| In this dissertation, two methods for estimating tree height using Shuttle Radar Topography Mission (SRTM) and ancillary data are developed and tested. Since the SRTM data set is near-global, the methods developed could be applied to large portions of the world, although the scale-up process is beyond the scope of this work. Previous tree height estimation algorithms have been confined to much smaller data sets.; Development of the methods began with validation of the SRTM height data for selected test areas in Southeastern Michigan. Validation statistically characterized the noise in the SRTM height data. The next stage in the development of the first tree height estimation algorithm was the construction of a simplified forward model to relate the observed SRTM data to tree heights, which model relied on a forward model specific to the red pine tree structure. The simplified model was inverted using standard methods to yield tree height as a function of SRTM data. The second algorithm was much more general, not requiring a structure-specific forward model. The resulting tree height estimation algorithm were tested on actual and simulated SRTM data. The simulated data was produced by a simulator from the literature. Two SAR/INSAR simulators, one high-resolution and the other low-resolution, were also developed as a part of this work. Simulated images from the low-resolution simulator were used as an input to further test the first tree height estimation algorithm. |
