Forest Investment Account (FIA) - Forest Science Program
FIA Project Y091064

    Effects of variable retention on planted and natural regeneration in Coastal BC
Project lead: Smith, N.J. (Western Forest Products Inc)
Contributing Authors: Smith, Nick J.; Iles, Kim; Raynor, Kurt
Subject: Forest Investment Account (FIA), British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
This project is a continuation of the two year Y082001. In that project critical long-term data from a set of 10 large area variable retention experiments were collected. In addition statistical and practical fundamentals of ‘sector sampling’ were developed resulting in the publication of two articles in Forest Science (Iles and Smith, 2006, Smith, et al., 2008). In this next phase we plan to collect critical data from additional established sites and continue the development of sector sampling methodology and statistical properties.

This study is designed to examine the effects of retained forest edge on the growth of regenerating trees in Coastal BC. Seven large scale (~100 ha) and three smaller scale (~40ha) experimental areas were recently established (2001+) across Western Forest Products landbase (Beese et al. 2005). The intent is to examine the effects of different amounts and patterns of variable retention on the response of retained and planted trees and natural regeneration. This includes measuring establishment, survival and growth over the long-term (20 years +). A variety of other indicators are tracked in complementary Adaptive Management and Monitoring projects (Beese et al. 2001) such as plant and animal populations (Beese et al., 2005). The experimental sites are established on homogenous areas, with clearcut and uncut controls and up to 3 treatments comprised of different types and levels of variable retention, all randomly allocated. Note that each site examines only one of these treatment types. Treatment types are group retention, group size, dispersed retention, group removal and mixed retention. Within each site the treatments are varied at three levels, for instance: 5%, 10% and 30% dispersed retention. Each site permanently monitors 3000- 5000 planted trees, 3000+ tagged natural regeneration trees and 500-1500 retained trees measured on a 1,3,5,8 and 10 year cycle (+/- 1 year).

Within these experimental areas ‘sector plots’ have been established to examine the effects of forest edge on growth (Iles and Smith, 2006, Smith et al., 2008). The sector plots are designed to sample growth response gradients across variable retention edges; planted, natural and retained trees are spatially located within the sector plots. The sector plots are randomly placed and oriented but sample N,S,E and W facing edges from a central pivot-point outward to the extent of the sampled strata. In dispersed experiments, clearcuts and uncut areas 0.1ha ‘sector’ plots with a central angle of 36 degrees are established randomly. For group retention or group removal treatments four 9 degree sectors tied to a central 'pivot-point' are established with random angles in a minimum of 3 patches per treatment. The sector plots overcome the edge effect biases entailed by establishing traditional fixed area or prism plots( Iles, 2003 p614).

An analysis will examine natural and planted seedling and retained tree responses at the two sites including previously collected data. Results to date show that seedling and natural regeneration growth peaks at a distance (generally 15-20m) from edges, patches or individual trees. A light model run alone does not fully explain this response but it is hypothesized that it may be additionally linked to moisture or other below ground effects. Data collected using separate funding include light (direct and diffuse), soil moisture and soil temperature. These variables and distance from edge will be used to examine the responses of the trees and seedlings measured as part of this project. Note that Y081210 is a complementary project aimed at measuring additional experimental and edge sites and includes a more thorough and comprehensive analysis of growth responses across all sites measured to date. (See Experimental Design).

Forest management issues addressed: effects of variable retention on the survival and growth of the next crop across a wide range of sites and conditions. An analysis and modelling framework that integrates numerous components including retained trees, natural regeneration and planted trees. We expect our results to have some influence on timber supply sensitivity analyses and silvicultural prescriptions at stand edges. The development of a new unbiased sampling system, sector sampling was specially designed to sample variable retention.

As part of this project we request assistance in :
1) measuring two large scale (~100ha) group amount studies (random allocation of clearcut, uncut, 10%, 20% and 30% group retention ): HD216 (Queen Charlottes old-growth hemlock/cedar) and GI100 (Goat Island, Powell River second growth Douglas-fir). Both sites are due for 3 year post–planting measurements of planted, natural and retained trees.
2) extending our investigations into the statistical properties of sector-sampling to fixed area rather than variable area plots: amoung other things the fixed areas should make field layout easier.

There have been considerable changes in ownership of the experimental areas recently so the funds will help us keep core assets intact. These sites cover over 1000 ha, with permanently tagged 30,000 planted seedlings and a large number of retained and natural regeneration trees all sampled in a statistically defensible manner. Measurements will help maintain the sites and continue to make them useful for future science research. The size and scope of the experimental areas appears to be unique.

Other relevant and complementary similar experimental studies on the Coast include: STEMS (Silviculture Treatments for Ecosystem Management) at two sites on Vancouver Island ( ). This study examines management types such as aggregated retention, dispersed retention, group selection and patch cuts at the same site rather than amounts within a given treatment as are examined in this study. This is a Canadian replicate of a US design implemented at 3 sites in Washington (Curtis et. al, 2004). The MASS (Montane Alternative Silvicultural Systems) is a complementary experiment that examines two levels of dispersed retention and also patch cuts but at only one site (Beese and Arnott 1999).

Literature cited. Beese, W.J. and J.T Arnott. 1999. Montane Alternative Silvicultural Systems (MASS): Establishing and managing a multi-disciplinary, multi-partner research site. Forestry Chronicle 75(3):413-416. Beese, W.J., B.G. Dunsworth and J. Perry. 2001. The Forest Project: three-year review and update. J. Ecoforestry 16(4): 10-17. Beese, W.J., B.G. Dunsworth, and N.J. Smith. 2005. Variable retention adaptive management experiments: testing new approaches for managing British Columbia’s coastal forests. Paper In: Eds., C. E. Peterson and D.A. Maguire. Balancing Ecosystem Values: innovative experiments for sustainable forestry, USDA Forest Serv., PNW-GTR-635: 55-4. Curtis, R.O., D. Marshall and D. S. DeBell, 2004. Silvicultural options for young-growth Douglas-fir forests: the Capitol Forest Study-establishment and first results. Iles, K. 2003. A Sampler of Inventory Topics. K. Iles and Assoc, Canada. Iles, K. and N.J. Smith. 2006. A new type of plot that is particularly useful for sampling small clusters of objects. For. Sci., 52 (2):148-157. Smith, N.J., K. Iles and K. Raynor 2008. Some statistical properties of sector sampling. For. Sci.,in press Feb 2008.
Related projects:  FSP_Y082001FSP_Y102064FSP_Y113064


Executive summary (0.5Mb)
Presentation - Sector sampling (3.2Mb)
deliverable 6 (11Kb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated April 29, 2011 

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

Please direct questions or comments regarding publications to