Discontinuous Low Density Urban Fabric
This urban fabric comprises the area of historical site which
is in UNESCO’s word heritage list and special linear settle-
ments along the road in
SUA
.
This class covers 153.71 and 250.23 ha, respectively, for
SUA
and
CUA
.
PD
values are 5.97 and 4.88; whereas
ENN_MN
values are 37.80 and 36.32, respectively. Higher values of
PD
and
ED
values in
SUA
indicates that these areas are more
heterogeneous and complex patches.
Discontinuous Very Low Density Urban Fabric
This class covers 113.20 and 128.77 ha for
SUA
and
CUA
. In
SUA
, this fabric is clustered in the middle of two new settle-
ments. This area is historical settlement area with high
amount of green space. Whereas, in
CUA
region, these fabrics
are located outside of the dense residential areas.
PD
values
are 2.83 and 2.78.
ENN_MN
values are 48.20 and 55.39, respec-
tively, for
SUA
and
CUA
indicating more compact areas for
SUA
.
Green Urban Areas
The total area of green urban areas are 42.31 ha and 126.71
ha for
SUA
and
CUA
, respectively.
PD
and
ED
values are also
higher, but
LPI
value is lower in
CUA
. This indicates that urban
green areas have a greater amount and density with high spa-
tial heterogeneity and relatively consist of smaller patches in
CUA
.
ENN_MN
values are 55.36 and 47.15, respectively, for
SUA
and
CUA
. This indicates a higher connectivity of urban green
patches in
CUA
. The configuration of green urban areas in
CUA
is more preferred in urban planning. Moreover, these results
show that
CUA
has a more planned urbanization structure
than
SUA
.
Industrial, Commercial, Public, Military and Private Units
This class covers 219.33 ha and 741.70 ha and
PD
values
are 2.97 and 3.3 for
SUA
and
CUA
, respectively. Densities of
these units within two settlements are close to each other but
the amount is specifically higher in
CUA
.
ENN_MN
values are
123.71 and 71.57 for
SUA
and
CUA
, respectively. Karabuk Iron
and Steel Factory located in
CUA
compose the largest patch
of this class.
LPI
value (2.84) specifies the percentage value
of this patch in the total urban area. University, public and
commercial areas, small industrial regions are the other land
uses that compose this class. In
SUA
,
LPI
is 2.52 where military
space and small industrial regions are the main regions of this
class in
SUA
.
Natural Land Covers
Agricultural areas, semi-natural areas and wetlands cover
more than half of the
SUA
and nearly half of the
CUA
.
PD
val-
ues illustrated that density of natural land cover in Safranbolu
is higher than Karabuk city center.
For districts, the class level metric results are shown in
Table 5. Agricultural and semi-natural areas are denser in
the Safranbolu district, while the
LPI
is higher in the Center
district. This indicates that these areas consist of more and
more large areas in the Center. In Safranbolu, agricultural and
semi-natural areas are more fragmented. However, the
FRAC_
AM
index is higher in Safranbolu, indicating that the shapes
of the areas are more complex. Also the
ENN_MN
index shows
that the connection between agricultural and semi-natural
areas is more in Safranbolu.
It is seen from the
PD
results that the density of forest areas
is higher in Safranbolu and closely distributed. The
LPI
results
indicate that forest areas are formed from more aggregated
areas in the Center. The
ENN_MN
results indicate that these ag-
gregated areas of the forest class in the Center are scattered in
different parts of the district. Although forest areas have more
proximity to each other in Safranbolu, other metric results
also indicate that these areas are composed of more fragment-
ed and smaller areas.
CONTAG
gives information about how fragmented or how
aggregated the landscapes are. The values of
CONTAG
indices
for urban areas are 65.29 and 64.27 ha in Safranbolu and
Center, respectively (Table 6). Whereas, 85.70 and 84.80 ha for
Safranbolu and Center districts, respectively (Table 7).
CON-
TAG
metric value calculated for the whole district area shows
that the heterogeneity is slightly higher compared to urban-
ized parts of these two districts.
Conclusions
High resolution urban
LCLU
maps are efficient source of
geo-information for variety of urban related studies such as
creation of master planning, monitoring of transportation
infrastructures, generation of location based services, and
support street navigation. Remote sensing technology could
be used as a versatile source to create urban related geospatial
information due to the variability of high resolution sensors
and advances in image processing algorithms. This study
proposed the usage of decision-tree based object oriented
Table 4. Metrics results for Safranbolu and Center urban areas.
Class
PD
LPI
ED
FRAC_AM ENN_MN
SUA CUA SUA CUA SUA CUA SUA CUA SUA CUA
Continuous Urban Fabric (S.L. > 80%)
2.46 2.23 0.15 0.06 9.43 10.41 1.07 1.09 84.48 34.30
Discontinuous Dense Urban Fabric (S.L. 50% - 80%)
6.34 3.07 0.15 0.04 24.58 13.68 1.08 1.09 32.62 69.60
Discontinuous Medium Density Urban Fabric (S.L. 30% - 50%) 4.99 4.16 0.08 0.11 21.63 19.73 1.11 1.12 54.17 39.29
Discontinuous Low Density Urban Fabric (S.L. 10% - 30%)
5.97 4.88 0.22 0.23 34.83 26.06 1.15 1.13 37.80 36.32
Discontinuous Very Low Density Urban Fabric (S.L. < 10%)
2.83 2.78 0.15 0.13 18.38 14.18 1.12 1.13 48.20 55.39
Industrial, commercial, public, military and private units
2.97 3.31 2.52 2.84 19.73 27.08 1.12 1.10 123.77 71.57
Other roads and associated land
10.50 7.15 9.13 4.64 197.43 139.64 1.56 1.54 9.26 14.24
Railways and associated land
* 0.01 * 0.28 *
6.62 * 1.54 *
N/A
Mineral extraction and dump sites
0.51 0.07 0.43 0.17 6.06 1.42 1.16 1.13 249.15 793.98
Construction sites
0.43 0.22 0.08 0.04 2.96 1.32 1.17 1.12 371.73 326.53
Land without current use
0.54 0.65 0.35 0.13 3.17 4.01 1.09 1.09 225.48 47.44
Green urban areas
1.99 2.58 0.11 0.07 10.06 15.00 1.13 1.16 55.36 47.15
Sports and leisure facilities
0.97 0.20 0.11 0.04 3.30 1.18 1.11 1.06 85.32 961.29
Agricultural areas, semi-natural areas and wetlands
5.74 4.17 19.24 14.70 114.10 67.18 1.27 1.22 25.11 31.41
Forests
1.83 2.60 0.80 10.67 34.39 39.82 1.21 1.24 94.33 42.50
Water
0.06 0.33 0.43 0.35 4.04 8.78 1.31 1.22 10.82 12.60
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
November 2018
719
