Forest inventory: multiple data sources

How can a meaningful inventory of forest resources be undertaken cost-effectively across a land area as extensive and generally hard to access as Canada’s Northwest Territories (NWT)?

Gathering accurate and up-to-date inventory information has long been a challenge in many of Canada’s remote regions, yet such information is essential to our understanding of the state of Canada’s forests and to our ability to report on these resources at regional, territorial and national levels.

Estimating forest inventory from many sources: The overview

A project now underway in the NWT is applying a novel approach to this challenge of estimating forest inventory attributes. Using a variety of new methods developed by NRCan-CFS researchers and several partner agencies, information is being collected from multiple data sources:

  • field plots
  • LiDAR (Light Detection and Ranging) units mounted on aircraft and satellites
  • Landsat Thematic Mapper satellite images
  • existing forest inventory, where available

These various sets of data are then combined and used to produce a digitally mapped inventory referred to as the Multi-source Vegetation Inventory (MVI). The MVI extends over a larger area than that covered by the NWT’s existing Forest Vegetation Inventory (FVI).

The MVI project is being undertaken by NRCan-CFS in a partnership collaboration with the NWT Department of Environment and Natural Resources (GNWT). Early stages of the project were also supported by the Government Related Initiatives Program of the Canadian Space Agency, and the Northern Oil and Gas Science Research Initiative.

The MVI project: A closer look

Creating the MVI starts with information collected from four key components, including remote sensing technologies:

Field plots are visited and measured.

Field data are typically collected from individual tree measurements (such as diameter, height and species) and used to estimate height, total and merchantable volume, and aboveground biomass at plot- and stand levels.

For the NWT project, the above attributes were estimated, as was crown closure from measurements taken within each plot. These data were used to calibrate LiDAR models of stand structure. Tree increment cores from dominant and co-dominant trees were also collected for estimating tree and stand age.


Information on stand structure can be derived from airborne and satellite LiDAR data

LiDAR sensors mounted on airborne and satellite platforms emit laser pulses to the Earth’s surface, providing data that can be used to generate detailed structural information about the forest. A combination of samples of airborne LiDAR data over small portions of the NWT are being used with satellite LiDAR data from ICESat. The Geoscience Laser Altimeter System (GLAS) onboard ICESat sends laser pulses that illuminate footprints nominally 70 meters in diameter, and distributed at 170-meter intervals that resulted in a wide network of satellite LiDAR data that spanned across the NWT.

The LiDAR measurements made in the NWT project were used to create models of height, crown closure, volume and aboveground biomass. These models described the relationship between the vertical profile and the range of field-collected stand attribute values.


Land cover data representing various forest and non-forested classes

Land cover is the basis for understanding the extent and type of forest species information across the landscape.

In the case of the NWT work, new land cover information was generated from Landsat (satellite) Thematic Mapper images acquired during 2006 to 2008. The resulting land cover map shows the location of coniferous, deciduous, mixedwood and wetland-treed forested areas and various types of non-forest. (For more information on the forest mapping system, read about the Earth Observation for Sustainable Development of Forests [EOSD] project.). A data sharing arrangement with Ducks Unlimited resulted in ground information that combined with GNWT forest information, was used to generate the land cover map.


Continuous estimation of stand volume over a forested region

Image maps of forest inventory data

For the NWT project, these maps were subsequently generated from a spatial modeling and mapping exercise that scaled the LiDAR-estimated inventory attributes with Landsat Thematic Mapper and other biophysical data.


The satellite-derived data were translated into a format referred to as the Satellite Vegetation Inventory (SVI), to resemble a conventional forest inventory dataset. Over an area of interest, the MVI consists of the FVI and the SVI in locations where FVI data does not exist. The MVI data can be easily viewed and manipulated in a Geographic Information System (GIS) by forest technicians and managers.

Example MVI map over a portion of the NWT.

In the Northwest Territories, the MVI was completed over a pilot study area and is now being extended to much of the southern Taiga Plains Ecozone, encompassing an area about 200 000 km2 in size.

Larger image


Numerous benefits anticipated from MVI

Combining several remote sensing technologies in this way to estimate forest inventory attributes will greatly improve resource assessment and reporting by the NWT’s Department of Environment and Natural Resources as well as by CFS.

The more up-to-date inventory information generated through the MVI method will also be invaluable to the National Forest Carbon Monitoring, Accounting and Reporting System (which calculates greenhouse gas emissions and removals from Canada’s forests) and for use in updating the NWT’s part of the National Forest Inventory.

Contact: Ron Hall