Published on 8 January 2024

LiDAR data acquisition

In 2024 swisstopo has launched another LiDAR survey of Switzerland and the Principality of Liechtenstein. Airborne laser scanning is one of the most effective methods for measuring territory in 3D. It provides valuable, high-quality data in terms of both detail and accuracy. As the basis for digital terrain models, these data also contribute to other products and are used by many applications.

The image shows a classified LiDAR point cloud around the castle of Spiez.

 

Acquisition campaign to be repeated

swisstopo has been using LiDAR technology since the 2000s with the production of the cadastral survey digital terrain model(DTM-AV) – the predecessor to swissALTI^3D. Since the laser penetrates into the forest, it is possible to produce a homogeneous and high-quality digital terrain model for the entire territory. In 2024, swisstopo launched a new acquisition campaign to obtain LiDAR data for the whole of Switzerland and Liechtenstein, spread over a six-year period (see below). The annual figures correspond to the year of the aerial survey. On average, the data are available about 12 months after acquisition.This campaign will be concluded with the publication of the data by the end of 2030.

Planned acquisition stages of swissSURFACE3D. Approximately 12 months after the acquisition, the data will be publicly available

LiDAR data form a classified point cloud. It is used in-house to update terrain and landscape models and is also available in our online shop with the product swissSURFACE^3D.

The current availability of LiDAR data is given by the layer “Division swissSURFACE^3D” in the federal geoportal map.geo.admin.ch.

Characteristics of LiDAR data

• No foliage or snow during acquisition to facilitate ground information and production of a digital terrain model
• Density of points: minimum 10 pulses per m²
• Classification: unclassified, ground, vegetation, building roofs, water, high-voltage and traffic lines, structural pylons/towers, bridge deck, docks and viaducts, building facades, bridge piers and viaduct stay cables
• Synthetic points on water bodies and under bridges (class water and ground, flagged as synthetic)
• Normalised intensity values
• Planimetric accuracy: 20 cm (1 sigma)
• Altimetric accuracy: 10 cm (1 sigma)
   

Cantonal LiDAR data

In addition to its own planned acquisitions, swisstopo sometimes purchases cantonal LiDAR data if they meet its criteria in terms of quality and acquisition period. These data are then also incorporated into the products distributed by swisstopo.

• Learn more about LiDAR technology

  LiDAR, an acronym for Light Detection And Ranging, comprises both a transmitter and a light source receiver that can measure distances. The wavelengths used are in the invisible spectrum and are of no consequence to the human eye. The laser pulses emitted are reflected by the environment and are captured again by the system. The distance is calculated based on the time of flight taken by the pulses to make the round trip.

  

  Illustration of a survey by airborne LiDAR

  The laser scanner emits hundreds of thousands of laser pulses per second, creating a detailed 3D map of the environment. The pulses can penetrate vegetation and even haze under optimal conditions.

  With a density of several points per square meter, an accuracy of 10 cm and high acquisition speed, airborne LiDAR is one of the best methods to determine the topography of our environment. This data is used, in particular, for the following applications:

  • Automatic 3D modelling of buildings and other landscape objects.
  • Study of the forest: height, width and inclination of trees, map of cuts, calculation of biomass, planning of cuts, digital canopy models.
  • 3D modelling of ground-based elements under vegetation cover such as forest roads or watercourses.
  • Monitoring of overhead lines (cables, high-voltage power lines).
  • Creation of digital terrain models.
  • Creation of digital surface models.
  • Visibility analysis.
     

  From raw points to finished products

  The volume of measurements acquired by LiDAR requires several processes before it can be used for commercial applications.

  Direct georeferencing

  Simply knowing the distance between the measuring device and the ground is insufficient for topographical surveys. All measured points must be referenced in a terrestrial coordinate system. The position and altitude of the aircraft are recorded using GNSS, roll, pitch, and yaw angles via an Inertial Measurement Unit (IMU). By synchronizing GNSS and IMU data and subsequently calculating the exact flight trajectory, the sensor's position and orientation in 3D space are precisely defined. Distortions and aircraft movements are corrected, and the time delay enables the calculation of 3D coordinates for all measured points. Finally, flight strip adjustments eliminate possible height and position differences between individual flight strips before the data is transformed into the Swiss coordinate and elevation reference system.

  Classification

  Known as a “point cloud”, the mass of points obtained from LiDAR acquisition describes all the objects that have returned a laser signal and is extremely complex to interpret. To make this information easier to understand and useful, each point is assigned to a predefined class that provides information on the nature of the latter. There are as many classes as there are applications. They depend in fact on users’ requirements. However, most projects contain at least the following classes: unclassified points, ground, vegetation, buildings and surface waters.

  

  Cross-section of a LiDAR pointcloud coloured by classes

  Classified point cloud

  The classified point cloud is already a finished product. More difficult to handle than some more simplified products (see below), it nevertheless gives more freedom and is suitable for more applications. swisstopo sells a classified point cloud with the product swissSURFACE^3D.

  The most widely used format for this data is LAS. Developed by the American Association of Photogrammetry and Remote Sensing (ASPRS) in open format, it is very effective for managing large data file sizes and retains information specific to LiDAR points.

  Digital terrain and surface model

  A classified point cloud has a lot of information but is still complex to use. Two simplified products are then often produced with lidar data: the digital terrain model (DTM) and the digital surface model (DSM). In both cases, the information is represented homogeneously in space. The irregular measurements of a point cloud are interpolated on a regularly spaced grid. This operation makes it possible to use a raster format (image) and facilitates processing. Nevertheless, the density of the data is often reduced and the classification lost. 

  The DTM represents the bare ground, only the “ground” points of the cloud are used. swisstopo distributes such a model with the swissALTI^3D product. The DSM uses all the classified points and thus models the surface of the territory with the presence of vegetation and buildings.

  

  Example of a DSM (left) and a DTM (right)