tiiÜÜfc rii 
Due to natural and anthropogenic activities the earth land 
changes, and this change plays vital role in the social and 
economic development of one area. In case of Egnatia highway, 
considering the large scale of the construction project and the 
impact that already has and will have in the future, the need to 
exploit the highway's contribution to the development and 
cohesion of its impact area, is becoming more and more 
essential. To achieve this task in this study, the estimation of 
two indicators regarding to land use changes over the decade 
1998-2007 is carried out, with change detection methods. 
5.2.1 The first indicator: The first indicator estimates (a) 
the rate of change of agricultural land (cultivated areas) into 
urban land, (b) the change of natural areas into urban land, and 
(c) the change of natural areas into agricultural land. This is 
accomplished with the intersection between the shapefiles that 
are shown in the following table: 
The change detection issue is one of the most interesting 
applications of Remote Sensing and numerous methods have 
been proposed by the scientists over the decades. The post 
classification comparison method is selected to perform land 
cover change detection in this study. 
5.1 The indicators 
a) 
As it is mentioned above, two indicators are estimated in order 
to assess the interaction between the Egnatia highway and the 
developmental procedure and physical planning in its impact 
area: 
The first indicator monitors the pressure for land use 
changes that have occurred over the decade 1998- 
2007 by estimating the change from rural to non-rural 
land use. 
The second indicator estimates the rate of change of 
the urban’s land density, based on three categories: 
continuous, linear and discontinuous urban land. 
Intersection 
Final 
product 
The shape file “agricultural land” from the 
SPOT vector image (Base year 1998) with the 
shape file “Urban land” from the IKONOS 
vector image (Check year 2007) 
Areas that 
changed 
from 
agricultural 
into urban 
land 
The shape file “Natural areas” from the SPOT 
vector image (Base year 1998) with the shape 
file “Urban land” from the IKONOS vector 
image (Check year 2007) 
-► 
Areas that 
changed 
from natural 
into urban 
land 
The shape file “Natural areas” from the SPOT 
vector image (Base year 1998) with the shape 
file “agricultural land” from the IKONOS 
vector image (Check year 2007) 
-> 
Areas that 
changed 
from natural 
into 
agricultural 
land 
Table 10. The intersections for the first indicator 
b) 
The areas that result from the above intersections depict the 
land use changes that have occurred over the decade 1998-2007 
in the selected areas. The following table presents the 
percentage of change that is also calculated. 
5.2 The change detection procedure 
The first step, in order to assess the two indicators, is to 
transform the classified images from raster format into vector 
format in order to introduce the images into the ArcGIS 
environment and to proceed in further processing. 
IOANN IN A 
THESSALONIK 
I 
KOMOTINI 
percentage of change 
Natural 
areas& 
Agricultural 
land 
-3.53% 
-19.49% 
-1.33% 
Urban land 
+12.20% 
+21.79% 
+8.22% 
Table 11. The percentage of land use change in the study areas. 
As it can be concluded from the above table, all the areas have 
an increase of urban land and simultaneously a degrease of 
natural and agricultural land over the past decade. 
5.2.2 The second indicator: The second indicator estimates 
the rate of change of the urban’s land density. Before 
performing the intersections for this indicator, it is necessary to 
determine the areas of continuous, linear and discontinuous 
urban land. 
Figure 9. Raster to vector: The initial orthoimage (a), the 
corresponding classified image (b) and the produced vector 
image (c). 
The shapefiles derived from the vector images, represent each 
class from the classification process. This enables to identify 
the areas that have been changed over the decade 1998-2007 by 
executing an intersection between the shapefiles of the SPOT 
vector images with the corresponding shapefiles of the 
IKONOS vector images. 
Therefore, the continuous land is determined as the urban land 
that is located inside the areas of the established borders of the 
Greek settlements of the study areas, the linear is defined as the 
urban land located in a buffer zone of 500m along the basic 
road axis of the study areas, and finally the discontinuous land 
is defined as the urban land that is located in areas that do not 
belong either at linear nor at continuous urban land. After the 
appropriate intersections between the shape files of these three 
categories of urban land, result the areas that depict the change 
of urban land, and the percentages of change that are calculated 
for the study areas are presented in the table that follows. 
18