Digital photogrammetric mapping of canopy height in a radiata pine plantation
thesisposted on 2023-05-27, 14:42 authored by Wijanarto, AB
Tree height is a crucial stand characteristic in forest inventories and has been used widely to estimate timber volume, to monitor growth rate, and as an indicator of site quality. Current tree heighting relies on field measurements, with sampling rates, which are typically less than one percent. Low sampling rates produce high variation in measured parameters and reduce statistical reliability. Digital photogrammetry provides an opportunity to automate the measurement of tree heights and to generate digital models of forest canopies. A sufficiently dense canopy model, with tree height data at nearly 100% sampling rates, has the potential to improve significantly the reliability of strategic and operational forest inventory. This study investigates the utility of digital photogrammetric techniques to map tree height within radiata pine plantations. A digital photogrammetric workstation, the Leica Helava DPW 770, was used to generate digital canopy models of both closed canopy and thinned stands. The performance of the image-matching algorithm was assessed using scanned aerial photos of different film types and scales, including colour, black and white, colour infrared, and black and white infrared, at 1:5,000, 1:10,000, and 1:15,090 scales. Canopy height data were obtained by subtracting a DTM from the photogrammetric data. Various filtering strategies were used to process the canopy height data including median, mean tallest tree, adaptive wavelet, adaptive Fourier filtering, and NDVI-thresholding techniques. A method for improving automatically derived tree height using a limited number of manually measured tree heights using a DPW was also developed. The reliability of the derived data was assessed by comparing it with a reference DCM produced using an analytical stereoplotter at a lOm grid, and field measurements. The study demonstrates that tree heights in a coniferous forest can be mapped using digital photogrammetric methods and appropriate post-processing. Filtered digital photogrammetric data are shown comparable to analytical stereoplotter data and field-measured data, particularly for dense stand. With the efficiency and high sampling rates afforded by digital photogrammetric techniques, this study demonstrates the potential application of photogrammetrically measured tree height data to forest inventory.
Rights statementCopyright 2000 the author - The University is continuing to endeavour to trace the copyright owner(s) and in the meantime this item has been reproduced here in good faith. We would be pleased to hear from the copyright owner(s) Thesis (PhD)--University of Tasmania, 2000. Includes bibliographical references