Improving Canada's National Forest Biomass Estimates

Download the PDF version (104KB)

Objective(s)

The goal of this Natural Resources Canada (NRCan) project is to improve Canada's national forest biomass estimates.

Specific project objectives include the following:

• Assign biomass data to every record in Canada's Forest Inventory (CanFI 2001) database.
• Report on Canada's forest biomass resources on a periodic basis in conjunction with the National Forest Inventory (NFI) reporting.
• Report on the application of the methodology to the new plot-based NFI as well as provincial and industrial inventories.

Description

For information on Canada's
National Forest Inventory (CanFI)
database, please visit:
http://nfi.nfis.org/

There is significant demand for current information on Canada's forest biomass resources. A forest biomass inventory is necessary for several activities including estimating biomass availability for bioenergy generation, modeling carbon budgets, and determining the contribution of Canada's forests to the global carbon cycle. Canada's Forest Inventory 1981 (CanFI 1981) was reworked into a forest biomass inventory in 1984 using tree biomass and volume equations and volume to biomass conversion factors. Canada's Forest Inventory has been updated to 2001 (CanFI 2001). This provided the opportunity to replace the 1984 biomass database with an updated version based on the inventory information in CanFI 2001, as well as to address some of the shortcomings identified with the 1984 biomass inventory.

Canada is also in the process of implementing a new plot-based National Forest Inventory (NFI) design consisting of permanent observational units located on a national grid rather than a periodic compilation of information across the country. The new NFI is designed to assess and monitor the extent, state, and sustainable development of Canada's forests in a timely and accurate manner, establishing a baseline of where the forests are, and how they are changing over time. Both volume and biomass will be included in the baseline report.

Given the demand for biomass information from two separate national inventory designs, the challenge was to design a methodology that could be applied to CanFI, NFI, and provincial/territorial and industrial inventories.

The project consists of:

• Assigning biomass values to every record in the CanFI 2001 database;
• Reporting on Canada's forest biomass resources;
• Revising the methodology for application to the new NFI as well as provincial and industrial inventories;
• Technology transfer through technical notes; and
• Publication of the Canadian biomass equations.

The project is supported by provincial, territorial, and federal agencies through the provision of basic tree data contained in temporary and permanent plot data, models (single tree volume and biomass, growth and yield equations, and model parameters), by providing guidance and advice, and through the review of technical reports and findings.

The CanFI 2001 database has been expanded to include biomass estimates (total and by tree component). The report on Canada's forest biomass resources is included in "Canada's Forest Inventory 2001" by Katia Power and Mark Gillis. In addition, a scientific article was produced and published in the Canadian Journal of Forest Research, entitled "Canadian national tree aboveground biomass equations", by M.-C. Lamber, C.-H. Ung, and F. Raulier, 2005. Finally, the biomass modeling methods are reported in "Model-based, volume-to-biomass conversion for forested and vegetated land in Canada" by P. Boudewyn, X. Song, S. Magnussen, and M.D. Gillis.

The CanFI database (including biomass) provides information on Canada's forests to support policy and trade decisions, science initiatives, and to assist in meeting regional, national, and international inquiries. Data models and information products have been incorporated into the National Forest Carbon Monitoring, Accounting, and Reporting System.

The focus of this project has now shifted to the new plot-based NFI. A number of revisions to the methodology are required. Methodology must consider that biomass estimates are required for both photo and ground plots, and measured and non-measured (modelled) plots. Methodology must also consider that photo plot data may be from the interpretation of aerial photography or the classification of satellite imagery.

Canadian single tree biomass equations developed as part of this project will be used to generate biomass estimates for the ground plots in most provinces (except BC and YT). Although the basic tree data have been recovered to allow national single tree biomass equations to be constructed for selected tree species in BC and YT, equations from previous studies will be used for ground plots in BC. The development of new equations for BC and YT will be included in next year's work plan. To estimate the biomass of small trees, a ratio-of-means estimator has been developed from existing sample plot data to adjust estimates provided by biomass equations for large trees.

For the photo plots where data is derived from the interpretation of aerial photographs, models developed for CanFI 2001 are being modified to use total volume as a predictor (rather than merchantable volume used in the CanFI 2001 estimation). For photo plots where the data is derived from the classification of satellite image data, procedures to assign volume and biomass to individual pixels have been developed and tested on a small scale study, and are now being fully implemented.

Finally, methods are required to predict volume and biomass for plots that are not measured as part of the annual re-measurement program. An analysis of existing growth and yield projection system was undertaken to determine gaps; and a set of preliminary 'one measurement-based' models were developed to address these gaps. Work continues on the development of these models.

Significant Outcomes

The expanded and updated version of the CanFI 2001 database incorporates significant improvements to individual tree biomass equations (adoption of new national equations) and to the model development process (model fitting and selection), which result in more robust and accurate estimates of forest biomass in Canada. This database will be used to develop all summaries and reports on Canada's biomass resources. Many requests from federal government clients, academia, and Environmental Non-Government Organizations (ENGOs) for relevant information on volume and biomass in Canada have also been made.

Work continues on the development of methodology to allow reporting on Canada's forest biomass resources on a periodic basis in conjunction with NFI reporting. The initial work on small tree biomass has provided an insight and a means to quantify errors in estimating small tree biomass particularly for biomass in northern regions. Work on calibrating remote-sensing models has also provided a method to expand land cover products based on remote-sensing classifications to include both volume and biomass estimates. These land cover products will provide photo plot data for northern Canada as well as auxiliary datasets to help improve the estimates derived from the sample based NFI. Finally the growth and yield work will provide tools to predict volume and biomass for plots that are not measured as part of the annual NFI re-measurement program. These tools will also be of value to provincial/territorial inventory programs.