چکیده:
In order to assess the satellite data for soil investigation, ASTER digital data 20 June 2006, field study and phisiochemical
properties of soil, were analyzed. All landcover classes including soils are classified based on
morphological and physico-chemical characteristics. Images were geocorrected and photomorphic units were selected
based upon visual interpretation and sampling in study area. The images was classified using maximum likelihood
algorithm; with eight approaches. The classified image was compared with the ground truth map. The lowest
classification accuracy was achieved by optimum index factor (OIF) approaches and hence application of OIF for
discrimination of soils was not effective way. The results showed that best index is not only efficient and other
information such as DEM (digital elevation model) with the spectral combination increase the accuracy of
classification and Kappa coefficient. Salinity Indexes (SI) and Normalized Soil Index (NDSI) and Brightness Index
(BI) were useful for discrimination of the soils in the study area. Typic Haplocambids showed the maximum
reflection due to bright color. In addition, the minimum value was related to the Typic Torriorthents class, because of
dark gravels. The result showed ASTER data can differentiate Typic Haplocambids from Typic Torriorthents, Typic
Haplosalids and Typic Haplogypsids in arid lands.
خلاصه ماشینی:
Different soils in arid regions vary widely in the chemical characteristics and their reflectance contributes significantly to the overall spectral response from the surface area when vegetation cover is below 25- 35% (Vinogradov, 1984, Tueller, 1987).
Metternicht and Zink (2003) identified the salinity classes and Constraints on the use of remote sensing data for mapping saltaffected areas are shown related to the spectral behaviour of salt types, spatial distribution of salts on the terrain surface, temporal changes on salinity, interference of vegetation, and spectral confusions with other terrain surface.
(2006) evaluated data from ASTER (Advance Spaceborne Thermal Emission and Reflection), LISIII, MSS (Multi Spectral Scanner), TM (Enhanced Thematic Mapper), and ETM+ (Enhanced Thematic Mapper plus) sensors to identify soils of Kashan area in Iran.
The results of this study indicate that ASTER, TM, and ETM+ sensors classified soils in three classes: Typic Aquisalids, Typic Haplosalids, and Gypsic Aquisalids, due to capability of thermal band in discrimination between saline soil and Gypsic -saline soils.
Given the above description, we can conclude that OIF index is not effective for discrimination of soil classes in an arid zone like Damghan (Table 3)(View the image of this page)3.
The lowest classification accuracy was achieved in OIF approaches and hence application of OIF for discrimination of soils of this investigated zone was not found effective, that is in conformity with results of Matinfar (2006).
In the first time PCA approach of all bands plus salinity indexes SI, BI, and NDSI, 72 percent accuracy was achieved that indicates usefulness of these indices for discrimination of soils in the investigated zone.