email:
jhomeie <at> gwdg.de
Faculty of Resource Management
University of Applied Sciences and Arts (HAWK)
37077 Göttingen
Germany
Individual:
Oliver Limberger
Contact:
email:
limbergo <at> staff.uni-marburg.de
Faculty of Geography
Deutschhausstraße 10
Philipps University of Marburg
Laboratory for Climatology and Remote Sensing
35032 Marburg
Germany
Individual:
Jörg Bendix
Contact:
email:
bendix <at> staff.uni-marburg.de
Faculty of Geography
Deutschhausstraße 10
Philipps University of Marburg
Laboratory for Climatology and Remote Sensing
35032 Marburg
Germany
The co-evolutionary arms race between herbivores and plants forces plants to evolve protection strategies that reduce the palatability of the plant modules attacked by the herbivores. These characteristics of traits have consequences for both the survival of plant individuals and the composition of plant communities. Thus, correlating traits of for instance leaves with herbivory is an important step toward understanding the dynamics of plant populations and communities. Traits can either be measured using conventional lab methods or recently developed spectral sensing techniques. We examined whether leaf traits of trees are related to herbivory in a multispecies approach. Furthermore, we explored whether leaf traits characterized by spectral sensing provide similar relations to herbivory as lab-based leaf traits. We established nine 1-ha square plots evenly distributed over three different forest types in Ecuadorian tropical montane rainforests where we estimated herbivory as the leaf area loss (in square centimeters) of 20 (±5) leaves sampled from the canopies of 380 tree individuals belonging to 51 tree species (7 ± 1 individuals/species) using lab- and spectral-sensing-based methods. For each methodological approach, we ran 100 linear mixed-effects models with all respective leaf traits as predictor and herbivory as response variables for data subsets containing one randomly selected tree individual of each species to estimate the range of the regression coefficients for each trait. Automated stepwise backward selections determined the frequency of each trait having an important influence on herbivory. We found no clear relations between leaf traits and herbivory for neither lab- nor spectral-sensing-based traits. A nested variance component analysis demonstrated that the observed variability was mainly due to the variation in trait concentrations between tree individuals of a species. Our results suggest that snapshot data lead to a mismatch between herbivory and the concentrations of traits during the peak of herbivory. Another explanation could be that environmental conditions or processes along the food web are more important in structuring herbivory than leaf traits.
