Urbanization continues to shape modern societies, with its pace accelerating globally at four times the previously estimated rate (Meyfroidt et al., 2022; Winkler et al., 2021; Liu et al., 2020). While urban development drives economic progress, it simultaneously introduces environmental and social challenges such as urban heat islands (Zhou et al., 2004), water-induced soil degradation (Borrelli et al., 2020), declining air quality affecting public health (Hong et al., 2019; Wu et al., 2024), and disparities in infrastructure access (Pandey et al., 2022). Rapid urban expansion has consumed most flat and low-elevation land, pushing agricultural, industrial, and residential activities into mountainous and marginal zones (Bren d’Amour et al., 2017; Chen et al., 2020; Yang et al., 2022). Given finite land availability, farmland protection policies, and sustainability goals, extensive land use models are no longer viable. Enhancing land use intensity and managing spatial development systematically are now critical to resolving human-land conflicts and supporting high-quality regional growth. A central challenge arises between economic expansion and ecological preservation: cities must grow to sustain economic activity, yet unchecked sprawl causes irreversible environmental harm, undermining long-term prosperity. Improving land use efficiency (LUE) offers a strategic path to limit unnecessary expansion and reduce ecological strain.
Boosting LUE is essential for balancing urban economic advancement with responsible land stewardship. LUE is measured as economic output per unit of urban land (He et al., 2020; Li et al., 2014; Luo and Wu, 2003; Wu et al., 2017; Yang and Luan, 2024). China’s 14th Five-Year Plan and Vision 2035 emphasize improving resource utilization, making LUE research imperative. Current studies focus on measurement techniques, spatial scales, and influencing variables. Quantitative tools like data envelopment analysis (Chen et al., 2016; Zhu et al., 2019) and stochastic frontier analysis (Wang et al., 2021) dominate the field. Research typically operates at macro levels—national, regional, or metropolitan. Key drivers include economic development (Chen et al., 2016), population density (Xue et al., 2022), industrial composition (Liu et al., 2021; Wang et al., 2023), clustering of industries (Zhang et al., 2022), urban layout (Liao et al., 2023; Wu et al., 2022), regional integration (Zhao et al., 2021), and transport connectivity (Wu et al., 2017).
Existing LUE assessments face several constraints. First, urban land area measurement often relies on built-up area statistics, which may not reflect actual conversion from non-urban to urban land during development. Second, intra-city spatial variation in LUE is underexplored due to reliance on aggregated administrative data, limiting fine-scale analysis. Third, the process of urban land maturation and efficiency enhancement (ULMEE) receives insufficient attention, despite its importance in understanding how land value evolves over time. Fourth, while LUE reflects land productivity, its role in driving urban economic growth remains inadequately studied.
China’s coastal cities, housing over 20% of the population and generating 30% of national GDP, feature advanced industries and strong innovation capacity, granting them strategic significance. These areas exhibit the country’s fastest urban land growth (Liu et al., 2015; Shi et al., 2015), high urbanization rates, extensive redevelopment (Cao et al., 2023; Huang et al., 2023), and greater economic exposure than inland regions (Chan et al., 2021). Amid nationwide land reform (Han et al., 2023), coastal zones face acute land scarcity. Addressing gaps in LUE research and leveraging the uniqueness of these cities, this study aims to: (i) use land use records and high-resolution satellite imagery to accurately assess urban land extent; (ii) develop a LitPop method combining nighttime light and population data to estimate LUE at micro scales with enhanced spatial detail; (iii) analyze the ULMEE process by tracking economic output and efficiency gains across land parcels of varying ages (based on conversion year from non-urban to urban land relative to 2020); and (iv) project future LUE under shared socioeconomic pathways (SSPs) to evaluate whether ULMEE alone can sustain a 5% annual urban economic growth rate. This approach transcends administrative boundaries and supports dynamic land monitoring and efficient use. By assessing ULMEE’s capacity to meet growth targets, the study informs strategies for balanced development in coastal regions.
The study covers 51 coastal cities (excluding Hainan), including 4 super cities, 4 megacities, 3 Type I large cities, 19 Type II large cities, 18 medium-sized, and 3 Type I small cities (Fig. 1). These urban centers serve as economic engines, concentrating population, GDP, and high-tech industries, while leading national urbanization. Their expansion differs from inland areas: export-oriented economies drive diversified, market-responsive spatial growth; coastal geography imposes rigid land limits, prompting solutions like land reclamation. However, rapid growth brings challenges—overcrowding, land shortages, and ecological vulnerability—exacerbated by sea-level rise and coastal erosion. Regional imbalances and urban dysfunction make these areas vital for studying sustainable land use, economic potential, and development optimization.
The analytical framework (Fig. 2) consists of four components: (A) mapping LUE patterns using multi-source data and the LitPop method; (B) analyzing ULMEE by incorporating land age; (C) projecting future LUE under SSP scenarios to assess alignment with 5% growth targets; and (D) exploring ULMEE mechanisms via POI data and kernel density estimation.
Data sources include China’s land use datasets (https://www.resdc.cn/DOI/DOI.aspx?DOIID=54), nighttime lights (DMSP/OLS, NPP/VIIRS) from NOAA (https://www.ngdc.noaa.gov/), and the China City Statistical Yearbook (http://www.stats.gov.cn/). To adjust for inflation, 1995 was used as the base year, with secondary and tertiary GDP indices deflating nominal values from 2000–2020 to derive real GDP. A gridded dataset of sectoral value-added under SSPs (2020–2100) is available at https://doi.org/10.4121/14113706.v2 (Jing et al., 2022). POI data comes from Amap (https://www.amap.com/). All data is included in the manuscript or supplementary materials (Table S1).
The economic elasticity coefficient (EEC) evaluates the balance between economic growth and land expansion across city sizes (Huang et al., 2013; Xu et al., 2017). Six periods are analyzed: 1995–2000, 2000–2005, 2005–2010, 2010–2015, 2015–2020, and 1995–2020. The formula is:
$${\rm{EEC}}= \frac{\root{n}\of{{{GDP}}_{b}-{{GDP}}_{a}}-1}{\root{n}\of{{{UL}}_{b}-{{UL}}_{a}}-1}$$
(1)
Here, EEC reflects the relative pace of economic and land expansion; ${{GDP}}_{a}$, ${{GDP}}_{b}$ are secondary and tertiary GDP sums in base and target years; ${{UL}}_{a}$, ${{UL}}_{b}$ are urban land areas in those years; $n$ is the period length. EEC categories: contraction (EEC < 0), land expansion (0 ≤ EEC < 1), basic harmony (1 ≤ EEC < 5), and economic growth (EEC ≥ 5) (Ouyang and Zhu, 2020; Qiao and Huang, 2021). Nighttime light data correlates strongly with economic activity and is widely used in spatializing GDP and population (Wu et al., 2013; Li and Zhou, 2018; Tan et al., 2018; Zheng et al., 2023). It also supplements economic indicators in data-scarce regions (Chen and Nordhaus, 2011). Traditional methods allocate socioeconomic data proportionally by light intensity, but this overestimates suburban values and underestimates urban cores (Eberenz et al., 2020). Zhao et al. (2017) introduced the LitPop method, combining light and population data for more accurate distribution, later refined by Eberenz et al. (2020) and widely adopted (Li et al., 2024; Wang and Sun, 2022). This study applies an enhanced LitPop framework: (i) nighttime light images are masked by annual urban land boundaries to reduce spillover effects; (ii) improved LitPop (Eq. 2) generates light-population grids from 1995–2020; (iii) a relationship is established between city-level real GDP and light-population data; (iv) GDP is downscaled to 1 km × 1 km grids to compute LUE. This reduces saturation bias and better captures true economic activity: $${\rm{LitPop}}= \left\{ \begin{array}{ll}{\rm{Pop}} ; {\rm{Lit}}=0 {\rm{Lit}} , \cdot ,{\rm{Pop}} ,{\rm{Lit}} ,> ,0 , & ,{\rm{Pop}} > 0 {\rm{Lit}} ; {\rm{Pop}}=0 \end{array} \right.$$
(2)
$${{\rm{LUE}}}_{{LitPop}}= \frac{{{GDP}}_{i}}{{{SLP}}_{i}}* {{LitPop}}_{{pixel}}$$
(3)
where ${\rm{Lit}}$ is nighttime light pixel value, ${\rm{Pop}}$ is population grid value, ${{LitPop}}_{{pixel}}$ is the resulting light-population pixel, ${{SLP}}_{i}$ is total light-population pixels per city, ${{GDP}}_{i}$ is real GDP of secondary and tertiary sectors, and ${{LUE}}_{{LitPop}}$ is pixel-level LUE.
The ULMEE process builds on urban life cycle theory (Suarez-Villa, 1985), which describes cities evolving through birth, growth, maturity, and decline. This study adapts the framework to LUE, defining stages as initial, growth, maturity, and recession (Fig. 3A). In the initial phase, land is newly developed with low efficiency. During growth, resource concentration boosts productivity. At maturity, full development and optimized industrial structures push LUE toward its peak. In decline, inefficient use leads to idle land and falling output. However, proactive planning can renew land use, restarting the cycle.
To operationalize ULMEE, “age” is defined as the difference between 2020 and the year land was converted to urban use. For land converted in 1995, ages are 0, 5, 10, 15, 20, 25 (5-year intervals); for 2000, ages are 0, 5, 10, 15, 20; etc. As land ages, economic upgrading, policy guidance, and innovation enhance functionality and efficiency (Fig. 3B), constituting the maturation and efficiency improvement process.
Coastal cities added urban land from 8,351.82 km² in 1995 to 30,843.52 km² in 2020, growing at 899.67 km²/year. Expansion peaked between 2005–2010. Top cities by new land—Tianjin, Dongguan, Tangshan, Guangzhou, Shanghai, Ningbo, Weifang, Quanzhou, Dalian, Nantong—are all large or above, leveraging their scale advantages. Growth followed multi-center radial or gradual point-to-surface patterns, clustering along coasts and roads (SI, Fig. S1A). Land area varied by city size: Type II large > medium > super > mega > Type I large > Type I small (SI, Fig. S1B). Expansion intensity metrics revealed phased and fluctuating trends (SI, Fig. S1C, S1D).
Total economic output rose from 1,542.49 billion yuan (1995) to 17,917.18 billion yuan (2020). Annual new output grew from 1,147.94 billion (1995–2000) to 4,340.36 billion (2015–2020) (SI, Fig. S2A). Output by scale: Type II large > super > medium > mega > Type I large > Type I small. LUE was highest in city centers, declining outward, showing strong intra-city variation (SI, Fig. S2B). Average LUE rose from 203 million yuan/km² (1995) to 605 million yuan/km² (2020), increasing 16 million annually (SI, Fig. S2D). Over six periods, 22 to 27 cities exceeded the coastal average, indicating uneven development. All city sizes showed rising LUE, though at varying rates (SI, Fig. S2C).
From 1995–2020, older urban land generated higher total output, confirming maturation (Fig. 5). Output ranked by conversion year: 1995, 2020, 2010, 2005, 2015, 2000—reflecting area and maturity differences. Among city sizes, super, Type II large, and medium cities led in 1995-converted land output; mega, Type I large, and Type I small lagged.
Table 1 shows average LUE rising from 323 million yuan/km² (5-year-old land) to 695 million (15 years) and 922 million (25 years), confirming efficiency gains with age. New land (0-year-old) improved from 205 million (2000) to 531 million (2020), indicating higher initial efficiency in recent developments. This reflects compact planning and infrastructure reuse, which boost early productivity. For 1995 land, LUE rose gradually with urbanization, peaking as economies stabilized and industrial structures improved.
Urban scale influences growth trajectories, making ULMEE analysis vital for tailored strategies. Figure 6 shows LUE generally improves with age, but not uniformly. For 1995 land, 25-year LUE fell below 20-year levels in some cases. By scale, 2020 LUE ranked: super > Type I large > mega > Type II large > medium > Type I small—challenging the assumption that larger cities always have higher efficiency. For super cities, 0-year LUE rose from 261 million (1995) to 498 million (2020), confirming newer land starts more efficiently.
Can ULMEE sustain future growth? Table 1 and Figures 5–6 confirm the process exists. In 2020, 69% of urban land was under 15 years old, indicating substantial untapped potential. Using SSPs (O’Neill et al., 2014, 2017) to project LUE, a gridded dataset (2020–2100) forecasts growth trajectories. From 2025–2050, LUE follows a rising-then-falling “inverted U” pattern (SI, Fig. S3): rapid gains before 2035, slowing after. Across SSPs (sustainability, middle-of-the-road, regional rivalry, inequality, fossil-fueled), LUE differences are small in 2025 but widen over time. SSP5 yields highest LUE; SSP3, lowest. By 2050, LUE ranking by scale: super, Type I large, mega, Type I small, Type II large, medium.
China’s growth outlook (Bank of China Research Institute, 2024; Tang et al., 2020) suggests a 5% annual target aligns with long-term goals. Assuming positive momentum, projected LUE under SSPs indicates that, except under SSP3, all city sizes can meet this target via ULMEE until 2035. Beyond 2040, new land will be needed. Type I small cities can rely longest on ULMEE before requiring expansion.
Coastal cities were chosen for their unique pressures and innovations. Over 30 years, they contributed 20.96% of national population, 33.34% of GDP, and 55.91% of urban land. They face dual pressures: high economic demand and severe land constraints, often expanding via reclamation. Climate risks like sea-level rise further limit space. Yet, they pioneer land management reforms—stock renewal, vertical development, mixed-use zoning—offering models for other regions. Studying ULMEE here reveals how cities thrive under scarcity, providing insights for developing economies.
POI data, rich in spatial detail, aids functional zone mapping and urban analysis (Liu et al., 2024; Long et al., 2024; Xue et al., 2020). Its density and clustering help trace ULMEE. Kernel density estimation (KDE) of POIs correlates positively with LUE (r = 0.57**, p < 0.01). A 2.10 km² plot in Dalian (Fig. 8A) saw land expansion from 2012–2020. POI types diversified, adding residential to commercial, industrial, public services, science, and education (Fig. 8B). Total POIs rose from 114 to 290. KDE values increased from 34.17–90.53 (mostly <80) in 2012 to 35.68–275.91 (mostly >100) in 2020 (Fig. 8C), reflecting improved service capacity and infrastructure, supporting structural optimization and land productivity.
— News Original —
Can urban land maturation and efficiency enhancement meet future economic growth demands in China’s coastal cities
Urbanization is an inevitable trend as human society evolves. The global urbanization rate is unprecedented, at a level four times greater than previously estimated (Meyfroidt et al., 2022; Winkler et al., 2021; Liu et al. 2020). Urbanization not only supports economic development, but also causes environmental and social challenges, including climate change such as urban heat island effect (Zhou et al., 2004), soil erosion by water (Borrelli et al., 2020), deteriorating air quality and human health (Hong et al., 2019; Wu et al., 2024), and infrastructure inequality (Pandey et al., 2022). Due to the rapid advancement of urbanization and the increasing scarcity of land resources, available land in plains and low-altitude areas has gradually been occupied by urban construction, forcing agriculture, industry and residential activities to move to mountainous highlands and marginal farmland areas (Bren d’Amour et al., (2017); Chen et al., 2020; Yang et al., 2022). With the limited availability of land resources, related farmland protection policies, and urban sustainable development, an inefficient and extensive land use development pattern is no longer viable. Therefore, under the premise of ensuring the livability of the city, rationally increasing the level of intensive utilization and the orderly development of land resources are essential to resolving the “human–land conflict” and ensuring high-quality urban economic and regional coordinated development. As a direct consequence, a fundamental tension emerges between urban economic growth and sustainable development: while cities must expand to accommodate economic activities, unconstrained spatial expansion leads to irreversible ecological damage and ultimately undermines long-term economic viability. In this context, enhancing land use efficiency provides a fundamental pathway to reduce unnecessary spatial expansion and its associated environmental pressures. n nImproving land use efficiency is an important solution to balance urban economic growth and sustainable use of land resources. Considering the economic benefits of urban land, LUE is defined as the economic output per unit of urban land area (He et al., 2020; Li et al., 2014; Luo and Wu, 2003; Wu et al., 2017; Yang and Luan, 2024). The Outline of the 14th Five-Year Plan (2021–2025) for National Economic and Social Development and Vision 2035 of the People’s Republic of China emphasize the comprehensive improvement in resource utilization efficiency. Therefore, it is essential to conduct research related to LUE. Current, LUE research focuses on measurement methodologies, research scales, and driving factors. Employing mainly quantitative methods, such as data envelopment analysis (Chen et al., 2016; Zhu et al., 2019) and stochastic frontier analysis (Wang et al., 2021). The spatial scales are dominated by macro scales, such as countries, urban agglomeration, and developed-region cities. The driving factors include economic development level (Chen et al., 2016), population density (Xue et al., 2022), industrial structure (Liu et al., 2021; Wang et al., 2023), industrial agglomeration (Zhang et al., 2022), spatial structure (Liao et al., 2023; Wu et al., 2022), regional integration (Zhao et al., 2021), and transportation accessibility (Wu et al., 2017). n nExisting studies rely mostly on statistical data to measure LUE, presenting several limitations. The first is the measurement of urban land area: The conversion of non-urban land to urban land, as an important feature of the change in land use patterns, is a byproduct of the demand for urban land during urbanization and economic development. However, in the process of computing LUE, the built-up area in the statistical data is generally recorded as urban land area, resulting in a discrepancy between this value and the actual urban land area. Second, the spatial heterogeneity of LUE within cities has not been adequately considered: Socioeconomic statistical data, typically available as text at various levels of administrative divisions, are suitable for macro-analysis and the overall evaluation of cities (S. Liu et al., 2020; Song et al., 2022), but make it difficult to investigate the spatial heterogeneity of LUE within cities or at a fine scale. Third, there is a lack of focus on the urban land maturation and efficiency enhancement (ULMEE) process: Existing research focuses on the spatiotemporal evolution of LUE and the identification of influencing factors, whereas LUE characterization in the ULMEE process is relatively weak. Fourth, while LUE is a direct representation of the economic output capability of urban land, its impact on urban economic growth has not been fully explored in the literature. n nChina’s coastal cities, accounting for over 20% of the country’s population and 30% of its economic output, have a well-developed industrial structure and strong scientific and technological innovation capabilities, giving them substantial economic and strategic advantages. These cities have the highest rate of urban land expansion in the country (Liu et al., 2015; Shi et al., 2015), with a high level of urbanization, extensive land redevelopment (Cao et al., 2023; Huang et al. 2023), and significantly higher economic exposure than inland cities (Chan et al., 2021). China is currently undergoing substantial land use reform (Han et al., 2023). Coastal cities, as the core areas of China’s economic development, face more stringent land resource limitations. Therefore, in response to the abovementioned limitations of LUE studies and given the representativeness and uniqueness of China’s coastal cities, the main objectives of this study are as follows: (i) to use various land use data and high-definition remote sensing images to measure the urban land area, and gain an in-depth understanding of the current state of urban land in China’s coastal cities; (ii) to propose a LitPop method for measuring LUE that integrates nighttime light and statistical data, and investigate the spatial pattern of LUE at the micro-scale and with comprehensive spatial information; (iii) based on the process of increasing the total economic output of urban land of different ages (the difference between the urban land generation year (time of conversion from non-urban land to urban land) and 2020) and enhancing the LUE of land at different ages, to clarify the specific ULMEE process and reveal the characteristics of LUE differences in coastal cities and of different urban scales; and (iv) from the perspective of different urban scales, using shared socioeconomic pathways (SSPs), to predict future LUE at different ages and determine whether relying on the ULMEE process can meet the demands of a 5% urban economic growth rate. This method overcomes the limitations of administrative divisions and contributes to the dynamic supervision and intensive utilization of urban land. Simultaneously, based on the ULMEE process to determine whether it satisfies economic growth demands, this study helps clarify the imbalanced economic development of China’s coastal cities and contributes to decision-making support by revealing strategies to enhance urban economic development. n nStudy area n nChina’s coastal region comprises 51 cities (excluding Hainan Province), including 4 super cities, 4 megacities, 3 Type I large cities, 19 Type II large cities, 18 medium-sized cities, and 3 Type I small cities (Fig. 1). As the core engine of China’s economic development, coastal cities concentrate the nation’s primary economic output, population resources, and high-tech industries, while leading urbanization with their unique geographical advantages. Compared with other cities, coastal urban expansion demonstrates distinct characteristics: the outward-oriented economy drives more globalized and market-oriented spatial expansion with diversified land use demands; constrained by coastal geography, these areas face more rigid land resource limitations, often resorting to special development models like large-scale land reclamation. However, rapid urbanization has brought multiple challenges including population over-concentration, land resource shortages, and ecological fragility, particularly prominent environmental risks like sea-level rise and coastal erosion. Concurrently, the intertwining issues of regional development imbalance and “urban diseases” make these areas crucial for researching land use efficiency improvement, economic potential realization, and sustainable development optimization. n nTechnology framework n nThe technical framework is detailed in Fig. 2, with the specific process consisting of four key components: (A) deriving the spatiotemporal patterns of LUE in China’s coastal cities through multi-source data fusion and the LitPop method; (B) investigating the urban land maturation and efficiency improvement process by incorporating the “age” concept; (C) assessing whether urban land maturation and efficiency enhancement can support future economic growth requirements based on SSPs scenarios; and (D) examining the underlying mechanisms of urban land maturation and efficiency improvement through POI data analysis using kernel density estimation. n nData n nThe primary data utilized in this study includes the following: The land use data for China can be found at https://www.resdc.cn/DOI/DOI.aspx?DOIID=54. The nighttime lighting dataset (DMSP/OLS and NPP/VIIRS) can be found at https://www.ngdc.noaa.gov/. China City Statistical Yearbook can be found at http://www.stats.gov.cn/. To account for price changes affecting real GDP growth, this study selected 1995 as the base year and used officially released secondary and tertiary GDP indices to deflate nominal GDP from 2000 to 2020, thereby eliminating price factor interference and obtaining the real GDP of secondary and tertiary industries for each city. A gridded dataset comprising value-added of primary, secondary and tertiary industries in China under shared socioeconomic pathways from 2020–2100 can be found at https://doi.org/10.4121/14113706.v2 (Jing et al., 2022). Point of interest data can be found at https://www.amap.com/. All data are included in the manuscript and/or SI, Table S1. n nMethodology n nCalculation of the economic elasticity coefficient (EEC) n nTo assess the balance of economic development and urban land expansion in China’s coastal cities, as well as their degree of synergy, this study employs the economic elasticity coefficient (EEC) to examine the relative changes in economic development and urban land expansion at various urban scales (Huang et al., 2013; Xu et al., 2017). The analysis of the EEC contains six study phases: 1995–2000, 2000–2005, 2005–2010, 2010–2015, 2015–2020, and 1995–2020. The computation formula is as follows: n n$${ rm{EEC}}= frac{ root{n} of{{{GDP}}_{b}-{{GDP}}_{a}}-1}{ root{n} of{{{UL}}_{b}-{{UL}}_{a}}-1}$$ n n(1) n nIn the formula, ({ rm{EEC}} ) development and urban land expansion reflect the relative speed of economic growth and urban land expansion, respectively; ({{GDP}}_{a} ), ({{GDP}}_{b} ) are the sums of the GDP of the secondary and tertiary sectors of different urban scales in the base and target years, respectively; ({{UL}}_{a} ), ({{UL}}_{b} ) are the urban land area in the base and target years for different urban scales, respectively; and (n ) is the study period. The degree of synergy is divided into four categories based on the EEC: economic or urban land contraction (EEC < 0), urban land expansion (0 ≤ EEC < 1), fundamental economic–urban land harmonization (1 ≤ EEC < 5), and economic growth (EEC ≥ 5) (Ouyang and Zhu, 2020; Qiao and Huang, 2021). n nMeasurement of the LUE n nNighttime light images are closely associated with human economic activities and have been widely used in GDP and population spatialization studies at various spatial scales (Wu et al. 2013; Li and Zhou, 2018; Tan et al., 2018; Zheng et al., 2023). Furthermore, for countries and regions with poor quality or missing data, nighttime light data may be a useful supplement to economic indicators (Chen and Nordhaus, 2011). Previous spatialization of socioeconomic data has principally utilized nighttime light data to decompose data proportionally to each pixel. However, the application of this proportional relationship leads to overallocation in suburban areas and underallocation in urban areas (Eberenz et al., 2020). To alleviate this phenomenon, (Zhao et al., 2017) proposed the LitPop method with higher accuracy, which combines nighttime light data with population data to generate lighted population images and then decomposes them proportionally. (Eberenz et al., 2020) further enhanced the LitPop method, leading to its widespread application (Li et al., 2024; Wang and Sun, 2022). Therefore, this paper proposes a framework for measuring LUE based on the LitPop method. The process of measuring LUE includes the following: (i) corrected nighttime light images are cropped based on the urban land area in different years to avoid overestimation of light area caused by the spillover effect of nighttime light data; (ii) based on the improved LitPop method (Eq. (2)) to generate light population images from 1995 to 2020; and (iii) constructed a mapping relationship between the real GDP of each city’s secondary and tertiary industries and light population images; (iv) based on the mapping relationships (Eq. (3)), the real GDP of each city’s secondary and tertiary industries was distributed across a smaller regular grid (1 km × 1 km) to measure LUE in coastal cities. The above LUE measurement framework will help alleviate the saturation effect and more accurately reflect the real economic activities of the city. The formula is as follows: n n$${ rm{LitPop}}= left { begin{array}{ll}{ rm{Pop}} ; { rm{Lit}}=0 { rm{Lit}} , cdot ,{ rm{Pop}} ,{ rm{Lit}} ,> ,0 , & ,{ rm{Pop}} > 0 { rm{Lit}} ; { rm{Pop}}=0 end{array} right.$$ n n(2) n n$${{ rm{LUE}}}_{{LitPop}}= frac{{{GDP}}_{i}}{{{SLP}}_{i}}* {{LitPop}}_{{pixel}}$$ n n(3) n nwhere ({ rm{Lit}} ) is the pixel value of the nighttime light images, ({ rm{Pop}} ) is the pixel value of the population grid data, ({{LitPop}}_{{pixel}} ) is the pixel value of the ({ rm{LitPop}} ) images, ({{SLP}}_{i} ) is the sum of the pixels in the nighttime light population image of each city, ({{GDP}}_{i} ) is the sum of the real GDP of the secondary and tertiary industries in each city, and ({{LUE}}_{{LitPop}} ) is LUE for each pixel. n nDefinition of the ULMEE process n nIn 1985, the scholar Luis Suarez-Villa proposed the urban life cycle theory, which suggests that cities go through four successive stages: birth, growth, maturity, and recession. A city’s life cycle is determined by the superposition of the life cycles of various economic elements that make up the city’s occurrence and development (Suarez-Villa, 1985). This study expands the micro perspective of the urban life cycle theory based on LUE and applies the theory to the development process of LUE. The life cycle of LUE is a conceptual framework defined as the initial, growth, maturity, and recession stage. The progression of these four stages is the ULMEE process (Fig. 3A). The stages are as follows. (i) Initial stage: Land use types are transformed and urban land is developed from scratch, resulting in a relatively low degree of LUE. (ii) Growth stage: As urbanization progresses and economic development continues, a variety of resources are concentrated on urban land, and LUE shows an increasing trend. (iii) Maturity stage: Land resources are fully developed, the industrial system and structure have become more reasonable, and LUE is steadily expanding and gradually approaching its pinnacle. (iv) Recession stage: Because of the long-term rough growth of urban land, some urban land is idle or semi-idle, causing LUE to enter a decline stage. However, when urban planning prioritizes foresight and strategic planning for urban land transformation in advance (successful transformation of urban land), LUE may reach a new stage of development. n nTo further concretize ULMEE, the concept of “age” is introduced, defined as the difference between the year of urban land generation (time of conversion from non-urban land to urban land) and the year 2020. For urban land in 1995, the age of urban land from 1995 to 2020 is recorded as 0, 5, 10, 15, 20, and 25 years (with 5-year intervals). For urban land in 2000, the age from 2000 to 2020 is recorded as 0, 5, 10, 15, and 20 years (with 5-year intervals); and so on. As the age of urban land increases, the functions of urban land gradually become more complex through economic development (industrial structure upgrading, capital accumulation), policy intervention (intensive land management, spatial planning guidance), and transformation and innovation (improvement of production efficiency), which in turn leads to the improvement of LUE (Fig. 3B). The above process is the maturation and efficiency enhancement process. n nIncreasingly efficient and intensive urban land in China’s coastal cities n nUrban land provides the spatial location for economic development, which promotes its expansion of urban land. The rapid economic development of China’s coastal cities has gradually expanded the scale of urban land, making the degree of coordination between economic development and urban land increasingly important. The EEC of different urban scales has obvious stage characteristics, primarily of two types: economic–urban land basic coordination and economic growth (Fig. 4). Economic growth mainly occurs in three stages: 1995–2000 (super cities, type I large cities, type II large cities, and type I small cities); 2005–2010 (type I small cities); and 2010–2015 (super cities, megacities, medium-sized cities, and type I small cities). All except the economic growth type belong to the economic–urban land basic coordination type. With the continuous advancement of urbanization, industrial agglomeration, and industrial structure optimization processes, coastal cities’ economic growth has been considerably faster than urban land expansion in that they are developing more intensively and efficiently (Li et al., 2020). The decoupling of economic growth from urban land is critical for the sustainable development of coastal cities. To achieve more sustainable decoupled development, it is recommended to promote a three-dimensional development model for urban land, encourage functional replacement of existing land and improve efficiency; at the same time, promote the “multi-plan integration” spatial planning system and strengthen the control of land space use. n nExpansion characteristics of urban land n nThe urban land expansion of coastal cities grew from 8351.82 km2 in 1995 to 30843.52 km2 in 2020, with an average annual growth rate of 899.67 km2. Coastal cities’ urban land expansion has distinct phases, with the largest area of new urban land added between 2005 and 2010. During the study period, the top 10 cities in terms of new urban land area were Tianjin, Dongguan, Tangshan, Guangzhou, Shanghai, Ningbo, Weifang, Quanzhou, Dalian, and Nantong, which are all above the level of a large city, further capitalizing on the locational advantages of large cities. The new urban land exhibits radioactive expansion based on a multi-center ring, or gradual expansion from point to surface, with agglomeration characteristics along the coast and roads (SI, Fig. S1A). There is a gap in the urban land area of different urban scales (SI, Fig. S1B), indicating the pattern of type II large cities > medium-sized cities > super cities > megacities > type I large cities > type I small cities. The urban expansion intensity index and the urban expansion intensity difference index were used to disclose the vertical and horizontal expansion process of urban land, and it was found to have clear stages and fluctuation characteristics (SI, Fig. S1C, and S1D). n nCharacteristics analysis of LUE n nThe total economic output of coastal cities increases annually, from 1542.49 billion yuan in 1995 to 17917.18 billion yuan in 2020. The total amount of newly added economic output increased from 1147.94 billion yuan from 1995–2000 to 4340.36 billion yuan during 2015–2020 (SI, Fig. S2A). The total economic output of different urban scales varies as follows: type II large cities > super cities > medium-sized cities > megacities > type I large cities > type I small cities. The LUE of coastal cities is highest in the city center and gradually decreases toward the edge, with distinct spatial heterogeneity within the cities (SI, Fig. S2B). The average LUE of coastal cities continuously increased, from 203 million yuan/km2 in 1995 to 605 million yuan/km2 in 2020, with an average annual increase of 16 million yuan/km2 (SI, Fig. S2D). Comparing the LUE of each city in the six study periods with the average LUE of coastal cities, the LUE of 22, 24, 25, 27, 27, and 22 cities, respectively, is greater than the mean value of coastal cities, indicating spatial differences in LUE and the uneven economic development of coastal cities. Cities of all scales generally follow the rule of increasing LUE annually, although the degree of increase varies (SI, Fig. S2C). n nULMEE process n nFigure 5 depicts the total economic output from urban land of various ages between 1995 and 2020, focusing on the differences between coastal cities and diverse urban scales. When the age of urban land increases (the maturation process), whether in coastal cities or different urban scales, the pattern states that the older the urban land, the higher the total economic output. The total economic output of urban land at different ages in coastal cities, ranked in order of size, was urban land in 1995, 2020, 2010, 2005, 2015, and 2000. This result is directly related to differences in the area size and maturation processes. The total economic output of different urban scales does not entirely follow the order of urban scales. Taking urban land in 1995 as an example, by 2020, super cities, type II large cities, and medium-sized cities are the top three performers in terms of total economic output, while megacities, type I large cities, and type I small cities are the bottom three performers. n nAccording to Table 1, the average LUE of coastal cities increased from 323 million yuan/km2 at the age of 5 years to 695 million yuan/km2 at the age of 15 years, and finally to 922 million yuan/km2 at 25 years. These results verify that the maturation process of urban land was accompanied by efficiency enhancement. The LUE at 0 years continuously increased from 205 million yuan/km2 in 2000 to 531 million yuan/km2 in 2020. This demonstrates that the later urban land is developed, the higher the initial LUE. This pattern is strongly related to the urban land growth model, which states that new urban land develops along with the expansion of the original urban land (Chakraborty et al., 2022). On the one hand, when the urban planning department formulates a plan to optimize the urban functional layout, it prioritizes guiding commercial development in a more compact and intensive direction, which promotes the effective aggregation of resources and improves LUE. In contrast, new urban land is developed along previous urban land and can use the original urban land’s infrastructure (e.g., transportation networks), which can improve the LUE of new urban land. To demonstrate the efficiency enhancement process, taking urban land in 1995 as an example, when urban land was just formed, the LUE was at a low level. As urbanization progressed, various material and economic resources accumulated on urban land, and LUE improved gradually. Subsequently, the urban economy has developed steadily, the industrial structure has become more reasonable, and LUE has gradually reached a historical peak. n nUrban scale directly reflects urban economic development and has an evident impact on economic growth. Disclosing the efficiency improvement process of urban land is critical for designing development strategies at various urban scales. Comparing the polygons of different colors in Fig. 6 (LUE of different urban scales and at different ages), this study finds the following: (i) Different ages: Overall, there is a trend of continuous improvement in LUE as urban land matures, indicating that the maturation proceeds alongside efficiency enhancement. However, it does not completely conform to the shape of “concentric circles.” Taking urban land in 1995 as an example, the LUE of different urban scales at 25 years is lower than that at 20 years. (ii) Different urban scales: Using urban land in 1995 as an example, as of 2020, the LUE followed the pattern of super cities > type I large cities > megacities > type II large cities > medium-sized cities > type I small cities. This indicates that it does not completely follow the pattern in which the larger the urban scale, the higher the LUE. Comparing the red polygons in Fig. 6 (LUE of different urban scales at the same age), taking the super cities as an example, the LUE for 0 years of age increased from 261 million yuan/km2 in 1995 to 498 million yuan/km2 in 2020, indicating that the later the urban land is developed, the higher its initial LUE. n nCan ULMEE meet the demands of future economic growth? n nOn the one hand, the results in Table 1 and Figs. 5 and 6 confirm the ULMEE process; on the other hand, as of 2020, the percentage of urban land under 15 years is as high as 69%, leaving plenty of potential for ULMEE. This result provides a theoretical basis for determining whether the ULMEE process can meet the needs of future economic growth. Shared socioeconomic pathways depict different development models of the future socioeconomic system and provide multiple development scenarios for predicting LUE (O Neill et al., 2014; O Neill et al., 2017). Therefore, this study applies the SSPs method to predict the LUE of various urban scales at different ages and determines whether relying on the ULMEE process can meet the needs of future economic growth in China’s coastal cities. n nA gridded dataset comprising value-added primary, secondary and tertiary industries in China under shared socioeconomic pathways from 2020–2100 is used to predict the LUE of various urban scales at different ages. The LUE at different ages exhibits diverse degrees of growth between 2025 and 2050, with the total LUE shown mostly in the first half of an inverted “U”-shaped curve (SI, Fig. S3). In other word, LUE will be a higher growth degree before 2035, but its growth rate will gradually decline after 2035. From different scenarios of SSPs, in 2025, the difference in LUE predicted by SSP1–SSP5 (sustainability (SSP1), middle of the road (SSP2), regional rivalry (SSP3), inequality (SSP4), and fossil-fueled development (SSP5)) is not significant. As ULMEE progresses, the difference in LUE gradually increases. SSP5 has the highest LUE, whereas SSP3 has the lowest. As of 2050, the LUE of various urban scales at different ages, in descending order, is super cities, type I large cities, megacities, type I small cities, type II large cities, and medium-sized cities. n nAccording to the 2024 Economic and Financial Outlook Report released by the Bank of China Research Institute and the Analysis of China’s Economic Growth Prospects and Dynamics (2015–2050), an expected economic growth target of approximately 5% would meet the demands of China’s future economic development (Tang et al., 2020). A positive future economic development momentum is assumed. Further, based on the total LUE (total economic output of secondary and tertiary industries) predicted by SSPs, it is determined whether relying on the ULMEE process can meet the demands of a 5% urban economic growth rate. According to Fig. 7, except for scenario SSP3, all urban scales can meet economic growth needs before 2035 by relying on ULMEE. The longer it takes to rely on the ULMEE process to meet urban economic growth demands, the later it becomes necessary to add new urban land. After 2040, all urban scales would require additional new urban land to meet economic growth demands. In different scenarios, the time required for different urban scales to expand urban land to meet economic growth needs varies. Type I small cities would rely on the ULMEE process to meet economic growth demands for a longer time and require adding new urban land later. n nRepresentativeness and typicality of coastal cities n nCoastal cities were selected as research objects to explore whether the process of urban land maturation and efficiency enhancement can meet the needs of future economic growth, mainly based on their unique development conditions and demonstration value. In the past 30 years, the population, GDP and urban land area of 51 coastal cities in China accounted for 20.96%, 33.34%, and 55.91% of the national total, respectively. Compared with other cities, the urbanization process of coastal cities faces more complex resource and environmental constraints and development demands (Zhang et al., 2020): on the one hand, coastal cities are usually economically developed, densely populated, and extremely scarce land resources. Urban expansion is often limited by natural barriers such as oceans and mountains, and they have to expand space through special methods such as land reclamation; on the other hand, coastal cities bear the important functions of opening up to the outside world and economic growth engines, and the demand for urban land continues to be strong (Scherner et al., 2013; Huang et al., 2023). This contradiction is more prominent in the context of climate change, and risks such as sea level rise and coastal erosion have further compressed the development space. Simultaneously, coastal cities are often at the forefront of land management system innovation and intensive development. The development models they explored, such as stock renewal, three-dimensional development, and mixed functions, are of great reference significance to other regions. Research on land maturation and efficiency enhancement in coastal cities can reveal the inherent laws of urban development under tight resource constraints and provide practical samples for global implications for developing economies. n nMechanism analysis of ULMEE n nAs a dependable source of spatial data, point of interest (POI) data provides rich spatial location information and is widely used in urban functional area division and POI recommendation (Liu et al., 2024; Long et al., 2024; Xue et al., 2020). The increasing density and agglomeration characteristics of POI data make it possible to reveal the ULMEE process. The correlation coefficient is calculated based on the coastal city POI kernel density estimation (KDE) index and LUE. The result demonstrates an obvious positive correlation between the two, with a correlation coefficient of 0.57** (**0.01 level (two-tailed); the correlation is significant). Therefore, this study selected a plot with an area of 2.10 km2 in Dalian City, as shown in Fig. 8A. Between 2012 and 2020, the plot’s urban land area increased. The mechanism of the ULMEE process was analyzed using the number of POIs on the plot, the KDE index, and the percentage of their classification. The types of POIs on this plot have increased to residential categories in addition to commercial, industrial, public services, science, and education (Fig. 8B). The number of various POIs is gradually increasing, with the total number of POIs increasing from 114 in 2012 to 290 in 2020. Both the types and quantities of POIs show a trend of gradual improvement and enhancement. As Fig. 8C shows, the KDE index in 2012 ranged from 34.17 to 90.53 and was mostly less than 80. In 2020, it ranged between 35.68 and 275.91, with most values exceeding 100. The progressive increase in the POI KDE index supports the optimization and adjustment of the industrial structure, driving the service level and support capacity of urban land (Xue et al., 2020). Simultaneously, a denser POI indicates that the urban infrastructure layout is more complete, creating greater additional capacity and