AquiusData

Àrea Metropolitana de Barcelona

About

Snapshot of all non-tourist scheduled public transport within the Àrea Metropolitana de Barcelona (AMB). Only services between stops within AMB’s administrative boundaries are included. AMB includes the core of the Barcelona conurbation (that built on the coastal plain), plus the southern-most part of the Vallès (the valley immediately north of the Catalan coastal hills), but does not include some of the largest towns in the Vallès (notably Martorell and Sabadell). The snapshot period is the week starting 26 November 2018 - the first week of operation for Transports Metropolitans de Barcelona’s (TMB) complete “horizontal and vertical” bus network. The snapshot also contains 2018’s new “AMB Exprés” network of limited stop bus services, plus metro Line 10 as far as Foc.

The dataset is an amalgam of GTFS data from the two large public agencies that manage local bus concessions (AMB and Generalitat de Catalunya), municipal bus/metro operator TMB, national railway operator Renfe, autonomous community railway operator Ferrocarrils de la Generalitat de Catalunya (FGC), and tram operator Tramvia Metropolità SA (TRAM), plus a hand-crafted network for bus operator Bus Nou Barris (a TMB concession that neither TMB, nor TMB’s owner AMB, include within their data).

Among Barcelona’s urban operators, service patterns are remarkably consistent throughout the day: Baixbus, TMB and TUSGSAL (the three largest bus operators) run almost exactly the same number of buses each hour between 08:00 and 20:00. The only significant variations in service are between weekdays and weekends - especially Sunday, which only offers about 40% of the (bus) service level of a weekday. A completely different “Nitbus” bus network operates overnight (roughly between 22:00 and 06:00), which has been separately filtered (defined as routes prefixed “N”). The overlaps between the day-time and night-time networks are few: Most obviously TMB’s metro, which continues to operate after midnight at weekends (journey opportunities which are counted as a part of the regular metro service).

Caution: The Barcelona public transport network rarely changes from week-to-week, and only selectively changes from year-to-year, so long after December 2018 this network snapshot is likely to look similar to some future network. However, such a snapshot should not be used if current accuracy is important. The snapshot should not be used to plan an actual journey. And certainly not without checking the results against current operator/agency publicity.

Vortex Map

The Vortex map summarises the prior snapshot in a fixed coordinate grid. This allows strategic links between geographic areas to be visualised in a consistent manner, regardless of administrative boundaries or nuances of route. Each grid is 0.01 (global) degrees in width, dividing the public transport network into 463 nodes. Such a matrix is crude, but broadly effective. All network nodes (stops, stations) within each grid square are amalgamated into a single point. The (2016) population of each grid square derives from Idescat/ICGC’s Quadtree analysis, which locates the resident population by grid square of 250 metres or less. The service patterns showns are for a typical day, excluding dedicated (Nitbus) night services.

The Vortex map includes the 3 most commonly proposed network changes:

Peak trams and trains per hour per direction have been multiplied by 13 to represent an average daily total. For example a 12 minute headway, or 5 per hour, 10 in each direction, equals 130 daily.

Live Demonstrations

Of Interest

Known Issues

Services are missing: Tourist services have been excluded, such as city tours, Tibidabo bus/funicular/tram, and port cruise terminal buses. Although some of these are legally public transport services, their premium fares are not part of the integrated (ATM) ticketing system, and their offer is not popularly considered part of Barcelona’s public transport system. Any other deficiencies reflect data missing from the GTFS files. Renfe data appears to be missing Rodalies trips to remote locations, such as Puigcerda and Portbou, thus slightly undercounts the service total within the AMB area. Otherwise these deficiencies are most likely to occur in the Vallès, where local bus concessions may neither appear in AMB or Generalitat de Catalunya data.

Sunday shows a greater count of TMB metro train journeys than weekdays: TMB metro operates late night Friday-Saturday and overnight Saturday-Sunday, journeys which are counted along with the core (daytime) service. Consequently Sunday includes 4 or 6 additional hours of operation on a metro network that otherwise varies little. However, these counts should be read with caution, since the metro is (uniquely) defined by average frequency over fixed time periods, not defined as a collection of individual scheduled trips. Differences can emerge between the advertised total and the actual total operated, if, for example, the operator advertises a “minimum guaranteed frequency”, which their actual operations augment.

The vortex map may double (or more) count the same service between pairs of nodes: The underlying route may provide the same link two or more times, merely from different stops within the vortex grid “square”, so is intentionally counted more than once. In contrast, services whose original stop sequence serves the same vortex grid node repeatedly (providing links wholly within one node) are only counted as if serving that node once. The map thus counts only each link made between vortex grid nodes.

License

The Àrea Metropolitana de Barcelona dataset is a creative work of academic curiosity, a limited snapshot of one week in history. The original creator makes no claim of ownership to any data therein, nor should be held responsible for its accuracy. Such can therefore be used as “freely” as its source. All the data sources used to build the AMB dataset may be broadly considered “open”, generally only requiring citation and that information is not presented in a misleading manner, although precise licensing terms vary: