Leaf dark respiration (R) consumes significant proportion of photosynthetic carbon uptake, and has a profound influence on atmospheric CO2 concentrations. R increases exponentially with temperature, so understanding the temperature response of R is critical for predicting responses of forest ecosystem carbon balance to climate warming. Despite the significance of tropical forests for the global carbon balance, empirical data on R and its temperature sensitivity (Q10, the proportional increase in R with a 10°C rise in temperature) are scarce for tropical trees and lianas. This study aimed to fill this gap by quantifying R and its relationship to leaf temperature in situ for tropical trees and lianas in Panama, using a tower crane to access fully exposed, top canopy leaves. We tested the hypothesis that R and Q10 will vary considerably among 26 species of tree and liana in relation to their leaf functional traits, showing significant differences among plant functional types (PFTs) and between growth forms.
Results/Conclusions
Respiration standardized at 25°C (R25) was significantly higher for lianas than for trees (14.5 vs. 11.3 nmol g-1s-1; P<0.01), and higher in early successional than late successional species (14.4 vs. 9.0 nmol g-1s-1; P<0.01). R25 values for both trees and lianas were among the highest when compared to published data on tropical canopy leaves. Interspecific variation in R25 was best explained by variation in leaf mass per unit area, photosynthetic capacity and foliar content of nitrogen, phosphorus and potassium. Q10 values varied widely (range 1.4-3.5), but not systematically among species and PFTs and could not be predicted from other leaf traits. Theoretical models assuming temperature acclimation have predicted Q10 values in the tropics to be lower than the often assumed standard of 2.0, but instead we found the mean Q10 of trees (2.4) and lianas (2.2) to be significantly higher than 2.0. These results suggest that current models of carbon fluxes from tropical forests underestimate carbon loss from foliar respiration and underestimate the additional carbon lost with climate warming. More data on temperature responses of R for tropical forest ecosystem are urgently needed to adequately predict tropical forest net primary production under the warmer climate conditions predicted in near future.