FORESTRY, REMOTE SENSING AND CATCHMENT MANAGEMENT: OPTICAL IMAGERY FOR LONG TERM TREE HEIGHT MAPPING.

摘要

With the introduction of the Water Framework Directive in 2000 the demand for high quality information for catchment management has never been as pressing. In areas suffering from acidification, where the potential for the filtering of atmospheric pollutants by forestry increases with tree height it is vital to know not only the proportion of forestry within a given catchment but also an indication of its scavenging potential. Existing science has used optical satellite imagery to model tree height from forest establishment to canopy closure. This paper develops this work to produce a time series of forest height change from 1989 - 2005 using a series of five radiometrically calibrated optical images and tree height data from a single year (2003). The success of the methodology is validated against field measured data from two other years 1995, 2001 showing a mean difference from measured tree height of 1.43m and a standard deviation of 1.04. This compares well with both the error found between LiDAR measured field height and field measurements, and indeed the inherent measurement error of field measurements themselves. The success of this technique provides i) catchment managers a regional scale tool to relate long term changes in the forest filtering potential to its chemical and biological factors ii) the capability to monitor forest establishment in years where no ground truth data is available by using paired ground truth data and imagery from another year and a radiometrically corrected optical image from the year of interest.