Assessing Ecosystem Productivity Through the Integration of Micrometeorology and Optical Remote Sensing

• Author / Creator

  Castro, Saulo M.

• The main goal of this thesis is to contribute to the growing body of scientific work by exploring mechanisms of ecosystem productivity through the integration of remote sensing and micrometeorological data. This was done throughout by (i) identifying the environmental mechanisms affecting productivity in tropical dry forest during normal and drought conditions; (ii) assessing the use of proximal PRI sensors as a proxy of photosynthetic efficiency and use towards a complete remote sensing derived measure of ecosystem productivity; and (iii) evaluating the impact of temporal aggregation and phenology on LUE model parametrization and ecosystem productivity in two deciduous forests.
  Chapter 2 uses information on seasonal phenology and carbon fluxes derived from optical remote sensors and eddy covariance to identify key mechanisms of ecosystem productivity under normal and drought seasonal precipitation regimes. Precipitation was identified as the trigger for the initiation of the phenological cycle. Results also showed a substantial decrease in productivity, net ecosystem exchange, and respiration due to drought, but the Tropical Dry Forest remained a net carbon sink over the season. Relative importance analysis identified latent heat as the principal controlling factor of TDF productivity. However, during drought, soil moisture became the limiting variable of productivity.
  Chapter 3 evaluates continuous data collected from recently available autonomous PRI sensors as a proxy of light use efficiency (LUE) in an aspen (Populus tremuloides) forest. Quantum yield values were calculated from eddy covariance data and used to assess the ability for PRI measurements to track changes in canopy light use efficiency. Spectrometer measurements were also used to validate and calibrate the sensor’s PRI signal. Uncalibrated PRI data was unable to resolve diurnal patterns and resulted in an overestimation of LUE and productivity. An offline diurnal calibration procedure was proposed to resolve diurnal and seasonal LUE trends. Calibrated PRI data was then used to derive productivity through a LUE model parameterized solely by remote sensing data. Modeled productivity significantly correlated with measured GPP values from an eddy covariance system.
  Chapter 4 explores the effect of temporal aggregation and phenology on LUE model variables and productivity in a tropical dry forest (TDF) and deciduous boreal forest (DBF). This was done as part of developing data management protocols for remote sensing data integration. Results showed the different impacts of aggregation in seasonal analysis than when data was divided by phenological cycle, for some variables (e.g. fAPAR). Results suggest that temporal aggregation can significantly impact LUE model accuracy and should be considered as we explore the proper protocols for optical and flux data integration. Relative importance results showed differences in dominant variables of productivity between seasonal and phenological analysis. Findings confirm that physiological and structural contributions of the LUE model change between vegetation, environmental condition and phenology.

• Subjects / Keywords

  • Eddy Covariance
  • Remote Sensing
  • Deciduous Boreal Forest
  • Tropical Dry Forest
  • Phenology
  • LUE Model

• Graduation date

  Spring 2020

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-takm-cr31

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.