Hyperspectral remote sensing of conifer biochemistry in the Greater Victoria Watershed District, British Columbia
Date
2008-08-14T22:55:02Z
Authors
McDonald, Sarah Elizabeth Alice
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The biochemical composition of conifer foliage in the Greater Victoria Watershed District (GVWD), Vancouver Island, Canada, was examined using hyperspectral remote sensing data. Imagery acquired from the airborne sensor Advanced Visible/Infrared Imaging Spectrometer (AVIRIS) was compared to sampled foliar chemical measurements to provide regional maps of biochemical distribution. The biochemical concentrations of nitrogen, chlorophyll and moisture derived from AVIRIS data were analyzed to provide an analysis of the forest canopy, comprised predominantly of Douglas-fir.
The AVIRIS data were preprocessed to correct for atmospheric and geometric distortion, degradation, and noise inherent in the data in order to properly represent the forest canopy at the time of image acquisition. The AVIRIS data were used to investigate the relationship between the reflectance, absorbance and derivative values present in the imagery corresponding to the sampled chemical data. A total of 29 plots were used in a partial least squares regression analysis to analyze the relationship between the data sets to extract chemical constituents in the forest canopy. Nitrogen and total chlorophyll models have r2
values of 0.73 and 0.68 respectively. Due to the complexity of moisture interaction with hyperspectral data, regression models were unable to be computed for the AVIRIS data over the GVWD. Regression models were then applied to the entire AVIRIS dataset for regional mapping of the canopy biochemistry. The distribution of nitrogen and total chlorophyll in the forested areas of the GVWD was mapped.
Description
Keywords
Forests, Forestry, Remote sensing, British Columbia, Victoria