Skip to content

Building data origins

TESSA populates a district with buildings from one of several sources, depending on what is available for the study area. This page describes what each source provides, its known limitations, and how TESSA fills gaps when data is incomplete.

Swiss national building data (RegBL + TLM3D)

For Switzerland, TESSA draws on two national open datasets:

  • RegBL (Eidgenössisches Gebäude- und Wohnungsregister — Swiss Federal Building and Housing Register) — the primary source of building identifiers, addresses, construction periods, building types, number of floors, and heating systems. RegBL data is collected by Swiss cantons, and completeness varies.
  • Swisstopo swissTLM3D — a 3D topographic model that provides building footprint geometries and, where available, height and floor counts. Missing ERA values are filled using TLM3D footprints and floor counts.

Known limitations of RegBL

  • Only residential buildings are required by law to be recorded in detail. Service-sector buildings are included in most cantons but may have incomplete attributes.
  • In some cases only the residential portion of a mixed-use building was recorded as the building floor area, leading to underestimated ERA for the whole structure.
  • Data harmonisation between cantons can produce inconsistencies in attribute coding.

TESSA applies heuristics to complete and correct RegBL data using TLM3D. Floor areas are calculated from residence areas and building footprints with number of floors; missing values are filled from the TLM3D geometry.

Energy Reference Area (ERA)

ERA is the floor area of the building that is thermally conditioned — it excludes unheated spaces such as underground car parks, stairwells, and uninsulated attics. ERA is the single most important building attribute for demand calculation.

For residential buildings, statistical models derived from Swiss SIA standards relate total area to heated area (applying an a_s coefficient). For non-residential (service-sector) buildings, ERA is taken as the total floor area since the entire area is typically conditioned.

An open-access summary of the building stock model is available online.

OpenStreetMap (OSM) data

For study areas outside Switzerland, or to supplement Swiss data, TESSA can extract building footprints from OpenStreetMap. OSM provides:

  • Building footprint geometry
  • Number of floors (building:levels tag where available)
  • Building type (building tag)

OSM coverage is uneven — dense urban areas are generally well-mapped, but building attributes (floors, type) may be missing. Where floor count is absent, TESSA applies default assumptions based on building type.

ERA is derived from the footprint area and floor count using the same approach as for Swiss data. Demand values are then computed using the archetype model.

User-uploaded building tables

You can supply your own building data by uploading a CSV table. This path is used when:

  • You have measured energy data (EPC labels, metered consumption) that should override the statistical model.
  • You are working outside Switzerland with a bespoke building dataset.
  • You need to correct specific buildings where the automated data is known to be wrong.

The uploaded table can provide ERA, demand values, building type, construction period, and equipment, in whole or in part. Columns that are not supplied fall back to the archetype model. See Building data template for the expected column names and units.

When a value comes from the user upload, the corresponding origin field (era_origin, qh_origin, etc.) is set to input ERA or reflects the user-supplied source, distinguishing it from model-derived values in the export.

Comparing sources

Source Geography ERA quality Demand quality Notes
RegBL + TLM3D Switzerland Good for residential; variable for services Archetype-based Best available for Swiss studies
OSM Global Variable; floor count often missing Archetype-based Use when RegBL unavailable
User upload Any As good as your data Measured or archetype Highest confidence when measured data is provided

Where to go next