The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008
426
Figure 4-5 RGB composite image of band 7, 4, 3 on August 8 th
Figure 4-6 Suspected fire points on NDVI image of August 8 th
Figure 4-7 RGB composite image of band 7, 4, 3
on August 30 th
Though some detected high temperature points were not forest
fire points, however, the real forest fire points have been
detected with our method. The result shows that it is a good
choice to monitor forest fires with MODIS.
5. CONCLUSIONS AND DISCUSSIONS
In our experiment, LST is retrieved with a statistical model.
Though the accuracy of LST is to be improved, the method is
an effective in operation work. However, as time passed, the
statistical model will be refined with more prior knowledge
used in the regression. So does the SI model.
Since the physical model for LST retrieval has a better
precision. Our next study will aims on combining the physical
model and the statistical model together so as to get a precise
and effective inversion way.
Though channel 7 was not designed for forest fire detection,
higher temperature forest fire points show higher reflectivity in
this channel. Combined with channel 7 and 4 middle infrared
bands (band 20-23) that designed for forest fire, higher
temperature forest fire points are detected with a higher
precision.
It can be concluded that MODIS is an ideal data source in
agriculture drought monitoring and forest fire monitoring.
Furthermore, in order to get more objective information, it is a
better way to combine remote sensing data and meteorologic
operational data together.
REFERENCES
Becker F, Li Zhaoliang, 1990. Towards a Local SplitWindow
Method Over land surface. Int. J. Remote Sens., 3: 369-393
Becker F, 1987. The impact of spectral emissivity on the
measurement of land surface temperature from a satellite. Int. J.
Remote Sens, 8 (10): 1509-1522.
CooperD I, Asrar G, 1989. Evaluating atmospheric correction
models for retrieving surface temperatures from the AVHRR
over a tall grass prairie. Remote Sens. Environ, 27: 93-102.