Study on Spatial Pattern Changes and Driving Factors of Land Use/Cover in Coastal Areas of Eastern China from 2000 to 2022: A Case Study of Jiangsu Province

Abstract

Jiangsu Province is an important economic province on the eastern coast of China, revealing the spatial–temporal characteristics, dynamic degree, and transition direction of land use/cover change, and its main driving factors are significant for the effective use of land resources and the promotion of regional human–land coordinated development. Based on land use data of Jiangsu Province from 2000 to 2020, this study investigates the spatiotemporal evolution characteristics of land use/cover using the dynamics model and the transfer matrix model, and examines the influence and interaction of the driving factors between human activities and the natural environment based on 10-factor data using Geodetector. The results showed that (1) In the past 20 years, the type of land use/cover in Jiangsu Province primarily comprises cropland, water, and impervious, with the land use/cover change mode mainly consisting of a dramatic change in cropland and impervious and relatively little change in forest, grassland, water, and barren. (2) From the perspective of the dynamic rate of land use/cover change, the single land use dynamic degree showed that impervious is the only land type whose dynamics have positively increased from 2000 to 2010 and 2010 to 2020, with values of 3.67% and 3.03%, respectively. According to the classification of comprehensive motivation, the comprehensive land use motivation in Jiangsu Province in each time period from 2000 to 2010 and 2010 to 2020 is 0.46% and 0.43%, respectively, which belongs to the extremely slow change type. (3) From the perspective of land use/cover transfer, Jiangsu Province is mainly characterized by a large area of cropland transfer (−7954.30 km²) and a large area of impervious transfer (8759.58 km²). The increase in impervious is mainly attributed to the transformation of cropland and water, accounting for 4066.07 km² and 513.73 km² from 2010 to 2020, which indicates that the non-agricultural phenomenon of cropland in Jiangsu Province, i.e., the process of transforming cropland into non-agricultural construction land, is significant. (4) From the perspective of driving factors, population density (q = 0.154) and night light brightness (q = 0.156) have always been important drivers of land use/cover change in Jiangsu Province. The interaction detection indicates that the land use/cover change is driven by both socio-economic factors and natural geographic factors. (5) In response to the dual pressures of climate change and rapid urbanization, coordinating the multiple objectives of socio-economic development, food security, and ecological protection is the fundamental path to achieving sustainable land use in Jiangsu Province and similar developed coastal areas. By revealing the characteristics and driving factors of land use/cover change in Jiangsu Province, this study provides qualitative and quantitative theoretical support for the coordinated decision-making of economic development and land use planning in Jiangsu Province, specifically contributing to sustainable land planning, climate adaptation policy-making, and the enhancement of community well-being through optimized land use.

1. Introduction

Land serves as the fundamental material basis for human societal development, playing an irreplaceable role in ensuring food security, promoting regional economic growth, maintaining social stability, and safeguarding ecological health [1,2,3]. With the rapid development of China’s socio-economy, urbanization, and industrialization, the contradiction between land supply and demand has been deepening, causing more intense competition for land resource allocation, threatening the coordinated development of regional territorial space [4,5]. Driven by both anthropogenic activities (e.g., urban expansion, agricultural intensification, and infrastructure development) and natural factors (e.g., climate change, topographic constraints, and natural disasters), the huge transformation of land use/cover change (LUCC) not only alters surface landscape patterns, but also exerts profound impacts on key ecological processes such as the carbon cycle, regional climate systems, and biodiversity conservation [6,7,8,9,10]. With the in-depth promotion of China’s sustainable development strategy, the efficient and rational utilization of land resources has emerged as a critical focus in academic research. At the same time, the “Regulation on the Implementation of the Land Administration Law of the People’s Republic of China (2021 Revision)” put forward the idea that scientific and reasonable spatial planning of the national territory needs to clarify the current situation of regional land use/cover and its transformation. Therefore, identifying land use/cover patterns—such as land use composition, landscape fragmentation and aggregation, and the directional flows of land use conversion—and conducting spatiotemporal dynamic monitoring are of paramount importance for understanding human–land interactions and promoting the sustainable and efficient use of land resources. For instance, the sustainable restoration of degraded coastal wetlands transforms underutilized land into a functional asset that provides crucial services such as carbon sequestration and storm surge buffering.

Since 1995, the International Geosphere–Biosphere Programme (IGBP) and the Human Dimensions Programme on Global Environmental Change (IHDP) have cooperated to put forward research on “land use and land cover change (LUCC)”, and this pioneering program has established the study of land use/cover change as a central and interdisciplinary field within global environmental change research [11,12]. Substantial research on land use has been performed in China and the rest of the world, investigating spatial–temporal patterns [13], dynamic changes [14], and driving factors [15]. Relevant studies show that the factors that trigger changes in the spatial and temporal characteristics of land use/cover are mainly divided into socio-economic factors (e.g., GDP, population density) and natural factors (e.g., temperature, precipitation, DEM, soil type) [16,17]. Therefore, it is important to determine the driving factors that affect land use/cover change and their influence factors to fully understand the change pattern of land use/cover in the past, clarify the characteristics of the current state of land use/cover, and predict the trend in the future, to formulate effective decision-making on the sustainable use of land.

At present, the research is also no longer limited to the construction of land use/cover change patterns, and the quantitative analysis of land dynamic changes and driving factors has also become an issue of concern for scholars at home and abroad [18,19,20]. For the study of land use/cover change, scholars have extracted land category, area, shape, and patch spatial arrangement information by identifying features such as the color, texture, and patch size of long time series remote sensing images, obtaining timescale land use data, and establishing a dynamic database of land use/cover change to analyze the changes in its spatiotemporal pattern [21,22]. They used the single land use dynamic degree and comprehensive land use dynamic degree to have quantitative analysis, and employed the transfer matrix model to identify changes in the attitude [23,24]. These methods are particularly suited for the land use/cover studies in vulnerable coastal areas, because they can not only accurately delineate the evolution of key land classes, but also effectively reflect the aggregate pressure exerted by human activities, particularly urbanization, on the coastal ecological environment. Meanwhile, with the development of remote sensing, geographic information, and other technologies, multi-source high-precision data provide more technical support for land use/cover change research [25,26], such as 1 km resolution global MODIS land use data [27], 300 m resolution global ESA land use data [28], and global land 30 m from the China Bureau of Surveying and Mapping [29]. By combining multi-source data—including remote sensing images (e.g., Landsat, Sentinel-2), socio-economic statistics (e.g., GDP, population density), point-of-interest (POI) data, and environmental factors (e.g., elevation, slope)—and intelligent models such as the Patch-generating Land Use Simulation (PLUS) model, Cellular Automata (CA)-Markov chain, and Random Forest (RF) algorithms, it is possible to achieve accurate prediction of land use/cover change trends [30,31], thereby supporting differentiated and refined land spatial management policies and promoting the harmony between socio-economic development and ecological protection. Scholars at home and abroad mainly conduct comprehensive research, and the study area mainly focuses on typical areas, hotspots areas, and ecologically fragile zones [7,32,33,34]. In addition, the research methods mainly include three categories: qualitative analysis, statistical analysis, and modeling methods [18,35,36]. Due to the spatial heterogeneity of land use/cover change, the existing conventional statistical analysis methods—such as ordinary least squares (OLS) regression, principal component analysis (PCA), and correlation analysis—can only quantitatively explain the global relationships of its driving factors. Compared with the traditional statistical analysis methods, Geodetector can better detect the spatial location relationship between factors and land use/cover change, and at the same time, effectively reveal the impact of driving factor interactions on its spatial heterogeneity [35,37]. In addition, Geodetector has better applicability in long-term scale land use/cover change scenarios [37,38]. In view of this, this paper chooses the Geodetector method to explore the driving factors of land use change in the study area.

Since the reform and opening up, China’s eastern coastal region, with its advantageous conditions such as coastal location, resource endowment, and innovation level, has experienced rapid economic development. China’s eastern coastal region includes eight eastern coastal provinces, two cities, and one region (encompassing Liaoning, Hebei, Shandong, Jiangsu, Zhejiang, Fujian, Guangdong, Hainan, Tianjin, Shanghai, and Guangxi Zhuang Autonomous Region). By 2022, this region generated about 70% of the country’s total GDP [39]. However, with the accelerated population growth and urbanization process in the eastern coastal region, the land use structure and spatial–temporal distribution patterns in the region have changed dramatically. Intense competition for land resources in coastal areas has led to increasingly sharp conflicts between urban space, agricultural production space, and ecological protection space. Key ecological spaces, especially coastal wetlands, which have important buffering and purification functions, are at risk of significant degradation of their ecological services due to encroachment by urban and industrial land use [40]. Meanwhile, in the context of global climate change, extreme events such as sea level rise and frequent storm surges have put unprecedented pressure on the existing coastal disaster prevention and mitigation engineering system, directly threatening the ecological security and sustainable development of coastal cities and towns [41]. Therefore, an in-depth investigation of the patterns and driving mechanisms of land use/cover change in the region has become an urgent scientific need to address the above challenges, optimize the spatial pattern of land use/cover, and enhance regional resilience. From the existing literature, there is a lack of information on the characteristics of the response of the land use of the economically developed coastal cities to the rapid urban development and its driving factors. Jiangsu Province, as one of the economically developed provinces in the eastern coastal region, has experienced rapid economic growth since 2000. Its urbanization rate has consistently exceeded the national average, resulting in a substantial increase in urban land area. This rapid development has also introduced challenges such as an imbalanced land use structure and inefficient utilization of land resources [42], making Jiangsu an exemplary case for studying land use transformation. In view of this, this study investigates long-term land use/cover changes and their drivers across Jiangsu Province. It utilizes land use data from 2000, 2010, 2020, and 2022 and aims to (1) reveal the spatiotemporal evolution and long-term characteristics of land use transitions in Jiangsu Province; (2) identify the key driving factors behind these changes through a multi-dimensional analysis encompassing natural, social, and economic dimensions using the Geodetector method; and (3) provide theoretical and practical insights for sustainable land use planning and management, both within Jiangsu and in other economically developed coastal regions of eastern China.

2. Methodology

2.1. Study Area

Jiangsu Province is located in the center of the eastern coastal center of mainland China, with latitudes of 30°45′–35°20′ N and longitudes of 116°18′–121°57′ E (Figure 1), bordering Shanghai, Zhejiang, Anhui, and Shandong provinces, which makes it one of the important centers of economy, culture, and scientific and technological innovation in China’s eastern coast. The province has a total area of 107,200 km², with a topography dominated by plains and hills. It has a subtropical monsoon climate with an average annual precipitation of about 1200 mm, and an average annual temperature of 14–23 °C. Meanwhile, Jiangsu Province has sufficient natural resources, shaped by fertile plains, developed water systems, and mild climatic conditions. It has continuously played a significant role in ensuring China’s food security as one of the most economically developed and densely populated regions, with a population density exceeding 500 inhabitants per km² [43]. Since the reform and opening up, the rapid urbanization and complex natural environment have dramatically changed land covers, resulting in land use problems, including structural imbalance and inefficient utilization [44].

2.2. Research Data and Methods

2.2.1. Data Sources

The data information of this study includes land use data and driver factors data. The four periods of land use data in Jiangsu Province in 2000, 2010, 2020, and 2022 were obtained from the annual land cover dataset of 30 m in China from 1990 to 2022, published by Professors Yang Jie and Huang Xin of Wuhan University [45]. The classification system of this dataset encompasses nine land use categories: cropland, forest, shrub, grassland, water, snow/ice, barren, impervious, and wetland, where impervious includes artificial surfaces, such as roads, roofs, and squares, that prevent water infiltration.

For the driver factors data, because of the diversity and complexity of the drivers of land use/cover change, a total of 10 factors were selected from the natural, social, and economic dimensions based on the comprehensive consideration of the basic characteristics of the study area. The driver factors data sources are shown in

Table 1. Driver factors data sources.

Data Type	Data Name	Data	Data Source
Natural system	X1: Altitude	SRTMSLOPE 90 M	http://www.gscloud.cn
	X2: Elevation	SRTMDEM 90 M	http://www.gscloud.cn
	X3: Slope direction	-	Calculated based on DEM
	X4: Geomorphological type	1:1,000,000 the Geomorphological Atlas of the People’s Republic of China	http://www.geodata.cn
	X5: Temperature	Annual average temperature (MODIS LST)	https://earthengine.google.com
	X6: Precipitation	Annual average precipitation (CHIRPS)	https://www.chc.ucsb.edu
	X7: Soil type	1:4,000,000 Soil Map of China	http://soil.geodata.cn
Social system	X8: Population density	LandScan Global Population Database (2000; 2010; 2020)	https://landscan.ornl.gov
Economic system	X9: GDP	China GDP Spatial Distribution Kilometre Grid Dataset (2000; 2010; 2020)	https://www.resdc.cn/
	X10: Night light brightness	Global 2000–2022 NPP-VIIRS nighttime lighting data (2000; 2010; 2020)	http://www.geodata.cn

.

2.2.2. Land Use Dynamic Degree

The attitude of land use dynamics is a key indicator of land use/cover change, which can reveal the degree of change and evolutionary trend of different land use types within a specific time span [46,47,48]. According to the research needs and object differences, it can be further classified into single land use dynamics and comprehensive land use dynamics.

Single land use dynamic degree refers to the quantitative change of a certain land use type within a certain time span in the study area, reflecting the speed and magnitude of change of a certain land use type within the study time period [47,49]. Equation (1) is as follows:

$$\mathrm{K}={\displaystyle \frac{{\mathrm{S}}_{\mathrm{b}}-{\mathrm{S}}_{\mathrm{a}}}{{\mathrm{S}}_{\mathrm{a}}}}\times {\displaystyle \frac{1}{\mathrm{T}}}\times 100\%$$

(1)

where $\mathrm{K}$ represents the single dynamic degree of a particular land use type over the study period, $\mathrm{T}$ represents the time interval of land use/cover change, and ${\mathrm{S}}_{\mathrm{a}}$ and ${\mathrm{S}}_{\mathrm{b}}$ are the areas of a particular land use type at the beginning and end of the study period, respectively.

Comprehensive land use dynamic degree reflects the overall land use/cover changes in the region and is used to assess the degree of land use/cover change activity [48]. Meanwhile, the time–dynamic characteristics of land use were classified into four types based on the comprehensive land use dynamics degree, i.e., 21–24% (rapid change type), 13–20% (rapid change type), 4–12% (slow change type), and 0–3% (very slow change type) [50]. Equation (2) is as follows:

$$\mathrm{L}\mathrm{C}={\displaystyle \frac{\sum _{\mathrm{i}=1}^{\mathrm{n}}∆{\mathrm{L}\mathrm{U}}_{\mathrm{i}-\mathrm{j}}}{2\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{L}\mathrm{U}}_{\mathrm{i}}}}\times {\displaystyle \frac{1}{\mathrm{T}}}\times 100\%$$

(2)

where $\mathrm{L}\mathrm{C}$ represents the comprehensive land use dynamic degree, $\mathrm{T}$ represents the time interval of land use/cover change, ${\mathrm{L}\mathrm{U}}_{\mathrm{i}}$ represents the area of its land use type at the beginning of the study period, and $∆{\mathrm{L}\mathrm{U}}_{\mathrm{i}-\mathrm{j}}$ is the absolute value of the area of category $\mathrm{i}$ land use type converted to category $\mathrm{j}$ in the study period.

The land use transfer matrix, which was constructed based on the Markov chain concept, quantifies the magnitude and direction of changes between land use types over a given period and reveals the structural shifts in land area and the processes governing these transitions [48,49,51]. Equation (3) is as follows:

$${\mathrm{S}}_{\mathrm{i}\mathrm{j}}=\left\{\begin{array}{ccc}{\mathrm{s}}_{11}& \dots & {\mathrm{s}}_{1\mathrm{n}}\\ ⋮& \ddots & ⋮\\ {\mathrm{s}}_{\mathrm{n}1}& \dots & {\mathrm{s}}_{\mathrm{n}\mathrm{n}}\end{array}\right\}$$

(3)

where ${\mathrm{S}}_{\mathrm{i}\mathrm{j}}$ is the land area transfer matrix, $\mathrm{i}$ represents the initial land use type, $\mathrm{j}$ represents the final land use type, $\mathrm{n}$ is the number of land classes, and $\mathrm{S}$ represents the area transferred from LULC type $\mathrm{i}$ to type $\mathrm{j}$ during the study period.

2.2.3. Geodetector

The Geodetector model is a research tool used to analyze the intrinsic mechanism of spatial heterogeneity and reveal its complex relationship with different drivers, and consists of four levels [52], namely, factor detection, interaction detection, risk detection, and ecological detection, to build a comprehensive analytical framework [53]. In order to reveal the driving factors of spatial and temporal distribution differences of land use types in Jiangsu Province, the two levels of factor detection and interaction detection are selected for exploration.

(1) Factor detection aims to accurately quantify the influence of the independent variable (X) on the spatial heterogeneity of the dependent variable (Y), and the result is denoted by $\mathrm{q}$, and the value of $\mathrm{q}$ ranges from 0 to 1 [54,55]. When the value of $\mathrm{q}$ tends to be close to 1, it means that the influence of the independent variable on the spatial differentiation of the dependent variable is greater. Equations (4)–(6) are as follows:

$$\mathrm{q}=1-{\displaystyle \frac{\sum _{\mathrm{h}=1}^{\mathrm{L}}{\mathrm{N}}_{\mathrm{h}}{\mathrm{\sigma }}_{\mathrm{h}}^2}{\mathrm{N}{\mathrm{\sigma }}^2}}=1-{\displaystyle \frac{\mathrm{S}\mathrm{S}\mathrm{W}}{\mathrm{S}\mathrm{S}\mathrm{T}}}$$

(4)

$$\mathrm{S}\mathrm{S}\mathrm{W}=\sum _{\mathrm{h}=1}^{\mathrm{L}}{\mathrm{N}}_{\mathrm{h}}{\mathrm{\sigma }}_{\mathrm{h}}^2$$

(5)

$$\mathrm{S}\mathrm{S}\mathrm{T}=\mathrm{N}{\mathrm{\sigma }}^2$$

(6)

where $\mathrm{q}$ represents the explanatory power of the driver on the dependent variable, $\mathrm{h}$ = 1, 2…, $\mathrm{L}$ is the classification of the independent or dependent variable, $\mathrm{S}\mathrm{S}\mathrm{W}$ is the sum of the variances in each category, and SST is the total variance within the whole study area. ${\mathrm{N}}_{\mathrm{h}}$ is the number of samples in the study area of category $\mathrm{h}$, $\mathrm{N}$ is the number of samples in the whole study area, ${\mathrm{\sigma }}_{\mathrm{h}}^2$ is the variance of the dependent variable in the study area of category $\mathrm{h}$, and ${\mathrm{\sigma }}^2$ is the variance of the dependent variable for the whole study area.

(2) Interaction detection is used to reveal the interaction effect among different drivers, i.e., the combined effect on the dependent variable when different drivers act jointly [55]. To assess the direction of the effect of two-factor interaction on the dependent variable, $\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\cap {\mathrm{X}}_{\mathrm{j}}\right)$ is introduced as a measure, which is compared with the $\mathrm{q}$ values (i.e., $\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right)$ and $\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right)$ under the effect of a single factor, ${\mathrm{X}}_{\mathrm{i}}$ and ${\mathrm{X}}_{\mathrm{j}}$, to derive the effect of the interaction among different driving factors on the dependent variable. The types of independent variable interactions are shown in

Table 2. Types of interactions and basis of judgment.

Interaction Type	Basis of Judgment
Bi-factor enhancement	$\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\cap {\mathrm{X}}_{\mathrm{j}}\right)>\mathrm{m}\mathrm{a}\mathrm{x}(\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right),\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right))$
Single-factor non-linear reduction	$\mathrm{m}\mathrm{i}\mathrm{n}(\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right),\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right))<\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\cap {\mathrm{X}}_{\mathrm{j}}\right)<\mathrm{m}\mathrm{a}\mathrm{x}(\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right),\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right))$
Non-linear reduction	$\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\cap {\mathrm{X}}_{\mathrm{j}}\right)<\mathrm{m}\mathrm{i}\mathrm{n}(\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right),\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right))$
Independent	$\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\cap {\mathrm{X}}_{\mathrm{j}}\right)=\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right)+\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right)$
Non-linear enhancement	$\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\cap {\mathrm{X}}_{\mathrm{j}}\right)>\mathrm{q}\left({\mathrm{X}}_{\mathrm{i}}\right)+\mathrm{q}\left({\mathrm{X}}_{\mathrm{j}}\right)$

.

3. Results

3.1. Characterization of Spatiotemporal Land Use/Cover Changes in Jiangsu Province

Utilizing land use data from Jiangsu Province, ArcGIS 10.8 software was used to statistically analyze the three periods of land use (from 2000 to 2020), and the area covered by different land use types in the three periods was obtained (

Table 3. Area of each land use type and its change in Jiangsu Province during 2000–2020 (unit: km²).

Land Use Type	2000		2010		2020		2000–2010	2010–2020	2000–2020
	Area	Percentage	Area	Percentage	Area	Percentage	Change Area	Change Area	Change Area
Cropland	81,842.6606	76.35%	76,960.4627	71.79%	73,888.3588	68.93%	−4882.1980	−3072.1039	−7954.3018
Forest	2138.3513	1.99%	1979.7217	1.85%	1719.2756	1.60%	−158.6296	−260.4461	−419.0757
Grassland	38.3652	0.04%	39.5346	0.04%	4.3759	0.00%	1.1694	−35.1587	−33.9893
Water	12,297.5817	11.47%	13,236.1729	12.35%	11,951.7461	11.15%	938.5912	−1284.4268	−345.8356
Barren	7.5635	0.01%	2.6502	0.002%	1.1884	0.001%	−4.9133	−1.4618	−6.3751
Impervious	10,875.4768	10.15%	14,981.4570	13.98%	19,635.0552	18.32%	4105.9803	4653.5982	8759.5784

). The spatial distribution of land use in Jiangsu Province from 2000 to 2022 is shown in Figure 2. Given the limited and negligible distribution of shrubs in Jiangsu Province, this study only focuses on six primary land use types: cropland, forest, grassland, water, barren, and impervious. Meanwhile, because the change rate of each land use type in Jiangsu Province from 2020 to 2022 is less than 0.77%, the following analysis mainly focuses on the period of 2000–2020 with drastic area changes and obvious change characteristics. As shown in Figure 2 and Table 3, the land use types in Jiangsu Province primarily comprised cropland, water and impervious, with proportions of 76.35%, 11.47%, and 10.15% in 2000, and 68.93%, 11.15%, and 18.32% in 2020, respectively, which represents the aggregated area of the three land use types accounting for more than 97% of the total land area of the whole region. From the perspective of spatial distribution, the land use pattern of Jiangsu Province at all time points presents the distribution characteristics of large areas of cropland, scattered distribution of impervious surfaces in dots, and the concentration of water in the south and north of Jiangsu Province, while forest and grassland are concentrated in a small amount in the south of Jiangsu Province.

From the perspective of changes in the area of land use types (Table 3), there were significant differences in the changes of each land use type in Jiangsu Province from 2000 to 2020. In 2000, the area of cropland, forest, grassland, water, barren, and impervious land was 81,842.66 km², 2138.35 km², 38.37 km², 12,297.58 km², 7.56 km², and 10,875.48 km², respectively. In 2020, the corresponding areas of these land types were 73,888.36 km², 1719.28 km², 4.38 km², 11,951.75 km², 1.19 km², and 19,635.06 km², respectively. The data showed that the area changes of cropland, forest, grassland, water, barren, and impervious were −7954.30 km², −419.08 km², −33.99 km², −345.84 km², −6.38 km², and 8759.58 km² from 2000 to 2020, with the only increments of 8.17% in impervious, and decreases of −7.42%, −0.39%, −0.03%, −0.32%, and −8.17% in cropland, forest, grassland, water, and barren. Overall, the drastic changes were in cropland and impervious, while relatively minor changes were in forest, grassland, water, and barren. Additionally, it is worth noting that cropland exhibited a spatial pattern of increasing in the north and decreasing in the south, while the growth in impervious was concentrated in the southern and northern regions of Jiangsu province.

3.2. Characterization of the Land Use Dynamic Degree in Jiangsu Province

In terms of land use dynamics (

Table 4. Single dynamic degree and comprehensive dynamic degree of land use types in Jiangsu Province during 2000–2020 (unit: %).

Land Use Type	2000–2010		2010–2020		2000–2020
	Single Land Use Dynamic Degree	Comprehensive Land Use Dynamic Degree	Single Land Use Dynamic Degree	Comprehensive Land Use Dynamic Degree	Single Land Use Dynamic Degree	Comprehensive Land Use Dynamic Degree
Cropland	−0.60	0.47	−0.40	0.43	−0.49	0.41
Forest	−0.74		−1.32		−0.98
Grassland	0.30		−8.89		−4.43
Water	0.76		−0.97		−0.14
Barren	−6.50		−5.52		−4.21
Impervious	3.78		3.11		4.03

), cropland, forest, barren, and impervious all showed a single trend from 2000 to 2020, with −0.60%, −0.74% −6.50%, and 3.78% in 2000–2010 and −0.40%, −1.32%, −5.52%, and 3.11% in 2010–2020. For comparison, impervious is the only land use type that maintains a positive growth rate. In addition, unlike the single decreasing trend of cropland, forest, barren, and impervious, grassland and water showed an increasing and then decreasing trend, with the trend of 0.30% and −8.89%, 0.76%, and −0.97% from 2000 to 2010 and 2010 to 2020, respectively. Meanwhile, to analyze the overall change characteristics of land use in Jiangsu Province, the time–dynamic characteristics of land use were classified into four types based on the comprehensive land use dynamics degree, i.e., 21–24% (rapid change type), 13–20% (rapid change type), 4–12% (slow change type) and 0–3% (very slow change type). Specifically, the comprehensive land use dynamics degree was 0.47% and 0.43% in 2000–2010 and 2010–2020, respectively, in Jiangsu Province, which was lower than 4%, indicating that the land use change in Jiangsu Province belongs to the very slow change type.

3.3. Characterization of Land Use Type Shifts

From the view of the transfer of each land use type (

Table 5. Land use transfer matrix in Jiangsu Province during 2000–2010 (unit: km²).

Land Use Type		2010
		Cropland	Forest	Grassland	Water	Barren	Impervious
2000	Cropland	75,768.98	175.76	21.04	1810.56	0.25	4066.07	75,768.98
	Forest	305.04	1799.28	1.22	3.49	0.00	29.32	305.04
	Grassland	13.67	1.41	16.49	0.54	0.49	5.77	13.67
	Water	835.79	3.22	0.77	11,125.51	0.43	331.87	835.79
	Barren	1.26	0.00	0.02	3.43	1.46	1.38	1.26
	Impervious	35.72	0.06	0.00	292.63	0.01	10,547.06	35.72
		75,768.98	175.76	21.04	1810.56	0.25	4066.07	75,768.98

and

Table 6. Land use transfer matrix in Jiangsu Province during 2010–2020 (unit: km²).

Land Use Type		2020
		Cropland	Forest	Grassland	Water	Barren	Impervious
2010	Cropland	71,664.39	134.53	0.75	955.90	0.03	4204.86	71,664.39
	Forest	383.21	1581.48	0.79	2.34	0.00	11.89	383.21
	Grassland	13.47	2.84	2.76	0.05	0.14	20.27	13.47
	Water	1815.33	0.42	0.02	10,906.13	0.55	513.73	1815.33
	Barren	0.56	0.00	0.05	0.12	0.47	1.45	0.56
	Impervious	11.40	0.01	0.00	87.20	0.00	14,882.85	11.40
		71,664.39	134.53	0.75	955.90	0.03	4204.86	71,664.39

), the total amount of land use/cover change in Jiangsu Province between the periods 2000–2010 and 2010–2020 is relatively close (7941.22 km² vs. 8161.92 km²), but there are obvious differences in the conversion between each category. The transfer of land use is mainly characterized by the conversion of large-scale transfers out of agricultural land and large-scale transfers in from impervious, in which the cropland was primarily converted to water and impervious, with the areas of 1810.56 km² and 4066.07 km², respectively, in Jiangsu Province between 2000 and 2010. It indicates that cropland has become the main source of impervious expansion in Jiangsu Province, accounting for 91.69%. Meanwhile, the 2010–2020 period compared with 2000–2010 has a similar pattern: it is dominated by large transfers out of cropland and large transfers in from impervious, but differs in that the water contribution to the increase in impervious increases to 10.81%. It shows that the increase in impervious surfaces in Jiangsu Province is mainly attributed to the transformation of cropland and water, which indicates that with the acceleration of urban development, the demand for land resources is increasing, resulting in the expansion of impervious.

To visually analyze the transfer between different land types in Jiangsu Province from 2000 to 2020, the land transfer data were visually represented. The results show (Figure 3) that there is a significant amount of transfer of cropland, which is mainly transformed into impervious and water. Meanwhile, the large amount of cropland and water has shifted to impervious, the pattern of which not only highlights the trend of defarming cropland in Jiangsu Province, but also reveals that the shrinking area of ecological land use is largely due to the continuous expansion of impervious.

3.4. Analysis of the Driving Factors of Land Use/Cover Change in Jiangsu Province, 2000–2020

3.4.1. Factor Detection Analysis

A multitude of studies have demonstrated that nighttime lighting data can offer insights into the production and living conditions of human society. Conversely, Gross Domestic Product (GDP) provides insights into the level of local economic development. Similarly, climate and topography macroscopically determine the conditions of regional land resources and the spatial pattern of land use. Taking into account the accessibility and representativeness of the data, ten indicators were selected as independent variables. These include altitude (X₁), elevation (X₂), slope direction (X₃), geomorphological type (X₄), temperature (X₅), rainfall (X₆), soil type (X₇), population density (X₈), GDP (X₉), and night light brightness (X₁₀). The intensity of land use/cover change in Jiangsu Province was analyzed as the dependent variable.

The q-value in the Geodetector Factor Detection Module results provides information on the magnitude of influence of each driving factor. A large q-value indicates that the factor has a greater influence on the dependent variable. Factor detection can reveal the key elements affecting land use/cover change in Jiangsu Province and their relative influence. The detection results show (

Table 7. Detection results of land use/cover change drivers in Jiangsu Province.

Driving Factors	2000		2010		2020
	q	Rank	q	Rank	q	Rank
X3: Slope direction	0.001	10	0.002	10	0.002	10
X1: Altitude	0.008	6	0.009	8	0.009	9
X9: GDP	0.029	3	0.071	3	0.093	3
X6: Rainfall	0.004	9	0.006	9	0.010	7
X2: Elevation	0.020	4	0.019	6	0.015	6
X5: Temperature	0.007	8	0.022	4	0.023	4
X7: Soil type	0.008	7	0.021	5	0.023	5
X4: Geomorphological type	0.011	5	0.012	7	0.009	8
X10: Night light brightness	0.072	1	0.131	2	0.156	1
X8: Population density	0.069	2	0.131	1	0.154	2

) that there are differences in the influence of the driving factors on land use/cover change in Jiangsu Province in different periods.

For 2000, from largest to lowest, the q-values of the driving forces for the period were X₁₀ > X₈ > X₉ > X₂ > X₄ > X₁ > X₇ > X₅ > X₆ > X₃ in order, and the q-value of all the factors is lower than 0.1, among which X₁₀ (nighttime light luminance) and X₈ (population density) have a main influence on land use/cover change, and their q-values are 0.072 and 0.069, respectively. In 2010, the ranking of the explanatory power of each factor on land use/cover change in Jiangsu Province changed slightly, but X₈ (population density), X₁₀ (nighttime light brightness), and X₉ (GDP) remain solidly in the top rankings in 2010, but the role of X₇ (soil type) and X₅ (temperature) in influencing land use change has significantly increased, while the influence of X₁ (elevation), X₂ (slope), and X₄ (landform type) has diminished compared to the year 2000. Further analysis of the driving factors in Jiangsu Province in 2020 reveals that while the role of social and economic factors in driving land use/cover change remains stable, the status of natural factors such as X₆ (precipitation) is steadily increasing, which indicates that land use/cover change in Jiangsu Province has changed into a complex situation dominated by a variety of factors at this stage.

3.4.2. Interactive Detection Analysis

Interactive detection examines how two driving factors work together (Figure 4). It compares the explanatory power (q-value) of a factor pair against that of each single factor, revealing whether their combined effect strengthens or weakens the interpretation of the dependent variable. The analysis of interactions reveals the differences in the effects of the factors on changes in land use when they act together versus when they act alone. The results of interactive detection analysis show that the two-factor enhancement or non-linear enhancement is the main feature of the results in Jiangsu Province, and this indicates that land use/cover change is driven by more than one factor, and land use/cover change is the result of the interaction of various factors.

Specifically, the top-five driver interactions with high influence in 2000 were in the following order: X₁₀∩X₂ (0.0867) > X₈∩X₂ (0.0865) > X₁₀∩X₁ (0.086) > X₁₀∩X₇ (0.0854) > X₁₀∩X₄ (0.0843), and the top-five driver interactions change to X₁₀∩X₉ (0.1769) > X₈∩X₉ (0.1759) > X₁₀∩X₅ (0.1633) > X₁₀∩X₇ (0.1632) > X₈∩X₂ (0.1625) in 2020, of which socio-economic factors such as nighttime light brightness (X₁₀), population density (X₈), and GDP (X₉) are the main driving factors for land use/cover change in Jiangsu Province. With Jiangsu Province’s social progress and economic development, its synergistic influence gradually increases. By comprehensively analyzing the interaction between climate, topography, soil environment, and social factors and the change in q value of the single-factor contribution rate, it can be seen that the explanatory power of the interaction between factors on the land use/cover change has been enhanced to different degrees compared with that of the single-factor role, and the future evolution of the land use in Jiangsu Province will be presented as a multifactor synergistic influence.

4. Discussion

4.1. Geodetector Model Reveals the Drivers of Land Use/Cover Change

This study systematically investigates land use changes in Jiangsu Province by analyzing variations in the area of different land use types, their transfer characteristics, and the overall intensity of change. To explore the drivers, ten indicators—including altitude, elevation, slope direction, geomorphological type, temperature, rainfall, soil type, population density, GDP, and night lights—were selected as independent variables, with the intensity of land use/cover change serving as the dependent variable. The Geodetector method was employed to address the limitation of using the change amount or proportion of a single land use type as the independent variable, which often fails to adequately reveal the intrinsic mechanisms of regional land use/cover change. Through single-factor and interaction detection analyses, this study elucidates the driving mechanisms of land use/cover change, overcoming the shortcomings of conventional methods in explaining the interactive effects of multiple factors. These influencing factors, encompassing both natural and anthropogenic dimensions, lead to modifications in land use structure and pattern. Considering the accessibility and representativeness of the data and referring to the existing studies, ten indicators were selected for analysis. However, that other relevant factors—such as additional soil properties, distance to water bodies, proximity to roads, and policy impacts—were not included due to data constraints [56,57]. Further research is needed to develop methods for selecting more representative factors and identifying the dominant drivers of land use/cover change. The geographical detector is suitable for use in case studies where the independent variable is a categorical quantity and the dependent variable is a continuous variable, whereby the study needs to categorize the continuous type of independent variables [58]. However, the method exhibits some limitations: it cannot explain whether the effect of each factor on land use/cover change is positive or negative [53]. Therefore, this paper combines the land use transfer matrix and the spatiotemporal pattern evolution map to aid judgment when interpreting the results. In future studies, attempts can be made to combine Geodetector with statistical models that can determine the direction of influence, in order to construct a more comprehensive framework for analyzing the driving mechanisms.

4.2. The Drivers and Impacts of the Process of Non-Agriculturalization in Jiangsu Province

The non-agriculturalization of cropland refers to the process of transforming cropland into non-agricultural construction land, which is, in essence, a process of land resource allocation triggered by the competition between the two land use types of arable land and construction land [59]. The study reveals a significant process of “non-agriculturalization” in Jiangsu Province from 2000 to 2020, characterized by a systematic shrinkage of cropland and a substantial expansion of impervious. Specifically, the land use/cover changes in Jiangsu Province were characterized by a net decrease of 7954.30 km² of cropland and a net increase of 8759.58 km² of impervious from 2000 to 2020. Meanwhile, the land use transfer matrix further indicated that cropland was the primary source for impervious expansion. Particularly during 2000–2010, a substantial 4066.07 km² of cropland was converted to impervious, accounting for 91.69% of the total area transferred into impervious. Spatially, this non-agriculturalization trend aligned closely with the economic development pattern, being most pronounced in southern Jiangsu and other economically developed areas. Driving force analysis using Geodetector identified socio-economic factors as the fundamental drivers of this process. Population density and night light brightness consistently emerged as the most critical individual driving factors, and their interactions with other factors like GDP exhibited strong non-linear enhancement effects, collectively amplifying their explanatory power on land use/cover change.

It has been shown that the de-agrarianization of cropland is a phenomenon that is difficult to avoid completely in the process of regional development [4], but that a drastic transition to “de-agrarianization” can have a negative impact on the region. With 3.2% of the country’s cropland, Jiangsu Province produces 5.5% of the country’s grain, making it one of the country’s 13 main grain-producing provinces. The process of non-agriculturalization of cropland in Jiangsu Province may affect regional food production, which, in turn, poses a potential threat to food security. Furthermore, this process negatively impacts the ecological environment. Large areas of impervious replacing ecologically functional cropland and water bodies exacerbate the urban heat island effect and degrade ecosystem services values provided by cropland [60,61]. Notably, these impacts exhibit significant spatial heterogeneity within Jiangsu Province, forming a south–north gradient pattern closely aligned with regional economic development levels. Consequently, future land management policies must strive to balance “development” and “protection” to foster a virtuous cycle between the sustainable use of land resources and high-quality economic development.

4.3. Land Use/Cover Change Driven by the Dual Pressures of Climate Change and Urbanization

This study systematically analyses the characteristics of land use/cover change in Jiangsu Province from 2000 to 2020. The results indicate that socio-economic factors were the primary drivers of these changes over the two decades, which are manifested in the continuous expansion of impervious and the sharp reduction in cropland. However, against the macro-backdrop of the interplay between global climate change and regional urbanization, this transformation process exhibits more complex characteristics and challenges. Firstly, although the independent explanatory power of natural factors was relatively limited, our study found that the q-values of climatic factors such as temperature (X₅) and precipitation (X₆) showed an increasing trend from 2000 to 2020. This suggests that climate change—for instance, warming and altered precipitation patterns—may be gradually becoming a potential factor of land use/cover change by affecting the suitability of agricultural production [62,63], regional water resource distribution [64], and urban heat island effects [65]. Secondly, the land use transition is dominated by urbanization, particularly the conversion of ecological land (e.g., water bodies, forests) to impervious. This process not only intensifies regional habitat fragmentation, but may also undermine the ecosystem’s resilience to climate change [66]. For instance, the reduction in water bodies diminishes the region’s capacity for flood storage and drought resistance, thereby creating a negative feedback loop [67].

In response to the dual pressures of climate change and rapid urbanization, more practical land governance strategies are needed. Future territorial spatial planning should not only focus on controlling the scale of urban expansion, but also pay greater attention to its spatial form and ecological quality. For example, green infrastructure can be planned and rehabilitated within and on the fringes of cities [68], which can mitigate the ecological risks of urbanization while increasing the resilience of cities to extreme climate events. At the same time, greater emphasis must be placed on regional climate change risk assessments, particularly in ecologically sensitive areas, to facilitate more rational spatial planning [69]. In conclusion, coordinating the multiple objectives of socio-economic development, food security, and ecological protection is the fundamental path to achieving sustainable land use in Jiangsu Province and similar developed coastal areas.

5. Conclusions

Taking Jiangsu Province as a typical study area, utilizing land use data for the period 2000 to 2020, the characteristics of land use/cover changes in Jiangsu Province were systematically analyzed by employing methods that included changes in dynamics and transfer characteristics. At the same time, the driving factors affecting land use/cover change in Jiangsu Province were revealed by using Geodetector, leading to the following conclusions:

(1)

Land use structure and change trends: Cropland, impervious, and water are the dominant land use types in Jiangsu Province, collectively accounting for over 97% of the total area during 2000–2020. During the study period, the changes of cropland and impervious in Jiangsu Province are drastic, with the changes amounting to −8846.29 km² and 10,019.58 km², reflecting a strong trend of non-agricultural land conversion driven by urbanization and economic growth. The changes in forest, grassland, water, and barren are relatively small, with the changes amounting to only −478.76 km², −39.27 km², −647.54 km², and −7.72 km². Only the impervious presents a drastic expansion of the situation, and the change is mainly concentrated in the south of Jiangsu Province, showing a significant phenomenon of outward expansion of urban areas, while the forest and grassland shrinkage is mainly concentrated in the western region.

(2)

Land use dynamics: The single land use dynamic degree showed that impervious is the only land type whose dynamics degree is positive from 2000 to 2010 and 2010 to 2020, with values of 3.67% and 3.03%, respectively, whereas the dynamic degree of cropland, forest, and barren are negative from 2000 to 2010 and from 2010 to 2020, with values of −0.59% and −0.39%, −0.75% and −1.35%, and −6.50% and −5.67%, with barren showing the most drastic changes in comparison. Unlike the single trend of change in cropland, forest, and barren, grassland and watershed showed an increasing and then decreasing trend of change, with 0.29% and −9.00%, and 0.66% and −1.03% of motivation in the period 2000–2010 and 2010–2020, respectively. According to the classification of comprehensive motivation, the comprehensive land use motivation in Jiangsu Province in each time period from 2000 to 2010 and 2010 to 2020 is 0.46% and 0.43%, respectively, which belongs to the extremely slow change type.

(3)

Land use conversion pathways: The land use transfer matrix and Sankey diagram revealed that the study area is most active in the transfer of cropland and impervious from 2000 to 2020. The increase in impervious was primarily sourced from cropland and water. From 2000 to 2010, cropland contributed 91.69% of the newly added impervious area, highlighting a significant non-agricultural transition and a continuous encroachment on ecological land. This transformation pattern highlights the trend of farmland de-agriculturalization in Jiangsu Province, and also reveals that the shrinkage of ecological land area is mainly due to the continuous expansion of impervious surfaces.

(4)

Driving factors: Population density (q = 0.154) and night light brightness (q = 0.156) have always been important drivers of land use/cover change in Jiangsu Province. Meanwhile, the q-values of climatic factors such as temperature (X₅) and precipitation (X₆) showed an increasing trend from 2000 to 2020, which suggests that climate change may be gradually becoming a potential factor of land use/cover change under the macro-backdrop of global climate change and regional urbanization. In addition, the interaction detection indicates that the land use/cover change in Jiangsu Province is driven by both socio-economic factors and natural geographic factors, but in general, the former has a stronger explanation of the land use/cover change, and the interaction among the factors shows two-factor enhancement or a non-linear enhancement effect, and the two-factor interaction has a stronger explanation for the land use/cover change than the single-factor explanation.

(5)

Coping strategies: In response to the dual pressures of climate change and rapid urbanization, coordinating the multiple objectives of socio-economic development, food security, and ecological protection is the fundamental path to achieving sustainable land use in Jiangsu Province and similar developed coastal areas.