Ghost Roads | Roman Storyworld

I · The missing network

A province stitched together by roads—many of them still uncertain

Roman roads carried soldiers, messages, supplies and trade across Britain. Many now survive only in fragments: a slight bank in one field, a terrace beneath trees, a straight boundary on an old map—and then nothing. This investigation shows how those fragments are recognised and how carefully they must be interpreted.

LiDAR does not photograph the past. It measures the shape of the present ground with millions of laser returns. When trees and buildings are filtered away, the resulting terrain model can reveal a raised agger, a road cutting or the parallel ditches beside an abandoned route.

That power has transformed Roman-road research. It has exposed routes beneath woodland, challenged inherited maps and allowed independent researchers to search whole counties from a computer.

But a visible line is not a date. A medieval causeway, turnpike, railway embankment, field bank or processing artefact can look equally convincing. LiDAR reveals the question. Archaeology must still answer it.

The governing principle

LiDAR can show that a feature survives. It cannot, by itself, show that Rome built it.

II · How the laser sees

One landscape. Four ways of looking.

Archaeologists rarely trust a single visualisation. Change the rendering and a feature can emerge, disappear—or acquire a misleading shadow of its own.

Digital Surface Model

The recorded surface includes trees and buildings. Archaeological earthworks may be hidden beneath the canopy.

Illustrative terrain created for this investigation—not real LiDAR data.

Laser pulse

An aircraft records the time taken for light to return from the ground and objects above it.

Point cloud

Millions of measured points form a three-dimensional record of elevation.

Filtering

Algorithms classify and remove many trees and buildings to create a bare-earth model.

Interpretation

Researchers compare multiple visualisations, maps, aerial photographs and known Roman sites.

III · Reading the landscape

Four roads. Four different kinds of evidence.

Each case begins with what the landscape appears to show. Reveal the evidence to see how archaeologists move from visual impression to a measured conclusion. There are no questions to answer and no score to earn.

Case one · A line beneath woodland

A raised feature runs towards two known Roman sites

Bare-earth LiDAR reveals a straight bank and shallow parallel hollows where ordinary aerial photography sees only trees.

Case two · Blackstone Edge

A monumental road long celebrated as Roman

The road is straight, stone-built and dramatic—the image many visitors expect a Roman road to resemble.

Case three · The empty ploughed field

The terrain model shows no road at all

The expected route passes between known Roman places, and aerial cropmarks suggest disturbance—but the LiDAR image is silent.

Case four · Forton Hall Farm

The ground is opened on the line predicted by LiDAR

A drainage cut exposes a broad concentration of substantial stones precisely where the suspected road should pass.

What the four cases teach

The same technology can reveal, suggest, remain silent or mislead.

Archaeological judgement comes from combining the image with construction, chronology, route context, historic mapping, aerial evidence and excavation—not from rewarding the straightest line on the screen.

IV · Case files

Where the laser changed the map—and where it did not

Lancashire · Forton Hall Farm

When the ground answered the image

A faint alignment suggested a route towards Cockerham. When a field drain later exposed the proposed line, a broad concentration of large foundation stones appeared where the LiDAR evidence predicted it.

Physical exposure supports the LiDAR interpretation

South Downs · Chichester to Arundel

The road beneath the High Woods

A route long suspected from older mapping became traceable across kilometres of woodland when bare-earth LiDAR was combined with aerial photography. The forest had concealed the form, not destroyed it.

Strong remote-sensing and landscape evidence

Devon and Cornwall · Regional network

From fragments to a modelled system

Researchers mapped approximately 100 kilometres of additional probable road and used GIS movement models to propose connections through the gaps. North Tawton emerged as a more important hub than older maps suggested.

LiDAR observations plus predictive reconstruction

Pennines · Blackstone Edge

The beautiful impostor

One of Britain’s most celebrated “Roman roads” is now widely interpreted as a much later turnpike. LiDAR can describe its form with precision, but neither straightness nor monumentality provides a Roman date.

Disputed identification—probably post-medieval

V · A landscape case study

Nettleford Wood: where the national question becomes local

The Cheshire Sandstone Ridge offers a clear example of how a Roman road is studied in a landscape altered by woodland, quarrying, later roads and changing land use. Here, no single clue is enough on its own.

Interpretive route diagram. It does not replace Historic Environment Record mapping or field survey.

A layered corridor

Road, ridge, quarry, forest and memory

This case study brings together fieldwalking, LiDAR, historic maps, archaeology and local topography to show how the course of a Roman road is investigated rather than simply assumed.

As the route passes through Nettleford Wood and the Cheshire ridge, later quarrying, woodland and road-building complicate what survives. Some evidence remains visible; some is displaced, buried or preserved only in older records.

The aim is not merely to locate a line on a map. It is to understand how archaeologists distinguish a Roman road from later routes, natural features and the accumulated memory of the landscape.

Documented corridor

Deva–Condate route, recognised junction and archaeological records.

Landscape evidence

Cuttings, alignments, ridge topography, LiDAR and historic mapping.

Interpretive reconstruction

How surviving fragments connect through altered ground.

Landscape memory

Place names, quarry stories and later movement along the ridge.

Explore the evidence behind this case

VI · What the laser cannot see

Absence of relief is not absence of a road

≈

Ploughed away

Repeated cultivation can flatten the agger until no measurable surface form survives.

Use instead: cropmarks, geophysics, excavation

▤

Buried below deposits

Alluvium, peat or later soil can seal the Roman surface beneath the terrain LiDAR measures.

Use instead: coring, geophysics, stratigraphy

▥

Hidden by the city

Buildings, cellars, utilities and later roads make urban routes largely inaccessible to airborne terrain models.

Use instead: watching briefs, historic excavations

⌇

Too lightly built

Minor tracks may never have possessed a substantial raised agger capable of surviving as relief.

Use instead: route logic, field survey, soil evidence

◒

Wrong light angle

A feature aligned with simulated illumination can vanish from a standard hillshade.

Use instead: multi-directional hillshade, openness

≋

Processing artefact

Local relief models can create phantom banks or ditches around other forms.

Use instead: compare raw data and several renderings

LiDAR sees the skin of the landscape. Archaeology asks how that skin was formed.

VII · The evidence ladder

From visible line to Roman road

The map should never draw every line with the same authority. Each route belongs on an evidence rung.

Confirmed by excavation

Construction, stratigraphy or dating securely supports Roman origin.

Strong multi-method evidence

LiDAR, aerial evidence, alignment and network context converge.

Probable alignment

Convincing fragments survive, but important sections or dating evidence are absent.

Predictive reconstruction

GIS modelling proposes the most plausible corridor between evidence points.

Disputed or rejected

The feature is visible but may belong to another date, function or process.

No surviving surface expression

LiDAR is silent. Other evidence may still preserve the road.

VIII · Go deeper

Explore the evidence in more detail

The investigation above provides the main story. The papers below are optional next steps for visitors who want to examine the maps, field observations and archaeological arguments in greater depth.

New to the subject? Begin with Lost Lines, Living Land.

Interested in construction? Open the Nettleford Wood road report.

Want the fullest route study? Choose the comprehensive Watling Street report.

Start here · Accessible overview

Lost Lines, Living Land

Roman roads through Kelsall and Nettleford Wood, combining fieldwalking, LiDAR, historic maps and landscape memory.

Explore this paper → Best for · Road construction and junctions

Roman Road at Nettleford Wood

Construction, route use, junction evidence and the relationship between Margary 7a and 7aa.

Explore this paper → Best for · The complete route study

Watling Street: Nettleford Wood to Delamere

The route as military infrastructure, later corridor and remembered feature of the Cheshire landscape.

Explore this paper → Best for · Landscape and local memory

King’s Chair and Hangingstone Hill

Road, quarrying, forest law, place-name traditions and the accumulated meanings of the ridge.

Explore this paper → Best for · A focused location study

Archaeology at Watling Street, Kelsall

A focused assessment of evidence around the road alignment east of Nettleford Wood.

Explore this paper → Specialist reading · Cautious interpretation

Stoney Lane and Eddisbury

A cautious assessment of possible ritual indicators and reported features near the Watling Street corridor.

Explore this paper →

Archive Browse all Roman Britain research Open the research archive → Map Roman roads, forts and tribal landscapes Open the interactive map → Routes The Antonine Itinerary in modern geography Follow the itineraries → Investigation The People Beneath the Landscape Explore regional population history →

IX · Explore further

See the landscape through the official data

These official viewers are available for anyone who would like to explore further. They are not required to understand the investigation; they simply offer a closer look at the same kinds of terrain data used by researchers.

England

Defra Survey Data Download

Download Environment Agency terrain, surface and point-cloud products.

Wales

DataMapWales LiDAR Viewer

Explore Welsh Government and historic LiDAR products.

Scotland

Scottish Remote Sensing Portal

Follow new national LiDAR releases and historic datasets.

Archaeology

Aerial Archaeology Mapping Explorer

Compare mapped archaeology identified from aerial and remote-sensing sources.

Before exploring a possible site

Treat every apparent feature as a possibility, not a discovery. Compare several visualisations and consult Historic Environment Records and old maps. Never enter private land without permission, dig, disturb a feature or publish a vulnerable site carelessly. A line on a screen is a reason to learn more—not permission to excavate.

X · Sources and further reading

Where the evidence comes from

The main claims on this page are grounded in official datasets, archaeological organisations and published research. The list is provided for transparency; visitors do not need to read every source to follow the investigation.

Historic England — Lidar and airborne archaeological survey

Official guidance on the archaeological use of airborne laser scanning and its visualisation.

Open source →

Environment Agency — National LIDAR Programme and composite DTM

Official national elevation data at one-metre spatial resolution across England.

Open source →

DataMapWales — LiDAR viewer and downloads

Welsh Government and Natural Resources Wales terrain and surface data.

Open source →

Scottish Government — Scottish Land LiDAR Programme

National capture programme and first open data releases.

Open source →

Parcero-Oubiña, Smart and Fonte (2023)

Remote Sensing and GIS Modelling of Roman Roads in South West Britain.

Open paper →

Historic England — The Lost Roman Road from Chichester to Arundel

LiDAR and aerial-photographic evidence for the long-speculated road through the South Downs.

Open case study →

Roman Roads Research Association

National resources, gazetteers, lectures, geophysics and peer-reviewed road research.

Open RRRA →

Ratledge — Britain’s Roman Roads: A LiDAR Re-appraisal

County-scale research, imagery and route reassessment by a leading independent interpreter.

Open research site →

Ratledge — Blackstone Edge, Margary 720a

The evidence against one of Britain’s most familiar supposed Roman roads.

Open gazetteer entry →

Turton (2024) — GeoAI and automated road detection

A deep-learning experiment combining LiDAR, historic mapping and aerial imagery between the Roman frontiers.

Open paper record →

Roman Storyworld — LiDAR and the Roman Roads of Britain

The analytical report from which this interactive investigation was developed.

Browse the research archive →

What to take away

The road is not simply the line on the screen

The road is the bank, the ditch, the stone beneath the soil, the route between settlements, the later lane that borrowed its course and the argument that survives when the evidence breaks.

LiDAR has made the lost network newly visible. Its greatest value is not instant certainty, but a clearer way to recognise evidence, understand uncertainty and ask better questions of the land.