Forest Inventory and Monitoring Research Group at NMBU-MINA

The Forest Inventory and Monitoring Research Group (SkogRover) is in Faculty of Environmental Sciences and Natural Resource Management (MINA) at the Norwegian University of Life Sciences (NMBU). Using remote sensing technology, we develop measurement and analysis methods to increase knowledge about forest resources.

About us

We have been at the forefront of research since the first trials of airborne laser scanning for forest mapping in 1995. Since then, we have greatly expanded our portfolio of sensors, platforms and applications:

  • Sensors: laser and optical (RGB, multispectral and hyperspectral)
  • Platforms: drones, aircraft, satellites and terrestrial
  • Applications:
    • Forest management inventories
    • Precision forestry
    • Remote sensing for ecological applications
    • Global and regional forest mapping and monitoring
    • Short- and long-term monitoring of drivers, changes and effects on forests

Background

Reliable information on forest resources has always been essential for sustainable forest management, and Norway has long been at the forefront of this field. The methods used to obtain this information have changed considerably over time. Today, remote sensing, and airborne laser scanning in particular, is an established part of forest mapping in Norway and many other countries.

Read more

National forest monitoring

Systematic mapping and monitoring of forest resources at the national level began in 1919 with the establishment of the Norwegian National Forest Inventory (NFI). It was the world’s first national forest inventory based on statistical sampling methods. Today it is based on permanent sample plots that are measured every five years. The NFI provides the basis for national forest statistics and for Norway’s reporting of forest carbon under international climate agreements. Remote sensing data are increasingly combined with the field data to produce more precise estimates and maps of forest resources for the whole country.

Forest management inventories

Forest inventories at the property level support planning and decision-making for individual forest owners. These inventories used to rely on aerial photo interpretation and extensive field measurements, which made them costly. In the mid-1990s, we began investigating airborne laser scanning as a data source for forest inventory. From 2002 the method was adopted in operational inventories. Today, laser scanning-based forest inventory is standard practice in Norway and is widely used in the Nordic countries and internationally.

Remote sensing with lidar

A laser sensor measures distance by emitting laser pulses and recording their reflections from the ground and vegetation. Because the position of the sensor is known, each reflection can be located in three dimensions. The result is a point cloud describing the terrain and forest structure, which can be used to estimate attributes such as tree height, volume and biomass, for forest stands or for individual trees.

From pilot studies to nationwide data

Since the first experiments, the availability of remote sensing data has increased dramatically:

  • National laser scanning data: Through the National Detailed Elevation Model (NDH), airborne laser scanning data are now available for all of mainland Norway. Together with the NFI, these data form the basis for nationwide forest resource maps.
  • Advances in laser sensors: The first laser scanning surveys of forests in the 1990s recorded less than one measurement per square metre. Modern airborne sensors deliver many points per square metre, and new technologies such as single-photon lidar and multispectral laser provide both wider coverage and more information per measurement. These high point densities make it possible to map individual trees, tree crowns and understorey vegetation over large areas.
  • Aerial imagery: Regular aerial photography allows three-dimensional point clouds to be generated through image matching, making it possible to update forest data more frequently.
  • Drones: Drones carrying laser sensors and multispectral and hyperspectral cameras provide highly detailed data on individual trees, tree species and forest damage.
  • Satellites: Satellites such as Sentinel-2 and Landsat provide frequent imagery for change monitoring. Satellite laser (GEDI and ICESat-2) and radar satellites such as ESA’s BIOMASS mission make it possible to map forest structure and biomass globally.
  • Machine learning and artificial intelligence: Modern machine learning and AI methods make it easier to extract information from large and complex datasets.
Laser point cloud from an airborne laser scanner.

Looking ahead

Forests are expected to contribute to value creation, climate change mitigation and biodiversity conservation. This increases the need for precise and up-to-date information on the extent, condition and changes of forests in time and space.

At the same time, forest mapping has shifted from one-off snapshots to a continuous stream of new data. Satellites revisit the same area at short intervals. Together with regular aerial photography, repeated laser scanning and drones, this provides an ever-longer and more complex time series. Making full use of these data requires methods that can combine different data sources over time and update information as new data become available.

The combination of multiple sensors and platforms, statistically rigorous estimation techniques and new analytical approaches opens up new opportunities in precision forestry, ecological mapping, greenhouse gas reporting and forest monitoring from local to global scales.