FIRE CASE STUDIES
SPECIES: Larix occidentalis | Western Larch 



1. Lubrecht Experimental Forest: larch/Douglas-fir prescribed fire
2. Miller Creek-Newman Ridge: larch/Douglas-fir prescribed- and wildfire



1st CASE NAME : 
Lubrecht Experimental Forest: larch/Douglas-fir prescribed fire


REFERENCES : 
Norum, R. A. 1975 [88]
Norum, R. A. 1976 [89]
Norum, R. A. 1977 [90]
Reinhardt, E. D.; Ryan, K. C. 1988 [100]
Stark, N. 1977 [135]
Stark, N.; Steele, R. 1977 [138]


SEASON/SEVERITY CLASSIFICATION : 
Spring and fall/low to moderate severity


STUDY LOCATION : 
The study site is on the University of Montana Lubrecht Experimental
Forest located 41 miles (66 km) east of Missoula in the Garnet Mountains
(north half of Section 3, T. 13 N, R. 15 W, Principal Meridian,
Montana).


PREFIRE VEGETATIVE COMMUNITY : 
Study plots were in a Douglas-fir/globe huckleberry (Pseudotsuga
menziesii/Vaccinium globulare) habitat type, bearberry (Arctostaphylos
uva-ursi) phase as described by Pfister and others [95].  Overstory
trees ranged in age from 50 to 300 years with an average diameter of
about 8.6 inches (22 cm).  This stand was not vigorous and was composed
primarily of Douglas-fir and western larch (Larix occidentalis) with
small amounts of lodgepole pine (Pinus contorta) and ponderosa pine (P.
ponderosa).  The largest and best trees had been selectively logged
about 50 years earlier.  Subsequently, an uneven understory of dense
Douglas-fir had developed.  There were a large number of widely spaced,
large-diameter western larch stumps to indicate preharvest conditions.
In addition to conifers and bearberry, huckleberries (Vaccinium spp.),
birchleaf spirea (Spiraea betulifolia), and heartleaf arnica (Arnica
cordifolia) were common in the understory.


TARGET SPECIES PHENOLOGICAL STATE : 
NO-ENTRY


SITE DESCRIPTION : 
The site is located at about 4,800 feet (1,464 m) with east to northeast
exposures and slopes of 20 to 50 percent.  The area has warm summers and
cold winters with annual precipitation of about 18 inches (47 cm), most
of which falls as winter snow.  Soils in the area are thin, poorly
developed sandy loams in the Holloway Series.  They formed in residuum
weathered from quartzite, argillite and, on these sites, contain
sufficient rock to be considered talus slopes.  Calcium and phosphate
are limited in these soils.  The site had a substantial complex fuel
load including many dead small trees in understory thickets.  Total dead
fuel loadings ranged from 5.5 to 50 tons per acre (12.3-112.1 metric
tons/ha).  Over 100 parameters were measured in connection with these
fires.


FIRE DESCRIPTION : 
Out of the 20 test plots broadcast burned during 1973, nine were burned
from early May to the first of July.  The rest were burned from early
September to mid-October.  The average dead fuel moisture contents
ranged from 8.5 to 35 percent and windspeeds were from 0 to 15 miles per
hour (0-24 km/hr).  Strip ignition horizontally across plots was used
for most plots, and fire intensity on each strip was allowed to drop
before the next strip was ignited.


FIRE EFFECTS ON TARGET SPECIES : 
Three years after the fires the average cover of western larch seedlings
on lightly burned plots was 0.27 percent; on moderately burned plots was
0.23 percent; on hotly burned sites was 2.4 percent; and no western
larch seedlings had established on control plots.  Nutrient analyses
were done on prefire and postfire soil, soil water and plants.  Results
of comparison of nutrients from aboveground samples of western larch
seedlings from burned and control plots were:

                       micrograms/gram 
                   Control        Hot burn
Calcium             4,805           3,000
Copper                 37              18
Iron                  162             280a
Potassium           6,864          16,000a
Magnesium             967           1,300a
Manganese             343             198
Sodium                114              95
Phosphorus          4,457           7,190a
Zinc                   28              37a

(a) indicates a significant difference at the 5 percent level.

Eight years after the fires a tree growth analysis was performed limited
to radial and basal area growth.  Western larch's relative radial
increment on burned plots was less than on unburned plots in the first
year and greater on burned plots thereafter.  The difference in growth
of trees on burned plots compared to trees on control plots increased
each year for the first 4 years.  Western larch's response was more
positive than that of Douglas-fir and from the fourth to the eighth year
the average relative radial increment was 60 to 80 percent greater on
burned plots than on unburned plots.  The average unadjusted radial
growth increment of trees on burned and unburned plots for the first 8
years after treatment is given below:

Year               Burned              Unburned
              inches     cm       inches      cm
  1            0.044    .114       0.048     .121
  2            0.059    .150       0.054     .136
  3            0.076    .193       0.055     .139
  4            0.069    .175       0.036     .091
  5            0.068    .173       0.037     .093
  6            0.056    .143       0.034     .087
  7            0.067    .171       0.037     .093
  8            0.071    .181       0.042     .106


FIRE MANAGEMENT IMPLICATIONS : 
Underburning in similar western larch/Douglas-fir forests is feasible.
An average of 15 percent of the overstory trees were killed in the
plots.  Within the range of fuel loadings in this study, fires were most
manageable and still effective when the moisture content of 0 to 1 inch
(0-2.5 cm) dead fuels was around 15 percent.  Strip ignition helped
overcome control and ignition problems caused by discontinuous
concentrations of heavy fuels.  Underburning requires attention to the
form, moisture status, and amount of living vegetation.  Detailed
prescriptions for underburning are given in Norum [1976 and 1977].  

Western larch seedlings established best on sites burned by the hottest
fires.  Prescribed underburns in western larch stands can result in an
increase in individual tree relative radial increment.  However, growth
of western larch in these poorly growing stands continued to be slow.
Growth, even in trees with fire damage, was not reduced by the fire, and
fire may be a useful tool for fuel reduction or other purposes in such
stands.



FIRE CASE STUDIES
SPECIES: Larix occidentalis | Western Larch 



2nd CASE NAME : 
Miller Creek-Newman Ridge: larch/Douglas-fir prescribed- and wildfire

REFERENCES : 
Beaufait, W. R.; Hardy, C. E.; Fischer, W. C. 1977 [7]
DeByle, N. V. 1981 [20]
Shearer, R. C. 1975 [123]
Shearer, R. C. 1976 [124]
Shearer, R. C. 1982 [128]
Shearer, R. C. 1984 [129]
Shearer, R. C. 1989 [153]


SEASON/SEVERITY CLASSIFICATION : 
May through October/low to very severe


STUDY LOCATION : 
Two study locations were used.  The first contained 641 acres in the
Miller Creek and Martin Creek drainages of the Flathead National Forest
of northwestern Montana.  This is referred to as the Miller Creek area.
The second location consisted of 526 acres on Newman Ridge located
between Two Mile Creek and Ward Creek on the Lolo National Forest near
the border of western Montana and Idaho.  


PREFIRE VEGETATIVE COMMUNITY : 
Most of Miller Creek was considered to be in one of three phases of the
subalpine fir/queencup beadlily (Abies lasiocarpa/Clintonia uniflora)
habitat type.  The fool's huckleberry (Menziesia ferruginea) phase was
found on higher middle and upper north- and east-facing slopes.  The
beargrass (Xerophyllum tenax) phase was on drier south and west aspects
and the queencup beadlily phase on most other sites.  Stream bottoms
belonged to the western redcedar(Thuja plicata)/queencup beadlily
habitat type.  The dominant conifers were western larch (Larix
occidentalis), Douglas-fir (Pseudotsuga menziesii), and Engelmann spruce
(Picea engelmannii), with some lodgepole pine (Pinus contorta), grand
fir (Abies grandis), and subalpine fir.  The larch/Douglas-fir cover
type occupied over 50 percent of the area.

On Newman Ridge seven habitat types were identified.  The warmest and
driest was the Douglas-fir/ninebark (Physocarpus malvaceus) habitat type
on convex southwest slopes.  Other habitat types included grand
fir/queencup beadlily on concave east, northwest, and protected
south-facing slopes; grand fir/beargrass on upper west-facing slopes;
western redcedar/queencup beadlily h.t., fool's huckleberry phase on
concave north- and northeast-facing slopes; Douglas-fir/globe
huckleberry (Vaccinium globulare) h.t. beargrass phase on upper
south-facing slopes; subalpine fir/queencup beadlily h.t., fool's
huckleberry phase on north slopes along the ridge; and subalpine
fir/beargrass h.t., globe huckleberry phase on south slopes near the
ridge.  Dominant conifers were Douglas-fir, larch and lodgepole pine
with some ponderosa pine (Pinus ponderosa), grand fir, subalpine fir,
western white pine (Pinus monticola), Engelmann spruce and western
redcedar.  Prefire stands at Newman Ridge were 26 percent western larch.


TARGET SPECIES PHENOLOGICAL STATE : 
NO-ENTRY


SITE DESCRIPTION : 
Elevation at Miller Creek ranges from 4,200 to 5,000 feet (1,280-1,524
m) with slopes averaging 24 percent and ranging from 9 to 35 percent.
Soils are Andic Cryoboralfs that developed in glacial till from the
argillites and quartzites of the Wallace (Belt) formation.  Average
precipitation is about 25 inches (64 cm) annually; approximately
two-thirds falls as snow during the long cool winter.  Elevation at
Newman Ridge ranges from 4,400 to 5,400 feet (1,341-1,646 m) with slopes
averaging 55 percent and ranging from 44 to 76 percent.  Soils are Andic
Cryochrepts that have developed in place or in colluvium from argillites
and quartzites of the Belt formation.  There is a surface loess deposit
containing ash from the Mt. Mazama and Glacier Peak volcanic eruptions
at both sites which is 0.5 to 2.5 inches (1-6 cm) thick at Miller Creek
and 2 to 3 inches (5-8 cm) thick at Newman Ridge.  Average precipitation
is nearly 40 inches (102 cm) at Newman Ridge of which two-thirds falls
as snow.

Sixty 10-acre (4-ha) treatment units were established at Miller Creek
and 16 units, ranging in size from 20 to 58 acres (8-24 ha), were
established at Newman Ridge.  One-fourth of the units at each site
generally faced each of the four cardinal directions.  The units were
clearcut, slashed and burned.  Fuel loads after clearcutting and before
fire, excluding duff, ranged from 60 to 165 tons per acre (135-370
metric tons/ha).  Average fuel loads are described below:  

Average            Miller Creek                   Newman Ridge
Weight        tons/acre      kg/square m     tons/acre      kg/square m

Duff             26.3           59.1            22.9           51.3
Needles           1.5            3.5             1.4            3.1
0 to 1 cm         1.3            2.9             1.1            2.5
1 to 10 cm        9.8           22.1            10.7           24.0
> 10 cm         101.3          227.0            90.5          202.8
 Total          140.3          314.5           126.5          283.7


FIRE DESCRIPTION : 
Slash fuels were allowed to cure for an average of 9 months before
burning (2 to 18 months range).  Fuel moisture of 0 to 0.4 inch (0-1 cm)
branchwood ranged from 5 to 21 percent.  Burning patterns and fire
severity varied among the plots burned.  After broadcast burns at Miller
Creek, 75 percent of the fuels less than 3.9 inches (10 cm) burned and
60 percent of the larger fuels burned.  At Newman Ridge 89 percent of
the fuels less than 3.9 inches (10 cm) burned and 55 percent of the
larger fuels burned.  Greater surface soil heating occurred at Newman
Ridge than at Miller Creek because the duff layer was shallower and
water content of both duff and soil was lower.  The average duff
reduction ranged from 36 to 70 percent at Miller Creek and 44 to 99
percent at Newman Ridge.  In 1967 a wildfire burned five units that had
been clearcut and four units that were uncut forest at Miller Creek.
Average duff reduction from the wildfire was 93 percent with a range of
84 to 100 percent.


FIRE EFFECTS ON TARGET SPECIES : 
Seedbeds were varied.  Early 1967 and most 1968 fires did not completely
burn litter and duff or expose much mineral soil.  The 1967 wildfire
consumed most of the duff and killed most roots of sprouting species.
Other fires were spotty and exposed some mineral soil.  Very few western
larch seedlings established on unburned duff.  Unburned duff continued
to decrease for several years exposing bare soil on areas where the fire
had left charred duff.  The reasons for this decrease may include:
increased decomposition stimulated by warmer surface temperature during
May and June where adequate moisture was present; redistribution by
precipitation, runoff or wind; and oxidation.  However, since western
larch needs to establish quickly and become dominant, this may benefit
other conifer species more.  In addition to natural seeding, seeds were
sown in 1967 on test plots, and bareroot seedlings were planted on
Newman Ridge from 1970 through 1975 and on four clearcuts at Miller
Creek from 1970 through 1973.

Postfire seed dispersal into the clearcuts from western larch in the
timber around clearcut areas was good.  The best seed year for all
conifers was 1971; however, heavy frost in May of that year decreased
the potential western larch seed crop.  The cumulative average number of
sound seed of western larch from 1969 through 1974 on eight clearcuts on
Newman Ridge by distance from the source is listed below:

         Within         Distance from timber edge within clearcut
         Timber    0-200 ft   200-400 ft     400-600 ft     600-800 ft
                   (0-61 m)   (61-122 m)     (122-183 m)    (183-244 m)

/acre    53,200      7,500      3,700           2,000          800  
/ha     131,500     18,600      9,200           4,900        1,900  

Germination of western larch began before the snow completely melted and
was greater on mineral soil than on unburned duff more than 0.5 inch (13
mm).  Seed and seedling losses were caused by rodents, drought, frost
heaving, high temperatures at the soil surface, and migrating juncos
that ate emerging seedlings in 1968.  Drought was the leading cause of
death on south-facing slopes and second highest on other aspects.
Natural regeneration was lowest on south-facing slopes.  In 1978 at
Miller Creek, stocking of western larch seedlings averaged 54 percent on
burned units and 5 percent on unburned clearcuts.  By 1984 at Miller
Creek, stocking of established seedlings of western larch averaged 71
percent on burned units and 1 percent on unburned clearcuts.  In 1979 at
Newman Ridge, stocking of established western larch seedlings averaged
10 percent (6%-16%) on burned clearcuts.  The average number of
established (>1 foot (30.5 cm) in height) western larch seedlings per
acre (hectare) in 1979 and 1984 and the range in 1984 on 37 burned units
at Miller Creek and 7 burned clearcuts at Newman Ridge are given below:

                        1979            1984       1984  
                       Average         Average     Range
Miller Creek
   /acre                  610             931     92-4,003
   /hectare             1,507           2,301    227-9,892
Newman Ridge
   /acre                   30              91     10-228
   /hectare                74             225     25-563

At Miller Creek in 1978, the tallest western larch seedlings averaged
8.5 feet (2.6 m) on uncut plots burned by the wildfire, 5.3 feet (1.6 m)
on clearcuts that were burned by the wildfire, and 3.7 feet (1.1 m) on
clearcuts that were burned by prescribed fire in 1967.  This sequence
may be because a greater proportion of the western larch were older on
the uncut units burned by wildfire or because the evapotranspiration was
lower and the number of growing days greater on those units.  Height was
also related to habitat type and phase with the tallest trees on the
warmer and drier beargrass phase of the subalpine fir/queencup beadlily
habitat type and shorter trees on the cooler and moister fool's
huckleberry phase.  Evidently, soil moisture is not as limiting in early
stand development as it is later; also, trees on the warmer sites began
growth earlier in the spring and may have benefited from extra nitrogen
contributed by snowbrush ceanothus (Ceanothus velutinus).  At Newman
Ridge in 1979, the tallest western larch seedlings averaged 3.0 feet
(0.9 m).  Natural regeneration of western larch at Newman Ridge was
lower than at Miller Creek due to harsh site conditions, larger
clearcuts, and poor seed production which resulted in less stocking than
expected from the previous stand composition.


FIRE MANAGEMENT IMPLICATIONS : 
During late spring and early summer, duff is usually wet and fires do
not expose much mineral soil.  Late summer or early fall fires are more
effective at removing duff and exposing mineral soil for western larch
regeneration.  However, if precipitation occurs, fuels and duff need to
dry for several days.  At Newman Ridge, moderate intensity fires removed
most of the duff and prepared an adequate seedbed.  At Miller Creek, the
same intensity fire exposed less mineral soil because the duff was
thicker and wetter.  Habitat type and site conditions alter the amount
of duff removal needed.  On mesic habitat types hot fires that expose a
high proportion of mineral soil, followed by good seed years, lead to
overstocking.  On steeper slopes with drier conditions, such as at
Newman Ridge, residual duff layers have more adverse impact on the
survival of seedlings.  

Seed dispersal should be taken into account when deciding the time of
fall fires.  In a good seed year, dispersed seed could be destroyed by
fires after early September at lower elevations and a few weeks later at
higher elevations.  The light seeds of western larch enable it to
establish a high proportion of the seedlings in the center of larger
clearcuts such as those on Newman Ridge.  Planting may be necessary on
steep, harsher sites.  Expected survival for planted western larch on
such sites at Newman Ridge was high in the western redcedar/queencup
beadlily habitat type; moderate in the grand fir/queencup beadlily,
grand fir/beargrass, and Douglas-fir/globe huckleberry habitat types;
and low in the Douglas-fir/ninebark habitat type.  Actual survival by
1979 was lower than expected, only 38 percent for western larch overall.

Related categories for Species: Larix occidentalis | Western Larch