Home › Features › The fells the map can’t take you up

The fells the map can’t take you up

Access that exists on the ground, but not in the dataset built to describe it.

We checked the access mapping for 378 of the Lake District’s lesser-known fells — the Birketts, Synges, Outlying Fells and Fellranger tops beyond the famous 214 — against the two open datasets that between them are supposed to describe where you may walk.

For 346 of them — 92% — the answer is clean: summit and approach sit on mapped access land, and you can walk there because the law says so, in so many words. That is worth saying first, because it is the real headline about the English Lakes. The right to roam works, almost everywhere, even on the hills nobody writes guidebooks about.

This piece is about the other 32.

Where the mapping runs out

3 fells reach their summits on a recorded right of way across enclosed land. 5 more have a mapped-access summit but a direct approach that crosses enclosed ground with no recorded path — solvable with a different start point. And 24 fells ended the computed pass in a category we called access-unresolved: no mapped access land at the summit, and no recorded right of way within reach of it. The datasets are simply silent.

So we spent a day chasing the public record for each of those 24. That is when the story changed shape.

The Thirlmere finding

Four of the silent fells — Great How, Fisher Crag, The Benn and Castle Crag of Shoulthwaite — surround one of the most-walked valleys in the national park. They have made paths, waymarks and car parks. United Utilities, which manages the Thirlmere estate, publishes an open invitation to walk there.

None of that appears in the official access dataset — because permissive access is not CRoW access. The landowner’s invitation, however genuine and however well furnished with car parks, lives on the landowner’s website, not the government’s map.

The same is true of Forestry England’s Setmurthy forest, which holds Watch Hill, and in softer forms of well-walked hills like Dunmallet and Cunswick Scar — whose paths are documented everywhere except the one dataset built to answer the question.

After that research pass, our 24 silent fells became 14. Ten resolved to published invitations or long-established routes the official mapping does not carry — at three quite different strengths of evidence, which is worth showing rather than flattening:

Ulgraves is the honest weak case: recent walk reports describe open access, but we found no statement from a steward or landowner. We publish it as such, and we have not drawn a route over it.

The stile statistic

The same gap appears at a smaller scale, and we can put our own number on it. For the 22 fells whose approaches cross a wall or a fence, we measured how many of the 23 crossings have a stile or a gate mapped in OpenStreetMap within 150 metres: 4 of 23 — fewer than one in five.

The barriers themselves are on the map — 12 walls and 11 fences, all traceable from an aerial photograph. The ways through them mostly are not (2 stiles and 2 gates between them), because a stile has to be stood next to, and nobody surveys Bannisdale.

Which is why our fell pages say “no mapped stile” and never “no stile” — and why, when you meet a wall with no obvious crossing, the answer is to walk the wall until you find one, never to climb it. A drystone wall is somebody’s working boundary, and it loses to boots every time.

What “silent” means — and doesn’t

For the remaining 14 fells we publish the silence itself: the open datasets show no access land and no recorded right of way to this summit.

That is a statement about mapping, not about the ground, and emphatically not about the law. Permissive arrangements often go unmapped, as Thirlmere proves at estate scale. A fell in this list is very often walked without incident.

We have chosen not to publish approach detail for these fells. Pointing a reader at a line we cannot connect to either a public right or a landowner’s invitation is not guidance, it is buck-passing.

• Birch Fell
• Burthwaite Heights
• Caermote Hill
• Clints Crags (Wainwright summit)
• Finsthwaite Heights
• High Hows (Lamplugh)
• High Knott
• Kelton Fell
• Little Meldrum
• St. John's Hill [Caermote Hill N Top]
• Staveley Fell
• Stone Pike
• Watermillock Fell [Little Mell Fell South Top]
• nameless (School Knott)

If you know one of these fells — a permissive path, an access arrangement, a landowner’s position — we would genuinely like to hear from you. This is exactly the kind of local knowledge the datasets are missing, and correcting the public record is most of the reason we did this work.

What we think should change

One thing, and it is narrow. Permissive access should be recorded in the open data.

Every case above where the record turned out to be wrong was a case where access genuinely exists and the map does not know it. United Utilities and Forestry England already invite people in; they gain nothing from the national dataset implying otherwise, and neither does anyone planning a walk. This is not an argument about how much access there should be. It is an argument that the access we already have should be findable in the place people look.

Sources

1. Natural England — CRoW Act 2000 Access Layer (OGL) - the mapped open-access land this page tests against (opens in new tab)
2. OpenStreetMap contributors — OpenStreetMap (ODbL) - rights-of-way designation tagging and barrier data; derived stats published as a Produced Work (opens in new tab)
3. Natural England — Open Access restrictions service - where closures and dog restrictions are actually announced (opens in new tab)

Access classifications and the per-fell research findings, with sources and dates, are published in our data files. Contains public sector information licensed under the Open Government Licence v3.0, and data from OpenStreetMap contributors, licensed under the ODbL. Computed 29 July 2026.