Our recent research on mapping individual tree health was published in the Remote Sensing of Environment journal. Building on our previous work (see here), we classified individual tree dieback levels using machine learning and field-measured tree crown dieback as indicators of tree health.
We calculated LiDAR and imaging spectroscopy indices for segmented tree crowns and used as predictor variables in supervised object-oriented classification.
Our results showed that tree health of individual eucalypt trees can be classified with overall accuracy of 81%. Essentially, our model was built from less than 100 field-measured trees and allowed classifying millions of LiDAR-measured trees.
As a final step we overlaid the flooding frequency map derived from time series of Landsat imagery (1986-2011, see here) with the LiDAR-derived tree health. Our findings highlighted that trees located in infrequently flooded areas were most susceptible to dieback.
To learn more: Shendryk, I., M. Broich, M.G. Tulbure, A. McGrath, D. Keith and S. V. Alexandrov (2016). Mapping individual tree health using full-waveform airborne laser scans and imaging spectroscopy: A case study for a floodplain eucalypt forest. Remote Sensing of Environment 187: 202-217. [Download]