ly grown
enlarged
contains
patterns.
ormation
, LEI can
iore tem-
udy area
(LEI) are
and their
| used to
S section
of differ-
different
alue and
tribution
erent pe-
re 1.
had been
any non-
ere three
pe, edge-
a existing urban outlying type
edge-expansion
type
infilling type
Figure 1. Distribution of urban expansion types in different pe-
riods from 1993 to 2006
The areas and patch counts of different urban expansion types
are calculated and showed in Figure 2 and 3. For each period,
Figure 2 shows the percentage of area each urban expansion
type takes, e.g., outlying type takes 31% of the total expansion
area from 1996 to 1998, and Figure 3 shows the percentage of
patch counts for each urban expansion type, e.g., patch count
for edge-expansion type from 1993-1996 is 50% of the total ex-
pansion patches.
According to the comparison of areas or patch counts of differ-
ent urban expansion types (see Figure 2 and 3), the edge-
expansion type was the main urban expansion form. The change
of proportion of urban expansion types of the study area over
the study period reveal that urban expansion patterns had
changed from a messy, dispersed early development phase to
more compact and reasonable layout close to the end of the
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
study period.
1993~1996 1996~1998 1998~2001 2001--2006
Bloutlying D edge-expansion B infilling
Figure 2. Areas of three urban expansion types in different peri-
ods
19931996 1996 —1998 1998 —2001 2001--2006
[e outlying Bl edge-expansion B infilling
Figure 3. Patch counts of three landscape expansion types in
different periods
3.2 Urban Landscape Expansion Periods
The study results demonstrate that there were different urban
expansion modes in four different periods. In the first period
from 1993 to 1996, the edge-expansion and outlying types were
both main urban expansion forms. During this period, the ur-
banization reflected the characteristics of the early stage of rap-
id urban developments. In the second period from 1996 to 1998,
although edge-expansion type was still the main urban expan-
sion form, the outlying type had been increasing rapidly, which
indicates urban development is messy and dispersion. In the
third period from 1998 to 2001, the edge-expansion and infil-
ling type became the primary urban expansion forms. This peri-
od was the proliferation phase in the rapid development of ur-
banization. In the last period from 2001 to 2006, the edge-
expansion and infilling remained the main urban expansion
forms, which indicate urban expansion was shifted towards
compact development.
4. CONCLUSIONS
Quantitative studies of urban expansion forms help identify dif-
ferent types of urban expansions and expansion modes in dif-
ferent time periods. This paper used landscape expansion indi-
ces (LEI) to quantify the dynamic changes of urban landscape
in a selected study area over 13 years, which resulted in the fol-
lowing:
* Urban expansion patterns changed from a messy, dispersed
early development phase to more compact and reasonable
layout from 1993 to 2006 in Changsha-Zhuzhou-Xiangtan
metropolitan areas, Central China.
= From 1993 to 1998, the dominant form of urban expansion in
the study area was the edge-expansion type along with outly-
ing type, which shows messy and dispersed developments.
From 1998-2006, the dominant form of urban expansion in
the study area was the edge-expansion type along with infil-
ling type, which indicate urban expansion was moving to-
wards compact development.
China has experienced major urban developments from 1993 to
2006. The study results shows that the edge-expansion type of
urban expansion has continuously contributed around 60% of
the overall expansion, and the infilling type of expansion re-
mained the same level and the outlying type of expansion
slightly dropped. More meaningful interpretation of the study
results may require further investigation of related urban devel-
opment policies, laws and by-laws, as well as more analysis of
the areas with the similar characteristics.
