-Th at 131 observed in a mangrove estuary in northern Brazil, with a significant ± 15 ka by laser ablation and by U-Th at a maximum age of 135.8 ± marine tidal influence (37). 1.1 ka by conventional solution multicollector inductively coupled Sponges are only represented by Entobia boreholes in oyster plasma mass spectrometry (MC- ICP-M S; Fig. 1D). Higher up in shells (Fig. 2E). Cnidarians are documented by the octocorallian Unit A, quartz- rich silts located 1.7 m above the base of the Coswine gorgonian Pacifigorgia, at KOU- AR6-0 6 (Fig. 2G) and >1,700 Fm. at KOU- AR6- 06 were dated through optically stimulated specimens of a single scleractinian species, Astrangia rathbuni, luminescence (OSL) with a minimum age estimate of 104.6 ± 17.9 either growing as solitary corallites or small colonies (Fig. 2F). ka (sample OSL- 2). Coswine clays formed a transgressive sequence Astrangia rathbuni was recognized in all sampled marine levels, around the Middle–Late Pleistocene transition (31, 32), further with a much higher density at KOU-A R6- 06 than in other documenting the highest sea level during the LIG [MIS 5e: 128 trenches (SI Appendix, Table S8). to 116 ka (9)], and reaching 4 to 6 m above the modern sea level The trenches KOU-A R6-0 1, - 03, and -0 5 yielded 19 species of (8, 10). At the top of Unit A, OSL-1 , a sample of ocher to beige bryozoans, mostly typical of tropical shallow waters. Most of these clayey silts situated 1.1 m above OSL-2 , was constrained with a taxa also occur in the present- day coastal waters of Brazil (e.g., minimum age estimate of 119.7 ± 10.2 ka through OSL dating refs. 38 and 39). Warm- water genera (Biflustra, Steginoporella, (corresponding possibly to MIS 5e or to the glacial stadial MIS 5d). Antropora, and Nellia) are well represented in both recent and For Unit B, there is no radioisotopic dating available, and its age fossil Kourou records. The predominance of encrusting forms PNAS  2024  Vol. 121  No. 14  e2311597121 https://doi.org/10.1073/pnas.2311597121   3 of 11 Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. Fig. 2. Marine foraminifer and metazoan communities from Unit A (~130 to 115 ka, Last Interglacial) and associated taxa in KOU-A R6 sections, Kourou, FG. Foraminifers (A–D): (A) Quinqueloculina seminula (KOU- AR6-0 5 Top); (B) Eponides repandus (KOU- AR6- 03 base); (C) Ammonia veneta (KOU-A R6- 03 base); and (D) Globigerina bulloides (KOU- AR6- 05 Top). (E) Detail of an oyster shell, encrusted by bryozoans and a small colony of Astrangia rathbuni corals and perforated by Entobia sponge boreholes (KOU- AR6-0 3 base). Cnidarians (F and G): (F) Astrangia rathbuni, colony (KOU- AR6- 06 base), and (G) Pacifigorgia sp., basal portion (KOU- AR6-0 6 base, 2 mm). Bryozoans (H–J): (H) Biflustra arborescens (KOU- AR6- 01); (I) Steginoporella magnilabris (KOU-A R6- 03 Top); and (J) Conopeum loki (KOU- AR6-0 5 Top). (K) Brachiopods, Discradisca antillarum, in external view (KOU- AR6- 06 Top). Mollusks (L–Y). (L) Crassostrea sp., flat (Right) valve in inner view, perforated by pholadid bivalves and Entobia (KOU- AR6- 02); (M) Lunarca ovalis (KOU- AR6- 05); (N) Sheldonella bisulcata (KOU- AR6- 05); (O) Leptopecten bavayi (KOU-A R6-0 5); (P) Caryocorbula contracta (KOU- AR6-0 5); (Q) Vitta virginea (KOU- AR6-0 5); (R) Stigmaulax cayennensis (KOU-A R6- 05); (S) Costoanachis avara (KOU- AR6- 03); (T) Eulima bifasciata (KOU-A R6- 03); (U) Mulinia cleryana (KOU-A R6- 05); (V) Stramonita haemastoma (KOU-A R6- 05); (W) Chione cancellata (KOU-A R6- 05); (X) Crassostrea rhizophorae (KOU- AR6-0 3); (Y) Crassinella lunulata, under natural light (Left) and 395- nm wavelength UV light (Right). Crustaceans. (Z) Balanomorphs: Amphibalanus sp., individual with articulated wall plates (KOU-A R6- 01). Axiidean (A’–D’), anomuran (E’ and F’), and brachyuran decapods (G’ and H’). Neocallichirus sp., left cheliped dactyl, outer margin (A’) and left cheliped propodus, outer margin (B’). Callichiridae indet., left cheliped dactyl, inner margin (C’) and left cheliped pollex, outer margin (D’). Pachycheles sp., left cheliped, outer margin, showing the palm and pollex (E’). Petrolisthes sp., pollex of right cheliped, outer margin view (F’). Eriphioidea (Eriphia/Menippe), dactyl of right cheliped, inner margin view (G’) and distalmost part of pollex of right cheliped (H’). ?Persephona sp., merus of cheliped indet (I’). Portunidae indet., fragment of cheliped pollex (J’). Echinoderms. (K’) Arbacia punctulata (KOU- AR6- 06 Base), test fragment. Elasmobranchs (L’–O’). (L’) Isogomphodon oxyrhynchus (upper tooth); (M’) Rhizoprionodon sp. (lower lateral tooth); (N’) Ginglymostoma cirratum (lower tooth); (O’) Hypanus sp. Bony fish otoliths in rotate views (P’ and T’). (P’) Aspistor luniscutis (KOU- AR6- 05 Top); (Q’) Cathorops spixii (KOU- AR6- 05 Top); (R’) Thalassophryne sp. (KOU-A R6- 03 Base); (S’) Macrodon ancylodon (KOU- AR6- 03 Base); (T’) Stellifer rastrifer (KOU- AR6- 05 Top). Bony fish teeth and bones (U’–W’). (U’) Erythrinidae indet., tooth (KOU- AR6- 03 Base); (V’) unidentified freshwater siluriform, pectoral spine (KOU- AR6- 03 Base); (W’) Nettastomatidae indet., dentary (KOU-A R6- 03 Base). [Scale bars, 100 µm (A–D), 5 mm (E, F, Y, A’–D’, G’–I’, and M’), 2 mm (G, H, F’, J’, and V’), 1 mm (I–K, Z, E’, K’, L’, N’–Q’, R’–U’, and W’), and 10 mm (L–X).] 4 of 11   https://doi.org/10.1073/pnas.2311597121 pnas.org Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. Fig. 3. Pollen and phytolith assemblages and charcoal fragments from Unit A (ca. 130 to 115 ka, Last Interglacial; A–K) and Unit B [LGP (ca. 110 to ca. 50 ka; L–Y) in KOU-A R6 sections, Kourou, FG. (A) Pollen diagram of Unit A (KOU- AR6- PN9), with typical palynomorphs (B–I) and charcoals (J and K); (B) Rhizophora sp., red mangrove; (C) Acrostichum sp., mangrove fern; (D) Hedyosmum sp.; (E) Schefflera sp.; (F) Peltaea sp.; (G) Ilex sp.; (H) Attalea type; (I) Symphonia sp.; (J) cf. Rhizophora; (K) Chrysobalanaceae. (L) Pollen diagram of Unit B (KOU-A R6- PN15), with typical palynomorphs (M–Q); (M) Mauritia sp.; (N) Schultesia sp.; (O) Rubiaceae indet.; (P) Poaceae indet.; (Q) Poaceae indet.; (R) Phytolith diagram of Unit B with typical phytoliths (KOU- AR6- PN15; S–X) and charcoal (KOU- AR04 Base; Y); (S) Poaceae, rondel; (T) Poaceae, bilobate; (U) Poaceae, Bambusoideae; (V) Poaceae, Panicoideae; (W) Zingiberales, Heliconia; (X) Woody dicot; (Y) cf. Myrtaceae (charcoal). [Scale bars, 20 µm (B–E, G, H, M–Q, and S–X), 50 µm (F), or 100 µm (J, K, and Y).] suggests a shallow depositional environment affected by freshwater several species which retain colored patterns visible to the naked influxes associated with increased turbidity, as in mangrove and eye [e.g., Vitta (Fig. 2Q), Pilsbryspira] or revealed under UV light oyster- rich settings (40). [e.g., Crassinella, Olivella; Fig. 2Y]. Most molluscan taxa have Around 200 calcareous tubes of unidentified polychaete worms affinities to intertidal and shallow subtidal sands, muds, or rocks are documented in the marine sequence of all trenches. Siliceous and several species are characteristic of mangrove habitats (e.g., shells of a single brachiopod taxon (Discradisca antillarum) are Vitta virginea, Isognomon radiatus). recorded at Kourou, with hundreds of specimens over the entire The crustacean arthropods are particularly dominant at Kourou, marine unit and in all sampled trenches. with thousands of specimens retrieved from the sediments. All these Mollusks vastly dominate other phyla in both taxonomic diver- specimens belong to either barnacles (balanomorph cirripeds), crabs sity and specimen numbers (Fig. 2 L–Y). They include two species or shrimps (decapods). The barnacles are notably represented by a of scaphopods (rare), 35 species of bivalves, and 50 species of large amount of disconnected wall plates of Amphibalanus, and a gastropods. Bivalves and snails are recorded by thousands of indi- single complete specimen (Fig. 2Z). The decapods are represented viduals in all marine levels that were sampled, with shallow water mostly by hundreds of isolated claw fragments, mainly of mobile Costoanachis avara (Fig. 2S), Sheldonella bisulcata (Fig. 2N), and and fixed fingers (SI Appendix, Table S11). The decapods comprise Chione cancellata (Fig. 2W) most abundant. In terms of richness eight morphotypes, including two species of mud shrimps (Fig. 2 and evenness, KOU- AR6- 03 is most diverse with 59 species A’–D’), three species of false crabs, or anomurans (Fig. 2 E’ and F’), (SI Appendix, Table S10). The state of preservation is exquisite for and three species of true crabs, or brachyurans (Fig. 2 G’–J’). PNAS  2024  Vol. 121  No. 14  e2311597121 https://doi.org/10.1073/pnas.2311597121   5 of 11 Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. Anomurans include filter feeders found in reefs, under rocks, shell (Fig. 2 U’–W’), belonging to 35 species (SI Appendix, Table S14). beds, or mangroves. Small claw fragments further document a pos- Sciaenid perciforms (16 species, with five distinct Stellifer) and sible paguroid. The overall decapod association indicates proximity ariid siluriforms (eight species) widely outnumber other taxo- to mangroves, with soft sediments hosting Neocallichirus mud nomic groups in the sample. KOU- AR6- 03 is by far the richest shrimps (feeding on seagrass and algae) and purse crabs Persephona. locality, with 32 species; SI Appendix, Table S14). Thirteen species This association points to intertidal–subtidal tropical to temperate are recognized in two or three localities, pointing to a certain waters (0 to 50 m), with Western Atlantic, Caribbean, and tropical heterogeneity between the samples (Table 1). Eastern Pacific affinities (Persephona). Plant composition and diversity in the marine unit is revealed Echinoderms were retrieved in high numbers in all marine by fossil charcoals, pollen, spores, and phytoliths (Fig. 3 and s amples, nevertheless pointing to a low taxonomic diversity (three SI Appendix, Tables S15–S17). As for charcoals, red mangrove (cf. species). The echinoderm community is overdominated by the Rhizophora sp.), boarwood (cf. Symphonia globulifera) and two Atlantic purple sea urchin Arbacia punctulata (Fig. 2K’) in all sam- representatives of Chrysobalanaceae and Myrtaceae were recog- pled levels and trenches (SI Appendix, Table S12). In stark contrast, nized. At KOU-A R6- 06, the base of the same unit yielded phy- we retrieved only a few dozens of test fragments of two unidentified toliths referable to unidentified woody eudicot and Asteraceae, in heart urchins and two plates of an astropectinid sea star. PN9A and PN9C pollen samples, respectively. The corresponding No marine mammals or seabirds were preserved, but elasmo- palynological assemblage (Fig. 3 A–I), with a low pollen concen- branch (sharks and rays; Fig. 2 L’–O’) were identified in all tration (around 700 grains cm−3), is dominated by Rhizophora trenches: four species of rays (whipray, eagle ray, saw fish, and pollen (80%), followed by spores of the mangrove fern Acrostichum cownose ray) and seven species of sharks, including smalltail, dag- (3.5%). No Avicennia pollen grains were found. Pollen of tree gernose, sharpnose, and lemon sharks, as well as small scoophead species accounts for 9% of the pollen sum and reflects the influx hammer sharks and a nurse shark. Daggernose sharks and whiprays of hinterland and lowland (swamp) forest trees (41). Herb and dominate the elasmobranch fauna in terms of specimens and vine pollen is relatively rare (5%) and dominated by Poaceae and occurrences (SI Appendix, Table S13). Bony fish are dominantly Asteraceae. Asteraceae pollen grains were only found in the PN9C documented by otoliths (Fig. 2 P’–T’), but also by bones and teeth sample, also containing one Asteraceae phytolith. The top of this Table  1. Taxonomic diversity of marine and continental communities from the KOU-A R6 Pleistocene–Holocene sections, Kourou, FG Marine Continental Unit A ca. Unit B ca. 110 ka to Higher taxa 130 to 115 ka ca. 50 ka Unit C 47 to 1 ka Total (distinct taxa) Foraminifera (foraminiferans) 12 - - 12 Porifera (sponges) 1 - - 1 Cnidaria (corals and gorgons) 2 - - 2 Bryozoa (bryozoans) 19 - - 19 Annelida (serpulid worms) 1 - - 1 Mollusca (mollusks) 87 - - 87 Scaphopoda (scaphopods) 2 - - - Bivalvia (bivalves) 35 - - - Gasteropoda (gastropods) 50 - - - Brachiopoda (brachiopods) 1 - - 1 Arthropoda (arthropods) 13 - - 13 Cirripedia (cirripeds) 2 - - - Decapoda (decapods) 11 - - - Echinodermata (echinoderms) 3 - - 3 Echinoidea (urchins) 2 - - - Asteroidea (sea stars) 1 - - - Vertebrata (vertebrates) 46 - - 46 Elasmobranchii (rays and sharks) 11 - - - Actinopterygii (ray- finned fishes) 35 - - - Plantae (plants) 41 35 22 92 Charcoal 10 8 22 - Phytoliths 2 18 - - Pollen 30 11 - - Total per unit 226 35 22 277 Some plant taxa have been recognized from charcoal and/or pollen in the same unit or in distinct units, hence distinct figures in the last line (distinct taxa per unit) and the last column (distinct taxa identified at KOU- AR6, regardless of the yielding unit). 6 of 11   https://doi.org/10.1073/pnas.2311597121 pnas.org Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. Fig. 4. Hypothesized evolution of Late Pleistocene–Holocene landscapes in FG, using fossil proxies, sedimentary facies, and radioisotopic age constraints available at KOU-A R6 sections, Kourou. (A) During MIS 5e (ca. 130 to 115 ka; Unit A, marine and mangrove settings). (B) During MIS 3c–3b (47 to 39 ka; Unit C, base; continental settings). (C) Today (Unit C, Top, coastal-c ontinental settings). Although documenting fires at its base, Unit B was not sufficiently time-c onstrained to be satisfactorily depicted here. The presence of a mangrove landscape bordering the Great Amazon Reef System (GARS) at 47 to 39 ka is hypothetical. Main environmental and ecological features related to the sea- level changes, observed locally/regionally over the last climatic cycle, are summarized in the boxes in A–C. (D) Sea- level curve and marine isotopic stages, modified from refs. 4 and 9; sea- surface temperatures modified from ref. 45. The location of the GARS is hypothesized based on ref. 6. For further details regarding age constraints, sampling sections, fossil content, biogeographic/ecological affinities, and corresponding levels, see Datasets S1–S3. unit has been comprehensively sampled at KOU-A R6- 06 for pal- The phytolith assemblages counted in the basal dark peat at ynomorphs and phytoliths (samples PN10 to 14; SI Appendix, KOU-A R6-0 6 (Fig. 3 R–X) are dominated by grasses (65%) in both Table S16). PN10 to 13 only provided a few phytoliths, and PN10 PN15A and PN15B with 65% grass, 31 and 17% woody eudicots, to 14 was also devoid of palynological content. Grass phytoliths respectively, and almost no palm phytoliths (<1%). Most grass phy- first occur at PN12 (dated at 104.6 ± 17.9 ka, OSL-2 ), with a toliths encountered are from Panicoideae and Bambusoideae panicoid cross and a bilobate [C3 and C4 grasses (42, 43)], plus (SI Appendix, Table S16 and Fig. S12). Bilobates and rondels are also a fused and two rugose spheroids (woody eudicots). PN14 yielded common, produced by a wide array of monocot grasses (42, 43). phytolith assemblages dominated by grass phytoliths (70%), as in Phytoliths from Pooideae (wavy trapezoids) and Chloridoideae PN15A- B (base of Unit B, see below). (squat saddles) are rare (<1%). Strikingly, a high percentage of phy- toliths were burnt (28%), especially specimens of Cyperaceae, Unit B (ca. 110 –50 ka), Coastal Savanna and Dry forest. Only Heliconia and Zingiberales. This assemblage suggests that a savanna plant remains were retrieved in this unit. vegetation had started growing locally way before 50 ka and spread Tree charcoals were identified at KOU-A R6- 04 Base (Fig. 3Y around and settled sustainably. Previous phytolith studies showed and SI Appendix, Table S15). The assemblage comprises notably that the natural vegetation of seasonally flooded/coastal Holocene Hadroanthus cf. serratifolius (ipê) and a close relative, cf. Drypetes savannas in FG consisted of Cyperaceae, Marantaceae, and Heliconia sp., Pterocarpus- like Leguminosae, red mangrove, as well as uni- herbs and panicoid and oryzoid grasses, with an overall high abun- dentified Melastomataceae, Myrtaceae-l ike dicots. Today, these dance of grass phytoliths (44). taxa represent trees and shrubs from the primary, riverine or dry The pollen concentration of the PN15 sample is much higher forest, savanna, or mangrove and suggest distinct vegetation suc- than in Unit A (around 20,600 grains cm−3), with a high relative cession stages at ca. 47 ka cal B.P. abundance of Poaceae (49%) and Spermacoceae (36%) pollen, PNAS  2024  Vol. 121  No. 14  e2311597121 https://doi.org/10.1073/pnas.2311597121   7 of 11 Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. indicative of open and disturbed vegetation (Fig. 3 L–Q) prior to the ocean being open at the northeast, with terra-fi rma forests to 45 ka cal B.P. in the ELA4 area (Fig. 4B). Many Poaceae pollen the southwest (Figs. 1 and 4A). Detrital elements sourced from grains are relatively large (50 to 64 μm), furthering the presence the continent, through small freshwater streams, as shown by the of Panicoideae and Bambusoideae grass phytoliths. Mangrove presence of a synbranchid swamp eel, as well as erythrinid and (2.4%) and tree (3.5%) pollen grains are rare (SI Appendix, non- ariid catfish specimens in several loci (Dataset S2). Table S17). Conversely, the large amount of charred plant frag- By ca. 110 ka, the sea had already retreated, as the marine fossils ments (“microcharcoals”) in the pollen slides is notable. The high are reworked or oxidized in most trenches, in good agreement with number of macro-c harcoals and high percentage of burned phy- the global eustatic history (Fig. 4D). The macrofossil content of toliths indicate recurring fires at the site during the concerned the level topping the marine sequence, with only siliceous/phos- time interval, i.e., prior to 47 ka cal B.P. (age of the base of the phatic fossil remains of marine origin (Discradisca brachiopods and overlying Unit C; see below), and further consistent with a glacial fish teeth), points to a differential preservation and a post-b urial stadial (MIS 4: 72 to 58 ka; Fig. 4D). dissolution of calcareous specimens (including foraminifers) in marine deposits posteriorly pedogenetized. Phytoliths sampled in Unit C (47 to 1 ka), Dry/swamp Forest and Savanna to Coastal the very top of Unit A point to continental affinities, with the first Savanna and Chenier Plain. Only macroscopic charcoals were conspicuous occurrence of grass phytoliths recorded in the PN14 hand- picked at KOU- AR6-0 4, in several levels from Unit C, sample. The iron crust topping this unit at various loci further spanning the 47 to 1 ka time interval (MIS 3c–1). More than 60 suggests intense surface weathering during a short time (Fig. 1). fragments, some of them from tree stumps, were retrieved in a brown Afterward, the Kourou sites registered a strong continental signa- conglomerate (“Mid”), 14C-d ated at 47,053 ± 572 cal B.P. They ture through pollen, phytoliths, and charcoal, indicating extensive attest to the most speciose tree community uncovered here through plain savanna to dry forest fringed conditions (Units B and C; charcoals, with at least 15 distinct tree taxa (SI Appendix, Table S15). Fig. 4 B and C). The occurrence of natural (i.e., pre- human) fires Mouriri sp. is the most abundant tree, followed by Chaunochiton is revealed by burnt phytoliths (ca. 28%) and microcharcoals at kappleri and a close relative, two close allies of Stryphnodendron, two KOU- AR6- 06- PN15, and by macro- charcoals before 47 ka cal B.P. unidentified Chrysobalanaceae, Lecythidaceae, cf. Anacardiaceae/ This suggests that a dry interval may have occurred regionally, per- Burseraceae and ipê. Just above, floodplain deposits and a silty litter haps coinciding with a glacial interval (possibly MIS 4). dated at 43,091 ± 284 cal B.P. yielded charcoals of unidentified Charcoals suggest that tree diversity culminated ca. 47 ka cal affinities and bootlace tree, respectively. The top levels, dated from B.P., in the transition zone between a savanna and a coastal/ the last millennia (14C ages of 1,938 ± 120 and 804 ± 55 cal swamp/riparian forest, or a mosaic of habitats including dryland B.P.), yielded charcoals of unidentified Anacardiaceae/Burseraceae, forest. The presence of bootlace tree might attest to the presence hog plum, cf. Chrysobalanaceae, Mabea sp. in the older layer and of a swamp forest in the surroundings by 43 ka cal B.P., >100 km bootlace tree, Rubiaceae anatomically close to batahua, as well as away from the coastline (Fig. 4B). unidentified Chrysobalanaceae and Leguminosae in the younger After a gap in the charcoal record, the 2-k a old level at KOU- one. Pollen and phytoliths were neither sampled nor investigated AR6-0 4 yields clues of wet-p lain and riparian habitats, with Mabea in Unit C, except for the last millennia (35). and hog plums. The latter tree, with edible fruits (mombin), might also be related to human occupation, documented in the area at Local Landscape Evolution Since the LIG (Fig. 4 A–C). The marine that time (36). The youngest charcoal sample (ca. 800 cal B.P.) unit represented in Unit A (Fig. 4A) documents a very short high points to a swamp or riparian forest. The absence of phytolith and sea- level interval spanning the LIG [most likely MIS 5e, 128 to pollen record in Unit C impedes characterizing further the last 116 ka (9)]. Rhizophora trees and Acrostichum ferns nowadays only pre- Columbian seasonally flooded local savannas (35). thrive in stable and mature mangroves (46). Their dominance in the pollen assemblages (including pollen clumps) and as charcoals LIG Marine Communities from KOU-A R6: Taxonomic Diversity at the base of Unit A suggests that a mangrove ecosystem occurred and Ecological Affinities. The estimated sea surface temperature for in the close surroundings (SI  Appendix, Tables  S15 and S16). the Guiana Basin during the LIG was higher (28.9 °C) than today Pollen recovery further attests to the existence of montane (e.g., [28.1 °C (45)] (Fig. 4D). Warm periods (e.g., today and LIG) are Alnus, probably Andean- sourced) and lowland swamp- forest characterized by an equatorial depletion of marine diversity, with trees, herb, and vines. This is supported by the recognition of diverse-m ost areas shifting toward higher latitudes, due to equatorial dry-f orest and back- mangrove tree charcoals (Chrysobalanaceae temperatures being higher than the physiological tolerance of and Myrtaceae; Fig.  4A). Aquatic communities confirm the certain species (11, 15, 16): brachyurans, bivalves and gastropods proximity of a mangrove belt (e.g., mangrove oysters, decapods, are particularly thermo- sensitive, and today their species diversity and foraminifers), with shallow-w ater marine habitats occurring decreases in waters exceeding 20 °C (17). To assess the diversity of near the sampling points (around 5-m depth), as supported by the KOU- AR6 metazoan paleocommunities by the warm LIG, we used co-o ccurrence of many mollusks. A certain habitat disparity can a comprehensive survey performed in the 1950s for recent marine be inferred from coeval samples, with softer substrates at KOU- organisms of FG as a reference (47). We compared species and genus AR6-0 1 than at other loci (abundant spatangoid urchins) or diversity against depth range and type of substrate (mud, muddy more wave- or tide- related energy in KOU-A R6- 06 and - 05 Top sands, dead shells, and sands), for corals, mollusks (gastropods, than anywhere else—high proportion of broken specimens and bivalves, and scaphopods), brachyuran decapods, echinoderms (sea solitary corallites—thereby pointing to disturbed settings. This stars and urchins), and bony fish. We chose this 1950s survey as i) it is landscape strongly recalls the environments of Marajó Island in the unparalleled as a sampling effort and ii) it was undertaken before the Amazon delta and marginal islands with strong marine and tidal last decades’ massive erosion of marine biodiversity of anthropogenic influence, as furthered by the co-o ccurrence of various bony fish origin (48). As a result, only sea stars denote a lower cumulative taxa with ana-, amphi- or catadromous life cycles (SI Appendix, diversity in the LIG than in recent samples of compatible substrates Table S14). Contrastingly, this type of highly speciose shallow- and wider bathymetric range (SI Appendix, Table S18). Despite a water environments is not recorded in FG today (20, 47). The comparatively limited sampling effort, further restricted by a ca. local and regional topographies at our sites are consistent with 5- m-d eep depositional setting (unfavorable to the development of 8 of 11   https://doi.org/10.1073/pnas.2311597121 pnas.org Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. species- rich assemblages), corals, brachyurans, urchins, and bony fish higher vulnerability to overfishing than to warming for these taxa have a similar alpha- diversity in the Kourou LIG samples and in recent (15–17). As such, this study on past communities may open unique samples, while KOU- AR6 mollusk alpha- diversity widely exceeds perspectives regarding in- depth taxonomic studies, community FG’s recent one (twice to four times higher; SI Appendix, Table S18). ecology analyses, and extinction risks of the considered assemblages. This disparity in molluscan species richness may in part be due to Hopefully, future records would follow and help bridging stratigraphic the time-a veraging of skeletonized components in death assemblages, and biogeographic gaps on a regional scale (14). accumulating over hundreds to thousands of years along the FG Environmental conditions during the climax of MIS 5e strik- shoreline prior to anthropogenic impacts. Changing environmental ingly echo the most pessimistic scenarios for global warming and factors allowed different species to inhabit these estuarine and coastal sea- level rise in 2100 AD, notably in Central and South America areas as time passed, with those death assemblages accumulating an (45, 52). This analogy has a particular resonance, as the Guianese ever- increasing species richness (49, 50). coastal areas located at less than 10 m above modern sea level are Indeed, temperature does not explain the entire history of tropical critically concerned by the current sea-l evel and temperature rises marine diversity (14) and causes may be multiple for this coastal and subsequent cascading risks, such as vector-b orne disease epi- diversity drop between the LIG and the 1950s in FG, notably due demics, and drastic changes in Amazon biome dynamics (52–54). to the large- scale marine retreat that occurred meanwhile and Indeed, low- elevation coastal zones hosting around 20, 55 and exhumed most of the Guianese continental shelf during the LGP 80% of FG, Guyana, and Suriname inhabitants, respectively, are (Fig. 4). At a shorter timescale, the shallow marine areas of FG have at risk of being entirely flooded before 2100 AD (52, 55, 56), as also experienced deep changes in terms of coastal environment and are irremovable infrastructures of the area (e.g., airports and substrates over the last centuries, with mangroves and mudbanks— Europe’s Spaceport). As the future may be learned from the past characterized by low taxonomic richness and high substrate homo- in terms of marine biodiversity and coastal ecosystem dynamics geneity—spreading northwestward all over clear waters. This massive and fate (16, 48, 57), we hope that the present work would help siltation is mainly due to the AP, a huge coastal flux of warm, hypo- raise collective awareness of the major environmental upheavals saline, nutrient- rich, and turbid surface waters of Andean- Amazonian that the region may experience in the next century, especially for origin (25). This flow plays today a prominent role as a barrier decision-m akers, whether local or transnational. between Brazil and the Caribbean for a wide array of marine animals, including reef fish and gastropods (26, 27). Indeed, its influence on Materials and Methods biotic communities over the last interglacials is unknown (25–27). We used the recent geographic distribution of mollusk species rec- The material was collected through handpicking (charcoals) and screen-w ashing ognized at KOU- AR6 as a proxy for testing the ecological affinities of more than one metric ton of sediments [with 2 mm, 1 mm, and 0.7 mm meshes of this coastal Guianese fossil assemblage (Datasets S2 and S3 and for most fossil groups and smaller meshes for microvertebrates (0.4 mm) and Fig. 4). As a result, the closest relationships are retrieved between foraminifers (150 and 63 μm)]. The fossil specimens belong to the collections of KOU-A R6 and recent mollusk communities from the Guianas, with the Université de Guyane in Cayenne. When large numbers of specimens were 53 species in common. While these Guianese communities tightly available for a given species, other specimens have been further stored in the group with a Southwestern and South Caribbean cluster, with collections of the Université de Montpellier. Eastern Brazil as an offshoot (SI Appendix, Fig. S14), they have Age constraints were provided through U-T h datings on corals (Unit A), OSL restricted affinities with the Amazon and Northeastern Brazil living dating on quartz grains (Units A and B) and 14C datings (Unit C). Taxonomic identifications were undertaken by recognized specialists of each communities. This somewhat contrasts with today’s spatial pattern, group of interest, aiming at retaining the most accurate and conservative assign- where Guianese coastal communities are both impoverished and ment level. These assignments range from species to family level, highly depending diverging taxonomically from the Caribbean ones, under the major on completeness of the concerned KOU-A R6 records and/or on current knowledge influence of the AP (27). This result suggests that the spatial distri- discrepancies about recent taxa themselves (SI Appendix, Tables S7–S17). bution of Western Atlantic tropical mollusks was not fully shaped For comparing the diversity of KOU-A R6 past communities (five marine sam- by a barrier prefiguring the AP around LIG times, either related to ples over a 0.5-k m2 surface at a ca. 5-m depth; this work), we used a comprehen- salinity, turbidity, nutrient- balance, or temperature discrepancies. sive survey (47) performed in 1954 to 1957 on recent marine organisms on the The virtual lack of fossil record documenting this penultimate warm Guianese Continental Plate (400 samples over ca. 40,000 km2, including 110 period near the equator (Fig. 1A) impedes getting a broader picture samples for a 0 to 29- m depth range and 272 for a 20 to 49- m depth range) on on this very issue, but it clearly highlights where sampling efforts compatible substrates and a wider bathymetric range (mud, 0 to 30-m depth; should increase in the future. muddy sands, dead shells, and sands, 20 to 49- m depth). To define the ecological affinities of KOU- AR6 LIG mollusks at the Western Inputs for Conservation Biology and Perspectives for the Near Atlantic scale, a taxon/area matrix, widely inspired from that of a recent bioge- Future. About 30 recent species, among foraminifers, cnidarians, ographic analysis (27), was built for the 74 species having a well- defined dis- sharks, bryozoans, brachiopods, and mollusks (14 species), have tribution area today (Datasets S2 and S3). For that, we used the mapper tool of their first and/or earliest fossil record in LIG deposits at KOU- AR6 the Ocean Biodiversity Information System repository (https://mapper.obis.org), (Dataset S2), which significantly adds to their knowledge and may completed by an atlas of FG’s mollusks (58). We then ran UPGMA and parsimony help for conservation policies (13). Most vertebrate taxa recognized analyses with PAUP* 4.0a.169 (59). in KOU- AR6 still inhabit FG seawater today, as endemic species Data, Materials, and Software Availability. All study data are included in the of the Central-S outh American Atlantic coasts (e.g., Isogomphodon article and/or supporting information. oxyrhynchus, Carcharhinus porosus, and Sphyrna media; SI Appendix, Tables S13 and S14). Due to overfishing, the sawfish, the smalltail, daggernose, and scoophead sharks are critically endangered while the ACKNOWLEDGMENTS. We are grateful to Nils Stenseth (invited editor), but also Albula vulpes bonefish is nearly threatened and the Cynoscion acoupa to Moriaki Yasuhara and an anonymous reviewer for their constructive and stim- weakfish is vulnerable (Dataset S2). The presence of these species 125 ulating remarks on previous versions of the manuscript. We are deeply indebted ka ago in the same area and their current small range (Dataset S2) to François Catzeflis for paving the way to this French Guianese fossil record, to suggest low mobility on 105 to 106- year timescales, as predicted by Martijn van den Bel and everybody at INRAP in Cayenne- Matoury for granting spatial distribution models (13, 51). It would also be indicative of access to their facilities in 2018, to Michel Brossard, Michel Macarit, and Lionel PNAS  2024  Vol. 121  No. 14  e2311597121 https://doi.org/10.1073/pnas.2311597121   9 of 11 Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. Hautier for their participation to preliminary field and screen-w ashing operations, Laboratory, Center for Physical Sciences and Technology, Vilnius 10257, Lithuania; h to Carlos Jaramillo for preparing early samples and for fruitful discussion, and Geomar, Helmholtz Centre for Ocean Research Kiel, Kiel 24148, Germany; iInstitute of Geological Sciences, Oeschger Centre for Climate Change Research, University of Bern, to Henry Hooghiemstra for providing relevant suggestions on our manuscript. Bern 3012, Switzerland; jArchéozoologie et Archéobotanique—Sociétés, Pratiques et We are particularly thankful to the Centre National d’Etudes Spatiales (Sandrine Environnements, CNRS, Muséum National d’Histoire Naturelle, Paris 75005, France; k - Invertebrate Paleontology Department, Natural History Museum of Los Angeles County, Richard and Henri Brunet Lavigne), the Centre Spatial Guyanais, Eiffage (Henri Los Angeles, CA 90007; lEcosystem & Landscape Dynamics Department, Institute for Beny), and Sodexo (Patrick Raverat). This work was funded by the French Agence Biodiversity and Ecosystem Dynamics, Universiteit van Amsterdam, Amsterdam 1098 Nationale de la Recherche (ANR) in the framework of both the LabEx CEBA XH, The Netherlands; mGeoarchaeology and Archaeometry Research Group, Southern Cross GeoScience, Southern Cross University, East Lismore, NSW 2480, Australia; nCentre (ANR- 10- LABX-2 5-0 1), through the projects Source, NeotroPhyl, Timespan and for Anthropological Research, University of Johannesburg, Johannesburg 2092, South Emergence, and the GAARAnti project (ANR-1 7-C E31-0 009). O.A. acknowledges Africa; oArbeitsgruppe Mikropaläontologie, Institut für Geowissenschaften, Paläontologie, p funding from the Brazilian Council of Science and Technological Development Universität Bonn, Bonn 53115, Germany; Department of Zoology, Museum of Zoology, University of Cambridge, Cambridge CB2 3EJ, United Kingdom; qDepartment of Historical (CNPq 304693/2021-9 ). N.H.W. acknowledges funding from the European Geology- Paleontology, National and Kapodistrian University of Athens, Panepistimiopolis, Research Council (ERC 2019 StG 853394). R.J.- B acknowledges funding from Zografou, Athens 15784, Greece; rLaboratoire de Géologie de Lyon - Terre, Planètes, the Australian Research Council (ARC DP220100195 and LE200100022). P.G.V. Environnement, Université Claude Bernard Lyon 1, Ecole Normale Supérieure de Lyon, CNRS, Villeurbanne F-6 9622, France; sSection Systems Ecology, Amsterdam Institute for acknowledges funding from the French ANR- PIA program (ANR- 18- MPGA- 0006). Life and Environment, Vrije Universiteit, Amsterdam 1081 BT, The Netherlands; tEquipe This is ISEM-S ud article no. 2024- 051. Tectonique, Reliefs et Bassins, Institut des Sciences de la Terre, Université Grenoble Alpes, Université Savoie Mont Blanc, CNRS, Université Gustave Eiffel, Grenoble 38058, France; uDépartement Formation et Recherche Sciences et Technologie, Université de Guyane, Cayenne 97300, Guyane; vOnikha, Kourou 97310, Guyane; and wOffice de l’Eau de Guyane, Author affiliations: aEquipe de Paléontologie, Institut des Sciences de l’Évolution de Cayenne 97300, Guyane Montpellier, Univ Montpellier, CNRS, Institut de Recherche pour le Développement, Author contributions: P.- O.A. and A. Heuret designed research; P.-O .A., L.N.W., S.A., O.A., Montpellier 34095, France; bPaleoecology and Global Changes Laboratory, Marine S.C.B., S. Cairns, C.A.C.-V ., J.- J.C., N.O.G., S.G., A. Hendy, C.H., R.J.- B., M.R.L., J.L., L.M., P.M., K.N., Biology Department, Fluminense Federal University, Niterói 24210-2 01, Rio de Janeiro, F.Q., M.S., P.G.V., N.H.W., A.C., S. Clavier, P.B., M.G., M.R., and A. Heuret performed research; Brazil; cDepartment of Paleoanthropology, Senckenberg Research Institute, Frankfurt am P.- O.A., L.N.W., S.A., O.A., S.C.B., J.- J.C., Ž.E., J.F., N.O.G., S.G., A. Hendy, C.H., R.J.- B., M.R.L., Main 60325, Germany; dDepartment of Invertebrate Zoology, Smithsonian Institution, J.L., P.M., K.N., F.Q., J.Š., M.S., P.G.V., and N.H.W. analyzed data; P.- O.A., M.G., and M.R. National Museum of Natural History, Washington D.C. 20013-7 012; eDepartamento de performed fieldwork; L.N.W., O.A., S.C.B., S. Cairns, C.A.C.- V., Ž.E., J.F., N.O.G., S.G., A. Hendy, Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México, Coyoacán, C.H., M.R.L., J.L., P.M., K.N., F.Q., J.Š., and N.H.W. contributed to writing; L.N.M., S.A., J.- J.C., Ciudad de México 04510, México; fEquipe Dynamique de la Lithosphère, Géosciences M.S., A.C., S. Clavier, P.B., and A. Heuret performed fieldwork, contributed to writing; and Montpellier, Univ Montpellier, CNRS, Montpellier 34095, France; gMass Spectrometry P.- O.A. wrote the paper. 1. H. P. Comes, J. W. Kadereit, The effect of Quaternary climatic changes on plant distribution and 25. M. Pujos, J.- M. Froidefond, Water masses and suspended matter circulation on the French Guiana evolution. Trends Plant Sci. 3, 432–438 (1998). continental shelf. Contin. Shelf Res. 15, 1157–1171 (1995). 2. M. Dynesius, R. Jansson, Evolutionary consequences of changes in species’ geographical 26. S. R. Floeter et al., Atlantic reef fish biogeography and evolution. J. Biogeogr. 35, 22–47 (2008). distributions driven by Milankovitch climate oscillations. Proc. Natl. Acad. Sci. U.S.A. 97, 9115–9120 27. E. Giachini Tosetto, A. Bertrand, S. Neumann- Leitão, M. Nogueira Júnior, The Amazon River plume, a (2000). barrier to animal dispersal in the Western Tropical Atlantic. Sci. Rep. 12, 537 (2022). 3. W. B. Ludt, L. A. Rocha, Shifting seas: The impacts of Pleistocene sea- level fluctuations on the 28. M. Mestre, S. Delpech, Plateau des Mines (Excavation report, Saint-L aurent du Maroni, INRAP, evolution of tropical marine taxa. J. Biogeogr. 42, 25–38 (2015). Cayenne, 2008). 4. L. B. Railsback et al., An optimized scheme of lettered marine isotope substages for the last 1.0 29. G. Odonne et al., Long- term influence of early human occupations on current forests of the Guiana million years, and the climatostratigraphic nature of isotope stages and substages. Quat. Sci. Rev. Shield. Ecology 100, e02806 (2019). 111, 94–106 (2015). 30. M. G. Plew, L. Daggers, The Archaeology of Guyana (University of Guyana Press, ed. 2, 2022). 5. R. M. Spratt, L. E. Lisiecki, A Late Pleistocene sea level stack. Clim. Past 12, 1079–1092 (2016). 31. E. Palvadeau, Géodynamique quaternaire de la Guyane Française (BRGM, Université de Bretagne 6. P. Giresse et al., Beachrocks of the last low sea level, substrate of the Great Amazon Reef system Occidentale, 1999), p. 255. along the outer Guiana shelf. Geo-M arine Lett. 43, 10 (2023). 32. G. Brunier et al., Evolution of the French Guiana coast from Late Pleistocene to Holocene based on 7. A. Dutton, K. Lambeck, Ice volume and sea level during the last interglacial. Science 337, 216–219 chenier and beach sand dating. Region. Envir. Change 22, 122 (2022). (2012). 33. P. U. Clark et al., The last glacial maximum. Science 325, 710.e714 (2009). 8. A. Rovere et al., The analysis of Last Interglacial (MIS 5e) relative sea-l evel indicators: Reconstructing 34. C. J. Van Meerbeeck, H. Renssen, D. M. Roche, How did marine isotope stage 3 and last glacial sea- level in a warmer world. Earth- Sci. Rev. 159, 404–427 (2016). maximum climates differ?—Perspectives from equilibrium simulations. Clim. Past 5, 33–51 9. P. L. Gibbard, M. J. Head, The Quaternary period. In Geologic Time Scale 2020, F. M. Gradstein, (2009). J. G. Ogg, M. D. Schmitz, G. M. Ogg, Eds. (Elsevier, 2020), pp. 1217–1255. 35. J. Iriarte et al., Fire-f ree land use in pre- 1492 Amazonian savannas. Proc. Natl. Acad. Sci. U.S.A. 109, 10. K. Rubio- Sandoval et al., A review of last interglacial sea-l evel proxies in the western Atlantic and 6473–6478 (2012). southwestern Caribbean, from Brazil to Honduras. Earth Syst. Sci. Data 13, 4819–4845 (2021). 36. M. M. van den Bel, Nouveaux apports sur l’archéologie du littoral de Guyane: De la préhistoire à la 11. W. Kiessling et al., Equatorial decline of reef corals during the last Pleistocene interglacial. conquête. J. Soc. Am. 104, 105–152 (2018). Proc. Natl. Acad. Sci. U.S.A. 109, 21378–21383 (2012). 37. N. Sariaslan, M. R. Langer, Atypical, high- diversity assemblages of foraminifera in a mangrove 12. A. Rovere et al., The World atlas of last interglacial shorelines (version 1.0). Earth Syst. Sci. Data 15, estuary in northern Brazil. Biogeosci. 18, 1–18 (2021). 1–23 (2023). 38. A. C. S. Almeida, F. B. C. Souza, L. M. Vieira, Malacostegine bryozoans (Bryozoa: Cheilostomata) 13. S. Finnegan et al., Using the fossil record to understand extinction risk and inform marine from Bahia State, northeast Brazil: Taxonomy and non- indigenous species. Marine Biodiv. 48, conservation in a changing world. Ann. Rev. Marine Sci. 16, 1–27 (2023). 1463–1488 (2017). 14. M. Yasuhara et al., Hotspots of Cenozoic tropical marine biodiversity. Oceanogr. Mar. Biol. 60, 39. L. M. Vieira, A. E. Migotto, J. E. Winston, Synopsis and annotated checklist of Recent marine Bryozoa 243–300 (2022). from Brazil. Zootaxa 1810, 1–39 (2008). 15. O. Rama- Corredor et al., Parallelisms between sea surface temperature changes in the western 40. D. J. Hughes, J. B. C. Jackson, Do constant environments promote complexity of form?: The tropical Atlantic (Guiana Basin) and high latitude climate signals over the last 140 000 years. distribution of bryozoan polymorphism as a test of hypotheses. Evolution 44, 889–905 (1990). Clim. Past 11, 1297–1311 (2015). 41. T. Van der Hammen, A palynological study on the Quaternary of British Guiana. Leidse Geol. Meded. 16. M. Yasuhara et al., Past and future decline of tropical pelagic biodiversity. Proc. Natl. Acad. Sci. U.S.A. 29, 125–180 (1963). 117, 12891–12896 (2020). 42. D. R. Piperno, Phytoliths: A Comprehensive Guide for Archaeologists and Paleoecologists 17. M. Yasuhara, C. A. Deutsch, Paleobiology provides glimpses of future ocean: Fossil records from (Rowman Altamira, 2006). tropical oceans predict biodiversity loss in a warmer world. Science 375, 25–26 (2022). 43. C. Crifò, C. A. Strömberg, Small- scale spatial resolution of the soil phytolith record in a rainforest 18. The Paleobiology Database, Download Records. http://paleobiodb.org/data1.2/occs/list. and a dry forest in Costa Rica: Applications to the deep- time fossil phytolith record. Palaeogeogr. csv?datainfo&rowcount&base_name=Mollusca&interval=Middle%20Pleistocene,Late%20Pleistoc Palaeoclimatol. Palaeoecol. 537, 109107 (2020). ene&cc=AFR,EUR,NOA,SOA&latmin=-6 0&latmax=60. Accessed 29 November 2023. 44. J. Iriarte et al., Late Holocene Neotropical agricultural landscapes: Phytolith and stable carbon 19. J. Boggan et al., Checklist of the Plants of the Guianas (Museum of Natural History, Smithsonian isotope analysis of raised fields from French Guianan coastal savannahs. J. Archaeol. Sci. 37, Institution, Smithsonian’s Biological Diversity of the Guianas Program series 30, ed. 2, 1997). 2984–2994 (2010). 20. L.- F. Artigas et al., Marine biodiversity in French Guiana: Estuarine, coastal, and shelf ecosystems 45. J. B. C. Jackson et al., Historical overfishing and the recent collapse of coastal ecosystems. Science under the influence of Amazonian waters. Gayana 67, 302–326 (2003). 293, 629–637 (2001). 21. A. Sánchez Meseguer et al., Diversification dynamics in the Neotropics through time, clades, and 46. S. Guitet, R. Pélissier, O. Brunaux, G. Jaouen, D. Sabatier, Geomorphological landscape features biogeographic regions. Elife 11, e74503 (2022). explain floristic patterns in French Guiana rainforest. Biodiv. Conserv. 24, 1215–1237 (2015). 22. A. K. Gibbs, C. N. Barron, F. Tabeart, The Geology of the Guiana Shield (Oxford University Press, 47. J. Durand, Notes sur le plateau continental guyanais: Les éléments principaux de la faune et leurs New York, 1993), vol. 246. relations avec le fond. Cah. ORSTOM 3, 1–99 (1959). 23. P.- O. Antoine et al., Un inventaire fossile de la Guyane: Historique et nouvelles perspectives. 48. C. Chaudhary et al., Global warming is causing a more pronounced dip in marine species richness Géologues 206, 28–30 (2020). around the equator. Proc. Natl. Acad. Sci. U.S.A. 118, 1–6 (2021). 24. J. C. Lindeman, The vegetation of the Coastal Region of Suriname. Results of the scientific 49. G. M. Staff, E. N. Powell, The paleoecological significance of diversity: The effect of time averaging expedition to Suriname 1948—49 botanical series No. 1. Meded. Bot. Mus. Herb. Rijks Univ. Utrecht and differential preservation on macroinvertebrate species richness in death assemblages. 113, 1–135 (1952). Palaeogeogr. Palaeoclimatol. Palaeoecol. 63, 73–89 (2008). 10 of 11   https://doi.org/10.1073/pnas.2311597121 pnas.org Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210. 50. A. Tomašových, S. M. Kidwell, Fidelity of variation in species composition and diversity partitioning 55. A. Heuret, S. Bodin, L. Marivaux, P.- O. Antoine, Des étoiles aux fossiles. Quand Ariane 6 dévoile la by death assemblages: Time- averaging transfers diversity from beta to alpha levels. Paleobiol. 35, Guyane pléistocène. Une saison en Guyane 7, 4–9 (2021). 94–118 (2009). 56. G. McGranahan, D. Balk, B. Anderson, The rising tide: Assessing the risks of climate change and 51. C. J. Reddin et al., Victims of ancient hyperthermal events herald the fates of marine clades and human settlements in low elevation coastal zones. Environ. Urban. 19, 17–37 (2007). traits under global warming. Glob. Change Biol. 27, 868–878 (2021). 57. H. K. Lotze, L. McClenachan, “Marine historical ecology: Informing the future by learning from the 52. I. Hagen et al., Climate change- related risks and adaptation potential in Central and South America past” in Marine Community Ecology and Conservation, M. D. Bertness et al., Eds. (Sinauer Associates during the 21st century. Environ. Res. Lett. 17, 033002 (2022). Inc, Sunderland, 2014), pp. 165–200. 53. J. Barlow et al., The future of hyperdiverse tropical ecosystems. Nature 559, 517–526 58. D. Massemin, D. Lamy, J.- P. Pointier, O. Gargominy, Coquillages et escargots de Guyane (Publications (2018). Scientifiques du Muséum, MNHN, Paris, 2009). 54. B. Neumann, A. T. Vafeidis, J. Zimmermann, R. J. Nicholls, Future coastal population growth and 59. D. Swofford, PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods), version exposure to sea-l evel rise and coastal flooding- a global assessment. PLoS One 10, e0118571 4.0a.169 (Sinauer Associates, Sunderland, 2002). https://phylosolutions.com/paup-test/. Accessed (2015). 18 December 2023. PNAS  2024  Vol. 121  No. 14  e2311597121 https://doi.org/10.1073/pnas.2311597121   11 of 11 Downloaded from https://www.pnas.org by Pierre-Olivier ANTOINE on March 25, 2024 from IP address 2.4.10.210.