Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Action: Log/remove trees within forests: effects on understory plants Forest Conservation

Key messages

Read our guidance on Key messages before continuing

  • Five of ten studies (including four replicated, randomized, controlled studies) in Bolivia, Canada, India and the USA found that logging trees in forests increased the density and cover of understory plants. Five studies found no effect or mixed effects.
  • Four of seven studies (including one replicated, randomized, controlled study) in Australia, Canada and the USA found that logging trees in forests increased species richness and diversity of understory plants. Three studies found no effect.


Supporting evidence from individual studies


A replicated, randomized, controlled study in 1991-1993 in temperate coniferous forest in Oregon USA (Bates, Miller & Svejcar 2000) found that cutting western juniper Juniperus occidentalis trees increased total biomass and cover of understory perennial plants. The total biomass of understory perennial plants (cut: 329 kg/ha; uncut: 38 kg/ha) and their cover (cut: 4.3-4.8%; uncut: 1.4-1.5%) was higher in cut plots. In 1993, total biomass was sampled at 3 m intervals with 1 m2 quadrats along two 45 m transects. Cover of perennial plants was measured along five 30.5 m line transects in each plot of eight pairs of cut (all juniper trees were cut down in 1991) and uncut 0.4 ha plots.



A replicated, controlled study in 1996-1999 in tropical forest in Bolivia (Fredericksen & Mostacedo 2000) found that selective logging increased ground vegetation cover. Ground vegetation cover was higher in logged (99%) than unlogged plots (81%). Four logged (single tree selection in 1996 and 1997 on a diameter-limit basis) and four unlogged plots (1 × 1 m) were replicated in nine block over a 200 ha area. Data were collected 14 months after treatment.



A replicated, controlled study in 1988-2001 in temperate coniferous forest in British Columbia, Canada (Krzic, Newman & Broersma 2003) found that harvesting had a mixed effect on understory plant cover, but did not affect species richness or diversity. The cover of the most common herbaceous species bluejoint reedgrass Calamagrostis canadensis (harvested: 27.7%-38.9%; unharvested: 9.7%-17.9%) as well as of four more herbaceous species (harvested: 1.3%-5.9%; unharvested: 0.5%-3.7%) was higher in harvested than unharvested plots. In contrast, the cover of Mountain Sweet Cicely Osmorhiza berteroi (harvested: 0.0%; unharvested: 0.2%-0.7%) was higher in unharvested plots. Cover of regenerating trees was higher in harvested plots: trembling aspen Populus tremuloides (harvested: 54.9%-63.9%; unharvested: 3.5%-4.1%) and balsam poplar Populus balsamifera (harvested: 8.6%-12.5%; unharvested: 0.0%-0.1%) However, the cover of the shrub birch-leaved spirea Spiraea betulifolia was higher in unharvested plots (harvested: 0.7%-1.0%; unharvested: 7.7%). Numbers of species/5 m2 (harvested: 38-41; unharvested: 34-39) and plant diversity (Shannon's index harvested: 2.53-2.74; unharvested: 2.78-2.89) were similar between treatments. Data were collected in 2001 in three replicates of two harvested (in 1988-1989) and two unharvested mature aspen plots (5 ha). Species richness and diversity were calculated for 40 subplots of 0.125 m2 in each plot (total of 5 m2).



A replicated, controlled study in 1951-1998 in coniferous montane and subalpine forest in Wyoming, USA (Selmants & Knight 2003) found that harvesting increased species richness but not cover of understory plants. The number of species/forest unit was higher in harvested than control plots in both montane (harvested: 26; unharvested: 19) and subalpine forest units (harvested: 32; unharvested: 12). Total cover of understory plants was similar between treatments in both montane (harvested: 25%; unharvested: 23%) and subalpine units (harvested: 58%; control: 55%). Data were collected in 1997-1998 using 50 frames (50 × 100 cm) at each of 30 harvested (in 1951-1969) and 24 unharvested forest units (< 0.5 ha).



A replicated, before-and-after study in 1987-1993 in wet sclerophyll eucalypt forest in Tasmania, Australia (Wapstra et al. 2003) found that cutting treatments increased plant species richness. All treatments increased species richness/plot: clearcutting (before: 11; after: 15), group-selection (before: 13; after: 18) and partial-logging (before: 15; after: 22). Data were collected before (1987) and after (1995-1996) treatments in 44 group-selection (100 m diameter clearcut gaps) and 103 partial-logging (retaining 25%-50% of stems) plots (5 × 5 m), and in 25 clearcutting plots (20 × 20 m).



A site comparison study in 2000-2001 in tropical moist lowland forest in India (Padmawathe, Qureshi & Rawat 2004) found that selective logging had a mixed effect on the abundance of ferns and other epiphytic plants. Abundance (individuals/25 × 25 m plot) of ferns and non-orchid epiphytes were lower in logged (28 and 33 respectively) than in unlogged plots (121-128 and 170-208 respectively). Abundance of epiphytic orchids was similar between treatments (35 vs 28-44). In 2000-2001, non-orchid epiphytes were monitored in four logged (selective logging 1960-1996) and eight unlogged plots (25× 25 m).



A replicated, randomized, controlled study in 1999-2001 in temperate coniferous forest in Oregon, USA (Bates 2005) found that cutting western juniper Juniperus occidentalis trees increased total herbaceous cover and seed production of perennial grasses. Herbaceous plant cover in cut plots (16%) was higher than in uncut plots (4%). Seed production of perennial grasses was higher in cut plots (42 kg/ha) than in uncut plots (<1 kg/ha), while seed production of Sandberg's bluegrass Poa sandbergii was similar (5 kg/ha) in both treatments. In 2001, herbaceous cover was estimated using 0.2 m2 frames and seed production was estimated using five 9 m2 frames in four pairs of cut (all juniper trees cut down in 1998) and uncut plots (0.45 ha).



A replicated, controlled study in 1992-2001 in boreal forest in Ontario, Canada (Bebber et al. 2005) found that structural retention harvest increased herbaceous vegetation cover. Average herbaceous vegetation cover was 40% in harvested compared to 26% in unharvested plots. Harvesting was carried out in 1992. Residual trees were healthy seed bearers and declining quality trees. Six unharvested control plots and 12 harvested plots, spread over an area of approximately 1,200 ha were monitored during August and September 2001. Plot areas varied from 3 to 104 ha (average 26 ha). Fifty five sample points were placed within control plots and 89 within harvested plots (3–20 points/plot). Herbaceous vegetation was recorded inside a 5 m radius ring around plot centre.



A replicated, randomized, controlled study in 1952-1991 in temperate broadleaf forest in Wisconsin, USA (Kern, Palik & Strong 2006) found no effect of cutting on ground layer plant species richness and diversity. For spring and summer flowering plants, species richness (1-6 species/150 m2 and 1-18 species/1 m2 respectively) and diversity (Shannon's Index 0.57 and 0.71 respectively) were similar between treatments. Six treatments (1ha): diameter-limit cut (5.3 m2/ha residual basal area, applied in 1952); shelterwood cut (9.2 m2/ha residual basal area, applied in 1957); three levels of individual tree selection: light (20.6 m2/ha residual basal area), medium (17.2 m2/ha residual basal area) and heavy (13.8 m2/ha residual basal area), applied in 1952, 1962, 1972 and 1982; and uncut, were randomly replicated in three blocks. In 1991, spring ephemeral species were monitored in five 10 × 15 m plots and summer flowering species in eight 1 m2 plots in each treatment.



A replicated, controlled study in 1968-2002 in temperate mixed wood forest in Alberta, Canada (Kurulok & Macdonald 2007) found that salvage logging increased the number of shrub species in early successional forest. In early successional forest, shrub species richness/100 m2 was higher in logged plots (logged: 11; unlogged: 9). There was no difference between treatments for herbaceous species richness (logged: 22; unlogged: 18) or all plants (logged: 33; unlogged: 27), or understory plant cover (logged: 94%; unlogged: 108%). In mid-successional forest, species richness/100 m2 of shrubs (logged: 20; unlogged: 18), herbs (logged: 26; unlogged: 23) and of all plants (logged: 22; unlogged: 41), and understory plant cover (logged: 88%; unlogged: 99%) were similar between treatments. Data were collected in 2002 in five logged (common operational salvage-logging) and five unlogged forest units. Two logged and two unlogged plots were established in mid-successional forests (burned by wildfire in 1968) and the other six in early successional forests (burned by wildfire in 1999). Understory plant cover was evaluated in two plots (1 m2) and species richness was determined in one plot (100 m2) in each of 13-20 sites within each forest unit.



A replicated, randomized, controlled study in 1993-1996 in boreal forest in Manitoba, Canada (Kembel et al. 2008) found no effect of cutting on plant cover and diversity. Total plant cover (uncut: 89-132%; harvest to stump: 78-107%; full tree removal: 76-103%), and plant species richness/2 m2 plot (uncut: 12-19; harvested to stump: 12-18; full tree removal: 12-20) and diversity (Simpson's index uncut: 3.9-6.0; harvested to stump: 4.4-3.0; full tree removal: 4.5-6.3) were similar between treatments. In 1993, three plots (30 × 100 m) of each uncut, harvested to stump and full tree removal (harvested trees completely removed) treatments were randomly applied in each of six blocks. Plant cover was measured in six subplots (5 × 5 m) within each plot (total of 324 subplots). Species richness and diversity were determined in a 1 × 2 m quadrat in each subplot. Data were collected in 1996.



A before-and-after trial in 1992-2003 in boreal forest in Quebec, Canada (Grandpré et al. 2011) found that conifer cutting increased understory species richness, diversity and cover. Numbers of species/1 m2 plot increased in conifer cut plots (before: 4-9; after: 7-13) and remained similar in uncut plots (before: 5-10; after: 6-12). Species diversity (Shannon's index) increased in conifer cut plots (before: 0.7-1.3; after: 0.9-1.5) and remained similar in uncut plots (before: 0.7-1.3; after: 0.8-1.5). Cover increased in conifer cut plots (before: 70%-80%; after: 100%-170%) and remained similar in uncut plots (before: 90%-100%; after: 90%-120%). In 1992, confer cutting (all conifers cut and removed) and uncut treatments (100 m2) were replicated in three blocks (>625 m2) at each of two sites. Data were collected before (1992) and after treatments (2003) in 5-12 plots (1 m2) in each treatment.


Referenced papers

Please cite as:

Agra H., Schowanek S., Carmel Y., Smith R.K. & Ne’eman G. (2019) Forest Conservation. Pages 331-347 in: W.J. Sutherland, L.V. Dicks, N. Ockendon, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2019. Open Book Publishers, Cambridge, UK.