ABSTRACT

1

Effects of Prescribed Fire and Fire Surrogates on
Insect Pollinators of North Carolina
ABSTRACT Pollination by insects in forests is an
extremely important process that should be
conserved. Not only do pollinating insects
maintain a diversity of plants within forests,
but also aid in pollinating crops that are found
near forested land. Unfortunately, pollinating
insect populations have shown declines in recent
decades throughout the world. Currently, the
effects of various forest management practices on
floral visiting insect abundance or diversity is
unknown. In 2003 and 2004, floral visiting
insects were captured with malaise and colored
pan traps from burn, mechanical, mechanical plus
burn, and untreated plots of forest land in North
Carolina. We caught 7921 floral visitors from
four orders and 21 families. Hymenoptera was the
most abundant and diverse order, with Halictidae
being the most abundant family. Six families and
45 species of Hymenoptera were caught. We caught
seven families and 35 species of Lepidoptera, six
families and 33 species of Coleoptera, and two
families and 13 species of Diptera. The majority
of floral visitors were captured in the
mechanical plus burn treatments, while similar
numbers were caught on the mechanical only
treatments, burn only treatments and control
treatments. Overall species richness was also
higher on mechanical plus burn treatments
compared to other treatments.
Joshua W. Campbell1, James L. Hanula1, and Thomas
A. Waldrop21USDA Forest Service, Southern
Research Station, 320 Green St., Athens, GA
2USDA Forest Service, Southern Research Station,
Clemson, SC
RESULTS The majority of floral visitors were
captured in the mechanical plus burn treatments,
while similar numbers were caught on the
mechanical only treatments, burn only treatments
and control treatments (Fig 3). Overall species
richness was also higher on mechanical plus burn
treatments compared to other treatments (Fig
4). Hymenoptera were significantly higher in
abundance on mechanical plus burn treated plots
in 2003 and 2004 (Fig. 5). In 2003, Halictidae,
was the most commonly captured family of
Hymenoptera and both Halictidae (pAnthophoridae (p 0.085) were captured in
significantly higher numbers on mechanical plus
burn treatments in 2003. In 2004, Halictidae,
Apidae, and Sphecidae were significantly higher
in abundance on mechanical plus burn treatments.
Species richness of Hymenoptera was also higher
on mechanical plus burn. The mechanical plus
burn treated plots had significantly higher
numbers of Diptera in 2003 and 2004 compared to
the other treatments. Syrphidae was the most
common family of floral visiting Diptera
comprising 99 of the total caught. Coleoptera
were significantly higher (pon the mechanical plus burn treated plots in 2004
and 2003 (p0.079) (Fig. 6) Species richness was
also significantly higher on mechanical plus burn
treated plots. Lepidoptera were slightly higher
in abundance on mechanical plus burn treated
plots in both years but there were no significant
differences among treatments. Species richness
was significantly higher (p0.07) on the
mechanical plus burn treatment plots than on the
mechanical only or control plots. Regression
analyses of change in basal area and percent
plant cover found a relationship of increased
plant cover with decreasing basal area (Fig. 7).
Total pollinator abundance also increased as
basal area decreased (Fig. 8) or as percent plant
cover increased (Fig. 9).
METHODS 10 ha plots of mechanical plus
prescribed burn, prescribed burn only, mechanical
only, and control plots were designed in the
Green River Game Management Area near
Hendersonville, NC. malaise traps with blue,
white, yellow and red colored panels (Fig 1) and
colored pan traps (blue, white, yellow) (Fig 2)
were used to collect pollinating insects traps
were operated at 4-6 week intervals from
April-October of 2003 and 2004. we trapped five
times in 2003 and six times in 2004, with each
trapping period lasting seven days. basal area
and herbaceous plant cover was measured and
compared with the pollinating insect data.
Fig 1. Malaise trap with colored panels
r20.39, p0.029 y19-1.4x
r20.71, p0.0006 y573.9110x
Fig 2. Blue Pan Trap
DISCUSSION Mechanical plus burn treatments
caused the greatest changes in basal area of
trees because increased fuel loads on the plots
resulted in hotter fires and more tree mortality.
The combination of removing the dense understory
of mountain laurel and rhododendron combined with
some overstory tree mortality allowed more
sunlight to reach the forest floor resulting in
greater herbaceous plant cover. Although, it is
unclear what caused the increased numbers of
pollinators, a combination of factors
(nectar/pollen resources, suitable nesting
structure, etc.) is likely to be important in
sustaining pollinator populations in forests.
Clearly forest management practices can be used
to influence pollinating insect abundance and
diversity. In this study, the mechanical plus
burn treatments proved to be the best for
increasing pollinator abundance. However,
continued burning, periodic thinning or both over
time might be as effective as the mechanical plus
burn treatments in providing conditions that
favor pollinators.
Fig 8. Linear regression of total pollinators
and change in basal area. Negative numbers
indicate a post treatment decrease in basal area.
Fig 7. Linear regression of plant cover and
change in basal area. Negative numbers indicate
a post-treatment decrease in basal area.
Fig 3. Total mean pollinators captured during
2003 and 2004. Columns with the same letter are
not significantly different at p?0.05 (REGWQ, SAS
1985). Treatments were MBmechanical plus burn,
Mmechanical, Bburn, Ccontrol.
Fig 4. Species richness of floral visiting
insects captured in 2003 and 2004. Columns with
the same letter are not significantly different
at p?0.05, REGWQ, SAS 1985 MBmechanical plus
burn, Mmechanical, Bburn, Ccontrol
r20.58, p0.0042 y-239.644.8x
Fig 5. Mean abundance of Hymenoptera for the two
collecting years Columns with the same letter are
not significantly different at p?0.05 (REGWQ, SAS
1985). Treatments were MBmechanical plus burn,
Mmechanical, Bburn, Ccontrol. .
Fig 6. Mean abundance of Coleoptera for the two
collecting years Columns with the same letter are
not significantly different at p?0.05 (REGWQ, SAS
1985). Treatments were MBmechanical plus burn,
Mmechanical, Bburn, Ccontrol.
Fig 9. Linear regression of total pollinators
and herbaceous plant.