Presentations
from Statewide Landcover and Trees Study Remote Sensing of
Foliage Projective Cover Workshop
29th November 2005
![]()
Stems, Leaves and Canopies – Relationships within Trees Peter Scarth (SLATS)
This presents some simple theoretical models that attempt to link some of the more commonly used canopy metrics. Field measurements were used to calibrate and test these models over a range of communities with good results.
Derivation of FPC and Related Canopy Attributes using LIDAR John Armston (SLATS)
The derivation and validation of FPC from airborne LIDAR will be presented. These data are being used to calibrate and validate products derived from Landsat and MODIS.
Analysis of the SLATS Landsat FPC Time-series Tim Danaher (SLATS)
The latest method for mapping woody FPC using a time series of Landsat measurements will be described. Future plans for monitoring longer term changes using the FPC time series will be discussed.
The current method of using multiple linear regression between field measurements and Landsat data to predict FPC will be presented. Different regression modelling strategies were explored to maximise predictive accuracy however representative field data was the most important consideration.
A Bayesian model is being developed with the objective of incorporating the different surrogate measures for FPC; basal area, LIDAR, field transect estimates and Landsat imagery into the one model. A Bayesian approach is the best way to incorporate the uncertainties of all parameter estimates into the one model.
A technique developed by CSIRO
researchers that uses time series imagery to separate woody from herbaceous
vegetation and how it can be improved using additional information offered by
the MODIS sensor will be discussed. Early results for the
Page created 17 Jan
2006