Temporal dynamics and leaf trait variability in Neotropical dry forests

• Author / Creator

  Hesketh, Michael S

• This thesis explores the variability of leaf traits resulting from changes in season, ecosystem successional stage, and site characteristics. In chapter two, I present a review of the use of remote sensing analysis for the evaluation of Neotropical dry forests. Here, I stress the conclusion, drawn from studies on land cover characterization, biodiversity assessment, and evaluation of forest structural characteristics, that addressing temporal variability in spectral properties is an essential element in the monitoring of these ecosystems. Chapter three describes the effect of wet-dry seasonality on spectral classification of tree and liana species. Highly accurate classification (> 80%) was possible using data from either the wet or dry season. However, this accuracy decreased by a factor of ten when data from the wet season was classified using an algorithm trained on the dry, or vice versa. I also address the potential creation of a spectral taxonomy of species, but found that any clustering based on spectral properties resulted in markedly different arrangements in the wet and dry seasons. In chapter 4, I address the variation present in both physical and spectral leaf traits according to changes in forest successional stage at dry forest sites in Mexico and Costa Rica. I found significant differences in leaf traits between successional stages, but more strongly so in Costa Rica. This variability deceased the accuracy of spectral classification of tree species by a factor of four when classifying data using an algorithm trained on a different successional stage. Chapter 5 shows the influence of seasonality and succession on trait variability in Mexico. Differences in leaf traits between successional stages were found to be greater during the dry season, but were sufficient in both seasons to negatively influence spectral classification of tree species. Throughout this thesis, I show clear and unambiguous evidence of the variability of key physical and spectral leaf properties over various temporal scales, with the conclusion that an understanding of this variability must play a central role in the establishment of monitoring techniques for dry forests.

• Subjects / Keywords

  • Remote sensing
  • Temporal variation
  • Tropical dry forests

• Graduation date

  Spring 2014

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R33J3973R

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.