Providing evidence to improve practice

Action: Use mechanical thinning before or after planting Forest Conservation

Key messages

  • Five of six studies (including two replicated, randomized, controlled studies) in Brazil, Canada, Finland, France and the USA found that thinning trees after planting increased survival and size of the planted trees. One study found it decreased their density. One study found that the effects of thinning on the size and survival rate of planted trees vary between species.
  • One replicated study in the USA found that the survival rate of red oak seedlings increased with the size of the thinned area.


Supporting evidence from individual studies


A replicated, controlled study in 1993-1997 in boreal forest in Alberta, Canada (Man & Lieffers 1999) found that canopy cutting decreased the mortality of planted white spruce Picea glauca seedlings. Mortality was higher in uncut than i cut treatments (uncut: 22%: partial-cut: 8%-9%; clearcutting: 13%). Seedling height increase (uncut: 14 cm: partial-cut: 29-31 cm; clearcutting: 24 cm) and root-collar diameter (uncut: 5 mm: partial-cut: 7-8 mm; clearcutting: 8 mm) did not differ between treatments. Data were collected in 1997 in one uncut, two partial-cut (residual basal area of 9-16 m2/ha trembling aspen Populus tremuloides and 4 m2/ha white spruce) and one clearcutplots (150 × 150 m) in each of two blocks. Treatments and seedlings planting were in 1993-1994.



A replicated, controlled study in 2001-2008 in boreal forest in Finland (Den Herder, Kouki & Ruusila 2009) found that thinning had mixed effects on the height and mortality of different tree species. In both burned and unburned sites, the height of silver birch Betula pendula was higher in thinned and clearcut (130-270 cm) than in unthinned plots (40-70 cm). Birch mortality was lower in thinned (5-10%) than in clearcut and control (25-55%). The height of rowan Sorbus aucuparia was similar in all treatments in both burned and unburned plots (40-70 cm). Rowan mortality was higher in clearcut (50%) than in thinned and unthinned plots (5-15%) in unburned, and similar in all treatments in burned sites (5-15%). The height of Eurasian aspen Populus tremula was higher in thinned and clearcut (70-80 cm) than in unthinned (30-35 cm) in both burned and unburned. Aspen mortality was higher in unthinned (30%) than in thinned and clearcut (10-15%) in burned, and similar in all treatments in unburned sites (10-30%). Ten seedlings of each species were planted in 2002-2003 in each of three treatment plots (10 ×15 m): clearcut, thinned (50 m3/ha green-tree retention) and unthinned, replicated in three burned (in 2002) and in three unburned sites (total of 18 plots). Treatments were applied in 2001-2002. Data were collected in 2002-2008.



A replicated, randomized controlled study in 1993-2007 in boreal forest in Ontario, Canada (Man, Rice & MacDonald 2009) found that cutting increased the survival rate and size of planted trees. Survival rate (5-14 years after planting) of white spruce Picea glauca (uncut: 36%; cut: 69-74%) and jack pine Pinus banksiana (uncut: 6%; cut: 39-52%) was lower in uncut than in the three cut treatments. Height (cm) of white spruce  (uncut: 60; partial cut : 180; partial cut and removal: 230; clearcut: 250) and  jack pine (uncut: 70; partial cut: 300; partial cut and removal: 400; clearcut: 450) as well as root-collar diameter (cm) of white spruce  (uncut: 1; partial cut: 3; partial cut and removal: 5; clearcut: 6) and  jack pine (uncut: 1; partial cut: 4; partial cut and removal: 7; clearcut: 9) increased with increasing cutting intensity. In 1993-1994 four treatments: uncut, 50% partial cut, 50% partial cut with removal of residuals after three years, and clearcut were replicated in six blocks (112 × 56 m). Blocks were planted with white spruce and jack pine in 1994. Data were collected in 1998-2007.



A replicated study in 2001-2007 in temperate broadleaf forest in Indiana, USA (Morrissey et al. 2010) found that large gap size increased the survival rate of northern red oak Quercus rubra seedlings compared with medium size gaps, gap size also increased the height and diameter of seedlings planted without containers but not of container-planted seedlings. Survival was higher in large (52-60%) than in medium gap plots (20-41%), but did not differ to small gap plots (33-65%). For bare-root seedling, height (large gaps: 190-210 cm; medium gaps: 125-150 cm; small gaps: 75-100 cm) and diameter (large gaps: 2.0-2.1 cm; medium gaps: 1.6-1.1 cm; small gaps: 0.9-1.0 cm) increased with gap size. Height (190-330 cm) and diameter (2.5-3.2 cm) of container seedlings was similar in all treatments. Four large, four medium and three small gap plots (0.400, 0.100 and 0.024 ha clearcuts, respectively) were established in 2002 and planted with 60, 40 and 20 northern red oak seedlings (both bare-root and container seedlings) respectively. Data weref collected five years after planting.


A replicated, randomized, controlled study in 2007-2010 in temperate coniferous forest in Georgia and North Carolina, USA (Knapp et al. 2011) found that cutting treatments increased the height of planted longleaf pine Pinus palustris seedlings, and decreased their density at one of two sites. At one site, seedling height was lower in uncut (20 cm) than intensively cut (50 cm) and clearcut plots (58 cm) and similar to the last two in intermediate-cut plots (30 cm). The number of new germinants/ha was lower in clearcut (167) than intermediate cut and uncut plots (8,208-10,458) and similar to the other treatments in intensively cut plots (2,319). At a second site, seedling height was lower in intensively cut and uncut plots (30-42 cm) than in clearcut plots (85 cm) and similar to the other treatments in moderate-cut (46 cm). The number of new germinants/ha (32,083-329,167) was similar in all treatments. Monitoring was in May 2010 in four randomly assigned 1 ha treatment plots: uncut; intermediate cut (residual basal area 9 m2); intensive cut (residual basal area 6 m2); clearcut. Treatments were replicated seven times at the first site and three times at the second site. All plots were all planted with longleaf pine seedlings in January 2008. Treatments were applied in 2007. In January-April 2010 prescribed burns were conducted in all plots.



A replicated, controlled study in 2007-2010 in Mediterranean Aleppo pine Pinus halepensis woodland in France (Prévosto et al. 2011) found that shelterwood cutting increased the height, diameter and survival of planted holly oak Quercus ilex and downy oak Q. pubescens seedlings. Seedling height of holly oak (uncut: 10 cm; intermediate cut: 13 cm: intensively cut: 15 cm) and downy oak (uncut: 9 cm; intermediate cut: 8 cm: intensively cut: 9 cm) and stem diameter (uncut: 2.0 and 1.5 mm; intermediate cut: 2.7 and 2.0 mm: intensively cut: 3.3 and 2.4 mm, respectively) differed between all treatments. Number of holly oak seedlings/point was higher in intermediate cut (2.4) than uncut plots (uncut: 2.1) and similar to both in intensively cut plots (2.2). Numbers of downy oak seedlings/point was higher in intermediate cut (2.1) and intensively cut (2.3) than uncut plots (1.3). Data was collected in 2010 in four replicates of uncut, intermediate cut (33% of basal area removed) and intensively cut (66% of basal area removed) treatment plots (25 × 25 m). Plots were established in October and seeded in November 2007 with downy oak and holly oak sowing points of three acorns spaced 1 m apart.


A site comparison study in 1978-1984 in dry tropical forest in Brazil (De-Souza Gomes Guarino & Scariot 2012) found that logging increased the survival of newly planted local tree seedlings. Tree seedling survival was the higher in heavily (64%) than in intermediately logged plots (50%), and the lowest in unlogged plots (41%). Forty seedlings of each of three tree species: Amburana cearensis, Cedrela fissilis, and Sterculia striata were planted in each of three forest fragments (115-212 ha): heavily logged (in 1997), intermediately logged (in 1996) and unlogged. Mortality was determined one year after planting. Seeds were collected at the study site between June and July and were grown in a greenhouse until planted back in December 2002.


Referenced papers

Please cite as:

Agra, H., Schowanek, S., Carmel, Y., Smith, R.K. & Ne’eman, G. (2017) Forest Conservation. Pages 285-328 in:W.J. Sutherland, L.V. Dicks, N. Ockendon & R.K. Smith (eds) What Works in Conservation 2017. Open Book Publishers, Cambridge, UK.