The research team, led by biogeographers working at the intersection of plate tectonics and genomics, constructed a new palaeogeographic framework spanning the last 35 million years. Rather than relying on broad regional tectonic models, the authors stitched together high-resolution reconstructions of Indonesian New Guinea and Papua New Guinea, which track the movement of crustal fragments in one to two million year time slices. These reconstructions draw on multibeam sonar, seismic surveys, isotopic dating, thermochronology and sedimentary records. The result is a dynamic map of what the team calls proto-New Guinea, the shifting mosaic of continental and oceanic crust along the northern margin of the Australian Plate that would eventually fuse into the world’s largest tropical island.

The model reveals something unexpected. For decades, some biogeographers argued that New Guinea was essentially submerged until about 15 million years ago, while others proposed it emerged far earlier. The new reconstruction shows that substantial land was present surprisingly early: roughly 330,000 square kilometres of emergent terrain existed by 35 million years ago, following the Peninsular Orogeny, when a volcanic arc called the Sepik Terrane slammed into the Australian Plate. Even during high sea-level episodes, proto-New Guinea persisted as an archipelago of large islands that periodically coalesced as sea levels fell, with emergent land fluctuating between roughly 210,000 and 320,000 square kilometres through the Oligocene and Miocene. For comparison, today’s island covers about 790,000 square kilometres.

Sea level proved to be the critical accomplice. When Antarctica first glaciated 34 million years ago, global sea level dropped by at least 50 metres, exposing vast shallow carbonate reef platforms that had built up along the Australian margin. The researchers reason that corals kept pace with rising seas for most of the Cenozoic, generating atolls and reef flats that repeatedly emerged during sea-level falls of more than 20 metres, forming temporary land bridges and stepping stones. Only during extreme highstands, such as the Mid Miocene Climatic Optimum, when seas stood more than 50 metres above the present, did the platforms stay drowned, and during the Pliocene a phenomenon the authors describe as the Pliocene Reef Gap meant that even later lowstands could not re-expose them, because coral growth had collapsed permanently in those areas.

Armed with this geological scaffold, the team quantified the changing distances between micro-tectonic landmasses, the coherent crustal blocks that drifted independently before merging into modern islands. Using GPlates and geographic information software, they measured pairwise distances every million years across 20 million years of history, applying a 10 kilometre uncertainty buffer. The analysis identified two golden windows for dispersal between New Guinea and Halmahera, an island now lying roughly 250 kilometres to the west. The first came around 13.8 million years ago, when tectonic transport along the Sorong Fault Zone brought Halmahera within 20 kilometres of proto-New Guinea while the Middle Miocene Climate Transition dropped seas by up to 50 metres. The second came around 8.2 million years ago, when a dramatic Milankovitch-driven lowstand exposed carbonate platforms and shrank water gaps of 60 to 70 kilometres to something far more crossable.

Enter the birds. The team built new dated phylogenies of the Paradisaeidae using randomly subsampled 10 kilobase genomic window alignments analysed in BEAST2, calibrated against the geological windows. Their enigmatic focus was the distribution of three species on Halmahera and nearby Obi: the dusky and Obi paradise-crows, Lycocorax pyrrhopterus and L. obiensis, and Wallace’s standard-wing, Semioptera wallacii. Two independent lineages on the same remote islands demanded an explanation. When the tree was calibrated to the older 13.8 million year window, the resulting divergence for Waigeo and Batanta endemics, Paradisaea rubra and Diphyllodes respublica, landed around 3.3 to 3.8 million years ago, during the high seas of the Pliocene Climatic Optimum, when those islands were more than 100 kilometres from New Guinea. Geologically, the researchers judged this scenario very unlikely.

The younger calibration told a far more coherent story. Under the 8.2 million year anchor point, both the paradise-crow ancestor and Wallace’s standard-wing lineage reached Halmahera within the same narrow window between 8.2 and 7.4 million years ago, plausibly hopping across emergent platforms via Obi and Bacan. The separation of the Obi paradise-crow from its Halmahera relative then falls around 1.9 million years ago, when Pleistocene cooling pushed seas 60 metres below present levels, allowing intermittent contact. The red bird-of-paradise and Wilson’s bird-of-paradise, endemic to Waigeo and Batanta, diverged from mainland relatives around 2.0 to 2.3 million years ago, precisely when the Sorong Fault had carried Waigeo closer than ever to the Bird’s Head Peninsula and when glacial lowstands repeatedly exposed land bridges. Every event in this scenario matches the geological record, whereas the older scenario fails on nearly every count.

The implications stretch far beyond birds-of-paradise. Songbirds, or oscine passerines, make up half of all bird species on Earth, and genomic evidence places their origin in the Australo-Papuan region before their worldwide radiation. Whether New Guinea could have hosted that early diversification has been fiercely debated. The new model shows that extensive, intermittently connected land existed on the northern Australian Plate from 35 million years ago, complete with wet forest refugia as Australia itself dried out. Independent evidence from lizards, insects, monotremes and pigeons is consistent with early colonisation of proto-New Guinea, suggesting the emerging archipelago served as an engine of avian diversity long before the modern island took shape.

What makes the study methodologically distinctive is its iterative, mutualistic logic. Geological reconstructions generated hypotheses about when dispersal was physically possible; phylogenetic divergence times then tested those hypotheses and, where they clashed, flagged which geological scenarios were most plausible. The same trick has previously clarified the amalgamation history of Sulawesi, and the authors argue that molecular clocks can serve as a genuine constraint on tectonic models, not merely a downstream beneficiary of them. Biology and geology, treated as partners, turn out to be more powerful than either discipline alone.

The study also carries a cautionary note for future biogeographic work across the Indo-Pacific, the largest expanse of tropical islands on the planet. Shallow marine carbonate platforms, often invisible in simple tectonic maps, acted as critical corridors that appeared and vanished with orbital-scale sea-level swings. Understanding their emergence history, the authors conclude, is essential for decoding the distribution of virtually any weakly dispersing terrestrial organism in the region. For now, the flamboyant birds-of-paradise have done what their forebears could not have imagined: carried rocks, reefs and shifting plates along on their wings, and helped geologists redraw the map of a vanished world.

Subject of Research: Palaeo-tectonic reconstruction of New Guinea and its role in shaping Indo-Pacific avian diversification

Article Title: A new palaeo-tectonic model of New Guinea and its consequences for understanding Indo-Pacific avian diversity in space and through time

Article References: Jønsson, K. A., Webb, M., Fotheringham, J., Gunawan, I., Haryoko, T., Iova, B., Melia, S., Nagombi, E., Reeve, A. H., Blom, M. P. K., Irestedt, M., & Rijsdijk, K. F. (2026). A new palaeo-tectonic model of New Guinea and its consequences for understanding Indo-Pacific avian diversity in space and through time. Frontiers of Biogeography, 19, Article e200936. https://doi.org/10.21425/fob.19.200936

Image Credits: AI Generated

DOI: 10.21425/fob.19.200936

Keywords: biogeography, New Guinea, birds-of-paradise, plate tectonics, sea-level change, phylogenomics, dispersal, vicariance, passerines, Halmahera, palaeogeography, Sorong Fault Zone

The Grabfeld Formation, a late Ladinian to early Carnian rock sequence sandwiched between the Erfurt and Stuttgart formations in the Central European Basin, was never an obvious place to look for thriving amphibian faunas. It formed under a generally arid climate dominated by alternating sabkha and playa conditions, where ephemeral brackish to saline lakes repeatedly expanded and evaporated. The classic sedimentary cycle, or cyclothem, consists of a claystone-dolomite horizon marking an ephemeral lake or marine ingression phase, followed by an evaporitic sulphate horizon representing the playa phase, and capped by a nodular sulphate claystone representing ancient palaeosols. Fossils are scarce, restricted largely to conchostracans, bivalves and rare fish remains that track rapid salinity swings. Temnospondyls, otherwise the most abundant tetrapod group in the German Triassic, were long considered notoriously rare here, and their taxonomy had remained elusive for decades.

The new review, led by Raphael Moreno of the State Museum of Natural History Stuttgart together with Sanjukta Chakravorti, Samuel L. A. Cooper and Rainer Schoch, assembles a total of eight specimens from Baden-Württemberg and Bavaria, all recovered from the upper part of the sequence known as the ‘Estherienschichten’. Four distinct temnospondyl taxa are now recognised: Metoposaurus sp., an indeterminate capitosaur, Gerrothorax sp. and Plagiosternum sp. The metoposaurid material includes an isolated angular bone found near Affaltrach in 1935 by a student named G. Buck, a fragmentary clavicle from Altheim originally described by Rupert Wild in 1974 but subsequently overlooked or rejected by later reviews, and a newly described fragmentary interclavicle from a field near Neustadt an der Aisch.

The key to identifying these broken bones lies in their surface sculpture. Temnospondyl dermal bones carry distinctive ornamental patterns of ridges, grooves and pits that differ between major groups. The Affaltrach angular, about 82 millimetres long, bears sinuously folded ridges radiating from its posteroventral margin, transitioning into neat, straight, striated ridges further forward, a pattern consistent with Metoposaurus and clearly different from the heavy reticulated network of Mastodonsaurus. The clavicle and interclavicle share a diagnostic combination of small, shallow polygonal pits confined to a narrow zone of the bone surface, with slender, evenly spaced radiating ridges that occasionally bifurcate in a Y-shape. In capitosauroids, by contrast, the polygonal pits are much deeper and wider, bordered by taller, steeper ridges. The absence of pustules rules out the plagiosaurids, which also occur in the formation.

Precisely dating the fossil-bearing horizons was essential to the study’s most striking conclusion. The Ladinian-Carnian boundary in the basin is marked by the bivalve index fossil Myophoria kefersteini okeni in the Bleiglanzbank of the middle Grabfeld Formation, while the base of the early Carnian Cordevolian substage falls within the Estherienschichten, indicated by the spinicaudatan Laxitextella multireticulata. The Anatinenbank, which yielded the Affaltrach angular, is exceptionally rich in Laxitextella taxitexta, an index for the upper Cordevolian, and palynological evidence supports this age. Milankovitch cyclicity analysis of the fossil-bearing horizons places the base of the Carnian at around 237 million years ago. The metoposaurid bones therefore predate the Carnian Pluvial Event, whose onset recent zircon dating places at about 234 million years ago in the Northwest Tethys region and as late as 231 million years ago in the Central European Basin’s Schilfsandstein.

This chronology rewrites the family’s history. Previously, the oldest metoposaurids in Central Europe came from the Stuttgart Formation, dated to about 231 million years ago, with the family’s global record beginning in the middle Carnian. Metoposaurids were nonetheless considered valuable tools for correlating terrestrial Triassic strata because they combined a near-global distribution with a short stratigraphic range. Outside Europe, the group is known from the Carnian of Morocco, where the Argana Basin has yielded the diverse assemblage of Dutuitosaurus, Almasaurus and Arganasaurus, from Madagascar and India with Panthasaurus maleriensis, and from North America, where metoposaurids enjoyed their longest run from the middle Carnian to the Rhaetian with genera including Anaschisma, Buettnererpeton and Apachesaurus. Indeterminate material even extends the family’s survivors into the late Norian of Zimbabwe, around 209 million years ago, more than ten million years past their last known record in India.

With the German specimens firmly placed in the Cordevolian and Julian 1 substages, the authors propose a three-wave dispersal scenario across Pangea. Metoposauridae, they suggest, first appeared in the Central European Basin around the Ladinian-Carnian boundary, a hypothesis supported by the earlier presence of Callistomordax kugleri, the sister taxon to Metoposauridae, in the underlying Erfurt Formation. A first radiation then coincided with the onset of the Carnian Pluvial Event near the Julian-Tuvalian boundary, with dispersal along aquatic pathways connecting southern Europe to northern Africa and onward to India and Madagascar, where emerging endemism developed under diverging climatic conditions. A second wave carried the group from Europe or northern Africa into North America, where early late Carnian occurrences of Buettnererpeton and Anaschisma across the Newark Supergroup and Wyoming point to widespread connected waterways. A third, later dispersal, emerging from India or Madagascar, may explain the high-palaeolatitude Zimbabwean record.

Equally provocative is what the bones suggest about metoposaurid physiology. The family has traditionally been interpreted as an inhabitant of strictly freshwater fluvial and lacustrine environments, and earlier claims of marine migration, based on the Moroccan Argana assemblage, were rejected when inland connections and the group’s absence from marine strata were demonstrated. Yet the Anatinenbank that produced one Metoposaurus specimen shows clear marine influence, containing the euryhaline sauropterygian Nothosaurus and invertebrates whose salinity has been described as equivalent to regular seawater. The condition of the temnospondyl bones, with clean fractures and minimal abrasion, suggests they were not transported far and may even be autochthonous, matching the preservation of the Nothosaurus material in the same layer. One metoposaurid interclavicle also comes from the palaeosol phase of the cyclothems, the most freshwater-restricted and terrestrial-influenced interval.

Taken together, these occurrences argue that metoposaurids tolerated a far wider range of salinities than previously credited, a conclusion echoed by other German deposits such as the Lehrberg lakes, whose salinity has been reconstructed as ranging from freshwater to euhaline, and by the playa-like settings of the upper Timezgadiouine Formation in Morocco. The plagiosaurid finds reinforce the picture: Gerrothorax has been characterised as a tolerant pioneer taxon coping with broad environmental gradients, while Plagiosternum, previously known from shallow marine, coastal and lagoonal settings of the Ladinian, now extends its stratigraphic range into the Late Triassic. The capitosaur intercentrum from Westheim, meanwhile, documents that group’s presence in the basin facies rather than only the marginal channel sands near the Vindelician Highlands, where the historically first German temnospondyl, Capitosaurus arenaceus, was found in 1836.

The study also raises the possibility that temnospondyls were never as rare in the Grabfeld Formation as assumed, and that decades of insufficient sampling simply concealed a regular faunal component of this harsh ecosystem. Fragmentary though the material is, its geographic and stratigraphic spread across the upper formation demonstrates that salinity-tolerant amphibians persisted through repeated environmental crises, from hypersaline playas to brackish incursions carrying Tethyan marine reptiles. Whether metoposaurids were genuine euryhaline wanderers capable of bridging marine corridors between continents, or merely hardy residents of fluctuating inland lakes, the early Carnian Germans of the Grabfeld playa now stand as the world’s oldest known members of one of the Triassic’s most successful amphibian dynasties, and theories of salt tolerance in the group can no longer be dismissed.

Subject of Research: Early Carnian temnospondyl diversity in the German Grabfeld Formation and the palaeogeography of metoposaurids

Article Title: Unexpected temnospondyl diversity in the early Carnian Grabfeld Formation (Germany) and the palaeogeography of metoposaurids

Article References: Moreno, R., Chakravorti, S., Cooper, S. L. A., & Schoch, R. R. (2024). Unexpected temnospondyl diversity in the early Carnian Grabfeld Formation (Germany) and the palaeogeography of metoposaurids. Fossil Record, 27(3), 381-400. https://doi.org/10.3897/fr.27.121996

Image Credits: AI Generated

DOI: 10.3897/fr.27.121996

Keywords: temnospondyls, Metoposaurus, metoposaurids, Grabfeld Formation, Triassic, Carnian, palaeogeography, Germany, Carnian Pluvial Event, Central European Basin, Gerrothorax, Plagiosternum

The central finding is staggering in its implications for how prehistoric life is understood in the region. Land once roamed by hunter-gatherers and later tended by early farmers now lies deep beneath the sea. The present-day Venice Lagoon, famous for its island city built on marshy, muddy ground, was once dry land that lay kilometres from the coast. The same is true of other towns in the lagoons further east, such as Grado, and even present-day coastal cities like Trieste would once have stood in hills overlooking swamps and lagoons below. In other words, some of the most archaeologically significant terrain in the region is invisible to conventional, land-based survey, hidden beneath seabed sediments that have accumulated over millennia.

Technically, the study is a palaeolandscape reconstruction built from two complementary strands of evidence. The researchers combined existing data from the analysis of sediment cores with new, high-resolution geophysical data acquired by OGS. Sediment cores allow scientists to identify different types of deposits, each characteristic of a particular environment, whether open marine, coastal barrier, lagoon, marsh or floodplain. Geophysical surveys, in turn, reveal the buried architecture of these deposits across wide areas of the seabed. By identifying the sediment types and modelling successive flooding episodes, the team was able to trace the backward shifting of coastal barriers, deposits running parallel to the shore, and the back-barrier systems, the marshy areas that lay beyond them. The result is a sequence of maps showing how the coastline retreated inland, stage by stage, across five millennia.

The timeline that emerges is anything but smooth. The reconstruction identifies a rapid rise in sea level around 8,400 BC that created tidal channels, marshes, sandbanks and large lagoons, followed by a more stable phase in which these features could persist and develop. Then came a large rise, a total of eight metres in the 1,200 years leading up to 6,600 BC, during which the sea repeatedly created, subsumed and recreated lagoons and marshes, pushing kilometres inland. A further major rise of up to five metres after 6,000 BC again shifted coastal features landward. Finally, after about 5,800 BC, the origins of the lagoons and the landscape recognisable today began to form. At times, the rate of rise reached up to 25 millimetres per year, a pace that would have been plainly visible within a single human lifetime.

That pace matters for understanding what prehistoric communities actually experienced. A rise of 25 millimetres per year translates into roughly a metre of relative sea level change across two or three generations, enough to permanently abandon settlements, alter hunting grounds and redraw the geography of trade and movement. The study highlights the constantly changing environment that Mesolithic hunter-gatherers had to cope with, as rising water altered habitats, resources and the availability of food such as plants, fish and animal prey. Coastal barriers were overrun and rebuilt; lagoons formed, filled and re-formed. Successive shaping and reshaping of coastal features influenced the lives of Mesolithic and Neolithic people and the development of local communities, as the sea swallowed features it had previously formed, shaped new ones, and eventually produced the origins of the modern coast.

Samuele Ongaro, lead author and PhD researcher at the University of Southampton, an archaeologist by training, emphasises why these drowned terrains matter. Submerged landscapes, he notes, are an important missing piece in the study of prehistory, especially in the Northern Adriatic, because they hold many secrets about how ancestors lived and how they were connected to the land. According to Ongaro, the results of the research allow the reconstruction of these lost landscapes, the environments they hosted, how they changed over time, and what significance they might have held for prehistoric populations, which he describes as a fundamental step for future archaeological research in the area. The framing is significant: rather than treating the seabed as a blank between terrestrial sites, the study treats it as a continuous, inhabited landscape worthy of systematic investigation.

The environmental variety of the reconstructed coast also helps explain why people were drawn to it in the first place. As Ongaro points out, the variety of coastal environments, lagoons, river deltas, marshes and swamps, would have been an attractive landscape for foragers and hunters, offering diverse and productive ecological niches. Yet the flat nature of the landscape also made it prone to significant land loss, and communities were forced to adapt over time, choosing where to hunt, fish and settle. This tension, between ecological richness and geographic vulnerability, is central to the study’s archaeological implications. The team hopes further study will reveal more about how communities shifted their activities in response to coastal flooding, and how this eventually affected the advent of farming in the region by Neolithic people, one of the most debated transitions in European prehistory.

From the marine geoscience side, the collaboration demonstrates the value of integrating geological and geophysical data. Dr Federica Donda, a marine geologist from the Geophysics section of OGS and co-author of the paper, comments that the study shows how integrating geological and geophysical data makes it possible to reconstruct the environmental evolution of areas that are now submerged, while highlighting that the seabed is an integral part of archaeological heritage. In her view, a significant portion of the areas frequented by ancient populations is now underwater, and their history and evolution cannot be understood through terrestrial archaeology alone. This is a pointed methodological claim: any model of Mesolithic or Neolithic settlement in the Northern Adriatic that ignores the drowned plains is, by definition, incomplete, because it excludes the very terrain where much of early coastal life unfolded.

The paper, titled Palaeolandscape reconstructions for the Northern Adriatic: Inundation models, depositional environments, and potential archaeological implications, underscores the potential of palaeogeographic reconstructions for studying the evolution of past societies. By returning to archaeological research a piece of territory that is no longer visible today, but which for thousands of years was part of the European landscape, the work opens a path toward predictive models of where submerged prehistoric sites might survive. The method of research is described as computational simulation and modelling, combining inundation models with depositional interpretations, an approach that could be applied to other drowned coastlines around the world where post-glacial sea level rise has erased the record of early coastal adaptation.

For a region whose modern identity is bound up with the fragile balance between water and land, from the Venice Lagoon to the gravelly shores of Trieste, the study offers a deep-time perspective on an ongoing process. The coast that residents and visitors know today is merely the latest configuration in a long sequence of flooding, barrier migration and lagoon formation that began when the last ice age ended. What the Southampton and OGS team has shown is that this sequence is readable in the sediments beneath the seabed, and that within those sediments lie clues to how some of Europe’s earliest coastal communities lived, moved and adapted as the sea rose around them. The next stage of research, the authors suggest, will focus on pinning down how human activity responded to each successive inundation, turning a drowned landscape into a map of prehistoric resilience.

Subject of Research: Prehistoric sea level rise and coastal landscape change in the Northern Adriatic

Article Title: Study reveals how prehistoric coastal flooding shaped the Northern Adriatic

Article References: Study reveals how prehistoric coastal flooding shaped the Northern Adriatic. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: Northern Adriatic, sea level rise, Venice Lagoon, Mesolithic, Neolithic, submerged landscapes, palaeogeography, coastal barriers, lagoons, sediment cores, geophysics, Quaternary Science Reviews