Mean basal area of the tree layer was 25.6, 23.7, and
15.3 m2 ha 1 for control, light, and moderate damage classes,
respectively, and values were significantly different between control and moderate classes and between light and moderate classes.
Density of stems > 5 cm dbh for control, light, and moderate classes was 771, 666, and 531 ha 1. The quadratic mean diameter
(QMD) for each damage class was 20.6 cm for control, 21.3 cm
for light, and 19.2 cm for moderate. QMD was statistically different
between light and moderate classes. Species richness of the tree
layer (stems P 5 cm dbh) was 32, 37, and 33 for control, light
and moderate damage classes, respectively, and was 40 for all
treatments combined. Mean sapling richness per plot was statistically different between control (mean S = 3.4 plot 1) and moderate
(S = 5.0 plot 1) classes, and between light (S = 3.6 plot 1) and moderate classes. Species richness was not significantly different
between any damage categories in the seedling or tree layers,
when seedling and sapling layers were combined, or when all
woody stems were combined. Abundance statistically differed
between all three classes of the tree layer (stems P 5 cm dbh),
and within the sapling layer between the control and moderate
damage classes and between the light and moderate classes. We
documented no statistically significant differences in species evenness between damage classes. Mean Shannon diversity for the tree
layer of control plots was 1.80 ± 0.06 (SE), for light damage plots
was 1.85 ± 0.04, and moderate damage plots was, 1.70 ± 0.07. We
found no significant statistical differences in Shannon diversity
except in the sapling layer between light and moderate damage
classes, for which mean values per plot were 1.02 ± 0.07 and
1.32 ± 0.08 (P = 0.012), respectively.
Average live basal area per 0.04 ha plot was 1.00, 0.96, and
0.61 m2 for control, light, and moderate classes, respectively, and
these values were not significantly different. Average basal area
lost (removed by the storm) per plot was 0.01, 0.26, and 0.56 m2
control, light, and moderate damage classes, respectively. Mean
percent basal area lost in each damage class was 2%, 21%, and
48%, for control, light, and moderate classes, respectively, and
these values were significantly different (P < 0.05). The mean dbh
for all trees in the decay 1 class was 24.1 cm. The mean dbh of Q.
alba individuals damaged during the storm was 38 cm, and the
mean dbh of all Quercus individuals damaged was 38 cm. The mean
dbh of Carya individuals was 29 cm, and the mean branch size
measured was 23 cm dbh. Of stems within the decay class 1 category, 17% were snags, 53% were snapped stems, and 30% were
uprooted stems. Stems were more likely to be snapped than either
to be uprooted or to remain standing as snags (P < 0.001).
For the control plot trees >5 cm dbh, Ostrya virginiana (Mill.) K.
Koch, Q. alba, Acer saccharum Marsh., and Carya spp. occurred in the
highest densities. The species with the highest densities on light
damage class plots were O. virginiana, Q. alba, F. grandifolia,
and Magnolia macrophylla Michx. Ostrya virginiana, Q. alba,M. macrophylla, and A. saccharum were present in the highest densities on moderate damage class plots (Table 1 ). The five most
dominant taxa based on basal area contribution within the control
class were Q. alba, Carya spp., A. saccharum, Quercus rubra L. and
L. tulipifera. The five most dominant taxa on light damage plots
were Q. alba, Carya spp., L. tulipifera, and Quercus prinus L. and Q.
rubra. Similarly, the taxa with the highest relative dominance on
moderate plots were Q. alba, Carya spp., P. taeda, and F. grandifolia
(Table 1 ). The most common species damaged within the control
plots, which represented expected background mortality, based
on relative densities were A. saccharum, Q. alba, Cercis canadensis
L., and M. macrophylla (Table 2). Within the light disturbance plots,
Q. alba, Carya spp., O. virginiana, and Q. prinus were damaged in the
highest densities. Taxa with the highest rates (based on density) of
mortality on moderate damaged plots were O. virginiana, Carya
spp., Q. alba, and M. macrophylla.
We documented 14 canopy dominant trees ha 1 in control
plots, 13 ha 1 in light damage plots, and 5 ha 1 in moderate damage plots (Fig. 2). The density of canopy co-dominant trees ranged
from 129 ha 1 in control plots to 64 ha 1 on moderate damage
plots. Across all damage classes, the canopy was dominated by
Quercus and Carya species. We did not document any Acer species
or F. grandifolia stems with dominant positions in the canopy and
these taxa represented minor contributions to the canopy codominant stratum. Density of overtopped stems in control plots
far exceeded the abundance of such stems in light and moderate
damage plots (119% more overtopped stems in control plots than
in moderate damage plots). Quercus and Carya represented a small
component of overtopped stems across all damage