lue of the
ed by the
not due to
aracterised
calibration
ectively the
empirically
regression
root mean
ions of the
' than the
egory "real
sing the
9e used for
of the step
/ significant
dates. The
rried out by
tage of this
erencing of
of different
| accuracy
seful results
for end users. For these very important categories the
above described procedure can alternatively be
supported by visual interpretation methods. Herewith the
classification of the changes can be improved by taking
into consideration more complex contextual information
such as texture and shape which is difficult to derive from
operational classification algorithms to be used for this
application. However, this process can be very time
consuming and be therefore expensive, and should be
limited only to "critical" change categories such as
reforestration, deforestation and drastic forest damage
symptoms.
5. SUMMARY AND CONCLUSION
From the results of this study it is to be inferred that the
monitoring method developed within the projects can offer
an effective mean of assistance for forestry planning
tasks. In comparison to conventional methods satellite
remote sensing represents an ideal instrument for
objective and standardised monitoring of environmental
changes. Moreover, compared to other inventory systems
it offers a highly cost efficient alternative. Further
advantages are that satellite images can synoptically
record wider areas, and that using this data it is possible
to observe the same areas repeatedly which permits
monitoring over many years. The latter is essential when
dealing with a sensitive ecosystem such as the Alps.
However, some shortcomings with respect to data
processing methods have to be outlined.
Extremely high geometrical accuracy is one of the most
important prerequisites for successful realisation of
monitoring applications. Geocoding of the images,
therefore, has to meet extremely strict requirements. If
accuracies achieved from precise parametric geocoding
are still not sufficient for monitoring tasks, automated
image matching and co-registration procedures have to
be applied to the multitemporal data in order to obtain
geocoded image data sets with acceptable efficiency.
Efforts in the development of these methods have to be
undertaken to optimise monitoring applications.
For quantitative assessment of changes in
multisensoral/multitemporal remote sensing data sets
radiometric calibration is required. This necessitates an
atmospheric correction of the data. In respect of this
problem many relative and absolute image calibration
methods have been developed which are mainly
restricted to one sensor type. However, in reality different
remote sensing data sets have to be used
complementary. This is due to the fact that a large scale
approach cannot rely only on one sensor type due to
different revisit rates of the satellites and cloud cover
problems. Therefore, one has to concentrate on the full
range of this operational systems, but not simultaneously
in one test area due to data availability and time
constraints in the project execution. Taking this into
consideration it can be stated that methods for calibration
of multisensoral satellite images have to be developed in
order to be more flexible according to data acquisition.
6. REFERENCES
Civco D.L. (1991). Topographic normalization of Landsat
Thematic Mapper digital imagery. Photogrammetric
Engineering and Remote Sensing, Vol. 55, No. 9, pp.
1303-1309.
Colby J.D. (1991). Topographic normalization in rugged
terrain. Photogrammetric Engineering and Remote
Sensing, Vol. 57, No. 5, pp. 531-537.
Haefner H., Itten I. K., Meier E., Meyer P. and Nüesch D.
(1992). Correction of the impact of topography and
atmosphere on Landsat-TM, forest mapping of alpine
regions. Remote Sensing Series of the Department of
Geography, University of Zürich, Vol. 18, Zürich 1992.
Kenneweg, H., Schardt, M., Sagischewski, H., 1996.
Beobachtungen von Waldscháden im Gesamtharz mit
Methoden der Fernerkundung. Berlin, im Selbstverlag,
p.229.
McCormick, N., Kennedy, P., Folving, S., 1995. An
integrated methodology for mapping European forest
ecosystems using satellite remote sensing. Earsel
Advances in Remote Sens., Vol.4, No.3, p. 87-92.
Meyer P., Itten K.I., Kellenberger T., Sandmeier S. and
Sandmeier R. (1993). Radiometric correction of
topographically induced effects on Landsat TM data in an
alpine environment. ISPRS Journal of Photogrammetry
and Remote Sensing, Vol. 48, No. 4, pp. 17-28.
Olsson, H., 1993. Regression functions for multitemporal
relative calibration of thematic mapper data over boreal
forest. Remote Sens. Environ. 46, p.1-25.
Olsson, H., 1995. Reflectance calibration of thematic
mapper data for forest change detection. Int.J.Remote
Sensing, Vol.16, No.1, p. 81-96.
Richter R. (1990). Atmospheric correction of Landsat TM,
MSS and SPOT images: ATCOR user manual, Institut für
Optoelektronik, DLR Oberpfaffenhofen.
Sandmeier, S., 1997. Radiometrische Korrektur des
Topographieeffekts in optischen Satellitenbilddaten -
Vergleich eines semi-empirischen Verfahrens mit einem
physikalisch-basierten Modell. Photogrammetrie
Fernerkundung Geoinformation 1/1997, p.23-32.
Schardt, M., 1987. Expositionsabhängies
Reflexionsverhalten von Waldbestánden im Schwarzwald.
2. DFVLR-Statusseminar: ,Untersuchung und Kartierung
von Waldschäden mit Methoden der Fernerkundung“,
DFVLR Tagungsband, p. 328-347.
Schardt, M., 1990. Verwendbarkeit von Thematic Mapper-
Daten zur Klassifizierung von Baumarten und natürlichen
Altersklassen, Dissertation Forstwissenschaftliche
Fakultät der Unversität Freiburg.
Schardt, M. and Schmitt, U., 1996. Klassifikation des
Waldzustandes für das Bundesland Kärnten mittels
Satellitendaten. Österreichische Zeitschrift für
Vermessung & Geoinformation, Heft 1/96, 84.
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 271
