Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Next Article in Journal
Quantifying the Driving Forces of Water Conservation Using Geodetector with Optimized Parameters: A Case Study of the Yiluo River Basin
Previous Article in Journal
Additionality in Theoretical von Thünenian Models of Deforestation and Conservation Payments
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Land
Volume 14
Issue 2
10.3390/land14020273
land-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Spatiotemporal Relationship Between Landscape Pattern and Ecosystem Service Connectivity in Wetland Environment: Evidence from Yellow River Delta, China

by
Chaozhi Hao
1,2,3,
Shuyao Wu
4,5,*,
Wenjie Cheng
2,3,5,
Mengna Chen
2,3,5,
Yaofa Ren
6,
Xiaoqing Chang
2,3,5 and
Linbo Zhang
2,3,5,*
1
School of Environmental Science and Engineering, Shandong University, Qingdao 266237, China
2
Humanities Laboratory for the Theory and Mechanism Research on the Value Realizing of the Yellow River Ecosystem Products, Shandong University, Qingdao 266237, China
3
Center for Yellow River Ecosystem Products, Shandong University, Qingdao 266237, China
4
College of Geography and Remote Sensing, Hohai University, Nanjing 211000, China
5
Qingdao Institute of Humanities and Social Sciences, Shandong University, Qingdao 266237, China
6
Modelling, Evidence and Policy Research Group, School of Natural and Environmental Sciences, Newcastle University, Newcastle upon Tyne NE1 7RU, UK
*
Authors to whom correspondence should be addressed.
Land 2025, 14(2), 273; https://doi.org/10.3390/land14020273
Submission received: 24 December 2024 / Revised: 26 January 2025 / Accepted: 26 January 2025 / Published: 28 January 2025
Browse Figures
Versions Notes

Abstract

:
Ecosystem service connectivity (ESC) is the spatial and functional links among and within ecosystems that support unimpeded service flows, and that could play an important role in ecosystem stability enhancement and regional land planning. Understanding the relationships between landscape patterns and ESC is crucial to achieving certain sustainable development goals, but it has not yet received an adequate amount of attention. Here, we evaluated the changes and connectivity of five key types of ecosystem services from 2000 to 2020 and analyzed the correlations and spatial aggregations between the ESCs and landscape metrics in the wetlands of the Yellow River Delta, China. Various research methods, such as the InVEST model, spatial autocorrelation analysis, Spearman’s correlation, and self-organizing map, were applied. The results showed that water yield, water purification, and habitat quality showed high connectivity, but the overall ESC declined along with the restoration of the wetland area. Meanwhile, the High-High ESC cluster of water yield, water purification, and habitat quality had similar spatial distribution patterns, and both were dominated by tidal flats. Moreover, the ESC and landscape metrics showed significant correlations and spatial heterogeneity, and a potential connectivity between water yield and habitat quality was also found. These findings can assist decision-makers in developing effective ecosystem management strategies and provide a reference for future research on ecosystem service connectivity.

1. Introduction

Ecosystem services (ESs) have become increasingly important in sustainable environmental management and policy-making, as they provide a framework to understand and quantify the benefits that natural systems provide to human well-being [1,2]. However, the effective management of ESs requires a thorough understanding of not only individual services, but also their interconnections and spatial relations across landscapes [3,4]. Ecosystem service connectivity (ESC) refers to the spatial and functional links among and within ecosystems that support unimpeded service flows and provides a new perspective on understanding how these services interact within and across different contexts [4,5]. Water-related ESs provide a clear example for understanding the connectivity and movement of water, which affects the functional connectivity of different ESs such as water supply and water purification [6]. The ESC is particularly crucial as human activities and land-use changes continue to fragment natural landscapes, potentially disrupting the flow of ESs [7]. Understanding these connections is essential for configuring landscapes and planning spaces to maintain the resilience of social-ecological systems and ensure sustainable ES provision [8,9,10].
The simultaneous consideration of multiple ES relationships during landscape management has been sufficiently highlighted recently [11,12]. The interactions between different services can be characterized by trade-offs or synergies [13,14]. For instance, grain production in the same space may affect the water yield, soil retention, and carbon storage in southwest China [15]. However, these methods are not directly based on the potential ecological process theories to assess the interaction and may ignore how ESs occurring in one area may directly or indirectly affect ESs in other areas [16]. Current research on ecological connectivity is mainly focused on identifying potential spatial links across the landscape context using approaches like least-cost path analysis [17], circuit theory [18], and network flow models [19]. Nevertheless, these study results do not represent functional links between different ESs [20]. Only limited empirical evidence on ESC was found to demonstrate functional connectivity in regional [4] and agricultural landscape contexts [5], respectively. In general, although ESC with multiple elements and spatial scales is attracting increasing attention, direct research is still quite lacking. Thus, new approaches that link the ESC and landscape structure might be needed to better map and understand the spatial nature of ESs.
Landscape metrics are utilized to assess landscape patterns at three distinct levels: patch, class, and landscape. Landscape pattern affects the supply of various ESs; both landscape composition and configuration play a critical role in influencing ecological processes through their effects on the exchange of materials and energy flow [21,22,23]. Several approaches have been developed to analyze the relationship between landscape patterns and ESs such as correlation analysis [24], Bayesian belief networks [12], and the geographical weighted regression model [10]. Furthermore, the effect of landscape patterns on ESs is increasingly recognized to encompass more than just isolated services; it has been gradually broadening to include trade-offs, synergies, and interactions among bundles [15,25,26]. Nevertheless, existing research that examined both landscape connectivity and ESs often centered on a specific service (e.g., food or pollination) or considered multiple ESs simultaneously (e.g., crop production, water flow regulation, and landscape esthetics) [4,7]. The long-term spatiotemporal effects of the landscape patterns on ESC remain poorly understood, especially for the proposed metrics of landscape connectivity in relation to the role of ESC.
The Yellow River Delta (YRD) represents the largest and youngest estuarine wetland ecosystem in China. The unique sedimentary conditions at the mouth of the Yellow River have resulted in a distinctive wetland landscape within the YRD such as the intertwined distribution of mudflats, river channels, and depressions [27]. The favorable geographical location and complex hydrological conditions make the YRD a key region for ES provisioning, and it is emerging as a global hotspot for biodiversity, offering habitats to a variety of rare and endangered bird species [28]. However, the wetland ecosystem within the YRD is particularly delicate, with a poor ability to resist external disturbances and self-regulation [29]. On the one hand, the wetlands, which have developed over a relatively brief period, face sand transport and coastal processes, leading to the alternating erosion and accretion of coastal wetlands [30]. On the other hand, the rapid urbanization in the YRD has led to human disturbances to wetlands such as reclamation, sewage disposal, dam construction, etc. [31], and the tension between ecological preservation and economic growth has initiated a process of fragmentation and rapid transformation in the wetland landscape pattern [32]. These factors have seriously affected the stability of the wetland ecosystem and the sustainable provision of ESs in the YRD.
The proposal of ESC provides a perspective for incorporating ESs into decision-making, particularly in the development of ecosystem conservation and restoration plans aimed at enhancing stability. However, existing research on the ESC remains insufficient, and the relationship between landscape patterns and ESC is not well-understood, particularly in the context of fragile wetland ecosystems. To address the above-mentioned research gaps, this study aimed to reveal the spatial relationship between the wetland landscape patterns and ESC of the wetlands in the YRD as well as provide guidance for regional landscape management and land planning. The specific objectives of this study were as follows: (1) to explore the ESC in a wetland environment; (2) to understand the relationship between the landscape metrics and ESC; (3) to investigate the potential connectivity between different ESs.

2. Materials and Methods

2.1. Research Framework

The research flowchart of the study is shown in Figure 1. We first quantified five key ESs and analyzed the spatial distribution characteristics of the landscape metrics and ESC across time through a spatial autocorrelation analysis. Next, we identified the link between them through the Spearman correlation and bivariate local spatial autocorrelation methods. Finally, we analyzed the possible connectivity between different ESs through bundle identification in a wetland environment.

2.2. Study Area

The YRD is situated in the northeastern section of Shandong Province, China. To facilitate analysis and data collection, we used the administrative boundary of Dongying, the core city of the YRD, as the study area (36°55′~38°12′ N, 118°07′~119°10′ E) (Figure 2). This coverage encompasses Dongying, Hekou, Kenli, Guangrao, and Lijin Counties, spanning an overall land area of 8243.26 km2, and had a population of 2.19 million as of 2020. The landforms are dominated by micromorphological landscapes in the area, with a gently sloping topography extending the direction of the Yellow River, tilting from southwest to northeast. The Yellow River traverses regions like the Loess Plateau, transporting significant quantities of sediment that accumulate at its estuary, leading to the continual development of new land around the delta each year. The annual runoff, sand transport, and erosion rate of the Yellow River Basin were 4.70 × 1010 m3, 2.40 × 108 t, and 33.06% in 2020. The average annual precipitation, temperature, and potential evapotranspiration in the YRD are 556.2 mm, 12.8 °C, and 1885.0 mm, respectively. Peak temperatures typically occur between June and July, and the frost-free growing season lasts around 216 days. The soil types are mainly tidal soil and saline soil, and the vegetation distribution types are homogenous in the area, limited by soil and topography.

2.3. Data Source

Multi-source geographic data were used for wetland pattern changes and ES assessments in the YRD (Table 1). The LUCC dataset, featuring a resolution of 30 m from 2000 to 2020, was sourced from the Institute of Geographic Science and Natural Resources Research, China. This dataset underwent completion through visual interpretation of the Landsat remote sensing image, achieving an accuracy of over 91.2% for secondary land use classes [33]. The second-level wetland classification in the YRD comprised paddy fields, reservoir pits, rivers and canals, lakes, tidal flats, beach land, and marshland [32]. Shallow seas situated within administrative boundaries were excluded from the wetland classification due to data availability and the heterogeneity of the methods. The DEM was derived from the Geospatial Data Cloud, and meteorological information (e.g., precipitation, evapotranspiration, and air temperature) were gathered from the National Earth System Science Data Center. The soil dataset for this study was derived from HWSD v1.2, published by the FAO. To standardize the spatial resolution across the various datasets, all data were resampled to the WGS_1984_UTM_Zone_50N projection coordinate system at a resolution of 30 m.

2.4. Ecosystem Service Assessment

The widely applied InVEST tool was used to assess five key wetland ESs including water yield (WY, mm), carbon storage (CS, Mg), soil retention (SR, t), water purification (WP, kg), and habitat quality (HQ, none) [34]. These ESs have been widely explored in wetlands studies [28,35,36,37,38] and are representative for reflecting the ESC. Appendix A.1 contains the definition and specific equation for the ESs.

2.5. Landscape Pattern Metrics Calculation

Landscape metrics serve to illustrate the traits of landscape composition and structural arrangement, aiding in the interpretation of the interplay between ecological processes and spatial patterns. By consulting related research studies [3,4,12,39,40,41,42,43,44] and understanding the ecological meaning of different landscape metrics, eight landscape metrics that focused on reflecting landscape connectivity were calculated including the number of patches (NP, #), patch density (PD, #/100 ha), largest path index (LPI, %), connectance (CONNECT, %), patch cohesion index (COHESION, none), effective mesh size (MESH, ha), Shannon’s diversity index (SHDI, none), and aggregation index (AI, %), where “#” denotes the number of patches. We used Fragstats 4.2 software to quantify the landscape metrics at the landscape level and implemented spatial visualization using the moving window method. Moving window is a spatial analysis tool for landscape metrics, which moves a fixed-size window, calculates the landscape metrics in each window, and assigns them to the center grid of the window.

2.6. Spatial Autocorrelation Analysis

ESC was characterized as regions in the landscape where one ES supply area affected the provisioning of another through fundamental ecological processes [4]. Simply put, there was a functional interdependence between ES high supply areas, expressed as an aggregation of sets of neighboring high supply space grid cells [45]. ESs with strong aggregation typically exhibit higher connectivity and can imply less impeded service flows [46,47]. Spatial autocorrelation is an essential method for assessing the spatial aggregation of variables of interest. It proficiently demonstrates the extent of similarity in attribute values among pixels situated in neighboring or contiguous regional units. This study utilized univariate and bivariate spatial autocorrelation analyses (1 km × 1 km grid scale). For spatial analysis and mapping, the research employed GeoDa 1.22.0.8 software alongside ArcGIS version 10.8.
Univariate spatial autocorrelation analysis was used to evaluate the ESC by examining the similarity and agglomeration of ESs in adjacent spaces. The Global autocorrelation measures the overall connectivity of wetland ESs in the YRD, while local autocorrelation reflects the aggregation effect. If Moran’s I > 0, it indicates that the relevant components of the ES show a positive spatial correlation, with values nearing 1 signifying a stronger connection. In contrast, when Moran’s I < 0, the relevant components of ES demonstrate a negative spatial correlation [47]. A significant positive correlation allows for the further analysis of clustering characteristics among various ESs. For specific formulas, see Appendix A.2.
Bivariate analysis was employed to examine the spatial autocorrelation between the ESC and landscape metrics, thereby unveiling the degree of spatial correlation of the attribute variables at local and regional levels [48,49]. In the bivariate LISA clustering diagram, the High-High (H-H) area represented a positive relationship between the values of landscape metrics and ESC; on the other hand, the Low-High (L-H) and High-Low (H-L) areas demonstrated a negative correlation between these two variables. Additionally, the Low-Low (L-L) area showed that the values of both landscape metrics and ESC were low while remaining positively correlated [50].

2.7. Spearman Correlation Analysis

The Spearman rank correlation coefficient serves as a widely utilized quantitative and nonparametric approach for evaluating correlations among variables according to their ranks. This method does not impose stringent assumptions regarding data distribution, such as normality or homogeneity of variance, and is capable of identifying nonlinear relationships [51]. To investigate the numerical associations between landscape metrics and ESC at the grid scale, we employed Spearman’s rank correlation analysis. The calculations for this coefficient were performed using IBM SPSS Statistics v.27 software and visualized in Origin 2022.

2.8. Ecosystem Services Bundles Identification

A region in the landscape that generates various types of ESs may exhibit connections between similar ES types located in different areas, associations among diverse ES types within the same area, or relations between dissimilar ES types across various locations [4]. We employed the bundle method to detect the potential connectivity of different ESs in a wetland environment. ES bundles denote collections of services that frequently occur together spatially, implying that these services are interrelated, although the underlying mechanisms remain unclear [52]. We employed the self-organizing mapping (SOM) technique to recognize ES bundles, using an unsupervised neural network approach that could cluster each grid based on the similarity of ES co-occurrence in spatial contexts [14,15,53]. Initially, we assessed the cell size plasticity and spatial attributes of wetlands in the YRD, settling on 1 km × 1 km grid cells as the suitable scale through several comparisons to better capture the spatial interactions of various ESs. The “kohonen” package available in R version 4.3.3 was then employed for conducting the SOM analysis, with the Calinski criterion chosen to identify the ideal number of classifications [26].

3. Results

3.1. Wetland Landscape Patterns and Metrics from 2000 to 2020

Wetland distribution information was extracted from the LUCC dataset for the years 2000, 2005, 2010, 2015, and 2020 (Figure 3). Based on the pixel size and the number of pixels counted, the area of each type of wetland was calculated in this study. The total wetland area showed an increasing trend between 2000 and 2020, but the wetland structure underwent significant changes. The total area of wetlands was 1276.57 km2 in 2000, of which tidal flats, reservoir pits, and paddy fields were the dominant types with uniform distribution, accounting for more than 80%. In 2020, the total wetland area reached 2569.86 km2, but the proportion of the reservoir pit area exceeded 50%, with an increase of 307.62%. Correspondingly, the area of paddy fields decreased by 203.04 km2. In general, the wetland patterns in the YRD over the past 20 years have been dominated by a considerable decrease in paddy fields and an increase in reservoir and tidal flats. The results show that the wetland conservation plan and the Grain for Green Project have achieved remarkable results.
The wetlands in the YRD had obvious spatial distribution characteristics, mainly distributed in the eastern and northern coastal areas and fan-shaped along the delta (Figure 3). Tidal flats and reservoir pits were distributed in coastal areas and delta areas as the dominant wetland types. In contrast, other types of wetlands were mainly distributed around rivers in platform areas. In the past 20 years, the distribution of inland wetlands in the YRD has significantly decreased, and the overall distribution has shifted to coastal areas. Wetlands located in nature reserves have been well-protected and restored.
Calculations of the landscape metrics showed that wetland landscape aggregation has enhanced at the landscape level in the YRD over the past two decades (Figure 4). From 2000 to 2020, the increase in wetland area was accompanied by a decrease in NP and PD, indicating an increase in aggregation and a decrease in fragmentation. The NP decreased from 915 in 2000 to 531 in 2020, while the PD simultaneously declined from 0.72 to 0.21. The period between 2005 and 2010 was a critical phase for landscape aggregation enhancement, with smaller changes after 2010. The CONNECT and AI are the most direct indices of inter-patch functional connectivity, and both showed a significant increase from 2005 to 2010. The AI metric increased from 96.95% to 98.28% between 2005 and 2010. The COHESION, as an indicator reflecting the state of aggregation and dispersion, showed a similar trend of change with the AI. The LPI and MESH explained that the dominant patches and average wetland patch size fluctuated over the past 20 years and indirectly reflected the intensity and frequency of human interference. The SHDI effectively characterizes landscape diversity, revealing that the equilibrium of wetlands in the YRD declined from 1.54 to 1.20 between 2000 and 2020.

3.2. Spatiotemporal Changes of Wetland ESs from 2000 to 2020

The spatial distribution of ESs was mapped based on the InVEST model for the years 2000, 2005, 2010, 2015, and 2020. As shown in Figure 5, the analysis indicated that distribution patterns of wetland ES in the YRD presented spatial heterogeneity. Specifically, WY and HQ showed similar spatial distribution patterns, and the other three ESs showed another similar spatial pattern. The “High” or “Low” in the mapping of each ES was based on the median value of the results of the ES outputs. The high supply areas of WY and HQ were mainly located in the estuarine delta and the northern coastal region, and the distribution wetland types were tidal flats and marshland combined with the wetland pattern analysis. Additionally, paddy fields and beach land also exhibited good water yield capacity. Areas on the banks of the river and paddy fields were more advantageous in providing CS, SR, and WP because these areas had denser vegetation, sediment accumulation, and agricultural non-point source pollution input.
From a temporal perspective, the WY, CS, and SR of wetland environments showed an overall increasing trend over time from 2000 to 2020, while WP exhibited the opposite decreasing trend. First, the total value of WY and CS slightly decreased from 2010 to 2015, and all other periods showed significant growth. Second, the variability of SR and HQ was more volatile, but there were notable differences in their fluctuation trends. The increase in SR occurred between 2005–2005 and 2015–2020, while the HQ increased substantially from 2005 to 2015. Overall, with the restoration and increase in wetland area, the overall supply of wetland ESs had been optimized except for WP, but it also showed different degrees of volatility.

3.3. Spatial Autocorrelation-Based Connectivity Analysis of Wetland ESs

The Global Moran’s I for different ESs presented a significant positive spatial correlation of wetland ESs in the YRD for the years 2000, 2005, 2010, 2015, and 2020 (Figure 6). Water-related ESs in the study area presented higher connectivity, such as WY and WP, with an index of Moran’s I ranging from 0.51 to 0.83 (p < 0.001). HQ had a high dependence on water in wetland environments and exhibited high connectivity. The Moran’s I ranged from 0.22 to 0.38 for CS and SR (p < 0.001), and were less spatially autocorrelated than the other three ESs, indicating a low degree of connectivity. However, the Moran’s I of some ESs showed a downward trend in different years, indicating that the ecosystem service connectivity was declining, especially WY’s connectivity, which continued to decrease. The Z-score was a key metric for interpreting Moran’s I, and the Z-score was >2.58, indicating that the probability of randomly generating this connectivity pattern was <1%. It is a matter of concern that the increase in wetland areas did not result in a higher connectivity of ESs. The significant expansion of constructed wetlands, especially reservoir pits, may be a significant contributor to the overall decline in ESC in wetlands.
To enhance the visual assessment of local spatial connectivity for all wetland ESs, maps illustrating the distribution of significant (p < 0.05) LISAs were generated (Figure 7). The H-H clusters represented the spatially clustered distribution of high values of ESs of the same type, which implied higher localized connectivity for specific ESs. The H-H clusters that occurred for WY were mainly distributed in tidal flats, river canals, and some areas of paddy fields, while L-L clusters were mainly located in reservoir pits. WP had similar local aggregation characteristics to WY, and WP had a more significant aggregation effect in paddy fields and marshes. The reason is that paddy fields are a concentrated load area of agricultural nitrogen pollution, and a considerable amount of marshland was directly distributed downstream of the paddy fields. The H-H cluster distribution area of HQ was dominated by tidal flats and marshes, while the L-L clusters were scattered around the other watershed. The aggregation effect of HQ was due to tidal flats and marshes being protected by less human disturbance and were subject to a combination of hydrological conditions and vegetation. Paddy fields and marshes exhibited greater advantages in H-H aggregation in CS, which was closely related to the dense distribution of vegetation. The H-H cluster region of SR was less distributed and mostly intersected with the L-H cluster, reflecting the poor connectivity of SR locally. ESC presented spatial heterogeneity accompanying different wetland types.

3.4. Spatial Relationship Between Landscape Metrics and ESC

The results of the Spearman correlation analysis between the landscape metrics and ESC are shown in Figure 8. Unexpected results showed that the landscape metrics were not strongly correlated (>0.5) with ESC at the grid scale in the statistical assessment. In the first, the landscape metrics showed significant positive and negative correlations, especially the LPI, COHESION, MESH, SHDI, and AI (p < 0.001). The SHDI was strongly correlated with all other metrics, positively correlated with the LPI, COHESION, MESH, and AI, and negatively correlated with other metrics. Our findings were consistent with the ecological implications of these landscape metrics. Secondly, the relationship between the ESC and landscape metrics was significant, but the correlation coefficients were all below 0.5. The ESC was positively correlated with the aggregation metrics, mainly distributed in the range of 0.20–0.36. The highest correlation of WY, CS, and SR connectivity with AI was 0.31, 0.30, and 0.32, respectively. WP connectivity showed a correlation coefficient of 0.36 with COHESION, MESH, and AI. HQ connectivity had the highest correlation, with CONNECT showing a value of 0.33. Considering the consistent trend of correlation from 2000 to 2020, we selected the results of 2020 with the strongest significant performance for bivariate spatial autocorrelation between the landscape metrics and ESC.
Subsequently, the spatial relationship between the landscape metrics and ES connectivity was studied. As shown in Figure 9, landscape metrics and ESC presented obvious spatial heterogeneity characteristics. The connectivity of WY, CS, WP, and HQ was similar to the significant spatial aggregation distribution of NP, PD, and SHDI, dominated by H-L and L-L clusters. The connectivity of the four ESs was in contrast to the spatial aggregation distributions of metrics like LPI, COHESION, MESH, and AI, which were predominantly characterized by H-H and L-H clusters. The H-H clusters were mainly distributed in tidal flats in estuarine deltas and the eastern and northern coasts as well as inland marshes, while L-H clusters were mainly located in reservoir pits. The spatial relationship between SR connectivity and landscape metrics showed different characteristics, dominated by the L-L clusters with NP, PD, SHDI and the L-H clusters with LPI, COHESION, MESH, and AI. The CONNECT was mainly spatially distributed in the H-L clusters and N-S with all ESC. Overall, ESC showed different correlated aggregation relationships with the landscape metrics across the spatial distribution of wetlands. The core area of the northeastern estuarine delta may serve as an indicator area reflecting the relationship between the landscape metrics and ESC due to its significant correlation.

3.5. Bundle-Based Identification of Potential Connectivity Between Wetland ESs

This study identified four wetland ES bundles using the SOM method (Figure 10). Bundle 1 was composed of CS, SR, and WP, and the main wetland distribution types were paddy fields and marshes. However, from 2000 to 2020, the distribution area of Bundle 1 shrunk significantly. Bundle 2 was one of the important wetland bundles that consisted of WY and HQ in the YRD, with an area share of 25.92%~41.36% between 2000 and 2020, reaching a maximum in 2010. This was mainly distributed in the coastal areas in the east and north, and tidal flats and beach land were the dominant wetland types. This region had better synergy in providing WY and HQ and was also the main area designated by the nature reserve. This suggests that there may be better connectivity between WY and HQ in these areas. Bundle 3 was composed of CS and HQ, which accounted for the largest portion (59.38% in 2020) of the study area, and was mainly distributed in the reservoir pit area. The ES supply involved in Bundle 4 showed differences across time, which were distinct from the other bundles and may be related to the interactive distribution of other bundles. In summary, Bundle 2 and Bundle 3 had strong correlation in the distribution of wetland types. There may be good connectivity between WY and HQ in the tidal flat areas, and certain connectivity between CS, SR, and WP in the paddy fields and marshes.

4. Discussion

4.1. Mechanistic Analysis of ESC

ESC is closely related to the medium characteristics of ES flows. The flow and transfer of ESs may occur via specific carriers, whether tangible or intangible, with the flow path possessing directional attributes determined by the delivery medium [54]. Based on the various kinds of ESs, the primary transmission routes primarily include the atmosphere, soil, waterways, organisms, human mobility, etc. [55], while the transport types can be categorized as global non-proximity, local proximity, direction flow, in situ, and user movement [56]. Hydrology is undoubtedly the most critical medium for ESC in wetland environments, and water-related ESs tend to present high connectivity by behaving as “direction flow” and “local proximity”. Our results validate this hypothesis and demonstrate the application of spatial correlation analysis in assessing ESC, with WY and WP showing high connectivity. Although SR was influenced by vegetation and the river, erosion control was primarily expressed “in situ” [57]. The transport of CS is considered to be atmospherically relevant, and its connectivity is influenced by flow distance and the carbon exchange rate [55]. The decentralized distribution of vegetation and seawater erosion in the wetlands of the YRD may be factors affecting CS connectivity. The construction of HQ was considered to be “omni-direction” [58], but hydrologic conditions were the most important influencing conditions and therefore also exhibited high connectivity [59]. It is crucial to recognize that ES flows are not equivalent to ESC, which pertains to the functional connectivity relationships among various supply areas.
There are complex interrelations among different ESs, and there may be connections among different ESs in the same location [60,61]. Previous studies have confirmed the existence of a wide range of trade-offs or synergies between ESs, which are easily affected by changes in policy or other factors [62,63]. However, few studies have investigated the potential connectivity of ESs from the perspective of the spatial distribution of ES bundles. Our research results had a high agreement with previous studies on the spatial distribution and identification of bundles [28] and revealed that the connectivity of different ESs was relatively stable in temporal relation to the distribution of wetland types. The connectivity of the CS, SR, and WP situated areas represented by Bundle 1 was evident. As a major grain-producing region of Bundle 1, SR was a necessary condition for the growth of grain crops, CS was a reflection of photosynthesis in the grain growth process, and WP represented paddy fields as the main area for the input and output of nutrient non-point source pollution. The relation between hydrological and habitat quality has been sufficiently studied and emphasized so that the connectivity between WY and HQ in Bundle 2 can also be explained [64,65]. In particular, the area where Bundle 2 is located is a transit point for migratory birds and a biodiversity-rich area. Despite the superficial explanations we have given, the study of connectivity between specific ESs requires an integrated consideration of the transport and exchange of material, energy, and species flows.

4.2. Relationship Between Landscape Metrics and ESC

Examining the relationship between landscape metrics and ESs is a popular study that can provide insights into improving ES provision through optimizing spatial patterns [66]. Previous findings have demonstrated that ESs including trade-offs, synergies, and bundles are affected by landscape patterns, with landscape composition, in particular, being a major influence on ES change [10,22,67,68]. The significant correlation between landscape metrics and ESC has also been confirmed in this study, both statistically and spatially. In general, alterations in landscape patterns could influence source-sink dynamics, the composition of species, and the physical interdependence between areas, resulting in changes in ESC [69]. However, these landscape metrics do not currently serve as a substitute for characterizing ESC, as their strong correlation and spatial correlation still require further research. Despite being characterized by heterogeneity, the ESC of tidal flats in the northeastern coastal region showed good positive (landscape aggregation metrics, e.g., AI, etc.) or negative (e.g., NP, etc.) correlations, especially in the portions located in nature reserves. Therefore, this region can be selected as an indicator region reflecting the relationship between the landscape metrics and ESC to map the differences in other regions. The findings might provide an important basis for future research on ESC and attract more attention to ESC construction in the study area.

4.3. Limitations and Future Outlook

The relationship between the landscape metrics and ESC in a typical wetland environment was investigated, and we found significant correlations and spatial heterogeneity between the landscape metrics and ESC. This helps to deepen our understanding of ESC and promotes its integration into future landscape planning and management. However, as an emerging concept of ESC, there is still much work to be done for this study and future work. (1) ESC at multiple scales may have different characteristics, and exploring the optimal scale for connectivity improvement is an important basis for guiding its incorporation into decision-making and management. Considering the close link between ESC and transmission media, it is necessary to study connectivity at the watershed scale and at larger grid scales in the future. (2) Extensive research has been conducted on the interrelationships of ESs, and the inner connection between ESC and other trade-offs or synergies needs to be further developed. This could be more useful for guiding ecological planning if the trade-offs for ESs are consistent with the spatial distribution of the ESC. (3) This study only analyzed the correlation between the landscape metrics and ESC at the landscape level, and whether ESC at the class level has a stronger relationship with landscape metrics needs to be further identified. Meanwhile, it is necessary to go beyond the current research scope and incorporate more potential landscape metrics into the study or establish a new indicator of ESC. (4) Identifying the factors influencing the spatiotemporal changes in ESC is another important issue for the future. The ESC may differ due to various factors including changes in the natural environment and influences from socioeconomic and policy factors. Analyzing the influencing factors and predicting the future can help develop targeted management strategies.
The YRD wetlands represent fragile and continuously renewing ecosystems, making the integration of ESC into wetland conservation and restoration planning crucial to enhance stability. Despite an increase in wetland area over the past two decades, this expansion is predominantly characterized by artificial wetlands, particularly reservoir pits. Unfortunately, ESC has not been sufficiently prioritized in this process, presenting spatial heterogeneity that introduces new challenges for wetland conservation. In addition, ESC is more dependent on hydrological conditions within wetland environments. Therefore, future wetland conservation plans in the YRD should place greater emphasis on ESC and clearly define the boundaries and scope of both natural and artificial wetland restoration including the intensity of artificial interventions and measures. Additionally, it is essential to ensure adequate water resource input in the YRD and to effectively manage the water–sand relationship.

5. Conclusions

Understanding ecosystem service connectivity and its relationship to landscape patterns is critical for enhancing ecosystem stability and improving ecosystem management, particularly in the context of fragile wetland ecosystems. This research involved examining the spatial characteristics of ecosystem service connectivity and its relationship with landscape metrics in the Yellow River Delta through univariate and bivariate spatial autocorrelation, identified potential connectivity among different ecosystem services through the determination of bundles based on the SOM tool, and profiled future study directions on ecosystem service connectivity. The major conclusions are as follows:
(1)
The water yield, water purification, and habitat quality of wetlands showed high connectivity in the Yellow River Delta, while carbon storage and soil retention had low connectivity. The concern is that the overall ecosystem service connectivity has decreased to different degrees along with the restoration of wetlands from 2000 to 2020, which may be related to the expansion of reservoir pits.
(2)
The connectivity of water yield, water purification, and habitat quality was characterized by similar spatially localized aggregation, where the H-H cluster was dominated by tidal flats and marshes. Ecosystem service connectivity was more dependent on hydrologic conditions within wetland environments.
(3)
Ecosystem service connectivity and landscape metrics showed significant correlation and spatial heterogeneity, where LPI, CONNECT, COHESION, MESH, and AI presented a positive correlation with ecosystem service connectivity to varying degrees. In contrast, spatial H-H clustering was mainly distributed in the tidal flats, especially in the nature reserves.
(4)
Potential connectivity may occur for carbon storage, soil retention, and water purification in Bundle 1, which was distributed in paddy fields and marshes, while water yield and habitat quality in Bundle 2, in the tidal flats, showed stronger connectivity. Our findings may assist decision-makers in developing effective ecosystem management strategies and provide a reference for future research on ecosystem service connectivity. The research on ecosystem service connectivity is still at an early stage, and future studies should focus on the mechanisms, methods, and drivers.

Author Contributions

Conceptualization, S.W. and L.Z.; Methodology, C.H., S.W. and Y.R.; Software, W.C. and M.C.; Validation, C.H., S.W. and L.Z.; Formal analysis, C.H., W.C. and M.C.; Investigation, W.C. and M.C.; Resources, X.C. and Y.R.; Data curation, C.H., M.C. and X.C.; Writing—original draft preparation, C.H.; Writing—review and editing, S.W. and L.Z.; Visualization, C.H., M.C. and X.C.; Supervision, S.W. and L.Z.; Project administration, L.Z.; Funding acquisition, L.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Natural Science Foundation of Shandong Province (ZR2022MC056) and the National Social Science Fund of China (23BTJ030).

Data Availability Statement

The LUCC data were obtained from the RESDC, CAS (https://www.resdc.cn/ (accessed on 27 May 2024)). The DEM data were derived from the Geospatial Data Cloud (http://www.gscloud.cn/ (accessed on 27 May 2024)), and the meteorological information were gathered from the NESSDC (https://www.geodata.cn (accessed on 29 May 2024)). The soil data and the road network were obtained from the HWSD (https://www.fao.org/ (accessed on 15 June 2024)) and OSM (www.openstreetmap.org/ (accessed on 20 June 2024)), respectively.

Acknowledgments

The authors are deeply grateful to the anonymous reviewers for their valuable and insightful suggestions.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Appendix A.1. Methods of Accounting for Ecosystem Services

(1)
Water yield (WY)
The WY refers to the capability and process of ecosystems to enhance the water availability by capturing precipitation through their structural and functional features. In the InVEST model, the computation of WY is determined by water balance, which is defined as the difference between precipitation, evaporation, and soil infiltration. The formula is as follows:
W Y ( x ) = 1 − A E T ( x ) P ( x ) × P ( x )
where WY(x) represents the annual water yield of pixel x (mm); AET(x) and P(x) are the annual evapotranspiration and precipitation of pixel x (mm), respectively. The relationship between AET(x) and P(x) is calculated using the Budyko hydrothermal coupling equilibrium equation [70].
(2)
Carbon storage (CS)
Wetland ecosystems are one of the most important carbon pools on earth, contributing 12~24% of carbon storage in 4%~6% of the global distribution area [71]. Wetland carbon stocks come from fixation by vegetation photosynthesis and undecomposed organic carbon, while the soil carbon stock is much higher than that sequestered by vegetation [72,73]. The InVEST model for carbon storage and sequestration utilizes land use data alongside stocks from four carbon pools: aboveground biomass, belowground biomass, soil, and dead organic matter. This model estimates the carbon storage capacity within wetland landscapes.
C S = C a b o v e + C b e l o w + C s o i l + C d e a d
where the CS is the amount of the total carbon storage (Mg); Cabove, Cbelow, Csoil, and Cdead represent aboveground biomass, belowground biomass, soil, and dead organic matter, respectively, and these units are Mg∙ha−1.
(3)
Soil retention (SR)
SR denotes the capacity of ecosystems to mitigate soil erosion and is intricately linked to factors such as soil properties, types of vegetation cover, climatic conditions, and topographical features. The Revised Universal Soil Loss Equation (RUSLE) is used to calculate the SR, and is equal to the difference between potential soil erosion and actual soil erosion:
S R = R × K × L × S × ( 1 − C × P )
where SR denotes the soil erosion index (t∙ha−1∙a−1); R is the rainfall erosivity factor (MJ∙mm∙ha−1∙h−1); K is the soil erodibility factor (t∙ha∙MJ−1∙mm−1); L and S are the slope length factor and slope steepness factor, respectively; C is the vegetation cover factor; P represents the soil conservation practice factor.
(4)
Water purification (WP)
The nutrient delivery ratio (NDR) module in the InVEST model was selected to calculate water purification services. The NDR is based on the mass balance, which describes the movement of pollutants such as nitrogen and phosphorus in space through empirical relationships and then estimates the pollutants intercepted by each pixel as an indicator of WP. The formula is as follows:
W P = l o a d N i × ( 1 − N D R N i )
where WP represents the amount of purified total nitrogen; loadNi is the total nitrogen load, which depends on the input of non-point source pollution including domestic sewage, agricultural fertilizers, and aquaculture wastewater in the YRD; NDRNi is the transport ratio of total nitrogen.
(5)
Habitat quality (HQ)
HQ service refers to the capability of ecosystems to create a habitat conducive to the sustainable growth of various species and populations, serving as an indicator of biodiversity to a certain degree [74]. The InVEST model integrates land use mapping with data regarding human activity threats to provide an extensive overview of the status of habitat degradation and protection. HQ was calculated by the given equation:
H Q x j = H j 1 − D x j z D x j z + k z
where HQxj denotes the habitat quality index of grid unit x of landscape type j, ranging from 0 to 1; Hj is the habitat suitability score of the landscape j; k is the semi-saturated constant, and z represents the scale constant; DZxj is the habitat degradation index, which represents the total threat level in the grid unit x of landscape j.

Appendix A.2. Methods of Spatial Autocorrelation Analysis

(1)
The formula for the univariate spatial autocorrelation analysis is as follows:
I = ∑ i = 1 n ∑ j = 1 n w i j ( x i − x ¯ ) ( x j − x ¯ ) S 2 ∑ i = 1 n ∑ j = 1 n w i j
I i = ( x j − x ¯ ) S 2 ∑ i = 1 n w i j ( x i − x ¯ )
where I and Ii demote the univariate global and local Moran’s I for ESs, respectively; n signifies the total number of grid cells of 1 km × 1 km; wij refers to the weight coefficient matrix of grid cells i and j; xi and xj are the normalized values of different ESs of grid cells i and j; x ¯ represents the average value of ES, while S2 is the variance measure.
(2)
The formula for univariate spatial autocorrelation analysis is as follows:
I i , x y = z i ∑ j = 1 n w i j z j
where Ii,xy is the local spatial relationship between variables x and y, and zi and zj are the standardized variance values of the observations of variables x and y on the spatial units i and j, respectively.

References

  1. Reid, W.V.; Mooney, H.A.; Cropper, A.; Capistrano, D.; Carpenter, S.R.; Chopra, K.; Dasgupta, P.; Dietz, T.; Duraiappah, A.K.; Hassan, R.; et al. Ecosystems and Human Well-Being: Synthesis; Island Press: Washington, DC, USA, 2005; ISBN 978-1-59726-040-4. [Google Scholar]
  2. Ouyang, Z.; Song, C.; Zheng, H.; Polasky, S.; Xiao, Y.; Bateman, I.J.; Liu, J.; Ruckelshaus, M.; Shi, F.; Xiao, Y.; et al. Using Gross Ecosystem Product (GEP) to Value Nature in Decision Making. Proc. Natl. Acad. Sci. USA 2020, 117, 14593–14601. [Google Scholar] [CrossRef] [PubMed]
  3. Chen, H.; Yan, W.; Li, Z.; Wende, W.; Xiao, S. A Framework for Integrating Ecosystem Service Provision and Connectivity in Ecological Spatial Networks: A Case Study of the Shanghai Metropolitan Area. Sustain. Cities Soc. 2024, 100, 105018. [Google Scholar] [CrossRef]
  4. Field, R.D.; Parrott, L. Mapping the Functional Connectivity of Ecosystem Services Supply across a Regional Landscape. eLife 2022, 11, e69395. [Google Scholar] [CrossRef] [PubMed]
  5. Pashanejad, E.; Kharrazi, A.; Araujo-Gutierrez, Z.M.F.; Robinson, B.E.; Fath, B.D.; Parrott, L. A Functional Connectivity Approach for Exploring Interactions of Multiple Ecosystem Services in the Context of Agricultural Landscapes in the Canadian Prairies. Ecosyst. Serv. 2024, 68, 101639. [Google Scholar] [CrossRef]
  6. Chen, Y.; Vardon, M. Accounting for Water-Related Ecosystem Services to Provide Information for Water Policy and Management: An Australian Case Study. Ecosyst. Serv. 2024, 69, 101658. [Google Scholar] [CrossRef]
  7. Aguilera, M.A.; Tapia, J.; Gallardo, C.; Núñez, P.; Varas-Belemmi, K. Loss of Coastal Ecosystem Spatial Connectivity and Services by Urbanization: Natural-to-Urban Integration for Bay Management. J. Environ. Manag. 2020, 276, 111297. [Google Scholar] [CrossRef]
  8. Bai, Y.; Chen, Y.; Alatalo, J.M.; Yang, Z.; Jiang, B. Scale Effects on the Relationships between Land Characteristics and Ecosystem Services- a Case Study in Taihu Lake Basin, China. Sci. Total Environ. 2020, 716, 137083. [Google Scholar] [CrossRef]
  9. Farley, J.; Voinov, A. Economics, Socio-Ecological Resilience and Ecosystem Services. J. Environ. Manag. 2016, 183, 389–398. [Google Scholar] [CrossRef]
  10. Liu, S.; Wang, Z.; Wu, W.; Yu, L. Effects of Landscape Pattern Change on Ecosystem Services and Its Interactions in Karst Cities: A Case Study of Guiyang City in China. Ecol. Indic. 2022, 145, 109646. [Google Scholar] [CrossRef]
  11. Burgos-Ayala, A.; Jiménez-Aceituno, A.; Meacham, M.; Rozas-Vásquez, D.; Mancilla García, M.; Rocha, J.; Rincón-Ruíz, A. Mapping Ecosystem Services in Colombia: Analysis of Synergies, Trade-Offs and Bundles in Environmental Management. Ecosyst. Serv. 2024, 66, 101608. [Google Scholar] [CrossRef]
  12. Karimi, J.D.; Corstanje, R.; Harris, J.A. Understanding the Importance of Landscape Configuration on Ecosystem Service Bundles at a High Resolution in Urban Landscapes in the UK. Landsc. Ecol. 2021, 36, 2007–2024. [Google Scholar] [CrossRef]
  13. Cord, A.F.; Bartkowski, B.; Beckmann, M.; Dittrich, A.; Hermans-Neumann, K.; Kaim, A.; Lienhoop, N.; Locher-Krause, K.; Priess, J.; Schröter-Schlaack, C.; et al. Towards Systematic Analyses of Ecosystem Service Trade-Offs and Synergies: Main Concepts, Methods and the Road Ahead. Ecosyst. Serv. 2017, 28, 264–272. [Google Scholar] [CrossRef]
  14. Xia, H.; Yuan, S.; Prishchepov, A.V. Spatial-Temporal Heterogeneity of Ecosystem Service Interactions and Their Social-Ecological Drivers: Implications for Spatial Planning and Management. Resour. Conserv. Recycl. 2023, 189, 106767. [Google Scholar] [CrossRef]
  15. Li, W.; Kang, J.; Wang, Y. Identifying the Impacts of Landscape Structure on Ecosystem Services Trade-Offs among Ecosystem Services Bundles in the Mountains of Southwest China. Ecol. Eng. 2024, 209, 107419. [Google Scholar] [CrossRef]
  16. Mitchell, M.G.E.; Chan, K.M.A.; Newlands, N.K.; Ramankutty, N. Spatial Correlations Don’t Predict Changes in Agricultural Ecosystem Services: A Canada-Wide Case Study. Front. Sustain. Food Syst. 2020, 4, e539892. [Google Scholar] [CrossRef]
  17. Balbi, M.; Petit, E.J.; Croci, S.; Nabucet, J.; Georges, R.; Madec, L.; Ernoult, A. Title: Ecological Relevance of Least Cost Path Analysis: An Easy Implementation Method for Landscape Urban Planning. J. Environ. Manag. 2019, 244, 61–68. [Google Scholar] [CrossRef]
  18. Zhang, Z.; Wang, Q.; Feng, Y.; Sun, Y.; Liu, N.; Yan, S. The Spatio-Temporal Evolution of Spatial Structure and Supply-Demand Relationships of the Ecological Network in the Yellow River Delta Region of China. J. Clean. Prod. 2024, 471, 143388. [Google Scholar] [CrossRef]
  19. Borrett, S.R.; Scharler, U.M. Walk Partitions of Flow in Ecological Network Analysis: Review and Synthesis of Methods and Indicators. Ecol. Indic. 2019, 106, 105451. [Google Scholar] [CrossRef]
  20. Peng, J.; Yang, Y.; Liu, Y.; Hu, Y.; Du, Y.; Meersmans, J.; Qiu, S. Linking Ecosystem Services and Circuit Theory to Identify Ecological Security Patterns. Sci. Total Environ. 2018, 644, 781–790. [Google Scholar] [CrossRef]
  21. Dong, Y.; Liu, S.; Pei, X.; Wang, Y. Identifying Critical Landscape Patterns for Simultaneous Provision of Multiple Ecosystem Services—A Case Study in the Central District of Wuhu City, China. Ecol. Indic. 2024, 158, 111380. [Google Scholar] [CrossRef]
  22. Lamy, T.; Liss, K.N.; Gonzalez, A.; Bennett, E.M. Landscape Structure Affects the Provision of Multiple Ecosystem Services. Environ. Res. Lett. 2016, 11, 124017. [Google Scholar] [CrossRef]
  23. Mitchell, M.G.E.; Suarez-Castro, A.F.; Martinez-Harms, M.; Maron, M.; McAlpine, C.; Gaston, K.J.; Johansen, K.; Rhodes, J.R. Reframing Landscape Fragmentation’s Effects on Ecosystem Services. Trends Ecol. Evol. 2015, 30, 190–198. [Google Scholar] [CrossRef] [PubMed]
  24. Li, J.; Zhou, K.; Xie, B.; Xiao, J. Impact of Landscape Pattern Change on Water-Related Ecosystem Services: Comprehensive Analysis Based on Heterogeneity Perspective. Ecol. Indic. 2021, 133, 108372. [Google Scholar] [CrossRef]
  25. Huang, J.; Xia, Z.; Liu, L. Impacts of Landscape Configurations on Ecosystem Services and Their Trade-Offs across Different Landscape Compositions. Ecosyst. Serv. 2024, 70, 101666. [Google Scholar] [CrossRef]
  26. Li, Y.; Geng, H.; Luo, G.; Wu, L.; Wang, J.; Wu, Q. Multiscale Characteristics of Ecosystem Service Value Trade-Offs/Synergies and Their Response to Landscape Pattern Evolution in a Typical Karst Basin in Southern China. Ecol. Inform. 2024, 81, 102584. [Google Scholar] [CrossRef]
  27. Ning, Z.; Li, D.; Chen, C.; Xie, C.; Chen, G.; Xie, T.; Wang, Q.; Bai, J.; Cui, B. The Importance of Structural and Functional Characteristics of Tidal Channels to Smooth Cordgrass Invasion in the Yellow River Delta, China: Implications for Coastal Wetland Management. J. Environ. Manag. 2023, 342, 118297. [Google Scholar] [CrossRef]
  28. Zhang, T.; Zhang, S.; Cao, Q.; Wang, H.; Li, Y. The Spatiotemporal Dynamics of Ecosystem Services Bundles and the Social-Economic-Ecological Drivers in the Yellow River Delta Region. Ecol. Indic. 2022, 135, 108573. [Google Scholar] [CrossRef]
  29. Zhang, H.; Sun, Y.; Zhang, W.; Song, Z.; Ding, Z.; Zhang, X. Comprehensive Evaluation of the Eco-Environmental Vulnerability in the Yellow River Delta Wetland. Ecol. Indic. 2021, 125, 107514. [Google Scholar] [CrossRef]
  30. Yu, B.; Zang, Y.; Wu, C.; Zhao, Z. Spatiotemporal Dynamics of Wetlands and Their Future Multi-Scenario Simulation in the Yellow River Delta, China. J. Environ. Manag. 2024, 353, 120193. [Google Scholar] [CrossRef]
  31. Zhao, Q.; Bai, J.; Gao, Y.; Zhang, G.; Lu, Q.; Jia, J. Heavy Metal Contamination in Soils from Freshwater Wetlands to Salt Marshes in the Yellow River Estuary, China. Sci. Total Environ. 2021, 774, 145072. [Google Scholar] [CrossRef]
  32. Zhang, X.; Wang, G.; Xue, B.; Zhang, M.; Tan, Z. Dynamic Landscapes and the Driving Forces in the Yellow River Delta Wetland Region in the Past Four Decades. Sci. Total Environ. 2021, 787, 147644. [Google Scholar] [CrossRef] [PubMed]
  33. Liu, J.; Kuang, W.; Zhang, Z.; Xu, X.; Qin, Y.; Ning, J.; Zhou, W.; Zhang, S.; Li, R.; Yan, C.; et al. Spatiotemporal Characteristics, Patterns, and Causes of Land-Use Changes in China since the Late 1980s. J. Geogr. Sci. 2014, 24, 195–210. [Google Scholar] [CrossRef]
  34. Natural Capital Project. InVEST 0.0.; Stanford University, University of Minnesota, Chinese Academy of Sciences, The Nature Conservancy, World Wildlife Fund, Stockholm Resilience Centre and the Royal Swedish Academy of Sciences, 2024. Available online: https://naturalcapitalproject.stanford.edu/ (accessed on 14 June 2024).
  35. Mehvar, S.; Filatova, T.; Sarker, M.H.; Dastgheib, A.; Ranasinghe, R. Climate Change-Driven Losses in Ecosystem Services of Coastal Wetlands: A Case Study in the West Coast of Bangladesh. Ocean Coast. Manag. 2019, 169, 273–283. [Google Scholar] [CrossRef]
  36. Xia, H.; Liu, L.; Bai, J.; Kong, W.; Lin, K.; Guo, F. Wetland Ecosystem Service Dynamics in the Yellow River Estuary under Natural and Anthropogenic Stress in the Past 35 Years. Wetlands 2020, 40, 2741–2754. [Google Scholar] [CrossRef]
  37. Xu, X.; Chen, M.; Yang, G.; Jiang, B.; Zhang, J. Wetland Ecosystem Services Research: A Critical Review. Glob. Ecol. Conserv. 2020, 22, e01027. [Google Scholar] [CrossRef]
  38. Zhang, X.; Wang, J.; Yue, C.; Ma, S.; Hou, L.; Wang, L.-J. Impact of Wetland Change on Ecosystem Services in Different Urbanization Stages: A Case Study in the Hang-Jia-Hu Region, China. Ecol. Indic. 2023, 153, 110382. [Google Scholar] [CrossRef]
  39. Huang, C.; Zhao, D.; Liao, Q.; Xiao, M. Linking Landscape Dynamics to the Relationship between Water Purification and Soil Retention. Ecosyst. Serv. 2023, 59, 101498. [Google Scholar] [CrossRef]
  40. Kim, J.; Song, Y. Integrating Ecosystem Services and Ecological Connectivity to Prioritize Spatial Conservation on Jeju Island, South Korea. Landsc. Urban Plan. 2023, 239, 104865. [Google Scholar] [CrossRef]
  41. Liang, F.; Zhu, R.; Lin, S.-H. Exploring Spatial Relationship between Landscape Configuration and Ecosystem Services: A Case Study of Xiamen–Zhangzhou–Quanzhou in China. Ecol. Model. 2023, 486, 110527. [Google Scholar] [CrossRef]
  42. Lookingbill, T.R.; Minor, E.S.; Mullis, C.S.; Nunez-Mir, G.C.; Johnson, P. Connectivity in the Urban Landscape (2015–2020): Who? Where? What? When? Why? And How? Curr. Landsc. Ecol. Rep. 2022, 7, 1–14. [Google Scholar] [CrossRef]
  43. Mitchell, M.G.E.; Bennett, E.M.; Gonzalez, A. Linking Landscape Connectivity and Ecosystem Service Provision: Current Knowledge and Research Gaps. Ecosystems 2013, 16, 894–908. [Google Scholar] [CrossRef]
  44. Spanowicz, A.G.; Jaeger, J.A.G. Measuring Landscape Connectivity: On the Importance of within-Patch Connectivity. Landsc. Ecol. 2019, 34, 2261–2278. [Google Scholar] [CrossRef]
  45. Field, R.D.; Parrott, L. Multi-Ecosystem Services Networks: A New Perspective for Assessing Landscape Connectivity and Resilience. Ecol. Complex. 2017, 32, 31–41. [Google Scholar] [CrossRef]
  46. Guan, X.; Xu, Y.; Meng, Y.; Xu, W.; Yan, D. Quantifying Multi-Dimensional Services of Water Ecosystems and Breakpoint-Based Spatial Radiation of Typical Regulating Services Considering the Hierarchical Clustering-Based Classification. J. Environ. Manag. 2024, 351, 119852. [Google Scholar] [CrossRef]
  47. Li, L.; Tang, H.; Lei, J.; Song, X. Spatial Autocorrelation in Land Use Type and Ecosystem Service Value in Hainan Tropical Rain Forest National Park. Ecol. Indic. 2022, 137, 108727. [Google Scholar] [CrossRef]
  48. Shao, Y.; Xiao, Y.; Sang, W. Land Use Trade-Offs and Synergies Based on Temporal and Spatial Patterns of Ecosystem Services in South China. Ecol. Indic. 2022, 143, 109335. [Google Scholar] [CrossRef]
  49. Zhu, C.; Zhang, X.; Zhou, M.; He, S.; Gan, M.; Yang, L.; Wang, K. Impacts of Urbanization and Landscape Pattern on Habitat Quality Using OLS and GWR Models in Hangzhou, China. Ecol. Indic. 2020, 117, 106654. [Google Scholar] [CrossRef]
  50. Ai, M.; Chen, X.; Yu, Q. Spatial Correlation Analysis between Human Disturbance Intensity (HDI) and Ecosystem Services Value (ESV) in the Chengdu-Chongqing Urban Agglomeration. Ecol. Indic. 2024, 158, 111555. [Google Scholar] [CrossRef]
  51. Yang, W.; Chen, J.; Ding, T.; Yan, X.; Gong, W. Supply-Demand Security Assessment of Water-Energy-Food Systems: A Perspective on Intra-City Coupling and Inter-City Linkages of Ecosystem Services. Sustain. Cities Soc. 2024, 117, 105964. [Google Scholar] [CrossRef]
  52. Xu, J.; Wang, S.; Xiao, Y.; Xie, G.; Wang, Y.; Zhang, C.; Li, P.; Lei, G. Mapping the Spatiotemporal Heterogeneity of Ecosystem Service Relationships and Bundles in Ningxia, China. J. Clean. Prod. 2021, 294, 126216. [Google Scholar] [CrossRef]
  53. Shen, J.; Li, S.; Liu, L.; Liang, Z.; Wang, Y.; Wang, H.; Wu, S. Uncovering the Relationships between Ecosystem Services and Social-Ecological Drivers at Different Spatial Scales in the Beijing-Tianjin-Hebei Region. J. Clean. Prod. 2021, 290, 125193. [Google Scholar] [CrossRef]
  54. Koellner, T.; Bonn, A.; Arnhold, S.; Bagstad, K.J.; Fridman, D.; Guerra, C.A.; Kastner, T.; Kissinger, M.; Kleemann, J.; Kuhlicke, C.; et al. Guidance for Assessing Interregional Ecosystem Service Flows. Ecol. Indic. 2019, 105, 92–106. [Google Scholar] [CrossRef]
  55. Su, D.; Cao, Y.; Dong, X.; Wu, Q.; Fang, X.; Cao, Y. Evaluation of Ecosystem Services Budget Based on Ecosystem Services Flow: A Case Study of Hangzhou Bay Area. Appl. Geogr. 2024, 162, 103150. [Google Scholar] [CrossRef]
  56. Costanza, R. Ecosystem Services: Multiple Classification Systems Are Needed. Biol. Conserv. 2008, 141, 350–352. [Google Scholar] [CrossRef]
  57. Li, K.; Hou, Y.; Andersen, P.S.; Xin, R.; Rong, Y.; Skov-Petersen, H. An Ecological Perspective for Understanding Regional Integration Based on Ecosystem Service Budgets, Bundles, and Flows: A Case Study of the Jinan Metropolitan Area in China. J. Environ. Manag. 2022, 305, 114371. [Google Scholar] [CrossRef]
  58. Vrebos, D.; Staes, J.; Vandenbroucke, T.; D׳Haeyer, T.; Johnston, R.; Muhumuza, M.; Kasabeke, C.; Meire, P. Mapping Ecosystem Service Flows with Land Cover Scoring Maps for Data-Scarce Regions. Ecosyst. Serv. 2015, 13, 28–40. [Google Scholar] [CrossRef]
  59. Ding, Q.; Chen, Y.; Bu, L.; Ye, Y. Multi-Scenario Analysis of Habitat Quality in the Yellow River Delta by Coupling FLUS with InVEST Model. Int. J. Environ. Res. Public Health 2021, 18, 2389. [Google Scholar] [CrossRef]
  60. Brauman, K.A.; Daily, G.C.; Duarte, T.K.; Mooney, H.A. The Nature and Value of Ecosystem Services: An Overview Highlighting Hydrologic Services. Annu. Rev. Environ. Resour. 2007, 32, 67–98. [Google Scholar] [CrossRef]
  61. Mach, M.E.; Martone, R.G.; Chan, K.M.A. Human Impacts and Ecosystem Services: Insufficient Research for Trade-off Evaluation. Ecosyst. Serv. 2015, 16, 112–120. [Google Scholar] [CrossRef]
  62. Ren, Y.; Zhang, L.; Wei, X.; Song, Y.; Wu, S.; Wang, H.; Chen, X.; Qiao, Y.; Liang, T. Evaluating and Simulating the Impact of Afforestation Policy on Land Use and Ecosystem Services Trade-Offs in Linyi, China. Ecol. Indic. 2024, 160, 111898. [Google Scholar] [CrossRef]
  63. Wang, H.; Liu, L.; Yin, L.; Shen, J.; Li, S. Exploring the Complex Relationships and Drivers of Ecosystem Services across Different Geomorphological Types in the Beijing-Tianjin-Hebei Region, China (2000–2018). Ecol. Indic. 2021, 121, 107116. [Google Scholar] [CrossRef]
  64. Gao, X.; Liang, J.; Zhu, Z.; Li, W.; Lu, L.; Li, X.; Li, S.; Tang, N.; Li, X. Drought-Induced Changes in Hydrological and Phenological Interactions Modulate Waterbird Habitats Dynamics. J. Hydrol. 2023, 626, 130228. [Google Scholar] [CrossRef]
  65. Singha, P.; Pal, S. Modelling Wetland Eco-Hydrological State and Its Role on Ecosystem Service Potentiality. Ecohydrol. Hydrobiol. 2024, in press. [Google Scholar] [CrossRef]
  66. Zhang, J.; Qu, M.; Wang, C.; Zhao, J.; Cao, Y. Quantifying Landscape Pattern and Ecosystem Service Value Changes: A Case Study at the County Level in the Chinese Loess Plateau. Glob. Ecol. Conserv. 2020, 23, e01110. [Google Scholar] [CrossRef]
  67. Fan, F.; Li, W.; Feng, Z.; Yang, Y. Combining Landscape Patterns and Ecosystem Services to Disclose Ecosystem Changes in Tropical Cropland-Forest Shifting Zones: Inspiration from Mainland Southeast Asia. J. Clean. Prod. 2024, 434, 140058. [Google Scholar] [CrossRef]
  68. Ran, P.; Hu, S.; Frazier, A.E.; Yang, S.; Song, X.; Qu, S. The Dynamic Relationships between Landscape Structure and Ecosystem Services: An Empirical Analysis from the Wuhan Metropolitan Area, China. J. Environ. Manag. 2023, 325, 116575. [Google Scholar] [CrossRef]
  69. Qiu, J.; Turner, M.G. Importance of Landscape Heterogeneity in Sustaining Hydrologic Ecosystem Services in an Agricultural Watershed. Ecosphere 2015, 6, art229. [Google Scholar] [CrossRef]
  70. Donohue, R.J.; Roderick, M.L.; McVicar, T.R. Roots, Storms and Soil Pores: Incorporating Key Ecohydrological Processes into Budyko’s Hydrological Model. J. Hydrol. 2012, 436–437, 35–50. [Google Scholar] [CrossRef]
  71. Liu, Y.; Xi, M.; Zhang, X.; Yu, Z.; Kong, F. Carbon Storage Distribution Characteristics of Wetlands in China and Its Influencing Factors. Chin. J. Appl. Ecol. 2019, 30, 2481–2489. [Google Scholar]
  72. Sun, S.; Song, Z.; Chen, B.; Wang, Y.; Ran, X.; Fang, Y.; Van Zwieten, L.; Hartley, I.P.; Wang, Y.; Li, Q.; et al. Current and Future Potential Soil Organic Carbon Stocks of Vegetated Coastal Ecosystems and Their Controls in the Bohai Rim Region, China. CATENA 2023, 225, 107023. [Google Scholar] [CrossRef]
  73. Zhao, Q.; Bai, J.; Wang, X.; Zhang, W.; Huang, Y.; Wang, L.; Gao, Y. Soil Organic Carbon Content and Stock in Wetlands with Different Hydrologic Conditions in the Yellow River Delta, China. Ecohydrol. Hydrobiol. 2020, 20, 537–547. [Google Scholar] [CrossRef]
  74. Ren, Q.; He, C.; Huang, Q.; Shi, P.; Zhang, D.; Guneralp, B. Impacts of Urban Expansion on Natural Habitats in Global Drylands. Nat. Sustain. 2022, 5, 869–878. [Google Scholar] [CrossRef]
Land 14 00273 g001
Figure 1. Research flowchart of the study.
Figure 1. Research flowchart of the study.
Land 14 00273 g001
Land 14 00273 g002
Figure 2. Location of the study area (Yellow River Delta, China).
Figure 2. Location of the study area (Yellow River Delta, China).
Land 14 00273 g002
Land 14 00273 g003
Figure 3. (a–e) Spatiotemporal changes in the wetland pattern and (f) area composition from 2000 to 2020 in the Yellow River Delta, China.
Figure 3. (a–e) Spatiotemporal changes in the wetland pattern and (f) area composition from 2000 to 2020 in the Yellow River Delta, China.
Land 14 00273 g003
Land 14 00273 g004
Figure 4. Landscape connectivity metrics for wetlands in the Yellow River Delta, China, from 2000 to 2020 at the landscape level. (#: the number of patches).
Figure 4. Landscape connectivity metrics for wetlands in the Yellow River Delta, China, from 2000 to 2020 at the landscape level. (#: the number of patches).
Land 14 00273 g004
Land 14 00273 g005
Figure 5. Spatiotemporal changes in wetland ecosystem services based on the InVEST model from 2000 to 2020 in the Yellow River Delta, China. (WY: water yield; CS: carbon storage; SR: soil retention; WP: water purification; HQ: habitat quality).
Figure 5. Spatiotemporal changes in wetland ecosystem services based on the InVEST model from 2000 to 2020 in the Yellow River Delta, China. (WY: water yield; CS: carbon storage; SR: soil retention; WP: water purification; HQ: habitat quality).
Land 14 00273 g005
Land 14 00273 g006
Figure 6. (a) Global Moran’s I and (b) Z-score of ecosystem services of wetlands from 2000 to 2020 in the Yellow River Delta, China (Moran’s I > 0, positive spatial correlation and nearing 1 signifying a stronger connection; Moran’s I < 0, negative spatial correlation; Z-score > 2.58 and p < 0.001, 99.9% confidence level clustered distribution).
Figure 6. (a) Global Moran’s I and (b) Z-score of ecosystem services of wetlands from 2000 to 2020 in the Yellow River Delta, China (Moran’s I > 0, positive spatial correlation and nearing 1 signifying a stronger connection; Moran’s I < 0, negative spatial correlation; Z-score > 2.58 and p < 0.001, 99.9% confidence level clustered distribution).
Land 14 00273 g006
Land 14 00273 g007
Figure 7. Localized spatial connectivity patterns of wetland ecosystem services based on autocorrelation analysis from 2000 to 2020 in the Yellow River Delta, China (H-H: High-High cluster; H-L: High-Low cluster; L-H: Low-High cluster; L-L: Low-Low cluster; N-S: not significant).
Figure 7. Localized spatial connectivity patterns of wetland ecosystem services based on autocorrelation analysis from 2000 to 2020 in the Yellow River Delta, China (H-H: High-High cluster; H-L: High-Low cluster; L-H: Low-High cluster; L-L: Low-Low cluster; N-S: not significant).
Land 14 00273 g007
Land 14 00273 g008
Figure 8. Spearman rank correlation coefficient between the landscape metrics and ecosystem service connectivity at the grid scale in (a) 2000, (b) 2005, (c) 2010, (d) 2015, and (e) 2020 (WY_C, CS_C, SR_C, WP_C, and HQ_C represent the connectivity of WY, CS, SR, WP, and HQ, respectively).
Figure 8. Spearman rank correlation coefficient between the landscape metrics and ecosystem service connectivity at the grid scale in (a) 2000, (b) 2005, (c) 2010, (d) 2015, and (e) 2020 (WY_C, CS_C, SR_C, WP_C, and HQ_C represent the connectivity of WY, CS, SR, WP, and HQ, respectively).
Land 14 00273 g008
Land 14 00273 g009
Figure 9. Spatial autocorrelation aggregation patterns of the landscape metrics and ecosystem service connectivity in 2020 in the Yellow River Delta, China (H-H: High-High cluster; H-L: High-Low cluster; L-H: Low-High cluster; L-L: Low-Low cluster; N-S: not significant).
Figure 9. Spatial autocorrelation aggregation patterns of the landscape metrics and ecosystem service connectivity in 2020 in the Yellow River Delta, China (H-H: High-High cluster; H-L: High-Low cluster; L-H: Low-High cluster; L-L: Low-Low cluster; N-S: not significant).
Land 14 00273 g009
Land 14 00273 g010
Figure 10. (a–e) Spatiotemporal patterns of ecosystem service bundles, (f) composition and relative magnitude of the ecosystem services in bundles, and (g) the proportion of the area of bundles from 2000 to 2020 in the Yellow River Delta, China.
Figure 10. (a–e) Spatiotemporal patterns of ecosystem service bundles, (f) composition and relative magnitude of the ecosystem services in bundles, and (g) the proportion of the area of bundles from 2000 to 2020 in the Yellow River Delta, China.
Land 14 00273 g010
Table 1. Data sources and spatial scales.
Table 1. Data sources and spatial scales.
Data TypeSpatial ScaleSourcesYear
Land use and land cover change (LUCC)30 mResource and Environment Science and Data Center,
Chinese Academy of Sciences (https://www.resdc.cn/
(accessed on 27 May 2024))		2000–2020
Digital elevation model (DEM)30 mGeospatial Data Cloud (http://www.gscloud.cn/
(accessed on 27 May 2024))		--
Meteorological data1 kmNational Earth System Science Data Center (https://www.geodata.cn
(accessed on 29 May 2024))		2000–2020
Soil data1 kmHarmonized World Soil Database (HWSD) v1.2 (https://www.fao.org/
(accessed on 15 June 2024))		--
Road network--OpenStreetMap (www.openstreetmap.org/
(accessed on 20 June 2024))		2000–2020
Nature reserves--Dongying Natural Resources Department--
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hao, C.; Wu, S.; Cheng, W.; Chen, M.; Ren, Y.; Chang, X.; Zhang, L. Spatiotemporal Relationship Between Landscape Pattern and Ecosystem Service Connectivity in Wetland Environment: Evidence from Yellow River Delta, China. Land 2025, 14, 273. https://doi.org/10.3390/land14020273

AMA Style

Hao C, Wu S, Cheng W, Chen M, Ren Y, Chang X, Zhang L. Spatiotemporal Relationship Between Landscape Pattern and Ecosystem Service Connectivity in Wetland Environment: Evidence from Yellow River Delta, China. Land. 2025; 14(2):273. https://doi.org/10.3390/land14020273

Chicago/Turabian Style

Hao, Chaozhi, Shuyao Wu, Wenjie Cheng, Mengna Chen, Yaofa Ren, Xiaoqing Chang, and Linbo Zhang. 2025. "Spatiotemporal Relationship Between Landscape Pattern and Ecosystem Service Connectivity in Wetland Environment: Evidence from Yellow River Delta, China" Land 14, no. 2: 273. https://doi.org/10.3390/land14020273

APA Style

Hao, C., Wu, S., Cheng, W., Chen, M., Ren, Y., Chang, X., & Zhang, L. (2025). Spatiotemporal Relationship Between Landscape Pattern and Ecosystem Service Connectivity in Wetland Environment: Evidence from Yellow River Delta, China. Land, 14(2), 273. https://doi.org/10.3390/land14020273

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop