Access to this full-text is provided by MDPI.
Content available from Remote Sensing
This content is subject to copyright.
Academic Editor: Dimitrios D.
Alexakis
Received: 9 November 2024
Revised: 9 January 2025
Accepted: 11 January 2025
Published: 14 January 2025
Citation: Liu, Y.; Wu, C.; Wu, J.; Zhang,
Y.; Bi, X.; Wang, M.; Yan, E.; Song, C.; Li,
J. Projected Spatiotemporal Evolution of
Urban Form Using the SLEUTH Model
with Urban Master Plan Scenarios.
Remote Sens. 2025,17, 270. https://
doi.org/10.3390/rs17020270
Copyright: © 2025 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license
(https://creativecommons.org/
licenses/by/4.0/).
Article
Projected Spatiotemporal Evolution of Urban Form Using the
SLEUTH Model with Urban Master Plan Scenarios
Yuhan Liu 1, Caiyan Wu 2, 3, *, Jiong Wu 2, Yangcen Zhang 2, Xing Bi 2, Meng Wang 2, Enrong Yan 1,4 ,
Conghe Song 5and Junxiang Li 2
1School of Ecological and Environmental Sciences, East China Normal University, Shanghai 200241, China;
52193903007@stu.ecnu.edu.cn (Y.L.); eryan@des.ecnu.edu.cn (E.Y.)
2Department of Landscape Architecture, School of Design, Shanghai Jiao Tong University,
Shanghai 200240, China; bigjohn59@sjtu.edu.cn (J.W.); zhangyangcen@sjtu.edu.cn (Y.Z.);
bixiaoxing@sjtu.edu.cn (X.B.); meng.wang@sjtu.edu.cn (M.W.); junxiangli@sjtu.edu.cn (J.L.)
3Department of Geography, Humboldt-Universität zu Berlin, 12489 Berlin, Germany
4
Shanghai Key Laboratory of Urban Ecological Processes and Eco-Restoration, East China Normal University,
Shanghai 200241, China
5Department of Geography and Environment, University of North Carolina at Chapel Hill,
Chapel Hill, NC 27599, USA; csong@email.unc.edu
*Correspondence: caiyanwu@sjtu.edu.cn
Abstract: Urban growth, a pivotal characteristic of economic development, brings many
environmental and ecological challenges. Modeling urban growth is essential for under-
standing its spatial dynamics and projecting future trends, providing insights for effective
urban planning and sustainable development. This study aims to assess the spatiotemporal
patterns of urban growth and morphological evolution in mainland Shanghai from 2016 to
2060 using the SLEUTH model under multiple growth scenarios based on the Shanghai Ur-
ban Master Plan (2017–2035). A comprehensive set of urban growth metrics and quadrant
analysis were employed to quantify the magnitude, rate, intensity, and direction of urban
growth, as well as morphological evolution, over time. We found that (1) significant urban
growth was observed across most scenarios, with the exception of stringent land protection.
The most substantial growth occurred prior to 2045 with an obvious north–south disparity,
where southern regions demonstrated more pronounced increases in urban land area and
urbanization rates. (2) The spatiotemporal patterns of the rate and intensity of urban
growth exhibited similar characteristics. The spatial pattern followed a “concave shape”
pattern and displayed anisotropic behavior, with the high values for these indicators pri-
marily observed before 2025. (3) The urban form followed a diffusion–coalescence process,
with patch areas dominated by the infilling mode and patch numbers dominated by the
edge-expansion mode. This resulted in significant alternating urban growth models in the
infilling, edge-expansion, and leapfrog modes over time, influenced by varying protection
intensities. These findings provide valuable insights for forward-looking urban planning,
land use optimization, and the support of sustainable urban development.
Keywords: urban form; urban growth; SLEUTH model; urban planning scenario;
spatiotemporal evolution; diffusion–coalescence process
1. Introduction
Urbanization is a dynamic process that reshapes the built environment, converting
rural areas into urban areas and relocating populations [
1
]. This transformation is evident
in the evolution of urban forms [
2
], including the spatial arrangement of land use and urban
Remote Sens. 2025,17, 270 https://doi.org/10.3390/rs17020270
Remote Sens. 2025,17, 270 2 of 21
landscapes, as well as their changes [
3
]. Meanwhile, urbanization is influenced by urban
planning and a variety of factors, such as infrastructure investments from both the public
and private sectors [
1
]. The interplay between top-down management or investments
and bottom-up processes creates the urban development dynamic [4]. Understanding the
evolution of urban forms is crucial for enhancing urban planning and identifying effective
urban development solutions [
5
]. While historical urban forms offer valuable insights into
past urbanization processes [
6
], our understanding of future urban forms, especially in
urban planning contexts, remains limited. This knowledge is crucial for predicting the
trajectory of urbanization and supporting sustainable urban development strategies.
Land use change (LUC) models are powerful tools for simulating and predicting
the spatiotemporal dynamics of urban areas, capturing the complex interactions between
natural and anthropogenic factors [
7
]. Urban growth is a key aspect of LUC, which can
be modeled with various approaches, including statistical models (e.g., linear and logistic
regression) [
8
], machine learning models (e.g., artificial neural networks, ANNs), tree-based
models (e.g., decision trees), and cellular automaton (CA) models [
9
]. CA-based models
are widely used to simulate urban growth dynamics at city, national, and global scales [
10
].
The SLEUTH model, a CA model [
11
], estimates two-dimensional urban growth [
12
]. A
unique advantage of the SLEUTH model is its ability for self-modification and adjustment
to shifting conditions in urban growth modeling [
13
]. Scenarios are essential for urban
growth modeling to inform planning policies, environmental protection measures, and
economic growth expectations [
14
]. Previous studies using the SLEUTH model have often
employed binary or more abstract scenarios like business-as-usual [
13
,
15
]. Research in
Shanghai has similarly relied on two to four predefined scenarios, mainly focusing on
short-term urban growth projections [
16
,
17
]. However, such scenarios may not adequately
capture the complexities of future urban development, underscoring the need for improved
scenario design [
18
,
19
]. Recent studies have incorporated objectives like maximizing eco-
nomic benefits and ecological benefits or balancing both ecological and economic outcomes
into scenario analysis [
20
,
21
]. Furthermore, landscape ecology principles [
22
] or habitat
quality [
23
] have been incorporated into the SLEUTH model, serving as exclusion layers to
regulate the development boundaries of new urban areas and prevent violations of ecologi-
cal redlines [
23
]. Despite these advancements, there remains a significant opportunity to
further enhance the model’s comprehensiveness by explicitly integrating urban planning
strategies. Such an integration would ensure that predictions of urban growth not only
reflect growth dynamics but also align with sustainability objectives and urban planning
goals, ultimately contributing to long-term urban sustainability.
Understanding the evolution of urban form deepens our insights into its spatiotem-
poral layouts and structure changes within cities, which is essential for informed urban
planning [
24
]. Previous studies have demonstrated that different urban growth mechanisms
significantly influence urban forms [
5
,
25
,
26
]. Three primary urban growth mechanisms,
i.e., infilling, edge-expansion, and leapfrog/outlying, lead to distinct urban morpholo-
gies [
27
,
28
]. Edge-expansion and leapfrog/outlying generally drive cities outward, result-
ing in a diffusion process characterized by outward urban growth. In contrast, infilling
signifies an inward growth, leading to urban coalescence [
5
,
6
]. The diffusion–coalescence
hypothesis underlying this process has been widely tested at city, regional, national, and
global scales [
6
,
27
,
29
–
31
], providing vital insights for urban planning, policymaking, and
sustainability efforts [
30
,
32
]. Despite these insights, future changes in urban forms remain
uncertain as urbanization processes continue to evolve, necessitating further investiga-
tion, particularly concerning the ongoing validity of the diffusion–coalescence hypothe-
sis [
24
,
33
,
34
]. During the urbanization process, newly grown urban patches emerge in
various directions at different rates, intensities, and patterns, resulting in anisotropic ur-
Remote Sens. 2025,17, 270 3 of 21
ban expansion [
35
,
36
]. Urban growth patterns are often driven by factors such as urban
planning, policy decisions, and economic development [
35
]. To effectively examine the
directional dynamics of urbanization, spatial analysis techniques, such as quadrant analy-
sis, are essential [
37
–
39
]. By mapping and examining the spatial patterns of future urban
growth across different directions, we can achieve a clearer understanding of the magni-
tude and trajectory of these changes. These insights will improve urbanization dynamic
monitoring and land use management, helping policymakers and planners better mitigate
the impacts of future land use and land cover changes on urban ecosystem functions [13].
Shanghai, as a metropolis with a distinct central city–satellite structure [
40
], has
experienced rapid urbanization and significant spatiotemporal changes in urban form over
recent decades [
6
,
41
]. Its urban evolution has been proven to follow the spiraling diffusion–
coalescence hypothesis [
6
]. However, despite extensive research on urban form, accurately
predicting its future evolution remains an issue. To project the spatiotemporal patterns of
urban growth in Shanghai in the future, we innovatively integrated the SLEUTH model
with the Shanghai Urban Master Plan into urban growth scenarios to simulate urban land
changes from 2016 to 2060. We further predict the urbanization rate (i.e., the proportion
of urban land area) and urban growth rate and intensity while testing the applicability of
the diffusion–coalescence hypothesis under varying land use protection scenarios. The
following questions are addressed: (1) What are the spatial patterns of urban growth in
Shanghai from 2016 to 2060 under the Shanghai urban planning scenarios? (2) What are
the spatial dynamics of urban growth in terms of rate, intensity, and directionality, and
do these patterns exhibit anisotropy? (3) Does the projected urban form evolution follow
the diffusion–coalescence process? Understanding these dynamics will provide valuable
insights for urban planners and decision makers in developing effective strategies for
sustainable urban development.
2. Materials and Methods
2.1. Study Area
Shanghai, situated in the Yangtze River Delta of East China, spans a total land area
of 6340.5 km
2
. The city is flanked by the East China Sea to the east and the mouth of
the Yangtze River to the north and is bordered by the Zhejiang and Jiangsu provinces to
the west. It rests on an alluvial plain that features a dense network of waterways, which
amounts to a density of 3.38 km/km
2
[
42
]. The region is approximately 2.19 m above sea
level on average and is characterized by several low hills. Over the past decades, Shanghai
has undergone significant urbanization, leading to substantial alterations in its landscape
and a marked increase in total urban land area [
6
,
27
,
43
]. The rapid growth has been
accompanied by a marked increase in population and a GDP of almost CNY 4.47 trillion
(nearly US $620.23 billion) [
44
]. Thus, quantifying the future spatiotemporal dynamics of
urban development in Shanghai is crucial for understanding the impacts of urbanization,
informing effective urban planning, and guiding policy decisions. This research focuses
on the mainland of Shanghai, excluding Chongming island (CM), and encompasses seven
districts within the urban center (UC) area and eight districts in the suburban and exurban
regions: Minhang (MH), Baoshan (BS), Jiading (JD), Pudong New Area (PD), Fengxian
(FX), Jinshan (JS), Qingpu (QP), and Songjiang (SJ) (Figure 1).
Remote Sens. 2025,17, 270 4 of 21
Remote Sens. 2025, 17, x FOR PEER REVIEW 4 of 22
Figure 1. The location of the study area.
2.2. Research Framework
This research is organized into three key stages: data preprocessing, scenario-based
simulation, and analysis of the spatiotemporal paerns in urban form. The SLEUTH
model (version 3.0_beta) was used to stimulate urban land change in Shanghai over a 45-
year period from 2016 to 2060. This timeframe corresponds to both short-term (2016–2035)
and long-term (2036–2060) urban development strategies for Shanghai, aiming to achieve
“carbon neutrality” by 2060. Achieving the carbon neutrality goal requires more stringent
urban planning criteria and constraints on urban growth [45]. Additionally, the Shanghai
Urban Master Plan outlines land use requirements, which essentially prescribe urban
growth scenarios for the city. To analyze the spatiotemporal changes in urban form,
quadrant analysis and urban land expansion indicators are utilized. Figure 2 provides an
overview of the main steps in this research framework.
Figure 2. The flowchart of the research framework.
Figure 1. The location of the study area.
2.2. Research Framework
This research is organized into three key stages: data preprocessing, scenario-based
simulation, and analysis of the spatiotemporal patterns in urban form. The SLEUTH model
(version 3.0_beta) was used to stimulate urban land change in Shanghai over a 45-year
period from 2016 to 2060. This timeframe corresponds to both short-term (
2016–2035
)
and long-term (2036–2060) urban development strategies for Shanghai, aiming to achieve
“carbon neutrality” by 2060. Achieving the carbon neutrality goal requires more stringent
urban planning criteria and constraints on urban growth [
45
]. Additionally, the Shanghai
Urban Master Plan outlines land use requirements, which essentially prescribe urban
growth scenarios for the city. To analyze the spatiotemporal changes in urban form,
quadrant analysis and urban land expansion indicators are utilized. Figure 2provides an
overview of the main steps in this research framework.
Remote Sens. 2025, 17, x FOR PEER REVIEW 4 of 22
Figure 1. The location of the study area.
2.2. Research Framework
This research is organized into three key stages: data preprocessing, scenario-based
simulation, and analysis of the spatiotemporal paerns in urban form. The SLEUTH
model (version 3.0_beta) was used to stimulate urban land change in Shanghai over a 45-
year period from 2016 to 2060. This timeframe corresponds to both short-term (2016–2035)
and long-term (2036–2060) urban development strategies for Shanghai, aiming to achieve
“carbon neutrality” by 2060. Achieving the carbon neutrality goal requires more stringent
urban planning criteria and constraints on urban growth [45]. Additionally, the Shanghai
Urban Master Plan outlines land use requirements, which essentially prescribe urban
growth scenarios for the city. To analyze the spatiotemporal changes in urban form,
quadrant analysis and urban land expansion indicators are utilized. Figure 2 provides an
overview of the main steps in this research framework.
Figure 2. The flowchart of the research framework.
Figure 2. The flowchart of the research framework.
Remote Sens. 2025,17, 270 5 of 21
2.3. The SLEUTH Model
2.3.1. Input Dataset of the SLEUTH Model
SLEUTH, an acronym representing slope, land use, exclusion, urban extent, trans-
portation, and hillshade, is the model developed by Dr. Keith C. Clark at the University of
California, Santa Barbara. It serves as a forecasting model for urban growth modeling and
land use modeling [46].
The SLEUTH model requires six input layer datasets: urban, land use, transportation,
exclusion, slope, and hillshade. All layers, except for hillshade, are essential for the model’s
functionality. The hillshade layer is used primarily for visualization purposes. The land
use data were derived from the China Landcover dataset (CLCD) with a resolution of
30 m [
47
], which has an accuracy ranging from 76.5% to 82.5% for the years from 1990
to 2019 [
47
]. The dataset is publicly accessible (https://zenodo.org/records/8176941,
accessed on
28 August 2021
). The original CLCD categories include cropland, forest, shrub,
grassland, water, snow/ice, barren, impervious, and wetland. To simplify this study and
align with the land use requirements in the Shanghai Urban Master Plan, we reclassified the
CLCD data into four types: urban land, farmland, water, and green land. The urban layer
and land use layer were extracted from the reclassified CLCD data. The road network data
were derived from the LULC dataset of Shanghai [
48
], which was initially classified using
2.5 m resolution images and then resampled to a 30 m resolution. Additionally, DEM data
used for the slope and hillshade layers were obtained from the Shuttle Radar Topography
Mission (SRTM) with a resolution of 30 m.
2.3.2. Scenario Settings for the SLEUTH Model
The SLEUTH model provides users with the flexibility to integrate planning strategies
tailored to their specific goals [
49
,
50
]. In this study, we incorporated the Shanghai Urban
Master Plan (2017–2035), which outlines specific land use requirements for green land,
water bodies, and farmland [
51
] into the SLEUTH model to simulate future urban growth in
Shanghai. To implement these strategies within the SLEUTH model, we assigned protection
intensities to each land use category. These intensities determine the level of protection
for each land use type, thus influencing their resistance to urban growth. For instance,
urban planning Scenario C is designed to preserve 1200 km
2
of farmland and achieve
23% forest coverage in accordance with the Shanghai Urban Master Plan. In keeping with
the requirement that green land in Shanghai should “only increase, not decrease”, the
protection intensity for green land is set at 100%. Similarly, water bodies are governed by
the Ecological Protection Redline and are assigned a protection intensity of 100%, ensuring
stringent protection. The assigned protection intensity values for each target land use type
in the model are outlined in Table 1.
Table 1. Protection intensity values for target land use types under different scenarios in the
SLEUTH model.
Scenarios Green Land (%) Water Body (%) Farmland (%)
A 100 100 86
B 100 100 51
C 100 100 17
D 63 100 10
E 27 100 4
We define five scenarios: A, B, C, D, and E, each representing distinct levels of pro-
tection intensity for the different land use categories. Scenarios A and E indicate two
extreme scenarios: Scenario A applies the most stringent protection measure with no
Remote Sens. 2025,17, 270 6 of 21
change allowed for the protected land use types, water body and green land. Scenario E
represents the least stringent protection measure, mirroring a business-as-usual scenario.
For the analysis, we employed the average historical growth rate of each land use category
from 1985 to 2015 as the baseline derived from the CLCD. The standard deviation (SD)
of these growth rates serves as the threshold value for the potential changes. The upper
and lower limits of change rates for all land use types in mainland Shanghai were defined
by the average growth rate adjusted by one standard deviation. The upper limit (+SD)
corresponds to Scenario A, while the lower limit (
−
SD) corresponds to Scenario E. The
specific algorithm for implementing Scenarios A and E is as follows:
r=a±SD (1)
where rrepresents the protection intensity,
a
denotes the average historical growth rate
of the corresponding land type, and SD indicates the standard deviation of the historical
growth rate. SD reflects the variability in historical growth rates across all land use types
in Shanghai. rranges from 0 to 100%, and any values exceeding 100% are capped at the
SLEUTH model’s maximum limit of 100%.
Scenario C is formulated according to the area restrictions for different land types in
the Shanghai Urban Master Plan (2017–2035). The equation below is utilized to calculate
the protection intensity level in this scenario:
r=
n
qS
S0
−1
a(2)
where rand
a
are the same as Equation (1), Sis the area of the corresponding land type
specified in the Shanghai Urban Master Plan (2017–2035), while S
0
denotes the area of the
corresponding land type in the predicted starting year, and nis the difference between the
predicted starting year and 2035. For example, when the forecast starting year is 2015, the
value of nis 20. Once the starting year is set, nbecomes a unique value used to calculate r.
This scenario quantified the probability that each land type would be protected during the
process of urbanization in alignment with the planning goal.
Scenarios B and D were designed as transitional scenarios between A and C and
between C and E, respectively. Both Scenarios B and D aimed to enhance the effectiveness
of urban growth modeling.
2.3.3. SLEUTH Model Calibration and Parameter Determination
Model calibration is an essential step in using the SLEUTH model to predict urban
growth and consists of three steps: coarse calibration, fine calibration, and final calibra-
tion. A “brute force” method is applied across the three calibration steps, progressively
narrowing the range of parameter values to identify the set that best matches the historical
data [
23
]. Each stage generates a dataset using the Lee–Salee method to establish the param-
eter ranges and step sizes for subsequent calibrations. The results of the final calibration
are then used to determine the optimal parameters.
In this study, the CLCD data from 1985, 1995, 2005, and 2015 were used for calibration.
After calibration, the finalized parameters of diffusion, breed, spread, slope, and road were
determined for the SLEUTH model and presented in Table 2. These parameters govern the
simulation of the urban growth process.
Remote Sens. 2025,17, 270 7 of 21
Table 2. The parameters determined by the SLEUTH model calibration.
Finalized Parameters END
DIFFUSION 4
BREED 1
SPREAD 39
SLOPE 56
ROAD 24
2.4. Accuracy Assessment for the Projected LULC Results
The stability and reliability of the SLEUTH model are evaluated through an accuracy
assessment of the projected LULC results, specifically urban land, water bodies, farmland,
and green land, using historical data from Yang and Huang [
47
]. Allocation disagreement
and quantity disagreement indices, as proposed by Pontius and Millones [52], are used to
evaluate the model’s performance. The accuracy assessment results, presented in
Table 3
,
reveal that the allocation disagreement ranges from 8% to 20%, while the quantity dis-
agreement falls between 2% and 11%. Additionally, for all scenarios, the standard kappa
coefficient exceeds 0.83, indicating a high level of accuracy. These results validate the
reliability of the SLEUTH model, integrated with urban planning scenarios, to accurately
simulate land use and land cover changes.
Table 3. Accuracy assessment of the projected LULC under five scenarios.
Year Scenarios Allocation
Disagreement (%)
Quantity
Disagreement (%)
Standard Kappa
Coefficient
2016 A 8 4 0.98
B 8 3 0.98
C 9 4 0.98
D 9 5 0.95
E 8 4 0.94
2017 A 11 5 0.94
B 12 5 0.96
C 12 6 0.97
D 12 7 0.94
E 13 8 0.93
2018 A 10 4 0.92
B 11 2 0.94
C 13 3 0.92
D 12 9 0.92
E 13 10 0.90
2019 A 11 5 0.91
B 10 5 0.92
C 12 7 0.88
D 16 6 0.88
E 15 5 0.87
2020 A 15 7 0.89
B 15 2 0.91
C 19 8 0.86
D 18 9 0.86
E 20 11 0.83
2.5. Quantification of the Spatiotemporal Dynamics of Urban Growth
To assess the spatiotemporal dynamics of urban growth under the five scenarios, a set
of urban growth indicators was established, including annual urban land area expansion,
annual urban land area relative growth rate, and annual urban land growth intensity [
53
,
54
].
Remote Sens. 2025,17, 270 8 of 21
We quantify the urban growth dynamics for approximately every 10-year interval during
the study period from 2016 to 2060: 2016–2025, 2025–2035, 2035–2045, and 2045–2060. These
intervals provide a detailed view of the urban growth patterns at different stages of the
urbanization process.
The average annual urban land expansion (R), which describes the annual urban land
area increase over the evaluation period, is calculated using the following formula:
R=ULAb−UL Aa
T(3)
The annual urban land relative growth rate (K), which quantifies the average percent-
age increase in urban land area relative to the base year urban land area over the evaluation
period, is calculated as follows:
K=ULAb−UL Aa
ULAa
×1
T×100% (4)
The annual urban land growth intensity (Q), which measures the percentage increase
of urban land growth over the total land area of the city in an evaluation period, is computed
as follows:
Q=ULAb−UL Aa
TLA ×1
T×100% (5)
where ULA
a
and ULA
b
are the urban land area at the start year of 2016 and the end year
of 2060, respectively. TLA is the total land area of the study area, and Trepresents the
duration of each study period.
The quadrant analysis is used for analyzing the anisotropy of urban growth. Mainland
Shanghai is divided into eight 45
◦
angle quadrants (Figure 3), with the origin of Shanghai’s
local coordinate system (121
◦
28
′
12
′′
E, 31
◦
13
′
48
′′
N) serving as the focal point for the analysis
(Figure 3). The east–west axis was taken as the reference for the horizontal direction, while
the north–south axis was designated as the vertical reference for the quadrant analysis. We
calculate the indicators, including urban land area expansion (R), urban land area relative
growth rate (K), and annual urban land growth intensity (Q) across the eight quadrants,
visualized with radar charts to effectively visualize the spatial variations in urban growth
across directions under various scenarios.
Remote Sens. 2025, 17, x FOR PEER REVIEW 8 of 22
and 2045–2060. These intervals provide a detailed view of the urban growth paerns at
different stages of the urbanization process.
The average annual urban land expansion (R), which describes the annual urban land
area increase over the evaluation period, is calculated using the following formula:
𝑅=𝑈𝐿
𝐴
−𝑈𝐿
𝐴
𝑇 (3)
The annual urban land relative growth rate (K), which quantifies the average
percentage increase in urban land area relative to the base year urban land area over the
evaluation period, is calculated as follows:
𝐾=𝑈𝐿
𝐴
−𝑈𝐿
𝐴
𝑈𝐿
𝐴
×1
𝑇×100% (4)
The annual urban land growth intensity (Q), which measures the percentage increase
of urban land growth over the total land area of the city in an evaluation period, is
computed as follows:
𝑄=𝑈𝐿
𝐴
−𝑈𝐿
𝐴
𝑇𝐿𝐴 ×1
𝑇×100% (5)
where ULAa and ULAb are the urban land area at the start year of 2016 and the end year of
2060, respectively. TLA is the total land area of the study area , and T represents the
duration of each study period.
The quadrant analysis is used for analyzing the anisotropy of urban growth.
Mainland Shanghai is divided into eight 45° angle quadrants (Figure 3), with the origin of
Shanghais local coordinate system (121°28′12″E, 31°13′48″N) serving as the focal point for
the analysis (Figure 3). The east–west axis was taken as the reference for the horizontal
direction, while the north–south axis was designated as the vertical reference for the
quadrant analysis. We calculate the indicators, including urban land area expansion (R),
urban land area relative growth rate (K), and annual urban land growth intensity (Q)
across the eight quadrants, visualized with radar charts to effectively visualize the spatial
variations in urban growth across directions under various scenarios.
Figure 3. Quadrant subdivision of the study area, with Q1–Q8 representing each equal-angle
quadrant (Q1 corresponds to the first quadrant, Q2 to the second, and so on through to Q8).
2.6. Quantification of the Changes in Urban Form
The landscape expansion index (LEI) is widely used to identify urban growth
paerns [28]. Based on previous studies, urban growth modes are typically classified into
three categories according to LEI values: infilling, edge-expansion, and leapfrog
expansion [28]. The LEI is computed as follows:
Figure 3. Quadrant subdivision of the study area, with Q1–Q8 representing each equal-angle quadrant
(Q1 corresponds to the first quadrant, Q2 to the second, and so on through to Q8).
Remote Sens. 2025,17, 270 9 of 21
2.6. Quantification of the Changes in Urban Form
The landscape expansion index (LEI) is widely used to identify urban growth pat-
terns [
28
]. Based on previous studies, urban growth modes are typically classified into three
categories according to LEI values: infilling, edge-expansion, and leapfrog expansion [
28
].
The LEI is computed as follows:
LEI =100 ×A0
A0+Av(6)
where A
0
is the area of the intersection between the buffer zone of newly developed
urban patches and the existing urban patches; A
v
is the intersection of the buffer of newly
developed urban patches with the non-urban areas. The buffer is set at 1 m, following the
approach used in previous studies [
28
]. The LEI value ranges from 0 to 100, with different
values reflecting various urban growth patterns. A value of 0 represents the leapfrog
expansion mode, while an LEI between 0 and 50 indicates the edge-expansion mode. An
LEI greater than 50 represents the infilling mode.
The mean expansion index (MEI) is calculated using the LEI, providing a detailed
quantitative analysis of urban morphological evolution [
28
]. The MEI is calculated
as follows:
MEI =∑N
i=1
LEIi
N(7)
where LEI
i
is the LEI for patch i, and Nis the total number of newly grown patches. A
larger MEI indicates a more pronounced aggregation of the landscape, reflecting a higher
compact of urban growth.
The area-weighted mean expansion index (AWMEI) is computed as follows:
AWMEI =∑N
i=1LEIi×ai
A(8)
where a
i
is the area of patch i, and Ais the total area of all new patches. A larger AWMEI
indicates a higher degree of aggregation in urban growth.
3. Results
3.1. Spatiotemporal Dynamics of the Projected Urban Land
Spatially, the projected urban land growth from 2016 to 2060 varied significantly across
four out of the five scenarios, except Scenario A (Figure 4). In Scenario A, there was minimal
growth and change in urban land. Scenario B saw moderate urban expansion to the north.
Scenarios C, D, and E displayed more extensive urban growth, with urbanization spreading
in all directions across Shanghai. The net increase in urban land areas for each scenario
was as follows: 8.6 km
2
for Scenario A, 316.0 km
2
for Scenario B, 1616.2 km
2
for Scenario C,
1773.8 km
2
for Scenario D, and 1876.1 km
2
for Scenario E. These results demonstrated that
the looser the protection measures, the greater the increase in urban land area.
Urban land area and urbanization rate changes were used to assess spatial heterogene-
ity across different directions from 2016 to 2060 (Figures 5and 6). Rapid urban growth
occurred before 2045, with the southern region, particularly the southwest, experiencing the
highest increase in urban land (Figure 5), while the northern region saw slower growth due
to already high urbanization rates (Figure 6). Throughout the study period, the northern
region consistently exhibited higher urbanization rates than the southern region.
Remote Sens. 2025,17, 270 10 of 21
Remote Sens. 2025, 17, x FOR PEER REVIEW 10 of 22
Figure 4. Spatial changes in the urban land area of mainland Shanghai from 2016 to 2060.
Urban land area and urbanization rate changes were used to assess spatial
heterogeneity across different directions from 2016 to 2060 (Figures 5 and 6). Rapid urban
growth occurred before 2045, with the southern region, particularly the southwest,
experiencing the highest increase in urban land (Figure 5), while the northern region saw
slower growth due to already high urbanization rates (Figure 6). Throughout the study
period, the northern region consistently exhibited higher urbanization rates than the
southern region.
Figure 4. Spatial changes in the urban land area of mainland Shanghai from 2016 to 2060.
Remote Sens. 2025, 17, x FOR PEER REVIEW 11 of 22
Figure 5. Urban land area changes across the eight quadrants of mainland Shanghai from 2016 to
2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C, D, and E, respectively). Q1 to Q8
represent the quadrants in numerical order.
Figure 6. Urbanization rate changes across the eight quadrants of mainland Shanghai from 2016 to
2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C, D, and E, respectively). Q1 to Q8
represent the quadrants in numerical order.
3.2. Spatiotemporal Paerns of Projected Urban Growth
The indicators of R, K, and Q showed significant temporal variations, except in
Scenario A (Table 4). The trends of these indicators were similar, showing high values
Figure 5. Urban land area changes across the eight quadrants of mainland Shanghai from 2016 to
2060 (a,b,c,d, and erepresent the changes in Scenarios A, B, C, D, and E, respectively). Q1 to Q8
represent the quadrants in numerical order.
Remote Sens. 2025,17, 270 11 of 21
Remote Sens. 2025, 17, x FOR PEER REVIEW 11 of 22
Figure 5. Urban land area changes across the eight quadrants of mainland Shanghai from 2016 to
2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C, D, and E, respectively). Q1 to Q8
represent the quadrants in numerical order.
Figure 6. Urbanization rate changes across the eight quadrants of mainland Shanghai from 2016 to
2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C, D, and E, respectively). Q1 to Q8
represent the quadrants in numerical order.
3.2. Spatiotemporal Paerns of Projected Urban Growth
The indicators of R, K, and Q showed significant temporal variations, except in
Scenario A (Table 4). The trends of these indicators were similar, showing high values
Figure 6. Urbanization rate changes across the eight quadrants of mainland Shanghai from 2016 to
2060 (a,b,c,d, and erepresent the changes in Scenarios A, B, C, D, and E, respectively). Q1 to Q8
represent the quadrants in numerical order.
3.2. Spatiotemporal Patterns of Projected Urban Growth
The indicators of R,K, and Qshowed significant temporal variations, except in Scenario A
(Table 4). The trends of these indicators were similar, showing high values before 2025,
followed by a decline after 2025, reflecting rapid growth in the early phase of the prediction.
As the land protection intensity increased, the average annual urban land expan-
sion (R) decreased. In Scenario B, Rvalues decreased from 14.02 km
2
/year during
2016–2025
to 0.67 km
2
/a during 2025–2035, stabilizing thereafter. In contrast, Scenar-
ios C, D, and E showed much higher rates, with values of 49.65 km
2
/year, 55.91 km
2
/year,
and 61.71 km2/year, respectively, from 2016 to 2025.
The annual urban land relative growth rate (K) consistently declined over time, with
higher protection intensities leading to lower Kvalues. From 2016 to 2025, Kvalues were
0.01% for Scenario A, 0.63% for Scenario B, 0.22% for Scenario C, 0.49% for Scenario D, and
2.74% for Scenario E. After 2025, Kvalues continued to decrease, remaining low and stable
after 2035.
The annual urban land growth intensity (Q) increased as protection intensities de-
creased. From 2016 to 2025, Qvalues ranged from 0.02% in Scenario A to 13.43% in
Scenario E. During 2025–2035, Qvalues remained relatively high in Scenarios C, D, and E,
while other scenarios exhibited lower values. After 2035, Qvalues remained stable.
The spatiotemporal patterns of the urban land expansion, urban land growth rate, and
urban land growth intensity followed a similar “concave shape” across different scenarios,
except in Scenario A (Figure 7). The first (Q1) and eighth quadrants (Q8) had lower values
for these indicators, although they had higher values in the early years of the study period
and gradually decreased over time. A reduction in protection intensity from Scenario B to
E resulted in a significant increase in these indicators.
Remote Sens. 2025,17, 270 12 of 21
Remote Sens. 2025, 17, x FOR PEER REVIEW 13 of 22
Figure 7. Spatiotemporal dynamics of urban land growth in the eight quadrants of mainland
Shanghai from 2016 to 2060. Q1 to Q8 represent the quadrants in numerical order.
Figure 7. Spatiotemporal dynamics of urban land growth in the eight quadrants of mainland Shanghai
from 2016 to 2060. Q1 to Q8 represent the quadrants in numerical order.
Remote Sens. 2025,17, 270 13 of 21
Table 4. The dynamics of the urban growth indicators from 2016 to 2060.
Indicators Year Scenario A Scenario B Scenario C Scenario D Scenario E
Urban land
expansion (R)
(km2/year)
2016–2025 0.11 14.02 49.65 55.91 61.71
2025–2035 0.10 0.67 31.99 33.77 33.35
2035–2045 0.10 0.67 0.79 0.82 0.85
2045–2060 0.10 0.67 0.78 0.81 0.83
Urban land
growth rate (K)
(%/year)
2016–2025 0.01 0.63 2.22 2.49 2.74
2025–2035 0 0.03 1.01 1.02 0.97
2035–2045 0 0.03 0.02 0.02 0.02
2045–2060 0 0.03 0.02 0.02 0.02
Urban land
growth intensity (Q)
(%/year)
2016–2025 0.02 3.05 10.81 12.17 13.43
2025–2035 0.02 0.15 6.96 7.35 7.26
2035–2045 0.02 0.15 0.17 0.18 0.18
2045–2060 0.02 0.15 0.17 0.18 0.18
3.3. Characteristics of Urban Form Changes
The infilling, edge-expansion, and leapfrog modes revealed wave-like patterns of
change from 2016 to 2060. Edge-expansion dominated the number of newly developed
urban patches, while the leapfrog and infilling modes fluctuated. Reduced protection
intensity (from Scenarios A to E) led to an increase in leapfrog expansion and a decrease
in infilling expansion, particularly before 2050 (Figure 8). In terms of patch area pro-
portion, the infilling mode dominated the newly developed urban patches, followed by
edge-expansion, with leapfrog expansion contributing minimally (Figure 9). As protection
intensities increased (Scenarios E to A), infilling and edge-expansion became the two major
modes of nearly balanced development.
Remote Sens. 2025, 17, x FOR PEER REVIEW 14 of 22
3.3. Characteristics of Urban Form Changes
The infilling, edge-expansion, and leapfrog modes revealed wave-like paerns of
change from 2016 to 2060. Edge-expansion dominated the number of newly developed
urban patches, while the leapfrog and infilling modes fluctuated. Reduced protection
intensity (from Scenarios A to E) led to an increase in leapfrog expansion and a decrease
in infilling expansion, particularly before 2050 (Figure 8). In terms of patch area
proportion, the infilling mode dominated the newly developed urban patches, followed
by edge-expansion, with leapfrog expansion contributing minimally (Figure 9). As
protection intensities increased (Scenarios E to A), infilling and edge-expansion became
the two major modes of nearly balanced development.
Figure 8. The proportions of newly grown patch numbers for the three urban growth modes in
mainland Shanghai from 2016 to 2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C,
D, and E, respectively).
Figure 8. The proportions of newly grown patch numbers for the three urban growth modes in
mainland Shanghai from 2016 to 2060 (a,b,c,d, and erepresent the changes in Scenarios A, B, C, D,
and E, respectively).
Remote Sens. 2025,17, 270 14 of 21
Remote Sens. 2025, 17, x FOR PEER REVIEW 14 of 22
3.3. Characteristics of Urban Form Changes
The infilling, edge-expansion, and leapfrog modes revealed wave-like paerns of
change from 2016 to 2060. Edge-expansion dominated the number of newly developed
urban patches, while the leapfrog and infilling modes fluctuated. Reduced protection
intensity (from Scenarios A to E) led to an increase in leapfrog expansion and a decrease
in infilling expansion, particularly before 2050 (Figure 8). In terms of patch area
proportion, the infilling mode dominated the newly developed urban patches, followed
by edge-expansion, with leapfrog expansion contributing minimally (Figure 9). As
protection intensities increased (Scenarios E to A), infilling and edge-expansion became
the two major modes of nearly balanced development.
Figure 8. The proportions of newly grown patch numbers for the three urban growth modes in
mainland Shanghai from 2016 to 2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C,
D, and E, respectively).
Figure 9. The proportions of newly grown patch area for the three urban growth modes in mainland
Shanghai from 2016 to 2060 (a,b,c,d, and erepresent the changes in Scenarios A, B, C, D, and
E, respectively).
The MEI and AWMEI indices revealed the changes in urban form over time. In
Scenario A, the MEI remained relatively stable with small fluctuations. In Scenario B, the
MEI exhibited an initial rise followed by a slight downward trend. In Scenarios C, D, and
E, the MEI showed a trend of continuous decline to a small value, followed by a sudden
increase. The AWMEI displayed a dynamic pattern of decline followed by an increase in
Scenario A. In Scenario B, AWMEI exhibited multiple cycles of rise and fall. In Scenarios
C, D, and E, the AWMEI showed a slight increase until around 2050, suggesting a phase
of urban coalescence. This was followed by a gradual decline, reflecting the subsequent
shift toward urban diffusion. As protection intensity decreased from Scenarios C to E, there
were no significant changes in the MEI and AWMEI (Figure 10).
Remote Sens. 2025, 17, x FOR PEER REVIEW 15 of 22
Figure 9. The proportions of newly grown patch area for the three urban growth modes in mainland
Shanghai from 2016 to 2060 (a, b, c, d, and e represent the changes in Scenarios A, B, C, D, and E,
respectively).
The MEI and AWMEI indices revealed the changes in urban form over time. In
Scenario A, the MEI remained relatively stable with small fluctuations. In Scenario B, the
MEI exhibited an initial rise followed by a slight downward trend. In Scenarios C, D, and
E, the MEI showed a trend of continuous decline to a small value, followed by a sudden
increase. The AWMEI displayed a dynamic paern of decline followed by an increase in
Scenario A. In Scenario B, AWMEI exhibited multiple cycles of rise and fall. In Scenarios
C, D, and E, the AWMEI showed a slight increase until around 2050, suggesting a phase
of urban coalescence. This was followed by a gradual decline, reflecting the subsequent
shift toward urban diffusion. As protection intensity decreased from Scenarios C to E,
there were no significant changes in the MEI and AWMEI (Figure 10).
Figure 10. Temporal changes in the MEI and AWMEI of mainland Shanghai from 2016 to 2060 (a,
b, c, d, and e represent the changes in Scenarios A, B, C, D, and E, respectively).
4. Discussion
4.1. The Paerns of Urban Land Growth
Modeling urban growth and projecting future scenarios are crucial for guiding urban
planning and promoting sustainable development amid urbanization challenges [55,56].
Previous studies had developed scenarios by defining extreme case gradients [57–61],
incorporating local factors as exclusion layers [23], or applying baseline urban
development strategies [62]. In this study, we developed multiple scenarios by integrating
the Shanghai Urban Master Plan (2017–2035) into the SLEUTH model. The effectiveness
of this approach was validated by an accuracy assessment of the predictive result (Table
3). Our findings revealed that urban growth would experience significant changes,
especially under Scenarios C, D, and E (Figure 4). These results highlighted the varying
impacts of protection measures on urban growth and form changes, suggesting that urban
planning strategies, particularly land use regulations, were crucial in shaping the
trajectory of urban development. A previous study on urban predictions for Shanghai
indicated that by 2035, the citys urban land would expand nearly 13-fold over 50 years
[41]. Another study forecasted a 33.9% increase in urban area by 2030, within just 15 years
[63]. Our projected results suggested that under Scenario C (planning scenario), the urban
Figure 10. Temporal changes in the MEI and AWMEI of mainland Shanghai from 2016 to 2060 (a,b,c,
d, and erepresent the changes in Scenarios A, B, C, D, and E, respectively).
Remote Sens. 2025,17, 270 15 of 21
4. Discussion
4.1. The Patterns of Urban Land Growth
Modeling urban growth and projecting future scenarios are crucial for guiding urban
planning and promoting sustainable development amid urbanization challenges [
55
,
56
].
Previous studies had developed scenarios by defining extreme case gradients [
57
–
61
],
incorporating local factors as exclusion layers [
23
], or applying baseline urban develop-
ment strategies [
62
]. In this study, we developed multiple scenarios by integrating the
Shanghai Urban Master Plan (2017–2035) into the SLEUTH model. The effectiveness of
this approach was validated by an accuracy assessment of the predictive result (Table 3).
Our findings revealed that urban growth would experience significant changes, especially
under Scenarios C, D, and E (Figure 4). These results highlighted the varying impacts
of protection measures on urban growth and form changes, suggesting that urban plan-
ning strategies, particularly land use regulations, were crucial in shaping the trajectory of
urban development. A previous study on urban predictions for Shanghai indicated that
by 2035, the city’s urban land would expand nearly 13-fold over 50 years [
41
]. Another
study forecasted a 33.9% increase in urban area by 2030, within just 15 years [
63
]. Our
projected results suggested that under Scenario C (planning scenario), the urban area would
expand by about 1616.2 km
2
by 2060. This suggested that Shanghai’s urban development
was expected to experience rapid growth. By 2035, the urban land area was projected to
reach 3179.7 km
2
, closely approaching the 3200 km
2
target outlined in the Shanghai Urban
Master Plan. This indicated that to meet the target, effective control of urban growth would
be necessary.
This study highlighted significant spatiotemporal dynamics in urban growth across
the different directions of mainland Shanghai, with variations in urban land area and
urbanization rates (Figures 5and 6). Notably, urban land growth was expected to expand
significantly in most directions (Figure 5), particularly in southern Shanghai, with limited
growth in the north. This disparity can be attributed to the earlier urbanization in the
north, where the old urban districts were located, leaving less room for further growth [
64
].
Previous studies identified Shanghai’s early urbanization process: from 1979 to 2000,
growth occurred primarily along the north–south axis. After 2000, urban growth spread
more evenly in all directions [
65
]. Another study also revealed a significant shift in the
geographical center of urban growth in Shanghai over the past five decades, moving
from the northeast to the southwest [
41
]. These changes reflected a shift in Shanghai’s
development strategy, with urban planners intentionally directing growth toward the north
to achieve more balanced urban development [
66
]. The urbanization rate was expected
to continue growing from 2016 to 2060, such as in the second (Q2), third (Q3), fifth (Q7),
and sixth (Q8) quadrants (Figure 6). In Scenarios C, D, and E, the eighth quadrant (Q8)
showed the highest urbanization rates in mainland Shanghai during the late simulation
period (2045–2060). This may be attributed to government-driven urban transformation,
along with significant real estate and foreign investments, which stimulated urbanization
in these areas [64].
4.2. Characteristics of the Spatiotemporal Changes in Urban Growth
The analysis of the spatiotemporal dynamics of urban growth is crucial for under-
standing urban morphological evolution. Spatial indicators play a key role in quantifying
these dynamics [
53
,
67
]. Our results showed that the spatiotemporal patterns of the average
annual urban land expansion, annual urban land relative growth rate, and annual urban
land growth intensity highlighted the future changes in urban land from 2016 to 2060
under various scenarios. These patterns exhibited similar temporal trends and spatial
distribution (Table 4and Figure 7), with notable differences between the early (pre-2025)
Remote Sens. 2025,17, 270 16 of 21
and late (post-2025) periods of urban growth (Table 4). A study of urban growth in 18 cities
in China from 1980 to 2008 found that Shanghai’s historical urban growth dynamics fol-
lowed a fluctuating trend, with the growth intensity initially declining until 2000, before
increasing again after 2000 [
53
]. Other studies demonstrated that urban land growth rates
also fluctuated, alternating with increasing and decreasing trends [
6
,
68
]. For example, a
study of Shanghai from 1978 to 2015 showed an initial increase in growth followed by a
decline, with a peak annual area increase of 118.22 km
2
and annual growth rate of 7.56%
during 2005–2010 [
68
]. Our projections suggested that the peak annual rate and intensity of
urban growth would occur between 2016 and 2025, followed by a decline in the later stages
of future urbanization. The decline in urban growth rates after 2025 could be attributed to
two primary reasons. First, urban area growth followed a logistic pattern, which showed
that urban growth rate reached a peak before gradually declining to a stable state [
6
].
Our modeling of urban growth trajectories under different scenarios corroborated this
phenomenon. Second, during the early stage of the projection period, there was ample land
available for development. However, as land availability became increasingly constrained,
opportunity for expansion diminished, resulting in a deceleration of the growth rate, as
evidenced by the decreasing growth rates observed after 2025.
Spatially, the anisotropy of urban growth, characterized by a “concave shape”, was
clearly shown in Figure 7, particularly under Scenarios C, D, and E. Specifically, the av-
erage annual urban land expansion (R), annual urban land relative growth rate (K), and
annual urban land growth intensity (Q) exhibited significant variations across all areas
except the first (Q1) and eighth (Q8) quadrants, with differing degrees of change across
the quadrants. The southern and eastern regions experienced greater changes, surpassing
the northern region, which aligned with previous findings. Previous studies showed that
during the later stages of growth (2020–2035), urban development in Shanghai shifted from
the northeast to the southwest regions [
41
]. In each period, differences in average annual
urban land expansion (R) were not significant, but considerable differences in annual urban
land relative growth rate (K) were observed (Figure 7). These anisotropic patterns may be
influenced by factors such as policies, socioeconomic/natural conditions (e.g., topography
and environmental factors), and the evolutionary dynamics of each quadrant [
69
]. Since
Shanghai is predominantly flat with few mountainous regions and using the geographic
center as the center for quadrant analysis, we mitigated the potential bias from irregular
land distribution or density variations. As a result, Shanghai’s anisotropic urban dynam-
ics may be greatly shaped by policy and socioeconomic influences. Recent studies have
provided evidence that urban development in Shanghai was largely driven by top-down
government regulations and policies, including central government-led suburban indus-
trial zone construction and local government initiatives for developing new urban areas
and towns [40].
4.3. The Evolution of the Urban Form
Urban morphological changes are shaped by geographical factors and the level of eco-
nomic development [
70
]. As cities evolve, urban areas can undergo either scattered or com-
pact growth patterns [
71
]. The diffusion–coalescence hypothesis of urban morphology has
been widely tested through historical urban land use data across multiple scales
[6,27,29]
.
In this research, the landscape expansion index (LEI) was used to reveal changes in urban
growth patterns. Unlike previous studies that observed the alternating dominance of infill-
ing, edge-expansion, and leapfrog modes [
6
,
29
], the results of this study showed that under
different scenarios, urban growth would be dominated by edge-expansion in terms of patch
number proportion and by infilling in terms of patch area proportion (
Figures 8and 9
). In
Scenarios C, D, and E, however, infilling and leapfrog processes emerged as the secondary
Remote Sens. 2025,17, 270 17 of 21
modes, exhibiting alternating patterns. This may be attributed to variations in economic
development, planning, policies, or land use strategies [
6
]. Such changes indicated the
complexity and non-linearity of the urban growth process.
The changes in MEI and AWMEI provided more detailed evidence regarding urban-
ization form changes (Figure 10). In Scenarios B, C, D, and E, the overall trends of MEI and
AWMEI showed opposite patterns. This discrepancy was likely due to the emergence of a
larger, more aggregated urban patch. As urban growth progresses, larger urban areas influ-
enced the changes in AWMEI, causing an initial increase followed by a decline. It revealed
a shift from coalescence to diffusion in urban growth. This trend reflected the dominance of
the infilling mode, resulting in more of a coalescence urban form. In later stages, increased
edge-expansion led to a diffusion form. Similar patterns were observed in previous stud-
ies [
29
,
41
]. The existing research has identified a spiraling diffusion–coalescence process in
urbanization [
6
,
27
], which marked two distinct rounds of urbanization. In this study, we
found a gradual rise and decline pattern in AWMEI, suggesting that urban form followed
a coalescence to diffusion pattern characterized by a single round of urbanization. This
was likely due to the comparatively strong land protection policies, which constrained
the extensive expansion of urban land. Consequently, each diffusion or coalescence phase
lasted longer.
4.4. Limitations and Implications
Urban growth models have been developed and widely utilized to forecast urban
development [
56
]. In this study, we utilized the SLEUTH model to simulate urban growth.
However, there are still some limitations in its application. First, the SLEUTH model
simulates urban growth but cannot capture processes such as urban renewal [
64
] or green
infrastructure regeneration [
72
], which may reduce its accuracy in reflecting real-world
urban dynamics and trends. This limitation could potentially lead to biases in the projected
urban growth patterns, particularly in areas where regeneration and redevelopment play
significant roles in shaping urban form. Second, previous studies have incorporated
principles such as landscape ecology principles [
22
] and habitat quality [
23
] as exclusion
layers to restrict development in specific areas. In this study, we primarily focused on the
land use requirements outlined in the Shanghai Urban Master Plan to define development
scenarios, excluding those based on specific environmental or ecological constraints. This
approach allowed us to assess urban growth dynamics within the framework of planned
development but may not fully account for potential limitations imposed by specific
habitat protections, ecological redlines, and other environmental constraints. Therefore,
future work could integrate these factors to provide a more comprehensive and realistic
simulation of urban growth. Third, we used only four land use types, i.e., farmland, green
land, water bodies, and urban land, in the simulation. However, the Shanghai Urban
Master Plan outlines more detailed land use requirements for various categories by 2035,
such as public infrastructure and transportation facilities. Therefore, incorporating high-
resolution imagery to obtain more detailed land use classifications could provide more
precise information on the development of different land use types, enabling more refined
urban planning.
This study employs urban growth models to predict future urban growth under
various scenarios, providing insights into urban growth dynamics and the evolution of
different growth modes such as edge-expansion, infilling, and leapfrog modes, as well as
the associated urban form changes. By integrating urban planning strategies, the model
facilitates the clearer identification of urban land use distributions and growth patterns,
helping planners to assess the long-term impacts under different development scenarios.
The planning targets used to create the scenarios can, in fact, be substituted with other
Remote Sens. 2025,17, 270 18 of 21
relevant sources, such as ecological protection guidelines, provided they contain specific
data on the targeted land use restrictions. This flexibility allows the model to be tailored to
various contexts while maintaining its applicability. Thus, the framework developed for
predicting future urban growth and urban form is highly transferable and can be effectively
applied to various urban settings, offering decision makers valuable tools for planning,
monitoring, and addressing future urban challenges.
5. Conclusions
This study simulated the spatiotemporal dynamics of urban growth and form evo-
lution in mainland Shanghai from 2016 to 2060 under five urbanization scenarios, using
the SLEUTH model integrated with the Shanghai Urban Master Plan (2017–2035). The
results revealed significant variations in urban growth across different scenarios, with the
southern region experiencing the most growth and clear anisotropy. As protection intensity
decreased, both the rate and intensity of urban growth increased. Urban growth indicators,
including the average annual urban land expansion, annual urban land relative growth
rate, and annual urban land growth intensity fluctuated significantly over time, exhibiting
a “concave shape” pattern. Rapid growth occurred before 2025, followed by a decline.
Growth patterns exhibited wave-like patterns, with edge-expansion dominating the number
of newly developed patches, while leapfrog and infilling modes fluctuated. As protection
intensity decreased, leapfrog expansion increased and infilling decreased, particularly
before 2050. In terms of the patch area, the infilling mode was dominant, followed by edge-
expansion. These shifts indicated that the urban form followed a diffusion–coalescence
process. The findings provide valuable insights into the future dynamics of urban form evo-
lution in Shanghai, offering valuable information for urban planning and land management
policies, particularly in the context of rapid urbanization and ecological constraints.
Author Contributions: Conceptualization, J.L., C.W. and Y.L.; methodology, Y.L., C.W. and J.L.;
software, Y.L., C.W. and J.W.; validation, Y.L., J.W., Y.Z., X.B. and M.W.; formal analysis, Y.L. and
J.W.; investigation, Y.L., J.W., Y.Z., X.B. and M.W.; resources, Y.L. and J.L.; data curation, Y.L. and
J.W.; writing—original draft preparation, Y.L.; writing—review and editing, C.W., J.L., C.S. and
E.Y.; visualization, Y.L. and J.W.; supervision, J.L. and E.Y.; project administration, J.L. and C.W.;
funding acquisition, J.L. and C.W. All authors have read and agreed to the published version of
the manuscript.
Funding: This study was partly supported by the National Key R&D Project of China (Grant No.
2022YFF1301105 to J. Li), the Natural Science Foundation of China (Grant No. 31971485 to J. Li, Grant
No. 32001162 to C. Wu), the China Postdoctoral Science Foundation (2021M702131 to C. Wu), and
the Joint-PhD project of Shanghai Jiao Tong University and The University of Melbourne to J. Li and
A. Hahs.
Data Availability Statement: The data presented in this study are available on request from the author.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1. United Nations (UN). World Urbanization Prospects 2018; United Nations (UN): New York, NY, USA, 2019.
2.
Wentz, E.A.; York, A.M.; Alberti, M.; Conrow, L.; Fischer, H.; Inostroza, L.; Jantz, C.; Pickett, S.T.A.; Seto, K.C.; Taubenböck, H. Six
fundamental aspects for conceptualizing multidimensional urban form: A spatial mapping perspective. Landsc. Urban Plan. 2018,
179, 55–62. [CrossRef]
3.
Cai, Z.; Demuzere, M.; Tang, Y.; Wan, Y. The characteristic and transformation of 3D urban morphology in three Chinese
mega-cities. Cities 2022,131, 103988. [CrossRef]
4.
Krueger, E.H.; Constantino, S.M.; Centeno, M.A.; Elmqvist, T.; Weber, E.U.; Levin, S.A. Governing sustainable transformations of
urban social-ecological-technological systems. npj Urban Sustain. 2022,2, 10. [CrossRef]
Remote Sens. 2025,17, 270 19 of 21
5.
He, X.; Zhou, Y. Urban spatial growth and driving mechanisms under different urban morphologies: An empirical analysis of
287 Chinese cities. Landsc. Urban Plan. 2024,248, 105096. [CrossRef]
6.
Wu, C.; Li, C.; Ouyang, L.; Xiao, H.; Wu, J.; Zhuang, M.; Bi, X.; Li, J.; Wang, C.; Song, C.; et al. Spatiotemporal evolution
of urbanization and its implications to urban planning of the megacity, Shanghai, China. Landsc. Ecol. 2023,38, 1105–1124.
[CrossRef]
7.
Liu, Y.; He, Q.; Tan, R.; Liu, Y.; Yin, C. Modeling different urban growth patterns based on the evolution of urban form: A case
study from Huangpi, Central China. Appl. Geogr. 2016,66, 109–118. [CrossRef]
8.
Wang, Q.; Guan, Q.; Lin, J.; Luo, H.; Tan, Z.; Ma, Y. Simulating land use/land cover change in an arid region with the coupling
models. Ecol. Indic. 2021,122, 107231. [CrossRef]
9.
Shafizadeh-Moghadam, H.; Asghari, A.; Tayyebi, A.; Taleai, M. Coupling machine learning, tree-based and statistical models
with cellular automata to simulate urban growth. Comput. Environ. Urban Syst. 2017,64, 297–308. [CrossRef]
10.
Yeh, A.G.O.; Li, X.; Xia, C. Cellular Automata Modeling for Urban and Regional Planning. In Urban Informatics; Shi, W.,
Goodchild, M.F., Batty, M., Kwan, M.-P., Zhang, A., Eds.; Springer: Singapore, 2021; pp. 865–883.
11.
Clarke, K.C.; Hoppen, S.; Gaydos, L. A self-modifying cellular automaton model of historical urbanization in the San Francisco
Bay area. Environ. Plan. B Plan. Des. 1997,24, 247–261. [CrossRef]
12.
Zhou, Y.; Varquez, A.C.G.; Kanda, M. High-resolution global urban growth projection based on multiple applications of the
SLEUTH urban growth model. Sci. Data 2019,6, 34. [CrossRef]
13.
Liu, D.; Clarke, K.C.; Chen, N. Integrating spatial nonstationarity into SLEUTH for urban growth modeling: A case study in the
Wuhan metropolitan area. Comput. Environ. Urban Syst. 2020,84, 101545. [CrossRef]
14.
Dietzel, C.; Clarke, K. The effect of disaggregating land use categories in cellular automata during model calibration and
forecasting. Comput. Environ. Urban Syst. 2006,30, 78–101. [CrossRef]
15.
Varquez, A.C.G.; Dong, S.; Hanaoka, S.; Kanda, M. Evaluating future railway-induced urban growth of twelve cities using
multiple SLEUTH models with open-source geospatial inputs. Sustain. Cities Soc. 2023,91, 104442. [CrossRef]
16. Cui, L.; Shi, J. Urbanization and its environmental effects in Shanghai, China. Urban Clim. 2012,2, 1–15. [CrossRef]
17.
Zheng, K.; Xu, X.; Zhang, X.; Liu, L. Spatial-temporal characteristics and future prediction of urban expansion in Shanghai.
J. Geo-Inf. Sci. 2012,14, 490–496. [CrossRef]
18.
Chandan, M.C.; Nimish, G.; Bharath, H.A. Analysing spatial patterns and trend of future urban expansion using SLEUTH. Spat.
Inf. Res. 2020,28, 11–23. [CrossRef]
19.
Liang, Y.; Liu, L. Spatiotemporal Dynamics of Rapid Urban Growth on the Loess Plateau from 1995 to 2050. J. Resour. Ecol. 2023,
14, 567–580. [CrossRef]
20.
Gao, L.; Tao, F.; Liu, R.; Wang, Z.; Leng, H.; Zhou, T. Multi-scenario simulation and ecological risk analysis of land use based on
the PLUS model: A case study of Nanjing. Sustain. Cities Soc. 2022,85, 104055. [CrossRef]
21.
Chen, Y.; Wang, J.; Xiong, N.; Sun, L.; Xu, J. Impacts of Land Use Changes on Net Primary Productivity in Urban Agglomerations
under Multi-Scenarios Simulation. Remote Sens. 2022,14, 1755. [CrossRef]
22.
Akyol Alay, M.; Tunçay, H.E.; Clarke, K.C. SLEUTH modeling informed by landscape ecology principles: Case study using
scenario-based planning in Sariyer, Istanbul, Turkey. J. Urban Plan. Dev. 2021,147, 05021043. [CrossRef]
23.
Li, F.; Wang, L.; Chen, Z.; Clarke, K.C.; Li, M.; Jiang, P. Extending the SLEUTH model to integrate habitat quality into urban
growth simulation. J. Environ. Manag. 2018,217, 486–498. [CrossRef] [PubMed]
24.
Choudhury, U.; Kanga, S.; Singh, S.K.; Kumar, A.; Meraj, G.; Kumar, P.; Singh, S. Projecting Urban Expansion by Analyzing
Growth Patterns and Sustainable Planning Strategies—A Case Study of Kamrup Metropolitan, Assam, North-East India. Earth
2024,5, 169–194. [CrossRef]
25.
Rao, Y.; Dai, J.; Dai, D.; He, Q.; Wang, H. Effect of Compactness of Urban Growth on Regional Landscape Ecological Security.
Land 2021,10, 848. [CrossRef]
26.
He, Q.; He, W.; Song, Y.; Wu, J.; Yin, C.; Mou, Y. The impact of urban growth patterns on urban vitality in newly built-up areas
based on an association rules analysis using geographical ‘big data’. Land Use Policy 2018,78, 726–738. [CrossRef]
27.
Li, C.; Li, J.; Wu, J. Quantifying the speed, growth modes, and landscape pattern changes of urbanization: A hierarchical patch
dynamics approach. Landsc. Ecol. 2013,28, 1875–1888. [CrossRef]
28.
Liu, X.; Li, X.; Chen, Y.; Tan, Z.; Li, S.; Ai, B. A new landscape index for quantifying urban expansion using multi-temporal
remotely sensed data. Landsc. Ecol. 2010,25, 671–682. [CrossRef]
29.
Wu, J.; Wu, C.; Zhang, Q.; Zhuang, M.; Xiao, H.; Wu, H.; Ouyang, L.; Liu, Y.; Meng, C.; Song, C.; et al. Spatiotemporal Evolution
of Urban Agglomeration and Its Impact on Landscape Patterns in the Pearl River Delta, China. Remote Sens. 2023,15, 2520.
[CrossRef]
30.
He, Q.; Song, Y.; Liu, Y.; Yin, C. Diffusion or coalescence? Urban growth pattern and change in 363 Chinese cities from 1995 to
2015. Sustain. Cities Soc. 2017,35, 729–739. [CrossRef]
Remote Sens. 2025,17, 270 20 of 21
31.
Chakraborty, S.; Maity, I.; Dadashpoor, H.; Novotn
˙
y, J.; Banerji, S. Building in or out? Examining urban expansion patterns and
land use efficiency across the global sample of 466 cities with million+ inhabitants. Habitat Int. 2022,120, 102503. [CrossRef]
32.
Rybski, D.; Li, Y.; Born, S.; Kropp, J.P. Modeling urban morphology by unifying diffusion-limited aggregation and stochastic
gravitation. Findings 2021. [CrossRef]
33.
Jatayu, A.; Saizen, I.; Rustiadi, E.; Pribadi, D.O.; Juanda, B. Urban Form Dynamics and Modelling towards Sustainable Hinterland
Development in North Cianjur, Jakarta–Bandung Mega-Urban Region. Sustainability 2022,14, 907. [CrossRef]
34.
Haselsteiner, E.; Smetschka, B.; Remesch, A.; Gaube, V. Time-Use Patterns and Sustainable Urban Form: A Case Study to Explore
Potential Links. Sustainability 2015,7, 8022–8050. [CrossRef]
35.
Peng, W.; Wang, G.; Zhou, J.; Zhao, J.; Yang, C. Studies on the temporal and spatial variations of urban expansion in Chengdu,
western China, from 1978 to 2010. Sustain. Cities Soc. 2015,17, 141–150. [CrossRef]
36.
Yang, Y.; Liu, Y.; Li, Y.; Du, G. Quantifying spatio-temporal patterns of urban expansion in Beijing during 1985–2013 with
rural-urban development transformation. Land Use Policy 2018,74, 220–230. [CrossRef]
37.
Miles, L.S.; Carlen, E.J.; Winchell, K.M.; Johnson, M.T.J. Urban evolution comes into its own: Emerging themes and future
directions of a burgeoning field. Evol. Appl. 2021,14, 3–11. [CrossRef]
38.
Monteiro, J.; Sousa, N.; Coutinho-Rodrigues, J.; Natividade-Jesus, E. Challenges Ahead for Sustainable Cities: An Urban Form
and Transport System Review. Energies 2024,17, 409. [CrossRef]
39.
Woldesemayat, E.M.; Genovese, P.V. Monitoring Urban Expansion and Urban Green Spaces Change in Addis Ababa: Directional
and Zonal Analysis Integrated with Landscape Expansion Index. Forests 2021,12, 389. [CrossRef]
40.
Wang, S.; Luo, X. The evolution of government behaviors and urban expansion in Shanghai. Land Use Policy 2022,114, 105973.
[CrossRef]
41.
Gao, C.; Feng, Y.; Wang, R.; Lei, Z.; Chen, S.; Tang, X.; Xi, M. 50-Year Urban Expansion Patterns in Shanghai: Analysis Using
Impervious Surface Data and Simulation Models. Land 2023,12, 2065. [CrossRef]
42.
Li, J.; Wang, Y.; Song, Y.; Wang, H. GIS-based analysis of spatial characteristics of watercourse and its water pollution in Shanghai
City. China Environ. Sci. 2004,24, 632–635, (In Chinese with English abstract).
43.
Li, J.; Li, C.; Zhu, F.; Song, C.; Wu, J. Spatiotemporal pattern of urbanization in Shanghai, China between 1989 and 2005. Landsc.
Ecol. 2013,28, 1545–1565. [CrossRef]
44. SMBS (Shanghai Municipal Bureau of Statistics). Shanghai Statistical Yearbook 2023; SMBS: Shanghai, China, 2023.
45. SMBS (Shanghai Municipal Bureau of Statistics). Shanghai Statistical Yearbook 2022; SMBS: Shanghai, China, 2022.
46.
Silva, E.A.; Clarke, K.C. Calibration of the SLEUTH urban growth model for Lisbon and Porto, Portugal. Comput. Environ. Urban
Syst. 2002,26, 525–552. [CrossRef]
47.
Yang, J.; Huang, X. The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019. Earth Syst. Sci. Data 2021,13,
3907–3925. [CrossRef]
48.
Li, J.; Song, C.; Cao, L.; Zhu, F.; Meng, X.; Wu, J. Impacts of landscape structure on surface urban heat islands: A case study of
Shanghai, China. Remote Sens. Environ. 2011,115, 3249–3263. [CrossRef]
49. Xiang, W.; Clarke, K.C. The use of scenarios in land-use planning. Environ. Plan. B Plan. Des. 2003,30, 885–909. [CrossRef]
50.
Jantz, C.A.; Goetz, S.J.; Shelley, M.K. Using the SLEUTH urban growth model to simulate the impacts of future policy scenarios
on urban land use in the Baltimore-Washington metropolitan area. Environ. Plan. B Plan. Des. 2004,31, 251–271. [CrossRef]
51. SMPG (Shanghai Municipal People’s Government). Shanghai Urban Master Plan (2017–2035); SMPG: Shanghai, China, 2018.
52.
Pontius, R.G.; Millones, M. Death to Kappa: Birth of quantity disagreement and allocation disagreement for accuracy assessment.
Int. J. Remote Sens. 2011,32, 4407–4429. [CrossRef]
53.
Xu, X.; Min, X. Quantifying spatiotemporal patterns of urban expansion in China using remote sensing data. Cities 2013,35,
104–113. [CrossRef]
54.
Chen, Y.; Li, Z.; Li, P.; Zhang, Y.; Liu, H.; Pan, J. Impacts and Projections of Land Use and Demographic Changes on Ecosystem
Services: A Case Study in the Guanzhong Region, China. Sustainability 2022,14, 3003. [CrossRef]
55.
Kii, M. Projecting future populations of urban agglomerations around the world and through the 21st century. npj Urban Sustain.
2021,1, 10. [CrossRef]
56. Li, X.; Gong, P. Urban growth models: Progress and perspective. Sci. Bull. 2016,61, 1637–1650. [CrossRef]
57.
Xi, F.; Hu, Y.; He, H.; Shi, T.; Bu, R.; Wu, X.; Zhu, J. Urban planning based on SLEUTH model in Shenyang-Fushun metropolitan
area. J. Univ. Chin. Acad. Sci. 2009,26, 765–773. [CrossRef]
58.
Liu, Y.; Liu, X. Applying SLEUTH for simulating urban expansion of Hangzhou. In Proceedings of the Second International
Conference on Earth Observation for Global Changes, Chengdu, China, 25–29 May 2009; pp. 42–49.
59.
Liu, X.; Sun, R.; Yang, Q.; Su, G.; Qi, W. Simulating urban expansion using an improved SLEUTH model. J. Appl. Remote Sens.
2012,6, 061709. [CrossRef]
60.
Hua, L.; Tang, L.; Cui, S.; Yin, K. Simulating urban growth using the SLEUTH model in a coastal peri-urban district in China.
Sustainability 2014,6, 3899–3914. [CrossRef]
Remote Sens. 2025,17, 270 21 of 21
61.
Zheng, L.; Zhang, D.; Zhou, Y.; Zhang, X.; Shi, R.; Chen, M. Simulation of land use/cover change in Shanghai based on
SLEUTH model. In Proceedings of the Remote Sensing and Modeling of Ecosystems for Sustainability XV, San Diego, CA, USA,
22 August 2018; pp. 208–216.
62.
Zhang, Q.; Ban, Y.; Liu, J.; Hu, Y. Simulation and analysis of urban growth scenarios for the Greater Shanghai Area, China.
Comput. Environ. Urban Syst. 2011,35, 126–139. [CrossRef]
63.
Zhou, L.; Dang, X.; Sun, Q.; Wang, S. Multi-scenario simulation of urban land change in Shanghai by random forest and
CA-Markov model. Sustain. Cities Soc. 2020,55, 102045. [CrossRef]
64.
Yue, W.; Fan, P.; Wei, Y.D.; Qi, J. Economic development, urban expansion, and sustainable development in Shanghai. Stoch.
Environ. Res. Risk Assess. 2014,28, 783–799. [CrossRef]
65.
Yin, J.; Yin, Z.; Zhong, H.; Xu, S.; Hu, X.; Wang, J.; Wu, J. Monitoring urban expansion and land use/land cover changes of
Shanghai metropolitan area during the transitional economy (1979–2009) in China. Environ. Monit. Assess. 2011,177, 609–621.
[CrossRef]
66.
Hu, Y.n.; Connor, D.S.; Stuhlmacher, M.; Peng, J.; Turner Ii, B.L. More urbanization, more polarization: Evidence from two
decades of urban expansion in China. npj Urban Sustain. 2024,4, 33. [CrossRef]
67.
Ma, Y.; Xu, R. Remote sensing monitoring and driving force analysis of urban expansion in Guangzhou City, China. Habitat Int.
2010,34, 228–235. [CrossRef]
68.
Fei, W.; Zhao, S. Urban land expansion in China’s six megacities from 1978 to 2015. Sci. Total Environ. 2019,664, 60–71. [CrossRef]
[PubMed]
69.
Zhang, J.; Ling, Y.; Zhu, A.X.; Zeng, H.; Song, J.; Zhu, Y.; Qian, L. Incorporation of spatial anisotropy in urban expansion
modelling with cellular automata. Int. J. Geogr. Inf. Sci. 2022,36, 86–113. [CrossRef]
70.
Schneider, A.; Woodcock, C.E. Compact, dispersed, fragmented, extensive? A comparison of urban growth in twenty-five global
cities using remotely sensed data, pattern metrics and census information. Urban Stud. 2008,45, 659–692. [CrossRef]
71.
Deng, H.; Zhang, K.; Wang, F.; Dang, A. Compact or disperse? Evolution patterns and coupling of urban land expansion and
population distribution evolution of major cities in China, 1998–2018. Habitat Int. 2021,108, 102324. [CrossRef]
72.
Lin, B.; Meyers, J.; Barnett, G. Understanding the potential loss and inequities of green space distribution with urban densification.
Urban For. Urban Green. 2015,14, 952–958. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
Available via license: CC BY 4.0
Content may be subject to copyright.