Providing evidence to improve practice

Action: Use prescribed fire: effects on mature trees Forest Conservation

Key messages

  • Four of eight studies (including two replicated, randomized, controlled studies) from the USA found that prescribed fire decreased tree cover, density and diversity. One study found it increased tree cover and three found no effect or mixed effects of prescribed fire on cover and density of trees.
  • Seven studies from the USA (including one replicated, randomized, controlled study) found that prescribed fire increased tree mortality.
  • One of three studies from the USA (including one replicated, controlled study) found that prescribed fire increased tree size while two found no effect of prescribed fire on tree size.

 

Supporting evidence from individual studies

1 

A controlled study in 1991-1997 in temperate coniferous forest in Louisiana, USA (Haywood et al. 1998) found no effect of prescribed burning on longleaf pine Pinus palustris growth and yield. Average diameter at breast height (30-31 cm), total height (23 m in both) and basal area (23-24 m2/ha) of longleaf pine were similar between treatments. Data were collected in four replicates of 0.16 ha burned (prescribed burned in March 1991, February 1994 and March 1997) and unburned control treatment plots. Longleaf pine were sampled in February 1996 in one 0.09 ha subplot within each plot.

 

2 

A before-and-after trial in 1994-1995 in temperate mixed forest in North Carolina, USA (Elliott et al. 1999) found that prescribed fire decreased the density and diversity of canopy trees in one of three locations. In one site located on top of the ridge density of trees >5 cm diameter at breast height decreased after burning (before: 1,545; after: 913/ha) as well as their diversity (Shannon's H before: 1.92; after: 1.73). In two other sites located in the middle of the slope and next to the stream tree density remained similar (before: 1,167-1,448; after: 1,117-1,365/ha) as well as their diversity (Shannon's indexbefore: 1.85-2.25; after: 1.81-2.26). Data were collected before (1994) and after prescribed-burning (1995) in six plots (15 × 15 m) at one location (top of the ridge) and in three similar plots at two other locations (middle of the slope and close to the stream) and in additional 20 plots (10 × 10 m) after burning.

 

3 

A replicated, controlled study in 1984-1995 in temperate broadleaf woodland in Minnesota, USA (Peterson & Reich 2001) found that prescribed fires increased mortality of northern pin oak Quercus ellipsoidalis but not of bur oak Q. macrocarpa. Mortality of northern pin oak was higher in burned (50%, 560 trees) than unburned plots (27%, 293 trees), while mortality rate of bur oak was similar in burned (8%, 120 trees) and unburned plots (17%, 40 trees). Data were collected in 1995 in 11 burned (4-26 prescribed fires, 1984-1995) and eight unburned plots (0.38 ha).

 

4 

A replicated, randomized, controlled study in 1997-2001 in chaparral in Texas, USA (Ruthven et al. 2003) found that prescribed fire decreased woody plant cover and species diversity but not species richness. Woody plant cover was higher in unburned (44%) than winter (22%) and winter summer burn plots (26%). Diversity was higher in unburned (Shannon's index unburned: 2.55) than winter and summer burn plots (winter burn: 2.33; winter and summer burn: 2.13). Numbers of species was similar among treatments (unburned: 18; winter burn: 16; winter and summer burn: 13/2 ha plot). Data were collected in 2001 in five replicates of: unburned control, winter burn (winter 1997-1998 and 1999-2000) and winter and summer burn (burned in winter 1997-1998 and summer 1999) treatment plots (2 ha).

 

5 

A replicated, controlled study in 2000-2005 in temperate coniferous forest in California, USA (Stephens & Moghaddas 2005) found that prescribed fire decreased tree density, but not basal area, height and canopy cover. The density for all trees >2.5 cm diameter at breast height was lower in burned (441 trees/ha) than unburned plots (1,109/ha), while basal area (burned: 48; unburned: 56 m2/ha), height (burned: 18; unburned: 16 m) and canopy cover (burned: 65; unburned: 75%) were similar between treatments. Data were collected in 2005 in 25 plots (0.04 ha) in each of three burned (prescribed fire in October-November 2002) and three unburned control treatment unit (14-29 ha).

 

6 

A replicated, controlled study in 2001 in temperate broadleaf forest in Tennessee, USA (Jackson et al. 2006) found that prescribed fire increased tree canopy cover but not species diversity. Canopy cover was higher in burned plots (burned: 4%; unburned: <1%), while diversity (Simpson's index) of herbs (burned: 2.6; unburned: 3.1) and woody plants (burned: 3.1-3.3; unburned: 2.3-2.8) was similar between treatments. Data were collected in summer 2001 in two burned (prescribed fire in April 2001) and two control (unburned) treatment plots (0.8 ha) in each of four sites.

 

7 

A replicated, randomized, controlled study in 2001-2004 in temperate coniferous forest in Florida, USA (Varner III et al. 2007) found that prescribed fire increased longleaf pine Pinus palustris mortality when the leaf litter layer was dry on the day of burn, but not when it was moist. Longleaf pine mortality was higher following burns with dry leaf litter than following the other three treatments (dry litter: 16%; moist litter: 4%; wet litter: 2%; unburned: 0%). In 2001-2002, four treatment plots (10-50 ha) were established at each of four sites: dry, moist and wet leaf litter (55%, 85% and 115% litter moisture content, % of dry mass, on the day of burn) and an unburned control. Data were collected two years after treatments.

 

8 

A paired-sites study in 2003-2006 in temperate coniferous forest in Arizona and New-Mexico, USA (Breece et al. 2008) found that prescribed burns increased tree mortality. Mortality of ponderosa pine Pinus ponderosa increased from 0.6% in unburned to 8.4% in burned plots and mortality of all tree species increased from 0.6% in unburned to 9.6% in burned forest units by the end of the third growing season after burning. Trees with at least a 13 cm diameter at breast height were monitored for three growing seasons after burning (2004–2006). They were monitored in 25-40 circular plots (10 m radius within each burned (prescribed fire between 2003 and 2004) and unburned control forest units in each of four sites.

 

9 

A replicated, controlled study in 2002-2006 in temperate coniferous forest in Washington State, USA (Harrod et al. 2009) found no effect of prescribed fire on tree density, basal area, average diameter and height. Numbers of trees (burned: 525; unburned: 530/ha), tree basal area (burned: 34; unburned: 34 m2/ha), average diameter (burned: 31; unburned: 30 cm) and height (burned: 6; unburned: 5 m) were similar between treatments. Data were collected in 2006 in six plots (20 × 50 m) within each of six burned (prescribed fire in spring 2004) and six unburned treatment units (10 ha).

 

10 

A before-and-after trial in 2003-2006 in temperate coniferous forest in California, USA (Vaillant et al. 2009) found that prescribed fire increased tree height and diameter but did not affect their density and cover. Tree height at the base of the canopy (before: 3.7-4.3 m; after: 6.7-7.3 m) and average diameter (before: 36 cm; after: 38 cm) increased after treatments. However, the number of trees (before: 408-462; after: 351-356/ha) and canopy cover (before: 49-74%; after: 46-74%) remained similar. Data were collected in 42 plots (0.1 ha) before (2003) and after (2006) the prescribed fire in 2003.

 

11 

A randomized, controlled study in 2004-2008 in temperate coniferous forest in California USA (Fettig et al. 2010) found that prescribed burning increased tree mortality. The overall percentage of trees killed was higher following early and late-season burning (16% and 18%) than in unburned plots (1%). For trees between 10-20 and between 31-41 cm diameter at breast height, mortality was higher following late-season burning (65% and 5% respectively) than in unburned plots (1% in both categories) and similar to both following early-season burning (22% and 4% respectively). For trees >51 cm diameter at breast height mortality was higher following early-season burning (4%) than in unburned plots (0%) and similar to both following late-season burning (2%). For trees 21-30 and 41-51 cm diameter at breast height mortality was similar in unburned (1% and 0% respectively), early-season (11% and 3% respectively) and late-season burning plots (8% and 2% respectively). Three treatments: control (unburned); early-season prescribed burn (May 2005); late-season prescribed burn (October 2005) were each randomly assigned to three 4 ha plots. Tree mortality was monitored in 2005-2008.

 

12 

A replicated, controlled study in 2002-2006 in temperate coniferous forest in California, USA (Fettig, Borys & Dabney 2010) found that prescribed burning increased bark-beetle caused tree mortality. Cumulative percentage of trees killed by bark beetles were higher in burned (9%) than in unburned plots (3%). All trees killed by bark beetles were recorded in 2006 in three control (unburned) and three burned (prescribed burning in the autumn) treatment units (10 ha). Prescribed burning was in 2002.

 

13 

A replicated, controlled, randomized study in 2001-2005 in temperate coniferous forest in Montana, USA (Fiedler, Metlen & Dodson 2010) found no effect of prescribed fire on trees density and basal area. Density of trees (burned: 386; control: 400/ha) and tree basal area (burned: 22; control: 25 m2/ha) were similar between treatments. Data were collected in 2005 in ten 0.1 ha plots in each of three replicates of burned (prescribed broadcast burning in spring 2002) and control (unburned) treatments.

 

14 

A replicated, before-and-after study in 2006-2010 in temperate coniferous forest in Washington State, USA (Fonda & Binney 2011) found that prescribed burning decreased the density, but not basal area of trees. The density of all trees (pre-fire: 1,565; post-fire: 655/ha) and of trees <20 cm diameter at breast height (pre-fire: 845; post-fire: 190/ha) was lower three years post-fire. Basal area of trees was similar (pre-fire: 82; post-fire: 60 m2/ha). Data were collected in 2006 (pre-fire) and in 2010 (post-fire) in five 20 x 20 m plots that were burned in 2007.

 

15 

A replicated, controlled study in 2001-2009 in temperate coniferous forest in California, USA (Van Mantgem et al. 2011) found that prescribed burning increased tree mortality in the short-term but not in the long-term. Tree mortality rate was higher in burned plots in the first six years after treatment (unburned: 1.5% in all six years; burned: 47% in the first to 3% in the sixth year), but similar in the following two years (unburned: 1.5%; burned: 1.5-2.0%). Data were collected in 2001-2009 in five burned (prescribed-fire in 2001) and seven unburned control plots (0.9-2.5 ha).

 

16 

A replicated, controlled study in 2001-2003 in temperate coniferous forest in California, USA (Stark et al. 2013) found that prescribed burning increased bark beetle caused mortality of white fir Abies concolor trees but not of sugar pine Pinus lambertiana or ponderosa pine Pinus ponderosa. Mortality of white fir trees 11-25 cm diameter at breast height (burned: 4.6%; unburned: 0.2%) and trees 25-45 cm diameter at breast height (burned: 0.8%; unburned: 0.1%) was higher in burned than unburned plots. Mortality did not differ between treatments for sugar pine 11-25 cm diameter at breast height (burned: 2.9%; unburned: 0.0%) and 25-45 cm diameter at breast height (burned: 4.8%; unburned: 0.0%), and of ponderosa pine 11-25 cm diameter at breast height (burned: 1.8%; unburned: 0.0%) and 25-45 cm diameter at breast height (burned: 0.0%; unburned: 0.0%). For trees >45 cm diameter at breast height, mortality did not differ between treatments for either white fir (burned: 0.3%; unburned: <0.1%), sugar pine (burned: 0.0%; unburned: 0.0%) or ponderosa pine (burned: 0.1%; unburned: 0.0%). Bark beetle caused tree mortality was monitored in 2003 in 20 subplots (0.4 ha) in each of three unburned control and three burned (prescribed fire in November 2002) treatment plots (14-29 ha).

 

Referenced papers

Please cite as:

Agra H., Schowanek S., Carmel Y., Smith R.K. & Ne’eman G. (2018) Forest Conservation. Pages 285-328 in: W.J. Sutherland, L.V. Dicks, N. Ockendon, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2018. Open Book Publishers, Cambridge, UK.