Restoration of native plant communities in a Hawaiian dry lowland ecosystem dominated by the invasive grass Megathyrsus maximus

Abstract

Nomenclature Wagner et al. (1999) for all native Hawaiian species USDA ARS-GRIN (www.ars-grin.gov/ npgs/) for M. maximus Abstract Questions: How does a highly degraded Hawaiian tropical dry lowland ecosystem dominated by the non-native invasive Megathyrsus maximus (guinea grass) respond to different restoration treatments (three native species outplanting treatments; four native broadcast seed treatments)? What effect do restoration treatments have on invasive and native species groundcover, biomass and physiological activity, and volumetric soil water content?Location: Waianae Kai Forest Reserve, Island of Oahu, Hawaii, USA. Methods:The invasive grass M. maximus was suppressed by initial mowing and pre-and post-planting herbicide applications. Native species were added in three outplant and four broadcast seed treatments in a complete randomized block design. Native species and M. maximus growth and ecophysiology, and volumetric soil water content were quantified for 8 mo following treatment establishment.Results: Native species outplant survival ranged from 38% to 67%. Cover of M. maximus was significantly reduced in all outplant treatments compared with control and treated control (mowing and herbicide without native species additions), but did not differ across outplant treatments. Of the native species, Dodonaea viscosa biomass was higher than Cordia subcordata, while other native species did not differ. Maximum photosynthetic rates (A max ) did not differ across species in July. However, in August (drier period), M. maximus exhibited lower A max than all native species except T. populnea, indicating adaptive dormancy during drought. Broadcast seeding with native species was not an effective restoration treatment, as field germination ranged from 0.5% to 2.3%.Conclusions: Ecological restoration of highly invaded Hawaiian tropical dry lowland ecosystems can be mediated through aggressive invasive species suppression and native species outplanting. Recommendations for restoration include initial removal of invasive grasses, adaptive suppression of grasses postoutplanting, and utilization of diverse native species assemblages that are ecophysiologically adapted to local conditions and competitive with M. maximus.

Regeneration response of Brazilian Atlantic Forest woody species to four years of Megathyrsus maximus removal

“…Ammondt et al (2013) suggest the presence of dry season dormancy in this species, and an analogous mechanism for wet season shade tolerance may support the persistence of Guinea-grass in the understory even for over 20 years. Seasonal fluctuation in canopy cover was also recorded for the adjacent mature forest fragment (Bianchini et al, 2001).…”

The Contemporary Scale and Context of Wildfire in Hawai‘i

“…comm.). Native plants used in res to ra tion also vary in their abil ity to re cover from fu ture fi res (Ainsworth and Kauffman 2009, Loh et al 2009, Ammondt and Litton 2013, as well as their pro pen sity to burn. The com monly used shrub spe cies Dodonaea viscosa (L.) Jacq., for in stance, is ex ten sively planted at dry for est sites but has been found to have ex tremely low live fuel mois ture (Ellsworth 2012).…”

Future directions for forest restoration in Hawai‘i

“…In some instances where there has been good weed control and subsequent rainfall, native plants have been successfully restored by direct seedling (Brooks et al 2009;Loh et al 2004;McDaniel et al 2011), but in many other trials direct seeding has not worked because of drought, poor seed germination, and weed competition (Ammondt et al 2013). Direct seeding may be the only way to quickly restore large areas to native vegetation after wildfires, as most native seedlings take months to years to propagate.…”