R E S E A R CH A R T I C L E

Long-term monitoring and evaluation of land developmentin a reclamation area under rapid urbanization: A case-studyin Qiantang New District, China

Tangao Hu1 | Jinjin Fan1 | Hao Hou1 | Yao Li1 | Yue Li2 |

Kangning Huang3

1Zhejiang Provincial Key Laboratory of Urban

Wetlands and Regional Change, Hangzhou

Normal University, Hangzhou, PR China

2Division of Environment and Sustainability,

Hong Kong University of Science and

Technology, Kowloon, Hong Kong, PR China

3Research Application Laboratory, National

Center for Atmospheric Research, Boulder,

Colorado

Correspondence

Hao Hou, Zhejiang Provincial Key Laboratory

of Urban Wetlands and Regional Change,

Hangzhou Normal University, Hangzhou,

311121, PR China.

Email: houhao@hznu.edu.cn

Funding information

Science and Technology Program of Hangzhou,

Grant/Award Number: 20191203B19;

Zhejiang Provincial Natural Science Foundation

of China, Grant/Award Numbers:

LQ20D010008, LY19D010004,

LGF18D010005; Zhejiang Provincial Natural

Science Foundation

Abstract

Land reclamation has occurred extensively worldwide to accommodate urbanization

and economic development, especially in developing countries like China. However,

we have a limited understanding of the long-term dynamics and key drivers of land

use/cover change in the reclaimed area. In this study, we monitored the detailed spa-

tiotemporal evolution of land reclamation from 1973 to 2018 in Qiantang New Dis-

trict using time-series LANDSAT and SENTINEL-2A images and then compared the

differences of landscape changes between reclaimed, which are the new-built land

from Qiantang River, and inland areas. Key findings include: (1) A significant decreas-

ing trend for areas near the Qiantang River along the coastline (212.21 to 80.99 km2)

and an increasing trend for constructed land (10.05 to 120.89 km2) from 1973 to

2018 was detected; (2) The development modes of the inland area and reclaimed

area were significantly different. Development in the inland area was similar to other

Chinese cities, whereas the reclaimed area was relatively complex with two main

changing paths; and (3) Year 2008 was an important turning point in the perspective

of urbanization in the study area. Before 2008, urbanization was random and uncon-

trolled. After 2008, new governance on land appeared and changed the landscape

into a compact and uniform pattern. The proposed framework should reveal the

detailed trajectory of land reclamation in small areas and provide insights and tools

for better understanding the impact of human activities on the landscape pattern in

coastal regions under rapid urbanization.

K E YWORD S

land reclamation, long-time monitoring, Qiantang New District, remote sensing, urbanization

1 | INTRODUCTION

To deal with the scarcity of land in densely populated coastal areas,

sea reclamation (the process of creating new land from a coastal

waterbody) has become a popular way of accommodating rapid

urbanization around the world (Li et al., 2014). Many countries, includ-

ing Singapore (Glaser et al., 1991), Japan (Suzuki, 2003), Italy (Breber

et al., 2008), The Netherlands (Hoeksema, 2007), Saudi Arabia

(Niang, 2020), and China (Xu et al., 2019) alleviate the scarcity of land

by sea reclamation (Yi et al., 2018). Reclamation provides new space

for both urban expansion and industrial and agricultural development

in coastal areas (Li et al., 2020). Although reclaiming land is a global

phenomenon, China's high demands for land have resulted in land rec-

lamation for development at much larger scales compared to other

countries (Sengupta et al., 2019). In recent decades, all coastal prov-

inces and metropolises in China have experienced significant coastal

Received: 12 September 2020 Revised: 31 March 2021 Accepted: 4 May 2021

DOI: 10.1002/ldr.3980

Land Degrad Dev. 2021;32:3259–3271. wileyonlinelibrary.com/journal/ldr © 2021 John Wiley & Sons, Ltd. 3259

reclamation related to land scarcity caused by rapid economic growth

and urbanization (Tian et al., 2016).

Coastal reclamation is capable of creating abundant land to allevi-

ate the pressure from land shortages. However, it significantly

changes the land use/cover pattern with landscape fragmentation and

modification to vegetation distribution and type. Compared to climatic

factors, man-made reclamations have more direct and significant

effects on vegetation growth. Some land use changes, such as the

establishment of wetland parks, can promote local vegetation restora-

tion. However, other land use changes, such as urban expansion, may

exert pressure on the natural environment and contribute significantly

to vegetative degradation (Li, Lu, et al., 2018). The spatial landscape

pattern of land use/cover can reveal the urbanization processes and

shape the policies for social and economic development at regional

scales. Landscape pattern metrics are important indexes to measure

how ecological processes change in spatial and temporal scales. These

metrics have been widely used to describe the landscape structure

and characterize land use/cover and landscape pattern dynamics in

urban areas (Yi et al., 2018). Therefore, assessing the location, extent,

type, and landscape pattern of land use/cover changes is critical for a

better understanding of the dynamics and consequences of

reclamations.

Remote sensing-based change detection techniques, such as prin-

cipal component analysis, image differencing, and post-classification

comparison can be used to monitor land use/cover dynamics (Li, Lu,

et al., 2018). With a history of near 50 years, the LANDSAT series of

satellites provides the longest temporal record of space-born observa-

tions of the Earth's surface. Landsat data have demonstrated its capa-

bilities in mapping and monitoring land cover and land surface

biophysical and geophysical properties (Roy et al., 2014). Researchers

have used LANDSAT data to monitor long-term landscape change in

coastal zones by catching their dynamic status (Wu et al., 2018). Using

freely available LANDSAT imagery, Abdullah et al. (2019) elucidated

the spatiotemporal patterns of land use/land cover (LULC) across

coastal areas over 28 years (1990–2017) in the Heterogeneous

Coastal Region of Bangladesh. Hossain et al. (2019) assessed the envi-

ronmental impact of coastal reclamation activities based on LANDSAT

imagery (1994–2017) through mapping land cover distribution

changes in the Sungai Pulai estuary of Malaysia. Remote sensing tech-

niques have been deployed to monitor land reclamations and the sub-

sequent environmental impacts due to the rapid urbanization in

China's long coastlines in the past decades. For example, Wang

et al. (2014), Tian et al. (2016), Ren et al. (2019), and Li et al. (2020)

analyzed the changing spatial extent of coastal land reclamation at a

national scale. At regional scales, many researchers focused on the

reclamation of three major economic zones from north to south along

the coast, including Bohai Bay (Chen et al., 2019; Xu et al., 2019),

Yangtze River Delta (Li, Lu, et al., 2018; Li, Yin, et al., 2018), and Pearl

River Delta (Li & Damen, 2010). Many studies also focused on land

reclamation dynamics and its impacts on the ecological system at the

individual city scale, such as Yantai (Yu et al., 2013), Lianyungang

(Li et al., 2014), Shanghai (Wu et al., 2018), and Shenzhen

(Yi et al., 2018).

However, we have a limited understanding of the long-term

dynamics and key drivers of land use/cover change in the reclaimed

area. It is necessary to examine the detailed trajectory of land recla-

mation in small areas and compare the differences of landscape pat-

tern changes between reclamation and inland areas, to better

understand the impact of human activities on the landscape patterns

in coastal regions during rapid urbanization.

Taking Qiantang New District (QND) of Hangzhou, China as a case,

our study monitored and evaluated the detailed process of land reclama-

tion from 1973 to 2018 using time-series LANDSAT and SENTINEL-2A

images. The specific objectives of our study were to: (a) monitor spatial

and temporal changes in the landscape of QND from 1973 to 2018;

(b) evaluate the differences of landscape pattern changes between

reclaimed and inland areas, (c) map the land reclamation trajectory and

summarize the urban development mode in QND.

2 | MATERIALS AND METHODS

2.1 | Study area

Located in the North Pacific coastal area, Hangzhou Bay is a trumpet-

shaped tidal estuary playing a significant role in the sustainable devel-

opment of Hangzhou (Hu et al., 2020). The QND is an economically

important part of the Yangtze River Delta in Zhejiang Province, China.

QND extends from 30�120 to 30�240N, and from 120�230 to

120�440E, covering an area of approximately 524.6 km2 (Figure 1).

Traditionally, local economies mainly relied on agriculture and aqua-

culture, resulting in arable lands and aquaculture lands as the major

land uses. However, since the 1970s, due to the rapid industrializa-

tion, population growth, and urbanization, the demand for land has

increased, and land reclamation has gradually emerged. In 2020, the

QND is defined by the Hangzhou Government as the industrial

agglomeration area in the east of the Yangtze River. Therefore, con-

tinuous urban expansion with the extension of constructed land is

expected here in the future. According to the process of land reclama-

tion, the study area is divided into two parts: inland and reclaimed

area. The inner baseline was determined by the coastal line in 1973

(1973-coastline). The land areas in 1973 were defined as inland and

the waterbodies in that year were defined as reclaimed areas.

2.2 | Data collection and pre-processing

A total of 10 scenes from LANDSAT-1–4 MSS C1 Level-1,

LANDSAT-5 TM C1 Level-1, LANDSAT-8 OLI C1 Level-1, and

SENTINEL-2A MSI Level-1C products between 1973 and 2018 with

similar seasons (winter) were collected from the United States Geo-

logical Survey (USGS) Global Visualization Viewer (https://www.usgs.

gov/). The details of selected images are listed in Table 1. All images

were registered to UTM zone 50N, WGS84 using at least 35 manually

selected ground control points and nearest-neighbor interpolation.

The results of the geometric corrections had a relative root mean

3260 HU ET AL.

square error of less than 0.5 pixel. Then these images were clipped by

the administrative boundary to form the mapping unit. Image pre-

processing was performed using ENVI® 5.3 (Harris Corporation,

Melbourne, FL) (Hu et al., 2020).

Two different groups of reference data were collected to ensure

the accuracy assessment: the reference data for 2018 and the refer-

ence data for other historical years. (1) The reference data for 2018

were collected by field survey in November 2018. About 300 points

covered all land use/cover types were collected by a simple random

sampling method. For all the sampling points, the attributes were

examined by ground observations, and the location coordinates of a

plot centre were recorded using a portable GPS. (2) The reference

data for the historical years (1973, 1979, 1983, 1988, 1993, 1999,

2003, 2008, and 2013) were mainly acquired from the Zhejiang

Administration of Surveying Mapping and Geoinformation. The aerial

photographs (single-band image, but with a spatial resolution of 1 m)

from the 1970s were derived from the following website: https://

ditu.zjzwfw.gov.cn/resources-server/rescenter/index.html#/overview.

And the historical images were downloaded from Google Earth

(http://earth.google.com) (Hu et al., 2020). Similarly, about 500 points

per year covered all land use/cover types were collected by a simple

random sampling method. All the sampling points were carefully inter-

preted, some of the sample points which were difficult to determine

the attributes were deleted, and about 470 points were retained.

Through the above methods, the reference datasets for 1973–2018

were successfully established (Table 2).

2.3 | Overall workflow

The overall workflow consists of four parts: land use/cover classifica-

tion, landscape configuration analysis, land change detection, and

F IGURE 1 Location of Qiantang New District. The base image is a Landsat-8 OLI true colour composition image taken on November10, 2018 [Colour figure can be viewed at wileyonlinelibrary.com]

TABLE 1 List of remote sensing images in Qiantang New District between 1973 and 2018

Year Date Sensor Sun elevation (�) Cloud cover (%) Path/row Resolution (m)

1973 November 16, 1973 LANDSAT-1 MSS 35.34 0 128/39 60

1979 November 02, 1979 LANDSAT-3 MSS 37.61 6 127/39 60

1983 November 14, 1983 LANDSAT-4 MSS 35.73 10 119/39 60

1988 November 03, 1988 LANDSAT-5 TM 38.44 0 119/39 30

1993 November 01, 1993 LANDSAT-5 TM 38.05 0.01 119/39 30

1999 October 17, 1999 LANDSAT-5 TM 44.17 20.38 119/39 30

2003 October 28, 2003 -LANDSAT 5 TM 41.17 0 119/39 30

2008 February 28, 2008 LANDSAT-5 TM 43.17 0 119/39 30

2013 November 08, 2013 LANDSAT-8 OLI 40.14 0.27 119/39 30

2018 November 10, 2018 SENTINEL-2A 48.98 0.214 - 10

Note: The files were downloaded from the USGS (https://glovis.usgs.gov/)

HU ET AL. 3261

regional comparative analysis. First, time-series classification maps

from 1973 to 2018 were produced using support vector machine

(SVM) technology and post classification method. Next, six typical

landscape indices were calculated based on classified maps. Then, his-

torical dynamics of the different land use types were analyzed. Finally,

for regional comparative analysis, the landscape classification and con-

figuration results were divided into two parts according to the scope

of the inland and reclaimed areas.

2.4 | Land use/cover classification and accuracyassessment

The process of land use/cover classification includes four steps:

(1) land use/cover classification system determination, (2) image clas-

sification, (3) post-classification, and (4) accuracy assessment.

2.4.1 | Land use/cover classification systemdetermination

According to the classification systems of relevant researches, and

combining with the main feature of the study area, a two-level hierar-

chical classification system was determined (Shi et al., 2002). The defi-

nition of land use/cover categories is listed in Table 3.

2.4.2 | Image classification

Compared with the traditional classification methods, SVM technol-

ogy uses fewer training samples and can effectively improve the clas-

sification accuracy by integrating spectral and texture shape features

and converted spectral components. SVM is particularly suitable for

the classification of moderate resolution remote sensing images

(Khatami et al., 2016; Samadzadegan et al., 2012). In this study, SVM

was performed using ENVI® 5.3 by the Harris Corporation and was

used to classify land use/cover maps from time-series images of

1973, 1979, 1983, 1988, 1993, 1999, 2003, 2008, 2013, and 2018

respectively. The process of SVM includes three steps: (1) Training

samples selection. Previous research has generally found a positive

relationship between the number of training samples and the classifi-

cation accuracy for a wide range of classifiers. Some of the literature

suggests the use of a minimum of 10–30n samples per class for train-

ing, where nis the number of wavebands used (Pan et al., 2012).

Therefore, we use up to 30n training samples per class. Then training

samples were selected from each satellite image itself using the

Region of Interest (ROI) Tool in ENVI® 5.3. The components of

the training samples in our study are constructed land, arable land,

woodland, aquaculture land, and water. (2) Parameters setting. There

are different kernels through which the hyperplane boundary can be

defined. Here, a radial basis function kernel was used because the

LANDSAT and SENTINEL-2 bands are not linearly separable

(Abdi, 2020). All parameter settings used in the kernel function

(gamma parameter, penalty parameter, pyramid levels, and classifica-

tion probability threshold) are the default values provided by ENVI®

5.3. 3) SVM classifier running. Training samples, parameters, and satel-

lite images from the same year were used to run the SVM classifier.

After that, the classified maps of each year were obtained.

TABLE 2 The number of validation points for accuracy assessment between 1973 and 2018

Year Constructed land Arable land Waterbody Woodland Aquaculture land Qiantang River Total

1973 76 85 83 76 75 75 470

1979 85 72 86 70 79 81 473

1983 79 84 83 71 81 73 471

1988 83 77 84 73 73 83 473

1993 86 78 73 80 82 74 473

1999 77 84 83 73 87 68 472

2003 82 87 81 80 78 62 470

2008 84 79 83 86 79 59 470

2013 76 74 86 81 78 80 475

2018 62 53 52 45 56 46 314

TABLE 3 Land use/cover categories and their descriptions

Categorylevel I

Categorylevel II Description

Constructed

land

Constructed

land

Impervious surfaces including

buildings, roads, airports, parking

lots and sidewalks

Arable land Arable land Agricultural area, crop fields,

grassland and vegetable lands

Woodland Woodland Forest and shrubland with tree

canopy

Aquaculture

land

Aquaculture

land

Artificially excavated or naturally

formed pits and ponds

specifically for aquaculture

Water Waterbody River, lakes, irrigation canals and

ditches

Qiantang

River

Water surface of Qiantang River

3262 HU ET AL.

2.4.3 | Post classification

Based on the definition of land use/cover and the results of SVM clas-

sification, taking the coastline as the boundary, the water area was

manually divided into the waterbody and Qiantang River. A manual

visual interpretation based on reference data from Google Earth his-

torical images was used for post classification to improve the accuracy

of results (Liu et al., 2014).

2.4.4 | Accuracy assessment

A confusion matrix, which shows the relationship between the classi-

fied result and the ground truth data to obtain the percentage of cor-

rect classification of objects in different places, is one of the most

commonly used methods to calculate the classification accuracy

(Foody, 2002). Based on the classified results and validation data, four

indices were calculated: overall accuracy, kappa coefficient (kappa),

producer's accuracy, and user's accuracy (Mui et al., 2015). In this

study, the training data (from the images itself) and validation data

(from the reference datasets) were conducted independently.

2.5 | Landscape configuration analysis

At the level of features and landscape, there are more than

100 statistical methods of landscape structure assessment/

analysis (Cushman et al., 2008). Through the landscape metrics,

we can effectively and quantitatively analyze the composition

characteristics, spatial allocation, and dynamic changes of the

landscape. In this study, the analysis of landscape patterns

includes two levels: the spatial feature of landscape types (class-

level) and the characteristics of overall landscape diversity (land-

scape-level). The following six indices were selected to reveal

the characteristics of the landscape in size, shape, and complex-

ity (Table 4): patch density (PD), largest patch index (LPI), land-

scape shape index (LSI), Shannon's diversity index (SHDI),

contagion index (CONTAG), and aggregation index (AI) (Chen &

Yu, 2017; Cushman et al., 2008).

2.6 | Land change detection

The post classification change detection methods perform single-

phase classifications on all the data in the geographical range, com-

pare and analyze the differences between each phase, and obtain

the final time-series classification results (Yan et al., 2019). The

process of change detection algorithm includes two steps. (1) Post-

classification resampling. Because the spatial resolution of the clas-

sification results varies in different years, it is necessary to

resample the classification results. Since the spatial resolution of

most of the classification results is 30 m, we resampled the classifi-

cation results of 1973, 1979, 1983, and 2018 to 30 m. The

resampling operation was conducted using ENVI® 5.3 and the

TABLE 4 Landscape pattern index and its ecological significance

Metrics Formula Description

PD PD¼ NA

The ratio of the number of patches to the area, the number of patches

per km2

LPI LPI¼maxaA The proportion of the largest patch in a patch type to the whole

landscape area

LSI LSI¼ 0:25EffiffiffiA

p The complexity of a patch shape is measured by calculating the

deviation between a patch shape and a circle or square of the same

area

SHDISHDI¼�Pm

i¼1PilnPið Þ SHDI equals minus the sum, across all patch types, of the proportional

abundance of each patch type multiplied by that proportion

CONTAG

CONTAG¼ 1þ

Pmi¼1

Pmk¼1

Pið Þ gikPmk¼1

gik

0B@

1CA

264

375� ln Pið Þ gikPm

k¼1

gik

0B@

1CA

264

375

2ln mð Þ

8>>>>>>><>>>>>>>:

9>>>>>>>=>>>>>>>;

100ð Þ

Describe the degree of aggregation or extension trend of different

patch types in the landscape. A small value indicates that there are

many small patches in the landscape, and when it tends to 100, it

indicates that there are dominant patch types with high connectivity

in the landscape

AIAI¼ gii

max! gii

� �100ð Þ At landscape level, the index is computed simply as an area-weighted

mean class aggregation index, where each class is weighted by its

proportional area in the landscape. The index is scaled to account for

the maximum possible number of like adjacencies given any

landscape composition

Note: N is the total number of patches in the landscape; a is the patch area (m2); A is the total patch area (m2); E is the total length of all patch boundaries in

the landscape (m); Pi is the percentage of the total landscape area in i-type patches. gik is the number of i-type patches and k-type patches adjacent to each

other; m is the total number of patches in the landscape. gii is the number of like adjacencies (joins) between pixels of patch type (class) i based on the

single-count method; max-gii is the maximum number of like adjacencies (joins) between pixels of patch type (class) i based on the single-count method; Piis the proportion of landscape comprised of patch type (class) i

Abbreviations: AI, aggregation index; CONTAG, contagion index; LPI, largest patch index; LSI, landscape shape index; PD, patch density; SHDI, Shannon's

diversity index

HU ET AL. 3263

nearest neighbor algorithm was selected. (2) Change detection. A

multi-date post-classification comparison change detection algo-

rithm was used to determine the land use/cover changes in differ-

ent periodic intervals, with a pixel-based comparison of the

corresponding classes from two adjacent classified images, to iden-

tify and quantify areas of changes (Biro et al., 2011; Yuan

et al., 2005). Following the post-classification resampling results

from the individual years, a multi-date post-classification compari-

son change detection algorithm was used to determine the changes

in land use/cover for 10 intervals: 1973–1979, 1979–1983, 1983–

1988, 1988–1993, 1993–1999, 1999–2003, 2003–2008, 2008–

2013, and 2013–2018. Change detection maps were derived from

each of the 10 six-class maps.

3 | RESULTS

3.1 | Spatial–temporal changes in the landscape ofQND from 1973 to 2018

The time-series landscape maps from 1973 to 2018 obtained with the

SVM and modified post classification are shown in Figure 2. To vali-

date these maps, classification results were compared with the refer-

ence datasets. The overall accuracies were calculated as 92.13% for

1973, 82.03% for 1979, 84.93% for 1983, 83.93% for 1988, 90.27%

for 1993, 88.77% for 1999, 90.21% for 2003, 93.62% for 2008,

92.42% for 2013, and 92.04% for 2018. The kappa statistics were cal-

culated as 0.91 for 1973, 0.78 for 1979, 0.82 for 1983, 0.81 for 1988,

F IGURE 2 Time-series landscape maps of Qiantang New District from 1973 to 2018 [Colour figure can be viewed at wileyonlinelibrary.com]

3264 HU ET AL.

0.88 for 1993, 0.87 for 1999, 0.88 for 2003, 0.92 for 2008, 0.91 for

2013, and 0.90 for 2018. The quantitative accuracy assessment

results are shown in Table 5. This suggests that the maps showed an

acceptable range of agreement with the reference data used for the

accuracy assessment (Tana et al., 2013). It shows a statistically signifi-

cant shrinking trend for the Qiantang River (212.21 to 80.99 km2),

and conversely, a significant increase occurred in constructed land

areas (10.05 to 120.89 km2) from 1973 to 2018. The change in arable

land was relatively complex, the area increased continuously from

1973 and reached its maximum value (336.61 km2) in 1988, then

began to decrease and gradually changed to 200.96 km2 in 2018.

3.2 | Landscape configuration results

In class-level metrics, PD, LPI, and LSI, were calculated for each

land use/cover type. The changes of three main types (constructed

land, arable land, and aquaculture land) in QND are shown in

Figure 3. As can be seen from the figure, the changes of PD and LSI

of the three mainland use types were similar, increasing at first and

then decreasing. The LPI of arable land did not change much from

1973 to 1999, but then began to decline rapidly. The LPI of con-

structed land continuously increased, especially from 1999 to

2018. The LPI of aquaculture land increased from 1999 to 2008,

but began to decline gradually after 2008.

In landscape-level metrics, CONTAG, SHDI, and AI were calcu-

lated for the whole study area (Table 6). In QND, from 1973 to 2008,

the CONTAG and AI decreased gradually, from 70.54 to 47.40 and

from 97.42 to 89.37, respectively. After 2008, the CONTAG and AI

began to increase. The SHDI showed an increasing trend during the

period from 1973 to 2008, and remained stable after 2008.

3.3 | Historical change detection results from 1973to 2018

Figure 4 shows the spatial pattern of land use/cover changes using

post-classification change detection method (1973–1979, 1979–

1983, 1983–1988, 1988–1993, 1993–1999, 1999–2003, 2003–

2008, 2008–2013, and 2013–2018), including gains and losses for

TABLE 5 Summary of classification accuracies (%) from 1973 to 2018

Class Accuracies 1973 1979 1983 1988 1993 1999 2003 2008 2013 2018

Constructed land User 90.24 83.95 78.31 91.14 96.47 95.95 96.15 100 92.59 100

Producer 97.37 80 82.28 86.75 95.35 92.21 91.46 97.62 98.68 96.77

Arable land User 90.7 58.59 63.64 58.14 79.79 80.65 81 93.75 84.52 85

Producer 91.76 80.56 83.33 97.4 96.15 89.29 93.1 94.94 95.95 96.23

Waterbody User 92.96 76.83 100 87.72 93.1 81.71 82.5 83.72 93.33 84

Producer 79.52 73.26 78.31 59.52 73.97 80.72 81.48 86.75 81.4 80.77

Woodland User 93.33 96.67 89.04 100 98.7 91.89 96.2 95.45 96.25 95.35

Producer 92.11 82.86 91.55 95.89 95 93.15 95 97.67 95.06 91.11

Aquaculture land User 86.42 89.55 92.54 88.68 77.65 86.42 90.14 90.67 88 89.09

Producer 93.33 75.95 76.54 64.38 80.49 80.46 82.05 86.08 84.62 87.5

Qiantang River User 100 96.43 100 97.65 100 100 100 100 100 100

Producer 100 100 100 100 100 100 100 100 100 100

Overall accuracy 92.13 82.03 84.93 83.93 90.27 88.77 90.21 93.62 92.42 92.04

kappa statistic 0.91 0.78 0.82 0.81 0.88 0.87 0.88 0.92 0.91 0.9

F IGURE 3 Landscape indices changes of different land use/cover types in class-level metrics [Colour figure can be viewed atwileyonlinelibrary.com]

HU ET AL. 3265

the four mainland use/cover types (constructed land, arable land,

aquaculture land, and Qiantang River). Qiantang River and arable land

decreased most, constructed land increased most, and aquaculture

land increased, and decreased in different periods. Specifically, the

Qiantang River waterbody decreased year by year along the coastline.

Most of the reclaimed areas were used as aquaculture lands. In addi-

tion, the constructed land gradually increased along the with villages

and roads, occupying a large area of arable land in inland area and

aquaculture land in reclaimed area.

4 | DISCUSSION

4.1 | Main changes of landscape in QND and itsdiving factors

The results showed that the whole area experienced rapid urbaniza-

tion during the last 45 years. The process of urban expansion in this

area varies in different periods. Before the release of the 'Reform and

Open-up' policy (before 1978), limited urbanization phenomenon

could be detected. In contrast, in the first 5 years of the 'Reform and

Open-up' policy, the urban area almost tripled in the study area. How-

ever, after the first 5 years, the speed of urbanization decelerated

from 1983 to 1993. This deceleration was mainly due to the policy to

restrict the urban expansion in big cities in the 1980s

(Wu et al., 2014). Meanwhile, the population in Hangzhou increased

from 5.33 million to 5.87 million, with an annual increasing rate of

0.97%, far behind the paces of growth from 1978 to 1982 and from

1994 to 2018, when growth rates were 1.10% and 1.12%, respec-

tively (Hangzhou Municipal Bureau of Statistics, 2021). Later, with the

release of the national strategy of 'opening Shanghai's Pudong new

area to the international investment' in 1990 (Wu & Zhang, 2012) and

the new instruction from the Government to 'accelerate the reform

process' in 1992 (Wang, 2020), the urbanization of QND accelerated.

The results revealed a continuous urban expansion from 1993 to

2008, and the speed reached the peak at the beginning of the 21st

TABLE 6 Landscape indices changes of whole study area in landscape-level metrics

Metrics 1973 1979 1983 1988 1993 1999 2003 2008 2013 2018

CONTAG 70.54 69.60 65.78 67.55 64.62 56.88 51.97 47.40 48.31 50.43

SHDI 0.94 0.97 1.05 0.96 1.03 1.24 1.37 1.46 1.47 1.44

AI 97.42 97.28 95.85 95.22 94.08 92.41 91.29 89.37 90.59 91.96

Abbreviations: AI, aggregation index; CONTAG, contagion index; SHDI, Shannon's diversity index

F IGURE 4 The spatial patterns of land use/cover changes in the study area, including losses and gains for specific types: (a) Qiantang Riverloss; (b) constructed land gain; (c) arable land gain and loss; (d) aquaculture land gain and loss [Colour figure can be viewed atwileyonlinelibrary.com]

3266 HU ET AL.

century. However, the increase of constructed land was 54.46 km2

while the decrease of arable land was 111.91 km2 in this period. Com-

pared to other studies in Hangzhou, which reported that the loss area

of arable land stood for over 90% of the gain in area of constructed

land (Hou et al., 2019), the arable land in QND was less affected by

the urbanization compared to other districts in Hangzhou. Observing

the whole landscape change process, we recognized that the newly

constructed land mostly occupied the arable land, while the arable

land also expanded to the reclaimed area which resulted in the supple-

ment of the total area. This partly results from the Land Management

Law, which requires the local government to ensure that if cultivated

land is converted into constructed land, the same quantity of culti-

vated land needs to be developed (Meng et al., 2017). The loss of ara-

ble land also contributed to the increase of aquaculture land,

especially in the period of 1990–2000, which was consistent with a

common phenomenon observed in the whole coastal areas of China

(Ren et al., 2019). In addition, Qiantang River played an important role

in balancing the land resources during the urbanization since it pro-

vided the area for agriculture and fishery to reduce the pressure of

farmland loss.

Landscape configuration analysis provided another perspective

in understanding the development in QND. The general indices

(CONTAG, AI, and SHDI) indicated an important temporal turning

point: the year of 2008. Before this point, the landscape here con-

tinuously scattered, which implied less general control by the local

government in the development. On contrast, after 2008, all the

indices showed a clustering trend. Normally, the clustering trend

indicated an integrated development mode supported by the local

government. It is noticeable that in 2009, the municipal govern-

ment made a strategic plan to speed-up the development of

regional integration in the study area (http://qt.hangzhou.gov.cn).

This new strategy has resulted in the changes of general landscape

configuration. The indices (PD, LPI, and LSI) for individual land

cover categories provides detailed insights into the regional devel-

opment from the perspective of landscape configuration. Similar to

the general indices, these individual indices changed trajectories

around 2008, especially for arable land and aquaculture land. The

patches of these two categories were split and scattered during

1973 to 2008, while after 2008, the patches started to form a uni-

form pattern. These evidences proved that in around 2008, the

QND changed its developing pattern from a spontaneous mode to

an artificial controlling mode. Compared to a case-study done in

the Jiangsu coastal zone which detected a strongest human distur-

bance on the landscape during 1995–2002, the booming period in

QND started later (Zhou et al., 2018). However, the two studies

both emphasized the importance of regional developing policies on

regulating landscape configurations. In addition, the patches of

constructed land were reorganized and uniformed in advance. As

early as 1999, both the PD and LSI of constructed land turned from

an increasing trend to a decreasing trend. In this context, during

the urban development, constructed land was first reorganized.

Later, the other landscape resources were intended to be

reorganized in response to the change of constructed land.

4.2 | The difference between inland andreclaimed areas

In order to gain insights into the developing patterns in different

areas, the study area was divided into inland and reclaimed areas

based on the land use/cover map in 1973. According to the analysis

of the changes in landscape composition and configuration, there are

significant differences between the two areas (Figure 5). Generally,

regarding to landscape composition, inland areas showed an urbaniza-

tion pattern that constructed lands expanded at the cost of arable

lands. Similar patterns were detected in several Chinese cities in the

F IGURE 5 Land use/cover area change in inland area and reclaimed area [Colour figure can be viewed at wileyonlinelibrary.com]

HU ET AL. 3267

previous studies (Haas & Ban, 2014; Hou et al., 2019; Liu et al., 2005;

Zhang et al., 2016). In contrast, the dynamics of landscape composi-

tion were much more complicated in the reclaimed area. Although

coastal reclamation is generally considered as a way to meet the

demand of space for living and development in coastal areas (Wang

et al., 2014), previous study has demonstrated its complexity from the

evidence that land for urban use is sufficient in some coastal regions

as there is much unused land (Meng et al., 2017). In this study, we

showed that the reclaimed areas were not directly occupied by con-

structed land in QND. Reclamation from the Qiantang River mainly

provided the developing space for other landscapes rather than urban

lands, while the main 'consumer' changed during the whole time

period.

The whole process could be separated into three periods:

(1) before 1988, (2) 1988 to 2008, and (3) after 2008. In the first time

period, arable land occupied most of the developing space. In this

period, the 'Reform and Open-up' policy just started, and the eco-

nomic benefits in urban development were not prominent. In this con-

text, local residents preferred to develop arable lands to satisfy the

growing demands on food. In the second period, local residents

started to use reclaimed areas for aquaculture ponds. Compared to

traditional agriculture, reclamation for aquaculture has several advan-

tages, including low cost, simple process, and high economic benefit

(Meng et al., 2017). Similar patterns were common in other coastal

provinces of China, especially from 1990 to 2000 (Ren et al., 2019). In

the third period, the continuous loss of Qiantang River stopped and

the landscape became stable. These evidences proved the develop-

ment of QND became mature, and the local government made sys-

tematic urban planning after 2008.

In the perspective of landscape configuration, the differences

between inland area and reclaimed area were not obvious compared

to the results of landscape composition (Figure 6). The constructed

land and the arable land almost followed the same steps with two

important turning point: 1999 for constructed land and 2008 for ara-

ble land, respectively. The only noticeable difference appeared in the

category of aquaculture land (Figure 3). Compared to the other two

main categories, aquaculture land experienced two turning points dur-

ing the development. The first turning point appeared in 1993 and

continued to 1999, indicating that the expansion of aquaculture land

during 1993–1999 was not random, and the government has

promoted this process. The second turning point appeared in 2008, in

consistency with the developing pattern of arable land, supported the

conclusion that the local government has made a comprehensive

developing plan for the landscape resources here.

4.3 | Urban development mode in QND

In this study, we created land use/cover maps at 10 time points to fol-

low the land changing path for 45 years. The results showed that the

inland area and reclaimed area have followed two distinct paths. In the

case of the inland area, constructed land occupied arable land while

arable land occupied aquaculture land to make up for the loss of agri-

cultural productions due to urban expansion [Figure 7(a)]. This phenom-

enon is common in the process of urbanization in China, because

historical towns and cities are usually built on arable lands and urban

expansion tends to occur in proximity to existing development

(Su et al., 2014). However, the reclaimed area experienced a unique

changing path. Before 2008, the landscape here was relatively instable

with arable land and aquaculture land alternately dominating the dis-

trict. Generally, we summarized two main paths: (1) Qiantang River !arable land! aquaculture land! arable land [Figure 7(b)]; (2) Qiantang

River ! aquaculture land ! arable land [Figure 7(c)]. In this time

period, the constructed land has not expanded into the reclaimed area

because the pace of urbanization was relatively slower and there was

enough room in the inland area to accommodate the less rapid urban

expansion. In addition, the arable land and aquaculture land exchanged

frequently to seize a balance for the best economic outcome. When

aquaculture land dominated the area, the demand for arable land and

agriculture products naturally increased, causing local residents to

reclaim arable land, and vice versa. This phenomenon also proved that

the regional development was mostly decided by the local residents

other than the Government in this time period. After 2008, the regional

development stepped into a mature stage. Reclamations of land from

the Qiantang River slowed down and the conversions between aqua-

culture land and arable land became less haphazard. In addition, with

the accumulation of time, the reclamation lands started to show its

potential for urban use. Scattered but uniformed constructed land

started to appear in this area, indicating a good momentum of urbaniza-

tion under matured urban planning.

F IGURE 6 Landscape indices change in inland area and reclaimed area [Colour figure can be viewed at wileyonlinelibrary.com]

3268 HU ET AL.

5 | CONCLUSIONS

Under the pressure of rapid urbanization, land reclamation has

become an attractive option to solve the problem of land shortage

in coastal areas. In this study, we examined the spatial and temporal

dynamics of landscape in QND, and compared the differences in

developing patterns between the inland and reclamation areas. The

QND has experienced rapid urbanization since 1979, with an

annual increase of 2.82 km2 in constructed land until 2018. The

changing pattern mainly behaves as constructed land occupies

arable land while arable land decelerates its rate of reduction from

land reclamation from the Qiantang River. The comparison

between inland and reclaimed areas shows totally different devel-

oping modes. The inland area in QND developed as a typical Chi-

nese city that constructed lands expanded rapidly at the cost of

arable land. On-the-other- hand, considering the geological condi-

tions, reclamation lands mostly changed to arable lands or aquacul-

ture lands in the beginning, and the urbanization here mainly

started from 2008 with the termination of reclamation from the

Qiantang River. Our study reveals an important turning point in the

perspective of urbanization in QND: the year 2008. Before 2008,

the urbanization in this district was random and uncontrolled. After

2008, the governance on land obviously appeared and changed the

landscape into a matured and uniformed pattern. These observa-

tions proved that the QND is developing in a proper way and the

reclaimed area would start to face the urbanization pressure in the

future.

In this study, we utilized the LANDSAT time-series data from

1973 to 2018 to comprehensively monitor the landscape dynamics in

QND. Considering the size of the study area, higher resolution images

are preferred in the future to better extract the land information. In

addition, because of the data availability, we extracted the reclama-

tion land based on the land condition in 1973. Some lands reclaimed

before 1973 might be neglected. Finally, considering the observations

in this study, the urban development has just started in this area. Con-

tinuous monitoring is needed in the future to capture the general

developing pattern of reclamation lands to support the urban planning

of coastal areas.

ACKNOWLEDGMENTS

The authors are very grateful for funding provided by the Zhejiang

Provincial Natural Science Foundation of China (Grant Nos.

F IGURE 7 Landscape indices change in inland area and reclaimed area [Colour figure can be viewed at wileyonlinelibrary.com]

HU ET AL. 3269

LQ20D010008; LY19D010004; LGF18D010005) and the Science

and Technology Program of Hangzhou (Grant No. 20191203B19).

CONFLICT OF INTEREST

The co-authors of the above-named article declare that they have

read the article and approved it for submission to Land Degradation

and Development. They also declare no conflict of interests.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the

corresponding author upon reasonable request.

ORCID

Tangao Hu https://orcid.org/0000-0003-4448-2614

Jinjin Fan https://orcid.org/0000-0001-9684-9495

Hao Hou https://orcid.org/0000-0003-2444-133X

Yao Li https://orcid.org/0000-0002-5406-4494

Yue Li https://orcid.org/0000-0001-8163-6227

Kangning Huang https://orcid.org/0000-0001-6877-9442

REFERENCES

Abdi, A. M. (2020). Land cover and land use classification performance of

machine learning algorithms in a boreal landscape using Sentinel-2

data. GIScience & Remote Sensing, 57, 1–20. https://doi.org/10.1080/15481603.2019.1650447

Abdullah, A. Y. M., Masrur, A., Adnan, M. S. G., Baky, M. A., Hassan, Q. K.,

& Dewan, A. (2019). Spatio-temporal patterns of land use/land cover

change in the heterogeneous coastal region of Bangladesh between

1990 and 2017. Remote Sensing, 11, 790. http://dx.doi.org/10.3390/

rs11070790.

Biro, K., Pradhan, B., Buchroithner, M., & Makeschin, F. (2011). Land use/-

land cover change analysis and its impact on soil properties in the

northern part of Gadarif region, Sudan. Land Degradation & Develop-

ment, 24, 90–102. https://doi.org/10.1002/ldr.1116Breber, P., Povilanskas, R., & Armaitiene, A. (2008). Recent evolution of

fishery and land reclamation in Curonian and Lesina Lagoons.

Hydrobiologia, 611, 105–114. https://doi.org/10.1007/s10750-008-

9453-6

Chen, Y., Dong, J., Xiao, X., Ma, Z., Tan, K., Melville, D., Li, B., Lu, H.,

Liu, J., & Liu, F. (2019). Effects of reclamation and natural changes on

coastal wetlands bordering China's Yellow Sea from 1984 to 2015.

Land Degradation & Development, 30, 1533–1544. https://doi.org/10.1002/ldr.3322

Chen, Y., & Yu, S. (2017). Impacts of urban landscape patterns on urban

thermal variations in Guangzhou, China. International Journal of Applied

Earth Observation and Geoinformation, 54, 65–71. https://doi.org/10.1016/j.jag.2016.09.007

Cushman, S. A., McGariyal, K., & Neel, M. C. (2008). Parsimony in land-

scape metrics: Strength, universality, and consistency. Ecological Indi-

cators, 8, 691–703. https://doi.org/10.1016/j.ecolind.2007.12.002Foody, G.M. (2002). Status of land cover classification accuracy assess-

ment. Remote Sensing of Environment, 80, 185–201. https://doi.org/10.1016/S0034-4257(01)00295-4

Glaser, R., Haberzettl, P., & Walsh, R. P. D. (1991). Land reclamation in Sin-

gapore, Hong Kong and Macau. GeoJournal, 24(4), 365–373. http://dx.doi.org/10.1007/bf00578258.

Haas, J., & Ban, Y. (2014). Urban growth and environmental impacts in

Jing-Jin-Ji, the Yangtze, River Delta and the Pearl River Delta. Interna-

tional Journal of Applied Earth Observation and Geoinformation, 30, 42–55. https://doi.org/10.1016/j.jag.2013.12.012

Hangzhou Municipal Bureau of Statistics. (2021). Hangzhou statistical year-

book 2020. http://tjj.hangzhou.gov.cn/art/2020/10/29/art_1229

453592_3819709.html (accessed 26 February 2021).

Hoeksema, R. J. (2007). Three stages in the history of land reclamation in

The Netherlands. Irrigation and Drainage, 56, S113–S126. https://doi.org/10.1002/ird.340

Hossain, M. S., Hashim, M., Bujang, J. S., Zakaria, M. H., & Muslim, A. M.

(2019). Assessment of the impact of coastal reclamation activities on

seagrass meadows in Sungai Pulai estuary, Malaysia, using Landsat

data (1994-2017). International Journal of Remote Sensing, 40, 3571–3605. https://doi.org/10.1080/01431161.2018.1547931

Hou, H., Wang, R., & Murayama, Y. (2019). Scenario-based modelling for

urban sustainability focusing on changes in cropland under rapid

urbanization: A case study of Hangzhou from 1990 to 2035. Science of

the Total Environment, 661, 422–431. https://doi.org/10.1016/j.

scitotenv.2019.01.208

Hu, T., Liu, J., Zheng, G., Zhang, D., & Huang, K. (2020). Evaluation of his-

torical and future wetland degradation using remote sensing imagery

and land use modelling. Land Degradation & Development, 31, 65–80.https://doi.org/10.1002/ldr.3429

Khatami, R., Mountrakis, G., & Stehman, S. V. (2016). A meta-analysis of

remote sensing research on supervised pixel-based land-cover image

classification processes: General guidelines for practitioners and future

research. Remote Sensing of Environment, 177, 89–100. https://doi.org/10.1016/j.rse.2016.02.028

Li, D., Lu, D., Wu, M., Shao, X., & Wei, J. (2018). Examining land cover and

greenness dynamics in Hangzhou Bay in 1985-2016 using Landsat

time-series data. Remote Sensing, 10, 32. https://doi.org/10.3390/

rs10010032

Li, F., Ding, D., Chen, Z., Chen, H., Shen, T., Wu, Q., & Zhang, C. (2020).

Change of sea reclamation and the sea-use management policy system

in China. Marine Policy, 115, 103861. https://doi.org/10.1016/j.

marpol.2020.103861

Li, M., Yin, P., Duan, X., Liu, J., Qiu, J., & Phung Van, P. (2018). Land-use

changes and human driving in and near the Yangtze River Delta from

1995-2015. Journal of Ocean University of China, 17, 1361–1368.https://doi.org/10.1007/s11802-018-3735-z

Li, X. J., & Damen, M. C. J. (2010). Coastline change detection with satellite

remote sensing for environmental management of the Pearl River

Estuary, China. Journal of Marine Systems, 82, S54–S61. https://doi.org/10.1016/j.jmarsys.2010.02.005

Li, Y., Shi, Y., Zhu, X., Cao, H., & Yu, T. (2014). Coastal wetland loss and

environmental change due to rapid urban expansion in Lianyungang,

Jiangsu, China. Regional Environmental Change, 14, 1175–1188.https://doi.org/10.1007/s10113-013-0552-1

Liu, G., Zhang, L., Zhang, Q., Musyimi, Z., & Jiang, Q. (2014). Spatio-

temporal dynamics of wetland landscape patterns based on remote

sensing in Yellow River Delta, China. Wetlands, 34, 787–801. https://doi.org/10.1007/s13157-014-0542-1

Liu, J., Liu, M., Tian, H., Zhuang, D., Zhang, Z., Zhang, W., Tang, X., & Deng, X.

(2005). Spatial and temporal patterns of China's cropland during 1990–2000: An analysis based on Landsat TM data. Remote Sensing of Environ-

ment, 98, 442–456. https://doi.org/10.1016/j.rse.2005.08.012Meng, W., Hu, B., He, M., Liu, B., Mo, X., Li, H., Wang, Z., & Zhang, Y.

(2017). Temporal-spatial variations and driving factors analysis of

coastal reclamation in China. Estuarine Coastal and Shelf Science, 191,

39–49. https://doi.org/10.1016/j.ecss.2017.04.008Mui, A., He, Y. H., & Weng, Q. H. (2015). An object-based approach to delin-

eate wetlands across landscapes of varied disturbance with high spatial

resolution satellite imagery. ISPRS Journal of Photogrammetry and Remote

Sensing, 109, 30–46. https://doi.org/10.1016/j.isprsjprs.2015.08.005Niang, A. J. (2020). Monitoring long-term shoreline changes along Yanbu,

Kingdom of Saudi Arabia using remote sensing and GIS techniques.

Journal of Taibah University for Science, 14, 762–776. https://doi.org/10.1080/16583655.2020.1773623

3270 HU ET AL.

Pan, Y., Hu, T., Zhu, X., Zhang, J., & Wang, X. (2012). Mapping cropland

distributions using a hard and soft classification model. IEEE Transac-

tions on Geoscience and Remote Sensing, 50, 4301–4312. https://doi.org/10.1109/TGRS.2012.2193403

Ren, C., Wang, Z., Zhang, Y., Zhang, B., Chen, L., Xi, Y., Xiao, X.,

Doughty, R. B., Liu, M., Jia, M., Mao, D., & Song, K. (2019). Rapid

expansion of coastal aquaculture ponds in China from Landsat obser-

vations during 1984–2016. International Journal of Applied Earth

Observation and Geoinformation, 82, 101902. https://doi.org/10.1016/

j.jag.2019.101902

Roy, D. P., Wulder, M. A., Loveland, T. R., Woodcock, C. E., Allen, R. G.,

Anderson, M. C., Helder, D., Irons, J. R., Johnson, D. M., Kennedy, R.,

Scambos, T., Schaaf, C. B., Schott, J. R., Sheng, Y., Vermote, E. F.,

Belward, A. S., Bindschadler, R., Cohen, W. B., Gao, F., … Zhu, Z.

(2014). Landsat-8: Science and product vision for terrestrial global

change research. Remote Sensing of Environment, 145, 154–172.https://doi.org/10.1016/j.rse.2014.02.001

Samadzadegan, F., Hasani, H., & Schenk, T. (2012). Simultaneous feature

selection and SVM parameter determination in classification of hyper-

spectral imagery using ant colony optimization. Canadian Journal of

Remote Sensing, 38, 139–156. https://doi.org/10.5589/m12-022

Sengupta, D., Chen, R., Meadows, M. E., Choi, Y. R., Banerjee, A., &

Zilong, X. (2019). Mapping trajectories of coastal land reclamation in

nine deltaic megacities using Google Earth engine. Remote Sensing, 11,

2621. https://doi.org/10.3390/rs11222621

Shi, Z., Wang, R., Huang, M. X., & Landgraf, D. (2002). Detection of coastal

saline land uses with multi-temporal landsat images in Shangyu City,

China. Environmental Management, 30, 142–150. https://doi.org/10.1007/s00267-001-2645-8

Su, S., Ma, X., & Xiao, R. (2014). Agricultural landscape pattern changes in

response to urbanization at ecoregional scale. Ecological Indicators, 40,

10–18. https://doi.org/10.1016/j.ecolind.2013.12.013Suzuki, T. (2003). Economic and geographic backgrounds of land reclama-

tion in Japanese ports. Marine Pollution Bulletin, 47, 226–229. https://doi.org/10.1016/s0025-326x(02)00405-8

Tana, G., Letu, H., Cheng, Z. K., & Tateishi, R. (2013). Wetlands mapping in

North America by decision rule classification using MODIS and ancil-

lary data. IEEE Journal of Selected Topics in Applied Earth Observations

and Remote Sensing, 6, 2391–2401. https://doi.org/10.1109/JSTARS.2013.2249499

Tian, B., Wu, W., Yang, Z., & Zhou, Y. (2016). Drivers, trends, and potential

impacts of long-term coastal reclamation in China from 1985 to 2010.

Estuarine Coastal and Shelf Science, 170, 83–90. https://doi.org/10.1016/j.ecss.2016.01.006

Wang, W., Liu, H. I., Li, Y. Q., & Su, J. L. (2014). Development and manage-

ment of land reclamation in China. Ocean & Coastal Management, 102,

415–425. https://doi.org/10.1016/j.ocecoaman.2014.03.009

Wang, Y. (2020). Urban land and sustainable resource use: Unpacking the

countervailing effects of urbanization on water use in China,

1990-2014. Land Use Policy, 90, 104307. https://doi.org/10.1016/j.

landusepol.2019.104307

Wu, J., Xiang, W.-N., & Zhao, J. (2014). Urban ecology in China: Historical

developments and future directions. Landscape and Urban Planning,

125, 222–233. https://doi.org/10.1016/j.landurbplan.2014.02.010Wu, K. Y., & Zhang, H. (2012). Land use dynamics, built-up land expansion

patterns, and driving forces analysis of the fast-growing Hangzhou

metropolitan area, eastern China (1978-2008). Applied Geography, 34,

137–145. https://doi.org/10.1016/j.apgeog.2011.11.006Wu, W., Yang, Z., Tian, B., Huang, Y., Zhou, Y., & Zhang, T. (2018). Impacts

of coastal reclamation on wetlands: Loss, resilience, and sustainable

management. Estuarine Coastal and Shelf Science, 210, 153–161.https://doi.org/10.1016/j.ecss.2018.06.013

Xu, J., Li, F., Suo, A., Zhao, J., & Su, X. (2019). Spatio-temporal change and

carrying capacity evaluation of human coastal utilization in Liaodong

Bay, China from 1993 to 2015. Chinese Geographical Science, 29, 463–473. https://doi.org/10.1007/s11769-019-1044-0

Yan, J., Wang, L., Song, W., Chen, Y., Chen, X., & Deng, Z. (2019). A time-

series classification approach based on change detection for rapid land

cover mapping. ISPRS Journal of Photogrammetry and Remote Sensing,

158, 249–262. https://doi.org/10.1016/j.isprsjprs.2019.10.003Yi, L., Chen, J., Jin, Z., Quan, Y., Han, P., Guan, S., & Jiang, X. (2018). Impacts

of human activities on coastal ecological environment during the rapid

urbanization process in Shenzhen, China. Ocean & Coastal Management,

154, 121–132. https://doi.org/10.1016/j.ocecoaman.2018.01.005

Yu, X., Zhang, A., Hou, X., Li, M., & Xia, Y. (2013). Multi-temporal remote

sensing of land cover change and urban sprawl in the coastal city of

Yantai, China. International Journal of Digital Earth, 6, 137–154.https://doi.org/10.1080/17538947.2011.653995

Yuan, F., Sawaya, K. E., Loeffelholz, B. C., & Bauer, M. E. (2005). Land

cover classification and change analysis of the twin cities (Minnesota)

metropolitan area by multitemporal Landsat remote sensing. Remote

Sensing of Environment, 98, 317–328. https://doi.org/10.1016/j.rse.

2005.08.006

Zhang, Z., Li, N., Wang, X., Liu, F., & Yang, L. (2016). A comparative study

of urban expansion in Beijing, Tianjin and Tangshan from the 1970s to

2013. Remote Sensing, 8, 496. https://doi.org/10.3390/rs8060496

Zhou, Y., Ning, L., & Bai, X. (2018). Spatial and temporal changes of human

disturbances and their effects on landscape patterns in the Jiangsu

coastal zone, China. Ecological Indicators, 93, 111–122. https://doi.org/10.1016/j.ecolind.2018.04.076

How to cite this article: Hu, T., Fan, J., Hou, H., Li, Y., Li, Y., &

Huang, K. (2021). Long-term monitoring and evaluation of

land development in a reclamation area under rapid

urbanization: A case-study in Qiantang New District, China.

Land Degradation & Development, 32(11), 3259–3271. https://

doi.org/10.1002/ldr.3980

HU ET AL. 3271