Woher jedes Feld stammt.

Fünf Quellen, jede mit eigener Lizenz, dazu eine redaktionelle Schicht obendrauf. Diese Seite nennt, woher ein Ortsdatensatz stammt und was seine Lizenz verlangt, damit du uns prüfen kannst. Können wir nicht sagen, woher ein Ort stammt, veröffentlichen wir ihn nicht.

3,474 places in the map data · 3,492 written up

01 · Die Quellen

Eine Ebene pro Quelle.

Die Zahl auf jeder Kachel ist die Anzahl der Datensätze, die aktuell in den veröffentlichten Kartendaten dieser Quelle zugeordnet sind, kein Ziel.

02 · Unsere ODbL-Erklärung

Die Kartenebene ist eine Derivative Database.

The OpenStreetMap-derived layer published on this site, including the GeoJSON files served under /map/, is a derivative database of OpenStreetMap and is made available under the Open Database License 1.0. The underlying data is © OpenStreetMap contributors.

Unsere redaktionellen Seiten sind ein Produced Work im Sinne dieser Lizenz und werden selbst nicht unter der ODbL angeboten.

We keep each source in its own layer and never deduplicate across layers. That is deliberate. Merging the OpenStreetMap layer with a differently licensed one would create a single derived database and pull the other source into ODbL's share-alike, which we have no right to do. Every feature we publish carries its own source property so the layers stay separately identifiable.

03 · Was wir nicht verwenden

Nutzereinträge aus anderen Apps.

Wir scrapen die kommerziellen Camping-Apps nicht. Ihre Einträge sind ihre eigenen Datenbanken, gesammelt von ihren eigenen Nutzern, und sie zu übernehmen wäre sowohl ein Lizenzbruch als auch schlicht Diebstahl fremder Arbeit. Vor allem aber ist ein Nutzereintrag keine offizielle Ausweisung. Wir behandeln so einen Eintrag als Hinweis, den man bei der Gemeinde nachprüfen sollte, nie als Daten zum Veröffentlichen.

Alles hier stammt aus offenen Daten oder von einer offiziellen Seite, die wir nennen. Keine Affiliate-Links, keine bezahlten Platzierungen, und nichts auf dieser Karte ist käuflich.

04 · Bekannte Lücken

Was diese Daten dir nicht sagen.

Gezählt nach dem, was heute veröffentlicht ist. Wir zeigen lieber die Lücke, als sie mit einer Vermutung zu füllen.

Die Ausstattung ändert sich ohne Vorwarnung. Jede Ortsseite trägt das Datum, an dem wir sie zuletzt mit der Quelle abgeglichen haben, und ein Schild an der Einfahrt schlägt alles, was wir veröffentlicht haben.

Die Land-für-Land-Übersicht, welche Quelle welche Ebene speist und wie weit jede davon reicht, folgt weiter unten.

05 · Land für Land

Was die Ebene jedes Landes wirklich abdeckt.

Welche Quelle welches Land speist, wie viele Orte sie liefert, und wo sie aufhört.

What our data covers, and what it does not

Our French layer at launch is OpenStreetMap only.

The DATAtourisme national dataset, which is the obvious second source for French municipal aires, is not in our pipeline. It requires a diffuseur account and an API key that we have not set up. Until that changes, anywhere French mappers have not recorded a place, we do not have it. We would rather say that plainly than let a map imply national coverage it does not have.

France is the country with the best official data, and it is what the map runs on. DATAtourisme is the French public tourism dataset, published under the Licence Ouverte by the regional tourism boards themselves.

  • 1,890 aires are on the map, each with a written page: 1,866 from DATAtourisme and 24 from OpenStreetMap. The map carries nothing else.
  • 2,041 more sit in the source layers, unwritten and unmapped, and are the queue. Filtered to fee=no, OpenStreetMap alone claims 2,066 free French places, 2,018 of them named, but a fee=no tag is a contributor’s assertion and the commune’s own page is what we write from.

Every record in that set carries a needs_designation_check flag, and it stays set until a person confirms the designation against an official source: the commune, the intercommunalité, or the départemental tourism authority that lists the place. An OSM record earns a place a spot in the count. It does not earn it a written page with coordinates on it. Places that fail the check stay in the data and out of the pages.

Two consequences worth stating. The data skews toward what mappers bothered to record, which in practice means well-signed communal aires and away from newer private networks. And an absent amenity in our data means nobody recorded one, not that the place lacks it. A blank is a blank, never a no.

You can see the whole layer on the France map.

Where there is enough data to write a region page

Eleven French regions have enough named free places for a region page to be worth reading. Counts are named places from OpenStreetMap at launch:

RegionNamed free places
Auvergne-Rhône-Alpes364
Nouvelle-Aquitaine336
Occitania221
Bourgogne-Franche-Comté179
Grand Est178
Centre-Val de Loire160
Pays de la Loire127
Brittany117
Normandy114
Provence-Alpes-Côte d’Azur99
Hauts-de-France47

Two regions are too thin to write about honestly: Corsica with 3 named places and Île-de-France with 6. That is a statement about our data, not about Corsica. Corsica has motorhome infrastructure; OpenStreetMap contributors have not recorded much of it as free, and we are not going to pad a page to cover the gap.

What our data covers, and what it does not

Be clear-eyed about this one, because Germany is our thinnest large country.

Our German layer is OpenStreetMap only. There is no national German open dataset equivalent to France’s DATAtourisme, so there is no official government feed to cross-check against and nothing to balance the crowd-sourced layer with. That makes Germany more OSM-dependent than France and, honestly, less complete.

The numbers as they stand:

  • 356 places are on the map, across seven states, and every one has a written page behind it checked against the Gemeinde’s or the operator’s own listing. The map carries nothing else. A pin with no page behind it would make this site a slower way of opening OpenStreetMap.
  • 682 more sit in the source layer, unwritten and unmapped. They are tagged fee=no in OpenStreetMap, which is a contributor’s assertion rather than a municipality’s, and each one carries needs_designation_check until a person has verified it against an official source. That check is the work; the tag is only the lead.
  • So a blank stretch of German map means nobody has written that place up yet, not that there is nothing there. Where a Gemeinde publishes a rate and OpenStreetMap says free, the Gemeinde is right and the page says so.

So the map is a starting point and the written pages are the finished work, and there is a large gap between the two in Germany right now. We would rather say that than pad the country out.

Regions with enough data behind them for a written region page: Bavaria (169 named places), North Rhine-Westphalia (168), Baden-Württemberg (161), Lower Saxony (127), Hesse (107), Rhineland-Palatinate (71), Schleswig-Holstein (57), Saxony (42), Brandenburg (37) and Thuringia (35). Three are too thin to write up honestly: Saxony-Anhalt (20), Saarland (10) and Mecklenburg-Vorpommern (8). Those last three will stay on the map and stay unwritten until the data supports more.

If you want the broader picture of how Germany sits next to its neighbours, start with free camping in Europe. If the vocabulary is the problem, wild camping vs free camping vs aires untangles three words that get used as one.

What our data covers, and what it does not

Be clear-eyed about this before you plan around our Spain pages.

Spain has no national dataset, and that is not a gap we can close. Tourism is a competence of the comunidades autónomas under article 148.1.18 of the Constitution, and stopping and parking on urban roads belongs to the ayuntamientos. There is no Spanish body that could publish a national list of áreas de autocaravanas. We checked rather than assumed: the official INSPIRE register returns nothing for autocaravana, the national spatial catalogue returns a single record (a car park in Pamplona), and the CNIG download centre has no camping or rest-area product at all. The national open data portal carries five relevant datasets for the whole country, none of them national.

Two sources, two shapes. Castile and León comes from one register. Asturias comes from 17 town halls, one at a time. The first: Junta de Castilla y León’s áreas de servicio para autocaravanas, published on the regional tourism portal. It holds 126 áreas, with coordinates and a page per record, and it states for each one whether it is an área pública, an área privada or an área en camping. That last field is why we can use it: it answers “did a public body designate this” from the source rather than from our guesswork.

We publish 98 of the 126. The 27 the register marks privada or en camping are businesses rather than town stopovers, and one record has no coordinates. The 19 it marks only Parking, Área en ruta or Área multiservicios are included, and their pages say the register does not state who runs them, because it does not.

Treat the pública label as a strong signal, not a guarantee. Two of the 80 records tagged área pública are plainly roadside businesses: Área de Petrovicma on the N-620a, and Béjar Caravan Park. That is why our pages report what the register says rather than promising you a town owns the tarmac.

The register prices nothing. It carries no fee, no capacity, no stay limit and no services for any área in the region, so all of those stay blank on our pages and every one of them tells you what to do instead. Bring coins, arrive self-sufficient, and read the sign at the entrance.

Reuse. The Junta’s general conditions permit reuse for commercial and non-commercial purposes, and require the citation Origen de los datos: Junta de Castilla y León, which every place page carries. We take no images from the portal, which its conditions exclude.

Asturias: 20 áreas, one town hall at a time

Asturias has no register to read. The Principado publishes no áreas dataset anywhere: we checked its whole open data catalogue and the SITPA-IDEAS spatial portal, and there is no camper layer in either. Decreto 61/2022 does make “áreas especiales de acogida de autocaravanas en tránsito” a registrable category, so the government holds a list, but it does not publish it.

There is one good regional directory, run by the Principado’s tourism marketing company. We do not use it. Its aviso legal expressly prohibits reproducing the site’s contents “con fines comerciales” without permission, and this site carries display advertising. It was never opened during this work.

So each Asturian área here was found on its own ayuntamiento’s page and read there. 47 concejos were checked. 20 áreas across 17 of them could be published; the rest either publish nothing, or publish an área nobody can locate. Where the town publishes coordinates we use them and cite the town. Where it does not, the coordinate comes from OpenStreetMap under ODbL and the municipal page is cited as the evidence that a public body designated the place.

Four real áreas are named on the region page but not pinned, because no coordinate exists for them in any source we trust: Llanes (86 bays, the largest in the region), Tapia de Casariego (the one clearly priced área, 9 euros a day in high season), Noreña and Salas.

The rest of Spain, and why it is not here yet

We looked at all seventeen comunidades autónomas. What we found:

RegionWhat existsWhy it is not published here
Castile and León126-área register, coordinates, per-record pagesPublished: 98 places
AsturiasNo regional dataset. 78 concejos publishing separatelyPublished: 20 places, each read on its ayuntamiento’s own page
Castilla-La Mancha22 genuinely municipal áreas, CC-BY-SANo coordinates and no per-record page. Needs a geocoding pass
Catalonia7 áreas in the tourism registerThe register covers only paid areas by law. Free municipal àrees are excluded from it
Comunitat Valenciana13 áreas de pernoctaSame: the decree defines these as areas open a canvi de preu
GaliciaStructured área records on the Xunta’s portalNo complete register we can work through
Cantabria11 to 14 areas listed by the regional tourism boardNo reuse terms published at all, and the only coordinate source is licensed non-commercial
Euskadi2 motorhome areas, CC-BY 4.0, per-record pagesReal but tiny
Canarias5 caravan zones on Tenerife, CC-BYTent and caravan zones inside forest parks, not town áreas
NavarraRegister category created in 2024The network is still being built. Nothing published yet
OthersNothing usable foundAragon files with 33 comarcas; Andalusia, Extremadura and Murcia publish no list

Where we have no places, the region page tells you the method instead: look up the ayuntamiento’s own page and its ordenanza municipal. That is slower, and outside Castile and León it is what actually works.

Known gap: the map and the pages disagree

The Spain map is still drawn from an OpenStreetMap extract, which is a separate pipeline from the pages. So the map shows more pins than we have written pages for, and those extra pins are contributor records we have not checked against a municipal source. Trust a page over a pin: a page means we opened the register entry behind it and it resolved.

Everything we map is a designated stopover with a coordinate. Wild camping is not on the map and will not be, in Spain or anywhere else. If you want the distinction spelled out, see wild camping, free camping and aires, and for how Spain fits the wider picture, free camping in Europe.

Browse what we have on the Spain map.

What our data covers, and what it does not

Italy is our smallest written layer and the one that took the most work per page, and it is worth understanding why before you plan around it.

There is no national Italian dataset. Nothing equivalent to France’s DATAtourisme or Denmark’s GeoFA exists. The Italian national open data catalogue holds 20 camper datasets in total, from four publishers, and not one of them is a register of municipal stopovers. Searching every regional catalogue that would plausibly hold one returns nothing: Veneto’s 6,606 datasets, Emilia-Romagna’s 2,880, Trento’s 1,380, Bolzano’s 917, Friuli-Venezia Giulia’s 810, Lazio’s 407. Where a region does publish an accommodation register, aires are structurally absent from it, because those registers exist for businesses filing a prices declaration and a comune does not file one.

So the Italian pages are built one comune at a time. OpenStreetMap supplies the candidate and the coordinate. Then somebody opens the page where a named public body says the place exists: the comune’s own website, an ordinanza or regolamento, a provincial tourism agency, an unione di comuni, or a national park authority. Only the ones that survive that check get written up. Each page names the body, links the page, and carries the date it was read, plus a line of the original Italian so you can check our summary against it.

The numbers. 65 written places across four regions: Emilia-Romagna 21, Piedmont 20, Veneto 15, Tuscany 9. On price, 19 are free, 6 are free to park with paid water or electricity, 14 charge, and 26 have no published price at all.

That last figure is the honest one to sit with. For 26 of 65, the comune documents the stopover and says nothing about money, so we say nothing about money either and tell you to bring coins.

RegionWritten places
Emilia-Romagna21
Piedmont20
Veneto15
Tuscany9

The map shows the 65 and nothing else. There is a wider and weaker layer behind it, 327 more Italian places drawn from OpenStreetMap and filtered to fee=no, and none of them are on the map. Those are contributor claims rather than comune ones, and the checking work above found several already wrong: Firenzuola has charged since August 2025, Feltre charges 4 euros per 12 hours, and Verona’s Porta Palio is 15 euros rather than the 10 the tag says. Publishing them as pins would pass that error on to you wearing our name, so they stay a work queue: leads to check, not places we are telling you to drive to.

Regions we have not written yet. Everywhere else in Italy. That includes places people genuinely drive to, and it is not because we did not look: the whole of the south and the islands yielded one municipally-run area with a licence and a coordinate. We would rather leave the gap visible than fill it with guesses.

Everything we map is a designated stopover with a coordinate. Wild camping is not on the map and will not be, in Italy or anywhere else. If the vocabulary is the problem, wild camping vs free camping vs aires untangles three words that get used as one, and free camping in Europe puts Italy next to its neighbors.

What our data covers, and what it does not

Denmark is our strongest country, because it is the one with a real public dataset behind it. GeoFA, the Danish public geodata that also feeds udinaturen.dk, carries the state shelter network with a named responsible authority on each record.

  • 1,050 places are on the map, each with a written page: 1,045 from GeoFA and five from OpenStreetMap. The map carries nothing else.
  • 1,169 more sit in the source layer, unwritten and unmapped, and are the queue.
  • 502 of the written ones must be booked, free but reserved, which the page and the card both say. A card reading only “Free” would send you to a shelter somebody else has reserved for tonight.

An earlier version of this page described a Danish layer that was OpenStreetMap only, 537 places and no shelters at all, which was true before the GeoFA import and is the reason that import happened.

That last line is the one that matters. The state shelter network is not represented in our Danish data at all, so any impression of shelter coverage from our map would be false.

The 501 campsite records are OpenStreetMap tourism=camp_site features tagged as free. They range from municipal free campsites to small nature campsites a contributor recorded, and what a given record actually is on the ground varies more than the tag suggests. The 36 aires are motorhome stopovers, the same category France built its network from.

Every record carries a designation check that stays open until a person confirms it against an official source: the kommune, the agency, or the operator that lists the place. An OSM record earns a spot in the count. It does not earn a written page with coordinates on it. Records that fail the check stay in the data and out of the pages.

And an absent amenity in our data means nobody recorded one, not that the place lacks it. A blank is a blank, never a no.

You can see the whole layer on the Denmark map.

Where there is enough data to write a region page

All five Danish regions have enough named free places to be worth a page. Counts are named places from OpenStreetMap at launch:

RegionNamed free places
Central Denmark Region121
Region of Southern Denmark104
North Denmark Region51
Capital Region43
Region Zealand31

Those numbers are small compared with France, and that is partly a mapping artifact and partly real: Denmark is a small country that has put its free overnight capacity into a state system rather than into thousands of municipal aires. Reading our count as the total supply of free places in Denmark would understate it badly.

What our data covers, and what it does not

Our US layer comes from the Recreation Information Database (RIDB), the open data behind Recreation.gov, published under CC-BY-4.0 and credited as “Data Source: Recreation.gov.” We filter it to free sites.

Here is the honest state of it:

  • 60 records total, across 11 states.
  • Of those, only 15 are actual campgrounds. The other 45 are facilities: trailheads, staging areas, wilderness area polygons. Places on public land, not places you sleep.
  • The states represented are Colorado, Arizona, Oregon, Wyoming, Utah, California, Montana, Alaska, Nevada, Washington and Georgia. Ten of those 11 are western. Georgia is the single eastern entry, with one record.
  • The BLM layer described in our planning notes is not yet in the pipeline. When people talk about American free camping, they mostly mean BLM land, and we do not have it yet.

So: there is no US coverage yet, and the United States map is empty on purpose. The map only carries places with a written page behind it, and none of these 60 has one.

That is not a formality we could waive. Checking the 60 found that only two of them have a public page on recreation.gov at all: the database is the government’s own record, but the consumer-facing page exists only for sites you can reserve, and a free first-come site usually is not one. The federal API answers everyone else with an authentication error, so the link we would have cited is a link no reader can open. Publishing 60 places behind 58 dead links is the mistake we already made once in Denmark and had to undo.

Writing these up therefore means going to the managing BLM or Forest Service office one site at a time. The database gives us a real starting point for that, including a description, directions, the managing agency and a phone number, so it is work rather than a dead end. It is simply not done.

Until it is, treat this country as orientation only. For a specific site, the district office that manages the land is the authority.