Forest Investment Account (FIA) - Forest Science Program
FIA Project Y092279

    Early survival and growth of natural regeneration and planted seedlings under seven silvicultural systems on the Coast
Project lead: de Montigny, Louise (Ministry of Forests and Range)
Author: de Montigny, Louise E.
Subject: Forest Investment Account (FIA), British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
Silviculture Treatments for Ecosystem Management in the Sayward (STEMS) is a large-scale, multi-disciplinary experiment that compares forest productivity, economics, and public perception of seven silvicultural regimes replicated at three sites in the Sayward Forest. The seven silvicultural systems and treatment regimes include: Extended Rotation (non-treatment control), Extended Rotation with Commercial Thinning, Uniform Dispersed Retention, Aggregate Retention, Group Selection, Modified Patch Cuts and Clearcut with Reserves. These silvicultural systems and treatments to create diversity in forest structure that results in a variety of canopy layers (vertical structure) and spatial patchiness (horizontal structure) that emulate natural variation in forest structure and promotes healthy ecosystem functioning.

The study is a randomized complete block design having seven treatments and a minimum of three replicates in the Sayward (de Montigny, 2004). There are another three replicates as part of the “Silvicultural Options for Harvesting Douglas-fir Young-Growth Production Forests” in the Capitol Forest in Washington, excluding the aggregate retention treatment (Curtis et al, 2004). The decision to replicate the treatments in the Sayward Forest was based on the similarity of knowledge gaps in the Sayward Forest and the Capitol Forest and because replication in Washington and B.C. results in greater statistical power and ability to extrapolate over a wide geographic area
Stand growth and yield information is based on repeated measurements on a grid of permanent plots, maintained for the life of the experiment. Supplementary short-term studies of harvesting costs and visual impacts are being done in cooperation with other organizations. In each of the replications (STEMS 1, 2 and 3) the seven treatment regimes were randomly assigned to one of seven treatment areas, which is the basic experimental unit. The minimum acceptable area required for each treatment unit varies with treatment, being less for those that produce homogeneous stand conditions (10-ha clearcut) than for those that result in patches or sparse distribution of residual trees (35 ha for the patch cut).
Using large operational units as treatment areas is the only way to understand forest dynamics (Franklin 1989) and has several important advantages over using small research plots: it is easier to generalize management results to the watershed and landscape because the experimental units accurately represent the spatial variation; direct observation of the treatment on the landscape can determine visual acceptance; there is a demonstration that the management treatments are both economical and feasible to implement because, by definition, they are operational. Although a series of smaller, short-term projects may seem more economical, they typically can only study one or two factors, and the process mechanisms are out of context. Also, they are not conducive to a long-term study in which causal links can be determined between a manipulated variable and some measured response (Monserud, 2002). The STEMS project is multi-agency and interdisciplinary; its statistical design will ensure long-term opportunities for research that address land management issues and ecosystem productivity.
This technical report describes the establishment of the first replication of STEMS in 2001 in the Snowden Demonstration Forest. Ongoing studies include:
• Tree growth and stand development, including understory vegetation
• Regeneration and light availability
• Windthrow, mortality, and coarse woody debris recruitment
• Harvesting production and impacts of residual tree damage and soil disturbance (in partnership with the Forest Engineering and Research Institute of Canada (FERIC))
• Visual quality and public response.

The results of the multi-disciplinary studies within the STEMS experiment will be used to improve forest management and policies because results from many studies done on the same sites can be directly interpreted operationally due to the large-scale, replicated experimental design.
STEMS 2 (the second replication of six), located near Elk Bay in the Sayward Forest, was harvested by Interfor in 2004, and planted with Douglas-fir and western redcedar in March, 2005. All seedlings within the regeneration plots were tagged and measured immediately after planting. Surveys of planted and natural regeneration under each of the silvicultural systems will be conducted to determine the early survival, growth, mortality, health and browsing damage over 3 years (four years since planting). As well, a vegetation survey will be done in the first year to provide baseline information that will be used to assess the effects of residual stand structure on vegetation growth and competition in the future. Windthrow surveys will be done annually to assess seedling and natural regeneration growth and mortality associated with fallen trees and increased light availability. The results of the surveys will be summarized and reported on yearly. Hemispherical photos (which should be completed at STEMS 2 by March 2007), will be calibrated for light availability and used to correlate light availability to seedling and natural regeneration survival, growth and health in 2008/09.
Seedling growth data will be of great interest for complex model development because there is little data for the early stages of regeneration and most light/growth studies on the coast have been performed on saplings. Measured yearly increments will be directly compared with simulated increments using least squares regression. Complex response surfaces of real and simulated values may be compared to reveal biases at high or low light or to identify confounding factors. As the dataset develops, the growth response of the conifer species will be compared to data and models tested in former studies.
Related projects:  FSP_Y081279FSP_Y103279

Executive summary (64Kb)
Progress Report (0.8Mb)

Updated August 16, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to