A second inquiry · Submerged landscapes

Another world
lies offshore.

The sea covered coastlines, rivers and places people once lived.

What survives—and what can today’s data reveal?

Begin with the shape of the land

Look beyond today’s coast.

Modern relief · depth experiment.
No ancient date assigned.

World relief from NOAA ETOPO1; present non-negative elevations in light green, negative elevations in blue — Present relief · an older global compilation

The same world relief with elevations from minus 60 metres to zero highlighted in gold; no date assigned — Gold = cells within the selected depth interval

Modern non-negative elevationsSelected shallow elevationsDeeper water

World view · static 60 m depth screen. Interactive controls require JavaScript.

Markers locate project regions. Map areas are not directly comparable.

Gold marks a depth interval on modern relief, including inland depressions. Land motion, sediment, erosion and past ice cover are not modelled. It is not an ancient coastline.

What would a map of 12,000 years ago require?

A dated regional sea-level model and a reconstructed former land surface, with uncertainty. The Last Glacial Maximum was earlier than 12,000 years ago; those shorelines differ.

North Sea peat records offer a starting point. “12,000 years BP” means before 1950, approximately 10,050 BCE; inundation dates vary by place.

Sea level and global ice volumes from the Last Glacial Maximum to the Holocene · Global sea-level rise in the early Holocene revealed from North Sea peats · North Sea peats: supplementary information and data

Prototype: NOAA ETOPO1 (2009), subsampled at 20′ globally and 2′–6′ regionally. Exact requests, extents, hashes and limitations. Source NetCDF files are retained in the repository.

The initial assessment

The missing landscape is real.
Its contents are a research question.

Documented submerged sites justify the search. How much evidence survives, and what it contains, remain open questions. Pavlopetri Underwater Archaeology Project: research impact record · Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in northwest Australia

Start where past habitability, preservation and usable surveys overlap.

Assessment updated 2026-09-20.

New · a test on actual survey data

What a coarse grid erases.

The Baltic alignment’s 0.5 m survey lets us measure how averaging removes fine relief. At 115 m, this test retains just 0.3% of local relief variance.

A data-resolution result, not an AI detection or a wall-recovery score.

Inspect the survey and measurements ↗

The field, in six starting places

What has come into view?

Discoveries, disputed claims and landscapes to investigate.

Active programme + published landscape evidence

Doggerland & the North Sea

A landscape we can begin to reconstruct

Peat records, buried channels and surveys constrain former landscapes. SUBNORDICA is developing methods to identify archaeological targets.

Evidence, limits & next investigation

Programme / record: SUBNORDICA · April 2024–March 2030
What remains open: A mapped river or dated peat bed does not by itself identify a settlement. Modern seabed relief may overlie a buried land surface.
A useful next step: Start with the open sea-level database and source-quality layers; reproduce a regional reconstruction before ranking candidate habitats.
Map precision: regional anchor

SUBNORDICA: beyond submerged landscapes · Global sea-level rise in the early Holocene revealed from North Sea peats · North Sea peats: supplementary information and data · SUBNORDICA: AI for Stone Age archaeology

Explore the regional depth view ↗

Disputed structure interpretation

Gulf of Khambhat / Cambay

A claim that needs its original context

NIOT researchers report sonar geometries and dredged samples. A dated, in-situ settlement has not been independently verified here.

Evidence, limits & next investigation

Programme / record: NIOT reports from the early 2000s; a current excavation programme was not verified
What remains open: The source’s grab/dredge samples do not, on their own, establish that dated material belongs to a human-built feature. This edition has not independently verified the reported city interpretation.
A useful next step: Seek original georeferenced sonar lines, processing history, repeat surveys and sample-context records. Separate a natural-feature test from a chronology test.
Map precision: regional anchor, not an exact feature position

Major marine archaeological discovery in the Gulf of Cambay and its significance

Explore the regional depth view ↗

Documented archaeological site

Pavlopetri, Greece

A submerged settlement, recorded in detail

Buildings, streets and ceramics document a settlement spanning the Final Neolithic and Bronze Age.

Evidence, limits & next investigation

Programme / record: Published survey and excavation programme begun in 2009; ongoing work not asserted
What remains open: This is a much later settlement than 12,000 years ago. Its submergence history cannot be applied to every shelf or monument.
A useful next step: Use the survey as a positive control for recording methods: which architectural features survive in bathymetry, sonar and diver photography?
Map precision: approximate site

Pavlopetri Underwater Archaeology Project: research impact record · Pavlopetri: 2010 field report

Explore the regional depth view ↗

Published archaeological discoveries

Murujuga’s submerged country

Stone artefacts in a drowned coastal landscape

Underwater stone artefacts at Cape Bruguieres and Flying Foam Passage show the potential of surveyed drowned landscapes.

Evidence, limits & next investigation

Programme / record: Deep History of Sea Country research reported in 2020; current field status not checked
What remains open: Transport, erosion and site formation matter. The inferred exposure history is not direct dating of every stone artefact.
A useful next step: Follow the published landscape-prediction method and engage the relevant Traditional Owners before any field-oriented extension.
Map precision: regional anchor

Aboriginal artefacts on the continental shelf reveal ancient drowned cultural landscapes in northwest Australia

Explore the regional depth view ↗

Published structure; function inferred

The Baltic stone alignment

A testable interpretation beneath 21 metres of water

Hydroacoustic, optical and sediment surveys document a roughly kilometre-long stone alignment. The authors interpret it as hunting architecture.

Evidence, limits & next investigation

Programme / record: Geersen and colleagues · 2024 study
What remains open: The existence and layout of stones are observations. Hunting function and the timing of construction depend on contextual arguments.
A useful next step: The local resolution audit now measures detail loss. Next: independent feature labels and natural look-alikes for a detection benchmark.
Map precision: regional anchor

A submerged Stone Age hunting architecture from the Western Baltic Sea · Baltic AUV survey · 0.5 m grid

Explore the regional depth view ↗

Reconstructed landscape / targeting opportunity

The Sunda shelf

A very large landscape with a much thinner archaeological record

Lower sea levels exposed extensive terrain between today’s islands. Published models examine inundation and population history.

Evidence, limits & next investigation

Programme / record: Published sea-level and population modelling, including a 2023 study
What remains open: Exposed terrain is not evidence of a lost city or of one connected civilisation. Local preservation, sediment burial and survey coverage remain decisive.
A useful next step: Identify accessible seismic and sediment datasets, then test palaeochannels and freshwater margins against a dated regional model.
Map precision: regional anchor

Prehistoric human migration between Sundaland and South Asia was driven by sea-level rise · Sea level and global ice volumes from the Last Glacial Maximum to the Holocene

Explore the regional depth view ↗

An inventory for investigation

Data we can begin with.

Sources, access and resolution, checked before analysis.

Blinkerwall, Baltic Sea