Spatiotemporal Dynamics of Ecosystem Service Value and Its Linkages with Landscape Pattern Changes in Xiong’an New Area, China (2014–2022)

Abstract

As China’s third national-level new area, Xiong’an New Area plays a pivotal strategic role in relocating non-capital functions from Beijing while serving as a model for sustainable urban development. This study investigates the spatiotemporal evolution of ecosystem service value (ESV) and landscape patterns in Xiong’an before (2014–2016) and after (2017–2022) its establishment, assessing the policy-driven impacts of green development initiatives. Using remote sensing data, random forest classification, and landscape pattern analysis, we quantified land use dynamics, landscape index, and ESV variations. Key findings reveal significant land use transformations, with cultivated land declining by 7.51% and coniferous forest expanding by 189.84%, driven by urbanization and afforestation efforts. The comprehensive land use dynamic degree reached 4.96% (2014–2022), while the land use intensity index decreased by 20.95%. Concurrently, the fragmentation index increased significantly (Diversity Index (SHDI) +45%; Edge Density (ED) +66.23%). Despite these changes, ESV surged by 57.51% (CNY 334.63 billion), primarily due to wetland and forest expansion. Statistical analysis revealed positive correlations between ESV and the fragmentation index (ED, NP, and SHDI), whereas the aggregated index (CONTAG and AI) exhibited negative correlations. The findings substantiate the policy effectiveness of Xiong’an’s ecological initiatives, revealing how strategic landscape planning can balance urban development with ecosystem protection, offering valuable guidance for sustainable urbanization in Xiong’an and comparable regions.

1. Introduction

Ecosystem services encompass the diverse material resources and environmental conditions that ecosystems provide to sustain global life systems. These services, categorized into provisioning, regulating, cultural, and supporting functions, constitute fundamental prerequisites for organism survival and human wellbeing enhancement [1]. Ecosystem service value (ESV), quantitatively representing the ecological benefits humans obtain from ecosystems [2], has emerged as a pivotal metric for evaluating ecosystem security and facilitating sustainable development. The methodological framework for ESV assessment has undergone substantial refinement since Costanza et al. (1997). [3] pioneered the economic valuation approach. Subsequent landmark work by Xie et al. (2003). [4] established a China-specific terrestrial ecosystem evaluation system that has become a benchmark for domestic studies [5]. Contemporary research has expanded the methodological repertoire, investigating spatio-temporal dynamics [6,7] and underlying drivers [8,9] while increasingly integrating ESV with ecological risk assessment [10] and ecosystem health evaluation [11].

Landscape pattern, serving as a crucial indicator of regional ecological integrity, characterizes the spatial configuration, scale, and geometry of landscape components [12]. Its temporal evolution embodies the complex interplay between anthropogenic activities and natural systems, predominantly mediated through land use/cover changes [13]. Structural variations in landscapes directly modulate the typology, magnitude, and quality of ESV, as heterogeneous landscape units deliver differentiated ecosystem functions. Land use, being the spatial manifestation of human socioeconomic activities, has exerted progressively intensifying pressures on ecological systems amid technological progress and urban expansion [14,15]. The accelerated urbanization process has significantly altered landscape patterns and ecosystems functionalities, thereby reshaping ESV distributions [16]. Deciphering these intricate relationships is paramount for achieving equilibrium between ecological conservation and socioeconomic advancement.

The spatial heterogeneity of landscapes governs material flows and energy transfer within ecosystems, fundamentally regulating ecological processes [17,18]. Perturbations to landscape architecture may compromise ecosystem functionality and consequently modify ESV [19,20]. This causal linkage has stimulated growing scholarly interest in landscape–ESV interactions [21,22], with recent investigations spanning diverse geographical contexts including urban agglomerations [23,24,25,26], river basins [27,28,29], wetland systems [30,31,32], and coastal regions [33]. These studies reveal that different landscape types contribute unevenly to ESV and that landscape changes can have either positive or negative impacts on ecosystem value.

Since its establishment in 2017, Xiong’an New Area assumes strategic importance in decentralizing Beijing’s non-capital functions and fostering coordinated development within the Beijing–Tianjin–Hebei megalopolis [34]. Xiong’an possesses distinctive ecological assets, most notably the Baiyangdian Wetland—the largest freshwater wetland in North China. However, intensive urbanization has precipitated dramatic transformations in Xiong’an’s landscape configuration, raising pertinent questions about its capacity to realize ecologically sustainable development. Current scholarship on Xiong’an has predominantly examined land use change characteristics [35], ESV assessment methodologies [36], and landscape pattern dynamics [37] as isolated phenomena, leaving significant knowledge gaps regarding their synergistic relationships and mutual feedback mechanisms. The unique policy environment, combining green development mandates with Beijing’s functional relocation, has fundamentally reshaped land use patterns through demographic redistribution and industrial reorganization, with cascading effects on ESV and landscape structure—aspects that remain underexplored in the existing literature.

Through integrated application of a random forest classification algorithm, landscape index, and ESV quantification for the 2014–2022 period, this study establishes spatial correlations between landscape patterns and ESV, ultimately proposing differentiated ecosystem management strategies across ESV gradient zones. This study aims to (1) systematically characterize the evolutionary patterns of land use, ESV, and landscape patterns along with their interrelationships, thereby providing scientific support for Xiong’an’s urban planning and management, and (2) analyze policy impacts on ESV and landscape patterns to establish an empirical foundation for sustainable development strategies in the New Area.

2. Study Area and Data Sources

2.1. Study Area

Strategically positioned in the core of the Beijing–Tianjin–Hebei urban agglomeration, Xiong’an New Area (38°43′27″–39°10′20″ N, 115°38′03″–116°19′56″ E) represents China’s third generation of special economic zones following Shenzhen and Pudong. Established in 2017, this approximately 2000 km² jurisdiction encompassing the Xiong, Rongcheng, and Anxin counties (Figure 1) embodies the national strategy for decentralized development, particularly the relocation of Beijing’s non-administrative functions. The region prioritizes ecological civilization principles, aspiring to establish an international model of hydro-urban integration and sustainable habitation [34].

The land cover composition features a predominance of cultivated land, urban construction land, and aquatic systems, with fragmented distributions of woodland and grassland. Characterized by a warm temperate monsoon climate (mean annual temperature: 12.4 °C; precipitation: 495.1 mm) [38], the terrain exhibits gentle northwest-to-southeast topographic gradients (elevation range: 5–26 m) [39]. Predominant soil types include chernozems and cambisols, with extensive hydrological networks comprising the Nanjuma and Daqing rivers, Baigou Irrigation Channel, and, the crown jewel, Baiyangdian Lake—North China Plain’s largest freshwater body occupying the southeastern sector [40].

2.2. Data Sources

This study integrates multi-source data encompassing Landsat images, high-resolution remote sensing images, net primary productivity (NPP), soil conservation data, precipitation records, grain output, and price statistics (

Table 1. Data information.

Data Type	Spatial Resolution	Expiration Date	Data Source
Landsat Images	30 m	2014, 2016, 2019, 2022	www.usgs.gov (accessed on 6 May 2023)
High-Resolution Images	1 m	2014, 2016, 2019, 2022	Land Satellite Remote Sensing Application Center, China
Net Primary Productivity	500 m	2000–2020	NASA
Soil Conservation	250 m	2020	www.gis5g.com (accessed on 6 May 2023)
Precipitation	-	1990–2020	Meteorological Station in Baoding
Grain Output and Prices	-	2020, 2021	Government Statistical Report

).

3. Methods

3.1. Random Forest Classification Method

The random forest algorithm, an ensemble learning method, constructs multiple decision trees through bootstrap sampling of training data and aggregates their predictions for final classification. By leveraging random feature selection during tree construction, the algorithm demonstrates superior accuracy and robustness in remote sensing classification compared to other machine learning methods [41].

For land use classification, this study acquired four high-resolution (1 m) satellite images (2014, 2016, 2019, and 2022) to collect >200 classification samples. These samples were randomly partitioned into training (60%) and validation (40%) datasets for supervised classification of Landsat images. To improve phenological discrimination, sixteen Landsat images (four seasonal images per year) were preprocessed through radiometric calibration, atmospheric correction, and mosaicking, resulting in 15 m resolution imagery.

3.2. Land Use Dynamics

Land use dynamics represent a crucial quantitative indicator for assessing the intensity of spatiotemporal patterns of land use change. These dynamics are typically categorized into two distinct metrics: single land use dynamic degree (SLUDD) and comprehensive land use dynamic degree (CLUDD) [28].

SLUDDs quantify the overall intensity of land use changes in a specific land use type over a given period, with positive and negative values indicating an increase or decrease in land use intensity, respectively. The absolute magnitude of the SLUDD quantitatively reflects the transformation magnitude, with larger values indicating more substantial land use changes [42]. The formula is as follows:

$$\mathrm{S}\mathrm{L}\mathrm{U}\mathrm{D}\mathrm{D}={\displaystyle \frac{{\mathrm{U}}_{\mathrm{b}}-{\mathrm{U}}_{\mathrm{a}}}{{\mathrm{U}}_{\mathrm{a}}}}\times {\displaystyle \frac{1}{\mathrm{T}}}\times 100\mathrm{\%}$$

(1)

where ${\mathrm{U}}_{\mathrm{a}}$ and ${\mathrm{U}}_{\mathrm{b}}$ denote the area of the land use type at the beginning and end of the study period, respectively, and $\mathrm{T}$ is the study period in years.

CLUDDs quantify the overall intensity of land use changes in a region [43]. The formula is as follows:

$$\mathrm{C}\mathrm{L}\mathrm{U}\mathrm{D}\mathrm{D}={\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{∆\mathrm{U}}_{\mathrm{i}\mathrm{j}}}{2{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{U}}_{\mathrm{i}}}}\times {\displaystyle \frac{1}{\mathrm{T}}}\times 100\mathrm{\%}$$

(2)

where ${∆\mathrm{U}}_{\mathrm{i}\mathrm{j}}$ is the absolute value of the conversed area from land use type $\mathrm{i}$ to type $\mathrm{j}$; ${\mathrm{U}}_{\mathrm{i}}$ is the area of land use type $\mathrm{i}$ at the beginning of the study period; and $\mathrm{T}$ is the study period in years.

3.3. Land Use Degree

The land use degree (LUD), originally developed by Liu [44], provides a comprehensive assessment integrating both natural land attributes and anthropogenic–environmental interactions. This widely adopted framework [45,46] classifies land use types into four hierarchical grades (

Table 2. Land use degree grading assignment.

	Unused Land	Ecological Land	Agricultural Land	Urban Settlement Land
Land Use Type	Unused Land, Snow and Ice	Coniferous Forest, Broad-leaved Forest, Grassland, Wetland, Water body	Cultivated Land	Construction Land
Grade	1	2	3	4

), each assigned a quantitative weight reflecting its ecological transformation intensity. The formula is as follows:

$$\mathrm{L}\mathrm{U}\mathrm{D}=100\times {\sum }_{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{A}}_{\mathrm{i}}\times {\mathrm{C}}_{\mathrm{i}})$$

(3)

$$\mathrm{R}={\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{A}}_{\mathrm{i}}\times {\mathrm{C}}_{\mathrm{i}\mathrm{y}})-{\sum }_{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{A}}_{\mathrm{i}}\times {\mathrm{C}}_{\mathrm{i}\mathrm{x}})}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{A}}_{\mathrm{i}}\times {\mathrm{C}}_{\mathrm{i}\mathrm{x}})}}$$

(4)

where ${\mathrm{A}}_{\mathrm{i}}$ refers to the index of LUD at grade $\mathrm{i}$, and ${\mathrm{C}}_{\mathrm{i}}$ represents the percentage of grade $\mathrm{i}$. $\mathrm{R}$ refers to the change rate of LUD. ${\mathrm{C}}_{\mathrm{i}\mathrm{x}}$ and ${\mathrm{C}}_{\mathrm{i}\mathrm{y}}$ refers to the percentage of grade $\mathrm{i}$ in period x and $\mathrm{y}$.

3.4. Landscape Pattern Index

Land use transformations, driven by the combined effects of natural and socio-economic factors, induce substantial modifications to landscape spatial configuration [47,48,49]. To systematically characterize these changes in Xiong’an, seven landscape-level indices were selected to describe the patch distribution patterns (

Table 3. List of landscape pattern indices.

Name	Unit	Value Range	Name	Unit	Value Range
Patch Density (PD)	pcs/100 ha	(0, +∞)	Shannon’s Diversity Index (SHDI)	-	[0, +∞)
Edge Density (ED)	m/ha	(0, +∞)
Division Index (DIVISION)	%	(0, 100]	Aggregation Index (AI)	%	(0, 100]
Mean Area (AREA_MN)	ha	(0, +∞)	Contagion Index (CONTAG)	%	(0, 100]

).

3.5. Ecosystem Service Value

Based on the ecosystem classification proposed by Xie et al.(2008) [50] and distinct ecological attributes, Xiong’an was divided into seven ecosystem types: grassland, cultivated land, construction land, broad-leaved forest, wetland, water body, and coniferous forest. ESV was calculated as follows:

$$\mathrm{E}\mathrm{S}\mathrm{V}=\sum _{\mathrm{i}=1}^{\mathrm{n}}({\mathrm{A}}_{\mathrm{i}}{\times \mathrm{e}\mathrm{v}}_{\mathrm{i}}\times \mathrm{v}\times {\mathrm{F}}_{\mathrm{i}})$$

(5)

where $\mathrm{n}=7$ refers to the number of ecosystems, ${\mathrm{A}}_{\mathrm{i}}$ represents the area of ecosystem type $\mathrm{i}$, ${\mathrm{e}\mathrm{v}}_{\mathrm{i}}$ refers to the equivalent value of ecosystem type $\mathrm{i}$, $\mathrm{v}$ refers to the ESV of the equivalent factor of standard unit in Xiong’an, and ${\mathrm{F}}_{\mathrm{i}}$ refers to the adjustment factor of equivalent.

3.5.1. Equivalent Factor of Standard Unit

According to the official agricultural statistics, the annual grain output of Xiong’an is 710,400 tons, and the average selling price of the main grain products in China is 141.5 CNY/50 kg [51,52]. Accounting for both the direct economic output and associated human inputs within natural ecosystems, the equivalent factor of a standard unit was determined as 1/7 of the annual crop value per hectare (13,182.56 CNY/ha).

3.5.2. Adjustment Factors

Ecosystem structures and morphologies exhibit significant spatiotemporal heterogeneity across regions and seasonal periods, resulting in dynamic variations in their ecological service functions and corresponding values. Previous research [53] has demonstrated distinct relationships between specific ecosystem services and environmental factors: The internal structure and external morphology of ecosystems constantly vary across different regions and during different periods within the same year, consequently leading to continuous changes in their ecological service functions and associated values. Based on previous studies [53], ecosystem services including (1) food production, raw material production, gas regulation, climate regulation, environment purification, maintenance of nutrient cycles, maintenance of biodiversity, and provision of esthetic landscapes (i1) generally exhibit positive correlations. (2) Water resource supply and horologic regulation (i2) are associated with precipitation variations. (3) Soil conservation (i3) shows strong dependencies on terrain slope and soil properties. Comprehensive analysis revealed NPP, precipitation, and soil conservation as the three primary dynamic regulatory factors. The equivalent table was adjusted using the following formula [54,55]:

$${\mathrm{F}}_{\mathrm{i}\mathrm{j}}=\left\{\begin{array}{c}{\displaystyle \frac{{\mathrm{N}\mathrm{P}\mathrm{P}}_{\mathrm{x}}}{{\mathrm{N}\mathrm{P}\mathrm{P}}_{\mathrm{c}}}}\times {\mathrm{F}}_{\mathrm{i}1}\\ {\displaystyle \frac{{\mathrm{P}}_{\mathrm{x}}}{{\mathrm{P}}_{\mathrm{c}}}}\times {\mathrm{F}}_{\mathrm{i}2}\\ {\displaystyle \frac{{\mathrm{S}\mathrm{C}}_{\mathrm{x}}}{{\mathrm{S}\mathrm{C}}_{\mathrm{c}}}}\times {\mathrm{F}}_{\mathrm{i}3}\end{array}\right.$$

(6)

where ${\mathrm{F}}_{\mathrm{i}\mathrm{j}}$ is the equivalent factor per unit area for services type $\mathrm{j}$ of ecosystem type $\mathrm{i}$; ${\mathrm{N}\mathrm{P}\mathrm{P}}_{\mathrm{x}}$ and ${\mathrm{N}\mathrm{P}\mathrm{P}}_{\mathrm{c}}$ are the average annual NPP of Xiong’an and China, respectively; ${\mathrm{P}}_{\mathrm{x}}$ and ${\mathrm{P}}_{\mathrm{c}}$ are the average annual precipitation of Xiong’an and China, respectively; ${\mathrm{S}\mathrm{C}}_{\mathrm{x}}$ and ${\mathrm{S}\mathrm{C}}_{\mathrm{c}}$ are the average soil conservation of Xiong’an and China, respectively; and ${\mathrm{F}}_{\mathrm{i}}$ refers to the ESV equivalent factor of ecosystem type $\mathrm{i}$.

3.5.3. The Equivalent of ESV

The adjusted equivalent factors per unit area of Xiong’an are shown in

Table 4. The equivalent of ESV per unit area in Xiong’an.

Type of Ecosystem	Supply Service			Regulating Service				Support Service			Cultural Service
	Food Production	Raw Material Production	Water Resource Supply	Climate Regulation	Gas Regulation	Environment Purification	Horologic Regulation	Soil Conservation	Maintenance of Nutrient Cycles	Maintenance of Biodiversity	Provision of Esthetic Landscapes
Grassland	1.06	1.57	0.11	5.51	14.56	4.81	1.33	0.04	0.52	6.10	2.69
Cultivated Land	3.88	1.83	0.01	3.06	1.64	0.46	0.15	0.03	0.55	0.59	0.27
Construction Land	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Broad-leaved Forest	1.32	3.01	0.19	9.90	29.66	8.81	2.70	0.08	0.91	11.00	4.84
Wetland	2.33	2.28	1.48	8.67	16.43	16.43	13.81	0.07	0.82	35.91	21.58
Water Body	3.65	1.05	4.73	3.51	10.45	25.33	58.28	0.03	0.32	11.64	8.62
Coniferous Forest	1.00	2.37	0.15	7.76	23.14	6.80	1.90	0.06	0.73	8.58	3.74

.

4. Results

4.1. Analysis of Land Use Dynamic Changes in Xiong’an

4.1.1. Land Use Classification

This study systematically monitored the transformation of Xiong’an New Area across four temporal nodes (2014, 2016, 2019, and 2022; Figure 2). The land use classification scheme incorporated seven distinct categories: grassland, cultivated land, construction land, wetland, water bodies, coniferous forest, and broad-leaved forest. This classification was achieved through an integrated approach combining a random forest algorithm processing with manual visual interpretation to ensure accuracy. Validation results demonstrated consistently high classification accuracy exceeding 80% across all study periods (2014: 83.38%; 2016: 84.93%; 2019: 89.16%; 2022: 88.22%), confirming the robustness of the adopted methodology.

4.1.2. Analysis of Land Use Dynamics

• Single Land Use Dynamics

The SLUDD of different land use types in Xiong’an are quantitatively presented in

Table 5. Single land use dynamics in Xiong’an.

Types	2014–2016	2016–2019	2019–2022	2014–2022
Grassland	30.98%	−17.69%	164.24%	43.81%
Cultivated Land	−2.79%	−11.62%	−11.68%	−7.51%
Construction Land	12.70%	1.06%	2.85%	5.06%
Broad-leaved Forest	−5.96%	6.42%	6.99%	3.38%
Wetland	5.82%	37.22%	1.34%	18.22%
Water Body	−7.03%	−17.09%	17.65%	−4.49%
Coniferous Forest	5.24%	324.57%	12.15%	189.84%

. Combined with the results of classification and area transfer, spatiotemporal patterns of land use changes were analyzed through the following aspects:

(1)

Grassland Changes. Grassland coverage exhibited a net increase (dynamic degree: 43.81%; annual rate: 5.48%) during 2014–2022, following an initial decline-then-recovery pattern. Remote sensing analysis revealed that this trend resulted from vegetation colonization on abandoned construction sites near the start-up area of Xiong’an in Rongcheng County, where the demolition of villages and the suspension of constructions created transitional grasslands (Figure 3).

(2)

Construction Land Changes. Construction land grew steadily (dynamic degree: 5.06%; annual rate: 0.63%), peaking during 2014–2016 due to village expansions in Rongcheng’s Zhang Town region (Figure 4). However, following the official establishment of Xiong’an New Area, the growth rate was moderated significantly due to two key factors: (i) stringent government land use regulations, implemented to ensure planned development, and (ii) pandemic-related disruptions to construction activities during the COVID-19 outbreak period. This dual regulatory and exogenous shock resulted in a notable deceleration of urban expansion compared to the pre-establishment phase.

(3)

Afforestation Changes. Forest cover changes revealed divergent trends: (i) broad-leaved forests saw a gradual increase (dynamic degree: 3.38%; annual rate: 0.42%), and coniferous forests saw dramatic expansion (dynamic degree: 189.84%; annual rate: 23.73%, particularly during 2016–2019). The “Millennium Forest” project, launched in 2017, achieved remarkable results. By 2022, it had successfully afforested an area of 30,000 hectares with 23+ million trees, elevating forest coverage from 11% to 32% (Figure 5).

(4)

Cultivated Land Changes. Cultivated land demonstrated a consistent decline (dynamic degree: −7.51%; annual rate: −0.94%), with accelerated losses post-2016. This transformation primarily supported two development priorities: (i) urban expansion around the start-up area in Rongcheng County (Figure 4) and (ii) ecological afforestation initiatives in Xiong County (Figure 5). Approximately 30% of converted farmland transitioned to coniferous forests, aligning with Xiong’an’s eco-city planning objectives.

(5)

Water Body and Wetland Changes. Water bodies fluctuated seasonally (dynamic degree: −4.49%; annual rate: −0.56%), with precipitation and image acquisition timing significantly influencing measurements. Wetlands expanded markedly (dynamic degree: 18.22%; annual rate: 2.28%), especially around Anxin County’s Yangdiko Village, where climate-driven rainfall increases facilitated substantial areal growth during 2016–2019.

2.

Comprehensive Land Use Dynamics

Analyzing the CLUDD the four periods in Xiong’an revealed distinct pattern (

Table 6. CLUDD in Xiong’an.

Period	2014–2016	2016–2019	2019–2022	2014–2022
Dynamic Degree	2.63%	7.88%	4.48%	4.96%

). During the entire study period (2014–2022), the total CLUDD reached 4.96%, corresponding to an annual change rate of 0.62%. This progression followed a characteristic “slow-fast-slow” trajectory:

(1)

Initial phase (2014–2016): exhibited relatively modest land use changes, reflecting baseline development conditions prior to major policy interventions.

(2)

Accelerated phase (2016–2019): marked by the most substantial transformations, directly coinciding with the formal establishment of Xiong’an New Area and its subsequent initial development surge.

(3)

Moderated phase (2019–2022): showed a gradual deceleration in land use dynamics, attributable to external constraints including the COVID-19 pandemic’s impacts on construction activities.

4.1.3. Analysis of Land Use Degree Changes

From 2014 to 2022, the land use degree in Xiong’an declined by 20.95%, with an annual reduction rate of 2.62% (

Table 7. LUD and Rate of Change.

Land Use Degree	2014	2016	2019	2022
	226.36	216.80	195.93	178.92
Rate of Change	2014–2016	2016–2019	2019–2022	2014–2022
	−4.22%	−9.63%	−8.68%	−20.95%

). The most significant decrease occurred between 2016 and 2019, while the slowest decline occurred between 2014 and 2016. After the establishment of Xiong’an New Area, the areas of grassland, wetland, and coniferous forest experienced substantial expansion, contributing to an overall enhancement of the ecological environment. The ecological spatial pattern of the region, characterized as “one marsh, three belts, nine patches, and multiple corridors,” has begun to take shape. The ongoing implementation of the “Millennium Forest” project, the construction of landscape belts, and the management of the Baiyangdian wetland have collectively fostered a more sustainable and improved ecological environment in Xiong’an. The decrease in the LUD reflects Xiong’an’s ecological transition, where expansive afforestation and wetland restoration (lower-intensity grades) outweighed limited construction growth (higher-intensity grade).

4.2. Analysis of Landscape Pattern Changes in Xiong’an

Seven landscape-level pattern indices, including the AREA_MN, SHDI, DIVISION, ED, CONTAG, NP, and AI, were systematically analyzed (Figure 6), yielding the following key findings:

• The NP in Xiong’an showed a marked increase, while the AREA_MN significantly decreased, indicating enhanced landscape fragmentation and finer patch compositions. Combined with SLUDD analysis, these patterns reveal frequent land use conversions, fragmented originally contiguous patches into smaller, more heterogeneous spatial units. Quantitatively, the AREA_MN declined from 1.44 ha to 0.85 ha, while the NP increased from 69.26 pcs/ha to 118.24 pcs/ha, demonstrating growing spatial heterogeneity per unit area.
• The ED, reflecting patch shape complexity, increased by 66.23% (from 127.42 to 211.81 m/ha) during 2014–2022. This pronounced growth indicated that intensified anthropogenic activities altered patch geometries substantially, strengthening ecological interactions between adjacent patch types through more convoluted boundaries.
• The AI showed an overall decreasing trend, while the DIVISION showed a slightly increasing trend, signaling progressive dispersal of landscape patches. These complementary metrics indicate deteriorating spatial cohesion, where formerly aggregated patches became increasingly fragmented, compromising landscape structural integrity.
• The SHDI is sensitive to the non-equilibrium distribution of landscape patches. Between 2014 and 2022, the SHDI increased by 45%, signifying a more balanced proportion of different landscape types. This shift was accompanied by a weakening of the dominant role of cultivated land, with portions of it being converted into other patch types. The CONTAG further supported this conclusion. With a 33.79% decrease, the CONTAG indicated a decline in the connectivity of dominant patches, consistent with the ongoing subdivision of large patches into small ones.

4.3. Analysis of ESV in Xiong’an

The ESV scores of Xiong’an in 2014, 2016, 2019, and 2022 are presented in

Table 8. ESV of Xiong’an.

Land Use Type	ESV of 2014 (108 Yuan/a)	ESV of 2016 (108 Yuan/a)	ESV of 2019 (108 Yuan/a)	ESV of 2022 (108 Yuan/a)
Grassland	4.48	7.25	3.40	20.17
Cultivated Land	159.98	151.04	98.38	63.91
Construction Land	0.00	0.00	0.00	0.00
Broad-leaved Forest	117.18	103.20	123.08	148.88
Wetland	135.85	151.65	320.97	333.85
Water Body	148.57	127.69	62.23	95.18
Coniferous Forest	15.72	17.36	186.44	254.40
Total	581.77	558.20	794.50	916.40

and Figure 7. The analysis revealed a consistent upward trajectory in ESV from 2014 to 2022, with a total increase of CNY 33.46 billion, representing 57.51% growth, and an annual growth rate of CNY 4.18 billion. This substantial enhancement in ESV primarily resulted from significant expansions in wetland, grassland, and coniferous forest, reflecting a clear improvement in both the ecological pattern and environmental quality of Xiong’an. Regarding ESV per unit area, the value of different land use types was as follows: water body > wetland > broad-leaved forest > coniferous forest > grassland > cultivated land > construction land. Taking 2022 as an example, wetlands and water bodies collectively contributed CNY 42.9 billion to total ESV, accounting for 46.82%. Meanwhile, grasslands, broad-leaved forests, and coniferous forests together generated CNY 42.35 billion, constituting 46.21% of total ESV. In contrast, cultivated lands and construction lands made minimal contributions, amounting to merely CNY 639 million and representing just 6.97% of total ESV.

4.4. Correlation Analysis Between ESV and Landscape Pattern Indices

The correlation between ESV and the landscape pattern indices in Xiong’an from 2014 to 2022 is shown in Figure 8. The ED, NP, SHDI, and DIVISION showed a positive correlation with ESV, while the CONTAG, AI, and AREA_MN showed a negative correlation. These findings collectively indicate that increased landscape heterogeneity and fragmentation in Xiong’an—characterized by greater patch-type diversity and finer spatial subdivision—are associated with enhanced ecosystem service provision.

5. Discussion

5.1. Changes in ESV and Landscape Pattern of Xiong’an

Previous studies [56,57,58,59,60,61,62] have demonstrated that regional disparities in economic development and land use patterns exert heterogeneous impacts on ESV. In Xiong’an, our analysis reveals a distinct temporal pattern: ESV declined from 2014 to 2016, followed by a consistent upward trend from 2016 to 2022. The peak ESV observed in 2014 can be attributed to two primary factors: (i) national agricultural policies [63], particularly the increased minimum purchase price for rice and corn supply shortages caused by livestock industry demands, triggered substantial grain price surges in Hebei Province; and (ii) meteorological records from Baoding Station indicate exceptionally high cumulative rainfall between 2012 and 2014 (peaking at 601.4 mm/year in 2013—double the 2019 level), which significantly expanded the water area of Baiyangdian. The period of 2016–2022 witnessed a remarkable ESV increase of CNY 35.82 billion (57.82% of total growth), averaging CNY 5.97 billion annually. This surge principally stemmed from dramatic area expansions in coniferous forests (+93.18%), grasslands (+64.06%), and broad-leaf forests (+30.79%), collectively enhancing Xiong’an’s ecological quality and spatial patterns. Increased wetland and forest coverage (ESV +57.51%) enhance air/water purification, reducing respiratory diseases, and the absorption capacity of PM2.5 in Baiyangdian has increased by approximately 22% since 2017 [38]. Cultural ESV surged by 278%, and in 2022, Baiyangdian received 5.3 million tourists and generated CNY 1.2 billion in revenue [51]. Wetland restoration boosted fisheries (+15% yield since 2017), directly benefiting 12,000 local fishers [64].

Studies on China’s two other national-level new areas demonstrate that their establishment exerts significant and lasting impacts on landscape patterns. For instance, Gan [65] and Wu [66] investigated landscape changes in the Shenzhen Special Economic Zone and Pudong New Area before and after their establishment, respectively. Both studies observed an initial rise followed by a decline in the NP and ED, alongside a decrease-then-increase trend in the AI and CONTAG. This pattern suggests that new areas typically undergo a phase of disordered expansion (landscape fragmentation) followed by policy-driven consolidation (landscape aggregation). Our analysis of Xiong’an reveals that construction activities have reshaped its spatial configuration, though the region currently remains in the early stage of unregulated development with rapidly escalating fragmentation. Since its inception, government-led infrastructure projects have increased the NP and ED while reducing the AREA_MN. Concurrently, ecological interventions such as roadside greening, afforestation, and the Millennium Forest Project have enhanced landscape heterogeneity, as indicated by declining AI and CONTAG values coupled with rising DIVISION and SHDI values. Rapid urbanization, while driving economic growth, tends to exacerbate landscape fragmentation and subsequently induce a cascade of negative ecological impacts. In Xiong’an, urban expansion, transportation network construction, and infrastructure development will fragment contiguous habitats into isolated patches. Concurrently, wetland degradation in the Baiyangdian Basin may shrink the living space of waterbirds, fish, and other species, reduce biodiversity, and increase the edge-to-area ratio of habitat patches. These changes could further aggravate water eutrophication.

5.2. Impacts of Landscape Pattern Changes on the Ecosystem in Xiong’an

Changes in landscape patterns significantly influence the composition, structure, functioning, and biogeochemical processes of regional ecosystems, ultimately leading to fluctuations in ESV [25]. The relationship between landscape patterns and ESV exhibits distinct regional characteristics. For instance, in the Yanbian area, Yu et al. [67] observed a strong positive correlation between ESV and AI and CONTAG values, but a negative correlation with the PD, LSI, and SHDI. Conversely, Zhang et al. [68] reported that ESV in karst regions of northwest Guangxi was negatively associated with the DIVISION and SHDI. In contrast, Cen et al. [69] found that in the southern Hangzhou Bay area, greater land use diversity, fragmentation, and landscape heterogeneity contributed to higher overall ESV.

In Xiong’an (2014–2022), cultivated land remained the dominant matrix type. However, rapid urbanization and large-scale afforestation accelerated the conversion of other land use types into construction land and coniferous forests, increasing landscape fragmentation and shape complexity. Spatiotemporal analysis revealed that the ED, NP, SHDI, and DIVISION were positively correlated with ESV, whereas the CONTAG, AI, and AREA_MN exhibited negative correlations. Notably, ESV in Xiong’an increased alongside rising landscape fragmentation.

Following the establishment of Xiong’an New Area in 2017, the shift from traditional urban sprawl to an ecologically prioritized development model significantly enhanced regional sustainability. The “Millennium Forest Project” played a pivotal role in improving ecological quality, particularly in southeastern Xiongxian County—a finding consistent with Liu Liqun et al.’s [70] assessment of ESV gains (2014–2020). While conventional urban expansion typically degrades ecological quality [71], Xiong’an’s afforestation of unused and cultivated lands established a robust green infrastructure system, substantially boosting regional ESV [72].

5.3. Suggestions for the Zoned Protection and Development of ESV in Xiong’an

• The high-ESV areas in Xiong’an are predominantly concentrated in and around Baiyangdian. As an ecological protection red-line zone centered on Baiyangdian, this region demonstrates exceptional capabilities in water retention, soil conservation, and biodiversity preservation [73]. The ecological fragility of this area makes it the cornerstone of regional ecological security, necessitating the implementation of the most rigorous environmental protection measures [74]. Enhanced protection of aquatic and wetland ecosystems should be prioritized, along with comprehensive watershed management. Only ecologically beneficial human activities that preserve and potentially enhance wetland ecosystem functions should be permitted, including sediment removal, water quality assessment, aquatic vegetation restoration, and water purification initiatives [75].
• The medium-ESV zones in Xiong’an are primarily situated in Xiongxian County, where forest and grassland ecosystems prevail. Sustained efforts should focus on optimizing forest ecosystem architecture, enhancing woodland ecological functions, and reinforcing conservation measures [76]. Through the “Project of Millennium Forest”, targeted ecological management should be implemented to maintain and improve ecosystem services, including forest landscape enhancement, therapeutic forestry projects, trail system development, cultural landscape restoration, and irrigation infrastructure improvement [77].
• The low-ESV regions in Xiong’an are chiefly distributed across Rongcheng County, characterized by agricultural and urban land uses. For cultivated areas, Xiong’an should develop phased implementation strategies to systematically transform all permanent basic farmland into contiguous high-standard farmland, adhering to principles of spatial consolidation [78]. For urbanized areas, Xiong’an should facilitate synergistic development across primary, secondary, and tertiary sectors by attracting advanced agricultural technologies and enterprises relocating from Beijing. Building upon the region’s resource endowments and agricultural foundations, the New Area should nurture competitive local enterprises [79], with particular emphasis on developing specialty industries including premium vegetable production, authentic Chinese herbal medicine cultivation, and lotus leaf tea processing [80].
• Landscape pattern indices serve as a robust quantitative framework for assessing ecological impacts in Xiong’an. Through systematic monitoring and policy integration, these indices enable dynamic optimization of territorial spatial planning to achieve ecologically prioritized development, thereby circumventing the conventional “develop-first, remediate-later” urbanization paradigm. The implementation of an index threshold system facilitates a closed-loop governance cycle encompassing monitoring, evaluation, and policy response. Elevated SHDI values suggest landscape heterogeneity degradation, necessitating immediate land use restructuring [81]; declining CONTAG values mandate the incorporation of ecological corridors to enhance connectivity among fragmented green spaces and wetland patches; and sustained increases in ED require establishing protective buffers around construction zones to mitigate anthropogenic impacts on core ecological areas [82].

6. Conclusions

• Changes in land development patterns and reductions in intensity improve the ecological environment. The land use dynamics in Xiong’an exhibited a phased growth pattern characterized by “slow–fast–slow” progression, accompanied by a consistent annual decline in land resource development intensity. Following Xiong’an’s establishment, substantial expansions were observed in grassland, wetland, coniferous forest, and construction land areas, contrasting with a persistent reduction in cultivated land. These transformations have collectively enhanced the regional ecological environment, facilitating the preliminary formation of the “one lake, three belts, nine patches, and multiple corridors” ecological spatial framework.
• The establishment of Xiong’an New Area has significantly contributed to population decentralization from Beijing and industrial agglomeration. From 2014 to 2022, Xiong’an experienced a population increase of 190,000 (+17%) and a GDP growth of CNY 13 billion (+60%), with the tertiary sector expanding notably by CNY 13.5 billion (+278%) [51,83]. The influx of new residents and industries has accelerated urban construction, altered land use patterns, and, consequently, impacted ESV and landscape configuration.
• ESV has been significantly improved. From 2014 to 2022, Xiong’an experienced a consistent upward trajectory in ESV, with a cumulative increase of CNY 33.46 billion (57.51% of total growth) and an average annual growth of CNY 4.183 billion. These results demonstrate marked improvements in both ecological structure and environmental quality.
• Landscape index analysis revealed positive correlations between ESV and the ED, NP, SHDI, and DIVISION indices, whereas negative correlations emerged with the CONTAG, AI, and AREA_MN indices. These findings indicate that increasing landscape fragmentation corresponds with elevated ESV in Xiong’an.
• In providing enlightenment for improving the rigor of ESV evaluation, this study employs the equivalent factor method to conduct ESV calculation. While this approach estimates value based on fixed coefficients, the results inherently carry certain margins of error. Subsequent research will refine the ESV calculation methodology by enhancing uncertainty analysis and adopting a tripartite framework—‘dynamic monitoring, local calibration, and policy simulation’—to improve the scientific rigor of ESV assessment.