airphotos in the scale 1: 5 000. The average size of the
area pertaining these individual buildings was 0.125 ha. In
densely built-up areas these buffers merged into bigger
polygons. The contours of individual buffers dissolved in
that process and uniform areas of these polygons were
taken into account for further elaboration (Fig. 1).
The centroids of houses from June 1993 were buffered
and the same procedure was repeated with the centroids
of houses from June 1997. The built-up areas for these
two years were obtained separately. Overlaying the layers
with buffered centroids, the locations of pulled-down
houses were identified. Next, the buffers that were
common for both layers were removed. The remaining
presents the augmented built-up areas in Slovenia in the
period from June 1993 to June 1997.
Figure 1: Schematic presentation of buffered centroids.
- black buffered centroids: state '93
- gray buffered centroids: new built-up
from '93 to '97
The calculation of the augmented built-up areas by first
subtracting centroids and buffering the remaining would
be a much faster method but would not enable the
merging of adjacent buffers, thus leading to an
overestimation of the augmented built-up areas. In our
case the calculated augment of built-up area would be
greater by 41.5%.
3. ESTIMATION OF LAND COVER CHANGES DUE TO
AUGMENTED BUILT-UP AREAS
The data layer of augmented built-up areas was overlaid
onto the compiled Statistical Land Cover/Land Use GIS of
Slovenia-state '93. The areas of wooded and agricultural
land cover categories that turned into built-up areas have
been determined (Fig. 2). When the new built-up areas
were detected within existing built-up areas, they were not
considered as land cover/use changes since these buffers
did not change the cover/use categories. The buffered
centroids that were located with more than 50% of their
areas over water, railways or roads were considered as
mistakes and removed from the data-layer.
In the final stage the layer of augmented built-up areas
was merged with the layer of administrative boundaries
and thus the augmented built-up areas for each statistical
region and for the whole Slovenia were obtained (Tab.1).
In Slovenia in the period from June 1993 to June 1997
most augmented built-up areas have expanded on
account of agricultural areas (Fig.3). The statistical
regions number 6. and 9. have the highest change of
agricultural land cover to built-up.
Legend:
A = agricultural area
F = wooded area
W = water
B = buffered centroids:
black = state '93
gray = new built-up from '93 to '97
Figure 2: Schematic presentation of new built-up areas
overlaid onto the Land Cover/Land Use GIS of Slovenia -
state '93.
But it is the graphical presentation of this data that gives
us the essential information, i.e. the geographical
distribution of the phenomena. As expected, the built-up
areas increased mostly around urban centres. However,
an exception is the area of Savinja valley where the
augmented built-up areas are nearly evenly distributed
over the whole valley, which is categoriesed as I st quality
agriculture land. The geographic distribution of the
augmented built-up areas, using a relatively simple and
time efficient method, renders new qualitative information,
which would never be revealed by presenting these data
only in the classical tabulated form.
4. CONCLUSIONS
The Statistical Land Cover/Land Use GIS of Slovenia is
the first numerical GIS of Slovenia that comprises a
uniform thematic layer of the whole territory. The
Statistical Land Cover/Land Use GIS of Slovenia - state
'93 officialy confirmed that much more area of Slovenia is
under forest than it has been officialy reported. The
improved resolution of satellite scanned data used in the
Statistical Land Cover/Land Use GIS of Slovenia-state '97
will enable a better delineation of land cover categories
down to 15 hectares of minimum mapping unit. Since the
geographical heterogenity as well as the small field sizes
are the characteristics of Slovenia, this improvement is an
important contribution to a more operational use of the
compiled Statistical Land Cover/Land Use GIS of
Slovenia.
The Statistical Land Cover/Land Use GIS of Slovenia-
state '97 will enable us to estimate the land cover/use
change in the period from 1997 to 2001 due to augmented
built-up areas more accurate. In addition, the three level
of roads will be included as well as the built-up areas of
larger industrial objects, warehouses, parking places, etc.,
which exceed the size of 20 m buffered centroids.
