Skip to content
Frontiers
Search

Original Research

Frontiers in Environmental Science ·

Volume 14 - 2026 | https://doi.org/10.3389/fenvs.2026.1989806

Spatiotemporal associations between ecosystem service value and human activity intensity: a case study of the Chengdu metropolitan area, China

  • Lin Wang1
  • Li Luo1*corresponding author
  • Weijing Zeng1
  • Guanghao Ren1
  • Xiaojun Luo1
  1. School of Tourism Management, Sichuan Polytechnic University, Deyang, Sichuan, China
Article metrics0 viewsView details
PDFCiteMetrics

Abstract

Understanding the spatiotemporal association between ecosystem service value (ESV) and human activity intensity (HAI) is essential for coordinating regional ecological protection with sustainable socioeconomic development. Using geospatial data for the Chengdu Metropolitan Area from 2000 to 2020, this study estimated ESV and HAI with an ESV accounting model and an HAI assessment model, and examined their spatiotemporal patterns and associations using bivariate spatial autocorrelation and coupling coordination degree models. The results showed that: (1) Temporally, total ESV initially increased and subsequently decreased, with an overall decline of CNY 37 million. Spatially, ESV decreased outward from the Longmen Mountains in the west and Longquan Mountains in the central area, with high-value zones occurring mainly along the Tuojiang and Minjiang rivers. (2) HAI increased continuously. Low values were concentrated in the western Longmen Mountains and eastern hilly areas, whereas high values were centered in Chengdu’s main urban area and expanded into surrounding areas. (3) HAI and ESV exhibited a significant and increasingly strong negative spatial correlation. High-low clusters were concentrated in Chengdu’s main urban area and adjacent built-up areas, low-high clusters remained stable in the densely vegetated western Longmen Mountains, and low-low clusters were widespread across the eastern hilly agricultural areas. (4) Overall coupling coordination remained low and was dominated by low coordination. Areas of severe and moderate imbalance continued to expand, forming a contiguous, outward-expanding pattern in the central region. By revealing the spatiotemporal association between ESV and HAI at the metropolitan-area scale, this study provides a theoretical basis for constructing ecological security patterns and sustainable planning in similarly urbanizing regions.

Keywords

1 Introduction

Ecosystem services are the life-supporting goods and services that people obtain directly or indirectly from natural ecosystems. They include provisioning, regulating, supporting, and cultural services (Costanza et al., 1998; Daily et al., 2000; Fisher et al., 2009) and underpin human survival and development (Carpenter et al., 2009). Since the beginning of this century, however, sustained population growth and rapid economic development have accelerated natural-resource depletion and placed unprecedented pressure on natural ecosystems (Aguilera et al., 2020). Approximately 60% of global ecosystem services are currently diminishing, a situation largely attributed to the fact that human activities have affected approximately 75% of the land and 66% of the oceans (Vitousek et al., 1997; Halpern et al., 2008; Coscieme et al., 2014). Intensive, large-scale human activities alter ecosystem structure and function (Lazarina et al., 2019; Perry and Alvarez-Filip, 2019), weaken the supply of ecosystem services (Moreno-Llorca et al., 2020; Han et al., 2022), and ultimately cause ecosystem degradation (Nematollahi et al., 2020), biodiversity loss (Veldhuis et al., 2019), and increased pollution (Adeyeri et al., 2022). At the same time, constraints imposed by ecosystem services on human activities are becoming increasingly prominent and affect progress toward sustainable development goals (Su et al., 2012). Unraveling the complex linkages between human activities and ecosystem services has thus become a major research agenda for regional sustainable development.

Quantifying ESV makes explicit the products and services provided by ecosystems, thereby increasing public awareness of ecosystem conservation and supporting regional sustainable development (Xie et al., 2017). ESV is commonly assessed using biophysical valuation (Finnveden and Moberg, 2005), emergy valuation (Coscieme et al., 2014), and monetary valuation methods (Costanza et al., 1998; Xie et al., 2017). Costanza et al. (1998) initially proposed the monetary valuation approach, which estimates ESV by constructing a per-unit-area equivalent-value table from land-use data. Its suitability for large-scale and long-term comparative studies has led to widespread adoption. Xie et al. (2017) subsequently developed a localized equivalent-value table based on the biogeographic characteristics and biomass parameters of China’s terrestrial ecosystems. This approach provided an important methodological basis for the present study.

HAI is a composite indicator describing the extent to which human production and daily activities affect the terrestrial surface to meet subsistence and development needs (Xu et al., 2016; Liu et al., 2023). It is widely used to quantify human intervention in natural ecosystems (Sun et al., 2020). Early studies characterized HAI using individual indicators such as nighttime lights (Zhao et al., 2022), population density (Cunningham and Beazley, 2018), or land use (Roth et al., 2016). A single indicator, however, cannot adequately capture regional differences and therefore provides limited assessment accuracy (Su et al., 2012). Other studies have developed multidimensional indicator systems. For example, Guo et al. used 27 indicators covering agriculture, industry, the service sector, social development, and ecological engineering, and reported an increase in HAI during the study period (Guo et al., 2024). Correa Ayram et al. reached a similar conclusion using seven indicators (Correa Ayram et al., 2020). Nevertheless, composite multi-indicator systems face inconsistent selection criteria, limited availability of long time-series data, disagreement over weighting methods, and restricted cross-regional comparability (Duan and Luo, 2021). To address these limitations, this study constructed an HAI index from three representative and readily available data sources: population density, nighttime lights, and land use. This approach uses a concise indicator system and readily comparable data; it can clearly characterize the spatial differentiation of HAI and support regional ecological assessment and policy decision-making.

As research on human activities and ecosystem services has advanced, their interrelationship has become a central concern. Existing studies have examined this relationship at national (Wu et al., 2023), provincial (An et al., 2024), urban-agglomeration (Gong et al., 2025), and city scales (Yang and Zhu, 2024). They have commonly used spatial-analysis software such as GeoDa (Anselin et al., 2006) and ArcGIS (Dong et al., 2026) to conduct spatial autocorrelation (Yang and Song, 2021), regression (Let et al., 2024), and coupling coordination analyses (Huang et al., 2023). Quintas-Soriano et al. (2016), for example, investigated the relationship between land-use change and ecosystem services in Spain and showed that different human activities exerted diverse effects on these services. Ai et al. (2024) used bivariate local Moran’s I and spatial autocorrelation analysis to examine the relationship between HAI and ESV in the Chengdu-Chongqing Twin-City Economic Circle. Areas with higher ESV were associated with lower HAI, whereas areas with lower ESV were affected by higher human disturbance intensity (HDI). Using convergent cross-mapping and uncertainty analysis, Song et al. (2025) found that population, gross domestic product, and the normalized difference vegetation index significantly affected ESV in Sanya. Studies across these scales consistently identify human activity as a major driver and show that intensive development substantially weakens ecosystem-service provision (Zhang et al., 2023).

In summary, previous studies have focused primarily on countries, large urban agglomerations, or individual cities. Theoretical and empirical research on the metropolitan area, a spatial scale intermediate between the individual city and the urban agglomeration, remains relatively scarce, yet it is precisely the metropolitan area that constitutes the most active spatial unit of urbanization and the most concentrated site of human-land conflicts (Ouyang et al., 2021). Moreover, studies of the spatial association between HAI and ESV have rarely examined their coupling and coordination (Chen et al., 2023). To address these gaps, this study developed an integrated framework combining an ESV flow gain-loss model, bivariate spatial autocorrelation, and coupling coordination analysis. It thereby provides the first metropolitan-scale assessment of the spatiotemporal association between ESV and HAI.

The Chengdu Metropolitan Area lies within the core ecological barrier of the upper Yangtze River. It is a major growth pole of the Chengdu-Chongqing Twin-City Economic Circle and one of the most urbanized and densely populated economic regions in western China (Meng et al., 2022). While supporting rapid regional socioeconomic development, the area also provides critical ecological functions, including soil retention, climate regulation, water conservation, and biodiversity maintenance (Hou et al., 2025). Since the beginning of the 21st century, urban and industrial expansion has intensified the human footprint, reshaping land cover and ecological systems. The resulting shrinkage of agricultural and wooded lands has raised degradation risks and intensified resource competition. How to harmonize development with nature thus poses a major sustainability challenge.

Against this background, the Chengdu Metropolitan Area was selected as the study area. Using geospatial and socioeconomic data for 2000, 2010, and 2020, this study aimed to: (1) estimate ESV with the equivalent-factor method and construct an ESV flow gain-loss model based on a land-use transition matrix to identify the sources and destinations of ESV gains and losses; (2) develop an integrated HAI assessment model combining nighttime light intensity, population density, and land-use conversion coefficients to characterize the spatiotemporal evolution of HAI; and (3) apply bivariate spatial autocorrelation and coupling coordination degree models at a 1 km grid scale to identify spatial associations between HAI and ESV and evaluate their coupling coordination. The findings of this study offer a methodological reference for coordinating ecosystem services in rapidly urbanizing regions. They also provide a scientific underpinning for ecological conservation and sustainable development in the Chengdu Metropolitan Area.

2 Materials and methods

2.1 Study area

The Chengdu Metropolitan Area is located in central Sichuan Province (29°24′–31°42′N, 102°49′–105°45′E). As the first national-level metropolitan area in central and western China, it comprises Chengdu, Deyang, Meishan, and Ziyang and covers approximately 33,100 km2 (Figure 1). The area occupies a core section of the ecological barrier in the upper Yangtze River. The Longmen Mountains in the west are important for biodiversity conservation and water retention (Xia et al., 2022), whereas the Chengdu Plain in the central region is a major grain-producing area. The Minjiang and Tuojiang rivers traverse the area, supporting soil and water conservation on the Chengdu Plain and maintaining ecological security in the upper Yangtze River. The metropolitan area is also the principal engine of the Chengdu-Chongqing Twin-City Economic Circle and the socioeconomic center of Sichuan Province, with population density and economic activity among the highest in western China’s urban agglomerations (Meng et al., 2022). From 2000 to 2020, its urbanization rate increased rapidly from approximately 40%–70.3%, marking a critical phase of urbanization and industrial expansion. The region is therefore both a focal area of conflict between economic development and ecological protection and an appropriate case for examining metropolitan-scale associations between ESV and HAI.

Location of the study area.
Figure 1. Location of the study area.

2.2 Data sources

The geospatial data used in this study include land use data, nighttime light data, and population density data (Table 1). All raster data were obtained for three years—2000, 2010, and 2020—and were uniformly projected to WGS_1984_UTM_Zone_48 N (Figure 2). The administrative boundary data of the study area were obtained from the National Geomatics Center of China (www.ngcc.cn/ngcc). The grain yield data for each city were sourced from the Sichuan Statistical Yearbook, and the commodity price data were obtained from the National Compilation of Cost and Benefit Information of Agricultural Products.

TABLE 1. Data description.

Data typeData nameYearsSpatial resolutionData source
Land use dataChina multi-period land use remote sensing monitoring dataset2000, 2010, 202030 m,1 kmResource and environment science and data center, Chinese academy of sciences (https://www.resdc.cn/DOI/DOI.aspx?DOIID=201)
Night light dataA prolonged artificial nighttime-light dataset of China (1984–2020)2000, 2010, 20201 kmNational Tibetan plateau data center (https://data.tpdc.ac.cn/zh-hans/data/e755f1ba-9cd1-4e43-98ca-cd081b5a0b3e)
Population dataPopulation density/Unconstrained individual countries 2000–2020 UN adjusted (1 km resolution)2000, 2010, 20201 kmWorldPop (https://hub.worldpop.org/geodata/listing?id=77)
Visualization of the geospatial datasets.
Figure 2. Visualization of the geospatial datasets.

2.3 Methods

2.3.1 ESV estimation

ESV was estimated using the equivalent-factor method proposed by Xie et al. (2015). The value of an equivalent factor was determined from the grain-sown area and grain prices in the study area. One standard ESV equivalent equals one-seventh of the annual economic return from grain production on one hm2 of cropland with the national average yield (Xie et al., 2017; Long et al., 2022). To account for price fluctuations, the mean grain price from 2000 to 2020 was used (Jin et al., 2023; Ai et al., 2024). Because the correction coefficient directly affects estimation accuracy, the per-unit-area ESV equivalent table was adjusted using the biomass factor for farmland ecosystems in Sichuan Province (Xie et al., 2005; Jiang et al., 2023; Zhao et al., 2026). The resulting ESV equivalent-factor value for the Chengdu Metropolitan Area was CNY 2,423.9 h m-2, and the corresponding equivalent factor table was established (Table 2). Following previous studies that did not assign ESV to built-up land (Li et al., 2023; Jia and Chen, 2025), its ESV was set to zero. ESV was calculated as follows:

E S V = ∑ i = 1 n A i × V C i (1)

E S V f = ∑ i = 1 n A i × V C f i (2)

TABLE 2. ESV coefficients per unit area in the Chengdu Metropolitan Area (CNY hm-2).

First-level serviceSecond-level serviceCroplandForest landGrasslandWater bodiesBuilt-up landUnused land
Provisioning servicesFood production2,690.53702.93921.081939.120.000.00
Raw-material production605.981,599.771,357.38557.500.000.00
Water supply−3,175.31824.13751.4120,094.130.000.00
Regulating servicesGas regulation2,157.275,259.864,775.081866.400.0048.48
Climate regulation1,139.2315,755.3512,628.525,550.730.000.00
Environmental purification339.354,678.134,169.1113,452.650.00242.39
Hydrological regulation3,635.8511,489.299,259.30247,819.540.0072.72
Supporting servicesSoil retention1,260.436,423.345,817.362,254.230.0048.48
Nutrient cycling387.82484.78436.30169.670.000.00
Biodiversity412.065,841.605,284.106,180.950.0048.48
Cultural servicesAesthetic landscape193.912,569.332,326.944,581.170.0024.24

In Equation 1, ESV is the total ecosystem service value, Ai is the area of land-use type i, VCi is the ecosystem service value coefficient for land-use type i, and n is the number of land-use types. In Equation 2, ESVf is the total value of ecosystem service f, and VCfi is the coefficient of service f for land-use type i.

2.3.2 ESV flow gain-loss model

An ESV flow gain-loss model was constructed from the land-use transition matrix. In addition to quantifying changes in ESV, the model identifies the sources and destinations of gains and losses caused by conversions among land-use types during each period (Guo et al., 2022). Conversion areas were first calculated from the transition matrix and were then used to estimate ESV gains and losses as follows:

P i j = V j − V i × A i j (3)

In Equation 3, Pij is the change in ESV caused by conversion from land-use type i to type j; Vi and Vj are the ESV coefficients for land-use types i and j, respectively; and Aij is the area converted from type i to type j.

2.3.3 HAI assessment model

HAI reflects the extent to which human production and daily activities develop and utilize the terrestrial surface within a region (Xu et al., 2016). Given the diversity and complexity of human activities, nighttime lights, population density, and land-use data were combined to construct the HAI model. Following established HAI index models (Xu et al., 2016; Cheng et al., 2022; Zhu et al., 2022), HAI in the Chengdu Metropolitan Area was calculated as follows:

X j = X − X min / X max − X min (4)

H A I = a × N T L + b × P O P + c × H A I L S (5)

H A I L S = S C L E / S (6)

S C L E = ∑ i = 1 n S L i × C L i (7)

In the equations, X *j* is the normalized value of indicator j; X is the original value of indicator i; X min and X max are the minimum and maximum values of the indicator across the study area, respectively. The normalization was performed using the min–max method to rescale each indicator to [0, 1], and no translation (offset) operation was applied to any dataset. All zero values are the direct results of the calculation (Ma et al., 2026). HAI is human activity intensity; NTL, POP, and HAILS are the normalized nighttime light index, population density, and human activity intensity of the land surface, respectively; S is the total study-area extent; S *CLE* is the equivalent area of built-up land; CL *i* is the built-up-land conversion coefficient for land-use type i; SL *i* is the area of land-use type i; and n is the number of land-use types. Based on previous studies (Xu et al., 2016; Zhang et al., 2025), the weights of indicators a, b, and c were determined by more than 40 experts using a Likert five-point scale, yielding weights of 0.3, 0.3, and 0.4, respectively. The results were examined using Kendall’s tau-distance, with a p-value <0.001, indicating significant consistency in the experts’ ratings and thus confirming the reliability of the results (Kaya et al., 2026). The built-up-land conversion coefficients for cropland, forest land, grassland, water bodies, built-up land, and unused land were 0.2, 0, 0.067, 0, 1, and 0, respectively.

2.3.4 Bivariate spatial autocorrelation model

Bivariate spatial autocorrelation reveals the spatial relationship between two variables and identifies the clustering of geographic attributes (Li et al., 2022). A bivariate spatial autocorrelation model was used to calculate the global Moran’s I and identify spatial clusters of ESV and HAI as follows:

I x y = n ∑ i = 1 n ∑ j = 1 n w i j · ∑ i = 1 n ∑ j = 1 n w i j x i − x ¯ y j − y ¯ ∑ i = 1 n x i − x ¯ 2 ∑ i = 1 n y i − y ¯ 2 (8)

I x y i = x i − x ¯ σ x ∑ j = 1 n w i j y j − y ¯ σ y (9)

In the equations, I *xy* and I *i* *xy* are the bivariate global and local Moran’s I, respectively; n is the number of spatial units (1 km grid cells); x *i* and y *j* are the corresponding values at unit i and unit j, respectively; x̄ and ȳ are their means; σ *x* and σ *y* are their standard deviations; and w *ij* is the row-standardized Queen contiguity spatial weight matrix.

2.3.5 Coupling coordination analysis

Coupling degree measures interactions among systems or system components, whereas coupling coordination degree is widely used to evaluate and interpret coordinated interdependence between coupled systems (Ouyang et al., 2025). Here, the coupling coordination degree model was used to assess interactions between ESV and HAI in the Chengdu Metropolitan Area. The coupling degree and coupling coordination degree were calculated as follows:

g y = 1 − H A I (10)

f x = E S V i − E S V min / E S V max − E S V min (11)

C = 2 × f x · g y f x + g y 2 1 2 (12)

T = α f x + β g y (13)

D = C · T (14)

In the equations, C is the coupling degree between ESV and HAI; f(x) is the standardized ESV; g(y) is the reverse-processed HAI; T is the comprehensive coordination index of ESV and HAI; α and β are their respective weights; and D is the coupling coordination degree. D ranges from 0 to 1, with values closer to one indicating a higher level of coordinated development between the two systems (Gao et al., 2025). Because ESV and HAI were considered equally important, α and β were both set to 0.5.

3 Results

3.1 Spatiotemporal changes in ESV

3.1.1 Temporal changes in ESV

Total ESV, absolute change, and percentage change in the Chengdu Metropolitan Area were calculated for 2000–2020 (Table 3). Total ESV decreased slightly by CNY 37 million, or 0.05%, with an initial increase followed by a decline. ESV increased from CNY 78.348 billion in 2000 to CNY 78.583 billion in 2010, an increase of CNY 235 million (0.30%). It then decreased by CNY 272 million (0.35%) to CNY 78.311 billion in 2020. Contributions by land-use type ranked as forest land > cropland > water bodies > grassland > unused land. ESV provided by water bodies increased by CNY 1.407 billion (9.31%). Forest-land ESV initially increased and then declined, but showed an overall increase of CNY 397 million (1.16%). ESV from cropland and grassland decreased continuously by CNY 1.538 billion (6.78%) and CNY 307 million (4.80%), respectively. Unused land accounted for a negligible proportion of the study area and made a correspondingly negligible contribution to ESV.

TABLE 3. ESV and changes by land-use type in the Chengdu Metropolitan Area, 2000–2020 (unit: CNY 100 million).

IndicatorCroplandForest landGrasslandWater bodiesUnused landTotal
2000226.67341.763.99151.120783.48
2010217.52348.161.42158.750.04785.83
2020211.29345.6760.92165.190.04783.11
Change, 2000–2010−9.156.4−2.577.630.042.35
Change, 2010–2020−6.23−2.43−0.56.440−2.72
Change, 2000–2020−15.383.97−3.0714.070.04−0.37
Change rate, 2000–2010 (%)−4.041.87−4.025.05—0.3
Change rate, 2010–2020 (%)−2.86−0.7−0.814.060−0.35
Change rate, 2000–2020 (%)−6.781.16−4.89.31—−0.05

Among the four first-level ecosystem-service categories (Figure 3), values ranked as regulating > supporting > provisioning > cultural services. Regulating services made the largest contribution, with values of CNY 57.395, 57.643, and 57.547 billion in 2000, 2010, and 2020, respectively, accounting for 73.26%, 73.35%, and 73.49% of total ESV. Cultural services contributed the least, with corresponding values of CNY 2.573, 2.583, and 2.567 billion and shares of 3.28%, 3.29%, and 3.28%. Provisioning-service value increased continuously by a total of CNY 88 million. Regulating- and cultural-service values initially increased and then decreased, resulting in a net increase of CNY 152 million and a net decrease of CNY 6 million, respectively. Supporting-service value declined continuously by a total of CNY 271 million.

Values and changes in first-level ecosystem-service categories in the Chengdu Metropolitan Area, 2000–2020.
Figure 3. Values and changes in first-level ecosystem-service categories in the Chengdu Metropolitan Area, 2000–2020.

3.1.2 Spatial changes in ESV

A 1 km × 1 km grid was established, and the natural breaks method in GIS was used to classify ESV into five categories for each year: low, medium-low, medium, relatively high, and high (Figure 4). ESV displayed marked spatial heterogeneity. Values were highest in the western Longmen Mountains and central Longquan Mountains and decreased progressively toward surrounding areas. High-value zones occurred primarily along the Tuojiang and Minjiang rivers. Relatively high values were concentrated in mountainous areas with dense vegetation and favorable ecological conditions, including western Chengdu, northwestern Deyang, and southwestern Meishan. Medium values occurred mainly in the eastern and southern hilly areas, where cropland was the dominant land-use type. Low and medium-low values formed extensive patches in the principal urban areas, where built-up land was widespread and ESV was lowest. This pattern was most pronounced on the Chengdu Plain.

Spatial distribution of ESV in the Chengdu Metropolitan Area, 2000–2020.
Figure 4. Spatial distribution of ESV in the Chengdu Metropolitan Area, 2000–2020.

3.2 ESV gains and losses

ESV gains and losses during 2000–2020 were estimated from the land-use transition matrix and the ESV provided by each land-use type (Figure 5). Total gains were CNY 5.340 billion, total losses were CNY 5.377 billion, and the net loss was CNY 37 million, indicating a broadly stable overall ESV flow. Conversions from cropland to forest land and water bodies generated CNY 3.857 billion, accounting for 72.17% of total gains. Conversion from forest land to water bodies generated a further CNY 608 million, or 11.38%. The largest loss, CNY 1.484 billion (27.58% of total losses), resulted from conversion of cropland to built-up land. This finding reflects rapid urbanization and extensive expansion of built-up land over the 20-year period. Conversion from forest land to cropland caused the second-largest loss, CNY 1.463 billion (27.19%), indicating continued pressure on forest land in areas where human-land conflicts remained acute. Together, these two conversion pathways generated CNY 2.947 billion in losses, or 54.77% of total ESV losses, and represented the principal pathways through which human activities affected ESV.

ESV gains and losses in the Chengdu Metropolitan Area, 2000–2020.
Figure 5. ESV gains and losses in the Chengdu Metropolitan Area, 2000–2020.

3.3 Spatiotemporal changes in HAI

3.3.1 Temporal changes in HAI

HAI was calculated using the assessment model and classified by the natural breaks method into five categories: low, medium-low, medium, relatively high, and high. The area share of each category was then calculated for 2000–2020 (Figure 6). The shares of low- and medium-low-HAI areas declined continuously by 1.25 and 6.72 percentage points, respectively. The shares of medium- and high-HAI areas increased continuously by 4.67 and 3.70 percentage points, respectively. The share of relatively high-HAI areas initially decreased and then increased, with an overall change of −0.39 percentage points. Low and medium-low HAI remained dominant throughout the study period, but the increasing share of areas with medium or higher HAI indicated a sustained intensification of human activities.

Area shares of HAI categories in the Chengdu Metropolitan Area, 2000–2020.
Figure 6. Area shares of HAI categories in the Chengdu Metropolitan Area, 2000–2020.

3.3.2 Spatial changes in HAI

From 2000 to 2020, HAI showed a strongly centralized spatial pattern (Figure 7). High-value zones were concentrated in Chengdu’s main urban area and gradually expanded, particularly during 2010–2020. As the capital of Sichuan Province, Chengdu has a dense population, a strong economy, and a markedly higher urbanization rate than surrounding cities, resulting in intensive human activity. A notable high-HAI corridor also formed northeastward from Chengdu toward Deyang. This industrially concentrated corridor has close intercity commuting links and a strong foundation for economic cooperation. It connects the metropolitan core, accommodates industrial spillover from Chengdu, and serves as an important growth axis for coordinated development in the northern metropolitan area. Additional high-HAI zones occurred as discrete patches in built-up areas around Chengdu, including Yanjiang, Dongpo, Shifang, Jianyang, and Renshou. Relatively high HAI was concentrated around cities, reflecting the outward spillover of human activities during rapid urbanization. Medium-low values occurred mainly in the eastern hilly areas, which were dominated by agriculture and had low levels of industrialization and relatively slow urbanization. Low HAI was concentrated in the western Longmen Mountains, where high elevations, dense forest cover, and sparse populations limited human disturbance. These low-impact areas remained spatially stable.

Spatial distribution of HAI in the Chengdu Metropolitan Area, 2000–2020.
Figure 7. Spatial distribution of HAI in the Chengdu Metropolitan Area, 2000–2020.

3.4 Spatial differentiation of the association between HAI and ESV

Moran’s I scatterplots were generated in GeoDa (Figure 8). The bivariate spatial autocorrelation indices between HAI and ESV were −0.281, −0.283, and −0.291 in 2000, 2010, and 2020, respectively (p < 0.05). These values indicated a negative spatial correlation: areas with higher HAI generally had lower ESV. The progressive decline in Moran’s I showed that this negative relationship strengthened over time and shifted from a relatively dispersed to a more clustered spatial pattern.

Moran’s I for HAI and ESV in the Chengdu metropolitan area, 2000–2020.
Figure 8. Moran’s I for HAI and ESV in the Chengdu metropolitan area, 2000–2020.

The rank order of the five cluster types remained stable across 2000–2020 (Figure 9), with non-significant cells the largest group (38.44%–41.09%) and low–low the most extensive significant type. The significant types together contracted from 61.56% to 58.91%, a 2.65 percentage point decline matched by the corresponding expansion of the non-significant class. Within the significant types the change was asymmetric: high–low and high–high clusters contracted by 13.67% and 15.41%, whereas low–low clusters grew by 2.10%, lifting their share of the four significant types from 34.86% to 37.20%. Nearly all of this change occurred before 2010, after which every type stabilized.

Area proportions of spatial clusters of HAI and ESV in the Chengdu Metropolitan Area.
Figure 9. Area proportions of spatial clusters of HAI and ESV in the Chengdu Metropolitan Area.

In spatial terms (Figure 10), low–low clusters occurred primarily in the eastern hilly areas, where limited built-up land and extensive cropland resulted in both low HAI and medium-low ESV. Low–high clusters were concentrated in the densely forested mountains of the west and remained broadly stable over the study period. High–low clusters were concentrated in Chengdu’s main urban area and its surroundings, tracking the outward spread of built-up land. High–high clusters occurred mainly along the Minjiang and Tuojiang rivers; although urban development and agricultural activities were also concentrated on the plains along these rivers, the exceptionally high ESV of water bodies dominated the grid-cell averages and produced a high–high pattern.

Spatial clusters of HAI and ESV in the Chengdu Metropolitan Area, 2000–2020.
Figure 10. Spatial clusters of HAI and ESV in the Chengdu Metropolitan Area, 2000–2020.

3.5 Coupling coordination between ESV and HAI

The coupling coordination degree was calculated and classified into six categories following previous studies (Zhu et al., 2022): severe imbalance [0, 0.2), moderate imbalance [0.2, 0.4), low coordination [0.4, 0.5), primary coordination [0.5, 0.6), good coordination [0.6, 0.8), and high-quality coordination [0.8, 1]. Area shares were calculated for each category from 2000 to 2020 (Figure 11). Overall coupling coordination was limited. Low coordination covered almost half of the study area, followed by good coordination, whereas high-quality coordination had the smallest share. The category distribution remained relatively stable over the 20-year period. Changes in the six categories were 1.56, 4.21, −5.15, −0.17, −0.54, and 0.09 percentage points, respectively. Areas of severe and moderate imbalance expanded continuously, whereas the shares of low, primary, and good coordination declined. Thus, continued urbanization intensified human disturbance of ecosystems and affected ecosystem-service functions, while overall coupling coordination showed no substantial improvement.

Area shares of ESV-HAI coupling coordination categories in the Chengdu Metropolitan Area, 2000–2020.
Figure 11. Area shares of ESV-HAI coupling coordination categories in the Chengdu Metropolitan Area, 2000–2020.

Spatially, coupling coordination exhibited a pattern of lower values in the center and higher values toward the periphery (Figure 12). High-quality coordination was concentrated around the Heilongtan Reservoir in eastern Meishan, where extensive water bodies coincided with limited human activity. Good coordination was concentrated in the western Longmen Mountains, where dense vegetation, low HAI, and high ESV formed contiguous patches. Areas of primary coordination occurred mainly in mixed forest-cropland zones surrounding areas of good coordination. Low coordination was widespread across the central Chengdu Plain. Severe imbalance formed clusters in Chengdu’s main urban area and surrounding built-up areas, where high HAI coincided with markedly reduced ESV. During the 20-year period, the most pronounced changes occurred in the central study area. Areas of severe imbalance spread outward from the main urban area, whereas areas of moderate imbalance expanded across agricultural land on the Chengdu Plain. These changes indicated that urbanization increasingly disturbed ESV in the core development zone and intensified the conflict between the two coupled systems.

Spatial distribution of HAI-ESV coupling coordination categories in the Chengdu Metropolitan Area, 2000–2020.
Figure 12. Spatial distribution of HAI-ESV coupling coordination categories in the Chengdu Metropolitan Area, 2000–2020.

4 Discussion

4.1 Changes in ESV and their driving mechanisms

From 2000 to 2020, total ESV in the Chengdu Metropolitan Area initially increased and then decreased, producing a net decline of CNY 37 million (0.05%). This trajectory reflected the competition between ecosystem conservation and rapid urbanization. During 2000–2010, implementation of the Grain for Green Program substantially increased forest area and raised forest-land ESV by CNY 640 million. This gain offset ESV losses caused by urban development and produced a net increase in total ESV. During 2010–2020, ecological policies such as the Grain for Green Program remained in place, but ESV contributed by forest land declined by CNY 243 million. Urbanization also accelerated, and the conversion of cropland to built-up land intensified, jointly reducing total ESV. The Grain for Green Program exerted a positive ecological influence during both periods, but urbanization was considerably more intensive in the second decade. Built-up land expanded only moderately during 2000–2010. After the Chengdu-Chongqing Economic Circle became a national strategy during 2010–2020, however, rapid industrial and population agglomeration expanded the spatial footprint of human activities. Accelerated occupation of cropland and forest land by built-up land caused ESV losses to exceed gains. These findings suggest that, in rapidly urbanizing regions, intensive human activity can progressively offset the benefits of ecosystem-conservation policies. Limiting ecosystem disturbance, enforcing urban growth boundaries, and promoting a more sustainable form of urbanization are therefore necessary.

4.2 Mechanisms underlying the negative spatial association between ESV and HAI

Bivariate spatial autocorrelation analysis yielded Moran’s I values of −0.281, −0.283, and −0.291 for 2000, 2010, and 2020, respectively. All values indicated significant negative spatial correlations, and the relationship strengthened over time. This pattern was consistent with findings for other rapidly urbanizing regions, including the Pearl River Delta (Huang et al., 2023) and Shandong Peninsula urban agglomerations (Gong et al., 2025), suggesting that a negative spatial association between HAI and ESV is common in such settings. High HAI in Chengdu’s main urban area and surrounding built-up areas formed stable high-low clusters. Restricted development in the western Longmen Mountains maintained low-high clusters, whereas cropland-dominated eastern hills exhibited low-low clustering. Directional land-use conversion was the direct driver of the strengthening negative association. Conversion of cropland to built-up land caused an irreversible ESV loss of CNY 1.484 billion. Conversion of forest land to cropland caused a further loss of CNY 1.463 billion, indicating that deforestation and cultivation continued to offset and disrupt the restoration benefits of the Grain for Green Program. Together, these conversions accounted for 54.77% of total ESV losses and represented the principal pathways through which HAI affected ESV. Landscape fragmentation may have further strengthened the negative association. As high-HAI areas expanded outward from Chengdu’s urban core, contiguous ecological patches were divided into isolated fragments, weakening supporting and regulating services even where land-use conversion did not occur. The increasing absolute value of Moran’s I reflected the cumulative effects of these interacting mechanisms.

4.3 Spatial differentiation and evolution of coupling coordination

Coupling coordination was generally low and was dominated by low coordination. Areas of severe and moderate imbalance expanded continuously during the 20-year period. The spatial pattern of lower coordination in the center and higher coordination toward the periphery further supported the negative association between ESV and HAI. Coupling conflicts were most pronounced in Chengdu’s main urban area and surrounding built-up areas, whereas the western Longmen Mountains ecological conservation area and the Heilongtan Reservoir in the south showed better coordination. Severe imbalance spread outward from Chengdu’s urban core, while moderate imbalance expanded across agricultural land on the Chengdu Plain. Ecosystem disturbance during urbanization was therefore transmitted beyond built-up areas through the expansion of construction land and the occupation and compensation of cropland. As the first national-level metropolitan area in central and western China, the Chengdu Metropolitan Area continues to urbanize rapidly. Industrial spillover from the core has increased human disturbance of surrounding ecosystems, progressively reducing agricultural and ecological spaces with previously high coupling coordination and lowering coordination across the region.

Coupling coordination also declined along the industrial corridor between Chengdu and Deyang. Intensive development progressively weakened ecosystem-service functions in this area. Without advance designation and restoration of ecological space, continued coordinated corridor development could create a new concentration of coupling conflicts. This evolution indicates that regional planning should extend ecological protection beyond existing ecological-function zones. It should also anticipate and regulate the transmission of impacts from development expansion to prevent ecological conflicts from spreading into surrounding areas.

4.4 Policy implications and study limitations

The results of this study can provide a reference for territorial spatial planning and ecological conservation in the Chengdu Metropolitan Area. First, areas with high ecosystem service value (ESV) are highly concentrated in two major ecological barriers—the Longmen Mountains in the west and the Longquan Mountains in the centre—forming a spatial gradient in which ESV declines outward from these “Two Mountains” as the core. This spatial pattern corresponds closely to the “three zones and two mountains” spatial structure outlined in the territorial spatial plan of the Chengdu Metropolitan Area. Accordingly, this study suggests that ecological conservation and restoration projects targeting the Two Mountains be linked with the results of the ESV–HAI coupling coordination analysis, and that differentiated spatial regulation strategies be implemented along the gradient of declining ESV from the Two Mountains toward the periphery, with ecological hotspot areas accorded priority in spatial planning hierarchies (Karadeniz et al., 2026). In the plain transition zone between the Two Mountains, namely, the Chengdu Plain, regulation should be guided by the principle of “structural optimization and prevention of encroachment”. In this zone, ESV remains at a moderate level while HAI is relatively high, and cultivated land is interspersed with construction land, making it the area where ESV is most strongly and negatively affected by HAI within the study region. Enhancing ecosystem service value per unit area through high-standard farmland construction and the redevelopment of existing construction land is therefore recommended. Second, the high-HAI corridor extending from the main urban area of Chengdu to Deyang indicates that ecological risks are transmitted across administrative boundaries during the process of metropolitan-area integration. Building on the eco-environmental collaborative governance mechanisms already established among Chengdu, Deyang, Meishan and Ziyang, an integrated ecological compensation and cross-regional collaborative governance system should be established for the metropolitan area. Taking the high-HAI corridor as the spatial unit, cross-city collaborative regulation pilots can be carried out, with coordinated control of construction-land increments applied at high-HAI nodes along the corridor. An ecological compensation scheme grounded in the ESV–HAI spatial relationship should also be developed, under which compensation funds flow from the central urban areas and other high-HAI zones to high-ESV areas such as the Longmen and Longquan Mountains, and compensation standards are tied to verified ESV maintenance outcomes, thereby putting cross-regional collaborative governance into practice. Finally, the red lines for permanent basic farmland and for ecological conservation should be strictly enforced to curb the conversion of cultivated land into construction land, particularly in the high-quality farmland areas of the Chengdu Plain.

This study has several limitations. The indicator weights in the HAI assessment system (a, b, and c = 0.3, 0.3, and 0.4) were adopted from previous studies, yet the optimal weights may differ across stages of urbanization. Moreover, this study focused on the spatial associations between ESV and HAI and did not establish their underlying causal relationship or characterize their non-linear responses. In the cross-period comparison spanning 2000–2020, the inflation component embedded in the mean nominal grain price used to derive the ESV equivalent factor may amplify or attenuate the absolute change in ESV; future research could therefore adopt CPI-deflated per-unit-area ESV coefficients to correct for this effect. In addition, future studies could use multiscale geographically weighted regression (MGWR), random forests, and other machine-learning methods to quantify the contributions of individual drivers to ESV change and the spatial heterogeneity of those contributions. Extending such modeling to finer spatial resolutions and linking it to operational decision-support platforms, as illustrated by recent building-scale applications that combine geo-information modeling with machine learning (Kaya et al., 2025), would further help translate these findings into management action.

5 Conclusion

Based on geospatial data for the Chengdu Metropolitan Area from 2000 to 2020, this study estimated ecosystem service value (ESV) and human activity intensity (HAI). Bivariate spatial autocorrelation and coupling coordination degree models were then used to examine their spatiotemporal evolution and association patterns. The main conclusions are as follows.

  • Total ESV declined slightly and exhibited marked structural differentiation. ESV initially increased and then decreased, with an overall loss of CNY 37 million. Contributions varied among land-use types: ESV from forest land and water bodies increased by CNY 397 million and CNY 1.407 billion, respectively, whereas ESV from cropland and grassland decreased by CNY 1.538 billion and CNY 307 million. Regulating services remained dominant, provisioning services were the only category to increase continuously, and supporting services declined throughout the study period.
  • ESV gains originated primarily from a limited set of land-use conversions. Total gains were CNY 5.340 billion, total losses were CNY 5.377 billion, and the resulting net loss was CNY 37 million. Conversions from cropland to forest land and water bodies accounted for 72.17% of total gains. Losses caused by conversions from cropland to built-up land and from forest land to cropland jointly accounted for 54.77% of total losses and represented the two principal pathways of ESV decline.
  • ESV and HAI showed a significant negative spatial correlation that strengthened continuously. Bivariate global Moran’s I declined from −0.281 to −0.291. Spatial clusters exhibited a clear gradient: high-low clusters were concentrated in Chengdu’s main urban area and surrounding built-up areas, low-high clusters remained stable in the densely vegetated western Longmen Mountains, and low-low clusters were concentrated in the eastern hilly agricultural areas.
  • Overall coupling coordination remained low, and the core imbalanced areas expanded contiguously outward. Low coordination was the dominant category. From 2000 to 2020, areas of severe and moderate imbalance expanded continuously, whereas areas of primary and good coordination contracted. Spatially, coordination was lower in the center and higher toward the periphery. Severe imbalance spread outward from Chengdu’s main urban area, while moderate imbalance expanded across agricultural areas of the Chengdu Plain, producing contiguous zones of declining coordination.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. During the preparation of this manuscript, the authors used generative AI (ChatGPT) solely for language polishing and proofreading to improve readability and fluency of the text. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the final version of the 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.

Footnotes

Edited by: Maomao Zhang, Huazhong University of Science and Technology, China

Reviewed by: A.Y. Kaya, Firat University, Türkiye

Qian Cheng, Shenyang Agricultural University, China

Statements

Conflict of interest

The 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.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Ministry of Education of China Humanities and Social Sciences Youth Foundation (Grant No. 23YJCZH157) and the High-level Talent Research Project of Sichuan Polytechnic University (Grant No. XJ2026HT-18). The funders had no role in study design, data collection, analysis, decision to publish, or preparation of the manuscript.

Data availability

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

Author contributions

LW: Writing – review and editing, Conceptualization, Writing – original draft, Methodology. LL: Funding acquisition, Conceptualization, Writing – review and editing. WZ: Data curation, Formal Analysis, Writing – review and editing. GR: Writing – review and editing, Software, Visualization. XL: Writing – review and editing, Visualization.

References

  1. Adeyeri O. E., Laux P., Ishola K. A., Zhou W., Balogun I. A., Adeyewa Z. D. (2022). Homogenising meteorological variables: impact on trends and associated climate indices. J. Hydrology 607, 127585. doi:10.1016/j.jhydrol.2022.127585
  2. Aguilera M. A., Tapia J., Gallardo C., Núñez P., Varas-Belemmi K. (2020). Loss of coastal ecosystem spatial connectivity and services by urbanization: Natural-To-Urban integration for Bay management. J. Environ. Manag. 276, 111297. doi:10.1016/j.jenvman.2020.111297
  3. Ai M., Chen X., Yu Q. (2024). Spatial correlation analysis between human disturbance intensity (HDI) and ecosystem services value (ESV) in the chengdu-chongqing urban agglomeration. Ecol. Indic. 158, 111555. doi:10.1016/j.ecolind.2024.111555
  4. An Q., Yuan X., Zhang X., Yang Y., Chen J., An J. (2024). Spatio-temporal interaction and constraint effects between ecosystem services and human activity intensity in Shaanxi Province, China. Ecol. Indic. 160, 111937. doi:10.1016/j.ecolind.2024.111937
  5. Anselin L., Syabri I., Kho Y. (2006). GeoDa: an introduction to spatial data analysis. Geogr. Anal. 38, 5 – 22. doi:10.1111/j.0016-7363.2005.00671.x
  6. Carpenter S. R., Mooney H. A., Agard J., Capistrano D., DeFries R. S., Díaz S. (2009). Science for managing ecosystem services: beyond the millennium ecosystem assessment. Proc. Natl. Acad. Sci. U.S.A. 106, 1305 – 1312. doi:10.1073/pnas.0808772106
  7. Chen S., Liu X., Yang L., Zhu Z. (2023). Variations in ecosystem service value and its driving factors in the nanjing metropolitan area of China. Forests 14, 113. doi:10.3390/f14010113
  8. Cheng Y., Liu H., Chen D., Liu H. (2022). Human activity intensity and its spatial-temporal evolution in China’s border areas. Land 11, 1089. doi:10.3390/land11071089
  9. Correa Ayram C. A., Etter A., Díaz-Timoté J., Rodríguez Buriticá S., Ramírez W., Corzo G. (2020). Spatiotemporal evaluation of the human footprint in Colombia: four decades of anthropic impact in highly biodiverse ecosystems. Ecol. Indic. 117, 106630. doi:10.1016/j.ecolind.2020.106630
  10. Coscieme L., Pulselli F. M., Marchettini N., Sutton P. C., Anderson S., Sweeney S. (2014). Emergy and ecosystem services: a national biogeographical assessment. Ecosyst. Serv. 7, 152 – 159. doi:10.1016/j.ecoser.2013.11.003
  11. Costanza R., d’Arge R., De Groot R., Farber S., Grasso M., Hannon B. (1998). The value of the world’s ecosystem services and natural capital. Ecol. Econ. 25, 3 – 15. doi:10.1016/S0921-8009(98)00020-2
  12. Cunningham C., Beazley K. F. (2018). Changes in human population density and protected areas in terrestrial global biodiversity hotspots, 1995–2015. Land 7, 136. doi:10.3390/land7040136
  13. Daily G. C., Söderqvist T., Aniyar S., Arrow K., Dasgupta P., Ehrlich P. R. (2000). The value of nature and the nature of value. Science 289, 395 – 396. doi:10.1126/science.289.5478.395
  14. Dong P., Zuo R., Gong J., Tang M., He C., Li J. (2026). Scenario simulation of tea expansion impacts in karst ecosystems. Ecosyst. Health Sustain 12, 0521. doi:10.34133/ehs.0521
  15. Duan Q., Luo L. (2021). Summary and prospect of spatialization method of human activity intensity: taking the Qinghai-Tibet Plateau as an example. J. Glaciol. Geocryol. 43, 1582 – 1593. doi:10.7522/j.issn.1000-0240.2021.0039
  16. Finnveden G., Moberg Å. (2005). Environmental systems analysis tools – an overview. J. Clean. Prod. 13, 1165 – 1173. doi:10.1016/j.jclepro.2004.06.004
  17. Fisher B., Turner R. K., Morling P. (2009). Defining and classifying ecosystem services for decision making. Ecol. Econ. 68, 643 – 653. doi:10.1016/j.ecolecon.2008.09.014
  18. Gao Y., Zhang P., Xu Y., Li Z., Liu K. (2025). A study on the spatiotemporal characteristics of the Xi’an metropolitan area based on the coupling and coordination of ecosystem services and human well-being. Land 14, 500. doi:10.3390/land14030500
  19. Gong D., Dong D., Du H., Zhou Y., Fu S., Fujioka Y. (2025). Spatiotemporal coupling mechanisms and driving forces of ecosystem services and human activity from a multidimensional perspective. Int. J. Digital Earth 18, 2512061. doi:10.1080/17538947.2025.2512061
  20. Guo S., Wu C., Wang Y., Qiu G., Zhu D., Niu Q. (2022). Threshold effect of ecosystem services in response to climate change, human activity and landscape pattern in the upper and middle yellow river of China. Ecol. Indic. 136, 108603. doi:10.1016/j.ecolind.2022.108603
  21. Guo X., Yang Y., Zhang Y., Kalantari M., Sun J., Sun W. (2024). The spatiotemporal correlation between human activity intensity and the evolution of ecosystem service value in the songnen plain, China. Land 13, 1158. doi:10.3390/land13081158
  22. Halpern B. S., Walbridge S., Selkoe K. A., Kappel C. V., Micheli F., D’Agrosa C. (2008). A global map of human impact on marine ecosystems. Science 319, 948 – 952. doi:10.1126/science.1149345
  23. Han Z., Cui S., Yan X., Liu C., Li X., Zhong J. (2022). Guiding sustainable urban development via a multi-level ecological framework integrating natural and social indicators. Ecol. Indic. 141, 109142. doi:10.1016/j.ecolind.2022.109142
  24. Hou L., Hu H., Liu T., Ma C. (2025). Ecological security pattern construction for carbon sink capacity enhancement: the case of chengdu metropolitan area. Sustainability 17, 4483. doi:10.3390/su17104483
  25. Huang Z., Chen Y., Zheng Z., Wu Z. (2023). Spatiotemporal coupling analysis between human footprint and ecosystem service value in the highly urbanized pearl river Delta urban agglomeration, China. Ecol. Indic. 148, 110033. doi:10.1016/j.ecolind.2023.110033
  26. Jia H., Chen H. (2025). Spatiotemporal dynamics and optimization management of ecosystem service flows in the yangtze river Delta urban agglomeration, China. Sustainability 17, 4727. doi:10.3390/su17104727
  27. Jiang Z., Gan X., Liu J., Bi X., Kang A., Zhou B. (2023). Landscape ecological risk assessment and zoning control based on ecosystem service value: taking Sichuan Province as an example. Appl. Sci. 13, 12103. doi:10.3390/app132212103
  28. Jin T., Chen Y., Shu B., Gao M., Qiu J. (2023). Spatiotemporal evolution of ecosystem service value and topographic gradient effect in the Da-Xiao liangshan Mountains in Sichuan Province, China. J. Mt. Sci. 20, 2344 – 2357. doi:10.1007/s11629-023-7986-9
  29. Karadeniz E., Sekeroglu A., Kaya A. Y., Zhang M., Adiguzel F. (2026). A spatial planning approach for the sustainability of provisioning ecosystem services with economic value: the case of Türkiye. J. Environ. Manag. 399, 128652. doi:10.1016/j.jenvman.2026.128652
  30. Kaya A. Y., Sajjad M., Dedekorkut-Howes A. (2025). Downscaling urban disaster resilience: toward building-scale geo-information modeling and machine learning-assisted resilience knowledge systems. Trans. GIS 29, e70153. doi:10.1111/tgis.70153
  31. Kaya A. Y., Imren E., Giyik C., Karadeniz E., Adıgüzel F., Özel H. B. (2026). Assessing the spatial suitability and adequacy of emergency assembly areas for urban disaster resilience using GIS and the best-worst method (BWM): the case of Malatya, Türkiye. Appl. Sci. 16, 5206. doi:10.3390/app16115206
  32. Lazarina M., Devalez J., Neokosmidis L., Sgardelis S. P., Kallimanis A. S., Tscheulin T. (2019). Moderate fire severity is best for the diversity of Most of the pollinator guilds in mediterranean pine forests. Ecology 100, e02615. doi:10.1002/ecy.2615
  33. Let M., Pal S., Let M., Ghosh R., Debanshi S. (2024). Anthropogenic impact on ecosystem service value of urban blue space in old malda municipality of eastern India. Environ. Monit. Assess. 196, 976. doi:10.1007/s10661-024-13046-7
  34. Li L., Fan Z., Feng W., Chen Y., Qin K. (2022). Coupling coordination degree spatial analysis and driving factor between socio-economic and eco-environment in northern China. Ecol. Indic. 135, 108555. doi:10.1016/j.ecolind.2022.108555
  35. Li G., Wu Z., Cheng G., Yuan Y., He Y., Wang H. (2023). Research on the spatiotemporal dynamic relationship between human activity intensity and ecosystem service value in the three gorges reservoir area. Sustainability 15, 15322. doi:10.3390/su152115322
  36. Liu H., Fan J., Zhou K., Xu X., Zhang H., Guo R. (2023). Assessing the dynamics of human activity intensity and its natural and socioeconomic determinants in Qinghai–Tibet Plateau. Geogr. Sustain. 4, 294 – 304. doi:10.1016/j.geosus.2023.05.003
  37. Long X., Lin H., An X., Chen S., Qi S., Zhang M. (2022). Evaluation and analysis of ecosystem service value based on land use/cover change in Dongting Lake wetland. Ecol. Indic. 136, 108619. doi:10.1016/j.ecolind.2022.108619
  38. Ma M., Wang Y., Du Y., Wu X., Song Y. (2026). Impacts of human activities on ecosystem services in watersheds: evidence from the taihu Lake basin, China. Sustainability 18, 7129. doi:10.3390/su18147129
  39. Meng B., Wang X., Zhang Z., Huang P. (2022). Spatio-temporal pattern and driving force evolution of cultivated land occupied by urban expansion in the chengdu metropolitan area. Land 11, 1458. doi:10.3390/land11091458
  40. Moreno-Llorca R., Vaz A. S., Herrero J., Millares A., Bonet-García F. J., Alcaraz-Segura D. (2020). Multi-scale evolution of ecosystem services’ supply in sierra Nevada (spain): an assessment over the last half-century. Ecosyst. Serv. 46, 101204. doi:10.1016/j.ecoser.2020.101204
  41. Nematollahi S., Fakheran S., Kienast F., Jafari A. (2020). Application of InVEST habitat quality module in spatially vulnerability assessment of natural habitats (case study: Chaharmahal and Bakhtiari province, Iran). Environ. Monit. Assess. 192, 487. doi:10.1007/s10661-020-08460-6
  42. Ouyang X., Tang L., Wei X., Li Y. (2021). Spatial interaction between urbanization and ecosystem services in Chinese urban agglomerations. Land Use Policy 109, 105587. doi:10.1016/j.landusepol.2021.105587
  43. Ouyang J., Zhang P., Yu H., Zhang N., Liu Y. (2025). Analysis on coupling coordination degree between livelihood strategy for peasant households and land use behavior in ecological conservation Areas—A case study of the chang-zhu-tan ecological greenheart area. Sustainability 17, 3996. doi:10.3390/su17093996
  44. Perry C. T., Alvarez-Filip L. (2019). Changing geo-ecological functions of coral reefs in the anthropocene. Funct. Ecol. 33, 976 – 988. doi:10.1111/1365-2435.13247
  45. Quintas-Soriano C., Castro A. J., Castro H., García-Llorente M. (2016). Impacts of land use change on ecosystem services and implications for human well-being in Spanish drylands. Land Use Policy 54, 534 – 548. doi:10.1016/j.landusepol.2016.03.011
  46. Roth D., Moreno-Sanchez R., Torres-Rojo J. M., Moreno-Sanchez F. (2016). Estimation of human induced disturbance of the environment associated with 2002, 2008 and 2013 land use/cover patterns in Mexico. Appl. Geogr. 66, 22 – 34. doi:10.1016/j.apgeog.2015.11.009
  47. Song P., Yang Q., Wu W., Su T., Jiang Y., Lu J. (2025). Assessing and predicting the value of ecosystem services in sanya city, Hainan island, China. Ecol. Indic. 170, 112967. doi:10.1016/j.ecolind.2024.112967
  48. Su C., Fu B.-J., He C.-S., Lü Y.-H. (2012). Variation of ecosystem services and human activities: a case study in the yanhe watershed of China. Acta Oecol. 44, 46 – 57. doi:10.1016/j.actao.2011.11.006
  49. Sun Y., Liu S., Shi F., An Y., Li M., Liu Y. (2020). Spatio-temporal variations and coupling of human activity intensity and ecosystem services based on the four-quadrant model on the Qinghai-Tibet Plateau. Sci. Total Environ. 743, 140721. doi:10.1016/j.scitotenv.2020.140721
  50. Veldhuis M. P., Ritchie M. E., Ogutu J. O., Morrison T. A., Beale C. M., Estes A. B. (2019). Cross-boundary human impacts compromise the serengeti-mara ecosystem. Science 363, 1424 – 1428. doi:10.1126/science.aav0564
  51. Vitousek P. M., Mooney H. A., Lubchenco J., Melillo J. M. (1997). Human domination of Earth’s ecosystems. Science 277, 494 – 499. doi:10.1126/science.277.5325.494
  52. Wu Y., Wu Y., Gao B., Li C., Zheng K., Li C. (2023). Spatial relationship between ecosystem service value and human activity intensity in the chengdu-chongqing urban agglomeration. Res. Soil Water Conservation 30, 173 – 182. (In Chinese). doi:10.13869/j.cnki.rswc.2023.01.018
  53. Xia X., Zhang Y., Shi X., Chen J., Rao T. (2022). Simulation of dynamic urban growth boundary combining urban vitality and ecological networks: a case study in chengdu metropolitan area. Land 11, 1793. doi:10.3390/land11101793
  54. Xie G., Xiao Y., Zhen L. (2005). Study on ecosystem services value of food production in China. Chin. J. Eco-Agriculture 13, 10 – 13.
  55. Xie G., Zhang C., Zhang L., Chen W., Li S. (2015). Improvement of the evaluation method for ecosystem service value based on per unit area. J. Nat. Resour. 30, 1243 – 1254. (In Chinese). doi:10.11849/zrzyxb.2015.08.001
  56. Xie G., Zhang C., Zhen L., Zhang L. (2017). Dynamic changes in the value of China’s ecosystem services. Ecosyst. Serv. 26, 146 – 154. doi:10.1016/j.ecoser.2017.06.010
  57. Xu Y., Xu X., Tang Q. (2016). Human activity intensity of land surface: concept, methods and application in China. J. Geogr. Sci. 26, 1349 – 1361. doi:10.1007/s11442-016-1331-y
  58. Yang Y., Song G. (2021). Human disturbance changes based on spatiotemporal heterogeneity of regional ecological vulnerability: a case study of qiqihaer city, northwestern songnen plain, China. J. Clean. Prod. 291, 125262. doi:10.1016/j.jclepro.2020.125262
  59. Yang Y., Zhu X. (2024). Distribution and change characteristics of ecosystem services in highly urbanized areas along gradients of human activity intensity: a case study of shenzhen city, China. Sustainability 16, 2543. doi:10.3390/su16062543
  60. Zhang X., Zheng Z., Sun S., Wen Y., Chen H. (2023). Study on the driving factors of ecosystem service value under the dual influence of natural environment and human activities. J. Clean. Prod. 420, 138408. doi:10.1016/j.jclepro.2023.138408
  61. Zhang J., Xue D., Li Z., Chen Y., Yan Q. (2025). Spatiotemporal correlation analysis between ecosystem service value and human activity intensity in the chang-zhu-tan urban agglomeration. J. Central South Univ. For. Technol. 45, 159 – 172. (In Chinese). doi:10.14067/j.cnki.1673-923x.2025.10.016
  62. Zhao Y., Qu Z., Zhang Y., Ao Y., Han L., Kang S. (2022). Effects of human activity intensity on habitat quality based on nighttime light remote sensing: a case study of northern Shaanxi, China. Sci. Total Environ. 851, 158037. doi:10.1016/j.scitotenv.2022.158037
  63. Zhao H., Yang W., Zhang Y., Luo C., Li X., Zhang Y. (2026). Assessment of ecosystem service value and analysis of driving factors in the giant panda national park in China. Land 15, 302. doi:10.3390/land15020302
  64. Zhu W., Muhammad A., Han M., Li Y., Kong X., Kong F. (2022). Spatial distribution and aggregation of human-environment coordination and optimal paths in the yellow river Delta, China. Ecol. Indic. 143, 109380. doi:10.1016/j.ecolind.2022.109380

Citation

Wang, L., et al. (2026). Spatiotemporal associations between ecosystem service value and human activity intensity: a case study of the Chengdu metropolitan area, China. Frontiers in Environmental Science, Volume 14. https://doi.org/
Received
Accepted
Published
Volume
14 - 2026

Copyright

Copyright © 2026 Wang, Luo, Zeng, Ren and Luo.

License: https://creativecommons.org/licenses/by/4.0/

*Correspondence: Li Luo

Article metrics

Views
0
Downloads
0
Citations
0
Shares
0

Related articles

Original ResearchHeterogeneous effects of climate change on real GDP per capita: evidence from selected MENA countries and TürkiyeHüseyin YılmazOriginal ResearchPromoting chiral metalaxyl degradation by bioaugmentation with a host-guest soil microbial consortiumChao Xu, Haibin Gu, Jiandong Sheng, Pingan Jiang, Ze WangOriginal ResearchRisk assessment of water treatment chemical supply disruptions in China using a synthetic water quality dataset and the Best-Worst MethodPan Zhao, Yuhui Lei