Operational Use of Airborne Laser Scanning for Forestry Applications in Complex Mountainous Terrain

Graz

2007

S. 23-30

4 Abb., 23 Lit. Ang.

Artikel aus einer ZeitschriftUnselbständiges Werk

200147223

Today, airborne laser scanning (ALS) is the standard method for detailed topographic data acquisition, which can complement, or partly replace, other existing geo-data acquisition technologies, and open up new exciting areas of applications. With hydrology as the main driving force extensive ALS fl ight campaigns have been carried out since the disastrous fl oods 2002 in Austria and therefore, ALS data are available for large areas today. For forestry applications ALS is currently one of the most promising remote sensing techniques for quantitative retrieval of forest parameters. While ALS has reached an operational status for mapping of boreal forests, its large area application over mountainous environments is lacking behind. During the last years a few studies have been published, that successfully use small footprint ALS data for forestry applications in mountainous areas. However, it needs to be recognised that these studies were limited to small test sites, the reference data were measured during extensive fi eld campaigns, and the used ALS data were acquired with homogenous system confi gurations. For operational applications of large areas, ALS data are normally acquired during several fl ight campaigns with varying system confi gurations. Thus, the algorithms used to derive forest attributes have to handle these different conditions. This paper is a short summary of two studies (Hollaus et al., 2006a; Hollaus et al., 2006b), which have investigated the applicability of ALS data for a 128 km2 test site in the western part of the Austrian Alps. On the basis of the derivation of canopy heights and of the estimation of stem volume it is discussed what can be reached with current ALS data and already available algorithms. The ALS data were acquired within the framework of an operational DTM mapping activity. The acquisition of the ALS data took place in two different seasons with varying ALS system confi gurations. For processing the ALS data only methods that are already implemented in commercially available software packages are used. These two studies have shown that for large area applications the applied ALS preprocessing algorithm and the hierarchic robust fi ltering technique are adequate methods to derive high accurate topographic models for this complex environment. The validation of the ALS derived canopy height products (e.g. single tree heights and Loreyęs mean heights) with operationally used forest inventory data and ground control points have shown good correlations using (1) three dimensional fi rst-echo points and (2) a grid-based canopy height model. However, some inaccuracies of the canopy heights arose in areas covered by homogeneous meadows with high grass or scrubs due to the technical limitations of current ALS systems. For the estimation of stem volume a simple linear approach has been used to combine forest inventory and ALS data. The validation of the estimated stem volumes has shown that even for ALS data with varying properties (point densities, different acquisition times e.g. leaf-off and leaf-on) reliable results of high accuracies could be achieved. Concluding, it can be stated that ALS has now reached the maturity not only for topographic data acquisition but also for operational forestry applications within complex alpine environments.