|
|
 |
Validation of Heights Derived From Interferometric SAR and LIDAR over the Temperate Forest Site Nationalpark Bayerischer Wald
Thomas Aulinger(1) , Tobias Mette(1)
, Kostas Papathanassiou(1)
, Irena Hajnsek(1)
, Marco Heurich(2)
, and Peter Krzystek(3)
(1)
German Aerospace Center,
P.O.Box 1116,
82234 Wessling,
Germany
(2) Bavarian Forest National Park, Freyunger Strasse 2, 94481 Grafenau, Germany
(3) Munich University of Applied Sciences, Karlstrasse 6, 80333 München, Germany
Abstract
Detecting forest parameters with the aid of remote sensing is getting
more and more important as new sensor technologies allow higher data
quality and better resolutions. The determination of forest height has
been shown to be feasible from both Lidar and interferometric SAR, and
can serve as an input parameter to estimate forest biomass, which is
the most important forest structural parameter. Within the frame of
the HTO project 33-8, different new remote sensing techniques were
tested and analysed at the test site “Nationalpark Bayerischer Wald”,
which is located in south-eastern Germany. Three major forest zones
dominated by Norway Spruce are located in a mountainous region ranging
from 750-1130 m a.s.l.. The stand structures are highly divergent
ranging from natural to semi-natural to disturbed.
The goal of this study was to validate and compare the (1) actual
forest height from ground measurements, (2) forest height retrieved
from Lidar and (3) forest heights from three interferometric SAR
products: X-band DEM (with subtracted ground DEM), X-band DEM (with
subtracted P-band DEM), and model-based height inversion from Pol-
InSAR L-band data.
Lidar and SAR data were acquired with airborne sensors over an area of
2x6 km. Inside this area, ground data were based on two sources: (a)
in situ measurements at 30 smaller test sites between 0.2-6.0 ha,
where stand conditions were homogeneous in species composition, height
distribution and density; (b) 200 points from the standard forest
inventory. A special focus was put upon the relation of the extracted
heights to the h100 as an approximation of the upper canopy height,
and the influence of forest type and density.
Full paper
|
