Forest transpiration on Kilauea volcano, Hawai'i

In Hawai'i and other oceanic islands, the invasive tree Psidium cattleianum is encroaching on native forest. Previous studies in Hawai'i found native Metrosideros polymorpha forest transpiration was lower than non-native trees, and the sub-canopy tree fern, Cibotium glaucum , was found to be the primary source of stand transpiration in a native forest. In this study, sap flow measurements were used to assess the effects of P. cattleianum invasion, and the role of C. glaucum on forest transpiration at native (Thurston) and invaded (Ola'a) forest sites on the Island of Hawai'i. Site differences in stand structure, successional stage, and environmental conditions, complicated the site comparison. To better assess the effect of P. cattleianum invasion, the relationships derived from measurements at the invaded site were used to estimate transpiration at six additional forest plots with differing levels of invasion by P. cattleianum.;Annual transpiration was 180 mm y-1 at the native site and 317 mm y-1 at the invaded site. Transpiration peaked in July and declined by 40-45% of peak values at the minimum in February. Tree fern transpiration was 8% and 25% of stand transpiration at the native and invaded sites, respectively. Wet conditions occurred a majority of the time and reduced transpiration by 40%. Transpiration as a fraction of potential evaporation was higher at the invaded site (Mann Whitney p<0.0001). Water use per basal area was significantly lower for M. polymorpha than P. cattleianum (p<0.0001). In the additional plot simulation, modelled transpiration increased with P. cattleianum dominance across three plots at one site and declined across the three plots at the other site. If stand basal area remains constant over the course of invasion, transpiration becomes significantly higher when P. cattleianum equals or exceeds 50% of the stand basal area (ANOVA p=0.0002 and 0.0011). Tree transpiration would increase by about 20% if M. polymorpha forest were converted to P. cattleianum cover and basal area remained unchanged. The key issue requiring further study, how stand basal area changes in response to P. cattleianum invasion, must be resolved in order to determine whether invasion increases transpiration.