Title of Project: Direct and indirect effects of key
alien species in a Hawaiian dryland forest community
Sponsoring Organization: National Tropical Botanical
Garden, 3530 Papalina Road, Kalaheo, HI 96741 (tel: 808-332-7324)
Investigators: David H. Lorence, PI, National
Tropical Botanical Garden, Kalaheo, HI; Stephen G. Weller, Co-PI, University of
California at Irvine; Robert J. Cabin, Co-PI, USDA Forest Service, Hilo, HI.
Funding Agencies: National Science Foundation:
Conservation and Restoration Biology Program (Grant DEB-9610414)
Location of Project: Ka`upulehu Preserve, North Kona
District, Island of Hawaii
Period of Project: August 15, 1997-July 31, 2000
Tropical dryland forests are one of the most threatened and endangered ecosystems. Over 90 percent of dryland forests in Hawaii have been destroyed by human activities, burning, grazing, and alien plant species, and the remaining 10 percent are heavily degraded. One of the largest remaining areas of dryland forest is in the North Kona region on the island of Hawai`i. The major threats to the dryland forest here are the alien fountain grass (Pennisetum setaceum, an African bunch grass), rodents, cattle, and feral goats. Fountain grass poses a severe threat because it carries fire into areas not usually prone to fires, increases the likelihood of fires, and interferes with native plant regeneration by killing seedlings and adults.
The site for this project is the Ka`upulehu Preserve, a 2.3 hectare (5.8 acre) enclosure fenced since the early 1950’s to prevent damage from grazing cattle and goats. The Preserve is located at 622-640 m elevation with an average annual rainfall of about 500 mm. It consists of remnant dryland forest dominated by lama (Diospyros sandwicensis) and mamane (Sophora chrysophylla) over rough a`a lava. Extensive clearing for cattle grazing has resulted in the loss of many endemic Hawaiian plant species, and the National Tropical Botanical Garden (NTBG) began to manage the site in the early 1970’s to protect the plant species that persisted in the area. Although fenced and protected from agricultural conversion, the species in the Preserve continue to be threatened by the increased dominance of alien plant species and the presence of rats which consume fruit and seeds of the native species and prevent natural regeneration.
A floristic survey undertaken in 1995 by David Lorence and Tim Flynn of the NTBG revealed that of the 67 vascular plant taxa occurring on the Preserve, 35 were native, 28 were naturalized alien species, 2 were planted alien trees, and 2 were planted native trees. The survey also revealed that few of the native species were successfully reproducing, primarily as a result of the dense, smothering cover of fountain grass (a fire resistant species), and the consumption of seeds and fruits by rats. Two native tree species which had been categorized as common in 1973, maua (Xylosoma hawaiiense) and kolea (Myrsine lanaiense) had nearly disappeared by 1995. The historically documented high levels of diversity and the exclusion of grazing animals from the site make the Ka`upulehu Preserve an ideal site to study techniques to restore dryland forest habitat.
In 1996-97, approximately 96 percent of the fountain grass cover had been removed throughout the preserve by first weed-whacking and then repeatedly applying grass specific herbicide (Fusillade) over a period of 20 months (Figures 1-3).
The objectives of this project are to test methodology for restoring and regenerating Hawaiian dryland forests by controlling the threats of fountain grass and other alien weeds, fire, and feral ungulates, coupled with supplemental outplanting of native seeds and/or established plants subjected to various water, shad, and weeding treatments. Regeneration of native and alien seedlings was studied following fountain grass removal at the site. After the removal of the fountain grass a total of 7680 seeds of six dryland forest tree and shrub species were planted into 64, one-square meter plots. These species included: Colubrina oppositifolia, Kokia drynarioides (both federally listed as endangered), Santalum paniculatum var. paniculatum (rare), Diospyros sandwicensis, Dodonaea viscose, and Sophora chrysopylla (all common).
The experimental plots were establhished under factorial combinations of sub-canopy shad/inter-canopy full sunlight, weeding of emerging non-native species/ no weeding, and supplemental water/ambient water treatments. The supplemental water treatment consisted of 5.3 gallons per plot (20 liters) three times a week for the first six months of the experiment and once a week thereafter. In February 2000, we destructively harvested all of the plants within these plots (except for saving as many of the endangered species as possibly) and are presently drying and weighing them to get biomass data.
We found significantly more native species in the shade vs. sun plots (7.0 and 2.2 plants per plot, respectively, p<0.01, t-test). Conversely, there were slightly but not significantly fewer alien species in the shade plots (8.3 and 11.3, respectively, p>0.71). Surprisingly, we found slightly more native species in the non-weeded vs. weeded plots (5.4 and 3.8 plants per plot, respectively, p>0.38). Despite initially promising germination of the seeded species in the ambient water plots, a sever drought in the summer of 1998 eventually led to virtually 100% mortality of these seedlings.
Three of the six native species produced considerable numbers of seedlings surviving to one year after sowing (there were 37, 34, and 246 individuals of Colubrina, Kokia, and Sophora, respectively), while seeds from the remaining three species virtually or completely failed to germinate. Although there were more surviving plants of all native species in the shade vs. sun plots, there were no significant differences in the height, width, or number of leaves of plants growing within these two treatments. There were also fewer surviving individuals of each species in the seeded vs. non-weeded plots, but with the exception of Sophora (mean height was 9.0 and 6.7 inches, mean number of leaves 17.1 and 13.3 in the weeded and non-weeded plots, respectively, p<0.05 in each case), there were no significant plant size differences between the seeded and non-weeded treatments.
Our results to date suggest that if adequate water and shade are provided, al least some native Hawaiian dryland forest species may be successfully regenerated using in situ direct-seeding techniques. Weeding emerging alien species in general did not appear to significantly affect the growth or survival of the natives, although it is important to point out that this experiment began under initially wee-free conditions. Nevertheless, we found this result surprising, given the aggressiveness of some of the alien species that invaded the plots (e.g. lantana (Lantana camara) and milkweed (Asciepias physocarpa) and the alleged competitive inferiority of native Hawaiian species. Indeed, in some plots we observed apparently healthy kauila and koki`o individuals surrounded by dense alien species foliage. This result suggests that given the right micro-environmental conditions and care, successful regeneration of rare native dry forest species can occur, and Hawaii’s remaining dry forest fragments need not be gradually lost to senescence.
The National Science Foundation grant is also helping fund dryland forest restoration efforts on a larger scale in an adjacent 70-acre parcel situated just below the Preserve. This larger parcel, which also supports remnant dryland forest invaded by fountain grass, was recently fenced to exclude cattle and feral goats. An experiment was designed to test additional methods of fountain grass control and restoration of native species. The Kamehameha Schools-Bishop Estate, U.S. Fish and Wildlife Service, USDA Forest Service-Institute of Pacific Islands Forestry, Hawaii Forest Industry Association, and NTBG are partners in the joint, collaborative effort.