Evaluating the effects of urbanization on land use and landscape patterns in southern China (1990–2020)

Abstract

Evaluating the effects of urbanization on land use and landscape patterns is crucial for protecting resources and promoting environmental sustainability. This study examined urbanization in southern China from 1990 to 2020, focusing on urban area growth, GDP, population density, and the Landscape Expansion Index (LEI). It also assessed the impact on land use and landscape patterns through land use structures and metrics. The results reveal that Urban built-up areas have consistently increased over the past three decades, especially between 2010 and 2020. The extent of urban land increased from 20,758 km² in 1990 to 42,939 km² in 2020, reflecting an average annual growth rate of 3.56%, largely driven by the conversion of cultivated land (74.5%) and forest land (18.4%) compared to other land cover types. This expansion coincided with increased population density and GDP. The LEI also indicates a transition from compact urban growth (1990–2010) to a more dispersed pattern (2010–2020). Landscape metrics indicate decreased dominance of a single land-use type, leading to a more balanced structure and greater fragmentation. This emphasizes the need to tackle urbanization’s environmental challenges and the importance of sustainable development in urban planning for harmonious coexistence.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-26368-4.

Introduction

Urbanization has experienced substantial growth on a global scale in recent decades. In 2018, the global urban land area reached 7.97 × 10⁵ km², 1.5 times larger than in 1990¹. From 1985 to 2015, the urban land area increased by an average of 9.7 × 10³ km² per year². Urban land is projected to continue to expand, and by 2030, the global urban land area is estimated to be three times larger than in 2000³. In 2018, China and the United States (US) were the leading countries for impervious area/urban land area expansion, collectively representing around 50% of the global total¹. Specifically, China’s built-up areas exceeded those of the US in 2015¹.

Research demonstrated that geophysical, social, and institutional factors play a significant role in land allocation for urban development and urbanization^4–6. These land allocations involve changes in the physical and biological characteristics of the landscape, such as shifts in vegetation cover, urban area expansion, and intensified agricultural activities^7,8.

Urban expansion has transformed land use and landscapes, posing major challenges to biodiversity and human well-being⁹. It threatens human health and well-being, hinders progress toward Sustainable Development Goals (SDGs)¹⁰, and disrupts ecosystem services by damaging the integrity of nearby urban ecology and changing the microclimate of urban built-up areas^11,12. Urban expansion predominantly encroaches upon agriculture, depleting cropland resources that worsen urban food security¹³. Moreover, forest cover loss is linked to urban population growth, resulting in decreased forest productivity and an intensified heat island effect in cities⁹. This abrupt land use and landscape change may affect worldwide environmental issues, including carbon dioxide emissions and deforestation, drawing attention from international scientific studies and global affairs¹⁴. Moreover, changes in landscape patterns due to urbanization profoundly impact ecological systems, often resulting in habitat loss, fragmentation, and biodiversity decline¹⁵.

Therefore, a more comprehensive understanding and more precise evaluation of urbanization on land use and landscape changes is crucial for informing policy decisions regarding land use regulation, protecting natural resources, and promoting environmental sustainability^7,8.

This study was designed to conduct a comprehensive investigation of the impact of urbanization on land use and landscape change in the southern regions of China, which can help formulate effective land-use policy and land-use management and provide insights and support for sustainable development. This study region is one of China’s five ecological security strategic patterns¹⁶. It is also the world’s largest and most intact subtropical forest ecosystem and an important ecological security barrier in southern China. Therefore, conducting a detailed study on urbanization-driven land use and landscape changes in this region is critical for understanding ecological risks, guiding sustainable urban planning, and ensuring the long-term preservation of its unique ecological functions.

Initially, we measured the degree of urbanization during the past three decades using economic, demographic, and urban morphology indicators from 1990 to 2020. Population and Gross Domestic Product (GDP) have been identified as essential and influential determinants for urbanization measures. For instance, from 1995 to 2020, the China population grew from 29.04% to 63.89%, contributing to a significant increase in the trend of urban lands and landscape changes^9,17,18. Furthermore, urban morphology and urban area expansion also examined the level of urbanization¹⁹. For example, the Landscape Expansion Index (LEI) is commonly used to reflect urban patch components and configuration patterns²⁰, which are important in identifying the level of urban land expansion.

Subsequently, we examined the impact of urbanization on land use and landscape patterns by utilizing land use structure and landscape indices. Changes in landscape patterns driven by urbanization were assessed using Shannon’s entropy²¹and key landscape metrics, including edge density and aggregation²². These metrics are widely employed to diagnose and quantify spatial variations in the composition and configuration of landscapes. Additionally, they facilitate analyses of fragmentation, diversity, and shifts in landscape patterns, offering insights into the structural transformations induced by urbanization.

Materials and methods

Study area descriptions

The study area is situated in the southern regions of China, encompassing an estimated area of 982,110 km². It spans parts of the Zhejiang, Fujian, Jiangxi, Hunan, Guangxi, Shanghai, Anhui, Hubei, and Guangdong provinces (Fig. 1). Shanghai, Fujian, and Guangdong are among the largest coastal cities in eastern China. They are the main economic, financial, trade, and shipping centers and are home to some of China’s most important industrial centers. The climate belongs to the subtropical monsoon climate, with distinct seasons, abundant rainfall, flourishing vegetation, and high biodiversity in the ecosystem. The influence of topography and climate results in significant regional differences in agricultural production and vegetation distribution²³. Forestry is the primary land cover type in hilly and mountainous areas, while cultivated land and cash crops dominate the valley plain areas.

Fig. 1.

Location of the study regions in Southern China: Map created using QGIS 3.32 (QGIS Development Team, 2023; https://qgis.org)²⁴.

Data sources

Data on land use and land cover (LULC) changes for 1990, 2000, 2010, and 2020 at a 1 km resolution were obtained from the Resource and Environmental Science Data Center of the Chinese Academy of Sciences (https://www.resdc.cn/DOI). A 1 km resolution Gross domestic product (GDP) data for 2000, 2010, and 2019, as well as population data for 1990, 2000, 2010, and 2020, were also sourced from https://www.resdc.cn/DOI/DOI.aspx?DOIID=33/32²⁵.

A 1 km resolution of LULC data was used to ensure consistent spatial resolutions for socioeconomic indicators (GDP and population density). The land-use types comprised six primary categories: cultivated land, forest land, grassland, water area, impervious surfaces, and unused land. The LULC data were primarily derived from Landsat TM/ETM remote sensing images, and subsequently, the quality was enhanced through manual and visual analysis of Landsat 8 remote sensing images²⁶.

Urbanization measures and indicators

Urbanization is a significant human factor that directly and indirectly influences land use change, leading to alterations in landscape patterns²⁷. Population density, GDP, urban area expansion, and landscape expansion index are used to assess the degree of urbanization (Fig. 2).

Fig. 2.

General framework for the analysis of urbanization and its effect on land use-landscape patterns: Note: LPI = Largest Patch Index, ED = edge density, LSI = landscape shape index, AI = aggregation index and Eq.= equation.

GDP and population density

The GDP data for each year was obtained by extracting information within the defined research area boundaries. The collected data were then pooled and evaluated in two distinct time intervals: 2000–2010 and 2010–2019. The data is 1 km grid data; each grid represents the total GDP output value within the grid range (1 km²), and the unit is 10,000 yuan/km² (Table 2, Fig.S1).

Table 2.

GDP and population density per square Km in the Southern regions of China.

Population density per square km					GDP in 10,000 yuan per square km
Year	Min	Max.	Average	Ten years difference	Year		Min.	Max.	Average	Ten years difference
1990	0	40832.77	263.64	28.38	2000		7.89	51389.14	266.04	459.245
2000	0	94112.49	292.02	26.93	2010		17.1	24595.15	725.285	1800.318
2010	0	114862.65	318.95	29.06	2019		95	1,731,925	2525.6	2259.563
2020	0	160439.85	348.01	84.37

Similarly, 1 km grid China’s population data, including the periods in 1990, 2000, 2010, and 2020, was extracted according to the specified region of interest, and the geographical distribution of population density was evaluated (Table 2, Fig.S1). Each grid represents the number of people within the grid range (1 km²).

Urban area expansion

Urbanization necessitates significant land allocation for industrial development and to accommodate the growing population. The spatial pattern can be assessed by examining the ratio of the urban area to the total land area, which provides insights into the rate and magnitude of urban expansion²⁸. The annual urban growth rate (AGR) was also calculated to verify the dynamic changes in individual land use/land cover categories²⁸ (Eq. 1).

1

where AGR is the annual urban area growth rate for an individual category (%), U_a and U_b are the urban areas at the beginning and end of a specific period, and T represents the total number of years spent in the study.

Urban area expansion modes and types

The study of landscape patterns is strongly linked to the urban development process, and it plays a crucial role in monitoring and evaluating the ecological impact of urbanization^29,30. Landscape indices in analyzing land use dynamics and urban growth help to provide information about the spatial heterogeneity of land use and the morphological characteristics of urban areas²⁰. Urban expansion modes can be calculated using the Landscape Expansion Index (LEI) and categorized as edge expansion, infilling growth, or outlaying growth²⁰ (Fig. 2). The LEI equation is expressed as follows (Eq. 2):

2

A₀ is the buffer zone between the new and old urban land, and A_γ is the intersection between the buffer zone and the non-urban lands. The land cover data collection has a spatial resolution of 1 km; we initially transformed all the data into vector format and then implemented a buffer of 500 m. The buffer distance should be equal to or smaller than the spatial resolution of remote sensing data²⁰. The values of LEI range from 0 to 100. The LEI = 0, outlaying growth: 0 < LEI ≤ 50 = edge-expansion growth and 50 < LEI ≤ 100 = infilling growth.

Moreover, the interest of this study was on the entire landscape; the mean expansion index (MEI) and an area-weighted mean expansion index (AWMEI) were further calculated as indicated in Eqs. 3 and 4, respectively²⁰. The values of MEI and AWMEI are the compact and diffusion natures of urban growth. A larger MEI and AWMEI value a more compacting trend of urban growth.

3

4

where LEI_i is the LEI for a new patch, N is the total number of new patches, a_i is the area of the new patch, and A is the total area of all the newly grown patches.

Impacts of urbanization on land use and landscape elements

Changes in land cover types due to urbanization

Changes in the land cover (LC) area can show structural characteristics and the direction of regional LC type changes, clarify transfers between LC types, and accurately reflect the dynamic characteristics of LC changes³¹. To study the effects of urbanization on land cover, methods involving the land cover area transfer matrix were employed to examine the conversion of areas from one type of land cover to another between 1990 and 2020 (Eq. 6)³². The size of land change for each category (Gain, Loss, and Persistence) and the corresponding components (percentages of quantity, Exchange, and Shift) were calculated and depicted on maps. The detailed procedures and mathematical expressions to quantify land use change at the interval, transitional, and category levels were found in¹⁴.

5

Where Sij reflects the area shift from one land cover type (LC) to another (LC) across the entire study period. The variable n represents the total number of land cover types.

Changes in landscape structure and pattern due to urbanization

Land use structural changes reflect trends in human intervention regarding land resources and land use practices³³. These changes can be measured using information entropy, the balanced index, and the dominance index³⁴ (Fig. 2). Information entropy indicates land-use fragmentation, with higher values reflecting greater segmentation and indicating urban sprawl and the diversification of land-use types³⁵. The balanced index assesses land allocation equity among categories, ranging from 0 (unbalanced, dominated by one type) to 1 (fully balanced distribution). The dominance index measures the extent of a specific land-use type’s dominance; higher values indicate reduced diversity and a more uniform land-use structure. The formulas for Entropy (H), dominance index (DI), and balanced index (BI) are indicated in Eqs. 7, 7, and 8, respectively^21,35.

6

where, Pi​: Proportion of the area of land use type i relative to the total study area; n: Number of land use types

7

.

Where. H_max=ln_m shows the maximum diversity Index; m is the quantity of land-use type.

8

Landscape pattern metrics provide insights into land-use changes due to urbanization, helping evaluate ecological impacts²². Four key metrics are selected to capture landscape characteristics: Largest Patch Index (LPI), edge density (ED), landscape shape index (LSI), and aggregation index (AI). For instance, an increase in ED indicates greater fragmentation, while a decrease in AI reflects reduced connectivity among natural patches. Additionally, a rise in the LSI suggests more irregular urban growth patterns. These metrics were calculated using Fragstats 4.0 software³⁶ with a grid cell size of 4 km × 4 km to ensure precision in the study area. Although the input LULC data had a finer 1 km resolution, this coarser grid size was selected to reduce local-scale noise and emphasize broader landscape configurations across the study area. This approach involves a trade-off between spatial precision and pattern generalization, but it provides a more stable and interpretable assessment of large-scale urbanization impacts on landscape structure.Using the Create Fishnet tool in ArcGIS 10.8, we divided the land use and land cover maps into these grid cells, allowing us to obtain the four selected metrics (LPI, ED, LSI, and AI) at the class level for each cell.

Results

Spatial-temporal land cover change

The spatial-temporal characteristics of LC change in the study southern China region are indicated in Table 1; Fig. 3. Over the past 30 years, forest land consistently dominated as the primary land use type, comprising 57.28% in 1990, 57.49% in 2000, 57.38% in 2010, and 57.05% in 2020. Cultivated land was the second most prevalent LC type (Table 1; Fig. 3). Furthermore, it was discovered that the cultivated land, forest land, and grassland experienced a gradual decline, while the water and built-up areas exhibited a noticeable increase. Specifically, the area of cultivated land decreased by 19,451 km², and grassland decreased by 3646 km² over the past 30 years, representing a decline rate of 7.1% and 6.5%, respectively. In contrast, the built-up area increased by 107% (22181 km²) between 1990 and 2020.

Table 1.

Land use land cover statistics in Southern regions from 1990 to 2020.

	1990		2000		2010		2020
LC class	Km2	%	Km2	%	2010	%	Km2	%
Cultivated land	292,843	30.09	289,228	29.73	282,556	28.98	273,392	28.13
Forest land	557,421	57.28	559,401	57.49	559,499	57.38	554,513	57.05
Grassland	59,921	6.16	57,963	5.96	56,616	5.81	56,275	5.79
Waterbodies	37,903	3.90	38,792	3.99	40,270	4.13	40,695	4.19
Built-up	20,758	2.13	23,348	2.40	29,968	3.07	42,939	4.42
Unused land	2210	0.23	2173	0.22	2004	0.21	2081	0.21

Fig. 3.

The spatial and temporal distribution of LC between 1990 and 2020:Map created using QGIS 3.32 (QGIS Development Team, 2023; https://qgis.org)²⁴.

Urbanization measures and indicators

Urbanization by demographic and economic indicators

The 1 km×1 km grid (Fig. S1) was used to depict spatial patterns of urban population density and GDP. The population density has experienced a gradual growth over 30 years, rising from 263.64 persons/km² in 1990 to 348.01 persons/km² in 2020 (Table 2). The data clearly show that the population density experienced a modest increase from 1990 to 2000, followed by a gradual fall from 2000 to 2010 (Table 2). Subsequently, between 2010 and 2020, there was a further increase in population density. Concurrently, a marked increase in GDP exhibited a consistent upward trajectory from 2.6 million yuan/km² in 2000 to 25.25 million yuan/km² in 2019 (Table 2).

Urbanization by urban areas expansion

Table 3 displays the growth rate of artificial or built-up surfaces in the study regions, including townland areas, rural settlements, and construction areas. The townland, rural settlements, and construction area grew continuously between 1990 and 2020. Moreover, a substantial increase in urban development started in 2000 (Table 3). Consequently, the townland area, rural settlements, and construction area experienced rapid growth, with respective annual growth rates of 4.77%, 0.91%, and 18.36% between 2000 and 2010. The amount of land used for construction has significantly increased over the past 30 years, with an annual growth rate of 39.45%, equivalent to 329 km² per year. In addition, the amount of town land increased yearly by 6.44% compared to the town areas occupied in 1990–2000 (Table 3). Overall, built-up areas expanded significantly, from 20,758 km² in 1990 to 42,939 km² in 2020, reflecting an overall annual growth rate of 3.56%.

Table 3.

Annual built-up area growth rate between 1990 and 2020.

	Areas in km2				Annual expansion of built-up areas
years	1990	2000	2010	2020		1990–2000		2000–2010		2010–2020		1990–2020
Built-up areas	km2	Km2	km2	km2		km2/yr	%	km2/yr	%/yr	km2/yr	%/yr	km2/yr	%
Townland	4699	6084	8989	13,783		138.5	2.95	290.5	4.77	479.4	5.33	302.80	6.44
Rural settlements	15,225	16,033	17,488	18,452		80.8	0.53	145.5	0.91	96.4	0.55	107.57	0.71
Construction land	834	1231	3491	10,704		39.7	4.76	226	18.36	721.3	20.66	329.00	39.45

Urban expansion types and modes

Figures 4 and 5 illustrate the types and modes of urban expansion patches during 1990–2020. Urban area growth mostly took the form of edge expansion (Fig. 5). Figure 4a presents the percentage of three landscape expansion patterns in the three periods. Initially, between 1990 and 2000, urban growth was characterized by outlying and edge expansion, and outlying patches were the dominant mode of urban expansion. However, after 2000, three urban expansion modes were created. From 2000 to 2010, edge expansion patches exceeded the number of outlying and infilling patches. Subsequently, after 2010, both edge expansion and outlying patches became the primary modes of urban expansion. Specifically, from 2010 to 2020, the urban expansion area resulting from edge expansion and outlying was 93.99 × 10² km² and 135.42 × 10² km², respectively, accounting for 38.6% and 55.6% of the total expanded urban area (Fig. 4b).

Fig. 4.

Numbers of built-up patches and landscape expansion types between the study periods: (a) Patch numbers in percentage; (b) Patch areas in km².

Fig. 5.

Spatial distribution of three LEI types from 1990 to 2020:Map created using ArcGIS Desktop 10.8 (ESRI, 2020; https://desktop.arcgis.com/en/arcmap/index.html)³⁷.

Moreover, to further enhance our understanding of urban growth morphology and development trends in the study regions, we calculated the MEI and AWMEI for growth patterns between 1990 and 2020 (Table 4). By analyzing the variation of the MEI and AWMEI, we can gather additional information about the type of urban expansion that has occurred. Both the MEI and AWMEI exhibited an increase from the first period (1990–2000) to the second period (2000–2010), followed by a decline in the third period (2010–2020) (Table 4). The increasing trend from the first to the second period indicates a compact or coalescence pattern of urban growth. In contrast, the decreasing trend in the first and third periods shows a dispersion urban growth pattern.

Table 4.

MEI and AWMEI of newly grown urban patches from 1990–2020.

Index	1990–2000	2000–2010	2010–2020
MEI	1.85	15.03	6.63
AWMEI	2.88	21.91	12.29

Impacts of urbanization on land use and landscape patterns

Changes in land use and land cover

The cumulative land area that changed between 1990 and 2020 amounted to 256,888 km² (Table S1). The spatial distributions of the transition map are also indicated in Fig. S2. Table 6 summarizes the loss of the four main LC types due to the expansion of various impervious surfaces. The most significant LC changes occurred when cultivated land was converted to impervious surfaces, including townland, rural settlements, and construction land, accounting for 25,999 km² (74.5%) of all land cover changes (Table 5). This was followed by converting forest land into impervious surfaces, constituting 6429 km² (18.4%) of the total LC. Grasslands and waterbodies also experienced significant losses within the last 30 years. Hence, urban development can be regarded as the outcome of LC change, predominantly driven by the conversion of forest and agricultural land.

Table 5.

The contribution of major LC types to the growth of urban lands from 1990 to 2020.

1990–2020
LC class name	Townland	Rural settlements	Construction land	Total (km2)	Proportion (%)
Cultivated land	7746	12,762	5491	25,999	74.45
Forest land	1153	1645	3631	6429	18.41
Grassland	160	251	550	961	2.75
Waterbodies	577	513	368	1458	4.18
Total (km2)	9636	15,171	10,040	34,847	99.79

Changes in land use structure and landscape pattern

The combined analysis of Shannon Entropy (H), the Balanced Index (BI), and the Dominance Index (DI) from 1990 to 2020 provides a clear understanding of how urbanization has significantly influenced the land-use structure in the study area (Fig. 6). The Entropy (H) increased steadily from 1.037 in 1990 to 1.092 in 2020. Similarly, the BI increased from 0.579 in 1990 to 0.609 in 2020. This indicates a progressive reduction in the dominance of a single land-use type, leading to a more balanced land-use structure over time. In contrast, the Dominance Index (DI) revealed a consistent decline from 0.421 in 1990 to 0.391 in 2020. This trend indicates that urban expansion has diminished the prevalence of forest and cultivated land, creating more space for other land uses such as built-up areas.

Fig. 6.

Land-use structural indexes and changes in study regions.

Moreover, the results of changes in land-use patterns at the landscape and class levels over the past 30 years are shown in Fig. 7. The landscape-level analysis reveals a clear trend of increasing fragmentation and complexity due to urbanization over the past three decades. Specifically, the Largest Patch Index (LPI) indicates a declining trend in cultivated and forest land, suggesting fragmentation and conversion to other land-use types. In contrast, the Aggregation Index (AI) for built-up areas has increased, showing that urban expansion has created a more connected and continuous urban landscape. Additionally, the Edge Density (ED) values for built-up areas have risen, reflecting urban sprawl and more fragmented landscapes. Furthermore, the Landscape Shape Index (LSI) values have increased for built-up areas, indicating more irregular and complex urban forms, which are characteristic of rapid urbanization.

Fig. 7.

Landscape patterns by different landscape indices from 1990–2020.

Spatial-temporal distributions and heterogeneity of built-up areas

The spatial-temporal distributions and heterogeneity of the built-up areas during 1990–2020 are indicated in Fig. 8. The LPI spatial distribution maps show an increase in the dominance of large urban patches, particularly around Shanghai and Jiangsu, indicating urban expansion. ED exhibits a substantial rise in edge density, particularly in urbanizing areas (Shanghai, Hubei, and Anhui), highlighting increased fragmentation of the landscape. The AI maps reveal a decline in aggregation in many regions between 1990 and 2020, suggesting that urban patches have become more dispersed rather than clustered. The spatial distribution of LSI maps also demonstrates structural changes in land-use patterns, with urban expansion leading to more complex and irregular landscape shapes.

Fig. 8.

The spatial-temporal distribution of built-up land landscape pattern between 1990 and 2020: Map created using ArcGIS Desktop 10.8 (ESRI, 2020; https://desktop.arcgis.com/en/arcmap/index.html)³⁷.

Discussion

Urbanization and driving factors

Evaluating the effects of urbanization on land use-landscape patterns is crucial for protecting natural resources and promoting environmental sustainability. A recent study³⁸ identified urban population growth and area expansion as the primary drivers of rapid urbanization. Over the past 38 years, China’s urbanization rate has risen from 19.39% in 1980 to 59.58% in 2018³⁸. According to the current study, the population density in southern China has steadily increased over the past 30 years, rising from 263.64 persons/km² in 1990 to 348.01 persons/km² in 2020 (Table 2). Additionally, multiple studies have stated that GDP serves as a socioeconomic indicator for the level of urbanization and is the main driver of land use change^39–42. Our study of the economic trends in Southern China’s regions revealed a steady rise in GDP from 2000 to 2019 (Table 2).

Urbanization involves not only socioeconomic clustering and population activities but also impervious surface expansion, which are the most remarkable activities in rapid urbanization³⁸. The annual urban growth rate (AGR) and landscape expansion index (LEI) have the potential to capture the evolution of urban expansion and morphology¹⁹. The findings show that urban area expansion and development patterns had different effects during the study periods. A substantial increase in AGR occurred between 2000 and 2010. The town and construction land increased by 479.4 km²/year and 721.3 km²/year, respectively (Table 3). Similarly, LEI results indicate that a compact or coalescence pattern of urban growth was found from 1990 to 2010, while a dispersion urban growth pattern was found from 2010 to 2020 (Table 4). Several studies have examined the driving factors of urbanization on alterations in land use and landscape patterns^2,9,43–47.

How does urbanization affect land use and landscape patterns?

Rapid urbanization has significantly altered land use, land cover, and landscape patterns through the conversion of land cover types and landscape fragmentation. The current study in southern China shows that in the past 30 years, a built-up area growth of 107% is primarily due to the conversion of cultivated (74.5%) and forest land (18.4%). Similarly, a recent study in China revealed that urban built-up areas expanded 9.1 times between 1995 and 2020, transforming mainly from cultivated land, forests, and grassland⁹. In other studies in Yunnan, China, urbanization has led to significant land-use changes, particularly an increase in construction land at the expense of cultivated land, forest land, and grassland⁴⁸. Studies in Belgaum City, urbanization has led to a significant decline in vegetation cover, from 98.8% in 1989 to 91.74% in 2012, indicating a transformation of land use towards built environments⁴⁹.

Urbanization has also driven the homogenous or dominant natural landscape elements to a more diversified and fragmented landscape. Landscape fragmentation leads to a reduction in total habitat size, dividing the landscape into smaller patches and increasing the isolation of ecosystem patches²⁷. The results of landscape metrics LPI, ED, AI, and LSI, and land use structure (H, BI, DI) in the southern region of China indicate increasing landscape fragmentation and decreasing aggregation (Figs. 6 and 7, and 8). The findings also indicate that changes in landscape patterns at the class level have been linked to increased urbanization (Fig. 7). These findings align with recent studies, emphasizing that uncontrolled urbanization not only reshapes land-use patterns but also undermines biodiversity conservation and ecosystem resilience^50,51, biodiversity decline, ecosystem instability, and reduced agricultural resilience^52,53. Similarly, studies have shown that as urban areas expand, the landscape patterns become more diverse and uniform, indicating increased heterogeneity within urbanized regions^54,55. Similar findings have been reported in other studies, which demonstrate that urban sprawl leads to greater landscape fragmentation and biodiversity losses^27,56,57. A recent study reported that urbanization emergence of irregular urban patches and mixed land uses due to sprawling development⁵⁸.

Moreover, urbanization may negatively affect biodiversity and ecosystems primarily through habitat destruction, fragmentation, and degradation, which reduces overall biodiversity and leads to a decline in ecosystem services^50,59.

Implications of the study

The movement toward urbanization has significantly changed landscape patterns and land use types, profoundly impacting biodiversity, ecological functions, and environmental sustainability^47,60. For instance, deforestation contributes to habitat destruction, posing challenges to ecological security⁶¹. Additionally, the destruction of grasslands and forests has worsened the occurrence of heat islands and heat waves in cities, which has had a detrimental impact on human health and well-being¹⁴. Moreover, urban expansion primarily encroaches on agricultural land, depleting valuable cropland resources and exacerbating urban food insecurity¹³.

As urban growth and expansion continue in this study of Southern region of China, the predominant landscapes, such as forests and cultivated land, have gradually been replaced by urban areas, resulting in a more fragmented landscape pattern and a change in the natural ecosystem of the region.

Therefore, in the southern region of China, where forest land accounts for 57% and cultivated land for 28% of total land use, urbanization has triggered significant changes in land-use structure and spatial patterns. The reduction in cultivated land and fragmentation of forested areas pose risks to ecosystem stability, biodiversity conservation, and food security. The observed shifts in landscape metrics, such as increased edge density and landscape shape complexity, indicate intensifying human influence, leading to habitat fragmentation and altered ecological processes. As results, these findings highlight the urgent need for integrated urban planning that harmonizes economic development with ecological sustainability. The observed landscape changes—particularly the loss of cultivated and forested land—underscore the importance of aligning urban expansion with China’s key environmental policies. Specifically, the Ecological Red Line Policy serves to safeguard critical ecosystems and maintain ecological security⁵²; the Farmland Protection and Red Line Policy ensures the preservation of productive agricultural land and food security⁶²; and the Carbon Neutrality Goals promote sustainable land use and low-carbon urbanization⁶³. Coordinating these national strategies through spatially informed land-use management can effectively mitigate the adverse ecological impacts of urban growth while fostering a more balanced and sustainable development pathway for the region.

Conclusions

This study aimed to measure urbanization from 1990 to 2020 using economic, demographic, and urban morphology indicators, and to assess its effects on land use and landscape patterns. Results showed significant growth in urban areas, particularly from 2010 to 2020, with built-up areas expanding from 20,758 km² in 1990 to 42,939 km² in 2020, marking a 107% increase. The land expansion index (LEI) indicated a compact growth pattern from 1990 to 2010 and a dispersion pattern from 2010 to 2020. In the past 30 years, a built-up area growth of 107% is primarily due to the conversion of cultivated (74.5%) and forest land (18.4%) relative to others land cover types. These changes led to a change in the landscape pattern and have caused fragmentation and heterogeneity in the whole landscape. Landscape metrics results also revealed increasing fragmentation and decreasing aggregation, highlighting the need for integrated urban planning approaches that balance development with ecological conservation and environmental sustainability.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Y.R: Supervision, project management, Project administration: Z.P: Dataset preparations and classifications; A.M: Conceptualization, Methodology, Writing- Original draft, Formal analysis; S.Z: Writing - review and editing.

Funding

This work was supported by the National Key Research Program of China (2022YFF1303001), National Natural Science Foundation of China (42001210, 31972951, 31670645, 42171100, 41801182, and 41807502).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.