Biology Reference
In-Depth Information
already burned area heavily dominated by
M. minutiflora
and
S. condensatum
. The
site had burned originally in 1972 and had few native species (Hughes et al. 1991;
D'Antonio, unpublished). We started by introducing seeds of two species that were
fire-tolerant and perhaps fire-enhanced,
D. viscosa
and
Sophora chrysophylla
. The
latter is a small, N-fixing tree known to support native bird species. We seeded plots
both before fire and after fire to evaluate whether fire would stimulate or inhibit
their germination. We also seeded adjacent plots of identical age (previous wood-
land) that we did not burn in order to evaluate whether restoration could be achieved
in a previously burned, exotic-grass-dominated site without reburning it. We fol-
lowed seedling emergence and survival over the next 4 years. This initial fire was
a low-intensity fire consuming only 50% of the aboveground grass biomass (Mack
et al. 2001). The grass layer regenerated very quickly both from seed (
M. minutiflora
)
and from resprouting individuals (
S. condensatum
) greatly limiting the window of
time for native species to establish. Nonetheless, we found that both
S. chrysophylla
and
D. viscosa
germinated and grew within the burned plots (Fig. 7.1). The former
species also germinated and grew in the unburned plots but individuals did not
reach as large of a size. In the burned plots
S. chrysophylla
individuals reached
flowering age within 5 years. Fourteen years later some of these individuals were
very large (Fig. 7.2) and had seedlings of their own species growing nearby, sug-
gesting that further recruitment had occurred.
D. viscosa
established at a lower rate
than
S. chrysophylla
perhaps because its seeds benefit from higher intensity fire
events (Tunison et al. 1995). Despite this,
D. viscosa
establishment was higher in
the burned than in the control plots (Fig. 7.1). Some individuals of this species per-
sisted throughout the monitoring time and they too reached reproductive maturity
and were still present after 14 years (Fig. 7.2).
Kipuka Nene burn
120
No burn
Prescribed burn
100
80
D. viscosa,
p = 0.000
S. chrysophylla,
p = 0.251
60
40
20
0
control
preburn
postburn
treatment
Fig. 7.1
Example of results of revegetation burns in Hawaii Volcanos National Park. Shown here
are results of seeding in the first controlled burn conducted for revegetation of grass-invaded,
burned seasonal submontane habitats conducted by Tunison, D'Antonio, and Loh in January 1993.
Data are individuals of either
Dodonaea viscosa or Sophora chrysophylla
that were >10 cm in
height after 4 years. N = 5 per treatment. Error bars not shown but P values represent significant
treatment affects in a one-way ANOVA for each species. “Unburned” plots were grass dominated
and were like the burned plots that had burned originally in 1973
