Zasada 1987

Research Project Summary: Forest floor and plant responses to experimental fires in an Alaskan black spruce/feather moss community

This document summarizes information from a research project conducted in the late 1970s and early 1980s in black spruce forest communities near Fairbanks, Alaska. It is intended to:

1. provide concise information about the effects of particular fire treatments on a specific plant community and
2. supplement FEIS reviews of individual species with detailed information on specific treatments and effects in a particular location. The studies summarized in FEIS Research Project Summaries are selected based on their integration of fire effects information with relatively complete descriptions of burned and unburned vegetation, burning conditions, fire weather, and fire behavior.

Common names are used throughout this summary. For a complete list of the common and scientific names of species discussed in this summary and for links to FEIS species reviews, see the Appendix.

Citation for this Research Project Summary:
Fryer, Janet L., compiler. 2013. Research Project Summary: Forest floor and plant responses to experimental fires in an Alaskan black spruce/feather moss community. In: Fire Effects Information System, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Available: [].

Unless otherwise indicated, the information in this Research Project Summary comes from the following papers:

Dyrness, C. T.; Norum, Rodney A. 1983. The effects of experimental fires on black spruce forest floors in interior Alaska. Canadian Journal of Forest Research. 13: 879-893.

Zasada, John C.; Norum, Rodney A.; Teutsch, Christian E.; Densmore, Roseann. 1987. Survival and growth of planted black spruce, alder, aspen and willow after fire on black spruce/feather moss sites in interior Alaska. The Forestry Chronicle. 63(2): 84-88.

Zasada, John C.; Norum, Rodney A.; Van Veldhuizen, Robert M.; Teutsch, Christian E. 1983. Artificial regeneration of trees and tall shrubs in experimentally burned upland black spruce/feather moss stands in Alaska. Canadian Journal of Forest Research. 13: 903-913.

Black spruce/feather moss woodlands and forests typically burn with mixed surface-crown fires (for example, [2,6]). This study demonstrated that feather moss and lichen fuels on the forest floor carry the surface portion of mixed-severity fires in black spruce/feather moss types. Downed woody fuels were too sparse to have much impact on fire spread on the forest floor. The "upper portion of the forest floor readily carried fire across most of every unit" [3]. From there, low branches on black spruce easily carried the fires into tree crowns. These experimental fires created a mosaic similar to that created by the mosaic fires typical of black spruce/feather moss forests of interior Alaska. The authors concluded that "wildfires burning on sites similar to the experimental area and within the ranges of variables documented" in their study should produce similar fire behavior and effects on vegetation and soils [3].

Postfire vegetation growth varied with species and depth of the postfire forest floor. Across 3 years, total annual biomass production was greater on plots that burned at high severity than on plots that burned at low severity (58 g/m² vs. 37 g/m²). Species composition was markedly different on low-severity and high-severity plots. Sprouting plants, with their roots crowns or rhizomes in organic and/or mineral soi layersl, dominated low-severity plots, while species with wind-blown seeds or spores dominated high-severity plots [3].

Artificial regeneration (seeds, cuttings, and seedlings) established best on high-severity plots. Mineral soil proved a better seedbed than feather mosses for 8 artificially sown tree and tall shrub species. For these woody species, no artificial regeneration survived into the 3rd postfire year. Hardwood cuttings were especially sensitive to depth of residual organic matter, only surviving on high-severity plots [7]. Woody species on high-severity plots gained more height than those on plots that burned at lower severities (P<0.05 for survivorship and height variables) [8]. Since this study was not replicated at other locations, the authors emphasized that the positive relationship of fire severity to height may be applicable to only these study plots, and they called for further studies on this relationship [7].

The authors concluded that overall, relatively severe burning apparently increased germination, early postfire establishment, and growth of tree and tall shrub seedlings on this site. Although organic substrates might favor germination, tree and tall shrub seedlings often fail to survive on organic substrates because seedlings can desiccate quickly in summer, resulting in seedling death. However, seedlings in organic soils may survive in summers with ample rain [8].

This study demonstrated that artificial regeneration of black spruce and hardwoods is feasible on burned black spruce/feather moss sites. Hardwood species generally grew faster than black spruce, and they will likely continue to increase site productivity and provide species diversity in this black spruce/feather moss community for years to come.

Viereck and others [6] conducted an earlier study in the Washington Creek Fire Ecology Experimental Area on this black spruce/feather moss community. The study investigated the effects of prescribed fires of varying severities on the forest floor and on the short-term effect of fire on black spruce. See the Research Project Summary of their study for further information.

The study site was in the Washington Creek Fire Ecology Experimental Area, about 54 km north of Fairbanks, Alaska [7,8] (latitude 64.20 longitude -149.49 radius 10 km) [8].

The total area burned was approximately 13 ha. It was divided into 7 burn units of approximately 2 ha each [3]. Elevation at the study site ranges from 480 to 520 m. Slope is 10% to 20%, and the aspect is mostly south to southeast from a main ridge running southwest to northeast. The burned area was on the upper third of the slope. The site is somewhat poorly drained, with soil developed in shallow loess over schist bedrock. Bedrock is generally 75 cm below the surface. Before the fire, the organic forest floor was 20 to 30 cm thick [3,7]. It had 4 layers: upper feather moss (a green layer); lower feather moss (a brown layer); upper duff with dead but discernible plant parts; and lower duff with compact, decayed organic material. Upper and lower duff layers were separated by a thin sheet of fungal mycelia, and the lower duff layer was wet and cold. Mineral soil was a very dark-brown silt loam surface horizon underlain by a yellowish-brown, stony (30%-50% by volume) silt loam. The study site had no permafrost [7].

The prefire community was an open black spruce/Schreber's moss forest. The mature, black spruce tree canopy layer had 2 dominant age classes that were about 85 and 140 years old. Some of the 7 burn units had small groups of quaking aspen and scattered paper birch. The tall shrub layer was sparse (~5% cover), containing scattered American green alder, willows, and prickly rose. The most common species in the low shrub layer were mountain cranberry, bog blueberry, crowberry, and bog Labrador tea; cover was about 30% for this layer. Pinegrass and other herbs were few, but the moss layer was a "thick and vigorous" carpet of feather mosses that included Schreber's moss, splendid feather moss, and juniper haircap moss. Lichens occurred in patches on stony soils; they included reindeer lichens, arctic kidney lichen, and green dog lichen [3].

The study site is classified in the following plant community and probably historically experienced this fire regime:

Table 1. Fire regime information on the vegetation community studied in this Research Project Summary. Fire regime characteristics are based on the LANDFIRE Biophysical Settings data layer [5]. This vegetation model was developed by local experts using literature and expert estimate as documented in the PDF file linked from the BioPhysical Settings Group name listed below.
Vegetation Community Mean interval
Fire severity* Percent of fires
Western North American boreal black spruce wet-mesic slope woodland 76 >Replacement 46
Mixed 54
Surface or low 0
*Fire Severities:
Replacement=Any fire that causes greater than 75% top removal of a vegetation-fuel type, resulting in general replacement of existing vegetation; may or may not cause a lethal effect on the plants.
Mixed=Any fire burning more than 5% of an area that does not qualify as a replacement, surface, or low-severity fire; includes mosaic and other fires that are intermediate in effects.
Surface or low=Any fire that causes less than 25% upper layer replacement and/or removal in a vegetation-fuel class but burns 5% or more of the area [1,4].


Table 2. Fire weather for 7 experimental fires in interior Alaskan black spruce/feather moss-lichen. Data are means [3].
Unit Date of fire Time Temperature
Relative humidity
Wind direction Rate of spread
Flame length
1 7/19/78 14:37-15:42 23 42 2-6 SSW 0.4 0.6
2 8/1/78 14:23-16:20 21 33 8-13 SW 1.1 0.8
3 8/4/78 15:13-17:20 24 30 6-10 SW 1.2 1.1
4 8/8/78 14:50-16:09 21 54 2-3 SW 0.3 0.2
5 7/30/78 16:28-18:30 22 42 5-10 SSW 1.2 0.8
6 8/4/78 17:20-19:30 21 33 5-8 SW no data 0.9
7 7/31/78 15:27-16:28 26 36 5-8 SSW 1.8 0.8

Fuel moisture conditions were variable. Surface woody fuels were sparse. Downed woody fuels loads ranged from 4,484 to 22,364 kg/ha. About half of woody fuels were <3 inches (8 cm) in diameter [3].

Table 3. Fuel conditions and tree densities for 7 experimental units in interior Alaskan black spruce/feather moss-lichen. Data are means [3].
Unit Down, dead woody fuels loading (kg/ha)
Mean fuel moisture contents just before burning (%)
Tree species Density
Upper moss Lower moss Upper duff Lower duff
1 6,702 8.3 13.3 31.8 137.4 134.9 114.3 black spruce 2,240
quaking aspen 379
2 4,484 11.2 13.3 16.0 128.2 110.1 108.3 black spruce 2,406
3 8,198 8.0 10.0 20.6 60.1 98.7 98.6* black spruce 3,000
paper birch 78
quaking aspen 39
4 22,364 21.0 44.5 76.1 84.5 88.0 81.2 black spruce 917
paper birch 24
5 9,220 8.3 13.3 33.9 197.7 187.6 155.4 black spruce 2,811
6 5,562 8.0 10.0 20.6 60.1 98.7 98.6* black spruce 820
7 5,450 8.4 12.8 40.7 113.9 105.6 106.8 black spruce 5,694
*Units 3 & 6 were burned at the same time, and data for their fuel moisture samples were pooled.

The fire was conducted during the growing season.

Season: summer (19 July-8 August 1978)
Severity: mosaic of fire severity classes from unburned to high severity (see Table 4)

Table 4. Fire severity descriptions [6,8]
Severity class Description
Unburned plant parts green & unchanged
Scorched moss & other plants yellow or brown but species identifiable
Low lightly burned, plants charred but original form of mosses & twigs visible
Moderate shallow ash layer present, organic layer partially consumed, parts of woody twigs remain
High ash layer present, organic material in the soil consumed or nearly consumed, no discernible plant parts remain

The objectives of these prescribed, experimental fires were to measure fire behavior, subsequent fire effects on the forest floor and vegetation [3], and postfire artificial and natural regeneration of trees and tall shrubs [7,8] on sites that are representative of upland black spruce/feather moss forests of interior Alaska [3].

Across burned units, the full range of fire severity classes shown in Table 4 was achieved. Almost half of the 13 ha either did not burn, was only scorched, or burned at low severity; the other half burned at moderate to high severity. On average, the fires consumed about half of the prefire forest floor. This mosaic of unburned patches and lightly to severely burned areas was "apparently typical of that encountered for black spruce-feather moss fires in interior Alaska", with patterns similar to those occurring in wildfires [3].

Fire behavior within the units ranged from slowly moving, mixed surface-and-ground fires to rapidly advancing crown fires. The forest floor (largely mosses and their decomposition products) primarily carried the fire; down, dead, woody fuels contributed little to fire behavior or subsequent fire effects. Surface fires determined the rate of spread. The crown fires were passive, with trees torching after surface fires had passed. Crowning flames reached 15 m or higher. Most of the units were completely covered by fire and burned down to glowing combustion within an hour of ignition. Fire severity ranged from unburned, scorched, or low-severity (most of Unit 1) to moderate- (most of Units 4, 5, and 7) and high-severity (most of Units 2, 3, and 6) (see Table 4). Units 2 and 3 burned "vigorously, with rapidly spreading surface fires and rapidly igniting crown fires". The fire could barely propagate on Unit 4, probably due to the high moisture content of surface fuels (see Table 3) [3].

Table 5. Weather conditions and fire behavior during the experimental fires [3]
1 2 3 4 5 6 7
Temperature (°C) 23 21 24 21 22 21 26
Relative humidity (%) 42 33 30 54 42 33 36
Wind (km/h) 2-6 8-13 6-10 2-3 5-10 5-8 5-8
Wind direction SSW SW SW SW SSW SW SSW
Rate of spread (m/min) 0.4 1.1 1.2 0.3 4.0 no data 1.8
Flame length* (m) 0.6 0.8 1.1 0.2 0.7 0.9 0.7
*Of surface fires only.

The fire killed the entire overstory on some units and part of the overstory on others. Smoldering combustion of the feather moss and soil organic layers also occurred on some units [8].

Fire severity and consequent removal of the soil organic layer affected postfire responses of vegetation. After burning, forest floors comprised a small-scale mosaic of unburned, scorched, lightly burned, moderately burned, and heavily burned areas. Forest floor consumption was negatively correlated with the moisture content of lower moss and lower duff layers at the time of burning (P=0.05). From postfire years 1 to 3, total biomass production of vegetation was greatest on units that burned at high severity. In postfire year 3, biomass production averaged 110 g/m² and 175 g/m² on low- and high-severity units, respectively. In postfire year 3, units that burned at mostly low severity were dominated by sprouting species including pinegrass, bog blueberry, and bog Labrador tea. Severely burned units were colonized by species with wind-blown seeds or spores including fireweed, fire moss, and common liverwort. For severely burned plots, most aboveground biomass in postfire year 1 to 3 was mosses and liverworts, and almost none was shrubs or trees. Black spruce seedlings were mostly concentrated in high-severity units, but they were still small in postfire year 3 and had "inconsequential" biomass. Fireweed was mostly restricted to high-severity units in postfire years 1 and 2, reaching peak biomass in postfire year 2. It began colonizing less severely burned units in postfire year 3, but its total biomass was decreasing sharply by then. Neither fire moss nor common liverwort was apparent in the prefire vegetation. Fire moss biomass on high-severity units steadily increased during postfire years 1 to 3, while that of common liverwort remained steady, from 15 to 20 g/m², across 3 postfire years. Fire moss biomass increased from about 8 g/m² in postfire year 1 to 38 g/m² in postfire year 2 and almost 60 g/m² in postfire year 3. Pinegrass apparently began invading severely burned plots in postfire year 3.

Artificial regeneration sources were seeds, unrooted stem cuttings that were stored in a freezer prior to outplanting, and seedlings. In the fall after the prescribed fires and the subsequent spring and summer, plots on Units 1, 2, and 5 were hand-sown with local seed collected from 8 tree and tall shrub species native to the area: paper birch, black spruce, balsam poplar, quaking aspen, American green alder, Alaska willow, Bebb willow, and Scouler willow. Those units were selected for planting because they represented the full range of burning severities [8]. Seeds were sown during the time of natural seed rain for each species: fall—about 2 months after the last fire—for American green alder, black spruce, and paper birch; and summer for balsam poplar, quaking aspen, and the willows. In addition to seed sowing, black spruce seedlings were outplanted. The seedlings were started in a nursery from local seed (3 months), then hardened off outside (3 months) prior to outplanting in September 1978 [7]. Stem cuttings of Alaska willow and balsam popular were collected near the study sites and stored in a freezer until late spring planting. Natural regeneration of trees and tall shrubs was monitored on unplanted plots [8].

Rainfall at time of planting and during the 1st postfire growing season was below normal [7]. Weather data for subsequent growing seasons were not provided.

Establishment and survivorship of trees and tall shrubs were highest on high-severity plots [7,8].

Table 6. Mean cover (SD) and frequency of artificial tree and tall shrub regeneration by fire severity class, postfire year 3 [8]
Burn severity class
Unit 1
Unit 2
Unit 5
Cover (%) Frequency (%) Cover (%) Frequency (%) Cover (%) Frequency (%)
Unburned 0 0 0 0 0 0
Scorched 23 (4) 70 6 (2) 45 27 (3) 98
Low severity 44 (6) 85 17 (4) 53 30 (4) 88
Moderate severity 10 (2) 70 13 (2) 78 11 (2) 85
High severity 15 (4) 40 53 (5) 85 25 (4) 75
Other (mostly dead trees) 8 (1) 85 11 (1) 90 7 (1) 70

The units that burned at mostly moderate and high severity (Units 5 and 2, respectively) had highest establishment of both artificial and natural regeneration for all tree and tall shrub species. Among these species, only seeded Bebb willow occurred on low-severity plots, at a rate of 23 germinants, 5 first-year seedlings, and 0 third-year seedlings [8].

Table 7. Total number of germinants and seedlings on moderate- and high-severity plots [8]
1st-year seedlings
3rd-year seedlings
Moderate severity High severity Moderate severity High severity Moderate severity High severity
Artificial regeneration
American green alder 13 160 0 83 0 65
paper birch 72 875 16 595 6 527
black spruce 12 48 3 27 0 48
quaking aspen 8 215 0 94 0 64
balsam poplar 17 71 1 39 0 39
Alaska willow 1 400 0 278 0 181
Bebb willow 105 144 7 46 0 46
Scouler willow 29 189 0 47 0 42
Natural regeneration
paper birch 0 0 0 0 0 2
black spruce 3 45 0 31 0 293
quaking aspen 0 0 0 0 0 2
willow (Salix spp.) 0 0 0 0 0 13

Over half of artificially seeded germinants (all species) were apparent by 5 June 1979 (postfire year 1). Some black spruce seeds blew in from parent trees onto and established in artificially seeded plots. This natural black spruce regeneration established later than artificial black spruce regeneration, around 26 June 1979. Some summer-seeded willows germinated within 2 weeks of sowing; however, the earliest willow germinants had lowest 1st-year survivorship.

Survival of black spruce seedlings was significantly greater than that of hardwood species (P=0.05). Among hardwoods, no species showed consistently higher survivorship across units [7]. First-year survivorship averaged 46% for American green alder, 60% for paper birch, and 80% for black spruce. Overall, 3rd-year survivorship was highest on high-severity plots. Third-year survivorship averaged 34% for American green alder and 43% for paper birch. Since black spruce establishment on seeded plots was bolstered by natural seed rain, 3rd-year survivorship of artificial regeneration could not be calculated [8].

By postfire year 6, height growth varied among species and units, but black spruce was always shortest and American green alder always tallest. Among tree and tall shrub species, height was least in Unit 1, which had the lowest overall fire severity [7]. For black spruce, American green alder, and paper birch, seedling height was significantly greater on Unit 2—which burned at mostly high severity—than that on units 1 and 5. This was also true for Alaska willow and Bebb willow (P<0.05 for all species), but not Scouler willow [7,8]. For hardwoods, mean height of natural regeneration was always greater than that of planted seedlings, while mean height of planted black spruce seedlings was greater than that of natural seedlings. For hardwood seedlings, best survivorship and growth was on high-severity plots that had little to no residual soil organic matter and were colonized with fire moss, juniper haircap moss, and other mosses that are common in early postfire succession. Poor hardwood seedling growth was positively associated with residual soil organic matter (P=0.05). For example, height of grayleaf willow seedlings averaged 25 cm on low-severity and 135 cm on high-severity plots. The effects of soil organic matter depth were apparent even over small areas, so seedlings on adjacent low- and high-severity microsites showed vast differences in height. Overall survivorship of willow cuttings was poor. Alaska willow and balsam poplar cuttings survived only on high-severity plots [7].

Table 8. Mean height (cm (SD)) of artificial regeneration, postfire years 3 and 6. For spruce, alder, aspen, and grayleaf willow, data for both hand-sown and planted regeneration are given. For planted seedlings, height at planting averaged 10 cm for spruce and 15 cm for alder, aspen, and grayleaf willow [7]. For other species listed, all data are for regeneration from hand-sown seed only [8]. Cells are blank where data were not available.
Unit 1
Unit 2
Unit 5
3rd year 6th year* 3rd year 6th year 3rd year 6th year
American green alder sown: 5.0
sown: 27.3
sown: 0.6
planted: 18.8
planted: 40.4
planted: 37.2
balsam poplar not present not present sown: 2.4
  sown: 2.1
planted: 2.4
planted: 2.1
Bebb willow not present   16.4
  not present   [8]
black spruce sown: 21.0
sown: 4.5
sown: 2.4
planted: 14.8
planted: 17.1
planted: 14.6
Alaska willow not present not present sown: 36.8 (21)   sown: 5.5
planted: 36.8
planted: 5.5 (0.2)
grayleaf willow 16.8
paper birch sown: 2.5
  sown: 17.6
  sown: 18.0
planted: 2.5
planted: 17.6
planted: 18.0
quaking aspen sown: 2.0 35.0
sown: 19.2
sown: 0   [7]
planted: 18.6
planted: 40.9
planted: 73.2
Scouler willow 13.5
  not present   [8]
*For postfire year 6, data were provided for planted regeneration only [7].

Table 9. Mean height (cm (SD)) of natural regeneration, postfire year 6 [7]
Species Vegetative Seed
American green alder 195.1 (31.5) 152.0 (47.2)
black spruce none 17.0 (9.1)
grayleaf willow 243.0 (52.0) 40.2 (16.3)

The species below occurred within the black spruce/feather moss community described in this Research Project Summary. For additional information on these taxa, follow the highlighted links to the FEIS reviews.

Common name Scientific name
Cladonia spp. reindeer lichens
Nephroma arcticum arctic kidney lichen
Peltigera aphthosa green dog lichen
Ceratodon purpureus fire moss
Hylocomiaceae feather mosses
Hylocomium splendens splendid feather moss
Marchantia polymorpha common liverwort
Pleurozium schreberi Schreber's moss
Polytrichum juniperinum juniper haircap moss
Calamagrostis canadensis pinegrass
Chamerion angustifolium fireweed
Alnus viridis subsp. crispa American green alder
Empetrum nigrum crowberry
Ledum groenlandicum bog Labrador tea
Rosa acicularis prickly rose
Salix alaxensis Alaska willow
Salix bebbiana Bebb willow
Salix glauca grayleaf willow
Salix scouleriana Scouler willow
Salix spp. willows
Vaccinium uliginosum bog blueberry
Vaccinium vitis-idaea mountain cranberry
Betula papyrifera paper birch
Picea mariana black spruce
Populus balsamifera subsp. balsamifera balsam poplar
Populus tremuloides quaking aspen


1. Barrett, S.; Havlina, D.; Jones, J.; Hann, W.; Frame, C.; Hamilton, D.; Schon, K.; Demeo, T.; Hutter, L.; Menakis, J. 2010. Interagency Fire Regime Condition Class Guidebook. Version 3.0, [Online]. In: Interagency Fire Regime Condition Class (FRCC). U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior; The Nature Conservancy (Producers). Available: [2013, May 13]. [85876]
2. Begin, Yves; Marguerie, Dominique. 2002. Characterization of tree macroremains production in a recently burned conifer forest in northern Quebec, Canada. Plant Ecology. 159: 143-152. [45872]
3. Dyrness, C. T.; Norum, Rodney A. 1983. The effects of experimental fires on black spruce forest floors in interior Alaska. Canadian Journal of Forest Research. 13(5): 879-893. [7299]
4. LANDFIRE Rapid Assessment. 2005. Reference condition modeling manual (Version 2.1), [Online]. In: LANDFIRE. Cooperative Agreement 04-CA-11132543-189. Boulder, CO: The Nature Conservancy; U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior (Producers). 72 p. Available: [2007, May 24]. [66741]
5. LANDFIRE. 2008. Alaska refresh (LANDFIRE 1.1.0). Biophysical settings layer. In: LANDFIRE data distribution site, [Online]. In: LANDFIRE. Washington, DC: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory; U.S. Geological Survey; Arlington, VA: The Nature Conservancy (Producers). Available: [2013, April 10]. [86808]
6. Viereck, L. A.; Dyrness, C. T., tech. eds. 1979. Ecological effects of the Wickersham Dome fire near Fairbanks, Alaska. Gen. Tech. Rep. PNW-90. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest and Range Experiment Station. 71 p. [6392]
7. Zasada, John C.; Norum, Rodney A.; Teutsch, Christian E.; Densmore, Roseann. 1987. Survival and growth of planted black spruce, alder, aspen and willow after fire on black spruce/feather moss sites in interior Alaska. The Forestry Chronicle. 63(2): 84-88. [85354]
8. Zasada, John C.; Norum, Rodney A.; Van Veldhuizen, Robert M.; Teutsch, Christian E. 1983. Artificial regeneration of trees and tall shrubs in experimentally burned upland black spruce/feather moss stands in Alaska. Canadian Journal of Forest Research. 13: 903-913. [6991]

FEIS Home Page