4.b.1.a. The Somali Basin

The tectono-sedimentary evolution of the Somali Basin includes a pre-Jurassic intracontinental aborted rift stage, during which deposition consisted predominantly of terrestrial fluvial, lacustrine and swamp sediments. An incursion of palaeo-Tethys seawater along the axial part of the depression may have occurred during pre-Jurassic times (Salman and Abdula, Reference Salman and Abdula1995).

By the early–middle Jurassic, Gondwana breakup had entered its principal phase of intercontinental rifting, broadly coeval with emplacement of the main Karoo Large Igneous Province between ∼184 and 181 Ma (Duncan et al. Reference Duncan, Hooper, Rehacek, Marsh and Duncan1997; Jourdan et al. Reference Jourdan, Féraud, Bertrand, Watkeys and Renne2007). Marine transgression from the palaeo-Tethys produced a narrow embayment system in which barrier reefs isolated saline lagoons within the Ogaden, Mandawa and possibly Ruvuma basins (Salman and Abdula, Reference Salman and Abdula1995). The palaeogeographic development of this region has subsequently been reconstructed in detail through hydrocarbon exploration studies (Davidson et al. Reference Davidson, Arthur, Smith and Tubb2018; Geng et al. Reference Geng, Wen, Wang, Wang, He, Song, Ji and Li2023) (Figure 6). Continued extension ultimately progressed to organized oceanic spreading during the late Jurassic, transforming the Somali Basin into a fully marine basin system (Mueller and Jokat, Reference Mueller and Jokat2019; Thompson et al. Reference Thompson, Moulin, Aslanian, De Clarens and Guillocheau2019).

Figure 6.

Palaeogeographic development of the Somali Basin during the Jurassic. (a) Jurassic stratigraphy of offshore sub-basins in the Somali Basin; adapted and simplified from Davidson et al. (Reference Davidson, Arthur, Smith and Tubb2018). (b) Palaeogeographic reconstruction of the Somali Basin near the early–middle Jurassic boundary. Abbreviations are as in Figure 5. Adapted and simplified from Geng et al. (Reference Geng, Wen, Wang, Wang, He, Song, Ji and Li2023) under the Creative Commons Attribution 4.0 Licence.

4.b.1.b. The Davie Fracture Zone

The Davie Fracture Zone represents the first narrow gateway within the chain of basins that collectively formed the early trans-Gondwana Seaway (Figures 1 and 5). It constitutes a major transform fault system that accommodated the southward displacement of Madagascar relative to Africa during the breakup of Gondwana from ∼157 Ma onwards (e.g. Vormann and Jokat, Reference Vormann and Jokat2021; Roche and Ringenbach, Reference Roche and Ringenbach2022). This prominent structural feature extends for approximately 2300 km from the Kenya margin to the southwestern margin of Madagascar and continues southwards as an oceanic fracture zone towards the Southwest Indian Ridge near 45°S latitude (Vormann and Jokat, Reference Vormann and Jokat2021; Roche and Ringenbach, Reference Roche and Ringenbach2022).

During plate motion between ∼157 and 144 Ma, much of the Davie Fracture Zone developed within transitional-to-oceanic crust formed during the initial phase of continental breakup. Consequently, earlier pull-apart basins and the oldest oceanic crust formed along the East African margin were subsequently deformed and locally overprinted by later tectonic activity (Vormann and Jokat, Reference Vormann and Jokat2021).

Sedimentary successions on the African side of the zone are documented in the Rovuma Basin, a rifted margin basin initiated by extension along the Davie Fracture Zone and extending from southern Tanzania into northern Mozambique (Coster et al. Reference Coster, Lawrence and Fortes1989). On the Madagascar side, equivalent deposits occur within the Morondava Basin of western Madagascar (Giese et al. Reference Giese, Seward and Schreurs2012; Figures 5 and 7).

Figure 7.

General stratigraphic scheme of the Davie Fracture Zone and the Mozambique Basin based on results from oil exploration. (a) Southern Davie Fracture Zone and Morondava Basin; (b) coastal Mozambique; (c) Mozambique Channel and coastal Tanzania; and (d) coastal Morondava Basin. Compiled, redrawn and modified from Brownfield (2016, figs 5 and 7), with additional interpretation.

The Rovuma Basin constitutes part of the larger East African marginal basin system extending along the coastal plains and continental margins of Tanzania and Kenya. In this basin, the earliest marine sedimentation is represented by Bajocian–Bathonian shales of the Mandawa succession, when deposits of comparable age are also recognized in western Madagascar (Coster, Lawrence and Fortes, Reference Coster, Lawrence and Fortes1989; Luger et al. Reference Luger, Gröschke, Bussmann, Dina, Mette, Uhmann and Kallenbach1994).

In western Madagascar, the earliest recorded Jurassic marine transgression began during the early Toarcian, when marine waters invaded from the north into the Diego, Majunga and northern Morondava basins. In the southern Morondava Basin, marine conditions developed later, during the Toarcian–Aalenian interval (Luger et al. Reference Luger, Gröschke, Bussmann, Dina, Mette, Uhmann and Kallenbach1994). Early–middle Jurassic (Aalenian) sandstones in the Morondava Basin are of continental facies (Coffin and Rabinowitz, Reference Coffin and Rabinowitz1988; Geiger et al., Reference Geiger, Clark and Mette2004). During the late middle Jurassic (Bajocian–Bathonian), marine conditions dominated the northern part of the Morondava Basin, where a carbonate platform developed in shallow-marine to tidal environments (Giese et al. Reference Giese, Seward and Schreurs2012, and references therein). Farther south, a gradual transition to mixed marine–continental facies is observed, accompanied by intervals of non-deposition (Coffin and Rabinowitz, Reference Coffin and Rabinowitz1988). In the south-central sector, marginal marine to fluvial sedimentation reappeared as early as the Bajocian and, driven by a regressive trend, expanded both northwards and southwards during the Bathonian. A renewed and widespread transgression occurred in the early Callovian, and by the middle Callovian open-marine conditions were established across the entire Morondava Basin, persisting until the early Oxfordian (Luger et al. Reference Luger, Gröschke, Bussmann, Dina, Mette, Uhmann and Kallenbach1994).

During the late Jurassic, sedimentation in the northern Morondava Basin was predominantly marine. Carbonate successions reaching ∼200 m in thickness pass southwards into mixed marine and continental facies within the central and southern sectors of the basin (Coffin and Rabinowitz, Reference Coffin and Rabinowitz1988), whereas prolonged intervals of non-deposition are recognized farther south. Although marine conditions are inferred for the late Oxfordian, fossil evidence confirming this age remains lacking (Luger et al. Reference Luger, Gröschke, Bussmann, Dina, Mette, Uhmann and Kallenbach1994). Offshore boreholes have also recovered rhyolitic units interlayered with marine sediments (Coffin and Rabinowitz, Reference Coffin and Rabinowitz1988). Shallow-marine environments persisted through the Kimmeridgian and Tithonian, although local continental deposition occurred during the early Kimmeridgian (Luger et al. Reference Luger, Gröschke, Bussmann, Dina, Mette, Uhmann and Kallenbach1994). Upper Tithonian strata are restricted to isolated exposures in the Analavelona and Mikoboka Mountains of the southern Morondava Basin. Marine sedimentation continued into the Berriasian in the northern basin (Luger et al. Reference Luger, Gröschke, Bussmann, Dina, Mette, Uhmann and Kallenbach1994), followed by renewed marine transgression during Valanginian–early Hauterivian times. Deposits of this interval are only locally preserved in the central and southern sectors, although lower Cretaceous strata up to the Albian are absent across much of the basin (Giese et al. Reference Giese, Seward and Schreurs2012, and references therein).

In contrast to the variable depositional record of the Morondava Basin, published results from onshore and offshore hydrocarbon exploration may provide a clearer and more continuous record for the Davie Fracture Zone area. These data indicate that relatively uninterrupted marine deposition has persisted within the zone since the middle Jurassic (Brownfield, Reference Brownfield2016) (see Figure 7a).

4.b.1.c. The Mozambique Basin

The Mozambique Basin extends across the central and southern parts of the Mozambican coastal plain and continues offshore beneath the continental shelf and slope (Salman and Abdula, Reference Salman and Abdula1995). Along its landward margin, the basin is flanked by Precambrian crystalline basement, whereas its Mesozoic–Cenozoic sedimentary fill locally overlies Karoo strata and volcanic rocks (Salman and Abdula, Reference Salman and Abdula1995; Mussa et al. Reference Mussa, Flores, Ribeiro, Mizusaki, Chamussa, Filho and Kalkreuth2018). The post-Karoo succession comprises Upper Jurassic, Cretaceous and Cenozoic formations that rest unconformably on deeply eroded Karoo strata (Catuneanu et al. Reference Catuneanu, Wopfner, Eriksson, Cairncross, Rubidge, Smith and Hancox2005). Post-Karoo deposits in the marginal sedimentary basins of Mozambique encompass continental, transitional, deltaic and fully marine facies, reflecting progressive basin evolution and marine incursions (Salman and Abdula, Reference Salman and Abdula1995; GTK Consortium, 2006). In the western part of the Mozambique Basin, upper Jurassic strata occur predominantly as continental red beds, mainly preserved within buried graben structures in the southern sector. In the northwestern part of the basin, these deposits correlate with continental sandstones of the Lupata Formation (Salman and Abdula, Reference Salman and Abdula1995; GTK Consortium, 2006). According to Salman and Abdula (Reference Salman and Abdula1995), Tethyan transgression reached the Ruvuma Basin during Kimmeridgian–Tithonian times, and the sea likely penetrated into the northern part of the Mozambique Basin as well. It is generally accepted that farther offshore, beyond the marginal sedimentary basins along the Mozambican coast, fully marine conditions prevailed in the East Mozambique Basin (Figure 8).

Figure 8.

General stratigraphy of the early Cretaceous strata of the Mozambique Basin calibrated to the Geological Timescale 2020 (Gradstein et al. Reference Gradstein, Ogg, Schmitz and Ogg2020). The Neocomian Unconformity at the base of the Maputo Formation is aged at 136 Ma (late early Valanginian). Based on stratigraphic information presented by Mussa et al. (Reference Mussa, Flores, Ribeiro, Mizusaki, Chamussa, Filho and Kalkreuth2018, fig. 2; after ECL, 2000); independently redrawn and simplified in this study.

Figure 8. Long description

A stratigraphic diagram of the early Cretaceous strata in the Mozambique Basin, showing different geological formations and their ages. The diagram is divided into two main sections: West and East. The West section includes the Sena Formation (North), Red Beds (South), and Maputo Formation (South). The East section includes the Lower-Domo Shale and Lupata Formation (North). The diagram spans from the Late Jurassic to the Early Cretaceous, with ages ranging from 160 to 100 million years ago. Key geological periods such as the Albian, Aptian, Barremian, Hauterivian, Valanginian, and Berriasian are marked on the left side. The diagram also highlights significant geological events like the Neocomian Unconformity and Mid Cretaceous Unconformity. Different colors and patterns represent various depositional environments, including non-marine, paralic, shallow marine-paralic, and shallow marine-euxinic.

4.b.1.d. The Limpopo Corridor

The Limpopo Corridor constitutes a major strike-slip-dominated rifted corridor extending along the continental Mozambique Coastal Plain and into the North Natal Valley (Figures 1 and 3). It represents a large transform shear zone that developed during the breakup of Gondwana, accommodating the southward relative motion of Antarctica with respect to Africa (Evain et al. Reference Evain, Schnürle, Leprêtre, Verrier, Watremez, Thompson, De Clarens, Aslanian and Moulin2021; Roche et al. Reference Roche, Leroy, Guillocheau, Revillon, Ruffet, Watremez, D’acremont, Nonn, Vetel and Despinois2021). This transform margin is characterized by a complex continent–ocean transition zone connecting the Beira High with the North Natal Valley margin segments. It marks the landward limit of unequivocal oceanic crust and separates continent-derived volcano-sedimentary graben fills from the adjacent oceanic crustal domain (Evain et al. Reference Evain, Schnürle, Leprêtre, Verrier, Watremez, Thompson, De Clarens, Aslanian and Moulin2021; Roche et al. Reference Roche, Leroy, Guillocheau, Revillon, Ruffet, Watremez, D’acremont, Nonn, Vetel and Despinois2021).

Wide-angle and reflection seismic studies indicate that the Mozambique Coastal Plain and the North Natal Valley are underlain by a thick, heterogeneous and strongly intruded crustal domain containing substantial continent-derived components (Moulin et al. Reference Moulin, Aslanian, Evain, Leprêtre, Schnurle, Verrier, Thompson, De Clarens, Leroy and Dias2020; Leprêtre et al. Reference Leprêtre, Schnürle, Evain, Verrier, Moorcroft, De Clarens, Corela, Afilhado, Loureiro, Leroy, D’acremont, Thompson, Aslanian and Moulin2021; Wang et al. Reference Wang, Singh, Wang, Ruan, Tang, Dyment, Leroy, Wu, Li and Dong2025), which grades eastwards into unequivocal oceanic crust (Evain et al. Reference Evain, Schnürle, Leprêtre, Verrier, Watremez, Thompson, De Clarens, Aslanian and Moulin2021; Roche et al. Reference Roche, Leroy, Guillocheau, Revillon, Ruffet, Watremez, D’acremont, Nonn, Vetel and Despinois2021; Watremez et al. Reference Watremez, Leroy, d’Acremont, Roche, Evain, Leprêtre, Verrier, Aslanian, Dias, Afilhado, Schnürle, Castilla, Despinois and Moulin2021; Schnürle et al. Reference Schnürle, Leprêtre, Evain, Verrier, De-Clarens, Thompson, Dias, Afilhado, Loureiro, Leroy, D’acremont, Aslanian and Moulin2023; Wang et al. Reference Wang, Singh, Wang, Ruan, Tang, Dyment, Leroy, Wu, Li and Dong2025). These observations constrain the range of possible plate restorations but do not, by themselves, demonstrate that the surface of this domain remained continuously emergent during breakup.

In the reconstruction of Thompson et al. (Reference Thompson, Moulin, Aslanian, De Clarens and Guillocheau2019), the Limpopo Corridor occupies one candidate zone of final proximity between the African and Antarctic plates. Whether this proximity produced a continuous terrestrial connection would additionally have depended on extension-related subsidence, sediment accumulation, volcanic construction and erosion, and eustatic sea-level change. The Limpopo Corridor is therefore treated here as a potential palaeogeographic constriction rather than as a demonstrated land bridge. The timing of its transition to marine conditions is assessed primarily from the sedimentary succession rather than from crustal affinity alone.

The Maputo Formation is documented in the southern and central sectors of the Mozambique Basin. In its northern sector, however, Neocomian strata that are considered stratigraphic equivalents of the Maputo Formation are deeply buried and have not been intersected by any exploration wells to date (Salman and Abdula, Reference Salman and Abdula1995). Their presence has therefore been inferred at the base of the sedimentary cover primarily through seismostratigraphic correlation (Salman and Abdula, Reference Salman and Abdula1995). Lithologically, the Maputo Formation comprises sandstones, conglomerates and sandy limestones, reflecting deposition under coastal to open-marine conditions (GTK Consortium, 2006). Reported facies include conglomeratic sandstones and clayey-to-glauconitic marlstones (GTK Consortium, 2006), consistent with a transgressive setting. The formation unconformably overlies either eroded basaltic units of the Stormberg Group or underlying continental red beds. Seismic data document its presence beneath both the Mozambique Coastal Plain and the Limpopo transform fault zone (Roche et al. Reference Roche, Leroy, Guillocheau, Revillon, Ruffet, Watremez, D’acremont, Nonn, Vetel and Despinois2021; Figure 5). In the North Natal Valley, the Maputo Formation records the initial early Cretaceous flooding of the continental shelf (Schnürle et al. Reference Schnürle, Leprêtre, Evain, Verrier, De-Clarens, Thompson, Dias, Afilhado, Loureiro, Leroy, D’acremont, Aslanian and Moulin2023). Accordingly, constraining the age of the earliest deposits of the Maputo Formation is fundamental for determining the timing of the first marine incursion along the Mozambique Fracture Zone. Two major issues complicate attempts to constrain the age of the Maputo Formation: (1) stratigraphic definition and unit differentiation and (2) chronostratigraphic uncertainty.

The first issue concerns the ambiguous differentiation between the Maputo Formation and the so-called ‘Red Bed Formation’, neither of which has been formally and consistently defined in the literature (GTK Consortium, 2006). In some studies, the red beds are treated as a lower unit within the Maputo Formation, whereas in others they are regarded as a distinct stratigraphic formation (e.g. José et al. Reference José, Chaúque, Uanzo, Cabral, Manhiça, Juvane, Morais and Jokat2024). Detailed seismic interpretation from the southern sector of the Mozambique Basin indicates that, in several wells, the sedimentary unit directly overlying the basal basalts (SU-2 in José et al. Reference José, Chaúque, Uanzo, Cabral, Manhiça, Juvane, Morais and Jokat2024; U1 in Li et al. Reference Li, Tang, Moulin, Aslanian, Evain, Schnurle, Leprêtre and Li2021) can be subdivided into a Lower Unit and an Upper Unit, separated by the so-called ‘Neocomian Unconformity’. The Upper Unit corresponds to glauconitic–quartzose sands attributed to the Maputo Formation, which in some cases is reported to rest directly on the basaltic basement (GTK Consortium, 2006), suggesting lateral variability in depositional architecture. The Neocomian Unconformity marks the boundary between the Maputo Formation and the underlying red beds (Figure 8) and has been interpreted as recording a phase of uplift and the initiation of oceanic crust formation offshore in the Mozambique Basin (Schnürle et al. Reference Schnürle, Leprêtre, Evain, Verrier, De-Clarens, Thompson, Dias, Afilhado, Loureiro, Leroy, D’acremont, Aslanian and Moulin2023). The Lower Unit, which lies unconformably above the basalts, corresponds to the Red Bed Formation. According to José et al. (Reference José, Chaúque, Uanzo, Cabral, Manhiça, Juvane, Morais and Jokat2024), this Lower Unit can itself be subdivided into (i) a lower interval dominated by mixed marine–non-marine shales and (ii) an upper interval composed mainly of coastal clastics. This subdivision implies that at least part of the red-bed succession accumulated in a transitional marine–terrestrial setting rather than under strictly continental conditions.

The second major issue concerns the poorly constrained age of both the Maputo Formation and the Red Bed Formation. The red beds are generally barren of pollen and diagnostic microfossils, and their age is broadly assigned to the late Jurassic–early Cretaceous (GTK Consortium, 2006). Many age attributions are based on secondary citations, however, and direct verification is often hindered by the limited accessibility of biostratigraphic data from exploration wells, much of which remain unpublished within industry archives. The Maputo Formation was initially assigned a Barremian–Aptian/Albian age (e.g. Nairn et al. Reference Nairn, Lerche and Iliffe1991). Later, a significantly older age – generally Neocomian (Berriasian–Hauterivian) – became widely accepted (e.g. Salman and Abdula, Reference Salman and Abdula1995; Mussa et al. Reference Mussa, Flores, Ribeiro, Mizusaki, Chamussa, Filho and Kalkreuth2018). Nevertheless, inconsistencies persist among published stratigraphic charts. For example, Schnürle et al. (Reference Schnürle, Leprêtre, Evain, Verrier, De-Clarens, Thompson, Dias, Afilhado, Loureiro, Leroy, D’acremont, Aslanian and Moulin2023, fig. 9) adopt an early Berriasian age (∼142 Ma) for the onset of Maputo deposition, citing Said et al. (Reference Said, Moder, Clark and Ghorbal2015), who present a similar early Berriasian age, referencing Salman and Abdula (Reference Salman and Abdula1995). Salman and Abdula (Reference Salman and Abdula1995), in turn, describe a generalized Neocomian age and depict a Berriasian–Valanginian boundary age (∼138 Ma) in their stratigraphic chart (their figure 14). In fact, a basal Neocomian (basal Berriasian) age is illustrated in the chart of Coster et al. (Reference Coster, Lawrence and Fortes1989, fig. 6). This chain of indirect citation and graphical reinterpretation has resulted in progressive chronological drift, obscuring the primary evidentiary basis for the age assignment. As a consequence, the timing of the earliest marine incursion recorded by the Maputo Formation remains less robust than commonly assumed.

Figure 9.

Schematic palaeoenvironmental locator diagram and key drilling stratigraphic sequences analysed by Setoyama and Kanungo (Reference Setoyama and Kanungo2020) in the proto-Weddell Sea region. The diagram is not intended as a geometrically scaled plate reconstruction. DSDP Sites 327, 330 and 511 are located on the Falkland Plateau; ODP Site 692 is located in the Weddell Sea; and DSDP Site 249 is located on the Northern Mozambique Ridge. ODP Site 693 is discussed in the text but is not shown as a separate stratigraphic column. Their interpretation of the onset of marine sedimentation at the Northern Mozambique Ridge (Site 249) during the late Berriasian differs from that of Dunay et al. (Reference Dunay, Braham, Cooper, Lester and Tremolada2018), who concluded that the lowermost sedimentary section is no older than Barremian. However, it is more consistent with earlier biostratigraphic studies suggesting older ages for the basal sediments at DSDP site 249 (Valanginian: Riegraf, Reference Riegraf1989; Neocomian: Shipboard Scientific Party, 1974). Stratigraphic columns and palaeoenvironmental data adapted and simplified from Setoyama and Kanungo (Reference Setoyama and Kanungo2020), with permission from Elsevier.

Figure 10.

The formation of the Palaeo-Pacific Weddell Sea (PPWS) gateway through the Rocas Verdes Basin (RVB). (a) Schematic pathway of the marine corridor at ∼150 Ma. Abbreviations: NSR = North Scotia Ridge; SG = South Georgia; SSR = South Scotia Ridge; TdF = Tierra del Fuego. Adapted and simplified from Oldenhage, Schellart and Beniest (Reference Oldenhage, Schellart and Beniest2024, fig. 6; CC BY 4.0), after Eagles (Reference Eagles2010). (b) Schematic configuration of the deep sector of the Rocas Verdes Basin at ∼120 Ma. Adapted and simplified from Diraison et al. (Reference Diraison, Cobbold, Gapais, Rossello and Le Corre2000: fig. 3a), with permission from Elsevier.

Figure 10. Long description

Panel A: A schematic map showing the marine corridor at approximately 150 million years ago. The map includes labels for South America, Africa, Antarctica, the Palaeo-Pacific Ocean, and the Weddell Sea. Key features include the North Scotia Ridge (NSR), South Georgia (SG), South Scotia Ridge (SSR), Tierra del Fuego (TdF), and the Rocas Verdes Basin (RVB). Arrows indicate the direction of water flow through the PPWS gateway. Panel B: A schematic map depicting the deep sector of the Rocas Verdes Basin at approximately 120 million years ago. The map includes labels for South America, Africa, the Weddell Sea, the Antarctic Peninsula, Cordillera Darwin, South Georgia, and South Orkney. The map also shows the locations of the Rocas Verdes marginal basin and volcanic arcs, with rifts indicated by specific symbols.

Figure 11.

Opening of the Rocas Verdes Basin (RVB) during the Late Jurassic–earliest Cretaceous. Kinematic reconstruction snapshots at 154 and 140 Ma in an East Antarctica-fixed reference frame. AF = Africa; ANT = Antarctica; AP = Antarctic Peninsula; FPB = Falkland Plateau Basin; PAT = Patagonia; P-P = Palaeo-Pacific; SA = South America; WS = Weddell Sea. Adapted, simplified and reinterpreted from van de Lagemaat et al. (Reference Van De Lagemaat, Swart, Vaes, Kosters, Boschman, Burton-Johnson, Bijl, Spakman and Van Hinsbergen2021, fig. 6b, c) under the Creative Commons Attribution 4.0 Licence.

Figure 12.

Palaeobiogeographic reconstruction for the Oxfordian–Kimmeridgian interval illustrating the inferred distribution of second-order bivalve biochoremas. Adapted and simplified from Damborenea, Echevarría and Ros-Franch (Reference Damborenea, Echevarría and Ros-Franch2013, fig. 5.10), reproduced with permission from Springer Nature. This material is not covered by the CC BY licence of this article.

Figure 12. Long description

A paleobiogeographic reconstruction map illustrating the distribution of second-order bivalve biochoremas during the Oxfordian-Kimmeridgian interval. The map includes several labeled regions: Africa, S. America, E. African, Tethyan, SE Tethyan, India, Antarctica, Australia, S. Andean, and Maorian. These regions are color-coded to indicate different biochoremas. The map also shows latitude lines at intervals of 15 degrees, ranging from -75 degrees to 40 degrees. The labeled regions are distributed across the map, with specific areas marked for the Tethyan, SE Tethyan, E. African, S. Andean, and Maorian biochoremas. The map provides a visual representation of the inferred distribution of these biochoremas during the specified geological interval.

Figure 13.

Temporal evolution of the second-order palaeobiogeographic units recognized within Southern Hemisphere marine benthonic bivalve faunas. Bar width is proportional to the percentage of endemic genera through time and reflects changes in the relative rank of each unit. The thick dashed line marks the approximate boundary between the Tethyan Realm and the southern first-order palaeobiogeographic realm, represented by the South Pacific Realm during the Jurassic and the South Temperate Realm during the Cretaceous. Adapted and simplified from Damborenea, Echevarría and Ros-Franch (Reference Damborenea, Echevarría and Ros-Franch2013, fig. 5.11), reproduced with permission from Springer Nature. This material is not covered by the CC BY licence of this article.

Figure 14.

Palaeobiogeographic distribution of the belemnites Belemnopsis and Hibolithes during the upper Tithonian–lower Berriasian. Data localities: 1, Neuquén Basin; 2, Lago San Martín, Austral Basin; 3, Lago Argentino, Austral Basin; 4, Torres del Paine, Chile; 5, Tierra del Fuego Archipelago; 6, Alexander Island, Antarctic Peninsula; 7, Malvinas/Falkland Plateau; 8, Orville Coast and Ellsworth Land, Western Antarctica; 9, Eastern Weddell Sea; 10, Madagascar; 11, Kutchh; 12, Northwestern Pakistan; 13, Spiti Shales; 14, Australia; 15, New Guinea; 16, Moluccas and Sula Islands; 17, Misool, Indonesia; 18, New Zealand. Adapted and simplified from Vennari et al. (Reference Vennari, Aguirre-Urreta, Marin, Pellenard, Martinez and Tunik2023, fig. 6), with permission from Elsevier.

Figure 14. Long description

A map titled Tethyan Realm. The map shows suggested dispersal pathways, biogeographical province boundaries, and locations of Belemnopsis and Hibolithes. The map is divided into three main provinces: the South Pacific Province, the Ethiopian Subprovince, and the Tethyan Subprovince, which is further divided into the Ethiopian Subprovince and the Indo-Tethyan Subprovince. The suggested dispersal pathway is indicated by a blue line, and biogeographical province boundaries are indicated by pink dashed lines. The locations of Belemnopsis and Hibolithes are marked by blue dots and numbered from 1 to 18. The map includes the following data localities: 1, Neuquén Basin; 2, Lago San Martín, Austral Basin; 3, Lago Argentino, Austral Basin; 4, Torres del Paine, Chile; 5, Tierra del Fuego Archipelago; 6, Alexander Island, Antarctic Peninsula; 7, Malvinas/Falkland Plateau; 8, Orville Coast and Ellsworth Land, Western Antarctica; 9, Eastern Weddell Sea; 10, Madagascar; 11, Kutchh; 12, Northwestern Pakistan; 13, Spiti Shales; 14, Australia; 15, New Guinea; 16, Moluccas and Sula Islands; 17, Misool, Indonesia; 18, New Zealand.

Stratigraphic charts compiled by Mussa et al. (Reference Mussa, Flores, Ribeiro, Mizusaki, Chamussa, Filho and Kalkreuth2018, fig. 2) and the GTK Consortium (2006, fig. 9.2) indicate an age of approximately 136 Ma for the Neocomian Unconformity and the onset of Maputo Formation deposition (Figure 8). Roche et al. (Reference Roche, Leroy, Guillocheau, Revillon, Ruffet, Watremez, D’acremont, Nonn, Vetel and Despinois2021), using biostratigraphic data from the Funhalouro Well to calibrate seismic profiles across the Limpopo transform fault zone, similarly assigned an age of 136 Ma to the top of the Red Bed Formation succession. Taken together, these observations suggest that the initial deposition of the marine Maputo Formation occurred at ∼136 Ma, corresponding to the late early Valanginian. However, as illustrated in the depositional scheme of Figure 8, this age primarily applies to nearshore regions of the southern and central Mozambique Coastal Plain. Eastwards, where the monocline relief deepens and accommodation space increases, a more open-marine environment is inferred for the Mozambique Basin beyond the Limpopo Corridor. This spatial gradient implies that the onset of fully marine conditions likely propagated progressively eastwards, consistent with the development of the proto trans-Gondwana Seaway.

Indeed, seismic data along the Limpopo margin and the Mozambique Coastal Plain indicate that the underlying continental crust thins eastwards towards the Mozambique Basin, with a concomitant increase in sediment thickness and development of deeper basins towards the Indian Ocean margin (Evain et al. Reference Evain, Schnürle, Leprêtre, Verrier, Watremez, Thompson, De Clarens, Aslanian and Moulin2021). This structural gradient is consistent with the progressive formation of syn-rift sequences (Roche et al. Reference Roche, Leroy, Guillocheau, Revillon, Ruffet, Watremez, D’acremont, Nonn, Vetel and Despinois2021; Schnürle et al. Reference Schnürle, Leprêtre, Evain, Verrier, De-Clarens, Thompson, Dias, Afilhado, Loureiro, Leroy, D’acremont, Aslanian and Moulin2023), including the Maputo Formation, and supports the hypothesis of an early proto-trans-Gondwana Seaway. The eastward thinning of the continental crust would have facilitated the incursion of marine waters, whereas the thickening sedimentary sequences in the deeper parts of the basin record the progressive accommodation of these transgressive deposits (Evain et al. Reference Evain, Schnürle, Leprêtre, Verrier, Watremez, Thompson, De Clarens, Aslanian and Moulin2021). Consequently, these seismic observations provide independent geophysical evidence for the establishment of marine connectivity along the Limpopo transform margin, reinforcing the interpretation of the Maputo Formation as a marker of the initial flooding of the Mozambique Fracture Zone. Accordingly, in recent stratigraphic schemes, mainly based on unpublished technical reports from oil exploration, the initial exposure of a marine environment east of the Limpopo Corridor is assigned an age of ∼145 Ma (Brownfield, Reference Brownfield2016; Mussa et al. Reference Mussa, Flores, Ribeiro, Mizusaki, Chamussa, Filho and Kalkreuth2018) (late Tithonian; Figures 7a and 8). This older age likely reflects initial transgressive conditions in the more distal, offshore portions of the Mozambique Basin, consistent with a progressive eastward propagation of marine environments.

4.b.1.e. The proto-Weddell Sea area

The Weddell Sea, situated east of the Antarctic Peninsula, developed during the Jurassic–Cretaceous in association with opening of the South Atlantic and progressive separation of the South American, African and Antarctic plates (e.g. König and Jokat, Reference König and Jokat2006). Published estimates for the onset of oceanic accretion within the wider proto-Weddell Sea–Falkland Plateau system range from approximately 165 to 145 Ma. In the kinematic reconstruction used by Macgregor and Reeves (Reference Macgregor and Reeves2025), oceanic crust begins to form between the Malvinas Plateau and the Maurice Ewing Bank at approximately 165 Ma; the sedimentation record of the Outeniqua basins of South Africa is cited in support of this interpretation. König and Jokat (Reference König and Jokat2006) proposed that the precursor of the Weddell Sea may have originated along fracture zones between the Falkland Plateau and Antarctica. More recently, Eagles and Eisermann (Reference Eagles and Eisermann2020) demonstrated the presence of oceanic crust beneath the Falkland Plateau and identified magnetic anomalies M25 (∼153 Ma) and M22A (∼149 Ma), supporting an opening as early as the Kimmeridgian (late Jurassic).

The complex geological evolution of the region has obscured evidence for the sedimentary development of the proto-Weddell Sea. Erosion, non-deposition and the accumulation of contourite drifts generated by bottom currents – particularly active from the Cenozoic to the present – are largely linked to the Gondwana breakup, including the Mesozoic opening of the South Atlantic Ocean and the Cenozoic opening of the Drake Passage (Del Ben and Mallardi, Reference Del Ben and Mallardi2004). During the Mesozoic, the opening of the proto-South Atlantic and Indian oceans triggered widespread erosion and non-deposition, resulting in the absence of significant portions of the sedimentary cover (Del Ben and Mallardi, Reference Del Ben and Mallardi2004). Nevertheless, correlation between sediments recovered from DSDP sites 330 and 511 and unfaulted near-basement regional reflectors within the basin interior places the possible initiation of seafloor spreading related to plate divergence within the Callovian–Tithonian interval (166–145 Ma; Eagles and Eisermann, Reference Eagles and Eisermann2020, and references therein). These constraints document diachronous local oceanic accretion within individual sectors of the evolving system; they do not, by themselves, establish when all constrictions along the trans-Gondwana Seaway became simultaneously open. Within this temporal framework, the identified magnetic anomalies can be correlated with the M25–M22A reversal isochron sequence (153–149 Ma; Gradstein et al. Reference Gradstein, Ogg, Schmitz and Ogg2020) associated with early seafloor spreading.

Setoyama and Kanungo (Reference Setoyama and Kanungo2020) re-analysed samples from 17 DSDP and ODP sites spanning from the equatorial region to the southern South Atlantic and established a high-resolution Mesozoic chronostratigraphic and palaeoenvironmental framework based on new micropalaeontological data. Their dataset included sites 327, 330 and 511 on the Falkland Plateau, sites 692 and 693 in the Weddell Sea and site 249 on the Mozambique Ridge (Figure 9). Their results indicate deposition of widespread marine organic matter-rich sediments of middle Jurassic–lower Cretaceous age (Bajocian–Berriasian, ∼170–140 Ma) on the Falkland Plateau, and early Cretaceous (late Berriasian–Valanginian, 140–135 Ma) deposits in the Weddell Sea (site 692) and on the Mozambique Ridge (site 249). Integration of these deposits with those of the Kimmeridgian–Berriasian Nordenskjöld Formation on the eastern Antarctic Peninsula margin (Whitham, Reference Whitham1993) suggests that the proto-Weddell Sea likely functioned as a marine corridor connecting the Mozambique Basin with the western margin of the Palaeo-Pacific Ocean since at least the late Jurassic.

4.b.1.f. The Palaeo-Pacific Weddell Sea gateway

Oldenhage et al. (Reference Oldenhage, Schellart and Beniest2024) introduced the term ‘Palaeo-Pacific Weddell Sea gateway’ to describe a continuous extensional system with an estimated width of 100–300 km that initially connected the Palaeo-Pacific Ocean with the Weddell Sea via the Rocas Verdes Basin (RVB) (Figure 10a). Similar gateway configurations were proposed by several earlier studies (Oldenhage et al. Reference Oldenhage, Schellart and Beniest2024, and references therein). The location and timing of the connection between the Palaeo-Pacific Ocean and the Weddell Sea were dynamic; it likely originated within the RVB and was subsequently abandoned, with the connection shifting to a proto-Drake Passage located just south of mainland Patagonia, Argentina (Dalziel et al. Reference Dalziel, Lawver, Norton and Gahagan2013; Van De Lagemaat et al. Reference Van De Lagemaat, Swart, Vaes, Kosters, Boschman, Burton-Johnson, Bijl, Spakman and Van Hinsbergen2021).

The RVB is an SE–NW-trending back-arc basin known from outcrops in the Patagonian Andes that experienced progressive northward propagation and opening during the late Jurassic and early Cretaceous (Diraison et al. Reference Diraison, Cobbold, Gapais, Rossello and Le Corre2000; Calderón et al. Reference Calderón, Fildani, Hervé, Fanning, Weislogel and Cordani2007; Cao et al. Reference Cao, Guillot, Torres Carbonell, Bordese, Lovecchio and Dimieri2025). During the early–middle Jurassic, the basin developed as a narrow, elongated volcano-tectonic depression along the southwestern Gondwanan margin, without evidence for oceanic crust formation. Subsequent late Jurassic extension transformed this system into a volcano-tectonic rift accompanied by widespread silicic magmatism along the southwestern Gondwana margin (Chon Aike magmatic province). Volcanism and rifting within the RVB occurred at least between 152 and 142 Ma, when the basin became underlain by oceanic crust generated along the Patagonian subduction margin (Calderón et al. Reference Calderón, Fildani, Hervé, Fanning, Weislogel and Cordani2007; Figure 11). Marine sedimentation began in the basin during the middle Jurassic (Lemaire Complex; Cao et al. Reference Cao, Guillot, Torres Carbonell, Bordese, Lovecchio and Dimieri2025) and continued until final closure at ∼110–100 Ma (Van De Lagemaat et al. Reference Van De Lagemaat, Swart, Vaes, Kosters, Boschman, Burton-Johnson, Bijl, Spakman and Van Hinsbergen2021).

These characteristics suggest that the RVB constituted a plausible pathway for water-mass exchange between the Palaeo-Pacific and the proto-Southern Ocean. However, its wedge-shaped geometry, opening from southeast to northwest, has led several palaeogeographic reconstructions to depict it as a semi-enclosed basin rather than a fully open throughgoing gateway (e.g. Diraison et al. Reference Diraison, Cobbold, Gapais, Rossello and Le Corre2000; Muller et al. Reference Muller, Calderón, Fosdick, Ghiglione, Cury, Massonne, Fanning, Warren, Ramírez de Arellano and Sternai2021). Consistent with this uncertainty, Oldenhage et al. (Reference Oldenhage, Schellart and Beniest2024) marked the western termination of the Palaeo-Pacific Weddell Sea gateway with a question mark (‘?RVB?’) in their reconstruction (Figure 10a). If the RVB did not function as a fully open marine gateway, an alternative connection to the Palaeo-Pacific may have existed through an opening in front of the South Orkney Microcontinent, located at the tip of the Antarctic Peninsula. In several palaeogeographic reconstructions (e.g. Poblete et al. Reference Poblete, Roperch, Arriagada, Ruffet, Ramírez de Arellano, Hervé and Poujol2016; Van De Lagemaat et al. Reference Van De Lagemaat, Swart, Vaes, Kosters, Boschman, Burton-Johnson, Bijl, Spakman and Van Hinsbergen2021), this microcontinental block appears to have acted as a structural ‘valve’ regulating the opening and restriction of the Palaeo-Pacific Weddell Sea gateway throughout its evolution.

4.b.1.g. Evidence from marine faunal exchange

Strong evidence from marine faunal exchange indicates that the initial establishment of a trans-Gondwana Seaway and the first major terrestrial separation between East and West Gondwana occurred during the late Tithonian (∼145 Ma). The existence of such a marine corridor was proposed more than seven decades ago by Termier and Termier (Reference Termier and Termier1952; as cited in Enay, Reference Enay and Hallam1973), who suggested a seaway linking Madagascar and South America on the basis of similarities among late Tithonian ammonites. Prior to the late Tithonian, marine faunas along the southern peri-Gondwanan margin were characterized by pronounced provincialism. A migration barrier between Madagascar and the Andean margin of southern Gondwana was inferred from the low diversity and strong Tethyan affinities of ammonoids and bivalves (Riccardi, Reference Riccardi1991). This barrier appears to have persisted until the Kimmeridgian, consistent with strong generic similarities among faunas from Madagascar, Australasia and the Tethyan Realm (Geiger, Reference Geiger2004, and references therein).

By the end of the middle Jurassic, bivalve endemism dominated both the Somali Basin and the southern Palaeo-Pacific margin, reflecting strict provincialism. Four biochoremas can be recognized in the Southern Hemisphere: two South Pacific units (Austral and South Andean) and two Tethyan units (East African and southeastern Tethyan; Figure 12) (Damborenea et al. Reference Damborenea, Echevarría and Ros-Franch2013). The East African province constituted a subordinate unit of the Tethyan Realm, whereas the Austral Realm belonged to the South Temperate Realm. By the latest Jurassic (Tithonian) and earliest Cretaceous, this pattern changed markedly (Figure 13). The East African unit shifted from the Tethyan Realm to the South Temperate Realm. Damborenea et al. (Reference Damborenea, Echevarría and Ros-Franch2013) suggested that this transition occurred ‘possibly as a direct consequence of the opening of the trans-Gondwana seaway, which communicated this part of Africa with the South Pacific’.

Further evidence suggests the presence of a shallow epicontinental seaway, intermittently connecting East Africa with southern Patagonia during the latest Tithonian–Berriasian interval. This interpretation is supported by the occurrence of the bivalve Megacucullaea in both regions, a genus unknown elsewhere (Riccardi, Reference Riccardi1991, and references therein), as well as by the distribution of the bivalves Eoseebachia and Seebachia (Saha et al. Reference Saha, Paul, Das and Bardhan2024). During the Tithonian, ammonoid assemblages from the Baños del Flaco and Lo Valdés formations of central Chile exhibit Indo-Pacific affinities, consistent with an open trans-Gondwana Seaway (Salazar Soto, Reference Salazar Soto2012). Further ammonoid similarities (e.g. Corongoceras lotenoense, Aspidoceras andinum) suggest direct faunal exchange between Madagascar and southern South America during the late Tithonian (Riccardi, Reference Riccardi1991). The occurrence of the ammonite Chigaroceras in both the western Tethys and the Andean region of Argentina has likewise been interpreted as evidence for seaway connectivity at that time (Leanza, Reference Leanza1996). Late Tithonian migration of large-bodied Virgatosphinctes spp. into the Andean region from the Kutch Basin (India), together with the presence of the Himalayan genera Blanfordiceras and Spiticeras in Baluchistan, Madagascar and the Andean Province, further supports interregional marine connectivity (Bardhan et al. Reference Bardhan, Shome, Roy, Landman, Davis and Mapes2007). Belemnite faunas provide additional support for late Tithonian marine connections. Mutterlose (Reference Mutterlose1986) reported that during the Kimmeridgian and early Tithonian, belemnite assemblages from the Latady Formation (southern Antarctic Peninsula) show affinities with those of New Zealand, whereas late Tithonian faunas display stronger Mediterranean affinities. More recently, Vennari et al. (Reference Vennari, Aguirre-Urreta, Marin, Pellenard, Martinez and Tunik2023) demonstrated that Neuquén Basin belemnite assemblages in Argentina show affinities with Madagascan–East African associations (Ethiopian Subprovince, Tethyan Province), reinforcing evidence for an established trans-Gondwana Seaway from the late Tithonian onwards (Figure 14).

Vertebrate fossil distributions provide complementary support. The actinopterygian fish Ameghinichthys antarcticus (Dapediiformes) from the uppermost lower to lower upper Tithonian Ameghino (Nordenskjöld) Formation of the Antarctic Peninsula suggests a seaway opening, as dapediiforms were previously known only from the Northern Hemisphere (Gouiric-Cavalli et al. Reference Gouiric-Cavalli, Iglesias, Cariglino and Reguero2024). Similarly, affinities among late Jurassic (Tithonian) plesiosaurs from the Neuquén Basin, India, Antarctica and the eastern Tethys suggest dispersal via the trans-Gondwana Seaway (O’Gorman et al. Reference O’Gorman, Gouiric-Cavalli, Scasso, Reguero, Moly and Acosta-Burlaille2018, and references therein). The discovery of ichthyosaur remains in the Longing Member of the Ameghino Formation (early–late Tithonian, Antarctic Peninsula) may further support this interpretation, given that middle–late Jurassic ichthyosaurs are predominantly known from the Northern Hemisphere (Campos et al. Reference Campos, Fernández, Herrera, Talevi, Concheyro, Gouiric-Cavalli, O’Gorman, Santillana, Acosta-Burlaille, Moly and Reguero2021).

4.b.2.a. Sri Lanka basins

Sri Lanka’s continental margin is bordered by three principal sedimentary basins: the Mannar Basin to the west, the Lanka Basin to the south and east and the Cauvery Basin to the north, the latter of which is shared with the Indian margin (e.g. Kularathna et al. Reference Kularathna, Pitawala, Senaratne and Ratnayake2020; Figure 15). From a palaeogeographic perspective, the Mannar Basin represents the former contact zone between western Sri Lanka and southeastern India during pre-drift times, whereas the Lanka Basin records the former juxtaposition between Sri Lanka and Antarctica.

Figure 15.

Sedimentary basins of Sri Lanka and southeastern India. (a) Principal sedimentary basins surrounding Sri Lanka; the box labelled (b) indicates the area enlarged in panel (b). Adapted from Kularathna et al. (Reference Kularathna, Pitawala, Senaratne and Ratnayake2020; CC BY 4.0). (b) Principal sub-basins of the Cauvery Basin in southeastern India. Simplified and redrawn from Rao et al. (Reference Rao, Chidambaram, Bharktya and Janardhanan2010, fig. 1).

The Mannar Basin occupies northwestern Sri Lanka, southeast of India (Figure 15a). Although drilling has penetrated only to upper Cretaceous levels, seismic data indicate that marine sedimentation began in the early Jurassic during the initial syn-rift phase of basin development (Baillie et al. Reference Baillie, Barber, Deighton, Gilleran, Jinadasa and Shaw2004; Kularathna et al. Reference Kularathna, Pitawala, Senaratne, Senevirathne and Weerasinghe2015). Depositional environments include continental, restricted marginal-marine and fully marine systems. Part of the succession is equivalent to the Upper Gondwana Group of India (Sivaganga Beds; Baillie et al. Reference Baillie, Barber, Deighton, Gilleran, Jinadasa and Shaw2004; Kularathna et al. Reference Kularathna, Pitawala, Senaratne, Senevirathne and Weerasinghe2015). Seismic surveys in the southern and eastern Lanka Basin reveal sedimentary successions up to ∼8 km thick; however, their chronostratigraphic and lithostratigraphic characteristics remain unknown because no deep drilling has yet been conducted (Silva et al. Reference Silva, Gamage and Weerasinghe2018; Davis and Ratnayake, Reference Davis and Ratnayake2024). In the absence of these critical constraints, palaeogeographic reconstructions linking Sri Lanka and Antarctica (e.g. Kularathna et al. Reference Kularathna, Pitawala, Senaratne and Ratnayake2020) remain speculative. The Cauvery Basin comprises a complex system of sub-basins along the southeastern Indian margin that diachronously formed the boundary and contact zone between southeastern India and Sri Lanka (Lal et al. Reference Lal, Siawal and Kaul2009; Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017) (Figure 15b). Its palaeoenvironmental evolution is considered in the following subsection together with that of other basins along the eastern coast of India.

4.b.2.b. Pericratonic basins of the eastern Indian continental margin

The breakup of East Gondwana led to the development of a series of NNE–SSW-trending rift systems along the eastern margin of the Indian Plate (e.g. Lal et al. Reference Lal, Siawal and Kaul2009; Sinha, Reference Sinha2014). These pericratonic basins accumulated sedimentary successions of varying composition during successive rifting phases.

During the early syn-rift phase (middle–late Jurassic), marine incursions are generally interpreted as restricted, shallow-water transgressions within enclosed basins that were unsuitable for foraminiferal colonization (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017; Raju and Reddy, Reference Raju and Reddy2016). Palaeoenvironmental interpretations of this phase rely heavily on dinoflagellate cyst data, leading to palaeogeographic reconstructions that remain a subject of debate. One such interpretation, inferred from Permian marine palynofossils recovered from the Assam Basin, proposes a marine corridor west of the Mahanadi Basin connecting Australia and India with the Tethys as early as the Permian (e.g. Bastia et al. Reference Bastia, Reeves, Pundarika Rao, D’Silva and Radhakrishna2010). Additional isolated occurrences of Permian marine palynofossils in the Krishna–Godavari, Cauvery, and Palar basins (Figure 2) have been interpreted as evidence for a narrow palaeo-shelf along the eastern Indian margin prior to the breakup and drift of eastern Gondwana (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017, and references therein). Regional tectono-palaeogeographic reconstructions indicate that the early–middle Jurassic rift system of the Somali Basin extended southwards and linked with the rift zone between eastern India and Antarctica (Geng et al. Reference Geng, Wen, Wang, Wang, He, Song, Ji and Li2023) (Figure 6). Considering this configuration, together with evidence that marine waters reached the Coastal Morondava Basin by the early Jurassic (Figure 7d), sea-level highstands may have permitted episodic marine overtopping and limited incursions into the pericontinental basins of eastern India (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017). The early syn-rift fill is dominated by spore–pollen assemblages with sparse dinoflagellates, suggesting predominantly continental deposition with occasional tidal influence.

Dinoflagellate cyst data indicate that marine incursions during highstand intervals progressively penetrated eastwards into the rift basins, reaching the West Bengal Basin by the latest Jurassic (Tithonian; Prasad and Pundir, Reference Prasad and Pundir2020), the Krishna–Godavari Basin by the Neocomian (Prasad and Pundir, Reference Prasad and Pundir1999), the Mahanadi Basin during the Berriasian–Aptian (Shukla et al. Reference Shukla, Prasad, Ali, Basavaraju, Saxena, Pundir, Singh, Whiso, Pandey and Shanmukhappa2014), the Palar Basin from the Bajocian to the Tithonian (Basavaraju et al. Reference Basavaraju, Jaiprakash, Chidambaram and Ayyadurai2016) and the Ariyalur–Pondicherry and eastern Cauvery sub-basins since the early Oxfordian (Narsimha, Reference Narsimha2006; see Figures 2 and 5). Basavaraju et al. (Reference Basavaraju, Jaiprakash, Chidambaram and Ayyadurai2016) further correlated the Kimmeridgian–Albian marine palynoassemblages of the Palar Basin with those of the Nagapattinam sub-basin of the eastern Cauvery Basin (Figure 15b). The temporal progression of these incursions suggests gradual flooding of the basins from west to east and may reflect the direction of rift propagation. Raju et al. (Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017) noted that, from the Bajocian (middle Jurassic) to the Valanginian (early Cretaceous), deposition along the eastern Indian continental margin was dominated by continental facies over an interval of approximately 38 million years. These deposits are present in the adjoining Cauvery and Krishna–Godavari basins but are poorly developed in the Mahanadi Basin, largely due to Rajmahal volcanism (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017). Thus, early syn-rift marine incursions likely resulted from episodic tidal inflow from the Mozambique proto-ocean to the west. Given that the western Cauvery sub-basins (Figure 15b) remained continental (Terani Formation) without marine influence during this interval (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017), marine waters may instead have entered the rifted palaeo-shelf via the Lanka Basin south of Sri Lanka (e.g. see Jokat et al. Reference Jokat, Altenbernd, Eagles and Geissler2021: Figure 5a for an analogous basin configuration).

Late syn-rift sedimentation began during the Valanginian and is characterized by more frequent tidal incursions associated with continued crustal breakup and thermal subsidence (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017). In exploratory wells from the Thanjavur sub-basin of the eastern Cauvery Basin (Figure 15b), lithofacies equivalent to the continental Terani Formation of the western sub-basins contain abundant marine dinoflagellates, indicating a Valanginian–Aptian age (Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017). However, the absence of foraminifera and ostracods suggests the presence of shallow, restricted marine-water bodies, comparable with sheltered marginal-marine lagoons unsuitable for sustained foraminiferal communities. The Valanginian transgression is more clearly expressed in the Palar Basin, where moderate dinoflagellate abundance is accompanied by low-diversity agglutinated benthic foraminifera (Basavaraju et al. Reference Basavaraju, Jaiprakash, Chidambaram and Ayyadurai2016; Raju et al. Reference Raju, Reddy, Jaiprakash, Gilbert and Sangwar2017). Comparable Valanginian marine sedimentation has also been documented in wells from the northern Krishna–Godavari proximal margin (Sinha, Reference Sinha2014).

4.b.2.c. Frontier basins of the western Australian continental margin

The complete opening of the eastern Indian Ocean involved uninterrupted marine communication between the proto-Mozambique Basin and the Tethys Ocean via an eastward extension of a Tethyan limb between the northeastern India–Greater India sector and the western Australian continental margin. Rifting between India and Australia initiated offshore northwestern Australia in the region of the present-day Argo Abyssal Plain and propagated southwards through a series of en echelon basins, with the Perth and Mentelle basins marking its southernmost extent (Geoscience Australia, 2020). Because geological remnants along the northeastern Indian conjugate margin are scarce, the western Australian margin provides the most reliable record of this rift evolution.

From the middle Jurassic to early Cretaceous, NW–SE-directed extension preceded the breakup between Greater India and Western Australia (Powell, Roots and Veevers, Reference Powell, Roots and Veevers1988). Following this breakup, rapid subsidence along the western Australian margin triggered widespread marine transgression and deposition of a fining-upward marine succession (the Cretaceous Winning Group) above a regional Valanginian unconformity surface (Daniell et al. Reference Daniell, Jorgensen, Anderson, Borissova, Burq, Heap, Hughes, Mantle, Nelson, Nichol, Nicholson, Payne, Przeslawski, Radke, Siwabessy and Smith2010; Geoscience Australia, 2020). In the Exmouth Sub-Basin of the Northern Carnarvon Basin, marine siliciclastic deposition is recorded from the Jurassic through the early Cretaceous, whereas on the offshore Cuvier Plateau, marine sedimentation is evident from the late Jurassic onwards. Farther south, fully marine conditions were established later, with marine deposition beginning in the Valanginian (Daniell et al. Reference Daniell, Jorgensen, Anderson, Borissova, Burq, Heap, Hughes, Mantle, Nelson, Nichol, Nicholson, Payne, Przeslawski, Radke, Siwabessy and Smith2010; Geoscience Australia, 2020). In the Southern Carnarvon Basin, rapid post-breakup subsidence likewise resulted in widespread transgression and deposition of marine sediments above the Valanginian unconformity (Geoscience Australia, 2020).

Core and rock samples from the Houtman and Zeewyck sub-basins of the Perth Basin and from the Cuvier margin document predominantly siliciclastic marine deposits of early Cretaceous age. Facies analysis indicates a progressive environmental transition from terrestrial to restricted marginal-marine settings (Berriasian–Valanginian), then to proximal shallow-marine environments (late Valanginian–Barremian) and finally to open-marine conditions (Aptian–Albian), evolving into distal open-marine facies during the Palaeogene (Daniell et al. Reference Daniell, Jorgensen, Anderson, Borissova, Burq, Heap, Hughes, Mantle, Nelson, Nichol, Nicholson, Payne, Przeslawski, Radke, Siwabessy and Smith2010).