| Abstract |
Climate niche approaches predict widespread vulnerability of urban tree species in cities, but the accuracy of these predictions has limitations given that many species can survive in cities outside their native distributions. Predictions based on functional traits offer an alternative approach, through quantifying species’ physiological tolerances, to select stress-tolerant urban tree species for future climates. We aimed to investigate if climate niche-based approaches or functional traits can predict patterns of growth and survival for 61 urban tree species planted in common garden experiments in two Australian cities, Sydney and Melbourne. Climate niche was estimated using global occurrence records of tree species and mean and extreme rainfall and temperature variables. The species’ thermal and precipitation safety margins were calculated as the difference between the species’ climate niche limits and each city’s climate. Functional traits related to drought tolerance, structural allocation, water use efficiency, stomatal anatomy, and conductance were measured for a subset of 18–26 species from both cities. Climate safety margins did not predict species’ growth or mortality during the first two years after planting. Species from drier climates (mean annual precipitation of ∼700 mm) had higher growth, while species from warmer/tropical climates (mean annual temperature >21 °C) exhibited higher mortality in both Sydney and Melbourne. High structural allocation (i.e., high leaf dry matter content) and higher water use efficiency (i.e., less negative leaf carbon stable isotope composition) were associated with low growth and mortality across species. Additionally, species with lower stomatal density and larger stomata had higher growth. Drought tolerance (i.e., more negative leaf water potential for turgor loss point) was linked to low mortality across species in both cities. Functional traits were better predictors of species’ growth and survival than climate safety margins for newly planted urban trees.
|
