Regional-scale land use change based on multi-scenario simulation and its impact on carbon storage: a case study of southern Jiangsu region

Research on regional land-use changes and carbon storage is vital given climate change and human activities. First, the spatiotemporal characteristics of land use change and carbon storage in southern Jiangsu from 2000 to 2020 were analyzed. Using the MCCA model, future land use in southern Jiangsu for the year 2030 was simulated under various development scenarios, and the associated carbon storage was estimated. The following conclusions were drawn: (1) From 2000 to 2020, the land-use structure in the southern part of Jiangsu Province had a certain degree of stability but also underwent minor changes. Over the past 20 years, the area of cultivated land has decreased, whereas the areas of urban, rural, industrial, mining, and residential land have increased dramatically. (2) From 2000 to 2020, carbon storage in southern Jiangsu showed a trend of first decreasing and then increasing. The areas with relatively high carbon storage were primarily distributed in strips in the western part of southern Jiangsu and in the northern, central, and southern parts. Areas with relatively low carbon storage were mainly distributed in the northern part of southern Jiangsu Province, spreading outwards from multiple centers in a planar pattern. (3) The change in carbon storage of land use types in the southern Jiangsu region from 2000 to 2020 was positively correlated with changes in land use type area. (4) In 2030, the changes in land use structure and spatial distribution of land types under multiple scenarios in the southern Jiangsu region will be relatively stable. Cultivated land showed an increasing trend in the cultivated land protection scenario and a decreasing trend in the other scenarios. The research results can provide a theoretical basis for decision-makers to optimize land-use structures and rationally allocate land resources for territorial space planning in southern Jiangsu.

1 Introduction

With its expansive territory, substantial population, and rapid economic growth, China has emerged as the world's leading carbon emitter (Li et al., 2024). It is highly urgent to reduce and control carbon emissions and strengthen ecological civilization (Chen, 2015). Against the backdrop of global climate challenges and domestic demands for sustainable development, the Chinese government first explicitly set forth the national goal of peaking carbon dioxide emissions by 2030 and achieving carbon neutrality by 2060 in September 2020. Land is a fundamental natural resource of vital importance for human survival and wellbeing. With the continuous growth of the population and the constant advancement of science and technology, the impact of human activities on land resources is increasing (Winkler et al., 2021). Human activities directly change the structure and mode of land use, thereby affecting the structure and service functions of terrestrial ecosystems (de Andrés et al., 2017). During the rapid process of urbanization, ecological land is gradually eroded to meet the needs of urban development and construction and to accommodate the expansion of the urban scale. Excessive concentration and a large population size can also lead to the deterioration of the ecological environment of urban areas (Zhu et al., 2023). Unreasonable land use reduces the carbon sink capacity of terrestrial ecosystems (Dale, 1997).

The southern part of Jiangsu Province is located in the Yangtze River Delta region along the southeastern coast of China. It is adjacent to the international metropolis of Shanghai to the east, Anhui Province to the west, Zhejiang Province to the south, and the Yangtze River to the north. In 2013, the southern part of Jiangsu Province was designated a national demonstration zone for modernization. Additionally, the southern part of Jiangsu has established China's largest “Environmental protection model urban agglomeration” and “ecological urban agglomeration” and Ecological Urban Agglomeration. The southern part of Jiangsu Province boasts strong economic strength and a high degree of modernization. Its social development ranks among the highest in the country. It shoulders multiple development responsibilities, including promoting high-quality economic growth and constructing an ecological civilization in Jiangsu. Its strategic position in economic development and ecological protection is significant. In the past development process, the continuous increase in demand for land resources, caused by the concentration of the population in towns and cities, as well as the need for economic development, has also impacted land carbon storage. Accurate evaluation of regional carbon storage is crucial for understanding carbon cycles and formulating climate change mitigation strategies (Fu et al., 2010; Grafius et al., 2018; Hasanah and Wu, 2024; Kumar et al., 2023; Lai et al., 2016). Multiple methods have been developed to assess carbon storage at regional scales, leveraging field measurements, remote sensing, and modeling approaches. Traditional methods include field inventory-based approaches, such as sample plot surveys, which are accurate for small areas but difficult to scale regionally without extensive data. The soil type and vegetation type methods are also used, but their accuracy is limited by spatial heterogeneity and data availability. Remote sensing technology significantly enhances the estimation of large-scale carbon storage. Studies have combined satellite data (e.g., Landsat and Sentinel) with models such as the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model to quantify spatiotemporal changes in carbon storage (Chang and Ko, 2014; Firozjaei et al., 2020; Gong C. et al., 2023; Liu et al., 2017; Xia et al., 2019). For example, research in Hangzhou Bay used Landsat time-series data and the InVEST model to track a decline in carbon storage from 108.15 to 82.47 Mt over 40 years, linking the reduction to urban expansion and vegetation loss. Machine learning algorithms, such as random forests, improve the classification accuracy of land use and land cover changes, further refining carbon storage estimates (Gong W. et al., 2023).

Advanced statistical and simulation models, such as geographically weighted regression (GWR) and sequential Gaussian co-simulation (SGCS), account for spatial non-stationarity and improve estimation precision. Some studies have integrated ecosystem process models with multi-source remote sensing data to enhance mechanistic understanding and prediction under different climate scenarios.

Despite this progress, challenges remain, including scale discrepancies, data uncertainty, and model parameterization. Future efforts should focus on combining multi-source data, adopting higher-resolution remote sensing, and developing integrated models to reduce the uncertainties and support regional carbon management policies (Xiang et al., 2022). Some scholars have employed the carbon storage module in the InVEST model to analyze the changing characteristics of carbon storage under various land-use conditions (Liu et al., 2019; Seto et al., 2012; Tang et al., 2020; Zhang et al., 2015). They also utilized the PLUS and InVEST model carbon storage modules to reveal the relationship between the spatiotemporal variation characteristics of carbon storage and land use changes and made predictions on carbon storage (Li et al., 2023). However, relevant research on the southern Jiangsu Province is relatively scarce.

Therefore, researching land use changes and their impact on carbon storage in the southern Jiangsu region is conducive to understanding the changes in carbon storage of the region's terrestrial ecosystems in the past and future, providing reasonable suggestions for sustainable land use, giving full play to the leading role of the southern Jiangsu region in high-quality economic development, ecological civilization construction, and green and low-carbon urban development, and contributing to the realization of regional dual-carbon goals. It also establishes a theoretical analysis framework with practical significance for enriching research on the correlation between land-use change and carbon sequestration effects, as well as for carbon storage research at the urban agglomeration scale.

2 Materials and methods

2.1 Research area

Southern Jiangsu is located in the heart of the Yangtze River Delta in Eastern China. It comprises five cities: Nanjing, Zhenjiang, Suzhou, Wuxi, and Changzhou, covering a total area of approximately 28,095 km², which accounts for 27.17% of Jiangsu Province's total area. This area is situated in the subtropical humid monsoon climate zone, characterized by cold winters, hot summers, concurrent rainfall, and high temperatures. The average summer and winter temperatures are 30.8 and 5.1 °C, respectively. The annual precipitation is over 1,000 mm. Compared with regions at the same latitude, the interannual variation in precipitation is slight, with an annual precipitation variation rate of 12–24%.

The southern part of Jiangsu Province has a dense population and a high level of urbanization. It is a significant transportation hub and economic center on the eastern coast of China. In recent years, the industrial structure in southern Jiangsu Province has undergone continuous optimization, leading to enhanced economic competitiveness and sustainable development capacity. In 2022, the regional Gross Domestic Product (GDP) of southern Jiangsu amounted to 7028.411 billion yuan, representing 57.2% of the total regional GDP of Jiangsu Province. This figure reflects a 4.3% increase compared to that of the previous year. The added value mainly came from the secondary and tertiary industries. The per capita disposable income of urban permanent resident households was 74,413 yuan. The southern part of Jiangsu Province has continuously established innovation centers, enterprise technology centers, and other innovation carriers to support the development of high-end manufacturing, promote the deep integration of industry and technology, and accelerate the transformation to an innovation-driven development model. Meanwhile, southern Jiangsu is actively developing the digital economy, promoting the application of industrial Internet and the Internet of Things, and exploring the path of digital transformation. A geographical map of the southern Jiangsu region is shown in Figure 1.

Figure 1

Figure 1. Study area location map.

2.2 Data source

The data required for this study and their sources are listed in Table 1. Specifically, it includes land use data for Jiangsu Province in 2000, 2005, 2010, 2015, and 2020; natural data such as elevation, slope, slope direction, annual average temperature, annual average precipitation, and rivers; social and economic data such as gross domestic product and population density; and location and transportation data such as major roads and railways. Based on the first-level classification system of land use status of the Chinese Academy of Sciences, the second-level land categories in the land use data were reclassified into six first-level land categories: cultivated, forest, grassland, water, urban and rural industrial and mining residential, and unused land. The land use data space was then standardized, and the geographical coordinates were uniformly set to GCS_WGS_1984 with a spatial resolution of 30 × 30 m. Euclidean distance analysis was conducted on the main road and railway data to obtain the raster data required for the research.

Table 1

Table 1. Data source.

2.3 Research methods

2.3.1 Land use transfer matrix

The land-use transfer matrix is an application of the Markov model in land-use change. It sorts the transfer areas of various land-use types into a matrix, reflecting the direction and quantity of transformation from one land-use type to another within a specified period (Ariken et al., 2020; Bilintoh et al., 2024). Its expression is as follows.

$\begin{array}{ll}{S}_{ij}=\left[\begin{array}{c}{S}_{11}\&\cdots \&S_{1n}\\ ⋮\&\ddots \&⋮\\ S_{n1}\&\cdots \&S_{nn}\\ \end{array}\right]& (1)\end{array}$

In this context, S denotes the area, while i and j signify the initial and final land use types, respectively. The S_ij represents the number of land-use types. In this study, land-use distribution maps of any two stages were utilized. With the aid of ArcGIS 10.2 software, spatial superposition operations were performed to obtain the transfer matrix of land use types in the southern Jiangsu region from 2000 to 2020, thereby facilitating a more detailed analysis of the change process in the land-use structure of the southern Jiangsu region.

2.3.2 MCCA model

The MCCA model can simulate the continuous changes of multiple land-use components, overcoming the limitation of discrete cellular states in traditional CA models (Liang et al., 2021). This is a new method for advancing the simulation of CA models from the cellular scale to the subcellular scale. Compared to traditional cellular automata models, the MCCA model distinguishes itself by its ability to account for the coexistence of multiple land use types within each cell. It estimates changes in the proportions of land use types through competition among the land use components within each cell, operating at a sub-cellular scale. This enables the simulation of land use changes with enhanced precision in mixed land use units. This model comprises three components: land use conversion rules and demand forecasting, simulating changes in land use spatial structure, and verifying model accuracy (Gao et al., 2023; Zhao et al., 2022).

(1) Mining of land use conversion rules and demand forecasting

First, based on the starting and target years, the relationship between the change in the area proportion of each land use component within the cell and the driving factors was trained using the random forest regression (RFR) algorithm to explore the transition rule of land use change. Subsequently, the development potential of each land-use type was predicted, and a probability map of land-use change for the target year was generated. Among them, the RFR effectively overcomes the correlation problem of collinearity among multiple spatial variables. Its expression is as follows:

$\begin{array}{ll}{R}_k=RFR<uscore>train(Y_{i,k}^s,F_i^s)& (2)\end{array}$

$\begin{array}{ll}{P}_{i,k}=RFR<uscore>predict(R_k,F_i)& (3)\end{array}$

Where i represents a mixed cell, k represents the type of land use, s represents the sample threshold, and R represents the relationship between the driving factors and the changes in the proportion of class coverage in hybrid cells. RFR train represents the training process of RFR; Y represents the sample dataset of mixed cells; $F_i^s$ represents the sample datasets of each driving factor; P_{i, k} represents the development potential of each land use component; RFR predict represents the RFR prediction process; and F_i represents the dataset of driving factors.

(2) Simulation of changes in land use spatial structure

Based on the input each type of land within the mixed cell, land development probability, and field scope, the coverage ratio of each land-use type in the mixed cell unit was determined using the roulette competition mechanism. When the simulated land use ratio reaches the target value, the spatial distribution of each land use type is generated. Its expression is as follows:

$\begin{array}{ll}{O}_{i,k}^t=P_{i,k}\times Q_{i,k}^t\times D_k^t& (4)\end{array}$

Among them, $O_{i,k}^t$ represents the overall change probability of the coverage ratio of the k land use type at cell unit i at the t iteration; $O_{i,k}^t$represents the domain effect of the k land use type at cell unit i at the t iteration; $D_k^t$represents the demand feedback for the k type of land use at the t iteration.

(3) Model accuracy verification

The MCCA uses the sub-pixel confusion matrix (SCM) statistical tool and the mixed cell form index (mcFoM) to test the accuracy of the MCCA model in simulating changes.

The subpixel confusion matrix (SCM) can directly calculate the overall accuracy of the MCCA model, and its expression is as follows:

$\begin{array}{ll}OA={\displaystyle \sum _{k=1}^n}{P}_k& (5)\end{array}$

Where OA represents the total accuracy, P_k represents the diagonal consistency of the subpixel confusion matrix, and n represents the number of land cover categories within the mixed cell.

The Mixed-cell Figure of Merit (mcFoM) is a new index proposed for mixed cells based on the traditional Figure of Merit (FoM) index, which is applicable for verifying the accuracy of the MCCA model. Its expression is as follows.

$\begin{array}{ll}mcFOM=\frac{B}{A+B+C+D}& (6)\end{array}$

A indicates that the error caused by the change in land cover ratio was underestimated, and B represents the part where the simulated change direction is correct and consistent with the actual change in land cover ratio. C represents the error caused by the discrepancy between the direction of the change in the land cover ratio and the actual direction of change. D indicates that the error caused by changes in the land cover ratio was overestimated.

This study employed MCCA V2.0 and selected nine driving factors: elevation, slope, aspect, annual average temperature, annual average precipitation, population density, GDP, distance to the highway, and distance to the railway (Figure 2). The MCCA model integrates the influences of nature, social economy, and location transportation on land use to ensure the accuracy of simulating future land use changes in the southern Jiangsu region. Taking 2010 as the starting year, the transformation area between land use types from 2010 to 2020 was taken as the transition matrix and input into the MCCA model to obtain the simulation results of the spatial distribution of land use in southern Jiangsu in 2020. After verification, the model's calculation results showed that the Kappa coefficient was 0.81 and the mcFoM index was 0.19. This suggests that the model has a relatively high level of simulation accuracy and can be used in subsequent simulation studies.

Figure 2

Figure 2. Simulation driving factors of land use in southern Jiangsu Region by 2030.

2.3.3 InVEST model

Stanford University jointly developed the InVEST model in collaboration with the World Wildlife Fund and Nature Conservancy. It includes modules such as soil conservation, water purification, and carbon sequestration, and is primarily used to assess ecosystem services. The carbon storage module in the model calculates the carbon storage of each land type and simultaneously determines the total carbon storage of the terrestrial ecosystem within the study area as follows: The core data required for the operation of the model carbon storage module include land-use status and ecosystem carbon pool data. Among them, the carbon density data of the ecosystem carbon pool include aboveground, underground, soil, and dead organic carbon density. The formula for calculating carbon storage is as follows (Liu et al., 2019):

$\begin{array}{ll}{C}_i=C_{i-above}+C_{i-below}+C_{i-soil}+C_{i-dead}& (7)\end{array}$

$\begin{array}{ll}{C}_{i-total}=C_i\times A_i& (8)\end{array}$

Among them, C_i represents the total carbon density of Earth class i, C_i-above represents the surface carbon density of Earth type i, C_i-below indicates the underground carbon density of land type i, and C_i-soil represents the carbon density of soil organic matter in land type i. C_i-dead indicates the carbon density of dead organic matter in Earth class I, C_i-total represents the total carbon storage of land class I, and A_i represents the area of class I.

Based on the actual situation in the southern Jiangsu region, this study calculated the carbon storage distribution and total carbon storage data for the region in each period using the carbon storage module in InVEST 3.14.1 software. These data were analyzed in conjunction with land-use status data for the same period.

2.3.4 Multi-scenario simulation

Scenario simulation helps compare future environments and the potential consequences of different policies may bring about. Based on the resource and environmental carrying capacity of the southern Jiangsu region and in accordance with existing data and relevant literature, this study proposes three scenarios for land use in the southern Jiangsu region in 2030: natural development, cultivated land protection, and ecological protection. The specific scenario design was as follows:

(1) Natural development scenario

The natural development scenario refers to a simulation method that allows land use to develop without constraints based on the current trend of land use changes without considering any policy adjustments or their impact. This study states that in this scenario, the probability of land-use transfer remains unchanged, the rules of land-use conversion are not altered, and the impact of policies is not considered.

(2) Scenarios of cultivated land protection

To implement the national policy on farmland protection and adhere to the red line of farmland protection, this study established scenarios for farmland protection. In this context, the transfer rate of cultivated land to other land types should be slowed, and the expansion speed of construction land should be suppressed to protect cultivated land. Scenario setting for farmland protection: The probability of farmland being converted to construction land due to natural development was reduced by 60%.

(3) Ecological protection scenarios

The “14th Five-Year Plan for Ecological and Environmental Protection of Jiangsu Province” highlights that the “14th Five-Year Plan period” is a pivotal time for promoting the comprehensive green transformation of economic and social development, aiming to achieve qualitative improvement in ecological and environmental protection beyond quantitative gains. To ensure the construction of an ecological civilization in southern Jiangsu and promote high-quality economic development in this region, this study established ecological protection scenarios, focusing on protecting forests, grasslands, and cultivated land. This scenario was set as follows: based on natural development, the probability of transferring cultivated land to construction land was reduced by 30%. Except for construction land, all other land types can be converted into forests and grasslands. The probability of transferring forest and grassland to construction land was reduced by 50%.

3 Results

3.1 Spatio-temporal analysis of land use change

The changes in land use quantity in the southern Jiangsu region from 2000 to 2020 are shown in Figure 3. Overall, the proportion of cultivated land in the southern part of Jiangsu Province has consistently been the highest, accounting for approximately half of the total regional area, and is primarily distributed in the western part of the southern region of Jiangsu Province. However, during the study period, the area of cultivated land decreased from 16467.26 km² in 2000 to 12701.16 km² in 2020. Over the past 20 years, the area of cultivated land decreased by 3766.10 km², representing a 13.40% decrease. The water area is primarily located in the eastern part of southern Jiangsu Province and comprises Taihu Lake and the lower reaches of the Yangtze River. From 2000 to 2005, the water area was the second-largest terrestrial type in southern Jiangsu Province, and since 2010, it has become the third-largest terrestrial type in the region. Overall, over the 20 years from 2000 to 2020, the water area increased from 5544.87 to 5787.71 km², a rise of 242.83 km², representing a 4.37% increase in proportion. Specifically, from 2000 to 2010, the water area increased from 5544.87 to 5748.99 km² and then to 5980.47 km².

Figure 3

Figure 3. Land use structure in southern Jiangsu from 2000 to 2020.

The area increased period by period, but the growth rate decreased. Since 2010, the water area has decreased annually, and its proportion has continued to decline. Construction land is primarily distributed in the northern part of southern Jiangsu Province, mainly along waterways. From 2005 to 2005, construction land was the third major land category in southern Jiangsu Province, China. After 2010, its proportion gradually increased, and it became the second major land category in southern Jiangsu Province. During the research period, the urban and construction land increased. Overall, from 2000 to 2020, the construction land area increased from 3919.11 to 7408.52 km², representing a growth of 3489.41 km², with a growth rate of nearly 90%, and a corresponding increase in proportion of 12.42%. Forest land has always been the fourth major category of land, with a proportion consistently around 7%, and is mainly distributed in the northwestern and southwestern border areas of southern Jiangsu. From 2000 to 2020, the area of forest land decreased annually, and the proportion continued to decline. Over the 20 years, the area decreased by 78.55 km², and the proportion decreased by 0.28%. The changes in the above-mentioned land types reflect the fact that economic development in the southern part of Jiangsu Province from 2000 to 2020 increased the demand for construction land. The expansion of construction land inevitably encroaches on agricultural and ecological areas. During the study period, the proportion of grassland and unused land remained consistently low, and the combined area of the two did not exceed 1% of the total land area of the region. The increase in unused land by more than 500% from 2005 to 2010 was due to severe salinization in the Taihu Lake Basin and the expansion of the saline-alkali land area.

As shown in Table 2, from 2000 to 2005, cultivated land was transferred out the most, accounting for 92.4% of the total transferred area, followed by water areas. Unused land was transferred out the least and had a relatively low degree of mobility. From the perspective of transfer, construction land was transferred the most, accounting for 71.5% of the total transferred area, whereas forest land was transferred the least. During the 5 years from 2000 to 2005, the area of cultivated land transferred out for construction land was the largest, indicating that the demand for construction land in urban economic and social development increased, and urban construction land has continuously expanded. In this process, other types of land, mainly cultivated land, were occupied.

Table 2

Table 2. Land use transfer matrix (km²) in southern Jiangsu region from 2000 to 2005.

As shown in Table 3, from 2005 to 2010, the transfer-out of cultivated land was still the largest, covering an area of 2356.07 km², accounting for 83.9%, whereas the transfer-out of unused land was the smallest. From the perspective of transfer-in, the transfer-in of urban and rural industrial, mining, and residential land is the highest, followed by water areas. Additionally, it is worth noting that the transfer-in area of unused land is 67.83 km², representing an increase of 67.79 km² from the previous stage. Compared with other research stages, the area of land-use type transfer from 2005 to 2010 was the highest among all research stages, indicating that land type conversion was frequent and the degree of mobility was high during this stage.

Table 3

Table 3. Land use transfer matrix (km²) in southern Jiangsu region from 2005 to 2010.

As shown in Table 4, from 2010 to 2015, from the perspective of transfer-out, although the transfer-out of cultivated land remained the largest, the transfer-out area decreased significantly compared to the previous two stages. The transfer-out of urban and rural industrial, mining, and residential land remained second at 90.32 km², while the transfer-out of grassland was the smallest. From the perspective of transfer-in, the transfer-in area of construction land was the highest, reaching 345.90 km². Additionally, the transfer-in area of cultivated land exceeded that of the water area at this stage and was ranked second. The transfer-in area of unused land decreased from 67.83 km² in the previous stage to 2.13 km², marking the lowest level to date. The area of land-use transfer from 2010 to 2015 was the lowest among all the research stages, indicating a relatively low degree of land-use mobility in the southern Jiangsu region during this period.

Table 4

Table 4. Land use transfer matrix (km²) in southern Jiangsu region from 2010–2015.

As shown in Table 5, from 2015 to 2020, from the perspective of transfer-out, the area of transfer-out was cultivated land > urban and construction land > water area. During this period, the difference between the transfer-out of urban and rural industrial, mining, and residential land and water areas was minimal. From the perspective of transfer-in, the area of construction land transferred in remains the highest, followed by cultivated land, with the least amount of unused land.

Table 5

Table 5. Land use transfer matrix (km²) in southern Jiangsu region from 2015 to 2020.

As shown in Table 6, from 2000 to 2020, urbanization in southern Jiangsu continued to advance, with significant changes occurring in various regions. The inflow of urban and construction land, water areas, grasslands, and unused land was greater than the outflow. In contrast, the inflow of cultivated and forested land was less than the outflow. Cultivated and forested land are the primary sources of the increase in the areas of other land types. In the future development process, it is necessary to strictly implement the national cultivated land protection policy, control the conversion of cultivated land to other land-use types, and improve the sustainable utilization capacity of land.

Table 6

Table 6. Land use transfer matrix (km²) in southern Jiangsu region from 2000 to 2020.

3.2 Spatio-temporal variation analysis of carbon storage

This study combines the existing research results on the urban scale in the southern Jiangsu region (Liu et al., 2019), uses the carbon density data of land types in Suzhou City (Seto et al., 2012) as the basis, and refers to scholars' research on carbon density under different climate zone land use types in the Yangtze River Economic Belt (Tang et al., 2020). The carbon density data corresponding to different land types in southern Jiangsu were calculated, as shown in Table 7, where 1Mg = 1t:

Table 7

Table 7. Carbon density (Mg/hm²) of various land-use types in southern Jiangsu.

Based on land use and carbon density data for different land use types, the carbon storage of the southern Jiangsu region from 2000 to 2020 was calculated using the InVEST carbon storage model, as shown in Figure 4. Subsequently, the spatial distribution pattern of carbon storage in southern Jiangsu from 2000 to 2020 was determined using ArcGIS software. As shown in Figure 4, the total carbon storage in the southern Jiangsu region was 2851.27 × 10⁵ Mg, 2825.79 × 10⁵ Mg, 2734.61 × 10⁵ Mg, 2754.81 × 10⁵ Mg, and 2742.93 × 10⁵ Mg in 2000, 2005, 2010, 2015, and 2020, respectively. Overall, carbon storage in the southern part of Jiangsu Province showed a decreasing trend from 2000 to 2020. In particular, from 2005 to 2010, the decline in total carbon storage was the most pronounced, with the most significant reduction observed. However, from 2010 to 2015, the total carbon storage in southern Jiangsu Province increased. The reasons for such changes are as follows: From 2000 to 2020, the social economy of the southern Jiangsu region continued to develop, and the urbanization process accelerated, with the expansion of urban construction land occupying ecological land, resulting in reduced carbon storage. In particular, the urbanization process accelerated from 2005 to 2010. However, the urbanization process slowed thereafter. Coupled with the government's practical measures to strengthen ecological protection, this resulted in a slight increase in the total carbon storage.

Figure 4

Figure 4. Spatial distribution pattern of carbon storage in southern Jiangsu from 2000 to 2020.

Areas with relatively high carbon storage are primarily distributed in the western part of southern Jiangsu Province, with a distribution pattern extending from north to south, forming a strip. Additionally, they are rarely found in the Taihu Lake Basin in Southeast China. Areas with relatively low carbon storage were mainly distributed in the northern part of southern Jiangsu Province, spreading outwards from multiple centers in a planar pattern. From a temporal perspective, during the 20 years from 2000 to 2020, there was a continuous expansion trend in regions with lower carbon reserves, particularly from 2005 to 2010, when the expansion trend was most evident. From the perspective of land type, areas with higher carbon storage have a higher degree of overlap with forests and grasslands. In contrast, areas with lower carbon storage have a higher degree of overlap with urban, rural, industrial, mining, and residential lands. Moreover, the expansion range of these areas is consistent with the expansion range of urban, rural, industrial, mining, and residential lands.

3.3 Analysis of the impact of land use change on carbon storage

The carbon storage of each land-use type in the southern Jiangsu region from 2000 to 2020 was calculated using the InVEST model, as shown in Table 8. As shown in the table, the carbon storage in the southern Jiangsu region from 2000 to 2020 was as follows: cultivated land > water area > construction land > forest land > grassland > unused land. Among them, cultivated land showed a decreasing trend period by period. The carbon storage of cultivated land in 2000 and 2020 was 1724.34 × 10⁵ Mg and 1329.98 × 10⁵ Mg, respectively, representing a 22.8% decrease. The changing trend of carbon storage in urban and rural industrial, mining, and residential land is opposite to that of cultivated land, showing an increasing trend in each period. The carbon storage was 301.28 × 10⁵ Mg in 2000 and 569.53 × 10⁵ Mg in 2020, representing an increase of 89.03%. The carbon storage in grasslands first decreased and then increased. The carbon storage of grassland was 18.00 × 10⁵ Mg in 2000, decreased to 16.52 × 10⁵ Mg in 2015, and increased to 23.72 × 10⁵ Mg again in 2020. Carbon storage in water areas first increased and then decreased. The carbon storage of forest land exhibited a trend of decreasing, increasing, and then decreasing again. It decreased from 2000 to 2010 and then began increasing. By 2015, the carbon storage on forest land had increased to 327.89 × 10⁵ Mg, and by 2020, it had decreased to 324.40 × 10⁵ Mg. The unused carbon reserves increased significantly from 2005 to 2010, rising from 1.00 × 10⁵ Mg to 6.03 × 10⁵ Mg. The change in carbon storage of land use types in the southern Jiangsu region from 2000 to 2020 was positively correlated with the change in the area of land use types.

Table 8

Table 8. Carbon storage of various land-use types in the southern Jiangsu region from 2000 to 2020 (10⁵ Mg).

A spatial distribution map of carbon storage transfer from 2000 to 2020 was calculated using the InVEST model. The changes in carbon storage caused by the transfer of various types in the southern Jiangsu region from 2000 to 2020 were calculated using the Spatial Analyst—Regional Analysis—table-display zoning statistics function in ArcGIS 10.2 software, as shown in Table 9.

Table 9

Table 9. Carbon storage changes caused by the transfer of various types in the southern Jiangsu region from 2000 to 2020 (10⁵ Mg).

As shown in Table 9, from 2000 to 2020, when the land use type in the southern Jiangsu region changed from other land types to cultivated land, forest land, or grassland, the carbon storage of the terrestrial ecosystem showed an increasing trend. The increase in carbon storage caused by the conversion of construction land to cultivated land was the greatest, followed by that caused by the conversion to forestland. When land types are converted from other types to water areas, construction land, or unused land, the carbon storage of terrestrial ecosystems shows a decreasing trend. The reduction in carbon storage caused by the conversion of cultivated land to construction land was the greatest, reaching 93.84 × 10⁵ Mg. This suggests that changes in construction land, cultivated land, and forest land in the southern Jiangsu region significantly affect the total carbon storage of the local terrestrial ecosystem. In subsequent development, efforts should be made to enhance the protection of cultivated and ecological land, rationally allocate various types of land use, and strengthen local ecological civilization construction.

3.4 Multi-scenario simulation and carbon storage prediction

3.4.1 Simulation of land use change under multiple scenarios

The transfer matrix was established according to the scenario requirements. The land use status of southern Jiangsu in 2030, under scenarios of future natural development, cultivated land protection, and ecological protection, was simulated using the MCCA model (Figure 5). According to the simulation results, the land use structure and spatial distribution in the southern Jiangsu region in 2030 are relatively stable compared to 2020. The proportion of land types in each scenario was as follows: cultivated land > urban and construction land > water area > forest land > grassland > unused land. There were few differences in the spatial distribution of land types in each scenario.

Figure 5

Figure 5. Spatial distribution of land use in multiple scenarios in southern Jiangsu in 2030.

Under the natural development scenario, the area of change in urban and rural industrial, mining, and residential land is the largest, followed by cultivated land. This indicates that if no restrictions are imposed on land use during the development process, it will lead to the continuous expansion of urban and rural industrial, mining, and residential land, which may pose a threat to cultivated land and is not conducive to the protection of cultivated land and food security. The area of the changed grassland was only 50.43 km². Compared to other land types, the number of changes was relatively small, but the change range reached 24.39%, indicating a notable degree of change in the area. Although the areas affected by urban, rural, industrial, and mining residents, as well as cultivated land, are large, the change ranges are relatively small, at 9.76 and 4.16%, respectively. The degree of change was not as good as that of grasslands. Compared with them, the amount of unused land changed was also relatively small, at 2.14 km², but its change range was only 3.09%, which was the least obvious among all types of land.

Compared with the natural development scenario, under the cultivated land protection scenario, the change range of grassland was 24.52%, and the degree of change was still the most obvious. The changing trend of urban and rural industrial, mining, and residential land remains unchanged, but the change range has decreased to 1.6%, whereas the cultivated land area has increased. This indicates that, under this scenario, the expansion of construction land can be suppressed to a certain extent by imposing restrictions. To protect the cultivated land.

Under the ecological protection scenario, the areas of cultivated land, water, and unused land decreased. In contrast, the areas of forestland, grassland, and construction land increased. This trend is similar to that of the natural development scenario. In all three scenarios, forest and grassland areas increased. However, compared with the other two scenarios, the area in this scenario increased the most, with the highest growth rate and the most obvious degree of change. The increase in construction land in this scenario was also less than that in the natural scenario, indicating that in the ecological protection scenario, the ecological environment and economic development were effectively coordinated.

3.4.2 Prediction of changes in carbon storage under different development scenarios

Based on the prediction of land use demand in the southern Jiangsu region under multiple scenarios in 2030 using the MCCA model, the carbon storage of various regions and the total carbon storage in the southern Jiangsu region under multiple scenarios in 2030 were calculated, as shown in Table 10.

Table 10

Table 10. Carbon storage in the southern Jiangsu region under multiple scenarios in 2020 and 2030 (10⁵ Mg).

As shown in Table 10, compared with 2020, the total carbon storage in southern Jiangsu in 2030 was the highest under the farmland protection scenario and the lowest under the natural development scenario. The total carbon storage values in all three scenarios showed an increasing trend. The carbon storage of each region is as follows: cultivated land > urban and rural industrial, mining, and residential land > water area > forest land > grassland > unused land. However, these trends vary from country to country. The carbon storage of cultivated land decreased only under natural development scenarios. In contrast, the carbon storage of urban and rural industrial, mining, and residential land decreased under both the cultivated land protection and natural development scenarios. Meanwhile, the carbon storage of forestland and grassland showed an increasing trend under all three scenarios. Carbon storage in water areas and unused land showed a decreasing trend in all three scenarios.

4 Discussion

This study analyzed the spatiotemporal evolution of land use and its impact on carbon storage in southern Jiangsu from 2000 to 2020, projecting future trends under multiple scenarios for 2030. The observed decline in carbon storage, driven primarily by rapid urbanization, aligns with patterns documented in other economically dynamic regions of China, such as Guangdong and the Pearl River Delta. This consistency underscores a pervasive challenge across rapidly developing areas: the trade-off between urban expansion and ecosystem service depletion, particularly carbon sequestration.

The spatial patterns of carbon storage—with higher values concentrated in the western strip and lower values expanding from northern nuclei—reflect the region's underlying socioeconomic and topographic gradients. The most significant carbon loss occurred between 2005 and 2010, coinciding with a period of accelerated industrial and infrastructural development. The subsequent slight recovery in total carbon storage may be attributed to increased environmental awareness and policy interventions, such as the conversion of some built-up land to green spaces or the implementation of afforestation projects, though these forces were not sufficient to offset the overall declining trend.

Our multi-scenario simulations for 2030 reinforce the critical role of policy in shaping future outcomes. The cultivated land protection scenario yielded the highest carbon storage, followed by the ecological protection scenario, with natural development resulting in the lowest value. This hierarchy highlights that proactive land-use planning, especially the conservation of agricultural and forest lands, is a potent strategy for mitigating carbon loss. The finding that conversion of cultivated land to construction land caused the largest carbon depletion, while reversion of construction land to cultivated land led to the greatest gain, powerfully illustrates the carbon consequences of land conversion decisions. However, several limitations must be acknowledged to contextualize the findings. First, the parameter uncertainty, particularly regarding carbon density values which were not directly measured but drawn from literature, may affect the absolute estimates and cross-regional comparability. Future work should prioritize localized carbon density sampling or employ sensitivity analysis to quantify this uncertainty. Second, while the study compared natural development, cultivated land protection, and ecological protection scenarios, it did not include a dedicated economic development scenario, which limits the understanding of potential carbon storage outcomes under unabated growth pressures. Incorporating such a scenario in future research would provide a more comprehensive basis for policy deliberation. Finally, the discussion would benefit from deeper integration with the socioeconomic drivers (e.g., GDP growth, population migration, industrial policy) behind the observed land-use transitions to more fully explain the spatiotemporal patterns.

A notable source of uncertainty in this carbon storage assessment stems from the carbon density parameters. Ideally, these critical values should be derived from localized empirical measurements within the study area. However, due to data constraints, this study, like many others, relied on proxies from similar regions. This approach may introduce biases, as carbon densities can vary significantly with local soil, vegetation, and management practices. Future research should prioritize field-based measurements to obtain accurate, location-specific carbon density data, which would enhance the reliability of the estimates and strengthen the validity of spatial comparisons and policy inferences.

5 Conclusions

Based on the analysis of land-use change and carbon storage dynamics in southern Jiangsu from 2000 to 2020, and projections for 2030, the main conclusions of this study are as follows:

(1) Between 2000 and 2020, the land-use structure in southern Jiangsu was dominated by cultivated land, water areas, and construction land. Construction land expanded rapidly, becoming the second-largest category by 2010. The spatial distribution of major land types remained relatively stable despite these changes.

(2) Regional carbon storage exhibited a pattern of initial decrease followed by a slight increase, with the most pronounced decline occurring from 2005 to 2010. Spatially, higher carbon storage persisted in the western regions, while low-carbon-storage areas, originating from multiple nuclei in the north, continued to expand.

(3) Changes in carbon storage were directly linked to land-use conversions. The conversion of cultivated land to construction land resulted in the most significant carbon loss, whereas the reversion of construction land to cultivated land contributed the largest carbon gain.

(4) Simulations for 2030 under three scenarios (natural development, cultivated land protection, and ecological protection) show relatively stable future land-use patterns. The ranking of land-type proportions remains consistent across scenarios: cultivated land > construction land > water area > forest land > grassland > unused land.

(5) Total carbon storage in 2030 is projected to be highest under the cultivated land protection scenario and lowest under the natural development scenario. This underscores the importance of targeted land-use policies in enhancing regional carbon sequestration capacity.

This study provides a reference for territorial spatial planning and ecological-economic coordinated development in southern Jiangsu. Future research should focus on refining carbon density parameters, incorporating economic development scenarios, and conducting cross-regional comparative analyses to improve the generalizability and policy relevance of the findings.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Author contributions

WZ: Conceptualization, Data curation, Formal analysis, Resources, Writing – original draft. YL: Funding acquisition, Methodology, Supervision, Writing – review & editing. CT: Methodology, Project administration, Visualization, Conceptualization, Writing – review & editing. YM: Data curation, Visualization, Writing – review & editing. YZ: Investigation, Methodology, Resources, Software, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Qinghai Province “Kunlun Talent High-end Innovative and Entrepreneurial Talents” Project (2023).

Acknowledgments

The authors are particularly grateful to all the researchers for providing data support for this study.

Conflict of interest

YL was employed by Qinghai Century National Library Technology Service Co., Ltd.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Ariken, M., Zhang, F., Liu, K., Fang, C., and Kung, H.-T. (2020). Coupling coordination analysis of urbanization and eco-environment in Yanqi Basin based on multi-source remote sensing data. Ecol. Indic. 114:106331. doi: 10.1016/j.ecolind.2020.106331

Crossref Full Text | Google Scholar

Bilintoh, T. M., Pontius, R. G. Jr, and Liu, Z. (2024). Analyzing the losses and gains of a land ca3tegory: insights from the total operating characteristic. Land 13:1177. doi: 10.3390/land13081177

Crossref Full Text | Google Scholar

Chang, Y.-C., and Ko, T.-T. (2014). An interactive dynamic multi-objective programming model to support better land use planning. Land Use Policy 36, 13–22. doi: 10.1016/j.landusepol.2013.06.009

Crossref Full Text | Google Scholar

Chen, W. Y. (2015). The role of urban green infrastructure in offsetting carbon emissions in 35 major Chinese cities: a nationwide estimate. Cities 44, 112–120. doi: 10.1016/j.cities.2015.01.005

Crossref Full Text | Google Scholar

Dale, V. H. (1997). The relationship between land-use change and climate change. Ecol. Applic. 7, 753–769. doi: 10.1890/1051-0761(1997)007[0753:TRBLUC]2.0.CO;2

Crossref Full Text | Google Scholar

de Andrés, M., Barragán, J. M., and García Sanabria, J. (2017). Relationships between coastal urbanization and ecosystems in Spain. Cities 68, 8–17. doi: 10.1016/j.cities.2017.05.004

Crossref Full Text | Google Scholar

Firozjaei, M. K., Fathololoumi, S., Weng, Q., Kiavarz, M., and Alavipanah, S. K. (2020). Remotely sensed urban surface ecological index (RSUSEI): an analytical framework for assessing the surface ecological status in urban environments. Remote Sens. 12:2029. doi: 10.3390/rs12122029

Crossref Full Text | Google Scholar

Fu, Y., Liu, Y., and Wang, Y. (2010). Evaluation method and supporting system of low carbon cities. China Popul. Resour. Environ 20, 44–47. doi: 10.3969/j.issn.1002-2104.2010.08.008

Crossref Full Text | Google Scholar

Gao, Z.-Q., Tao, F., Wang, Y.-H., and Zhou, T. (2023). Potential ecological risk assessment of land use structure based on MCCA model: a case study in Yangtze River Delta Region, China. Ecol. Indic. 155:110931. doi: 10.1016/j.ecolind.2023.110931

Crossref Full Text | Google Scholar

Gong, C., Lyu, F., and Wang, Y. (2023). Spatiotemporal change and drivers of ecosystem quality in the Loess Plateau based on RSEI: A case study of Shanxi, China. Ecol. Indic. 155:111060. doi: 10.1016/j.ecolind.2023.111060

Crossref Full Text | Google Scholar

Gong, W., Duan, X., Sun, Y., Zhang, Y., Ji, P., Tong, X., et al. (2023). Multi-scenario simulation of land use/cover change and carbon storage assessment in Hainan coastal zone from perspective of free trade port construction. J. Clean. Prod. 385:135630. doi: 10.1016/j.jclepro.2022.135630

Crossref Full Text | Google Scholar

Grafius, D. R., Corstanje, R., and Harris, J. A. (2018). Linking ecosystem services, urban form and green space configuration using multivariate landscape metric analysis. Landsc. Ecol. 33, 557–573. doi: 10.1007/s10980-018-0618-z

PubMed Abstract | Crossref Full Text | Google Scholar

Hasanah, A., and Wu, J. (2024). Exploring dynamics relationship between carbon emissions and eco-environmental quality in Samarinda Metropolitan Area: a spatiotemporal approach. Sci. Total Environ. 927:172188. doi: 10.1016/j.scitotenv.2024.172188

PubMed Abstract | Crossref Full Text | Google Scholar

Kumar, P., Singh, A. B., Arora, T., Singh, S., and Singh, R. (2023). Critical review on emerging health effects associated with the indoor air quality and its sustainable management. Sci. Total Environ. 872:162163. doi: 10.1016/j.scitotenv.2023.162163

PubMed Abstract | Crossref Full Text | Google Scholar

Lai, L., Huang, X., Yang, H., Chuai, X., Zhang, M., Zhong, T., et al. (2016). Carbon emissions from land-use change and management in China between 1990 and 2010. Sci. Adv. 2:e1601063. doi: 10.1126/sciadv.1601063

PubMed Abstract | Crossref Full Text | Google Scholar

Li, L., Awada, T., Zhang, Y., and Paustian, K. (2024). Global land use change and its impact on greenhouse gas emissions. Glob. Chang. Biol. 30:e17604. doi: 10.1111/gcb.17604

PubMed Abstract | Crossref Full Text | Google Scholar

Li, P., Chen, J., Li, Y., and Wu, W. (2023). Using the InVEST-PLUS model to predict and analyze the pattern of ecosystem carbon storage in Liaoning Province, China. Remote Sens. 15:4050. doi: 10.3390/rs15164050

Crossref Full Text | Google Scholar

Liang, X., Guan, Q., Clarke, K. C., Chen, G., Guo, S., Yao, Y., et al. (2021). Mixed-cell cellular automata: a new approach for simulating the spatio -temporal dynamics of mixed land use structures. Landsc. Urban Plan. 205:103960. doi: 10.1016/j.landurbplan.2020.103960

Crossref Full Text | Google Scholar

Liu, J., Zhang, G., Zhuang, Z., Cheng, Q., Gao, Y., Chen, T., et al. (2017). A new perspective for urban development boundary delineation based on SLEUTH-InVEST model. Habitat Int. 70, 13–23. doi: 10.1016/j.habitatint.2017.09.009

Crossref Full Text | Google Scholar

Liu, X., Wang, S., Wu, P., Feng, K., Hubacek, K., Li, X., et al. (2019). Impacts of urban expansion on terrestrial carbon storage in China. Environ. Sci. Technol. 53, 6834–6844. doi: 10.1021/acs.est.9b00103

PubMed Abstract | Crossref Full Text | Google Scholar

Seto, K. C., Güneralp, B., and Hutyra, L. R. (2012). Global forecasts of urban expansion to 2030 and direct impacts on biodiversity and carbon pools. Proc. Nat. Acad. Sci. U.S.A. 109, 16083–16088. doi: 10.1073/pnas.1211658109

PubMed Abstract | Crossref Full Text | Google Scholar

Tang, L., Ke, X., Zhou, T., Zheng, W., and Wang, L. (2020). Impacts of cropland expansion on carbon storage: a case study in Hubei, China. J. Environ. Manage. 265:110515. doi: 10.1016/j.jenvman.2020.110515

PubMed Abstract | Crossref Full Text | Google Scholar

Winkler, K., Fuchs, R., Rounsevell, M., and Herold, M. (2021). Global land use changes are four times greater than previously estimated. Nat. Commun. 12:2501. doi: 10.1038/s41467-021-22702-2

PubMed Abstract | Crossref Full Text | Google Scholar

Xia, C., Li, Y., Xu, T., Chen, Q., Ye, Y., Shi, Z., et al. (2019). Analyzing spatial patterns of urban carbon metabolism and its response to change of urban size: a case of the Yangtze River Delta, China. Ecol. Indic. 104, 615–625. doi: 10.1016/j.ecolind.2019.05.031

Crossref Full Text | Google Scholar

Xiang, S., Wang, Y., Deng, H., Yang, C., Wang, Z., Gao, M., et al. (2022). Response and multi-scenario prediction of carbon storage to land use/cover change in the main urban area of Chongqing, China. Ecol. Indic. 142:109205. doi: 10.1016/j.ecolind.2022.109205

Crossref Full Text | Google Scholar

Zhang, M., Huang, X., Chuai, X., Yang, H., Lai, L., Tan, J., et al. (2015). Impact of land use type conversion on carbon storage in terrestrial ecosystems of China: a spatial-temporal perspective. Sci. Rep. 5, 1–13. doi: 10.1038/srep10233

PubMed Abstract | Crossref Full Text | Google Scholar

Zhao, Z., Fan, B., Zhou, Q., and Xu, S. (2022). Simulating the coupling of rural settlement expansion and population growth in Deqing, Zhejiang Province, based on MCCA modeling. Land 11:1975. doi: 10.3390/land11111975

Crossref Full Text | Google Scholar

Zhu, K., Cheng, Y., Zhou, Q., Kápolnai, Z., and Dávid, L. D. (2023). The contributions of climate and land use/cover changes to water yield services considering geographic scale. Heliyon 9:e20115. doi: 10.1016/j.heliyon.2023.e20115

PubMed Abstract | Crossref Full Text | Google Scholar