Validation of Canopy Height Profile methodology for small-footprint full-waveform airborne LiDAR data in a discontinuous canopy environment

Karolina D. Fieber, Ian J. Davenport, Mihai A. Tanase, James M. Ferryman, Robert J. Gurney, Victor M. Becerra, Jeffrey P. Walker, Jorg M. Hackerf

    Research output: Contribution to journalArticlepeer-review

    133 Downloads (Pure)


    A Canopy Height Profile (CHP) procedure presented in Harding et al. (2001) for large footprint LiDAR data was tested in a closed canopy environment as a way of extracting vertical foliage profiles from LiDAR raw-waveform. In this study, an adaptation of this method to small-footprint data has been shown, tested and validated in an Australian sparse canopy forest at plot- and site-level. Further, the methodology itself has been enhanced by implementing a dataset-adjusted reflectance ratio calculation according to Armston et al. (2013) in the processing chain, and tested against a fixed ratio of 0.5 estimated for the laser wavelength of 1550nm. As a by-product of the methodology, effective leaf area index (LAIe) estimates were derived and compared to hemispherical photography-derived values. To assess the influence of LiDAR aggregation area size on the estimates in a sparse canopy environment, LiDAR CHPs and LAIes were generated by aggregating waveforms to plot- and site-level footprints (plot/site-aggregated) as well as in 5m grids (grid-processed). LiDAR profiles were then compared to leaf biomass field profiles generated based on field tree measurements. The correlation between field and LiDAR profiles was very high, with a mean R2 of 0.75 at plot-level and 0.86 at site-level for 55 plots and the corresponding 11 sites. Gridding had almost no impact on the correlation between LiDAR and field profiles (only marginally improvement), nor did the dataset-adjusted reflectance ratio. However, gridding and the dataset-adjusted reflectance ratio were found to improve the correlation between raw-waveform LiDAR and hemispherical photography LAIe estimates, yielding the highest correlations of 0.61 at plot-level and of 0.83 at site-level. This proved the validity of the approach and superiority of dataset-adjusted reflectance ratio of Armston et al. (2013) over a fixed ratio of 0.5 for LAIe estimation, as well as showed the adequacy of small-footprint LiDAR data for LAIe estimation in discontinuous canopy forests.
    Original languageEnglish
    Pages (from-to)144-157
    Number of pages14
    JournalISPRS journal of photogrammetry and remote sensing
    Early online date28 Mar 2015
    Publication statusPublished - 1 Jun 2015


    • full-waveform
    • LiDAR
    • small-footprint
    • canopy height profile (CHP)
    • effective leaf area index (LAIe)
    • discontinuous canopy cover


    Dive into the research topics of 'Validation of Canopy Height Profile methodology for small-footprint full-waveform airborne LiDAR data in a discontinuous canopy environment'. Together they form a unique fingerprint.

    Cite this