tanbul 2004 
‘een texture 
s effect. 
Performance 
1 wavelets, 
Engineering 
ire In digital 
and Remote 
73. Texture 
sactions on 
25. 
vised image 
robabilisatic 
s and Image 
egmentation. 
), pp.19-26. 
ution signal 
Trans. Patt. 
<yllônen J & 
for empirical 
Proc. 16th 
ion, Quebec, 
' for texture 
s. Patt. Anal. 
.M.J., Pardo, 
mentación 'de 
. Proc. of the 
p. 477-480. 
measures for 
sease. IEEE 
6. 
entation using 
£5 4( I 1), pp. 
e 
ck, D., 199. 
crete Wavelel 
sing, 8(4) P 
rovided by the 
nd the FEDER 
), as well as 9 
102-0627). 
STUDY ON SOIL EROSION AND SEDIMENTATION IN ALASHTAR WATERSHED USING 
IMAGE PROCESSING SOFTWARE 
a . . . . D . 
Parviz Zeaiean Firouzabadi, "Ali Davoodi 
a . . . . 
Department of Geography, Teacher Training University Tehran Iran 
parviz8(@hotmail.com 
phy, Teacher Training University Tehran Iran 
      
"Department of Geogra 
KEY WORDS: Remote Sensing, Soil, Modeling, Landsat, Spatial, Spectral, Land Cover, Land Use 
ABSTRACT: 
In this research, an attempted has been made to study soil erosion and sedimentation rates in Al 
of Iran using MPSIAC method. Different data includin 
used. All available data were first geo-referenced using 
ashtar watershed basin, Lorestan province 
£ TM image, climatic data, soil map, geology map and topographic maps were 
GEOMATICA image processing software. Land use/ land cover and NDVI maps 
were produced through image classification algorithm and EASI modeling language of this software. From climatic data and through 
interpolation technique an image containing distribution of precipitation was generated. All boundaries of soil categories and geological 
futures were digitized using the same software. Slope and aspect images were generated from digital contour map of the area. Using EASI 
modeling language of Geomatica, all the maps were given their specific weights and overlaid. By using formula Qs-38.77e??* erosion 
rates ( M’/km“/yr) were calculated. The resultant erosion map showed that the erosion rate is high in this area. The importance of this 
research work is that instead of using GIS software, all the spatial analyses were carried out using image processing software package. 
map. In this reseach an attempt has been made to study the rate of 
soil erosion in Dooab-e-Alashtar watershed in Lorestan province 
of Iran. The study area is located between the longitude 48°30'6" 
to 48 3113" north and latitude 33°44'24” to 34°30'15” east and has 
an area of 766.92 Km. 
I. INTRODUCTION 
Soil erosion is one of the natural processes that would cause many 
constrains to the environmental and regional planners. Erosion on 
farm fields reduces potential crop production, and sediment which 
leaves the field can result in subsequent sedimentation problems 
which, in turn, can cause off-site environmental problems (e.g., 
ASCE, 1975, 1982). An example is the redistribution of fallout 
plutonium. Erosion on other upland areas--such as construction 
sites, urban areas, mine developments, or other disturbed areas-- 
can also cause on-site and off site problems (e.g., AGU, 1977: 
2. MATERIALS AND METHODS 
Followings are the list of data and materials used in this research 
work 
I. Digital topographic maps of Lorestan province in the 
scale of 1:25000. 
2. Hardcopy topographic maps of Lorestan province in the 
ASCE, 1975). Channel erosion or deposition processes can cause scale of 1:50000. 
further problems because the stream channels are components of 3. Landsat digital data pertaining to the study area. 
the watershed system (ASCE, 1975, 1982). Although total 4. Rainfall data for S stations. _ 
prevention of this process is not possible at all, but reduction of its S. 
Hardcopy soil map of the study area in the scale of 
1:50000 
6. Hardcopy Geology map of the study area in the scale of 
1:1250000 
7. An AO color scanner 
8. image processing package (Geomatica Ver.8.2 ) 
volume, extent and speed may be practically possible. Many 
attempts have been made to use different models to estimate the 
volume of soils eroded every year. Models such as SEIM, USLE, 
EPM, PSIAC, MPSIAC and FAO are among the most widely used 
models. The soil erosion index model (SEIM) was used by Su- 
Chin Chen 2002 to assess and predict the potential areas of 
erosion in Taiwan. The SEIM model uses different factors such as 
rainfall, soil characteristics, slope, and cover above soil surface 
and land use. PSIAC model with its 9 possible parameters, among 
other model, is the most suitable model for arid and semi-arid 
regions. Soil erosion in different parts of Iran, as an arid and semi- 
ard region, has been studied by several researchers. Heydarian 
1996 and Tajbakhsh et. Al, 2003 used PASIC and Modified 
PSIAC (MPSTAC) to estimate erosion yield and erosion intensity 
The general steps followed in this study are as follows. 
A) Preliminary studies to define the objectives and methodology 
B) Preparation of different materials and the selection of software 
package 
C) Registration of all the available maps on each other 
D) Production of digital maps from different data 
E) Assignment of weights to different factors of MPSIAC model 
1115