Using a broad suite of traits, including hydraulics, for robust prediction of species and forest vital rates in space and time

Forests worldwide are facing climate change, necessitating prediction of the responses of species and ecosystems. Much research has applied functional traits for such prediction, yet numerous physiological traits have not been incorporated into the approach, including hydraulics traits that are expected to be crucial for plants to survive under the increased droughts impacting many ecosystems. We measured an extensive suite of leaf and wood traits related to water transport, drought tolerance, gas exchange, nutrient composition and resource economics for Hawaiian montane wet forest (MWF) and lowland dry forest (LDF) in four-hectare forest plots. We tested correlations among traits and relative growth rates (RGR) and mortality rates (m). We hypothesized that: (1) wet and dry forest species' traits would differ as expected from contrasting adaptation; (2) RGR and m would correlate with a number of specific traits, with (3) stronger relationships when controlling for tree size and, thus, (4) RGR and m across species could be explained by trait-based models, and further, that (5) traits could explain variation in vital rates across years varying in climate; and (6) species' distributions across communities statewide.

Species from the MWF and LDF differed strongly on average in numerous traits as hypothesized. On average, the MWF species' traits were associated with adaptation to high soil moisture and nutrient supply and greater shade tolerance, whereas the LDF species' traits were associated with drought tolerance. Thus, MWF species achieved greater heights than LDF species, and lower wood density, larger stomata, higher maximum stomatal conductance and CO₂ assimilation rate, higher carbon discrimination rate, higher phosphorus concentration, higher chlorophyll to nitrogen ratio, lower vein lengths per area, and less negative turgor loss points. The functional traits predicted species' RGRs and m across forests, with stronger relationships when accounting for tree size. Integrating physiological traits provides a novel and robust approach to predicting species' distributions and vital rates. This approach is being expanded across diverse temperate and tropical forests for mechanistic modeling of species responses to climate change.