In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Voi. XXXVIII, Part 7B
202
Figure 3. Vegetation changes detected by SAVI2 and NDVI
Among all other indices, only SAVI2 showed a reduction in the
fraction of vegetation. The rest of the indices were also returned
mixed pixels.
Next, the drought assessment was examined using three indices:
VSWI, PDI and MPDI. As table 1 reveals in 2002, annual
rainfall increased sharply. The results of PDI approved that the
drought severity decreased. Since the PDI is based on soil
moisture and the reflectance of the targets in the Red and NIR
bands, it is suitable for meteorological drought monitoring.
On the other hand, the results of VSWI, also, showed mixed
pixels. This is due to the fact that this index is based on the
NDVI (Carlson 1994) and as mentioned before, NDVI is not
appropriate for arid areas.
Ghulam et al, in 2007 used the NDVI to assess drought using
MPDI, but as the result showed, this index is not very
appropriate for arid areas. Consequently, we examined its
substitution with another index. Among all studied indices only
SAVI2 had well presented the vegetation fraction (F v ).
Therefore, the fraction of vegetation has been estimated using
the following index.
SAV12-SAVI2 Max
V ~ [ SAVI2 m -SAVl2 M J
(5)
The results of the revised MPDI indicated that the area with
higher drought severity (more than 0.4) has largely increased
(Figure 4). The regions with moderate drought severity were
located in the northern and eastern hillside. These regions,
keeps the moisture for a longer time. Therefore, the intensity of
drought is lower than the other hillsides especially in
comparison with the southern hillsides. The lowest values for
the revised MPDI appeared in the high-land regions. In the
high-land regions, there is snow until the end of the growing
season and provide a supply of water for the vegetation growth.
Figure 4. MPDI changes (1999-2002)
Studying the meteorological data reveals that drought did not
occurred in 2002 and the results of PDI confirmed that as well.
However, the results of the revised MPDI showed that the
drought severity has increased. Therefore, it can be concluded
that both meteorological data and the indices based on the
satellite images, are essential for an accurate assessment of
drought disaster.
In summary, this study had the following conclusions:
1. NDVI is not an appropriate index for vegetation assessment
in the arid areas.
2. Vegetation indices which consider soil background
reflectance such as SAVI2 can be more useful for vegetation
assessment.
3. The PDI is suitable for meteorological drought monitoring.
4. Meteorological data and Remote sensing data are both
essential for an accurate assessment of drought.
References:
Baaqide, M. (2007). "Drought monitoring using multi temporal
NOAA, NDVI and GIS in Isfahan province." P.H.D thesis,
Tarbiat moallem university, Tehran.
Baret, F., Clevers, J., Steven, M.D. (1995). "The robustness of
canopy gap fraction estimations from red and near-infrared
reflectances." Remote Sensing of Environment 54(3): 14-151.
Baret, F„ Guyot, G. and Major, D.J., 1989. TSAVI: A
vegetation index which minimizes soil brightness effects on
LAI and APAR estimation, Geoscience and Remote Sensing
Symposium, IGARSS'89. 12th Canadian Symposium on Remote
Sensing, pp. 1355-1358.
Carlson, T. N., Gillies, R ., Perry, E.M. (1994). "A method to
make use of thermal infrared temperature and NDVI
measurement to infer surface soil water content and fractional
vegetation cover." Remote Sensing Environment 9: 161-173.
Carpenter, G., Gopal, S ., Macomber, S .,Martens, S .,
Woodcock, C ., Franklin J. (1999). "A Neural Network Method
for Efficient Vegetation Mapping." Remote Sensing of
Environment 70: 326-338.
Chen, W., Xiao, Q ., Sheng, Y. (1994). "Application of the
anomaly vegetation index to monitoring heavy drought in
1992." Remote Sensing of Environment 9(2): 106-112.