Biology Reference
In-Depth Information
over time in these ecosystems (Hughes et al. 1991). A resurvey of these sites in
October 2007, unburned since 1973, confirms the lack of recovery of native species
(D'Antonio et al., unpublished). The prefire dominant tree,
M. polymorpha
, grows
very slowly as a seedling and it along with
L. tameiameia
, the dominant shrub, can-
not tolerate competition from the grasses (Hughes and Vitousek 1992). In addition,
they would readily be killed by the next, inevitable fire. For these reasons, it was
decided that any restoration efforts toward these species would be futile.
Surveys of burned sites in the coastal lowlands (Tunison et. al. 1994) and our own
and other observations (Tunison et al. 1995) suggest that there are fire-tolerant species
in the Hawaiian flora that can be found in these ecosystems. Their scarcity throughout
these sites is considered in part to be a function of previously high goat browsing,
which is known to have reduced many woody species (Mueller-Dombois and Spatz
1975; Loope and Scowcroft 1985). Only one of these native species,
D. viscosa
, was
common in burned sites (Hughes et al. 1991; D'Antonio et al. 2000). However, along
the roads leading to these sites, burned individuals of planted trees showed regenera-
tion from seed and stumps (Tunison et al. 1995, and personal observation). Also, the
occurrence of other native dry forest and shrubland species in fire-prone habitats
higher on Mauna Loa and Mauna Kea volcanoes suggested that there are fire-tolerant
species in the Hawaiian flora that could grow at these sites. We hypothesized that to
persist successfully in these habitats, a native species would have to tolerate growing
with the dense exotic grasses that form an almost continuous canopy within 2-3 years
after fire, and they must have mechanisms of resilience to regenerate rapidly after the
inevitable fire. The success of the introduced grasses in invading intact woodland
(D'Antonio et al. 2001a) suggested that no matter what plant communities exist on the
sites, they are likely to offer little resistance to grass invasion. Thus, we designed a
revegetation program that focused on developing fire-resilient native species assem-
blages that can coexist with grasses and persist with the new disturbance regime. It was
unlikely that resistance could be strong enough to significantly dampen grass invasion
in the near term especially since
S. condensatum
regenerates rapidly from root crowns
after fire and can reduce the growth of native species (D'Antonio et al. 1998).
Our approach to the management of these sites is appropriately termed ecosys-
tem rehabilitation (Bradshaw 1997). Ecosystem restoration implies the return of the
composition and functioning of a system to the predisturbance state (Jordan 2003).
Efforts to manage plant invasions in these sites focus on the process of resilience to
restore ecosystem functions to the sites, with little understanding of whether resil-
ience was a feature of pre exotic-grass ecosystems. It is possible that we are restor-
ing aspects of composition that occurred prior to the extensive goat grazing that
occurred before botanical records of the area were kept, but we do not know this.
7.2.4
Current Approaches to Management
In January 1993, we conducted our first experimental burn within Hawaii Volcanoes
National Park with a goal of reestablishing more native species by reburning an
