USDA Forest Service

Pacific Northwest Research Station

Pacific Northwest Research Station
333 SW First Avenue
Portland, OR 97204

(503) 808-2100

US Forest Service
Pacific Northwest Research Station logo.

Land and Ecosystem Management


Siuslaw Thinning and Underplanting for Diversity Study: Overview and Summary of Early Findings

Study Overview


Problem Statement

Since the 1960s, harvest activities in the Oregon Coast Range have resulted in conversion of hundreds of thousands of acres of natural forest to young Douglas-fir plantations. The consequences of timber-oriented plantation management for forest health and long-term ecosystem productivity have been of concern for the past two decades. In particular, concerns exist over the harvest of old-growth forest and replacement of old-growth with structurally less complex even-aged plantations. Many consider such conversions a threat to habitat quality and biological diversity. The 1991 listing of the northern spotted owl as an endangered species and the subsequent 1993 adoption of the Northwest Forest Plan restricting harvest of larger trees from Pacific Northwest coastal forests. Research and development was initiated on alternative, ecologically-based silvicultural systems for enhancing structural development in young stands.

Forest ecologists and wildlife biologists have begun to recognize and understand the relationships of stand structure and landscape patterns to forest ecosystem functions (Swanson and Franklin 1992). Although specific stand structure needs are still unknown for many species, structural features resembling old growth have been suggested as necessary to a wide variety of plant and animal taxa. A body of pre-Northwest Forest Plan research suggests that young stands can be manipulated to provide some of old-growth habitat characteristics in a relatively short timeframe (Newton and Cole 1987, O'Hara 1990). Typically, this involves thinning young stands to relatively low densities to stimulate the growth of dominant trees and facilitate development of the understory.

While the concept of thinning to enhance structural diversity has gained relatively wide acceptance, stand dynamics data are lacking to project stand development of thinned stands over time (Harrington 1990) as are broader sets of data verifying that thinning can initiate and sustain increased understory development and associated habitats and ecological processes.
The structural characteristics of forest stands can differ widely among stands of a given age. Stand basal area, density, canopy cover, species composition and crown class distribution can differ vastly among stands established at the same time. Through silvicultural manipulation, nearly all of these stand structure characteristics can be modified with relative rapidity if the stand is established but still at a young age. This situation particularly important in wildlife management and the conservation biology of endangered species because attributes of stand structure determine a stand's suitability as wildlife habitat. Stands of the same stand density may have very different canopy and microsite characteristics.

The Siulslaw Thinning and Underplanting for Diversity Study was undertaken to form the scientific basis needed to demonstrate that stands can be partially harvested and managed to create habitat for wildfire on old-growth or late-successional dependent species. The study will provide basic information on overstory and understory vegetation dynamics and microsite for any landowner who wishes to manage stands with increased structural diversity on long rotations. It is intended to provide forest managers with information to enable them to effectively employ an ecosystem approach to management focused on sustaining processes that restore and maintain diverse, healthy, and productive forest ecosystems.



The primary, long-term objectives of this study are to (1) demonstrate silvicultural methods to increase structural diversity in young, dense, even-age Douglas-fir stands of high site index in the Oregon Coast Range and (2) characterize the effects of structural manipulation on stand development, biodiversity, and productivity. These objectives will be accomplished by evaluating functional relationships among biotic and abiotic stand components as responses to a variety of thinning and underplanting regimes, including:

  • Projected growth and yield of overstory crop trees
  • Individual tree crown and stand canopy dynamics
  • Survival and growth of tree species of various shade tolerances planted in the understory
  • Temporal dynamics of germination, survival and growth of natural regeneration
  • Relative survival and growth performance of natural and planted seedlings of similar age
  • Abundance, species diversity and growth response of herb and shrub layers
  • Dynamics of downed woody material and snag development

To date, the primary study objectives have been addressed through a single, initial thinning entry followed by monitoring over an 8-year period (Phase I).  The initial treatment application has resulted in stands having various levels of structure and understory composition. Within the context of the long-term objectives, it is our purpose in this plan for continued research to evaluate the outcomes of the first entry and to prescribe and implement follow-up density management treatments as deemed necessary to develop or maintain structural complexity. We refer to this proposed second entry and associated monitoring and analysis as Phase II.

In addition to those objectives listed above, the original study plan included an objective relating habitat quality to the abundance and composition of amphibians. However, amphibian monitoring has not been followed through the duration of the study and is not included in the current plan for continued effort. Similarly, monitoring of other faunal taxa has not occurred although it would be encouraged if there were researchers having the interest and resources to do so.

The near-term, Phase II objectives of this study are to:

  1. Measure and evaluate the 14th-year condition of the stands in terms of structure, composition, growth, and abundance of overstory and understory vegetation, and associated microsite.
  2. Develop and implement density management prescriptions to maintain or enhance the development of structural heterogeneity initiated in the first thinnings.
  3. Monitor and evaluate the subsequent response relationships between structural manipulations of the overstory and stand development, biodiversity, and productivity.

Methodology–Phase I Treatment Implementation and Monitoring

The Siuslaw Thinning and Underplanting for Diversity Study was implemented as an operational-scale silviculture experiment at three sites (Cataract, Mapleton Ranger District; Yachats, Waldport Ranger District; and Wildcat, Hebo Ranger District) in the Oregon Coast Range. At each site 30- to 35-year-old Douglas-fir plantations were delineated into four experimental units that ranged in size from 7 to 10 acres.

Three units per site were thinned to residual densities of 120 trees per acre(tpa), 60 tpa, or 30 tpa, and the fourth was left unthinned having approximately 225 tpa. The thinnings, conducted in 1992 (Cataract) and 1993 (Yachats and Wildcat) were from below, removing predominantly trees occupying suppressed or intermediate crown positions within the overstory canopy. Nested within the overstory treatment units were two underplanting experiments: a regeneration trial and a species trial.

The regeneration trial consisted of a comparison of regeneration between underplanted and unplanted 1-acre subplots within each overstory treatment. Douglas-fir and western hemlock seedlings were planted in alternate rows at 15 by 15 foot spacing (194 tpa) in the underplanted subplot.  The abundance and growth of planted and natural regeneration were monitored in both the underplanted and unplanted subplots. The species trial consisted of planting seedlings of six conifer species, Douglas-fir, grand fir, western hemlock, western redcedar, Sitka spruce, and pacific yew, and two broadleaf species, red alder and bigleaf maple.

The conifers were planted as 4-seedling species clusters at 5 by 5 foot spacing (1,742 tpa). A 6 by 13 grid of species clusters provided 52 seedlings of each species for each treatment and site for survival and growth assessment. The deciduous species were planted at 8 by 8 foot spacing (680 tpa) in blocks consisting of 48 seedlings. Each block consisted of six 16-seedling rows, with species alternating between rows for a total of 64 seedlings per treatment and site. The species trial plantings were monitored through 8-years post overstory thinning for survival and growth performance (stem diameter, height, height:diameter ratio).

Overstory structure and stand development were evaluated in the two 1.0-acre subplots per overstory treatment unit. In each subplot, breast-height diameter was monitored for all trees while total height and height to base of live crown were monitored on a subset of 40 trees.

Understory vegetation abundance and composition, natural tree regeneration, and understory light conditions were monitored on a grid of 16 subplots within each 1.0-acre subplot. Nested circular plots were used in quantifying composition and percentage cover by shrub, forb, and fern taxa. Hemispherical photography was used to quantify percentage of visible skylight. Microclimate at 1-meter height above was characterized using point-in-time surveys of air temperature, relative humidity, photosynthetically active radiation, and net solar radiation.

Data were analyzed according to a split-plot treatment structure within a randomized complete block design with sites considered blocks. Repeated measures analysis of variance was used in providing tests of treatment effects as discerned from measurements made pre-treatment and at various intervals (depending on response variable) between 1 and 8 years following overstory thinning.


Results to Date


Stand Development

Prior to the initial Phase I thinning, stand conditions were relatively uniform across treatments with density ranging from 223 to 277 tpa, basal area ranging from 187 to 205 foot 2 acre-1, and relative densities ranging from 53 to 59. The thinning treatments decreased basal areas by 0, 51, 67, and 84 percent for the unthinned, high residual density, moderate residual density, and low residual density treatments, respectively. The resulting relative densities were 53, 27, 16, and 8, respectively (Chan et al. 2006).

In the 8 years following the Phase I treatment implementation, stand densities continued to decline as a result of various mortality agents including a limited amount of windthrow (Chan et al. 2006). Mortality ranged from 3.9 percent in the high residual density treatment to 9.6 percent in the unthinned stands, with the greatest percentage decline occurring within 4 years of treatment.

Regardless of posttreatment mortality, all stands grew following treatment as basal area increase ranged among treatments from 14 percent in unthinned stands to 43 percent in heavily thinned stands (Chan et al. 2006). These increases in basal area corresponded to increases in relative density ranging from 8 percent in unthinned stands to 38 percent in heavily thinned stands, with the greatest percentage of change occurring between the fourtht and eighth years posttreatment. Differences in stand structure among treatments remained visually apparent through 10 years post treatment.


Overstory Treatment
Stand structure 10 years following treatment at the Cataract site; (a) unthined control, (b) 30 tpa, (c) 60 tpa, (d) 120 tpa thining.

Individual tree growth varied significantly following Phase I treatment (Chan et al. 2006). On average, mean tree diameters were nearly 2.6 inches greater in thinned stands relative to unthinned stands after 8 years. Tree diameters in moderately and heavily thinned stands were approximately 1.2 inches greater than those in lightly thinned stands. Live crown ratios of trees in unthinned stands were approximately 13 percent less than those of thinned stands indicating crown recession in the absence of thinning. There were no effects of thinning treatment on mean tree height as trees grew from approximately 82 to 97 foot average height.


Canopy Cover/Understory Light

Postthinning percentage of visible skylight ranged approximately from 2 percent in unthinned stands to 48 percent in heavily thinned stands. Over the subsequent 8 years, percentage of visible skylight in thinned stands decreased approximately 3.5 to 5 percent, mostly after the third year posttreatment. By the eigth year, understory light levels in the lightly thinned stands had diminished to prethinning levels. The relationship between percentage of visible skylight and basal area was nonlinear with light levels decreasing asymptotically with increasing stand basal area. Overall there was a relatively rapid decline in percentage of visible light relative to immediate posttreatment values. In contrast, percentage of visible light in the unthinned stands increased about 2 percent over 8 years, even with increasing basal area as tree crowns continued to recede resulting in thinner canopies and lower leaf area densities.


Tree Regeneration

(a) Species Trial

Through 8-years postplanting, survival of underplanted seedlings was relatively high in thinned stands, exceeding 84 percent for all species except bigleaf maple (64 percent). In contrast, mortality of seedlings underplanted in unthinned stands was 100 percent for all species within four years of planting (Maas-Hebner et al. 2005).

While survival was generally high, species differences in growth were substantial (Maas-Hebner et al. 2005). By year 8 stem diameter and total height were greater for western hemlock than for Sitka spruce, Douglas-fir and grand fir. Western redcedar seedlings were typically smallest in both height and diameter, likely a result of heavy preferential browsing or damage by elk or deer.

Differences in height and diameter among overstory treatment were insignificant. In contrast, the height:diameter ratio (H:D) differed significantly with the ratio being greatest in the high residual density treatment and least in the low residual density treatment. While species differences in H:D were insignificant by year 8 the trends since thinning differed greatly among species with H:D of Douglas-fir, grand fir, and Sitka spruce increasing and that of western hemlock decreasing substantially. Although it is inappropriate to compare H:D values among species, the temporal trend of H:D within species strongly suggests that western hemlock seedlings became increasingly robust while seedlings of the other species became less robust.

When average seedling size, vigor and damage were considered, it was concluded from the species trial that through 8 years post thinning, conifer species preference for underplanting were ranked from highest to lowest as western hemlock, Sitka spruce, Douglas-fir, grand fir, and western redcedar (Maas-Hebner et al. 2005). Fast growth by red alder suggested that it was a good choice of broadleaved species for underplanting thinned stands. In contrast, heavy browse damage and associated slow growth rendered bigleaf maple a poor choice for underplanting.

(b) Regeneration Trial

Growth and development responses of planted Douglas-fir and western hemlock seedlings to the overstory treatments mirrored those observed for these two species in the species trial. Both species survived through 8 years in thinned stands but growth and seedling vigor after 8 years were better for planted western hemlock (Chan et al. 2006).

Natural regeneration was generally absent in the unthinned stands and variable among species and among thinning treatments through 8 years (Chan et al. 2006). Eigth-year natural seedling densities in thinned stands ranged from 11,400 to 26,100 for douglas fur, 6040 to 8903 for western hemlock, and 1,475 to 23,300 for red alder. While planted seedlings of western hemlock ranged from 3.5 to 5.2 meters tall and planted Douglas-fir ranged from 2.0 to 4.0 meters tall, naturally regenerating seedlings of these species averaged less than 0.5 meters in height, regardless of thinning intensity. Naturally regenerating red alder ranged from about 2.2 meters to 4.5 meters in height with mean height increasing with thinning intensity.



Density Trajectory
Observed and projected density development by treatment for the Siuslaw Thinning for Structural Development Study.
Basal Area Trajectory
Observed and projected basal area development by treatment for the Siuslaw Thinning for Structural Development Study.


Relative Density Trajectory
Observed and projected Curtis’ relative density development by treatment for the Siuslaw Thinning for Structural Development Study.


Mean Diameter Trajectory
Observed and projected quadratic mean diameter development by treatment for the Siuslaw Thinning for Structural Development Study.


Understory Vegetation

In unthinned stands, the percentage of cover by shrubs remained unchanged at slightly less than 15 percent over the 8 years of Phase I monitoring, while the percentage of cover by herbaceous species increased from approximately 22 to 53 percent. The increase in herbaceous cover was likely related to the concurrent increase in understory light in the unthinned stands. Among treated stands, both shrub and herbaceous cover was initially decreased following thinning. However within 3 years of treatment, percent cover by shrubs and herbs increased to levels greater than observed in the unthinned stands. By year 8, herbaceous cover in thinned stands was near 70 percent and shrub cover exceeded 40 percent, indicating a positive relationship between increased light and abundance of understory vegetation. Furthermore, the richness of shrub and herbaceous species increased with increasing thinning intensity.

Literature Cited

Chan, S.S.; Larson, D.J.; Emmingham,W.H.; Maas-Hebner,K.G.; Johnston, S.; and Mikowski, D. 2006. Thinning effects of overstory and understory development in young Douglas-fir stands in the Oregon Coast Range, USA. Canadian Journal of Forest Research. 36: 2696-2711.

Harrington, C.; Debell, D.; Raphael, M. ; Aubry, K. ; Carey, A.; Curtis, R. ; Lehmkuhl, J.; Miller, R. 1990. Stand-level information needs related to new perspectives: an analysis prepared by the Olympia Forestry Sciences Laboratory, USDA Forest Service. On file with: USDA Forest Service, Pacific Northwest Research Station, Olympia, WA 98512.

Maas-Hebner, W.H.; Emmingham, K.G. ; Larson, D.J. ; Chan, S.S. 2005. Establishment and growth of native hardwood and conifer seedlings underplanted in thinned Douglas-fir stands. Forest Ecology and Management. 208: 331-345.

Newton, M.; Cole,E.C. 1987. A sustained-yield scheme for old growth Douglas-fir. Western Journal of Applied Forestry 2: 22-25.

O'Hara, K.L. 1990. Twenty-eight years of thinning at several intensities in a high-site Douglas-fir stand in western Washington. Western Journal of Applied Forestry. 5(2): 37-40.

Swanson, F.J.; Franklin, J.F. 1992. New forestry principles from ecosystem analysis of Pacific Northwest forests. Ecological Applications 2: 262–274.

Tucker, G.F., Emmingham,W.H.; Johnston, S.; Chan, S.S.; Minore, D.; Owston, P.; McCreightR.W.; Hayes, J. P.. 1993. Adaptive COPE Study Plan: Commercial Thinning and Uunderplanting to Enhance Structural Diversity of Young Douglas-fir Stands in the Oregon Coast Range. Study Plan on file with the Siuslaw National Forest, the USFS Pacific Northwest Research Station, Corvallis, OR; and the Department of Forest Science, Oregon State University.







pnw > about us > programs & teams > LWM > Landscape and Ecosystem Management


US Forest Service - Pacific Northwest Research Station
Last Modified: Monday,16December2013 at14:18:38CST

USDA logo which links to the department's national site. Forest Service logo which links to the agency's national site.