Loss of green landscapes due to urban expansion in China

Full text

Resources, Conservation & Recycling 199 (2023) 107228
Available online 27 September 2023
0921-3449/© 2023 Elsevier B.V. All rights reserved.
Loss of green landscapes due to urban expansion in China
Yating He
a
, Youjia Liang
a
,
*
, Lijun Liu
b
, Zhangcai Yin
a
, Jiejun Huang
a
a
School of Resources and Environmental Engineering, Wuhan University of Technology, Wuhan 430070, China
b
College of Resources and Environment, Yangtze University, Wuhan 430100, China
ARTICLE INFO
Keywords:
Urban expansion
Cropland replacement
Crop production
Climate scenarios
Direct occupancy
Indirect losses
ABSTRACT
Revealing multi-scale cascading change in land use is crucial for developing urbanization strategies and
ecological restoration in China. This study used an integrated assessment of “landscape processes-ecological
services” to simulate landscape changes over historical (1995–2020) and climate (SSP1–2.6, SSP2–4.5,
SSP5–8.5) scenarios (2020–2100) in China. Moreover, direct and indirect impacts of green landscape loss due to
urban expansion were assessed via food production changes. The historical period exhibited a nearly threefold
urban expansion, mainly occupying croplands and forest-grasslands. The direct impact of urban expansion on
forest-grasslands was more signicant than the indirect impact. Between the future scenarios, the growth rate of
national grain production signicantly differed (SSP5–8.5 >SSP2–4.5 >SSP1–2.6). SSP5–8.5 showed a high
increase in grain yield and the largest indirect occupation of forest-grasslands by urban expansion. These ndings
reveal land use trade-offs in China, offering scientic support for regional land use planning and ecosystem
management policies.
Spotlight
Studying changes in land use aids carbon reduction strategies and
ecological restoration.
The ndings revealed that urbanization resulted in green land-
scape loss, but crop yields continued to increased.
Future predictions showed that, despite cropland losses, the total
national grain production maintained stable growth.
The indirect impact of urban expansion on croplands and forest-
grasslands is enormous.
The ndings can inform future regional land use planning and
ecosystem management policies.
Data availability
Data will be made available on request.
1. Introduction
The mechanisms behind landscape pattern changes in regional socio-
ecological systems are a hot topic in global environmental change
research (Foley et al., 2005; Song et al., 2018; Zhou et al., 2023).
Changes in land use (e.g., accelerated urbanization) severely threaten
carbon reduction policies (Popp et al., 2012; Zhao et al., 2018), cropland
security (Amato et al., 2016), and "green" (vegetated land) landscapes
(Ramankutty et al., 2008), and result in signicant challenges to
terrestrial biodiversity (Newbold et al., 2015) and the sustainable supply
of ecosystem services (Gomes et al., 2021). For example, urban expan-
sion can occupy surrounding green landscapes and highly productive
croplands (Avellan et al., 2012; Bren d’Amour et al., 2017). This
urban-cropland cascade can trigger regional cropland replacement
mechanisms (conversion of ectopic forest-grassland to additional crop-
land) to compensate for crop yield losses, resulting in direct and indirect
losses of forest-grassland (van Vliet, 2019). Loss of green landscapes
(Liu and Zhou, 2021; Wu et al., 2017; Zhou et al., 2021) has a long-term
negative impact on regional sustainable development (McDonald et al.,
2020; Meyfroidt et al., 2013). The demand for materials such as food and
fuelwood continues to increase due to climate change and population
growth, which may further increase green landscape losses.
Food security is a global requirement to achieve sustainable human
* Corresponding author.
E-mail address: yjliang@whut.edu.cn (Y. Liang).
Contents lists available at ScienceDirect
Resources, Conservation & Recycling
journal homepage: www.elsevier.com/locate/resconrec
https://doi.org/10.1016/j.resconrec.2023.107228
Received 19 June 2023; Received in revised form 2 August 2023; Accepted 22 September 2023

Resources, Conservation & Recycling 199 (2023) 107228
2
development (Rosegrant and Cline, 2003), and Sustainable Develop-
ment Goals (SDGs) such as "zero hunger" emphasize the critical contri-
bution of food security to individual well-being and socio-ecological
system development (Qiu et al., 2022). However, urban expansion at the
global scale is leading to approximately 3.2% of prime cropland
becoming occupied, resulting in a global crop yield loss of more than
3.7% (Bren d’Amour et al., 2017). In China, urban expansion has
become a key factor in the loss of cropland and uctuations in per capita
cropland resources (Kuang et al., 2022). The effective use and conser-
vation of cropland resources are essential for food security (Deng et al.,
2006), sustainable economic development, and social well-being (Xie
et al., 2018; Zhou et al., 2021). For this reason, the Chinese government
has implemented rigorous cropland protection policies, for example, the
policy to balance the occupation and replenishment of arable land by
providing a red line of 120 million hectares of cultivated land. However,
the compensated cropland is mainly achieved through the occupation of
forest and grassland and is accompanied by problems such as untimely
compensation and yield quality decline of the compensated cropland
(Chen et al., 2022b; Gerten et al., 2020). Therefore, sustainable agri-
cultural development at the global-regional scale still faces signicant
challenges.
The impact of climate change on landscapes is increasing, resulting
in about 50% of China’s land area becoming ecologically vulnerable
(Liu et al., 2015). Additionally, the effects of large-scale urban-cropland
cascade changes are often associated with increased carbon emissions
due to energy use (Chuai et al., 2015; Koerner and Klopatek, 2002) and
increase the complexity of assessing carbon loss from encroached
forest-grassland lands. Therefore, clarifying the direct and indirect los-
ses of forest-grassland is an important scientic issue facing the devel-
opment of climate-resilient low-carbon strategies.
Remote sensing and modeling technologies have made it possible to
monitor and assess land use dynamics at large scales and over long time
series. For example, high-precision land use pattern-process simulations
and driving force studies (He et al., 2020b; Wang et al., 2021) have
expanded modeling abilities following the development of landscape
metrics (Bürgi et al., 2004) and multiple Cellular Automata (Gong et al.,
2018; Halmy et al., 2015; Liu et al., 2017) to simulate dynamic changes
in land use. In addition, simulation models integrating land use and crop
growth (e.g., WOFOST) have been applied to study crop-specic growth
mechanisms and crop yield changes, but they require high model input
parameters. Statistical yield models based on crop productivity and
driving forces have been utilized in large-scale arable production studies
(Ewert et al., 2015; Ray et al., 2019).
Changes in the urban-cropland cascade not only affect food pro-
duction but also cause complex changes in services such as climate
regulation, carbon storage, and soil-water conservation. Therefore,
quantitative research on the comprehensive impact of urban sprawl on
green landscapes can reveal potential changes in regional ecosystem
services and provide scientic references for the Chinese government to
optimize land management policies and ecosystem management objec-
tives. The direct effects of urban expansion on forest-grassland land-
scapes have been conrmed in existing studies (Bren d’Amour et al.,
2017; Zhou et al., 2023), but the indirect effects remain unclear. To
answer this scientic question, we rst analyzed the spatiotemporal
processes of regional land use displacement to clarify the direct occu-
pation of green landscapes by urban expansion and the loss of food
production due to encroachment on cropland, then we calculated the
area of forest and grassland encroached to compensate for the loss of
food to quantify the indirect impact of urban expansion on green land-
scape. This study assessed the comprehensive impacts of urban-cropland
cascade changes on green landscapes in nine agricultural regions of
China, based on the land use and coverage change (LUCC) dataset from
1995 to 2100, with the following aims: (1) To use the land use transfer
matrix and spatial overlay methods to analyze the multi-scale dynamic
changes of land use patterns; (2) To set up three typical scenarios
(SSP1–2.6, SSP2–4.5, SSP5–8.5) coupled with climate-socioeconomic
change pathways, and use regression models to predict changes in
grain yield in China under different scenarios; (3) To simulate the pro-
cess of cropland displacement following urban expansion to reveal the
dynamic effects on grain yield; (4) To estimate the direct and indirect
impacts of urbanization and cropland displacement processes on green
landscapes.
2. Materials and methods
2.1. Data
Signicant spatiotemporal heterogeneity in climate, geographical,
and socioeconomic development in varied regions of China leads to a
complex landscape change pattern (Chuai et al., 2018; Zhang and Chen,
2017). The ranges of nine major agricultural zones (Fig. 1A) were ob-
tained based on the principles of agricultural production conditions,
developmental characteristics positioning and administrative unit
integrity, and were used to statistically characterize the landscape
changes in different regions. LUCC data for the historical period
(1995–2020) was obtained from the European Space Agency-Climate
Change Initiative (ESA-CCI), which has better overall classication ac-
curacy globally (90%) and in China (71%) (Hartley et al., 2017; Lauer
et al., 2017; Yang et al., 2017) than other data products of the same
resolution (e.g., MCD12Q1). The LUCC data for the scenario period
(2020–2100) were obtained mainly based on three typical scenarios,
SSP1-RCP2.6, SSP5-RCP8.5, and SSP2-RCP4.5, proposed by CMIP6,
representing scenarios with low vulnerability and low radiative forcing;
no intervention for rapid development, high vulnerability, and the
highest radiative forcing; and a moderate development pathway,
respectively (van Maanen et al., 2022; Zhang et al., 2019). The selected
Global-SSP-RCP-LUCC product had a complete land classication and
good type agreement with the ESA CCI product, which had a signicant
accuracy advantage (Overall kappa coefcient =0.87) in similar data-
sets (Chen et al., 2022a). Line mask extraction of LUCC data using
Chinese national boundary maps was performed and the data were
reclassied into seven land types (Table A.1, Fig. 1). Meteorological data
included monthly mean temperature and precipitation (Peng et al.,
2019), actual evapotranspiration (Lu et al., 2021), and solar radiation
(He et al., 2020a), and annual-scale data were obtained using the raster
calculation method. The processing tool used for the raster data was
ArcGIS10.4 (ESRI 2018), projected as Krasovsky_1940_Albers, and the
data were resampled to 1 km ×1 km resolution to consider spatial detail
and computational efciency. socioeconomic data were collected from
the Municipal and China Rural Statistical Yearbooks for 360
prefecture-level cities (excluding Taiwan, Hong Kong, and Macau),
including grain production (rice, wheat, cereals, potatoes, and beans),
agricultural population, total mechanical power, effective irrigated area,
and fertilizer application (Table A.2). Missing data were replaced by the
average value of the nearest adjacent years or estimated based on the
difference between the total amount in the province and other cities in
the same province.
2.2. Landscape pattern change analysis
Land use transfer matrix and land use dynamic degree (LUDD)
analysis methods were used to obtain landscape pattern changes at
separate periods (Eqs.1 and 2). The land use transfer matrix quantita-
tively describes the area interconversion of various land types between
specic time phases, which can improve understanding of the trends and
patterns of LUCC (Ning et al., 2018); the LUDD analysis reveals the in-
tensity and speed of LUCC (Zhou et al., 2020).
A=

S11 ⋯S1n
⋮ ⋱ ⋮
Sn1⋯Snn

(1)
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
3
Di=Ul−Up
UP
×1
t×100% (2)
where A is the land use transfer matrix; i and j are the land types before
and after the transfer, respectively; n indicates the number of land use
types; S
ij
represents the land area (10
4
km
2
) converted from land class i
to land class j. Diis the dynamic degree (%) of land type i; UPand Ul
denote the area of a specic land type at the initial period and the end
(10
4
km
2
); t denotes the number of years of study.
2.3. Grain production estimates
Logistic regression equations of grain yield in the nine agricultural
zones were established to simulate total grain yield at the municipal
level. The dependent variable was grain yield, and nine independent
variables inuencing grain yield were determined from the ndings of
existing studies and data (Chen et al., 2020c; Liu and Zhou, 2021; Lu
et al., 2019; Wang et al., 2022), including meteorological factors
(average annual temperature, annual precipitation, annual actual
evapotranspiration, and solar radiation), economic and social factors
(agricultural population, cropland area), and agrarian input manage-
ment factors (total mechanical power, effective irrigation area, and
fertilizer application). The municipal averages of the meteorological
raster data were calculated and entered into SPSS 26.0, along with other
data, to perform multiple linear regression. Using 2020 as the base
period, the parameter values for each scenario were determined by
considering the rate of change of indicators specied in diverse climate
scenarios (Table A.3) (Zhang et al., 2019), and the urban grain pro-
duction of each scenario was predicted based on the obtained regression
equation (Table A.4).
2.4. Occupation of green landscapes by urban expansion
First, annual average grain yields were calculated for each munici-
pality under different scenarios and combined with the rate of cropland
change for each municipality (Eq. (3)), using the "cropland-yield" index
to detect the relationship between the intensity of regional cropland
change and grain production (Eq. (4)).
CCRi=Cl−Cp
CP
(3)
CY =
n
i=1
|CCRi| × Yi
ymax
(4)
where, CCRi denotes the rate of change of cropland area (%), Cl denotes
the initial cropland area (km
2
), CP denotes the nal cropland area (km
2
),
CY denotes the "cropland area-yield level" index, Yi indicates the
average yield per unit area (t/km
2
) of each municipality during the
study period, and ymax indicates the maximum annual average yield per
unit area (t/km
2
) of each municipality. Larger values of CY indicate that
the cropland in areas with high food production has changed more
dramatically.
Then, the reduction in grain production resulting from urban
expansion was computed. Urban expansion directly occupies green
landscapes and surrounding high-quality cropland. It was assumed that
the policy of cropland acquisition and replenishment was strictly
enforced and new cropland was acquired through the occupation of
green landscapes. To measure the impact of this replacement mecha-
nism, the indirect impact of urban expansion on green land patterns was
assessed using food production as a mediating variable, as calculating
food loss due to urban expansion requires compensating for how much
forest and grassland was occupied (Eq. (5)). The LUCC data analysis
indicated that the urban area will not expand after 2050, therefore this
study assumed that the urban area will remain unchanged until 2050
and the direct and indirect impacts of urban expansion on forest-
grassland were only analyzed until 2050.
EI =n
1NiYi
y×n
1UiYi
n
1DiYi
×P(5)
EIT =n
1UiYi
y×P(6)
where EI represents the indirect impact of urban expansion on forest-
grassland landscape (km
2
); n represents the number of cities within
each agricultural zone; Yi represents the average production per unit
area in the city i every 5 years (t/km
2
); y represents the average food
yield per unit area in each agricultural zone every 5 years (t/km
2
);
Niand Di represent the expansion and contraction of cropland area in
each city every 5 years (km
2
), respectively; Ui represents the extent of
cropland encroached by urban land every 5 years (km
2
); P represents the
proportion of forest-grassland area encroached by the new cropland.
Food production from new cropland often fails to fully offset the yield
Fig. 1. Location of the nine major agricultural regions in China. (A) Land-use in 2020; (B) Geographical features; and (C) Percentage of land-use types in the nine
major agricultural regions in 2020.
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
4
decline, and assuming that it can fully compensate for the loss of food
production due to urban expansion, the indirect impact of theoretical
urban expansion on forest-grassland landscapes (EIT, Eq. (6)) was
further estimated, with the equation parameters above. When EIT >EI,
the supplementary cropland area is inadequate to offset the decline in
yield. The integrated assessment methodology used in this study was
shown in Fig. 2.
3. Results
3.1. National-regional scale LUCC
The most common land use type in China during the historical period
was grassland (39.00%–38.38%), followed by unused land (21.61%–
21.06%), forests (17.68%–17.66%), and cropland (17.07%–17.03%),
whilst all other types comprised less than 2.38% (Table A.5). Despite the
solid ecological conservation projects and cropland preservation policies
implemented by the Chinese government, the national areas grassland,
cropland, and forest were reduced by 586 Mha, 41 Mha, and 16 Mha,
respectively. Although urban land only accounts for 0.40%–1.51%
(Fig. 3) of land, urban area increased to 1459 Mha in 2020, a 2.83-fold
increase, and had the highest LUDD (13.31%, Fig. 4B), mainly occu-
pying cropland and grassland (Fig. 3A). Comparison of the changes in
landscape patterns between the three climate scenarios, with 2020 as
the baseline, showed that the area of grassland may continue to decrease
(115.10–185.76 Mha), resulting in cropland (70.40–83.77 Mha) and
forests (45.63–52.06 Mha) signicantly increasing (Table A.6). The
decrease in grassland area was the largest in the SSP1–2.6 scenario
(Fig. 3B), including the conversion of 46.05 Mha to unused land, but the
net loss of green landscapes will be compensated by the increase in forest
area (87.61 Mha). The growth rate of urban area decreased in all sce-
narios, and the urban area was the largest in SSP5–8.5, indicating that
the high urbanization rate and population growth in this scenario
require more urban land to meet development needs. The total urban,
cropland, and forest-grassland area changes in SSP2–4.5 were the me-
dian across all the scenarios. The urban-cropland-forest and grassland
occupation in the scenario period was signicantly different at the na-
tional scale. The smallest area of urban expansion was in the SSP2–4.5
scenario (Fig. 3C), directly occupying 186 and 163 Mha of cropland, and
forest and grassland, respectively, and the cropland occupied the
smallest amount of forest-grassland (8930 Mha). For SSP1–2.6, urban
expansion occupied 210 and 163 Mha of cropland and forest-grassland,
respectively, whilst the indirect encroachment of forest and grassland by
cropland was the largest (10,263 Mha). The largest area of urban
expansion was in the SSP5–8.5 scenario (Fig. 3D), occupying 213 and
176 Mha of cropland, and forest and grassland, respectively, whilst
encroachment of forest and grassland by cropland was 9657 Mha.
For agricultural regions, the increase in cropland area in the histor-
ical period (Fig. 4A) was mainly distributed in the Northern arid and
semiarid region (NS) and part of the Qinghai-Tibet Plateau (QT). In
contrast, the cropland area in Southern China (SC), the middle-lower
Yangtze Plain (MY) and other south-central regions was reduced and
mostly transformed into forest-grassland and urban land. The urban land
in the Yangtze River Delta and the Pearl River Delta increased signi-
cantly. Comparing the spatial characteristics of LUCC changes in the
nine agricultural regions (Fig. A.1, Table A.7), the NS and QT mainly
showed reciprocal transformation between unused land and grassland,
and were also the only regions in China with increased cropland area
(296.12 Mha). The urban area of NC and Huang-Huai-Hai Plain (HHH)
signicantly increased, accounting for 1.52% and 32.49% of the in-
crease in urban land area nationwide, respectively. The loss of grassland
area (338 Mha) was profound, while HHH had the highest loss of
cropland (148.49 Mha) nationwide, accounting for 45.07% of the total
loss. The forest in LP and SB continued to increase (163.38 Mha) and the
conversion of cropland to urban land continued, resulting in the loss of
cropland (84.68 Mha), accounting for 25.70% of the total loss. The MY
and SC exhibited the most severe losses of forest, mainly due to con-
versions to cropland and grassland. The total area loss reached 4.94
Mha, with MY and SC accounting for 62.89% and 36.89% of the total
forest loss, respectively. Moreover, the rapid economic development of
these two regions, the loss of cropland resulting from urban expansion
(270.05 Mha), and the increase in urban land in the two regions
accounted for 33.48% of the total increased area.
The scenario LUCC mainly occurs during the period 2020–2050 as,
after 2050, there will be no further urban expansion and consequential
loss of cropland or forest-grassland. The spatial distribution of LUCC in
the three scenarios is basically the same, but the area of change varies
and has different characteristics in different regions (Fig. A.2). The
SSP1–2.6 scenario shows extensive grassland degradation in NS and QT
and conversion mainly to unused land, but also of grassland to cropland
in NS (Fig. A.2, A and B), resulting in the highest grassland loss (8379.71
Mha) among the three scenarios (Fig. 3C). The most signicant con-
version of cropland to forest-grassland in the SSP2–4.5 scenario
(Fig. A.2, C and D) occurred in the Yunnan-Guizhou Plateau (YG) (1034
Mha), Sichuan basin (SB) (406.27 Mha), and SC (344.75 Mha). The
SSP5–8.5 scenario showed a large area of unused land converted to
grassland, mainly in NS and QT (Fig. A.2, E and F), resulting in increases
of 1516.48 and 1353.18 Mha, respectively.
Fig. 2. The integrated assessment methodology used in this study.
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
5
3.2. National-regional scale changes in food production
The obtained regression equations for grain yield estimation all had
R
2
greater than 0.8, and the overall t was good for the scenario yield
estimation (Table A.3). The geographical pattern of national grain
production in the historical period (Fig. 5A) showed that the high-value
regions were concentrated in the NC, HHH, and MY. The QT and NS are
constrained by climate, soil conditions, and other natural factors
(Fig. A.3), therefore grain production was relatively low. Rapid eco-
nomic development in SC has continuously compacted agricultural
space, resulting in reduced cropland area and low food production
nationwide. Even though the cropland area decreased by 0.41 ×10
4
km
2
, the national average grain yield increased from 284.62 t/km
2
in
1995 to 409.84 t/km
2
in 2020, resulting in an overall increase in grain
production, which continued to increase during the scenario period
(Fig. 5B). The SSP1–2.6 scenario had the slowest and lowest increase in
grain production among the scenarios, with lower population growth
and grain demand leading to signicantly lower yields compared to that
of 2020, with a national average grain yield per unit of about 370 t/km
2
by 2100, cropland area of 233.43 ×10
4
km
2
, and grain production of
86,542 ×10
4
t (Fig. 5, C and D), which is equivalent to 1.29 times the
base period. The SSP2–4.5 scenario had a national average grain yield of
439.86 t/km
2
, the lowest cropland area of all scenarios in 2100 (218.05
×10
4
km
2
), and a total grain production of 95,909.28 ×10
4
t, equiv-
alent to 1.43 times the base period. The SSP5–8.5 scenario exhibited the
highest growth rate in both grain yield and cropland area. By 2100, the
national average grain yield and the cropland area will reach 480 t/km
2
and 236.87 ×10
4
km
2
, respectively, and the grain production will reach
113,158 ×10
4
t, which is 1.69 times that of 2020, probably due to the
higher population growth rate and technological innovation in this
scenario.
During the scenario period, there was signicant spatial variation in
grain production across China (Fig. 6). Increased grain production was
observed across NC, HHH, and SC, whilst decreased grain production
was observed across the eastern and western parts of the NS and QT, and
southwestern parts of YG. In SSP1–2.6 (Fig. 6, A and B), the increases
and decreases in production in each city were small, and the increase in
total production from 2020 to 2050 was mainly in NC and SC (46.39%
and 50.22%, respectively) and grain production in NS and LP increased
by 8%. The remaining regions showed a reduction in production, with
the QT, YG, and SB accounting for approximately 17.28–34.73% of the
total production reduction. Only the MY showed an increase in pro-
duction from 2050 to 2100, which was only by 4.20%, and the total
national scale food production in 2100 was 1.29 times that of the base
period. In SSP2–4.5 (Fig. 6, C and D), more agricultural areas showed
increased yields compared to SSP1–2.6. The agricultural areas with the
highest increase in yields in 2020–2050 were the SC and NC (70.87%
and 64.80%, respectively); the yield increases in the remaining areas
(HHH and LP) were approximately 10%, and the main areas with
decreased yield were NS, QT and SB, with decreases of 15.47–35.12%.
By 2100, compared with 2050, the production of most regions will
decrease, but the NS and MY showed a clear increasing trend, with
production increasing to 1.43 times that of the base period. In the
SSP5–8.5 scenario (Fig. 6, E and F), urbanization development was not
strongly constrained, resulting in a polarized trend of increasing or
decreasing food production in different cities, with the highest increase
and decreases among the three scenarios and total decreases of 17.59%
and 57.34% in QT and NS, respectively, and pronounced decreases in
food production in some cities. Food production in the SB and YG
showed <1% reduction, while that of the remaining regions showed an
Fig. 3. Land use changes for 1995–2020 (Historical) and 2020–2100 (SSP1–2.6, SSP2–4.5, SSP5–8.5) in China (unit: 10
2
Mha).
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
6
increasing trend, especially the NC and SC, which increased by >90%.
Total grain production was 1.69 times that of the base period by 2100.
Despite the different trends of increasing and decreasing grain produc-
tion in the three scenarios for the nine agricultural regions, the total
national production showed an increasing trend compared with the base
period.
The CY index test showed that the cropland in the high production
areas changed the most in the SSP5–8.5, whilst those in the SSP1–2.6
scenario changed less (Fig. 5E). Comprehensive analysis of the changes
in land use and food production in each agricultural region revealed an
increasing trend in the national production, although there are obvious
differences between different agricultural regions. Cropland is gradually
moving towards the less productive Northwest, but national production
security comes mainly from the NC and HHH. Additionally, the NC has
been able to maintain a trend of steady growth in production, even with
the loss of cropland in the historical period. In the future scenario, as the
area of cropland increases, the production will always be at the leading
level in the country (Fig. 6), due to the advantages provided by the re-
gion’s nature, and making an important contribution to food security as
“China’s breadbasket”. The HHH, MY, SB, and SC, had median grain
yields, among which the decrease in grain yield and cropland area in SC
was remarkable. The average grain yield in four regions (NS, LP, QT, and
YG) was <200 t/km
2
in 2100. For the areas where production condi-
tions are clearly constrained, the area of cultivated land showed an
increasing trend under the inuence of the policy of levy and subsidy
balancing, but total production still declined. This developing trend,
while guaranteeing food security, does not take advantage of regional
strengths and remains detrimental to the sustainable development of
agriculture.
3.3. Impact of urban expansion on green landscape occupation
Analysis of the change in grain production in each city in each sce-
nario (Fig. 6) revealed that the change in food production in SSP5–8.5
was higher than that in the other scenarios, indicating that the cropland
change in this scenario was more drastic. In SSP1–2.6, the smaller
change in production indicated that the cropland change was relatively
gradual, impacting food production and green landscapes less, which
was generally in line with the outcomes of CY index analysis (Fig. 5E).
Further examination of the direct, indirect, and theoretical indirect
impacts of urban expansion on green landscapes in each agricultural
region (Fig. 7E) showed that the nationwide indirect impacts on forest-
grassland landscapes surpassed the direct impacts. The cumulative
urban expansion in the historical period directly encroached on
48,871.22 km
2
of green landscapes, and approximately 54,400 km
2
of
cropland, resulting in an indirect loss of 33,673.01 km
2
of green land-
scapes. Theoretically, the indirect impact area could reach 50,714.45
km
2
, making it much larger than the direct impact. The indirect impacts
(7363.33–14,249.84 km
2
) are much greater than the direct impacts
(3954–4868 km
2
) for all three scenarios from 2020 to 2045. SSP1–2.6
had the smallest indirect and direct impacts among the three scenarios,
likely due to the green development strategy. In contrast, SSP5–8.5 had
the most extensive loss of green landscapes and indirect loss, which was
about twice that of SSP1–2.6 and is a typical unsustainable development
pattern.
In the agricultural regions (Fig. 7), the largest indirect impact during
Fig. 4. Spatial distribution of LUCC (A) and land-use dynamic degree (B) in the nine major agricultural regions of China during the historical period
(1995–2020). “Other” represents areas that showed no change and land-use changes other than those shown in the legend.
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
7
the historical period was in MY, accounting for 43.27% of the national
indirect impact losses, followed by SC (32.20%). The total indirect loss
area of the two regions accounted for 75.47% of the national total,
indicating that the indirect loss of green landscapes due to urban
expansion nationwide was mainly in SC. This could be attributed to the
accelerated economic growth and the larger initial forest-grassland in
the southern region compared to other regions. The most signicant
direct impact area occurred in HHH, which comprised 43.20% of the
national area of forest-grassland converted to urban land. The loss of
forest and grassland in each agricultural area of the three scenarios was
relatively consistent. The area with the largest direct and indirect loss
was HHH, and the scenario largest indirect loss was SSP2–4.5 (3816.33
km
2
). The area with the lowest direct and indirect loss of forest and
grasslands is QT, where the large forest and grassland areas remain
relatively undisturbed by human activities, thus the comprehensive
impact of urban expansion on this area is negligible. In areas where
green landscapes are more affected by the combined effects of urban
expansion, food production is often threatened. For example, in both MY
and HHH, there was a reduction in food production in the future sce-
narios. In the YG area, where the loss of green landscapes was small,
there was also a decline in yields due to the small replenishment of
cropland (Fig. 6 and 7). The actual replenished cropland area cannot
fully offset the yield decline due to the series of ecological policies. The
theoretical indirect loss area is greater than the actual indirect loss area
in both historical and future scenarios (Fig. 7E), and the largest differ-
ence between the two indicators was reached in SSP1–2.6 (11,665.34
km
2
), followed by SSP5–8.5 (6615.91 km
2
). The slight difference in
SSP2–4.5 (2551.92 km
2
) reects the higher replenishment of cropland
and the low loss of food production under this scenario. The region with
the largest gap between the two indicators was SC, indicating that the
loss of cropland due to urban expansion in the region was not adequately
replenished, and food production was affected.
4. Discussion
4.1. Food security changes due to urban expansion
As the global economy and population have grown, food security has
become a global issue. The SDG goal of “zero hunger” has become a
governance priority. Despite increasing global food production, national
Fig. 5. Changes in cropland area and grain production in China. (A) Grain production level by city in China in 2020 (10
4
t); (B) Change in grain yield in China
(10
4
t), 1995–2100; (C) Change in cropland area in China (10
4
km
2
), 1995–2100; (D) Grain yield per unit area (t/km
2
) in China, 1995–2100; (E) Cropland-Yield
index for China and nine major agricultural regions.
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
8
food decits remain. Diverse contingencies (e.g., wars, epidemics)
(Carriquiry et al., 2022) and climate change (e.g. climate extremes) pose
severe challenges to food security. This study investigated a specic set
of climate-social-economic coupling parameters and combined them
with regional grain production to establish grain yield projection models
under different scenarios in China. We found that total grain production
was highest under SSP5–8.5 in 2050 (approximately 10,970 ×10
4
t).
Total grain production increased the least under SSP1–2.6 (8610 ×
10
4
t), and the production growth was 21.61–63.57% compared to the
three scenarios in the base period. By 2100, total production will reach
about 113,200 ×10
4
t under the SSP5–8.5; the lowest growth in total
food production was 86,500 ×10
4
t under the SSP1–2.6, with production
growth signicantly decreasing and ranging from 0.80 to 6.00%
compared to that in 2050, indicating an increased pressure to cope with
food shortage situations after 2050.
Under the SSP1–2.6 scenario, both ecological environment and food
security will be in relative equilibrium. In contrast, the SSP5–8.5 sce-
nario has higher food security but excessive loss of green landscapes,
which is not conducive to sustainable ecological environment develop-
ment. Therefore, despite the challenges of future environmental
resource changes, we expect to maintain a stable growth of food pro-
duction and recommend that the demand of local population growth be
met by macro-regulation of food production and internal mobility for
the NS, YG, and MY. We also recommend strengthening the manage-
ment and regulatory system of cropland and food production, such as by
improving the quality of additional cropland under the expropriation
and replenishment policy, and strengthening the protection of perma-
nent basic cropland to maintain the sustainability and stability of food
production.
4.2. Ecosystem management changes due to green landscape loss
The Chinese government has conducted substantial policy measures
for environmental protection and ecological restoration since the 1970s
(Sun et al., 2020), providing a solid foundation for economic develop-
ment (Shao et al., 2022). The overlay analysis of the coverage of the
eight ecological projects in China and the LUCC in the historical period
(Fig. A.5) reveals that the area of forest and grassland increased by the
eight ecological projects in the historical period accounts for 77.04% of
the national total increased area, and the area increased by Three-North
Fig. 6. Map of three scenarios (SSP1–2.6, SSP2–4.5, SSP5–8.5, 2020–2100) of grain yield (10
4
t) in China. The bar chart represents the value of change in total
food production in each agricultural region.
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
9
Shelter Forestation Project’s forest area is the largest (8.46´
I10
4
km
2
),
accounting for 45.24%. Long-term ecological projects carried out in
China are effective for the sustainable development of ecosystems, but
the competition for land between afforestation and food production is
persistent (Qiu et al., 2022). The policy of balancing cropland acquisi-
tion and replenishment further leads to the loss of forest and grassland
and threatens the long-term continuity of carbon sinks in terrestrial
ecosystems (Gao et al., 2022). Thus, the complex trade-off between food
security and ecological safety needs to be addressed through systemic
solutions that ensure both the growth and stability of food production
and supply, while preserving the sustainability of natural resources and
the ecological environment. This requires the joint efforts of govern-
ments, businesses, and citizens to develop and implement sustainable
development strategies and policies, promote the modernization and
green transformation of agricultural production, strengthen environ-
mental protection and ecological restoration, and improve ecosystem
stability and resilience.
4.3. Uncertainty
This study simulated the impacts of LUCC on food production under
typical development scenarios and developed a long-time series of
urban-green landscape cascade change processes at national-regional
scales based on changes in food production and identied the direct
and indirect occupations of forest and grassland by urban expansion.
The impact of urban expansion on China’s forest and grassland is un-
deniably signicant, aligning with the general ndings of a limited
number of multi-scale investigations (Ke et al., 2018; van Vliet, 2019;
Wang et al., 2022). Overall, the national cropland area and grain yields
are increasing, and food security remains intact, in line with the ndings
of earlier research (Liang et al., 2023). However, the signicant vari-
ability of different agricultural zones should be considered. For example,
grain production in NC will continue to increase in the future period,
while production in MY, HHH and YG shows a decreasing trend. This
paper integrates the analysis of functional changes in food production
with the effects of urban-cropland cascade changes. This line of research
can provide a research methodology and case reference for assessing
ecosystem services provided by green landscapes, as well as a research
basis for multiple ecosystem service trade-offs/synergy analysis. The
uncertainty of the study is as follows: First, the LUCC data in this study
passed the precision test, but the use of higher spatial resolution land use
maps (e.g., 30 m) and driver data matching their resolution would
improve the precision of the study and avoid abrupt changes in data
trends due to inconsistent data sources (e.g., 2020). Nevertheless, using
high spatial resolution data affects computational efciency, which
needs to be traded off. Second, regression analysis is used in yield
Fig. 7. Direct and indirect impacts of urban expansion on the forest-grassland landscape 1995–2045. EI represents the actual indirect impact, EIT represents
theoretical indirect impact, and DI represents direct impact. (A-D) Impacts of urban expansion in nine agricultural regions 1995–2045. (E) Impacts of urban
expansion in China 1995–2045.
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
10
simulation, which involves obtaining yield data from multiple sources
for cross-comparison or developing crop growth models based on bio-
physical processes to simulate yield, therefore is limited by the accuracy
of grain yield data, and the complexity of the models and data re-
quirements will further increase. The regression models established by
regional decomposition compensate for the limitations of data accuracy,
and all models show good accuracy and robustness. Finally, the pattern
of occupation of green landscapes by urban expansion primarily relies
on the precise simulation of land use patterns and grain yield, and the
accuracy of green landscape occupation areas can be effectively
improved by improving the accuracy of these two indexes in the future.
Meanwhile, comparing the impact of LUCC data from different models
on the accuracy of assessment results is also required.
5. Conclusion
This study analyzed the multi-scale urban-cropland cascade changes
through multi-scenario LUCC data and integrated assessment methods.
It revealed the direct-indirect impacts of urban expansion on forest-
grassland in historical periods and future scenarios. The main conclu-
sions of the study are as follows: (1) The continuous urban expansion
during the historical period has caused a loss of 544 Mha and 497 Mha of
cropland and forest-grassland, respectively, and the pattern of LUCC
change varies signicantly among different agricultural regions, with
the increase in cropland distributed in the northwest and the severe
degradation of forest in the central and eastern coastal regions; (2) In the
future scenario, even with the reduction of cropland, the total national
grain production increased. The grain production in the NC remains the
national leader, while the grain production in the northwestern region
continues to be low due to the environmental conditions, and the MY is
restricted in agricultural development space due to economic develop-
ment, leading to a substantial decline in grain production; (3) The
largest indirect impact of cascading changes in the historical period was
distributed in the southern region (MY and SC), resulting in a total loss
of forest-grassland of 25,411.06 km
2
, accounting for 75.47% of the
country’s loss. The greatest loss of forest-grassland was in the SSP5–8.5
scenario, of which the indirect loss reached 14,249.84 km
2
, nearly 30%
of the country’s indirect loss of forest and grassland. This loss occurs in
HHH, which needs further improvement of land planning policies in
future urban development to ensure that green landscapes are not
infringed upon; (4) The SSP1–2.6 scenario can better maintain the bal-
ance between ecological environment and food security, and achieving a
low emission development scenario requires future attention. In
conclusion, the inuence of urban expansion on cropland and forest-
grassland is extensive, and future government policy planning should
account for this and further integrate the impact of food production on
other ecosystem services to provide a guiding framework for attaining
sustainable and coordinated development of regional ecosystems
throughout the country.
CRediT authorship contribution statement
Yating He: Data curation, Software, Visualization, Writing – original
draft, Investigation. Youjia Liang: Conceptualization, Methodology,
Supervision, Data curation, Software, Visualization, Writing – original
draft, Investigation, Writing – review & editing, Funding acquisition.
Lijun Liu: Writing – review & editing, Funding acquisition. Zhangcai
Yin: Writing – review & editing. Jiejun Huang: Writing – review &
editing.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
Data will be made available on request.
Funding
This work was supported by the Science Foundation of Hubei Prov-
ince, China (2021CFB295), the China Postdoctoral Science Foundation
(2023M730363), the CMA Key Open Laboratory of Transforming
Climate Resources to Economy (2023016), and the National Natural
Science Foundation of China (42171415). The authors would like to
thank the anonymous reviewers and editors for their invaluable com-
ments to improve this paper.
Supplementary materials
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.resconrec.2023.107228.
References
Amato, F., Maimone, B.A., Martellozzo, F., Nol`
e, G., Murgante, B., 2016. The effects of
urban policies on the development of urban areas. Sustainability 8, 297. https://doi.
org/10.3390/su8040297.
Avellan, T., Meier, J., Mauser, W., 2012. Are urban areas endangering the availability of
rainfed crop suitable land? Remote. Sens. Lett. 3, 631–638. https://doi.org/
10.1080/01431161.2012.659353.
Bren d’Amour, C., Reitsma, F., Baiocchi, G., Barthel, S., Güneralp, B., Erb, K.-H.,
Haberl, H., Creutzig, F., Seto, K.C., 2017. Future urban land expansion and
implications for global croplands. Proc. Natl. Acad. Sci. U.S.A. 114, 8939–8944.
https://doi.org/10.1073/pnas.1606036114.
Bürgi, M., Hersperger, A.M., Schneeberger, N., 2004. Driving forces of landscape change
— current and new directions. Landsc. Ecol. 19, 857–868. https://doi.org/10.1007/
s10980-004-0245-8.
Carriquiry, M., Dumortier, J., Elobeid, A., 2022. Trade scenarios compensating for halted
wheat and maize exports from Russia and Ukraine increase carbon emissions without
easing food insecurity. Nat. Food 3, 847–850. https://doi.org/10.1038/s43016-022-
00600-0.
Chen, G., Li, X., Liu, X., 2022a. Global land projection based on plant functional types
with a 1-km resolution under socio-climatic scenarios. Sci. Data 9 (125). https://doi.
org/10.1038/s41597-022-01208-6.
Chen, H., Tan, Y., Xiao, W., Li, G., Meng, F., He, T., Li, X., 2022b. Urbanization in China
drives farmland uphill under the constraint of the requisition–compensation balance.
Sci. Total Environ. 831, 154895 https://doi.org/10.1016/j.scitotenv.2022.154895.
Chen, Y., Guo, F., Wang, J., Cai, W., Wang, C., Wang, K., 2020c. Provincial and gridded
population projection for China under shared socioeconomic pathways from 2010 to
2100. Sci. Data 7 (83). https://doi.org/10.1038/s41597-020-0421-y.
Chuai, X., Huang, X., Wang, W., Zhao, R., Zhang, M., Wu, C., 2015. Land use, total
carbon emissions change and low carbon land management in Coastal Jiangsu,
China. J. Clean. Prod., Carbon Emissions Reduction: policies, Technologies,
Monitoring. Assessment and Modeling 103, 77–86. https://doi.org/10.1016/j.
jclepro.2014.03.046.
Chuai, X., Qi, X., Zhang, X., Li, Jiasheng, Yuan, Y., Guo, X., Huang, X., Park, S., Zhao, R.,
Xie, X., Feng, J., Tang, S., Zuo, T., Lu, J., Li, Jianbao, Lv, X., 2018. Land degradation
monitoring using terrestrial ecosystem carbon sinks/sources and their response to
climate change in China. Land Degrad. Dev. 29, 3489–3502. https://doi.org/
10.1002/ldr.3117.
Deng, X., Huang, J., Rozelle, S., Uchida, E., 2006. Cultivated land conversion and
potential agricultural productivity in China. Land use policy 23, 372–384. https://
doi.org/10.1016/j.landusepol.2005.07.003.
Ewert, F., R¨
otter, R.P., Bindi, M., Webber, H., Trnka, M., Kersebaum, K.C., Olesen, J.E.,
van Ittersum, M.K., Janssen, S., Rivington, M., Semenov, M.A., Wallach, D., Porter, J.
R., Stewart, D., Verhagen, J., Gaiser, T., Palosuo, T., Tao, F., Nendel, C., Roggero, P.
P., Bartoˇ
sov´
a, L., Asseng, S., 2015. Crop modelling for integrated assessment of risk
to food production from climate change. Environ. Model. Softw. 72, 287–303.
https://doi.org/10.1016/j.envsoft.2014.12.003.
Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S.,
Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A.,
Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., Snyder, P.K.,
2005. Global Consequences of Land Use. Science 309, 570–574. https://doi.org/
10.1126/science.1111772.
Gao, R., Chuai, X., Ge, J., Wen, J., Zhao, R., Zuo, T., 2022. An integrated tele-coupling
analysis for requisition–compensation balance and its inuence on carbon storage in
China. Land Use Policy 116, 106057. https://doi.org/10.1016/j.
landusepol.2022.106057.
Gerten, D., Heck, V., J¨
agermeyr, J., Bodirsky, B.L., Fetzer, I., Jalava, M., Kummu, M.,
Lucht, W., Rockstr¨
om, J., Schaphoff, S., Schellnhuber, H.J., 2020. Feeding ten billion
Y. He et al.

Resources, Conservation & Recycling 199 (2023) 107228
11
people is possible within four terrestrial planetary boundaries. Nat. Sustain. 3,
200–208. https://doi.org/10.1038/s41893-019-0465-1.
Gomes, E., In´
acio, M., Bogdzeviˇ
c, K., Kalinauskas, M., Karnauskait˙
e, D., Pereira, P., 2021.
Future land-use changes and its impacts on terrestrial ecosystem services: a review.
Sci. Total Environ. 781, 146716 https://doi.org/10.1016/j.scitotenv.2021.146716.
Gong, J., Hu, Z., Chen, W., Liu, Y., Wang, J., 2018. Urban expansion dynamics and modes
in metropolitan Guangzhou, China. Land Use Policy 72, 100–109. https://doi.org/
10.1016/j.landusepol.2017.12.025.
Halmy, M.W.A., Gessler, P.E., Hicke, J.A., Salem, B.B., 2015. Land use/land cover change
detection and prediction in the north-western coastal desert of Egypt using Markov-
CA. Appl. Geogr. 63, 101–112. https://doi.org/10.1016/j.apgeog.2015.06.015.
Hartley, A.J., MacBean, N., Georgievski, G., Bontemps, S., 2017. Uncertainty in plant
functional type distributions and its impact on land surface models. Remote Sens.
Environ, Earth Observ. Essential Climate Variables 203, 71–89. https://doi.org/
10.1016/j.rse.2017.07.037.
He, J., Yang, K., Tang, W., Lu, H., Qin, J., Chen, Y., Li, X., 2020a. The rst high-resolution
meteorological forcing dataset for land process studies over China. Sci. Data 7 (25).
https://doi.org/10.1038/s41597-020-0369-y.
He, T., Xiao, W., Zhao, Y., Deng, X., Hu, Z., 2020b. Identication of waterlogging in
Eastern China induced by mining subsidence: a case study of Google Earth Engine
time-series analysis applied to the Huainan coal eld. Remote Sens. Environ. 242,
111742 https://doi.org/10.1016/j.rse.2020.111742.
Ke, X., van Vliet, J., Zhou, T., Verburg, P.H., Zheng, W., Liu, X., 2018. Direct and indirect
loss of natural habitat due to built-up area expansion: a model-based analysis for the
city of Wuhan, China. Land Use Policy. Land Use Rural Sustain. China 74, 231–239.
https://doi.org/10.1016/j.landusepol.2017.12.048.
Koerner, B., Klopatek, J., 2002. Anthropogenic and natural CO
2
emission sources in an
arid urban environment. Environ. Pollut. 116, S45–S51. https://doi.org/10.1016/
S0269-7491(01)00246-9.
Kuang, W., Liu, J., Tian, H., Shi, H., Dong, J., Song, C., Li, X., Du, G., Hou, Y., Lu, D.,
Chi, W., Pan, T., Zhang, S., Hamdi, R., Yin, Z., Yan, H., Yan, C., Wu, S., Li, R.,
Yang, J., Dou, Y., Wu, W., Liang, L., Xiang, B., Yang, S., 2022. Cropland
redistribution to marginal lands undermines environmental sustainability. Natl. Sci.
Rev. 9 (nwab091) https://doi.org/10.1093/nsr/nwab091.
Lauer, A., Eyring, V., Righi, M., Buchwitz, M., Defourny, P., Evaldsson, M.,
Friedlingstein, P., de Jeu, R., de Leeuw, G., Loew, A., Merchant, C.J., Müller, B.,
Popp, T., Reuter, M., Sandven, S., Senftleben, D., Stengel, M., Van Roozendael, M.,
Wenzel, S., Will´
en, U., 2017. Benchmarking CMIP5 models with a subset of ESA CCI
Phase 2 data using the ESMValTool. Remote Sens. Environ., Earth Observ. Essential
Climate Variables 203, 9–39. https://doi.org/10.1016/j.rse.2017.01.007.
Liang, Z., Sun, L., Tian, Z., Fischer, G., Yan, H., 2023. Increase in grain production
potential of China under climate change. PNAS Nexus 2. https://doi.org/10.1093/
pnasnexus/pgad057 pgad057.
Liu, J., Zou, C., Gao, J., Ma, S., Wang, W., Wu, K., Liu, Y., 2015. Location determination
of ecologically vulnerable regions in China. Biodiversity Sci. 23, 725–732. https://
doi.org/10.17520/biods.2015147.
Liu, X., Liang, X., Li, X., Xu, X., Ou, J., Chen, Y., Li, S., Wang, S., Pei, F., 2017. A future
land use simulation model (FLUS) for simulating multiple land use scenarios by
coupling human and natural effects. Landsc. Urban Plan. 168, 94–116. https://doi.
org/10.1016/j.landurbplan.2017.09.019.
Liu, Y., Zhou, Y., 2021. Reections on China’s food security and land use policy under
rapid urbanization. Land Use Policy 109, 105699. https://doi.org/10.1016/j.
landusepol.2021.105699.
Lu, J., Wang, G., Chen, T., Li, S., Hagan, D.F.T., Kattel, G., Peng, J., Jiang, T., Su, B.,
2021. A harmonized global land evaporation dataset from model-based products
covering 1980–2017. Earth Syst. Sci. Data 13, 5879–5898. https://doi.org/10.5194/
essd-13-5879-2021.
Lu, S., Bai, X., Li, W., Wang, N., 2019. Impacts of climate change on water resources and
grain production. Technol. Forecast. Soc. Change 143, 76–84. https://doi.org/
10.1016/j.techfore.2019.01.015.
McDonald, R.I., Mansur, A.V., Ascens˜
ao, F., Colbert, M., Crossman, K., Elmqvist, T.,
Gonzalez, A., Güneralp, B., Haase, D., Hamann, M., Hillel, O., Huang, K., Kahnt, B.,
Maddox, D., Pacheco, A., Pereira, H.M., Seto, K.C., Simkin, R., Walsh, B., Werner, A.
S., Ziter, C., 2020. Research gaps in knowledge of the impact of urban growth on
biodiversity. Nat. Sustain. 3, 16–24. https://doi.org/10.1038/s41893-019-0436-6.
Meyfroidt, P., Lambin, E.F., Erb, K.-H., Hertel, T.W., 2013. Globalization of land use:
distant drivers of land change and geographic displacement of land use. Curr. Opin.
Environ. Sustain. 5, 438–444. https://doi.org/10.1016/j.cosust.2013.04.003.
Newbold, T., Hudson, L.N., Hill, S.L.L., Contu, S., Lysenko, I., Senior, R.A., B¨
orger, L.,
Bennett, D.J., Choimes, A., Collen, B., Day, J., De Palma, A., Díaz, S., Echeverria-
Londo˜
no, S., Edgar, M.J., Feldman, A., Garon, M., Harrison, M.L.K., Alhusseini, T.,
Ingram, D.J., Itescu, Y., Kattge, J., Kemp, V., Kirkpatrick, L., Kleyer, M., Correia, D.L.
P., Martin, C.D., Meiri, S., Novosolov, M., Pan, Y., Phillips, H.R.P., Purves, D.W.,
Robinson, A., Simpson, J., Tuck, S.L., Weiher, E., White, H.J., Ewers, R.M., Mace, G.
M., Scharlemann, J.P.W., Purvis, A., 2015. Global effects of land use on local
terrestrial biodiversity. Nature 520, 45–50. https://doi.org/10.1038/nature14324.
Ning, Jia, Liu, J., Kuang, W., Xu, X., Zhang, S., Yan, C., Li, R., Wu, S., Hu, Y., Du, G.,
Chi, W., Pan, T., Ning, J., 2018. Spatiotemporal patterns and characteristics of land-
use change in China during 2010–2015. J. Geogr. Sci. 28, 547–562. https://doi.org/
10.1007/s11442-018-1490-0.
Peng, S., Ding, Y., Liu, W., Li, Z., 2019. 1km monthly temperature and precipitation
dataset for China from 1901 to 2017. Earth Syst. Sci. Data 11, 1931–1946. https://
doi.org/10.5194/essd-11-1931-2019.
Popp, A., Krause, M., Dietrich, J.P., Lotze-Campen, H., Leimbach, M., Beringer, T.,
Bauer, N., 2012. Additional CO
2
emissions from land use change — Forest
conservation as a precondition for sustainable production of second generation
bioenergy. Ecol. Econ. 74, 64–70. https://doi.org/10.1016/j.ecolecon.2011.11.004.
Qiu, S., Peng, J., Quine, T.A., Green, S.M., Liu, H., Liu, Y., Hartley, I.P., Meersmans, J.,
2022. Unraveling trade-offs among reforestation, urbanization, and food security in
the South China Karst Region: how can a Hinterland Province Achieve SDGs? Earth’s
Future 10. https://doi.org/10.1029/2022EF002867 e2022EF002867.
Ramankutty, N., Evan, A.T., Monfreda, C., Foley, J.A., 2008. Farming the planet: 1.
Geographic distribution of global agricultural lands in the year 2000. Global
Biogeochem Cycles 22, 30–48. https://doi.org/10.1029/2007GB002952.
Ray, D.K., West, P.C., Clark, M., Gerber, J.S., Prishchepov, A.V., Chatterjee, S., 2019.
Climate change has likely already affected global food production. PLoS ONE 14,
e0217148. https://doi.org/10.1371/journal.pone.0217148.
Rosegrant, M.W., Cline, S.A., 2003. Global Food Security: challenges and Policies.
Science 302, 1917–1919. https://doi.org/10.1126/science.1092958.
Shao, Q., Liu, S., Ning, J., Liu, G., Yang, F., Zhang, X., Niu, L., 2022. Assessment of
ecological benets of key national ecological projects in China in 2000-2019 using
remote sensing. Acta Geogr. Sin. 77, 2133–2153.
Song, X., Hansen, M.C., Stehman, S.V., Potapov, P.V., Tyukavina, A., Vermote, E.F.,
Townshend, J.R., 2018. Global land change from 1982 to 2016. Nature 560,
639–643. https://doi.org/10.1038/s41586-018-0411-9.
Sun, L., Chen, J., Li, Q., Huang, D., 2020. Dramatic uneven urbanization of large cities
throughout the world in recent decades. Nat. Commun. 11, 1–9. https://doi.org/
10.1038/s41467-020-19158-1.
van Maanen, N., Andrijevic, M., Lejeune, Q., Rosa, L., Lissner, T., Schleussner, C.F., 2022.
Accounting for socioeconomic constraints in sustainable irrigation expansion
assessments. Environ. Res. Lett. 17, 075004 https://doi.org/10.1088/1748-9326/
ac77a5.
van Vliet, J., 2019. Direct and indirect loss of natural area from urban expansion. Nat.
Sustain. 2, 755–763. https://doi.org/10.1038/s41893-019-0340-0.
Wang, B., Liang, Y., Peng, S., 2022. Harnessing the indirect effect of urban expansion for
mitigating agriculture-environment trade-offs in the Loess Plateau. Land Use Policy
122, 106395. https://doi.org/10.1016/j.landusepol.2022.106395.
Wang, L., Zheng, W., Tang, L., Zhang, S., Liu, Y., Ke, X., 2021. Spatial optimization of
urban land and cropland based on land production capacity to balance cropland
protection and ecological conservation. J. Environ. Manage. 285, 112054 https://
doi.org/10.1016/j.jenvman.2021.112054.
Wu, Y., Shan, L., Guo, Z., Peng, Y., 2017. Cultivated land protection policies in China
facing 2030: dynamic balance system versus basic farmland zoning. Habitat Int 69,
126–138. https://doi.org/10.1016/j.habitatint.2017.09.002.
Xie, H., Wang, W., Zhang, X., 2018. Evolutionary game and simulation of management
strategies of fallow cultivated land: a case study in Hunan province. China. Land Use
Policy 71, 86–97. https://doi.org/10.1016/j.landusepol.2017.11.050.
Yang, Y., Xiao, P., Feng, X., Li, H., 2017. Accuracy assessment of seven global land cover
datasets over China. ISPRS J. Photogramm. Remote Sens. 125, 156–173. https://doi.
org/10.1016/j.isprsjprs.2017.01.016.
Zhang, L., Chen, X., Xin, X., 2019. Short commentary on CMIP6 Scenario Model
Intercomparison Project (ScenarioMIP). Adv. Clim. Chang. Res. 15 (519) https://doi.
org/10.12006/j.issn.1673-1719.2019.082.
Zhang, P., Chen, Y., 2017. Polycyclic aromatic hydrocarbons contamination in surface
soil of China: a review. Sci. Total Environ. 605–606, 1011–1020. https://doi.org/
10.1016/j.scitotenv.2017.06.247.
Zhao, Y., Wang, M., Hu, S., Zhang, X., Ouyang, Z., Zhang, G., Huang, B., Zhao, S., Wu, J.,
Xie, D., Zhu, B., Yu, D., Pan, X., Xu, S., Shi, X., 2018. Economics- and policy-driven
organic carbon input enhancement dominates soil organic carbon accumulation in
Chinese croplands. Proc. Natl. Acad. Sci. U.S.A. 115, 4045–4050. https://doi.org/
10.1073/pnas.1700292114.
Zhou, Y., Li, X., Liu, Y., 2021. Cultivated land protection and rational use in China. Land
use policy 106, 105454. https://doi.org/10.1016/j.landusepol.2021.105454.
Zhou, Y., Li, X., Liu, Y., 2020. Land use change and driving factors in rural China during
the period 1995-2015. Land use policy 99, 105048. https://doi.org/10.1016/j.
landusepol.2020.105048.
Zhou, Y., Zhong, Z., Cheng, G., 2023. Cultivated land loss and construction land
expansion in China: evidence from national land surveys in 1996, 2009 and 2019.
Land use policy 125, 106496. https://doi.org/10.1016/j.landusepol.2022.106496.
Y. He et al.