forestlas

Python code for generating metrics of forest vertical structure from airborne LiDAR data. This code was developed as part of my PhD (completed in 2016, can be viewed <a href=https://www.researchgate.net/publication/290436021_Assessment_of_forest_canopy_vertical_structure_with_multi-scale_remote_sensing_from_the_plot_to_the_large_area>here) and was developed over the forests of Victoria, Australia. The aim was to develop a suite of metrics that are robust to forest type i.e. can be applied without prior information of forest structure.

There are a number of methods available, check this <a href=https://github.com/philwilkes/forestlas/blob/master/forestlas_intro.ipynb> Jupyter notebook for an introduction. Functions include reading .las files to numpy array, writing to .las as well as a number of methods to dice, slice and tile LiDAR data. The main set of functions found in forestlas.canopyComplexity. These allow you to derive metrics of vertical canopy structure such as Pgap and also estimate number of canopy layers. More information can be found in this paper <a href=https://doi.org/10.1111/2041-210X.12510>Wilkes, P. et al. (2016). Using discrete-return airborne laser scanning to quantify number of canopy strata across diverse forest types. Methods in Ecology and Evolution, 7(6), 700–712.

Funding

This research was funded by the Australian Postgraduate Award, Cooperative Research Centre for Spatial Information under Project 2.07, TERN/AusCover and Commonwealth Scientific and IndustrialResearch Organisation (CSIRO) Postgraduate Scholarship.