Modeling carbon-water-vegetation dynamics using remote sensing and climate data

• Author / Creator

  Jahan, Nasreen

• Carbon and water fluxes are essential components of biospheric processes which directly or indirectly influence climate, surface energy balance, hydrologic processes and hence influence the vegetation productivity, distribution and characteristics. In this research, promising techniques for simulating carbon (Gross primary production) and water fluxes (soil moisture and evapotranspiration) were developed using remotely sensed data to overcome our dependence on meteorological data which are often not available with sufficient accuracy for regional scale climate studies.

  The temporal responses of vegetation to climate were assessed using Artificial Neural Network (ANN) and two remotely sensed vegetation indices (VIs), normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI). The results demonstrated a promising potential for monitoring the patterns of terrestrial vegetation productivity from VIs and climate variables in a boreal mixedwood forest of western Canada.

  Next, the potential of using the newly available, quad-polarized, RADARSAT-2 synthetic Aperture Radar (SAR) data in retrieving near surface soil moisture in the Canadian Prairies was examined. Ten Radarsat-2 images have been acquired over the Paddle River Basin (PRB), Alberta, Canada and 2250 soil samples have been collected from 9 different sites within the same basin on those 10 days. Soil moisture was retrieved using the regressions, theoretical Integral Equation model (IEM) and two machine learning techniques: ANN and Support Vector Machine. The results show that combined radar and optical satellite data from the Moderate Resolution Imaging Spectroradiometer (MODIS) can be used to retrieve near surface soil moisture accurately.

  Finally, algorithms were developed to model vegetation carbon flux (Gross Primary Production, GPP) and evapotranspiration (ET) for the coniferous and deciduous forests using solely remote sensing data from MODIS. The remotely sensed GPP (R-GPP) and ET model (R-ET) were parameterized and validated using the observed data derived from the eddy covariance towers located in north-eastern USA. The proposed models attempt to exclude the use of ground data or climate data as model input by utilizing MODIS ecosystem and radiation budget variables. Considering the trade-off between sophisticated modeling approach and the uncertainties in obtaining regional scale reliable climate data, it can be concluded that these simple models (R-GPP and R-ET) are practical and promising in providing valuable inputs for regional scale hydrological modeling and water resource planning and managements.

• Subjects / Keywords

  • Gross Primary Production
  • Remote Sensing
  • Evapotranspiration

• Graduation date

  Spring 2012

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R3260J

• 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.