What's New
Research & Projects Publications
People
Data
Links
UCSB Biogeography Lab Publications Abstracts
EFFECTS OF VARIATION IN SOIL HEATING DURING FIRE ON PATTERNS OF PLANT ESTABLISHMENT AND REGROWTH IN MARITIME CHAPARRAL
ODION, DENNIS, 1995
PhD Dissertation, Department of Geography, University of California, Santa Barbara. 289 pp.I related patterns of chaparral regeneration to spatial variability of surface heating
during fire and its effect on seeds and resprout propagules. Three burns were
studied. All took place on the Burton Mesa in northern Santa Barbara County.
Sites were on level terrain where eolian sand supports maritime chaparral. This
chaparral is composed of the widespread dominant, Adenostoma fasciculatum,
capable of resprouting, and locally endemic Ceanothus and Arctostaphylos,
which can only regenerate from seed the first year after fire. The first burn in
1986 (Chapter I) was used to test procedures for the more detailed studies to
follow. Patterns in vegetation and seed assemblages along a 24 m transect
were investigated in relation to soil temperatures during burning. Maximum soil
temperatures were lowest in 1-2 m gaps in the pre-burn canopy. Post-burn
vegetation was very patchy, with 4-6 m wide former understory areas nearly
devoid of seedlings, and former gaps supporting high densities. Numbers of
germinable seed after fire were highest in gaps. Direct soil heating effects could
not be adequately determined because refractory seed in pre-burn samples was
not quantified. For the second and third fires in 1988 and 1989 (Chapters II-IV), I
established transects consisting of 47 contiguous m$/sp2$ plots. Pre- and
post-burn seed populations were studied in detail at two depths. I described
patterns of heating using several procedures along transects and in and around
20 $/sim$ 1.5 m diameter canopy gaps in the otherwise continuous shrub cover.
At five of these gaps, the distribution of fuel was reversed by moving adjacent
canopy into the gap. Seedlings were counted in all plots for several seasons
after fire. Thermocouples indicated that both fires produced high heating
compared to what has been reported in the literature. Collapse and prolonged
combustion of canopy material on the ground was common, as indicated by
thermocouples. This long duration, glowing combustion was largely responsible
for the dramatic spatial variation in heating detected at the soil surface. Variation
in total heating and duration were strongly related to pre-burn canopy cover.
Patterns of variation in overall heating reached its maximum at 3-5 m block
sizes, one-half the size of the 5-6 and 1-2 m patches of dense and sparse
canopy combined. Seed populations diminished greatly with fire, and much
more in areas of greater heating. This rearranged patterns in seed abundance
and reflected the heating pattern imprinted into the soil by glowing combustion.
Seedlings and resprouts in the field also exhibited this pattern. Depth of burial
was important to seed survival, for example, seeds of Arctostaphylos and
Ceanothus were deeper and suffered lower mortality than those of Adenostoma.
Experimental reversal of gap and adjacent understory fuel reversed patterns in
seeds and seedlings after fire, even though seeds were more abundant before
fire in gaps. Regeneration patterns were not appreciably modified during 5 years
of vegetation development following fires. Thus, species distributions after fire
are directly linked to the pre-burn vegetation through its effect on patterns in
seeds and resprouts during combustion, and these patterns are manifest in
long-term community structure.