Next Article in Journal
Methodological Framework for Evaluating Quarry Reclamation Based on the Reclamation Quality Index
Previous Article in Journal
Do Metropolitan Zoning Asymmetries Influence the Geography of Suburban Growth and Gentrification?
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Spatiotemporal Evolution and Driving Force Analysis of Habitat Quality in the Beibu Gulf Urban Agglomeration

1
Guangxi Key Laboratory of Environmental Processes and Remediation in Ecologically Fragile Regions, College of Environment and Resources, Guangxi Normal University, Guilin 541000, China
2
Key Laboratory for Earth Surface Processes of the Ministry of Education, College of Urban and Environmental Sciences, Peking University, Beijing 100871, China
*
Authors to whom correspondence should be addressed.
These authors contributed equally to this work.
Land 2025, 14(8), 1556; https://doi.org/10.3390/land14081556
Submission received: 17 June 2025 / Revised: 9 July 2025 / Accepted: 18 July 2025 / Published: 29 July 2025

Abstract

The ecological environment is crucial for human survival and development. As ecological issues become more pressing, studying the spatiotemporal evolution of ecological quality (EQ) and its driving mechanisms is vital for sustainable development. This study, based on MODIS data from 2000 to 2022 and the Google Earth Engine platform, constructs a remote sensing ecological index for the Beibu Gulf Urban Agglomeration and analyzes its spatiotemporal evolution using Theil–Sen trend analysis, Hurst index (HI), and geographic detector. The results show the following: (1) From 2000 to 2010, EQ improved, particularly from 2005 to 2010, with a significant increase in areas of excellent and good quality due to national policies and climate improvements. From 2010 to 2015, EQ degraded, with a sharp reduction in areas of excellent quality, likely due to urban expansion and industrial pressures. After 2015, EQ rebounded with successful governance measures. (2) The HI analysis indicates that future changes will continue the past trend, especially in areas like southeastern Chongzuo and northwestern Fangchenggang, where governance efforts were effective. (3) EQ shows a positive spatial correlation, with high-quality areas in central Nanning and Fangchenggang, and low-quality areas in Nanning and Beihai. After 2015, both high–high and low–low clusters showed changes, likely due to ecological governance measures. (4) NDBSI (dryness) is the main driver of EQ changes (q = 0.806), with significant impacts from NDVI (vegetation coverage), LST (heat), and WET (humidity). Urban expansion’s increase in impervious surfaces (NDBSI rise) and vegetation loss (NDVI decline) have a synergistic effect (q = 0.856), significantly affecting EQ. Based on these findings, it is recommended to control construction land expansion, optimize land use structure, protect ecologically sensitive areas, and enhance climate adaptation strategies to ensure continuous improvement in EQ.

1. Introduction

The ecological environment is the foundation of human survival and development [1]. It not only provides essential natural resources but also plays an irreplaceable role in climate regulation, biodiversity maintenance, and environmental purification. However, with the increasing impact of global climate change and human activities [2], ecological issues such as land degradation and water scarcity [3] have become more prominent, posing a severe threat to the stability of ecosystems and human living conditions. Particularly, the decline in ecological quality (EQ) caused by rapid urbanization [4] has become a global concern. Therefore, studying the spatiotemporal evolution of EQ and its driving mechanisms is of great significance for formulating scientific ecological protection policies and achieving sustainable development.
EQ refers to the capacity of ecosystems to support human well-being and socio-economic development within a specific spatial–temporal context. It serves as an important indicator of the degree of coordination between natural systems and human production activities [5]. In recent years, EQ evaluation methods can be broadly categorized into single-index and integrated model approaches. Single-index methods are typically based on specific ecological factors, such as the normalized difference vegetation index (NDVI) [6], vegetation coverage (FVC) [7], and enhanced vegetation index (EVI) [8]. These methods can quickly reflect the ecological environment status in a certain aspect. However, they fail to comprehensively reflect the complexity and diversity of ecosystems, particularly when facing the interactions of multiple factors, as a single index often provides insufficient evaluation results [9,10]. Integrated models, which combine multiple ecological factors, can more comprehensively reflect the overall status of the ecological environment. Common integrated models include the “pressure-state-response (PSR)” model [11], the coupling coordination degree model [12], and the “driving force-pressure-state-impact-response (DPSIR)” model [13]. These models can systematically analyze the interactions between ecological environments and socio-economic factors. However, the determination of weights in integrated models often relies on expert scoring, leading to subjectivity in the evaluation results.
In recent years, with the advancement of remote sensing, geographic information systems, and big data analysis, EQ evaluation has become increasingly quantitative and precise. Remote sensing technology provides extensive, multi-temporal, and multi-scale ecological data, greatly improving the efficiency of EQ assessments [14]. For instance, the remote sensing ecological index (RSEI) proposed by Xu [15,16] integrates remote sensing indices such as greenness, humidity, dryness, and heat (expressed by NDVI, WET, NDBSI and LST, respectively), providing a comprehensive and objective reflection of the spatiotemporal variation in regional EQ. The RSEI not only avoids the subjectivity of expert scoring in traditional integrated models but also enhances the scientificity and reliability of evaluation results through quantitative methods like principal component analysis [17]. As a result, RSEI has been widely applied in ecological environment evaluations in urban [18,19], watershed [20,21], and plateau regions [17,22], becoming one of the most important tools in current ecological environment research.
However, existing studies primarily focus on inland cities, watersheds, and plateau regions, while coastal EQ assessments and their driving force analyses are relatively scarce. Coastal regions, influenced by oceanic and atmospheric circulation, urban expansion, and human activities, present complex EQ evolution processes. These regions are driven by both natural factors (such as land–ocean interactions, typhoons, and tidal changes) and the profound impacts of human activities, including socio-economic development, land use change, and industrial restructuring. Therefore, traditional EQ evaluation methods struggle to comprehensively reveal the changes and driving mechanisms in coastal ecological environments. In terms of research methods and data sources, many studies use Landsat data for ecological evaluations [23,24,25], but due to the limitations of Landsat data in terms of temporal resolution and coverage it cannot meet the demands of large-scale, long-term ecological monitoring. In contrast, studies based on the Google Earth Engine (GEE) platform and MODIS data offer significant advantages for large-scale, long-term ecological environment monitoring. The GEE platform enables efficient processing of vast remote sensing data, supports multi-source data integration and analysis, and significantly reduces the time cost of data acquisition and processing [26]. Additionally, MODIS data, with higher temporal resolution, is well-suited for long-term ecological environment monitoring [27,28].
Thus, this study focuses on the Beibu Gulf Urban Agglomeration (BGUA), aiming to comprehensively reveal the spatiotemporal evolution characteristics, future development trends, and driving mechanisms of EQ, providing scientific evidence for ecological protection and sustainable development in the region. The selection of the BGUA as the study area is not only to address the limitations of previous research but also due to several other reasons: as a major economic growth hub in southern China, changes in EQ in the BGUA affect regional sustainable development and have significant implications for global climate change and biodiversity conservation. Despite the profound impact of rapid urbanization and economic development on the region’s ecological environment, EQ evaluation research in this area remains limited. Furthermore, a systematic analysis of the driving forces of EQ in the BGUA, particularly the interactions of various driving factors, has yet to be explored in depth.
To fill these gaps, this study constructs the RSEI for the BGUA based on the GEE platform and MODIS high-resolution data. Using methods such as Theil–Sen trend analysis, Mann–Kendall test, HI, and geographic detector, along with multidimensional driving factors such as natural, socio-economic, and topographical elements, the study aims to systematically reveal the evolution trends and future development of EQ, identify the dominant driving factors affecting ecological changes, and provide scientific evidence for regional ecological protection, environmental restoration, and sustainable development.

2. Materials and Methods

2.1. Study Area

The BGUA is located along the southern coast of China, encompassing key cities such as Nanning, Chongzuo, Fangchenggang, Qinzhou, Beihai, and Yulin (Figure 1). The geographical area spans approximately from 21°13′ N to 23°32′ N and from 106°19′ E to 110°40′ E. This region has a typical subtropical monsoon climate, with an average annual temperature ranging from 22 °C to 25 °C and annual precipitation between 1000 mm and 2000 mm. Vegetation is predominantly subtropical evergreen broadleaf forests, and the region’s ecological environment is complex and diverse. The BGUA coastal area serves as an important ecological barrier in southern China, hosting critical ecosystems such as mangroves, tidal wetlands, and coral reefs, which provide habitats for endangered species. However, with the rapid urbanization development of the BGUA, the ecological environment is facing mounting pressure. As a key gateway for cooperation between China and ASEAN countries, the BGUA plays a pivotal role in the Belt and Road Initiative. Therefore, studying the spatiotemporal evolution of EQ and its driving mechanisms in this region is vital for formulating effective ecological protection policies and achieving sustainable regional development.

2.2. Data Sources and Processing

This study is based on the GEE platform, utilizing MODIS data from the growing season (May to October) for each year from 2000 to 2022 (MOD13A1, MOD11A2, and MOD09A1) to extract NDVI, WET, NDBSI, and LST (representing greenness, wetness, dryness, and heat, respectively). These indices were coupled to construct the RSEI. The annual ecological indicator data were preprocessed through steps including cloud removal, water masking, clipping, mosaicking, and resampling. Then, the ecological indicator raster data for each year underwent median synthesis to eliminate the influence of meteorological factors such as clouds, atmosphere, and solar zenith angle.
Based on the geographical characteristics of the BGUA and existing research, this study selected four indicators of RSEI (NDVI, WET, NDBSI, and LST) as internal factors, along with nine external factors (natural, topographical, socio-economic, and land use factors) as listed in Table 1, to comprehensively analyze the region’s ecological changes [29,30]. After preprocessing, all data were resampled to a uniform spatial resolution of 500 m × 500 m.

2.3. Methodology

2.3.1. Construction of RSEI Based on the GEE Platform

Based on the GEE platform, this study calculates the following four key ecological environment indicators: vegetation greenness (NDVI), surface humidity (WET), surface dryness (NDBSI), and heat (LST). These indicators collectively form the core components of the RSEI. Specifically, NDVI is used to represent vegetation cover and reflects the health status of regional ecosystems; WET measures surface humidity, indicating changes in moisture conditions; NDBSI characterizes the extent of impermeable surface coverage, primarily reflecting the impact of urbanization and land development on the ecological environment; and LST represents land surface temperature, which indirectly reflects the urban heat island effect and thermal variations in the ecological environment. The specific calculation formulas for these indicators are shown in Table 2. The RSEI is constructed using principal component analysis (PCA) to combine these four indicators without applying any weighting, minimizing the influence of subjective human factors, and providing a more objective reflection of the overall regional EQ. Prior to PCA, all four indicators (NDVI, WET, NDBSI, and LST) were standardized. This step eliminates unit differences and ensures equal weighting in the PCA. The PCA method extracts the main feature information from the four indicators and uses the first principal component as a representative variable for comprehensive ecological environmental quality, thereby constructing the initial RSEI. This approach effectively integrates different dimensions of information, such as greenness, humidity, dryness, and heat, ensuring the scientific rigor and validity of the RSEI. Consequently, RSEI has been widely applied in the field of EQ evaluation [19,31,32,33] and is used for environmental monitoring and assessment in urban areas, watersheds, and different ecological regions. The formula for calculating the initial RSEI (RSEI0) is as follows (Equation (1)):
R S E I 0 = 1 P C 1 ( N D V I , W E T , N D B S I , L S T )
where PC1 represents the value of the first principal component. The RSEI0 is further normalized as follows to obtain the final value (Equation (2)):
R S E I = R S E I 0 R S E I M i n R S E I M a x R S E I M i n
where RSEIMin and RSEIMax represent the minimum and maximum values of the RSEI, respectively.

2.3.2. Theil–Sen and Mann–Kendall Trend Analysis

To assess the long-term trend of EQ in the study area, this research employs the Theil–Sen Median trend analysis method. The Theil–Sen Median method is a non-parametric statistical approach suitable for trend analysis of time-series data [34,35,36], and is known for its robustness against outliers. The slope calculation formula is as follows (Equation (3)):
β = Median R S E I j R S E I i j i 1 < i < j < n
where β represents the trend of the RSEI change (slope), with RSEIi and RSEIj denoting the RSEI values in the i-th and j-th years, respectively. When β > 0, it indicates an improving RSEI trend over time; when β < 0, it indicates a deteriorating trend; and when β = 0, it indicates no change in RSEI.
To verify the significance of the Theil–Sen Median trend analysis results, this study further employs the Mann–Kendall test method, and the calculation formula for the test statistic S is as follows (Equations (4) and (5)):
S = k = 1 n 1 j = k = 1 n sgn ( R S E I j R S E I k )
Z = s 1 n ( n 1 ) ( 2 n + 5 ) / 18 s > 0 0 s = 0 s + 1 n ( n 1 ) ( 2 n + 5 ) / 18 s < 0
where Z represents the standardized test statistic. When the absolute value of Z exceeds 1.65, 1.96, and 2.58, it indicates that the trend passes the significance test at the 90%, 95%, and 99% confidence levels, respectively. A significance level of α = 0.05 is used for the test, and based on the results the trends are classified as no significant change, marginally significant change, significant change, and highly significant change, with the specific classification criteria shown in Table 3.

2.3.3. Hurst Index

The Hurst index (HI) is a statistical method used to analyze the long-term persistence of time-series data and effectively assess the sustainability of future trends [37,38]. When 0 ≤ HI < 0.50, it indicates that the future trend of EQ will be opposite to the past trend. When HI = 0.5, it suggests that the future changes in RSEI will be random. When 0.5 < HI ≤ 1, it implies that the future trend of EQ will align with the past trend. This study utilizes the HI to explore the future trend of EQ in the BGUA. The Hurst index (HI) is calculated as follows (Equations (6) and (7)):
( R / S ) n = c × n H
log ( R / S ) n = log ( C ) + H log ( n )
where R is the range of cumulative deviations, S is the standard deviation, n is the time-series length and H represents the Hurst index.

2.3.4. Spatial Heterogeneity of EQ

The Moran’s index is a statistical method used to measure spatial autocorrelation, effectively revealing the clustering or dispersion characteristics of spatial data [39,40,41]. To analyze the spatial distribution characteristics and heterogeneity of EQ in the study area, this research employs both the global Moran’s I (Equation (8)) and local Moran’s I (Equation (9)) for spatial autocorrelation analysis. These two equations are as follows:
M o r a n s I = N i = 1 N j = 1 N W i j × i = 1 N j = 1 N W i j ( x i x ¯ ) ( x j x ¯ ) i = 1 N ( x i x ¯ ) 2 ( i j )
I i = N ( x i x ¯ ) j i N w i j ( x i x ¯ ) i = 1 N ( x i x ¯ ) 2
where Moran’s I represents the global Moran’s index, while Ii denotes the local Moran’s index; N is the number of spatial units; xi and xj represent the RSEI values of the i-th and j-th spatial units, respectively; x ¯ is the mean RSEI value; and Wij denotes the spatial weight matrix. The global Moran’s I ranges from [−1, 1]. When Moran’s I > 0, RSEI exhibits a positive spatial correlation, indicating a clustered distribution. When Moran’s I < 0, RSEI shows a negative spatial correlation, suggesting a dispersed distribution. When Moran’s I = 0, RSEI is randomly distributed in space. The local Moran’s index further reveals localized spatial clustering patterns, enabling the identification of high-value and low-value clusters within the study area.

2.3.5. Geodetector

Geodetector is a statistical method for identifying driving factors, effectively revealing the influencing factors behind geographic phenomena [18,42,43]. This study applies the Geodetector method, utilizing factor detection and interaction detection to identify the dominant factors and key interactions driving ecological environmental quality changes in the BGUA. The explanatory power of each driving factor on the spatial heterogeneity of ecological environmental quality is represented by q. The larger the value of q, the greater the influence of the factor, and its calculation formula is shown in Equation (10). The types of interactions between different driving factors and their corresponding criteria are presented in Table 4. Equation (10) is as follows:
q = 1 Σ h = 1 L N h σ h 2 N σ 2
where q represents the explanatory power of the factor, Nh and σ h 2 denote the sample size and variance of the h-th layer, respectively, while N and σ 2 represent the total sample size and overall variance, and L refers to the number of layers.
In this study, 13 factors were selected as independent variables, including RSEI (NDVI, WET, NDBSI, and LST), natural factors (TEM and PRE), socio-economic factors (GDP, POP, and NTL), terrain factors (DEM, Slope, and Aspect), and LULC. RSEI was chosen as the dependent variable. The selected driving factors were rigorously chosen based on their relationships with the RSEI components (NDVI, WET, NDBSI, and LST), systematically capturing both natural processes and anthropogenic influences on ecological quality [31,44,45]. Data was extracted by creating a 3 km × 3 km fishnet, and the natural breaks method was used to divide the independent variables into 5 categories. The data was then imported into the geographical detector for factor and interaction detection analysis. The larger the q value for single factors and interaction factors, the greater their explanatory power, indicating that the factor has a greater influence on RSEI.

3. Results and Analysis

3.1. Principal Component Analysis of the Four Indicators

The principal component analysis results of the RSEI for the BGUA (2000–2022) are shown in Table 5. The first principal component (PC1) accounts for 67.00%, 60.92%, 67.76%, 67.58%, and 65.51% of the variance in 2000, 2005, 2010, 2015, and 2022, respectively, indicating that PC1 is the dominant component for representing ecological index trends. In PC1, NDVI and WET have positive loadings, reflecting a positive contribution to the environment, while NDBSI and LST have negative loadings, indicating a negative impact. Other principal components (PC2, PC3, and PC4) show unstable signs and lower contribution rates, making them less reliable for capturing overall ecological changes. Thus, PC1 is selected for further analysis.

3.2. Spatiotemporal Analysis of EQ Changes

According to the “Technical Standards for Ecological and Environmental Quality Evaluation” issued by the Ministry of Ecology and Environment of China in 2015, and existing relevant research findings [10], the RSEI is classified into five levels with the following criteria: poor (0 ≤ RSEI < 0.2), fair (0.2 ≤ RSEI < 0.4), moderate (0.4 ≤ RSEI < 0.6), good (0.6 ≤ RSEI < 0.8), and excellent (0.8 ≤ RSEI ≤ 1).
In terms of spatial distribution (Figure 2), in 2000, regions with excellent EQ were mainly concentrated in Fangchenggang, the southern part of Chongzuo, and northern Nanning. Good quality areas were primarily in central Nanning, around Yulin, and the southern part of Chongzuo. Moderate areas were widely distributed, especially in the central areas of cities and the southeast of Fangchenggang. Poor and very poor regions, which had relatively small areas, were concentrated in the western part of Nanning, the eastern part of Chongzuo, the central-southern part of Qinzhou, and the southern part of Beihai, with lower EQ. In 2005, the area with excellent EQ shrank, but it remained in Fangchenggang, southern Chongzuo, and northern Nanning. Good areas expanded in the northern part of Chongzuo, the south of Fangchenggang, and the west of Qinzhou. Moderate areas notably increased in Chongzuo and Qinzhou, while poor and very poor areas increased in the central part of Nanning. In 2010, the area with excellent EQ significantly expanded, mainly in the central part of Fangchenggang and the northeast of Qinzhou, showing marked improvements. Good areas remained stable, while moderate and poor regions slightly reduced in size. In 2015, the excellent quality areas significantly decreased and were scattered, mainly in central Fangchenggang, northern Yulin, and northern Nanning. Good areas contracted, and the overall EQ declined. By 2022, the areas with excellent EQ had increased substantially, with significant recovery seen in northern Fangchenggang and central Nanning, reflecting the positive impact of ecological protection measures. Good areas also increased, while poor areas decreased, indicating an overall improvement in EQ.
In terms of temporal changes, as shown in Figure 3, from 2000 to 2005, the area of very poor EQ decreased from 627.51 km2 to 463.99 km2, and the area of poor EQ dropped from 10,050.14 km2 to 7274.05 km2. Areas with good EQ increased from 23,135.55 km2 to 27,380.27 km2, while the area with excellent EQ decreased by 548.95 km2, indicating localized ecological degradation. From 2005 to 2010, the area with excellent EQ grew from 2416.30 km2 to 3483.40 km2 and the area with good EQ rose to 27,772.58 km2, showing overall improvement. Poor and very poor areas remained stable. From 2010 to 2015, EQ declined. Very poor areas increased to 1214.68 km2, and poor areas expanded to 7977.74 km2. Excellent areas decreased to 1056.14 km2, accounting for 1.45% of the total. Moderate quality areas increased to 37,126.39 km2 (51.01%). From 2015 to 2022, EQ improved. Good areas rose to 28,672.55 km2, and excellent areas increased by 1200.43 km2 to 2256.57 km2. The proportion of moderate quality areas fell to 45.05%, while poor areas remained stable.

3.3. Trend and Stability of EG in the BGUA

According to Figure 4a and Table 6, the EQ of the BGUA has generally exhibited a positive trend. Specifically, the area showing significant improvement in EQ measures 1340.22 km2, representing 1.85% of the total area, while the area with minor improvements spans 3996.47 km2, or 5.52% of the total area. Additionally, 32,147.60 km2 (44.38%) of the total area demonstrated no significant improvement. Spatially, the regions with significant improvement are primarily concentrated in the southeastern part of Chongzuo, the northwest of Fangchenggang, the southwest of Nanning, and scattered areas in Qinzhou, especially where ecological restoration efforts have been particularly intensive. On the other hand, the area showing significant ecological degradation covers 1177.89 km2 (1.63%), while the area with minor degradation spans 3831.99 km2 (5.29%). These deteriorating regions are mainly found in the eastern part of Nanning, the southern part of Yulin, and scattered areas across other cities.
Furthermore, according to the spatial distribution of the HI shown in Figure 4b, the majority of the BGUA exhibits an HI between 0.5 and 1, suggesting a significant degree of persistence in the RSEI. This implies that the future trends of EQ are likely to follow the same trajectory observed over the past 23 years. However, a few areas with an HI below 0.5, predominantly in central Nanning, northern Fangchenggang, northern Chongzuo, and certain other scattered regions, may experience trends contrary to the past observations. Additionally, a small number of regions with an HI near 0.5 exhibit random fluctuations in RSEI, making future trends difficult to predict. In summary, the changes in EQ across the BGUA display strong persistence, offering a scientific foundation for further ecological management and decision-making.

3.4. Spatial Heterogeneity of EQ in the BGUA

To reveal the spatial differentiation characteristics and evolution patterns of EQ in the BGUA, this study conducts spatial autocorrelation analysis of the RSEI from 2000 to 2022 using global Moran’s I and local Moran’s I, as shown in Figure 5. Firstly, from the perspective of the global Moran’s I, the RSEI in the BGUA for 2000–2022 consistently showed values greater than 0, indicating significant positive spatial autocorrelation within the study area. This suggests that regions with similar EQ exhibit spatial clustering. From a temporal perspective, the global Moran’s I followed a trend of increasing, then decreasing, and then a slight increase, with it being 0.733 in 2000, peaking at 0.766 in 2005, decreasing to 0.734 in 2015, and slightly rising to 0.737 in 2022. This trend indicates that the spatial clustering of EQ in the BGUA strengthened from 2000 to 2005, weakened afterward, and then showed a slight recovery in 2022. Secondly, from the local Moran’s I analysis, the most significant distributions were high–high clusters and low–low clusters. High–high clusters were mainly concentrated in central Nanning, central Fangchenggang, southern Chongzuo, northern Yulin, and other scattered areas. In terms of area change, the high–high cluster expanded from 16,701 km2 in 2000 to 18,117 km2 in 2015, before declining to 17,671 km2 in 2022. This suggests that the spatial extent of high–high clusters increased from 2000 to 2015, reflecting the improvement in EQ in these regions and their positive radiative effect on surrounding areas. The reduction in area after 2022 may be related to the increased ecological pressure in local regions. Low–low clusters were mainly distributed in Nanning, Beihai, and Yulin. In terms of area change, the low–low cluster expanded from 16,812 km2 in 2000 to 17,057 km2 in 2010, before decreasing to 15,106 km2 in 2022. This indicates that the spatial extent of low–low clusters expanded from 2000 to 2010, reflecting the degradation of EQ in these regions and their negative impact on surrounding areas. The reduction in area after 2010 may be attributed to the implementation of local ecological environment management measures.

3.5. Analysis of the Driving Factors of EQ in the BGUA

3.5.1. Factor Detection Analysis

According to the factor detection results of the geographic detector in Table 7, all driving factors have p-values less than 0.01, indicating that the factors passed the significance test. In 2022, the primary driving factor of the RSEI in the BGUA was NDBSI, with a q-value of 0.806, indicating the most significant influence on ecological environmental quality, thus becoming the dominant factor. Following NDBSI, NDVI, LST, and WET had q-values of 0.667, 0.592, and 0.497, respectively, indicating that vegetation cover, surface temperature, and humidity also had a substantial impact on ecological environmental quality. Additionally, the q-values of DEM and LULC were 0.201 and 0.255, respectively, suggesting that these factors also contributed to the RSEI, although their influence was relatively weaker compared to others. Therefore, NDBSI, NDVI, LST, and WET are the main driving factors of ecological environmental quality changes in the BGUA, while the influence of other factors is comparatively minor.

3.5.2. Interaction Detection Analysis

According to the interaction detection results for the 2022 RSEI impact factors shown in Figure 6, the interaction between the driving factors generally provides a higher explanatory power for ecological environmental quality than the independent effects of individual factors, exhibiting a clear enhancement effect. For instance, the interaction between NDBSI and NDVI achieved the highest value of 0.856, which is significantly higher than the individual effect of NDVI at 0.667. The interaction between LST and NDVI reached a value of 0.845, also substantially higher than the individual effect of LST at 0.606. Additionally, the interaction between WET and DEM was 0.623, exceeding the individual effect of WET at 0.497. These findings suggest that the ecological environmental quality of the BGUA is driven by the combined interaction of multiple factors, with the synergy between NDBSI and NDVI being the most significant, thus becoming the dominant interaction factor influencing the region’s ecological environmental quality. Consequently, future ecological governance strategies should focus on comprehensively considering the combined impact of multiple factors.

4. Discussion

4.1. The Spatiotemporal Changes in EQ of the BGUA

The spatiotemporal changes in the EQ of the BGUA from 2000 to 2022 show significant phase characteristics, highly consistent with policy direction, ecological governance measures, and changes in climatic conditions. The results indicate that during this period, the EQ underwent a process of improvement (2000–2010), degradation (2010–2015), and recovery (post–2015). The trends in each stage were likely influenced not only by policies and governance efforts but also by the urbanization process, industrial structure adjustments, natural disasters, and other factors.
From 2000 to 2010, the overall EQ of the BGUA showed a steady improvement, particularly from 2005 to 2010, with significant increases in areas rated as excellent and good. The improvement during this period was primarily driven by the implementation of national and local government ecological protection policies, along with favorable climatic conditions. In 2008, the national government officially launched the “Beibu Gulf Economic Zone Development Plan,” which clearly emphasized strengthening ecological protection whilst also promoting regional economic development [46,47]. This policy encouraged local governments to intensify efforts in restoring and protecting ecosystems such as forests and wetlands, including coastal mangrove restoration, wetland park development, and improved water resource management. Additionally, local governments implemented policies like “Returning Farmland to Forest” and ecological compensation, which effectively expanded forest coverage (raising NDVI values), enhanced carbon sequestration capacity, improved water conservation, and stabilized WET, thus strengthening ecosystem regulation. Climatically, the study found that annual precipitation increased during this period, with relatively stable temperatures, providing favorable conditions for vegetation growth and ecosystem recovery. Higher precipitation promoted vegetation growth, enhanced soil moisture, and facilitated the restoration of forest and wetland ecosystems, improving ecosystem stability. Moreover, stable temperatures reduced the impact of extreme weather events on ecosystems, keeping LST stable and creating a favorable climate for ecological recovery. Furthermore, during this phase, urbanization in the BGUA was still relatively slow. The decrease in NDBSI indicates that the expansion of construction land had a minor negative impact on the environment, with reduced vegetation and heat island effects not yet becoming significant issues, thus allowing overall environmental quality to improve.
From 2010 to 2015, the EQ in the BGUA experienced some degree of degradation, marked by a reduction in areas rated as excellent and an increase in areas rated as moderate. The primary reasons for this decline include rapid urbanization, land use changes, ecological pressure from industrial structure adjustments, and the impacts of natural disasters. After 2010, the Beibu Gulf Economic Zone entered a phase of rapid urbanization, with a large-scale expansion of urban construction land. As industrial parks and infrastructure were built, parts of forests, farmlands, and wetlands were occupied, exacerbating ecological degradation. Urban expansion not only increased impermeable surfaces but also disrupted the continuity of natural ecosystems, compressing wildlife habitats. Furthermore, the urban heat island effect intensified, impacting the regional microclimate and contributing to the decline of local EQ. During this period, the BGUA also accelerated industrial upgrading and urbanization, leading to rapid land expansion and increased pollutant emissions, further damaging the surrounding ecological environment and negatively affecting overall environmental quality. Regarding climate change, the BGUA experienced several strong typhoons and extreme rainfall events during 2010–2015, such as Typhoon Haiyan in 2013, which caused severe damage to ecosystems along the Guangxi coastline. Storm surges and heavy rainfall damaged coastal wetland and mangrove ecosystems (WET declined). Additionally, extreme rainfall increased the risk of soil erosion, leading to vegetation degradation (NDVI decreased) and impacting ecological stability.
From 2015 to 2022, the EQ showed a gradual recovery, with significant increases in areas rated as excellent and good, largely attributed to the promotion of national ecological civilization policies, the implementation of pollution control measures, and strengthened ecological restoration projects. Starting in 2015, the national government elevated “ecological civilization construction” to an unprecedented strategic level, introducing a series of environmental protection laws and action plans. For example, the “13th Five-Year Plan for Ecological and Environmental Protection” explicitly set goals for improving EQ, and governments at all levels strictly enforced “ecological red lines” to limit development in ecologically sensitive areas. Additionally, coastal cities implemented the “Blue Bay” initiative to restore coastal wetlands and enhance coastal ecosystem functions, significantly improving vegetation coverage (NDVI) and strengthening the stability of coastal ecosystems. Moreover, the gradual improvement in EQ during this phase was related to the deepened implementation of ecological restoration projects. Local governments increased investment in ecological restoration and implemented a series of ecological protection and restoration projects. For instance, Fangchenggang City expanded the mangrove protection area and carried out mangrove restoration projects, restoring a significant amount of coastal wetland ecosystems. Additionally, forest protection and afforestation projects in the BGUA region yielded significant results, further increasing forest coverage (NDVI increased), enhancing regional ecosystem stability, and boosting carbon sequestration capacity.
In summary, the trend of EQ in the BGUA from 2000 to 2022, characterized by improvement, degradation, and recovery, reflects the combined influence of policy regulation, urbanization, climatic conditions, and ecological governance measures. Future ecological governance should focus on the coordinated management of multiple factors to enhance ecosystem stability and achieve sustainable regional development.

4.2. Impact of Driving Factors on RSEI

This study employs the geographical detector method to identify the key driving factors and their interactions influencing the changes in EQ in the BGUA, providing scientific evidence for understanding the mechanisms of ecological environmental evolution. The analysis results show that the dryness index (NDBSI) has the most significant impact on the RSEI, with a q-value of 0.806. This indicates that urbanization and land use changes are the core factors driving ecological environmental changes. Furthermore, NDVI and LST also play crucial roles, with q-values of 0.667 and 0.592, respectively, suggesting that vegetation degradation or restoration, along with temperature changes, also significantly influence the evolution of EQ. Additionally, humidity (WET) has a q-value of 0.497, indicating that although the impact is relatively lower, the influence of moisture conditions on EQ should not be overlooked.
Further analysis of the interactions between these driving factors reveals that the synergistic effects of multiple factors contribute more significantly to explaining EQ than the independent effects of individual factors. For instance, the interaction value between NDBSI and NDVI reaches 0.856, significantly higher than the individual impacts of these factors, indicating a strong coupling effect between the expansion of impermeable surfaces (increased NDBSI) and changes in vegetation cover (decreased NDVI) due to urbanization. This phenomenon demonstrates that in the process of urban expansion, large natural ecological spaces are replaced by built-up areas, reducing green spaces, weakening soil conservation capacity, and diminishing the stability of ecosystems. Moreover, the interaction value between NDBSI and LST is also high, reflecting that urbanization not only directly leads to vegetation loss but also exacerbates ecological degradation through changes in land surface temperature.
Additionally, the interaction value between LST and NDVI reaches 0.845, showing that the coupling effect between land surface temperature changes and vegetation cover conditions has a significant impact on EQ. Increased temperatures can enhance plant transpiration and soil moisture evaporation, thereby affecting regional WET and ecosystem health. Simultaneously, the reduction in vegetation cover diminishes the cooling effect, causing further temperature rises and creating a vicious cycle. Therefore, during urban expansion in the BGUA, special attention should be paid to the protection of urban green spaces and the restoration of vegetation to mitigate the adverse effects of high temperatures on the ecological environment.
Regarding the role of WET, although its q-value is relatively low (0.497), its interaction with other factors remains noteworthy. For example, the interaction between humidity and vegetation (NDVI) may influence the regional hydrological cycle, further affecting ecosystem stability. In coastal areas, humidity variations may also be linked to land–sea interactions and changes in precipitation patterns, which in turn impact EQ. Future research could explore the coupling relationship between humidity changes and the marine-terrestrial system and investigate how water resource management can optimize EQ.
In conclusion, whilst dryness (NDBSI) is the most significant driving factor, the interactions between various driving factors significantly enhance the explanation of EQ changes. In particular, the synergistic effect of increased impermeable surfaces (NDBSI rise) and decreased vegetation (NDVI decline) due to urban expansion has a particularly prominent impact on EQ. Therefore, future ecological governance strategies should focus on multi-factor collaborative management, optimize land spatial layout, strengthen ecological restoration, and improve the stability and sustainability of regional ecosystems, thereby achieving high-quality development and long-term ecological improvement in the BGUA.

4.3. Future Development Trends of EQ in the BGUA

The analysis based on the HI reveals the spatiotemporal evolution of ecological environmental quality in the BGUA, providing a scientific basis for future ecological management. By developing differentiated protection and restoration strategies tailored to the ecological evolution trends of various regions, the stability of ecosystems can be effectively enhanced, promoting sustainable regional development.
The results indicate that the HI in most areas of the BGUA ranges from 0.5 to 1, meaning that the future trend of EQ will generally follow the development pattern of the past 23 years, demonstrating strong persistence. In particular, regions where EQ has significantly improved, such as the northern part of Fangchenggang and central Nanning, have an HI close to 1, suggesting that the positive trend in EQ is expected to continue. This phenomenon is closely related to a series of ecological protection measures implemented in the region in recent years. For example, Fangchenggang has increased its efforts in mangrove ecosystem restoration, including the implementation of mangrove restoration projects, wetland protection, and degraded land restoration, which have enhanced the stability of coastal ecosystems. Similarly, the improvement in central Nanning’s ecological environment is closely linked to urban greening initiatives, the construction of ecological corridors, and water body management projects. The government’s promotion of “sponge city” construction has effectively improved rainwater resource utilization and enhanced the self-regulation capacity of urban ecosystems. In the future, these regions should further strengthen ecological restoration, leveraging smart environmental monitoring technologies to enable precise ecological management and ensure continued ecological improvement.
However, in some areas, the HI is below 0.5, indicating that the future EQ trend in these regions may reverse compared to the past, exhibiting significant uncertainty and volatility. Notably, in regions where EQ has significantly deteriorated, such as the eastern part of Nanning and the southern part of Yulin, the HI is significantly below 0.5, reflecting the vulnerability of ecosystems and increasing environmental pressure. Analysis shows that the decline in EQ in these areas is closely related to rapid urbanization, industrialization, and the expansion of agricultural activities. For example, industrial parks in eastern Nanning have been expanding in recent years, while land use changes in southern Yulin have been particularly intense, with forests and wetlands being converted into construction land, affecting the structure and function of ecosystems. Therefore, for these degraded areas, more proactive and targeted ecological restoration and environmental management measures are required. First, construction land expansion should be strictly controlled, and land use regulations should be strengthened, especially in ecologically sensitive areas. Ecological redlines should be demarcated to prevent excessive development.

4.4. Limitations of the Study

This study provides a comprehensive analysis of the spatiotemporal evolution and driving factors of ecological environmental quality in the BGUA from 2000 to 2022. However, there are certain limitations. Firstly, the 500 m resolution of the MODIS data, whilst suitable for large-scale analysis, lacks accuracy in localized areas. Future studies could incorporate higher-resolution remote sensing data to improve precision. Secondly, the use of a 3 km × 3 km grid—determined by Geodetector’s computational constraints—may overlook finer-scale heterogeneity; multi-scale analysis could help assess the spatial dependence of driving mechanisms. Thirdly, although the RSEI incorporates NDVI, WET, NDBSI, and LST, it does not fully reflect anthropogenic influences such as GDP, population density, or air quality, which future research should consider to enhance assessment robustness. Additionally, key factors such as policy, ecological governance, and soil conditions (e.g., soil organic matter) were not included but may significantly influence EQ. Finally, this study does not explicitly apply the emerging “polycrisis” framework, which emphasizes the interconnected and compounding nature of ecological, economic, and institutional risks [48,49],. Incorporating this perspective in future work could offer a more integrated understanding of ecological change under complex systemic pressures.

5. Conclusions

Based on the GEE platform and MODIS data from 2000 to 2022, this study constructs the RSEI for the BGUA. It reveals the trends and driving mechanisms of ecological environmental quality changes, incorporating multiple driving factors. The main conclusions are as follows:
(1)
EQ changes: From 2000 to 2010, EQ improved, especially between 2005 and 2010, driven by national policies and favorable climate conditions. From 2010 to 2015, quality declined, likely due to urban expansion and industrialization. After 2015, EQ rebounded due to stronger ecological policies and governance measures.
(2)
EQ trends: HI analysis shows a continued improvement in most areas, particularly in regions with effective ecological governance like southeast Chongzuo and northwest Fangchenggang.
(3)
Spatial differentiation: A significant positive spatial correlation exists in EQ. High-quality areas are primarily found in central Nanning, Fangchenggang, southern Chongzuo, and northern Yulin, indicating that improvements in these areas have a positive impact on surrounding regions. Low-quality areas are concentrated in Nanning, Beihai, and Yulin, influenced by urban expansion and industrial pollution. After 2015, the high-quality areas slightly shrank, while low-quality areas reduced in size, likely due to the implementation of ecological governance measures.
(4)
Driving factors: EQ changes are mainly driven by NDBSI (q = 0.806), with significant contributions from NDVI, LST, and WET. The combined effects of urban expansion leading to increased impervious surfaces (NDBSI rise) and vegetation loss (NDVI decline) (q = 0.856) and the interaction between LST and NDVI (q = 0.845) provided much greater explanatory power than individual factors, demonstrating that the synergy between urban expansion, vegetation loss, and temperature rise significantly impacts EQ.
These findings suggest that the BGUA should focus on controlling construction land expansion, optimizing land use structures, and reducing the adverse effects of increasing NDBSI on EQ. Strengthening ecological restoration to enhance vegetation cover and prioritizing the protection of ecologically sensitive areas, such as the northwest of Fangchenggang and southeast of Chongzuo, is crucial. Additionally, refining climate adaptation strategies to address precipitation changes and extreme weather will ensure ongoing ecological improvements, supporting the coordinated development of urbanization, ecological protection, and economic growth.

Author Contributions

This manuscript is the result of the research of J.J. and H.J. under the supervision of F.W. and Z.C., and the advising of J.X., L.X., Y.J. (Yu Jiang) and Y.J. (Yanhong Jia). All authors designed the study, developed the methodology, and discussed the results, and J.J. and H.J. wrote the paper. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded by the National Natural Science Foundation of China (42461015), the Guangxi science and technology base and talent special project (GuiKeAD23026194), the Open Fund of Key Laboratory of Coastal Science and Integrated Management, Ministry of Natural Resources (2024COSIM01), the Open Fund of key Laboratory for Earth Surface Processes, Ministry of Education, and the Guangxi Normal University’s National Training Program of Innovation and Entrepreneurship for Undergraduates (X2025106020260).

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding authors.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. An, M.; Xie, P.; He, W.; Wang, B.; Huang, J.; Khanal, R. Spatiotemporal change of ecologic environment quality and human interaction factors in three gorges ecologic economic corridor, based on RSEI. Ecol. Indic. 2022, 141, 109090. [Google Scholar] [CrossRef]
  2. Liu, H.; Fang, C.; Zhang, X.; Wang, Z.; Bao, C.; Li, F. The effect of natural and anthropogenic factors on haze pollution in Chinese cities: A spatial econometrics approach. J. Clean. Prod. 2017, 165, 323–333. [Google Scholar] [CrossRef]
  3. Wang, S.; Cao, Z.; Luo, P.; Zhu, W. Spatiotemporal Variations and Climatological Trends in Precipitation Indices in Shaanxi Province, China. Atmosphere 2022, 13, 744. [Google Scholar] [CrossRef]
  4. Deng, O.; Wang, S.; Ran, J.; Huang, S.; Zhang, X.; Duan, J.; Zhang, L.; Xia, Y.; Reis, S.; Xu, J.; et al. Managing urban development could halve nitrogen pollution in China. Nat. Commun. 2024, 15, 401. [Google Scholar] [CrossRef]
  5. Feng, Y.X.; He, W.; Wang, J.Y.; Liu, D.; Yao, Y.F. Temporal and Spatial Evolution Characteristics and Driving Factors of Ecological Environment Quality in Beibu Gulf Port Area. Environ. Sci. 2025. (In Chinese) [Google Scholar] [CrossRef]
  6. Zhang, L.Y.; Li, X.; Feng, J.H.; Rao, R.G.; He, T.Y.; Chen, Y. Spatial-temporal Changes of NDVI in Yellow River Basin and Its Dual Response to Climate Change and Human Activities During 2000–2018. Bull. Soil. Water Conserv. 2021, 41, 276–286. [Google Scholar]
  7. Yu, H.J.; Zhang, F.M.; Ma, H.; Lu, Y.Y. Spatio-temporal Evolution and Driving Factors of Ecological Environment Quality in the Huaihe River Basin Based on RSEI. Environ. Sci. 2024, 45, 4112–4121. (In Chinese) [Google Scholar]
  8. Mao, Z.C.; Song, Y.; Li, M. Research of the Desertification in Hetao Area Based on MODIS Inversion Data. Acta Sci. Nat. Univ. Pekin. 2015, 51, 1102–1110. [Google Scholar]
  9. Liu, W.; Cheng, Z.; Li, J.; Li, G.; Pan, N. Assessment of ecological asset quality and its drivers in Agro-pastoral Ecotone of China. Ecol. Indic. 2025, 170, 113072. [Google Scholar] [CrossRef]
  10. Ye, B.W.; Sun, B.; Zhao, Y.L.; Pang, J.Q.; Guo, Y.Y.; Yao, W.Z. Spatial-temporal Evolution and Driving Force Analysis of Habitat Quality in Inner Mongolia from 2000 to 2022. Environ. Sci. 2025, 46, 4473–4484. (In Chinese) [Google Scholar]
  11. Wang, D.; Li, Y.M.; Yang, X.D.; Zhang, Z.Y.; Gao, S.T.; Zhou, Q.H.; Zhuo, Y.; Wen, X.C.; Guo, Z.Y. Evaluating urban ecological civilization and its obstacle factors based on integrated model of PSR-EVW-TOPSIS: A case study of 13 cities in Jiangsu Province, China. Ecol. Indic. 2021, 133, 108431. [Google Scholar] [CrossRef]
  12. Zhao, J.L.; Zhao, H. Coupling coordination analysis of urbanization and eco-environment in Ningxia based on DMSP-OLS and MODIS data. Bull. Surv. Mapp. 2021, 10, 9. (In Chinese) [Google Scholar]
  13. Wang, B.; Yu, F.; Teng, Y.; Cao, G.; Zhao, D.; Zhao, M. A SEEC Model Based on the DPSIR Framework Approach for Watershed Ecological Security Risk Assessment: A Case Study in Northwest China. Water 2022, 14, 106. [Google Scholar] [CrossRef]
  14. Wang, X.; Meng, X.; Long, Y. Projecting 1 km-grid population distributions from 2020 to 2100 globally under shared socioeconomic pathways. Sci. Data 2022, 9, 563–576. [Google Scholar] [CrossRef]
  15. Xu, H.Q. A remote sensing index for assessment of regional ecological changes. China Environ. Sci. 2013, 33, 889–897. (In Chinese) [Google Scholar]
  16. Xu, H.Q. A remote sensing urban ecological index and its application. Acta Ecol. Sin. 2013, 33, 7853–7862. (In Chinese) [Google Scholar]
  17. Zhang, J.; Yang, G.; Yang, L.; Li, Z.; Gao, M.; Yu, C.; Gong, E.; Long, H.; Hu, H. Dynamic Monitoring of Environmental Quality in the Loess Plateau from 2000 to 2020 Using the Google Earth Engine Platform and the Remote Sensing Ecological Index. Remote Sens. 2022, 14, 5094. [Google Scholar] [CrossRef]
  18. Li, H.; Jing, H.; Yan, G.; Guo, H.; Luan, W. Long-Term Ecological Environment Quality Evaluation and Its Driving Mechanism in Luoyang City. Sustainability 2023, 15, 11866. [Google Scholar] [CrossRef]
  19. Che, L.; Yin, S.; Jin, J.; Wu, W. Assessment and Simulation of Urban Ecological Environment Quality Based on Geographic Information System Ecological Index. Land 2024, 13, 687. [Google Scholar] [CrossRef]
  20. Li, W.; Samat, A.; Abuduwaili, J.; Wang, W. Spatiotemporal Pattern, Evolutionary Trend, and Driving Forces Analysis of Ecological Quality in the Irtysh River Basin (2000–2020). Land 2024, 13, 222. [Google Scholar] [CrossRef]
  21. Yang, H.; Yu, J.; Xu, W.; Wu, Y.; Lei, X.; Ye, J.; Geng, J.; Ding, Z. Long-time series ecological environment quality monitoring and cause analysis in the Dianchi Lake Basin, China. Ecol. Indic. 2023, 148, 110084. [Google Scholar] [CrossRef]
  22. Airiken, M.; Li, S. The Dynamic Monitoring and Driving Forces Analysis of Ecological Environment Quality in the Tibetan Plateau Based on the Google Earth Engine. Remote Sens. 2024, 16, 682. [Google Scholar] [CrossRef]
  23. Nong, L.P.; Wang, J.L. Dynamic monitoring of ecological environment quality in Kunming based on RSEI model. Chin. J. Ecol. 2020, 39, 2042–2050. (In Chinese) [Google Scholar]
  24. Li, N.; Guo, Y.; Wang, L.; Wang, Q.; Yan, D.; Zhao, S.; Lei, T. Evaluation and quantitative characterization for the ecological environment impact of open pit mining on vegetation destruction from landsat time series: A case study of Wulishan limestone mine. Ecol. Indic. 2024, 158, 111371. [Google Scholar] [CrossRef]
  25. Zhang, J.; Li, J.; Bao, A.; Frankl, A.; Wang, H.; Bai, J.; Shen, Z.; Li, L.; De Maeyer, P.; Van de Voorde, T. Ecological restoration trajectory of the Taitema Lake wetland in arid northwest China: A 36-year wetland health assessment using Landsat time series data. Ecol. Indic. 2024, 161, 111956. [Google Scholar] [CrossRef]
  26. Aizizi, Y.; Kasimu, A.; Liang, H.; Zhang, X.; Zhao, Y.; Wei, B. Evaluation of ecological space and ecological quality changes in urban agglomeration on the northern slope of the Tianshan Mountains. Ecol. Indic. 2023, 146, 109896. [Google Scholar] [CrossRef]
  27. Kong, W.; Chang, W.; Xie, M.; Li, Y.; Wan, T.; Nie, X.; Mo, D. Long-Term Monitoring and Analysis of Key Driving Factors in Environmental Quality: A Case Study of Fujian Province. Forests 2024, 15, 1541. [Google Scholar] [CrossRef]
  28. Lei, X.; Liu, H.; Li, S.; Luo, Q.; Cheng, S.; Hu, G.; Wang, X.; Bai, W. Coupling coordination analysis of urbanization and ecological environment in Chengdu-Chongqing urban agglomeration. Ecol. Indic. 2024, 161, 111969. [Google Scholar] [CrossRef]
  29. Yang, Y.; Li, H.; Qian, C. Analysis of the implementation effects of ecological restoration projects based on carbon storage and eco-environmental quality: A case study of the Yellow River Delta, China. J. Environ. Manag. 2023, 340, 117929. [Google Scholar] [CrossRef]
  30. Zhang, L.; Liu, Q.; Wang, J.; Wu, T.; Li, M. Constructing ecological security patterns using remote sensing ecological index and circuit theory: A case study of the Changchun-Jilin-Tumen region. J. Environ. Manag. 2025, 373, 123693. [Google Scholar] [CrossRef]
  31. Chen, M.; Tang, C.; Yang, Y.; Xiong, J. Quantifying spatio-temporal changes in the ecological environment quality and their implications for surface mass movement after a high-magnitude earthquake. Glob. Ecol. Conserv. 2025, 58, e03454. [Google Scholar] [CrossRef]
  32. Lu, F.; Zhang, C.; Cao, H.; Wang, X.; Zheng, T.; Huang, Z. Assessment of ecological environment quality and their drivers in urban agglomeration based on a novel remote sensing ecological index. Ecol. Indic. 2025, 170, 113104. [Google Scholar] [CrossRef]
  33. Nie, X.; Hu, Z.; Zhu, Q.; Ruan, M. Research on Temporal and Spatial Resolution and the Driving Forces of Ecological Environment Quality in Coal Mining Areas Considering Topographic Correction. Remote Sens. 2021, 13, 2815. [Google Scholar] [CrossRef]
  34. Pi, G.N.; He, Z.H.; You, M.; Zhang, L.; Yang, M.K. Effects of Climate Change and Human Activities on Vegetation Change in Guizhou Province from 2001 to 2020. J. Soil. Water Conserv. 2022, 36, 160–167. (In Chinese) [Google Scholar]
  35. Zhang, S.; Zhong, Q.; Cheng, D.; Xu, C.; Chang, Y.; Lin, Y.; Li, B. Landscape ecological risk projection based on the PLUS model under the localized shared socioeconomic pathways in the Fujian Delta region. Ecol. Indic. 2022, 136, 108642. [Google Scholar] [CrossRef]
  36. Zhang, H.B.; Wang, W.; Song, Y.J.; Miao, L.G.; Ma, C. Ecological index evaluation of arid inflow area based on the modified remote sensing ecological index: A case study of Tabu River Basin at the northern foot of the Yin Mountains. Acta Ecol. Sin. 2024, 44, 523–543. (In Chinese) [Google Scholar]
  37. Chen, L.W. Spatio-temporal Variation of Water Use Efficiency and Its Responses to Environmental Factors in Yellow River Basin During 2001–2020. Bull. Soil. Water Conserv. 2022, 42, 222–230. (In Chinese) [Google Scholar]
  38. Umuhoza, J.; Jiapaer, G.; Tao, Y.; Jiang, L.; Zhang, L.; Gasirabo, A.; Umwali, E.D.; Umugwaneza, A. Analysis of fluctuations in vegetation dynamic over Africa using satellite data of solar-induced chlorophyll fluorescence. Ecol. Indic. 2023, 146, 109846. [Google Scholar] [CrossRef]
  39. Jing, J.; Wei, F.; Jiang, H.; Chen, Z.; Lv, S.; Li, T.; Li, W.; Tang, Y. Prediction of Land Use Change and Carbon Storage in Lijiang River Basin Based on InVEST-PLUS Model and SSP-RCP Scenario. Land 2025, 14, 460. [Google Scholar] [CrossRef]
  40. Shi, S.; Peng, S.; Lin, Z.; Zhu, Z.; Ma, D.; Yin, Y.; Lu, X.; Li, T.; Gong, L. Analysis of ecological environment quality heterogeneity across different landform types in Myanmar and its driving forces. Ecol. Indic. 2024, 168, 112755. [Google Scholar] [CrossRef]
  41. Liu, Y.; Liu, H.; Yan, C.; Feng, Z.; Zhou, S. Evaluation and dynamic monitoring of ecological environment quality in mining area based on improved CRSEI index model. Heliyon 2023, 9, e20787. [Google Scholar] [CrossRef]
  42. Geng, J.; Yu, K.; Xie, Z.; Zhao, G.; Ai, J.; Yang, L.; Yang, H.; Liu, J. Analysis of Spatiotemporal Variation and Drivers of Ecological Quality in Fuzhou Based on RSEI. Remote Sens. 2022, 14, 4900. [Google Scholar] [CrossRef]
  43. Zhang, X.; Jia, W.; Lu, S.; He, J. Ecological assessment and driver analysis of high vegetation cover areas based on new remote sensing index. Ecol. Inform. 2024, 82, 102786. [Google Scholar] [CrossRef]
  44. Sui, X.; Sun, Y.; Wang, X.; Yuan, J. Research on ecological quality and restoration of fragile mining areas in the Yellow River Basin—The case of Xiegou coal mine. Ecol. Indic. 2025, 174, 113426. [Google Scholar] [CrossRef]
  45. Zhang, L.; Li, X.; Liu, X.; Lian, Z.; Zhang, G.; Liu, Z.; An, S.; Ren, Y.; Li, Y.; Liu, S. Dynamic monitoring and drivers of ecological environmental quality in the Three-North region, China: Insights based on remote sensing ecological index. Ecol. Inform. 2025, 85, 102936. [Google Scholar] [CrossRef]
  46. Ling, Z.Y.; Li, Y.S.; Jiang, W.G.; Liao, C.M.; Ling, Y.R. Dynamic Change Characteristics of “Production-living-ecological Spaces” of Urban Agglomeration Interlaced with Mountains, Rivers and Sea: A Case Study of the Beibu Gulf Urban Agglomeration in Guangxi. Econ. Geogr. 2022, 42, 18–24. (In Chinese) [Google Scholar]
  47. Luo, S.F.; Yan, W.D. Evolution and driving force analysis of ecosystem service values in Guangxi Beibu Gulf coastal areas, China. Acta Ecol. Sin. 2018, 38, 3248–3259. (In Chinese) [Google Scholar] [CrossRef]
  48. Lawrence, M.; Homer-Dixon, T.; Janzwood, S.; Rockstöm, J.; Renn, O.; Donges, J.F. Global polycrisis: The causal mechanisms of crisis entanglement. Glob. Sustain. 2024, 7, e6. [Google Scholar] [CrossRef]
  49. Matlovic, R.; Matlovičová, K. Polycrisis in the Anthropocene as a key research agenda for geography: Ontological delineation and the shift to a postdisciplinary approach. Folia Geogr. 2024, 66, 5–33. [Google Scholar]
Figure 1. Location of the study area.
Figure 1. Location of the study area.
Land 14 01556 g001
Figure 2. RSEI grade distribution of the BGUA from 2000 to 2022. Notes: (a) 2000, (b) 2005, (c) 2010, (d) 2015, (e) 2022.
Figure 2. RSEI grade distribution of the BGUA from 2000 to 2022. Notes: (a) 2000, (b) 2005, (c) 2010, (d) 2015, (e) 2022.
Land 14 01556 g002
Figure 3. (a) Area and proportion of RSEI grades in the BGUA from 2000 to 2022. (b) the concentric circles correspond to the years 2000, 2005, 2010, 2015, and 2022, from inner to outer.
Figure 3. (a) Area and proportion of RSEI grades in the BGUA from 2000 to 2022. (b) the concentric circles correspond to the years 2000, 2005, 2010, 2015, and 2022, from inner to outer.
Land 14 01556 g003
Figure 4. RSEI trend of BGUA from 2000 to 2022. Notes: (a) RSEI trend categories in BGUA, (b) the spatial distribution of the HI.
Figure 4. RSEI trend of BGUA from 2000 to 2022. Notes: (a) RSEI trend categories in BGUA, (b) the spatial distribution of the HI.
Land 14 01556 g004
Figure 5. LISA clustering of the RSEI in BGUA from 2000 to 2022. Notes: (a) 2000, (b) 2005, (c) 2010, (d) 2015, (e) 2022.
Figure 5. LISA clustering of the RSEI in BGUA from 2000 to 2022. Notes: (a) 2000, (b) 2005, (c) 2010, (d) 2015, (e) 2022.
Land 14 01556 g005
Figure 6. Interaction detection results of RSEI influencing factors in 2022.
Figure 6. Interaction detection results of RSEI influencing factors in 2022.
Land 14 01556 g006
Table 1. Data sources.
Table 1. Data sources.
Data TypeData NameResolution/mData Sources
RSEIMOD13A1500https://modis.gsfc.nasa.gov/ (accessed on 15 December 2024)
MOD11A21000
MOD09A1500
Natural factorsAnnual average temperature (TEM)1000https://data.tpdc.ac.cn/ (accessed on 15 December 2024)
Annual average precipitation (PRE)1000
Socio-economic dataGDP1000http://www.resdc.cn/ (accessed on 15 December 2024)
Population (POP)1000
Night-time light intensity (NLI)500http://www.geodata.cn (accessed on 15 December 2024)
Topographical factorsDEM30http://www.gscloud.cn/ (accessed on 15 December 2024)
Slope30Extracted from DEM data (http://www.gscloud.cn/, accessed on 15 December 2024)
Aspect30
Land useLULC30http://www.resdc.cn/ (accessed on 15 December 2024)
Table 2. Calculation formulas for each indicator.
Table 2. Calculation formulas for each indicator.
IndicatorCalculation FormulaParameter Explanation
Greenness
(NDVI)
NDVI = ρ 2 ρ 1 / ρ 2 + ρ 1 In each formula ρ 1 ~ ρ 7 represents the reflectance of each MODIS band. DN refers to the pixel grayscale value. SI is the soil index, and IBI is the building index.
Humidity
(WET)
WET = 0.1147 ρ 1 + 0.2489 ρ 2 + 0.2408 ρ 3 + 0.3122 ρ 4 0.3122 ρ 5 0.6416 ρ 6 0.5087 ρ 7
Heat (LST) LST = 0 . 02 DN - 273 . 15
Dryness (NDBSI) SI = ρ 6 + ρ 1 ρ 2 + ρ 3 ρ 6 + ρ 1 + ρ 2 + ρ 3 IBI = 2 ρ 6 / ρ 6 + ρ 2 ρ 2 / ρ 1 + ρ 2 + ρ 4 / ρ 4 + ρ 6 2 ρ 6 / ρ 6 + ρ 2 + ρ 2 / ρ 1 + ρ 2 + ρ 4 / ρ 4 + ρ 6 NDBSI = SI + IBI 2
Table 3. Classification of Mann–Kendall test trends.
Table 3. Classification of Mann–Kendall test trends.
β ZTrend Characteristics
β > 0   2.58 < Z Highly significant improvement
1.96 < Z 2.58 Significant improvement
1.65 < Z 1.96 Marginally significant improvement
Z 1.65 No significant improvement
β = 0 ZNo change
β < 0   Z 1.65 No significant deterioration
1.65 < Z 1.96 Marginally significant deterioration
1.96 < Z 2.58 Significant deterioration
2.58 < Z Highly significant deterioration
Table 4. Types of factor interaction.
Table 4. Types of factor interaction.
Types of Factor InteractionCriteria for Judgment
Nonlinear weakening q ( X 1 X 2 ) < Min [ q ( X 1 ) , q ( X 2 ) ]
Single-factor nonlinear weakening Min [ q ( X 1 ) , q ( X 2 ) ] < q ( X 1 X 2 ) < Max [ q ( X 1 ) , q ( X 2 ) ]
Two-factor enhancement q ( X 1 X 2 ) > max [ q ( X 1 ) , q ( X 2 ) ]
Mutual independence q ( X 1 X 2 ) = q ( X 1 ) + q ( X 2 )
Nonlinear enhancement q ( X 1 X 2 ) > q ( X 1 ) + q ( X 2 )
Table 5. Principal component analysis results of RSEI for the BGUA.
Table 5. Principal component analysis results of RSEI for the BGUA.
YearPrincipal Component (PC)NDVIWETLSTNDBSIEigenvalueEigenvalue Contribution Rate (%)
2000PC10.4880.473−0.444−0.5830.06667.00
PC20.1610.2580.892−0.3350.01717.17
PC30.660−0.7430.060−0.0960.01413.84
PC4−0.548−0.397−0.062−0.7330.0021.99
2005PC10.4480.553−0.512−0.4810.05560.92
PC2−0.7910.374−0.4520.1740.01921.43
PC3−0.1960.5950.730−0.2740.01314.16
PC40.3680.4480.0400.8140.0033.49
2010PC10.4480.470−0.569−0.5050.06167.76
PC20.691−0.7010.099−0.1500.01517.20
PC30.1990.3910.815−0.3780.01112.41
PC4−0.532−0.368−0.047−0.7620.0022.64
2015PC10.4810.453−0.560−0.5000.06167.58
PC20.646−0.7360.111−0.1700.01516.69
PC30.1860.3780.817−0.3930.01213.23
PC4−0.563−0.332−0.081−0.7530.0022.49
2022PC10.4990.442−0.519−0.5350.06365.51
PC2−0.4320.8160.373−0.0900.01717.66
PC30.593−0.0620.766−0.2410.01313.71
PC4−0.461−0.3660.074−0.8050.0033.12
Table 6. Area and proportion of RSEI trend categories in BGUA from 2000 to 2022.
Table 6. Area and proportion of RSEI trend categories in BGUA from 2000 to 2022.
TypeSignificant DegradationMarginally Significant DegradationInsignificant DegradationNo ChangeInsignificant ImprovementMarginally Significant ImprovementSignificant Improvement
Area (km2)1177.893831.9929,582.26357.9032,147.603996.471340.22
Proportion (%)1.635.2940.840.4944.385.521.85
Table 7. Factor detection results of RSEI for BGUA in 2022.
Table 7. Factor detection results of RSEI for BGUA in 2022.
IndicatorsWETNDVINDBSILSTGDPPOPNTL
q0.4970.6670.8060.5920.0350.1430.139
p0.0000.0000.0000.0000.0000.0000.000
IndicatorsTEMPREDEMSlopeAspectLULC
q0.1550.0120.2010.0840.0570.255
p0.0000.0000.0000.0000.0000.000
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

Jing, J.; Jiang, H.; Wei, F.; Xie, J.; Xie, L.; Jiang, Y.; Jia, Y.; Chen, Z. Spatiotemporal Evolution and Driving Force Analysis of Habitat Quality in the Beibu Gulf Urban Agglomeration. Land 2025, 14, 1556. https://doi.org/10.3390/land14081556

AMA Style

Jing J, Jiang H, Wei F, Xie J, Xie L, Jiang Y, Jia Y, Chen Z. Spatiotemporal Evolution and Driving Force Analysis of Habitat Quality in the Beibu Gulf Urban Agglomeration. Land. 2025; 14(8):1556. https://doi.org/10.3390/land14081556

Chicago/Turabian Style

Jing, Jing, Hong Jiang, Feili Wei, Jiarui Xie, Ling Xie, Yu Jiang, Yanhong Jia, and Zhantu Chen. 2025. "Spatiotemporal Evolution and Driving Force Analysis of Habitat Quality in the Beibu Gulf Urban Agglomeration" Land 14, no. 8: 1556. https://doi.org/10.3390/land14081556

APA Style

Jing, J., Jiang, H., Wei, F., Xie, J., Xie, L., Jiang, Y., Jia, Y., & Chen, Z. (2025). Spatiotemporal Evolution and Driving Force Analysis of Habitat Quality in the Beibu Gulf Urban Agglomeration. Land, 14(8), 1556. https://doi.org/10.3390/land14081556

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