[8 
  
El 
The aims of this paper are: (1) monitoring changes on 
landuse-landcover of the area extending from Gamasa to 
Baltim towns (140 km) and extended about 20 km to the 
south. The study area is bordered by the Mediterranean Sea 
from the north and the old cultivated land from the south, 
(Fig. 1), (2) studying the impact of sea wave currents in the 
erosional and depositional actions along the coast line, and 
(3) studying the effect of active sand dunes on the 
infrastructure and human activities in the study region. 
2. MATERIALS AND METHODS 
2.1. Material 
1- Landuse-landcover map (1:100,000) based n SPOT 
XS data of 1991, produced by the Remote Sensing 
Unit of SWERI. 
2- Landsat TM image of 1990. 
3- SPOT XS image of 1995. 
4- Several field visits and soil sampling and analysis from 
selected sites using GPS. 
5- Aerial photographs of (1983) scale 1:20,000. 
2.2. Methodology 
1- Interpretation of Aerial photographs for producing 
geomorpho-logical maps at a scale of (1:20,000). 
2- Image processing includes; 
a) Unsupervised classification of Landsat TM and SPOT 
data using IMAGEN Package (ISODATA) 
program. 
b) Principle component Analysis (PCA) of Landsat 
TM data. 
c) Hybrid classification and statistical analysis of the 
classified units, (1990-1995). 
3- Ground measurements and laboratory analysis. 
RESULTS AND DISCUSSION 
3.1- Geomorphology of the study area 
The interpretation of aerial photographs and terrain 
analysis of the studied region produced a geomorphologic 
map that shows the major land forms represented by: 
beach, sand plan, dunes, old beach ridges, wetland, sabkha 
and evaporate crusts. Each land unit has been investigated 
in the field, and the main geomorphic processes such as 
wind erosion in the form of sand removal from the root 
zone of palm trees, or sand deposnition on the road, 
setelments and cultivated land. 
3.2.1-Landuse-landcover classification 
The application of hybrid image classification technique to 
Landsat TM data acquired in 1990 (Fig. 2a) reveals that the 
main landuse-landcover classes are; water bodies, shallow 
water, sabkhas, cultivated land, sand dune, bare soil, urban, 
and reclaimed land. However, the classification of the 
SPOT image of the same area aquired in 1995 (Fig. 2b) 
shows clearly the increase of the total area of some classes 
on the expense of others. 
The comparison between the statistical data of the used 
SPOT and Landsat TM images reveals that the landuse- 
landcover has been changed dramatically within five years. 
The data in Table (1) show that: 
a) There is an increase in the area of the wetland and 
waterlogging regions due to the establishment of fish 
farms, extension of rice growing fields and high water 
table level, (Eweida et al., 1997). 
b) The total area of cultivated land is approximately 
doubled, due to land reclamation activities. 
c) Vast areas of sabkhas and sand areas have been 
converted to productive land. 
Table (1) Changes in landuse-landcover in (km?) between 
(1990-1995). 
  
  
  
Land unit Landsat TM 1990 SPOT 1995 
(km?) (km?) 
Water bodies 10.8 11.6 
Agriculture 13.0 24.5 
Reclaimed land 20.9 30.3 
Sabkhas 524 41.5 
Sand dunes 265.6 177.8 
Windblown sand 189.8 167.0 
  
  
  
  
3.3. Erosion and accretion processes along the coast line 
Application of multi-temporal analysis of remote sensing 
data and the interpretation of the existing maps of the study 
area enabled in predicting changes along the coastline. The 
principle component analysis (PCA) of the Landsat 
Thematic Mapper data assisted in determining the newly 
formed areas (spils formation). 
In most of the studied sites (1.e. Baltiem, Ras El Bar and 
Gamasa), accretion occurs in the eastern side of the eroded 
areas due to the west to east direction of active wave 
current. 
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 
61