Fire as a Disturbance Agent

					Fire as a Disturbance Agent

Scott Stephens
UC Berkeley
ESPM 290, 9/23/03

Science of Fire as an Ecological Factor
       -Begins in southeast in the 1920’s (Chapman (1922) first fire paper, Longleaf
       - Spreads to west in the 1940’s and 50’s (Weaver, Cooper, Ponderosa Pine)
       -Harold Biswell, first professor specializing in fire in western North America
       UCB 1947-1970)
               - Biswell was trained as a range scientist, and career began in SE US

Disturbance Regime (Sousa 1984)
       -Areal extent
       -Sizes of fires
               - In the 1910-20's Rangers would map fire perimeters
               - Often could put out fires with pickup/shovels
       - Areas burned by low, moderate, and high intensity fires, not just perimeters.
       - Historical records of fires – perimeter data. Within the perimeter, fires are
       variable, and this is not recorded. Would be more useful to have a map.
       Remote sensing can assist.
               - Now have a normalized burned severity index; can map severity within
               perimeter using a metric such as chlorophyll

Prescribed fire size can have a big impact on fire effects.
       - Burning in native California grasslands in the central valley, fires ignited when
       current years seed mature but not dispersed (exotic Medusa Head). Fire intensity,
       ROS, etc. were kept as similar as possible.
                - Burn small units (1 ha) -no impact
                        - Edge effects, dispersal from matrix too great
                - Burn large units (50 ha)-reduce non-natives by 90%
                        - Natives also made a comeback
       - Stephens: surprising little research has been conduction on spatial scale


Intensity (measure of strength of disturbing force)
        1. Fireline intensity (kW/m). Heat produced from the flaming front (heat transfer)
                - Higher flames, more heat transfer to crown, buds
                - Heat transfer is dominated by convection – upward movement of heat
                - Intensity can also be measured by Reaction Intensity – kW/m^2
        2. Fuel Consumption (measure of strength of disturbing force)
                - Heat produced from burn-out of the forest floor (duff and litter) and large
                woody fuels.
              - Higher fuel consumption, more heat to bark, fine roots
              - A 20 cm flame length can girdle trees
              - Scorched trees with small amounts of surviving crown can survive,
              possibly because they are root grafted to neighbors.
       - 60 degrees C for 1 minute kills any plant tissue
              - Questions: Tropical plant surfaces are known to be exposed to similar
              temps in nature. Likely means internal plant temperature?

Magnitude continued

       - Severity (measure of the damage caused by the disturbing force)
       Interaction of fireline intensity and fuel consumption produces severity in many
               - High severity, majority of dominant vegetation killed (most grasslands,
               shrublands, some forests)
               - Quantify by percent of biomass killed. Need pre and post inventory data.
                       - Stephens: Never been done in fire literature
                       - Battles: Use soils, C, organic matter consumption
                       - Stephens: Is logging a form of severity?
                       - Battles: Yes, often measured as such.

              - Stand replacement vs. stand removal
                      - replacement: vegetation adapted to fire effects
                              - ex. serotinous cones

                      - removal: vegetation not adapted to fire effects

      - Number of disturbances per unit time

       - Low-moderate intensity fire regimes, sample fire scars to estimate point fire
       return intervals
               - In Western US, can often go back 350 years
               - Swetnum, using Giant Sequoias, has gone back to 2000 B.C.
               - Most conservative estimate of past frequency
                        - trees often do not record all fire events
               - Debate about random sampling (random point frequency) – fire data is
               clumped, some areas seem to record fire more than other areas. Should
               these areas be targeted for sampling?
               - Is the absence of evidence actually evidence of absence?

       - Fire scars commonly in patches. Approximately 10-30 percent of trees scarred,
       others of similar age never scarred. Not sure why.
               - Possible large woody fuel combustion near tree causes scar. After tree is
               a good recorder
                       - gravity accumulates debris on uphill side.
                       - upslope side of bole can receive 40x's the heat transfer
               - Different species have different flammability/ susceptibility to fire
                       - ex: Ponderosa Pine needles make fluffy fire fuel, White Pine
                       needles make a flat bed that is not as flammable
                       - cone morphology determines flammability: seeds in closed cones
                       don’t burn. Seeds in open cones can burn, and fire can potentially
                       cause stand removal because the seed bank is destroyed.
                       - Some species/ communities (ex: Chaparral) are adapted to high
                       severity fire regimes, and stand replacement occurs

       -Quantify Fire History
              - Samples taken to lab and sanded to 400 grit
              - Cross-dating allows exact colander year to be assigned to fire scar
              - Seasonality also estimated from scar position in ring
              - Need for more information on actual annual ring growth

Frequency continued

       - Regional frequency (total number of disturbances in an geographical area per
       unit time)
       - Composite fire return interval
               - 1-3 ha sampled, combined into composite
       - Low-moderate intensity fire regimes, sample fire scars to estimate mean and
       median fire return intervals (must define area sampled)
               - Same debate about random sampling
               - Composite area sampled should be small, average fire event should burn
               entire area
                        - Otherwise fire interval estimates will be in error
               - Common sizes 1-3 ha
               - In analysis treat essentially as a point

       - Inversely related to the variance in the time between disturbances (possible from
       fire scars)
               - Stephens: haven't seen this reported
       - Turnover rate or rotation period
       - Fire rotation. Number of years required to burn the area of interest. Not all ha
       have to burn.
               - Challenging to obtain field data (many plots-stratify)
               - Best for stand replacement regimes
                        - Minnich et al. 2000 in northwest Mexico
               - Mixed severity regimes can use both composite fire return intervals and
               fire rotation

       - Season of year when fire actually occurred
       - In California, prehistorically fire season was August- December
       - Can have a dramatic impact on fire effects depending on vegetation
       - Grasslands and shrublands
               - Seed banks in soil and contained in plant
               - Spring: seeds face difficult post-fire conditions during CA summer
               - Fall: better conditions after burn
       - In forests seasonality can impact immediate fire effects
               - New conifer growth killed
               - Possibly more susceptible to bark beetles in early season
       - New growth is more susceptible to fire than old growth
       - Wildlife may be disrupted during nesting
       - Impacts may be small in long-term, because of the long life of the trees involved
       - Example: Chaparral - when burned in the spring, the seeds are scarified at the
       dry time of the year, and they can’t germinate

- Example of Timing for 5 Western Forest Regions:

Region                       Growing (%)      Early (%)      Late (%)       Dormant (%)
San Pedro Martir             99               91             8              1
S Sierra Nevada              87               15             62             13
N Sierra Nevada              8                4              4              92
S Cascades                   20               9              11             80
Klamath                      16               4              12             84

- Most lightning strikes occur in July for both the SSPM and S Sierras
       - SSPM is drier and is ready to burn earlier in the year.
       - Table represents a N/S gradient.


       - Most difficult to understand and predict
       - Interaction of past stand replacement fires and blow down in Colorado
       (Kulakowski and Veblen 2002)
               - More blow down on ridges but less in areas with relatively young trees
               - Areas with old-growth more susceptible
               - In areas where blow down below 80%
                       - At this level and above other factors not important
                       - Threshold exceeded. Many systems seem to have this
               Battles: Are thresholds common in other fields? Or are we just identifying
                      - Power, Sousa agree to thresholds in certain instances (scour, log
                      battering), DEPENDS on disturbance agent, and the level of detail
                      that you look at.

       - Prescribed fire in bishop pine
               - Both fire size and season impacted severity
                       - Major mortality (Ips beetle came into burned area and wiped out
                       the trees), 2/3 killed in Spring, 3/4 Fall
                       - low intensity fire in a high intensity system
                               - importance of spatial scale – if the burned area was large,
                               insects may have been more dispersed, less concentrated,
                               and less damaging
                       - Stephens: small size = higher beetle mortality?
                       - Student: Bishop pine is adapted for crown fires, not surface fires.

       - Fire and Bark Beetles
               - Prescribed fire commonly induces attack from red turpentine beetle
               - Trees stressed and more susceptible
               - Does it increase severity? Maybe
                       - RTB's often disappear after initial attack

       - Fire Regimes in the Sierra San Pedro Martir, Mexico
               - Why did the fire regime change at 1790?
                      1. Grazing?
                              - Likely not enough cattle to have impacts
                      2. Intro of disease to native peoples?
                              - May not have used area, maybe concentrated at Mission.
                              - Anthropology of area is unknown
                      3. Climate Change?
                              - records show synchrony in fire regime throughout SW at
                              this time
                              - changes in El Nino SO at this time – lightning source may
                              have disappeared. Could look for a change in O18 isotope
                              to se if there is a change associated with El Nino rains
                              - Chagoopas (sp?) evidence
                                       - pine islands in lava beds, no grazing, no Natives
                                       - Show similar climate signal
               - What factors may have contributed?

Discussion about fire behavior models – Firesight
       - Fit to observed behavior is getting quite good ~85% correct
       - Links fire behavior models to fire effects
       - Estimates fuel consumption
       - Estimates post-fire mortality of dominant organisms (trees)
        - Aids in prioritizing areas for fire treatments – simulate fires in certain areas and
        predict the effects.
        - No comparable models exist for other systems


Grogan, et al. 2000. Fire effects on ecosystem nitrogen cycling in a Californian bishop pine
forest. Oecologia 122:537-544.

Kulakowski, D. and T.T. Veblen. 2002. Influences of fire history and topography on the pattern
of a severe blowdown in a subalpine forest in northwestern Colorado. Journal of Ecology 90:

Neary, D. G., C. C. Klopatek, L. F. DeBano, and P. F. Ffolliott. 1999. Fire effects on
belowground sustainability: a review and synthesis. Forest Ecology and Management 122:51-71.

Paine, R. T., et al. 1998. Compounded perturbations yield ecological surprises. Ecosystems

Stephens, S.L., C.N. Skinner, and S.J. Gill. 2003. Dendrochronology-based fire history of Jeffrey
pine-mixed conifer forests in the Sierra San Pedro Martir, Mexico. Canadian Journal of Forest
Research 33:1090-1101.

Turner and Romme. 1994. Landscape dynamics in crown fire ecosystems. Landscape Ecology

Shared By: