Consequences

1
Consequences correlates of fire in Florida
wetlands Gordon A. Fox1,2, David Brownlie3,
Monica Folk4, Sandy Woiak4, and Kinberly
Hum2 1Dept. of Biology, University of South
Florida 2Dept. of Environmental Science
Policy, University of South Florida 3U.S. Fish
Wildlife Service, Tall Timbers Research Station
4The Nature Conservancy, Disney Wilderness
Preserve
Results organic soil loss Predicted well by
KBDI from nearest station (several km) at larger
scales, predictive power was lost (Table 1)
Predicted poorly by monthly piezometer readings
(finer spatial but coarser temporal scale) (Table
1). Depended strongly on community and KBDI
(Figure 2).
Introduction Fires in wetlands sometimes lead
to organic soil fires these can generate
considerable smoke. As a result, land managers
are sometimes reluctant to conduct prescribed
fires in wetlands, and fire managers often take
special measures to keep wildfires out of
wetlands. Little is known about the conditions
under which organic soil fires actually occur.
Even less is known about the effects of fire on
wetland plant communities.

• Questions
• Can we predict when fires lead to substantial
  loss of organic soil?
• What kind of data will best predict this?
• What are the ecological consequences of wetlands
  fires?
• Specifically, what are the effects of fire on
• Diversity and richness
• Wetlands vs. uplands species

Table 1. Best-fitting soil-loss models
Figure 3. Mean effect of fire on (top) species
richness and (bottom) Simpson evenness, by
community.

• Methods
• Wild and prescribed fires at TNCs Disney
  Wilderness Preserve in 1994-2003.
• Used a set of 104 transects in 24 wetlands
  (Figure 1).
• Piezometers in each major plant community along
  each transect, read monthly. Calculated KBDI for
  each burn date at nearest station (up to several
  km), at main station on property (up to 12.5 km),
  and county-wide.
• Annual measurements of organic soil depth
  densities of each plant sp.

• Conclusions
• We can predict soil loss. This should help
  managers decide when they can burn wetland sites.
• Weather data on the scale of a few km predicts
  soil loss well.
• Consequences of burning wetlands depend strongly
  on wetland community type.
• Fire in wetland communities often has some
  important ecological benefits, including the
  following
• Increased species richness.
• This is important in many conservation settings.
• Increased cover of wetland species.
• This is important in many restoration settings.
• Many land managers should consider
  re-introduction of fire into wetlands.
• Fire sets the clock for changes in diversity in
  these wetland plant communities.

Figure 2. Mean effect of fire on soil loss, by
community. All error bars give 95 confidence
intervals.

• Results plant communities
• Richness is increased by fire in most
  communities but evenness is decreased (Figure 3).
• For burned sites, diversity is predicted much
  better by models using years-since-fire as the
  repeated measures unit, rather than calendar
  year.
• Fire always increased relative cover of wetland
  species in Cypress, Flatwood, Marsh, and Slash
  Pine communities. In other communities the effect
  varied over time, or was not estimable.

Figure 1. Fires (hatched areas) at DWP since
1998, and transects (red lines).

• Statistical analyses
• Data analyzed with mixed-model ANOVA, using
  wetland and transect as random effects all
  others are fixed.
• Where multiple models were fit, we used the
  Akaike Information Criterion (AIC) to select the
  best. The best is defined as the model with the
  lowest AIC. AIC trades off goodness-of-fit
  against number of parameters.