Mysteria

About the atlas

An atlas that shows its evidence.

Mysteria catalogues the places on Earth that still hold open questions — megaliths, pyramids, buried and sunken cities, geoglyphs, rock-cut temples — and says plainly, for each one, how much of the story the evidence actually supports.

places
4,496places
countries
137countries
UNESCO sites
670UNESCO sites
contested or unproven
183contested or unproven

What this is

One record per place, kept honest.

There are two kinds of books about ancient mysteries. One prints every legend as fact. The other leaves out anything that has not been settled. Both are unsatisfying, because the interesting part is precisely the gap: the thing that was built, the thing we can measure, and the thing nobody can yet explain.

So each entry here carries three layers. What is documented — coordinates, period, culture, excavation status, dimensions, UNESCO inscription. What is open — the questions archaeologists still argue about. And what people claim — including claims the field has rejected, kept in the atlas and labeled as such rather than quietly deleted.

Around that record sits an atlas you can actually use: a live map of every place, filters by kind, country, period and confidence, and a community that keeps lists, writes reviews from the ground, and argues about interpretations in the open.

The data

Where the record comes from.

The catalogue is built from MysteriaData, an open dataset of archaeological and contested sites compiled from published excavation reports, museum and national heritage registers, the UNESCO World Heritage List, and the peer-reviewed literature cited on each page. It is released under CC BY-SA 4.0 — you may reuse it, including commercially, if you credit Mysteria and share your derivative under the same licence.

4,496
Places catalogued
10,680
Mapped sub-features
3,479
Credited photographs
13,543
References & sources

How a record is kept current

Imports are idempotent. Every place has a permanent internal id, so re-importing the dataset updates the record in place rather than creating a second one; renamed entries keep their old URLs as aliases, and anything an editor has corrected by hand survives the next import. Near-identical entries within two kilometres of each other are merged, with the weaker record redirected to the stronger one. Places removed upstream are marked rather than deleted.

Records vary in depth. A full entry has a long description, sourced history and photographs; a stub is a real place we have little to say about yet. Both are labeled, and nothing is padded with invented detail.

How we verify

Six levels of confidence.

Every place carries a seal — on its card, its map pin and its page — describing how strongly the evidence supports what is claimed about it. The seal is about the claim, not about how interesting the place is. A stone circle whose date is settled and a submerged formation whose very artificiality is disputed both belong here; they simply carry different seals.

  1. Verified

    4,121 places

    Peer-reviewed consensus: inscribed, extensively excavated, dating corroborated.

  2. Widely accepted

    192 places

    Broad scholarly acceptance with minor debate on details.

  3. Contested

    43 places

    Active scholarly dispute on authenticity, date, or interpretation.

  4. Hypothetical

    74 places

    Plausible but insufficient published evidence; often known from remote sensing or legend.

  5. Unverified

    0 places

    Claimed but lacking peer-reviewed field verification.

  6. Pseudoarchaeology

    66 places

    Claims rejected by the scholarly community; included and labeled, never hidden.

Seals are assigned from the state of the published literature and reviewed by editors, not by vote. When a place moves between levels — as excavation, dating or a new survey settles a question — the change is recorded in its history. If you think a seal is wrong, say so on the place’s discussion tab with the citation that proves it.

Licenses & credits

Who made what you are looking at.

Mysteria is assembled from open material, and every layer keeps its own licence. The short version: the record is share-alike, the photographs belong to the people who took them, and the map belongs to OpenStreetMap.

The catalogue

Descriptions, coordinates, histories and verification seals. Reuse it with attribution to Mysteria, under the same licence.

Photographs

Images are served from Commons and credited individually: photographer, licence and a link back to the file page appear beside every photo. Licences differ per image — most are CC BY-SA or public domain. Check the credit before reusing one.

The map

Vector tiles are served by OpenFreeMap from OpenMapTiles schemas, built from OpenStreetMap data, which is licensed ODbL 1.0. Basemap styles are © OpenMapTiles.

Type & interface

Set in Fraunces for display, Instrument Sans for text and JetBrains Mono for coordinates and counts. Interface icons are Lucide (ISC).

Are you the rights-holder of a photograph you would rather not see here? Tell us and it comes down the same day — see below.

Corrections & contact

Tell us what we got wrong.

This atlas is wrong somewhere right now — a date, a coordinate, a seal, a photograph of the wrong hill. Corrections are the most useful thing you can send.

  • A specific place is wrong

    Open its page and use the discussion tab, or flag the entry. Include the source that settles it.

    Find the place
  • You have a source we are missing

    Submit a paper, book, documentary or news story and we will attach it to the places it covers.

    Submit a resource
  • Licensing, takedowns, press

    Email hello@mysteria.co. Rights-holder requests are handled first.

    Email us

Join the expedition

An account lets you keep lists, review places and post theories.

Search Mysteria

Search places, or jump to a section