Next Article in Journal
Feasibility of Struvite Crystals Obtained from Swine Wastewater and Urban Sewage Sludge Liquid Fraction as Fertilising Product
Previous Article in Journal
Dependent-Chance Goal Programming for Sustainable Supply Chain Design: A Reinforcement Learning-Enhanced Salp Swarm Approach
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Identifying Priority Bird Habitats Through Seasonal Dynamics: An Integrated Habitat Suitability–Risk–Quality Framework

1
Shanghai Investigation, Design & Research Institute Co., Ltd., Shanghai 200050, China
2
School of Ecological and Environmental Sciences, East China Normal University, Shanghai 200241, China
3
Department of Urban and Rural Planning, School of Architecture, Soochow University, Suzhou 215123, China
4
School of Ecology and Environmental Science, Yunnan University, Kunming 650500, China
5
Land Consolidation and Rehabilitation Center, Ministry of Natural Resources, Beijing 100035, China
6
School of Resource and Environmental Sciences, Wuhan University, Wuhan 430079, China
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(13), 6078; https://doi.org/10.3390/su17136078
Submission received: 14 May 2025 / Revised: 20 June 2025 / Accepted: 30 June 2025 / Published: 2 July 2025
(This article belongs to the Section Sustainability, Biodiversity and Conservation)

Abstract

A key challenge is how to effectively conserve habitats and biodiversity amid widespread habitat fragmentation and loss caused by global urbanization. Despite growing attention to this issue, knowledge of the seasonal dynamics of habitats remains limited, and conservation gaps are still inadequately identified. This study proposes a novel integrated framework, “Habitat Suitability–Risk–Quality”, to improve the assessment of the seasonal bird habitat quality and to identify priority conservation habitats in urban landscapes. The framework was implemented in Wuhan, China, a critical stopover site along the East Asian–Australasian Flyway. It combines the Maximum Entropy (MaxEnt) model to predict the seasonal habitat suitability, the Habitat Risk Assessment (HRA) model to quantify habitat sensitivity to multiple anthropogenic threats, and a refined Habitat Quality (HQ) model to evaluate the seasonal habitat quality. K-means clustering was then applied to group habitats based on seasonal quality dynamics, enabling the identification of priority areas and the development of differentiated conservation strategies. The results show significant seasonal variation in habitat suitability and quality. Wetlands provided the highest-quality habitats in autumn and winter, grasslands exhibited moderate seasonal quality, and forests showed the least seasonal fluctuation. The spatial analysis revealed that high-quality wetland habitats form an ecological belt along the urban–suburban fringe. Four habitat clusters with distinct seasonal characteristics were then identified. However, spatial mismatches were found between existing protected areas and habitats of high ecological value. Notably, Cluster 1 maintained high habitat quality year round, spanning 99.38 km2, yet only 46.51% of its area is currently protected. The remaining 53.16 km2, mostly situated in urban–suburban transitional zones, remain unprotected. This study provides valuable insights for identifying priority habitats and developing season-specific conservation strategies in rapidly urbanizing regions, thereby supporting the sustainable management of urban biodiversity and the development of resilient ecological systems.

1. Introduction

Habitats serve as the foundation of ecological systems, supporting essential ecological functions and providing critical resources for species survival and regional sustainability [1,2,3]. For birds, including both migratory and resident species, wetlands, forests, and grasslands offer critical habitats for foraging, breeding, and resting across different life stages and seasons [4]. However, these habitats are increasingly threatened by land-use changes, urban expansion, and climate variability, resulting in fragmentation, ecological degradation, and widespread biodiversity loss [5,6]. Currently, 40.7% of all the known species worldwide are still threatened, and 21% of the natural wetlands have been lost [7,8]. By 2100, global wetland loss is projected to reach 6000 km2 [9,10], further intensifying the risks of habitat degradation and biodiversity decline [11]. To address these challenges, the international community, including China, has undertaken extensive efforts in biodiversity conservation, such as ratifying international conventions, designating nature reserves, and developing ecological networks [12,13,14]. Recently, nations have committed to protecting 30% of Earth’s land and water by 2030 [15,16,17]. Nonetheless, it remains unclear whether the designation of these protected areas will result in more effective biodiversity conservation [18,19]. Therefore, it is essential to conduct comprehensive habitat assessments and identify priority areas for conservation to ensure the long-term sustainability of regional biodiversity.
Previous studies have primarily focused on analyzing the spatiotemporal dynamics or predicting the future trends in habitat patches’ characteristics at multiple scales and in various regions [20]. Some studies have analyzed changes in habitat quality or suitability using multi-source datasets [21,22]. Others have explored the spatiotemporal driving mechanisms of habitat quality or patterns [23]. Additionally, some studies have estimated the ecological impacts of anthropogenic threats on the habitats of specific species [24] or have assessed habitat quality in urban expansion scenarios using Cellular Automata and Markov chain models [25,26]. However, these studies have primarily concentrated on annual timescales, with limited attention given to seasonal dynamics [27]. In fact, seasonal fluctuations can significantly alter the abundance and distribution of natural resources, directly affecting habitat availability and the spatial distribution of species [28,29,30]. As a result, resource availability can fluctuate substantially between seasons. These seasonal dynamics are especially challenging for climate-sensitive bird populations, highlighting the need for conservation strategies that are responsive to seasonal changes [31]. For instance, the accurate identification of key habitats and the minimization of human disturbance during the breeding season are critical [32,33]. However, knowledge of seasonal variability in habitats remains limited. Therefore, assessing seasonal changes in habitat characteristics and patterns is essential for identifying priority habitats and developing effective conservation strategies that can adapt to both seasonal and climatic variability.
Identifying priority habitats in complex landscapes presents new opportunities due to the popularity and accessibility of various models, such as Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST), Morphological Spatial Pattern Analysis (MSPA), and the Maximum Entropy Model (MaxEnt) [34,35,36,37,38]. Typically, priority habitats are identified by assessing habitat conditions, such as habitat quality, suitability, and connectivity, and subsequently selecting areas that score highly in these aspects [39]. Among these models, the Habitat Quality (HQ) model in InVEST is widely utilized because it provides valuable information with relatively modest data requirements [40,41]. However, the accuracy of habitat quality assessments depends on the setting of the key parameters [25], particularly “habitat suitability” and “the sensitivity of the habitat to threats.” Previous studies have often relied on the literature or expert knowledge to determine these parameters rather than quantitative analyses [26,42]. For example, “habitat suitability” is typically assigned a fixed value based on the land-use type, with forest land often assumed to be optimal and given the highest suitability value of 1 [21]. Yet habitat suitability for any given species is inherently complex and spatially heterogeneous, influenced not only by environmental factors, such as land type, topography, climate, and vegetation abundance, but also by species-specific biological traits, including ecological preferences, population size, reproductive rates, and conservation status [43,44]. Similarly, the sensitivity of the habitat to threats can fluctuate depending on the stage of urban development and the geographic location [7]. For example, in areas dominated by agriculture, habitats may be more sensitive to cropland than to urban land. Furthermore, most previous studies have primarily focused on identifying priority habitats from the perspective of constructing ecological security patterns or ecological networks, while less attention has been given to incorporating the ecological characteristics and requirements of species in the study area [45]. Effective biodiversity conservation requires knowledge of the distribution and habitat preferences of the species present in the area. Therefore, identifying priority habitats based on habitat quality assessment remains challenging, particularly when incorporating spatially heterogeneous habitat suitability and quantitative analyses of habitat sensitivity to threats from the perspective of target species conservation.
To address the limitations discussed above, this study proposes a novel Habitat Suitability–Risk–Quality integrated assessment framework. The framework was implemented in Wuhan, Hubei, China, a key stopover along the East Asian–Australasian Flyway, one of the world’s major migratory bird routes that supports the annual movement of millions of waterbirds, to identify priority habitats with distinct seasonal characteristics. Specifically, the framework combines the MaxEnt model to evaluate the spatial heterogeneity of the habitat suitability for each season and the HRA model to quantify habitat risk and sensitivity to multiple threats. These results are integrated to refine the HQ model and improve the accuracy of the seasonal habitat quality assessment. Finally, K-means clustering is employed to identify priority habitat patches that maintain high quality across all seasons, thus supporting the development of differentiated conservation strategies. These findings can provide valuable information for biodiversity conservation and biodiversity-inclusive urban planning.

2. Materials

2.1. Study Area

Wuhan (113°41′–115°05′ E, 29°58′–31°22′ N), the capital of Hubei Province in central China, is located at the confluence of the Yangtze River and the Han River (Figure 1). The city covers an area of 8569.51 km2 and has a subtropical monsoon climate, characterized by distinct seasons, abundant precipitation, and considerable sunshine. Its ecosystems are primarily composed of freshwater and forest ecosystems, including 1380.42 km2 of water bodies and 2308.86 km2 of forested land, together accounting for 43.38% of the total area [46]. These natural resources provide essential habitats for a wide range of species, especially birds. Wuhan serves as a key stopover site along the East Asian–Australasian Flyway and supports rich avian biodiversity, with 381 species recorded, 74.01% of which are migratory birds. This includes Baer’s Pochard (Aythya baeri), listed as critically endangered on the IUCN Red List of Threatened Species, highlighting the ecological importance of this region.
Over the past two decades, Wuhan has undergone rapid urbanization. From 2000 to 2020, the population increased from 8.05 million to 12.45 million, while the urban built-up area expanded from 209.99 km2 to 885.11 km2, representing a 3.22-fold increase. To better understand the spatial structure of this expansion, it is important to distinguish between the city’s core and suburban areas. The main urban areas include Jiangan, Jianghan, Qiaokou, Hanyang, Wuchang, Qingshan, and Hongshan. The suburbs include Dongxihu, Hannan, Caidian, Jiangxia, Huangpi, and Xinzhou. The rapid urban expansion and associated human activities have led to habitat fragmentation and wetland loss, posing serious threats to local ecosystems and bird populations.

2.2. Data Sources and Preparation

The datasets utilized in this study are summarized in Table 1, including both annual and seasonal datasets. The annual datasets consist of eight categories. Land-use data, at a 30 m resolution, were classified into six types: cropland, forest, grassland, wetland, urban land, and bare land. Digital elevation model (DEM) data, at a 30 m resolution, were used to derive topographic parameters, such as elevation, slope, and aspect. Road data were categorized into high-level roads, including railways, highways, and urban expressways, and low-level roads, consisting of primary and secondary roads. GDP and population data, at a 1 km resolution, were adjusted using administrative district statistics from the Wuhan Statistical Yearbook. Building height data were categorized into four levels (<20 m, 20–50 m, 50–100 m, and >100 m) based on avian collision risk studies [39,47]. Water quality data were sourced from monitoring records reported in the Wuhan Water Resources Bulletin. Nature reserves were designated as strictly protected and inaccessible areas.
The seasonal datasets included bird observation data, Normalized Difference Vegetation Index (NDVI) data, temperature data, precipitation data, nighttime light (NTL), data, and urban boundary data. These data were synthesized according to the following four periods: spring (March, April, and May), summer (June, July, and August), autumn (September, October, and November), and winter (December, January, and February). Bird observation data were filtered to include 17 key and common bird species with long-term monitoring records and sufficient sample sizes, ensuring that the priority habitats would benefit the majority of the bird species in the region. The filtering process involved two steps: First, the top 30 bird species with the largest sample sizes and highest occurrence frequencies were selected; second, the screening results were validated against annual bird-monitoring reports and previous studies, resulting in a final set of 17 species. For these selected species, we also collected key biological and ecological parameters to inform subsequent habitat suitability modeling (see Table S1). The dataset included 2296 sample points, which were assigned to seasonal bird datasets based on their recorded months (see Table S2). The 250 m 16-day NDVI and 500 m monthly nighttime light data were processed using the Maximum Value Composite (MVC) method to extract seasonal maximum values, while the 1 km monthly temperature and precipitation data were processed using the Average Value Composition (AVC) method to calculate seasonal averages. Urban boundaries were identified using the Vegetation Adjusted NTL Urban Index (VANUI) [48,49], a composite index that integrates NDVI and NTL data. This method defines urban boundaries based on the extent of human activity and has proven to be highly accurate [50]. The identification process involved two steps: First, the VANUI index was calculated for each season; second, the index was classified into two categories using the natural breaks method, with high-value areas identified as seasonal urban boundaries [51].
All the data were uniformly projected to the UTM Zone 49N coordinate system to ensure spatial consistency. Raster data were resampled to a 30 m resolution, with building height data processed using the nearest neighbor method and other raster data using bilinear interpolation.

2.3. Framework

In this study, an analytical framework was developed that integrates the Maximum Entropy (MaxEnt) model and the Habitat Risk Assessment (HRA) model to enhance the Habitat Quality (HQ) model for assessing the seasonal habitat quality. Furthermore, the K-means clustering algorithm was applied to cluster the seasonal habitat quality and to identify priority bird habitat patches for conservation (Figure 2).
In this framework, the land cover types of the wetland, forest, and grassland were initially identified as habitat types, representing the major ecological environments utilized by target bird species. Next, the MaxEnt model was employed to predict bird habitat suitability in different seasons under complex environmental conditions. This process generated bird habitat suitability with clear spatial heterogeneity. Subsequently, the HRA model quantified habitat risk under both single and multiple threat factors, producing a habitat–threat risk matrix and habitat risk maps. The outputs from the MaxEnt and HRA outputs were then incorporated into the HQ model to enhance the evaluation of the seasonal bird habitat quality. Finally, the K-means algorithm was employed to cluster the seasonal habitat quality and identify priority bird habitats for conservation, thereby informing conservation strategies. Further methodological details are provided below.

2.3.1. Prediction of Habitat Suitability Using MaxEnt

In this section, the seasonal habitat suitability of birds in Wuhan was predicted using the MaxEnt model, a machine-learning algorithm based on the maximum entropy principle [52,53,54]. This model estimates a probability distribution that satisfies given constraints and has been widely used to predict species habitat suitability and occurrence probability across various regions [34,55,56]. The prediction process consists of three main steps: data input, model construction and validation, and result analysis.
The input data consisted of bird occurrence data and environmental variables. First, bird occurrence data were spatially rarefied by species for each season at a 100 m resolution using the SDM tool to reduce spatial autocorrelation and overfitting [57,58]. Each species was required to have at least five occurrence records per season to meet the model’s minimum data requirement [59]. The final sample sizes, defined as species-specific bird occurrence records, were 373 (spring), 421 (summer), 904 (autumn), and 598 (winter) for the 17 bird species (see Table S2). Importantly, habitat suitability modeling was conducted separately for each species to preserve ecological differences in habitat use. Species-specific MaxEnt models were run for each season, accounting for variations in habitat preferences, behaviors, and ecological traits. In addition, 16 environmental variables related to land use, topography, climate, vegetation, and distance factors were selected based on previous studies and expert opinions and are listed in Table 2 [60,61,62]. Among these, precipitation, temperature, NDVI, and distance to urban areas were the variables that reflected seasonal dynamics. To mitigate multicollinearity, Pearson correlation analysis was performed, ensuring no strong correlations among the variables (|r| < 0.85; see Figure S1).
To ensure model stability and result reliability, several parameters were configured [63,64]. First, ten-fold cross-validation was applied, and the area under the receiver operating characteristic curve (AUC) was calculated to evaluate the model performance [65]. Based on Swets’ classification for AUCs [66], model accuracy is categorized as follows: low (0.5–0.7), good (0.7–0.9), and excellent (>0.9). In this study, the AUC values for habitat suitability predictions across all bird species and seasons exceeded 0.7, indicating that the model exhibited good predictive and explanatory abilities (see Table S2). Second, bird habitat suitability was output in logistic format, representing species presence probability on a scale from 0 to 1, with higher values indicating greater habitat suitability [54].
To obtain a composite habitat suitability map that reflects the importance of habitats for multiple species, the seasonal suitability values of all the species were averaged for each grid cell. This method facilitated the identification of habitat hotspots while maintaining interspecific distinctions through species-specific modeling. Finally, the habitat suitability for the forest, wetland, and grassland was extracted and categorized into six classes using the natural breaks method. These classification results served as the foundation for subsequent habitat quality modeling.

2.3.2. Assessment of Habitat Risk Using the HRA Model

In this section, the cumulative risk to habitats from various external threats, typically human activities, in Wuhan was assessed using the HRA model. The results provided insights into habitat–threat interactions [67], which supported subsequent habitat quality modeling.
First, habitats and threats were defined by classifying land-use data based on the biological and ecological characteristics of the target species (Table S1). Habitat types included forest, wetland, and grassland, while threat types consisted of cropland, urban land, and bare land, all associated with agricultural and industrial activities. Next, a habitat risk assessment framework was established by selecting a set of criteria from both the exposure and consequence dimensions (Table 3). The selected criteria were then graded, scored, rated for data quality, and weighted based on previous research and expert opinions [68,69,70,71]. Exposure refers to the degree to which habitats are threatened and includes three criteria: spatial overlap, temporal overlap, and intensity. Spatial overlap quantifies the proportion of the habitat area overlapping with threats; temporal overlap indicates the duration of the exposure to threats, and intensity reflects the strength of the threats. The classification of these criteria was based on the HRA model documentation and previous studies [68,72,73]. Consequence refers to habitat sensitivity and resilience to threats, represented by four criteria: change in area, connectivity, water quality, and NPP. Change in area reflects habitat sensitivity, indicating the degree of habitat encroachment by threats, with thresholds were defined based on the HRA model documentation and Caro et al. (2020) [67,73]. Connectivity, water quality, and NPP reflect habitat resilience. Among them, connectivity supports ecological processes, and its classification was determined according to the HRA model documentation and Bastos et al. (2023) [67,74]. Water quality and NPP indicate habitat health. The classification for water quality was based on China’s National Surface Water Environmental Quality Standard (GB 3838–2002), where Class III water is suitable for fish overwintering and spawning [75]. The classification thresholds for NPP were defined with reference to the values reported by Li et al. (2018) and Chen et al. (2020), which indicated that the mean NPP of forests in Wuhan is approximately 0.5 kgC/m2/a, while non-forest and grassland types are generally below 0.2 kgC/m2/a [76,77]. Subsequently, each criterion was classified into three levels and assigned scores from 1 to 3, with higher scores indicating greater exposure or consequence [78]. Similarly, data quality and criterion weights were categorized into three levels, with scores ranging from 1 to 3, where higher scores denoted lower data accuracy or greater weight [79]. Finally, habitat risk was calculated using the Euclidean-distance-based risk function, as described in Equations (1)–(4) [68,74].
R h s = E 1 2 + C 1 2
E = i = 1 N e i d i · w i i = 1 N 1 d i · w i
C = i = 1 N c i d i · w i i = 1 N 1 d i · w i
where R h s represents the risk value of habitat h exposed to threat s , and E and C denote the habitat exposure index and consequence index, respectively, which are derived from the weighted average of the exposure value ( e i ) and the consequence value ( c i ) for each criterion ( i ), respectively. The exposure value ( e i ) and consequence value ( c i ) are assigned according to Table 3; d i and w i represent the data quality rating and importance weight for criterion i , respectively, both ranging from 1 to 3.
The cumulative risk value of all the threats to habitat h is given by
R h = s = 1 S R h s
where R h denotes the cumulative risk value of all the threats to habitat h , which is the sum of all the risk values for the combination of habitat h and threat s . Since the maximum risk value was defined as 3 in the model, all the calculated risk values were normalized to a 0–3 scale and classified into four levels: no risk (0), low risk (0–1), medium risk (1–2), and high risk (2–3).
The model generated a habitat–threat interaction table containing exposure, consequence, and risk values for each habitat–threat pair. The risk value of each pair can represent the sensitivity of the habitat to the corresponding threat. These results were adjusted based on previous research and expert opinions and served as input data for the habitat quality assessment model. Furthermore, the results presented the spatial distribution of risk values for individual habitat types, with risk values ranging from 0 to 3, where higher values indicated greater risk. In addition, the assessment of the habitat risk was conducted using annual data rather than seasonal data, primarily due to the difficulty in obtaining seasonal land-use data. Although other seasonal datasets, such as NDVI and nighttime lights, could be used for adjustments, their coarser spatial resolution was insufficient to match the land-use data.

2.3.3. Evaluation of Habitat Quality Using the Improved HQ Model

The seasonal habitat quality for birds in Wuhan was assessed using an improved habitat quality (HQ) model. To refine this model, two key parameters were adjusted based on the results from the MaxEnt and HRA models: habitat suitability and the relative sensitivity of habitats to threats [26]. These modifications aimed to enhance the accuracy of the habitat quality assessment, thereby ensuring the effective conservation of species with distinct habitat requirements [67].
Specifically, (1) the seasonal habitat suitability values in the HQ model were derived by integrating land-use data with seasonal suitability rankings from the MaxEnt model for land-use types classified as habitat in each season. The assigned values ranged from 0 to 1, with non-habitat land-use types assigned a suitability value of 0, indicating unsuitability. For land-use types classified as habitat, suitability values were assigned based on their seasonal ranking. Given that habitats with lower suitability still provide ecological functions [80], the initial value was set at 0.5 and increased in increments of 0.1, with higher values indicating greater habitat suitability (Table 4). (2) The relative sensitivity of habitats to threats was calculated as a weighted average of the interaction intensity between habitat–threat pairs in the HRA model and expert opinions (Table 5) [39]. Additionally, two other important parameters in the HQ model were determined based on previous studies and expert opinions: the maximum impact distance of threats, and the weights and distance-decay function of each threat (Table 6) [23,25]. Finally, the seasonal habitat quality of birds in Wuhan was evaluated by converting habitat degradation values using the half-saturation function, as described in Equation (5).
Q x j = H x j 1 D x j z D x j z + k z
where Q x j [ 0 , 1 ] denotes the habitat quality of grid cell x in land-use type j , with higher values indicating higher habitat quality. H x j [ 0 , 1 ] represents the habitat suitability of grid cell x in land-use type j , where higher values indicate higher suitability. Additionally, z is a scaling parameter with a default value of 2.5, k is a half-saturation constant that equals half of the highest degradation value, and D x j z denotes the total threat level in grid cell x in land-use type j , which is expressed as follows:
D x j = r = 1 R y = 1 Y r c r r = 1 R c r r y z r x y β x S j r
z r x y = 1 d x y d r   m a x
z r x y = exp 2.99 d r   m a x d x y
where R and r represent the number of threats and a specific threat, respectively. Y r is the set of all the grid cells in the raster map of threat r , and y denotes a specific cell in that set. Additionally, r y indicates the threat intensity of threat r in cell y ; c r [ 0 , 1 ] denotes the impact weight of threat r , with higher values indicating a greater threat to habitat integrity, meaning a higher degree of habitat degradation; z r x y denotes the impact of threat r in grid cell y on the habitat in grid cell x , modeled by a linear or exponential decay function based on distance, as represented by Equations (7) and (8); d x y is the distance between the habitat grid ( x ) and the threat grid ( y ); d r   m a x is the maximum impact distance of threat r ; β x [ 0 , 1 ] denotes the level of accessibility in grid cell x , where 1 indicates complete accessibility (e.g., urban areas), and 0 indicates complete inaccessibility (e.g., nature reserves); and S j r [ 0 , 1 ] indicates the sensitivity of land-use type j to the threat ( r ), with values closer to 1 indicating higher sensitivity.

2.3.4. Identification of Seasonal Priority Habitats Using K-Means Clustering

In this section, the seasonal bird habitat quality, at the pixel scale, in Wuhan was clustered using the K-means clustering algorithm to identify priority habitat patches for protection and to develop a tiered conservation strategy. K-means clustering is an unsupervised learning algorithm that groups habitat patches based on the similarity of their seasonal habitat quality values [81,82,83,84,85]. First, the habitat quality values for each season were standardized using Z-score normalization to eliminate scale differences between seasons. Then, the optimal number of clusters (K = 4) was determined using the elbow method and silhouette score [86,87]. The sum of squared errors (SSE) was plotted against K-values ranging from 2 to 9, and the resulting curve showed an inflection point between K = 3 and 5, indicating a potential elbow. These results were further validated through silhouette analysis, which confirmed that K = 4 was the optimal number of clusters (Figure S2). Next, the standardized seasonal habitat quality feature data were assigned to the nearest cluster centers using the Euclidean distance metric. Subsequently, to validate the ecological relevance of the clusters and identify existing conservation gaps, a spatial overlay analysis was conducted between the clustering results and the distribution of nature reserves, forest parks, and natural scenic areas in Wuhan. Finally, based on the clustering results and the characteristics of each cluster, corresponding conservation strategies were developed, with habitat patches maintaining high quality across all seasons identified as priority areas for protection.

3. Results

3.1. Bird Habitat Suitability

3.1.1. Seasonal Variation in Bird Habitat Suitability

The results indicate significant seasonal variation in bird habitat suitability (BHS) in Wuhan (Table 7). For all the habitats, suitability was highest in autumn (0.1393) and winter (0.1123), with greater spatial heterogeneity (standard deviation (SD): 0.16 and 0.11, respectively). In contrast, spring (0.0745) and summer (0.0504) had the lowest BHS values and more spatially uniform distributions (SD: 0.10 and 0.08). Among the different habitat types, the wetland exhibited the highest suitability, particularly in autumn (0.1854) and winter (0.1405), indicating its importance as foraging and roosting habitats during these seasons. The forest showed the lowest suitability throughout the year (ranging from 0.0385 to 0.0665), while the grassland displayed moderate suitability, with peaks in spring (0.0939) and autumn (0.0922). At the administrative level, BHS in urban areas was significantly higher than in suburban areas, with seasonal means ranging from 1.30 to 4.68 times greater. Seasonal fluctuation in urban habitats was also more pronounced. For instance, in Hongshan District, wetland suitability reached 0.37 in autumn, nearly three times higher than in summer (0.13), highlighting strong seasonal variation in wetland suitability within urban areas.
Furthermore, percentage contribution analysis identified six key variables influencing seasonal BHS: land-use type, high-level roads, low-level roads, NPP, distance to urban areas, and distance to forest. These variables jointly accounted for 58.27% to 81.68% of the variation (Table S3), providing a critical reference for selecting threat factors in habitat quality assessments.

3.1.2. Habitat Suitability Levels and Spatial Distribution

BHS was classified into six levels using the natural breaks method: very low (<0.049), low (0.049–0.126), medium (0.126–0.226), medium–high (0.226–0.354), high (0.354–0.516), and very high (>0.516) (Figure 3). Overall, low-level BHS dominated the spatial distribution, primarily in the southern and northern parts of Wuhan. The total area and proportion of low BHS in each season were 987.59 km2 (63.59%) in spring, 1121.16 km2 (72.19%) in summer, 554.20 km2 (35.69%) in autumn, and 601.65 km2 (38.74%) in winter.
In comparison, habitats with medium and higher suitability (i.e., medium, medium–high, high, and very high levels) were mainly concentrated in the urban area and exhibited clear seasonal shifts. In spring, these areas covered 263.12 km2 (16.94% of the total habitat area), primarily distributed along the Yangtze River and large lakes. In summer, the area declined sharply to 146.39 km2 (9.43%) and became highly fragmented. In autumn, it expanded significantly to 528.82 km2 (34.05%), forming more contiguous patches in the main urban area and the urban–suburban fringe, with the highest suitability observed along the Yangtze River and its adjacent lakes. In winter, the area slightly decreased to 490.69 km2 (31.60%) but remained stable at a relatively high level, particularly in locations such as Chenhu in Caidian District, the Fu River in Dongxihu District, and Donghu Lake in Hongshan District. Notably, habitats with medium and higher suitability were predominantly composed of wetlands, especially in autumn and winter, when wetlands accounted for up to 91.50% of these areas. In contrast, forests and grasslands accounted for smaller proportions, which aligns with the dominance of waterbird species in Wuhan and highlights the critical role of wetlands in providing suitable habitats.

3.2. Bird Habitat Risk

3.2.1. Threat Intensity and Risk to Bird Habitats

The habitat risks associated with both multiple and single threats at the city and district levels are illustrated in Figure 4a and Figure 4b, respectively. These results reveal the interactions between habitat types and different sources of anthropogenic threats.
At the city level, all the habitats in Wuhan had risk values below the upper threshold for medium risk (value = 2), indicating that they were subject to moderate or lower pressure (Figure 4a). Among the habitat types, the wetland exhibited the lowest risk value, followed by the forest, while the grassland experienced the highest level of risk under both single and multiple threats. In terms of individual threats, the cropland and urban land exerted relatively strong pressure on habitats, particularly on the grassland, leading to medium risk values of 1.28 and 1.11, respectively. In contrast, the bare land posed minimal ecological pressure, with a risk value of 0.1 for the grassland, mainly due to its limited and fragmented spatial distribution in the study area.
At the district level, both habitat risk and threat intensity varied significantly between the main urban and suburban areas (Figure 4b). Overall, habitat risk was higher in the main urban districts than in the suburban ones. For instance, under multiple threats, the forest risk value in Jianghan District reached 1.56, which is 3.13 times higher than that in Huangpi District (0.49). In terms of individual threats, the urban land was the dominant stressor in the urban districts, while the cropland posed greater risks in most suburban areas. Notably, a reversed pattern was observed at the urban–suburban interface. In Dongxihu District, for instance, the risk to forest habitats from the urban land reached 2.21, exceeding that from the cropland (2.05). This is largely due to Dongxihu’s designation as the Wuhan Airport Economic Development Zone, a national-level development area that has experienced rapid urban expansion driven by industrial growth and infrastructure upgrading, significantly intensifying ecological pressure in the region.

3.2.2. Spatial Patterns of Bird Habitat Risk Levels

The spatial distribution of bird habitat risk levels in Wuhan is shown in Figure 4c. The areas of medium-, low-, and no-risk habitats were 176.30 km2 (11.35%), 1020.72 km2 (65.73%), and 355.99 km2 (22.92%), respectively. This indicates that most bird habitats in Wuhan are characterized by low or negligible ecological risk. Low-risk habitats were primarily composed of large wetlands (668.12 km2) and forest areas (323.83 km2), which are widely distributed in outer suburban districts, such as Huangpi, Jiangxia, and Caidian Districts, as well as in parts of Hongshan District within the main urban area. No-risk habitats were typically located at the cores of large habitat patches, where exposure to surrounding threats was minimal. Medium-risk habitats were generally found along the edges of forest and wetland patches or in isolated and fragmented areas, with a scattered distribution across the city.

3.3. Bird Habitat Quality

3.3.1. Seasonal Variation in Bird Habitat Quality

Bird habitat quality in Wuhan exhibited significant seasonal variation, with higher average values in autumn (0.6063) and winter (0.5938) and lower values in spring (0.5418) and summer (0.5208) (Table 8). The standard deviations indicated greater spatial variability in habitat quality during autumn and winter and more uniform distributions in spring and summer. In terms of habitat types, there was little difference in the quality among the wetland, forest, and grassland in spring and summer. However, in autumn and winter, wetland habitats consistently exhibited higher quality than the forest and grassland. In addition, seasonal fluctuations in the wetland quality were the most pronounced, with autumn (0.6475) and winter (0.6228) being significantly higher than spring (0.5400) and summer (0.5224). The forest quality ranged narrowly from 0.5180 to 0.5443, with low standard deviations, indicating relatively high temporal stability and limited sensitivity to seasonal variation. The grassland quality also remained relatively stable, with the lowest quality in summer (0.5108).

3.3.2. Spatial Patterns and Seasonal Dynamics of Bird Habitat Quality

Bird habitat quality in Wuhan showed significant spatial heterogeneity across all four seasons. Quality values were consistently low in the northern and southern parts of the city, higher in the central transitional zones, and the lowest in the urban core (Figure 5). The low-quality areas in the north and south were mainly located in the outer suburban districts, such as southern Jiangxia, northern Huangpi, and northern Xinzhou. Although these regions have abundant forest resources, they are suboptimal for the dominant waterbird species in Wuhan. The highest-quality habitats were concentrated in the central and south-central urban–suburban transitional zones (Figure 5e–h), including Houguan Lake and Chen Lake in Caidian District, East Lake and South Lake in Hongshan District, and Tianxing Island and Wuhui Dike in Qingshan District. Despite being adjacent to the urban core, these areas maintained relatively high habitat quality due to the presence of large water bodies, nature reserves, and extensive urban green spaces. Collectively, these areas formed a peri-urban ecological belt that encircles the urban core and serves as a key ecological buffer. In contrast, the lower-quality core encompasses the central urban districts, including Jianghan, Qiaokou, and Jiang’an, which exhibit degraded habitat quality and severe landscape fragmentation as a result of long-term intensive urban development.
The spatial pattern of the bird habitat quality also showed notable seasonal dynamics. In spring (Figure 5a,e), the ecological belt was mainly distributed across central and southern parts of the city. By summer (Figure 5b,f), this pattern contracted, especially at the junction of Caidian and Hanyang, where habitat conditions declined while improving in the suburbs. These changes may reflect the combined effects of heat stress, reduced water availability, and intensified human activity. In autumn and winter (Figure 5c,d,g,h), wetland and forest habitats reached their annual peak in quality, and the ecological belt expanded considerably, especially in wetland-rich districts, such as Caidian, Jiangxia, and Hongshan Districts, where wetland habitats exhibit exceptionally high quality, suggesting stronger ecological stability and higher carrying capacity in these regions.

3.4. Clustering of Bird Habitats Based on Seasonal Quality Dynamics

K-means clustering identified four distinct clusters of bird habitat based on seasonal quality dynamics at the grid scale (Figure 6). Specifically, Cluster 1 (C1) exhibited consistently high habitat quality across all seasons (centroid values: 0.754 in spring, 0.713 in summer, 0.850 in autumn, and 0.724 in winter), indicating strong seasonal adaptability. It covered 99.38 km2 (6.40% of the total habitat area) and was mainly distributed in the south-central region, particularly in Hongshan (40.58 km2), Jiangxia (21.68 km2), and Caidian (14.75 km2) Districts, which, together, account for 77.49% of the total area of C1. Wetlands dominated this cluster (80.01 km2, 80.51%), followed by forests (17.49 km2) and grasslands (1.88 km2).
Cluster 2 (C2) was characterized by moderate and stable habitat quality throughout the year (centroid values: 0.504, 0.489, 0.532, and 0.519 in spring, summer, autumn, and winter, respectively), suggesting that it was less affected by seasonal changes. It was the largest cluster, covering 843.15 km2 (54.29%) and was primarily located in Jiangxia, Huangpi, and Jiangan Districts. The cluster was mainly composed of wetlands (422.04 km2) and forests (394.78 km2), with most of it consisting of small, fragmented patches influenced by urbanization and agricultural activities.
Cluster 3 (C3) showed pronounced seasonal fluctuation, with the highest habitat quality in winter (0.730), moderate in autumn (0.644), and lower in spring (0.510) and summer (0.526). It spanned 387.60 km2 (24.96%) and was overwhelmingly dominated by wetlands (360.39 km2). Spatially, it was mainly distributed in suburban areas, such as Jiangxia, Caidian, and the southern parts of Huangpi and Xinzhou Districts.
Cluster 4 (C4) exhibited from moderate to high habitat quality overall, with peaks in autumn (0.713) and spring (0.644) and lower values in winter (0.581) and summer (0.546), showing significant seasonal fluctuations. It covered 222.89 km2 (14.35%), mainly composed of wetlands (169.95 km2) and forests (45.94 km2), with the majority located in Jiangxia, Caidian, and Hongshan Districts.

4. Discussion

4.1. Importance of Identifying Priority Habitats Using the “Habitat Suitability–Risk–Quality” Framework

Effective biodiversity conservation depends on the accurate identification of priority habitats, especially in dynamic urbanizing areas and along migratory bird flyways [36,88,89]. However, traditional conservation-planning approaches, such as designating nature reserves based solely on forest cover or using single-dimensional assessments, like habitat suitability, quality, or connectivity, are typically based on annual data and tend to remain static [90]. Although some studies have incorporated temporal dynamics, such as tidal influences in coastal stopover sites [19], these efforts have rarely extended to highly urbanized inland migration stopovers [91,92,93]. As a result, the key habitats identified often do not adequately account for seasonal adaptability or exposure to complex environmental pressures [94]. This issue is particularly acute in cities, like Wuhan, where rapid urbanization and strong seasonal changes constantly alter habitat availability.
To address these limitations, we proposed the integrated assessment framework “Habitat Suitability–Risk–Quality”, which combines spatially explicit habitat suitability maps from the MaxEnt model, habitat risk matrices from the HRA model, and refined parameters within the HQ model. This integration enables a more realistic evaluation of the seasonal habitat quality while accounting for species-specific ecological requirements. In traditional HQ models, fixed suitability values are typically assigned to broad land-use types (e.g., 1 for forests and 0 for urban areas) [21], which overlooks the spatial heterogeneity and seasonal dynamics. In contrast, our framework replaces these static values with dynamic suitability rankings derived from MaxEnt predictions, which account for seasonal variations in climate, vegetation, and human activity. For example, in Wuhan, wetland suitability varied from 0.0561 in summer to 0.1854 in autumn, which may be due to seasonal hydrological conditions. In 2020, precipitation in Wuhan during summer and autumn was significantly higher than the multi-year average, with rainfall from June to November exceeding historical levels by 48.9% to 214.2% [95]. Meanwhile, seasonal groundwater levels remained relatively stable, ranging between 21.71 and 23.11 m [96]. These findings suggest that abundant rainfall and steady groundwater storage in summer and autumn provided favorable hydrological conditions that helped to maintain high wetland habitat suitability in the autumn and winter seasons. In addition, habitat sensitivity to threats was recalibrated quantitatively using HRA outputs in this study, replacing traditional reliance on expert opinion, to better reflect interactions between habitats and anthropogenic pressures [97]. For instance, although grasslands are commonly assumed to be more vulnerable to urban expansion, our results revealed that croplands exerted a greater threat to grasslands than the urban land, with risk values of 1.28 under cropland pressure compared to 1.11 under urban pressure. This is largely due to the landscape configuration of grasslands in Wuhan, which are highly fragmented and frequently situated adjacent to agricultural zones, increasing their exposure to agricultural disturbances.
By incorporating these refined parameters into the HQ model, we achieved a more accurate assessment of the seasonal habitat quality. The results indicate that habitat quality in Wuhan is generally higher in autumn and winter than in spring and summer. This pattern is consistent with local bird field-surveys [98,99], suggesting that seasonal habitat conditions influence the spatiotemporal distribution of the avian diversity. Similar seasonal patterns have been observed in other East Asian flyway cities. For instance, long-term monitoring in Shanghai’s Dongtan wetlands has revealed significantly larger wintering bird populations than in summer [100]. In South Korea, waterbird species’ richness peaked during spring and autumn migrations but declined sharply in winter due to widespread wetland freezing [101,102]. In contrast, Wuhan’s milder winter climate with minimal wetland freezing allows high habitat suitability throughout the autumn and winter seasons. These comparisons suggest that despite regional climatic differences, the ecological importance of autumn and winter habitats is broadly consistent across East Asian flyway cities. In addition, our results showed that high-quality habitats were concentrated along a southwest axis (approximately 214° from the north), closely aligning with the Yangtze River. This spatial pattern matched migratory bird flight corridors identified through weather radar monitoring [103], further validating the effectiveness of our framework. These findings are consistent with previous studies emphasizing the value of parameter refinement in improving habitat quality assessments [25]. Unlike integrated approaches, such as InVEST combined with circuit theory, which primarily focus on migration routes and ecological connectivity [36], our framework emphasizes seasonal habitat conditions. By integrating seasonally differentiated habitat suitability from the MaxEnt model, ecological risk from the HRA model, and refined parameters in the HQ model, our approach provides a reliable and effective tool for evaluating the seasonal habitat quality and identifying conservation priorities, particularly in rapidly urbanizing inland regions.

4.2. Conservation Gaps and Seasonal Strategies Based on Habitat Clusters

The current protected area in Wuhan exhibits significant spatial mismatches with habitats of the highest ecological value, revealing critical conservation gaps (Figure 6). The differences in protection levels among the four clusters are statistically significant (χ2 = 64.72, df = 3, p < 0.001) (Table 9). Cluster 1, which maintains high-quality habitat throughout the year, has the highest protection rate (46.51%) but the smallest total protected area (46.22 km2), leaving 53.16 km2 of ecologically valuable areas unprotected. These areas include key habitats, such as Qingshan Wetland, Wuhui Dyke, and the riparian wetlands along the Yangtze River between Yingwuzhou and Baishazhou Bridges. Notably, 60.59% (32.21 km2) of these unprotected areas is located within 5 km of the main urban areas, including Baishazhou in Wuchang District and Wugang in Qingshan District, making them particularly vulnerable to industrial infrastructure and intensive human activities. Cluster 3, which consists of habitats with high quality during winter, also exhibits a relatively high protection rate (46.30%) and a much larger total protected area (179.44 km2). Critical wetlands in this cluster, such as Chenhu Wetland in Caidian District, Zhangdu Lake in Xinzhou District, and East Lake in Hongshan District, are formally protected. Cluster 2, characterized by medium habitat quality year round, contains the largest area under formal protection (187.47 km2), accounting for 22.23% of its total area. Protection areas include Anshan National Wetland Park in Jiangxia District and Mulan Mountain Nature Reserve in Huangpi District. Cluster 4, which reaches peak habitat quality during migration seasons, has a protection rate of 31.50% (70.21 km2 out of 222.89 km2). Unprotected areas in this cluster, such as Bafen Mountain in Jiangxia, Guanlian Lake, and the Tongshun and Maying Rivers in Caidian, continue to face pressure from agricultural activities and infrastructure development.
This spatial mismatch highlights a broader challenge in global conservation efforts [88,104,105]. In rapidly growing cities, like Wuhan, these mismatches are often caused by urban sprawl, especially urban-planning decisions and urban functional zoning [106]. For example, industrial developments along the Caidian–Hannan border and in western Jiangxia District have been largely driven by economic priorities, often overlooking the ecological roles of these areas, such as supporting seasonal bird migration and sustaining wetland biodiversity. Moreover, conservation strategies still tend to focus on large, remote habitat patches, while smaller, high-value habitats within or near urban areas are frequently neglected [92,107]. Although modest in size, these habitats are crucial for maintaining ecological connectivity, particularly as stepping stones, along migratory routes. Yet their scattered distribution and exposure to development pressure have left many without formal protection. Taken together, the effectiveness of our framework in capturing seasonal ecological dynamics and urban complexity enables more accurate identifications of habitat clusters and critical conservation gaps. This is essential for balancing ecological integrity with long-term sustainable urban development.
To address these gaps, conservation strategies should be developed based on the characteristics and seasonal dynamics of habitat clusters [106]. First, Cluster 1 should be legally designated as an ecological priority protection zone to prevent urban encroachment and overexploitation. For instance, land reclamation in East Lake should be strictly prohibited, and a 1 km buffer zone should be established to mitigate edge effects. This measure is consistent with China’s Ecological Protection Red Line policy [108,109,110]. In addition, hydrological restoration, such as dredging blocked waterways in the Houguan Lake, Jinlong Lake, and Suozichang River wetland complexes in Caidian District, should be implemented. This measure has been shown to be effective, as restoring hydrological connectivity in fragmented wetland systems can significantly improve ecological function and biodiversity [111,112]. Second, for Cluster 2, ecological corridors 100–300 m wide should be constructed along major rivers, such as the Han River, to connect medium- and high-quality habitats, promoting species movement and genetic exchange [113]. This type of riparian habitat restoration has proven to be effective in both Beijing, China, and the Colorado River Delta, Mexico, demonstrating its broader applicability across urbanizing and degraded river systems [114,115]. These findings underscore the role of riparian corridors in mitigating the negative effects of urbanization on avian communities. Additionally, low-impact agricultural practices, such as reduced pesticide use, should be adopted in adjacent croplands to alleviate long-term ecological pressures. Third, Cluster 3 requires seasonal management [19]. From November to February, activities such as fishing and boating should be restricted in key overwintering habitats, such as Chenhu and Fuhe wetlands, to minimize disturbances to species like Baer’s Pochard (Aythya baeri). Meanwhile, restoration efforts should focus on planting native vegetation along degraded wetland edges. These plants not only improve microhabitat conditions but also provide essential shelter and food resources for wintering birds [116], thereby enhancing the overall habitat quality and ecological function. Fourth, Cluster 4 should also implement seasonal management strategies [117] but with focuses on spring and autumn. Construction and recreational activities should be prohibited during peak migration and breeding periods. This aligns with conservation guidelines by the U.S. Fish and Wildlife Service, which recommend limiting human disturbances, such as construction and noise, during sensitive biological periods, including migration seasons [118,119]. In addition to formally protected areas, Other Effective area-based Conservation Measures (OECMs) offer a complementary approach to biodiversity conservation, particularly in urbanizing regions [120]. OECMs are defined as areas that are not designated as protected areas but are governed and managed in ways that deliver long-term biodiversity benefits [121]. In the context of Wuhan’s habitat clusters, certain urban wetlands, and riparian buffer zones that exhibit ecological value and limited human disturbance, could potentially qualify as OECMs [122]. Identifying and integrating these areas into the broader conservation network may enhance ecological connectivity, support seasonal species needs, and promote long-term sustainability goals.
In summary, this study highlights the necessity of incorporating seasonal dynamics into conservation planning. Traditional approaches, which emphasize static protected areas, often fall short in responding to seasonal habitat shifts and pressures at urban edges. In contrast, the cluster-based seasonal conservation strategy proposed in this study offers a replicable and flexible framework for balancing ecological protection with human development in rapidly urbanizing regions.

4.3. Limitations and Future Outlook

Although the proposed framework provides a valuable contribution to seasonal habitat assessments, several limitations remain. First, the spatial and temporal resolutions of the input datasets remain key constraints. While 250 m NDVI imagery was used to explore seasonal vegetation dynamics, this resolution may not be sufficient to detect fine-scale or short-term habitat changes, such as temporary wetlands formed during the rainy season [123]. Additionally, due to the absence of seasonally explicit land-use data, the HRA model relied on annual inputs. This approach may fail to reflect critical seasonal threats, such as irrigation in agricultural areas or tourism pressure in wetlands, thereby limiting the temporal accuracy of risk assessments. Future research should incorporate higher-resolution, temporally explicit spatial data (e.g., monthly land use or 10 m imagery) to more accurately delineate both temporary and permanent habitats as well as dynamic urban boundaries. Second, although all the MaxEnt models performed well, with AUC values exceeding 0.7, residual spatial autocorrelation may still influence the model’s performance. Spatial thinning was applied to reduce clustering, but the limited sample size for some species precluded the use of more rigorous spatial validation techniques, such as spatial blocking or independent test datasets. This limitation may have led to a modest overestimation of the predictive accuracy. Future applications should explore more robust spatial validation strategies to enhance the model’s reliability. Third, this study employed a multi-species approach, based on 17 bird species, to identify regional habitat priorities. While this method is effective for revealing general patterns, it does not account for species-specific seasonal responses or ecological preferences. This limitation is particularly relevant for rare or endangered species, which habitat requirements may differ markedly from those of common species and may be more sensitive to seasonal or habitat changes. Future research should consider more refined, species-level analyses to support targeted conservation strategies for these vulnerable groups. Moreover, the study’s focus on birds may have overlooked the habitat requirements and seasonal responses of other taxa. For example, mammals may be more dependent on forest connectivity during migration and breeding [124,125], which may not coincide with bird conservation priorities. To ensure more inclusive biodiversity conservation, future work should adopt a multi-taxon approach by integrating species distribution models for birds, mammals, amphibians, and invertebrates [34,126,127]. Fifth, conservation gaps were identified by overlaying habitat clusters with various protected areas (e.g., nature reserves, wetland parks, and forest parks) but lacked field validation. Future research should incorporate on-the-ground surveys to verify these gaps and support the identification of overlooked conservation sites. Finally, incorporating climate variables in diverse emission scenarios could enhance the framework’s ability to predict long-term habitat shifts and identify climate-resilient areas [26,128], which could represent a valuable direction for future research.

5. Conclusions

This study proposed a novel integrated “Habitat Suitability–Risk–Quality” framework to improve the seasonal assessment of habitat quality and identify priority bird habitats in rapidly urbanizing areas. By integrating the MaxEnt, HRA, and HQ models, the framework incorporates spatially heterogeneous habitat suitability and quantified habitat sensitivity into the HQ model, thereby enhancing the accuracy and reliability of seasonal habitat quality assessments. Based on this, K-means clustering was applied to identify four distinct habitat clusters, each with unique seasonal characteristics, providing a foundation for the development of differentiated conservation strategies. The framework was applied to Wuhan, a key stopover along the East Asian–Australasian Flyway, and demonstrated strong potential for supporting biodiversity conservation and sustainable land-use planning in complex urban landscapes.
The results revealed that (1) wetlands provide the most suitable and highest-quality habitats in autumn and winter, especially along urban–rural fringes, while forest and grassland habitats exhibit lower suitability and quality across seasons. (2) Grasslands are subject to higher ecological risks, particularly from cropland expansion, while wetlands remain relatively less vulnerable. (3) There are substantial mismatches between existing protected areas and ecologically important habitats, especially those with year-round high quality or those important during migration seasons. These findings highlight the limitations of static conservation strategies and underscore the necessity of incorporating seasonal habitat dynamics and species-specific requirements into conservation planning.
To address conservation gaps, this study proposes cluster-specific, seasonally adaptive management strategies, such as prioritizing legal protection for year-round high-quality habitats, enhancing habitat connectivity for habitats of moderate quality, implementing seasonal management strategies for breeding and migratory habitats, and promoting habitat restoration at urban edges. These measures aim not only to safeguard avian biodiversity but also to enhance long-term ecological resilience and support sustainable urban development. Overall, the proposed framework offers a replicable tool for biodiversity-inclusive planning and habitat protection, contributing to integrated ecological management in rapidly urbanizing inland regions.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/su17136078/s1. Figure S1: Pearson correlation matrices of the 16 environmental variables for each season. Figure S2: Determination of optimal number of clusters using the Elbow method and Silhouette Score. Table S1: Bird species selected in this study and their biological and ecological parameters. Table S2: Bird species, sample sizes, and test AUC values of the MaxEnt habitat suitability model for spring, summer, autumn, and winter. Table S3: Percent contribution of environmental factors to bird habitat suitability in four seasons.

Author Contributions

Conceptualization, J.W. and Y.Z.; data curation, C.L.; funding acquisition, Y.T., C.L., Y.Z. and H.Y.; investigation, C.L.; methodology, J.W. and Y.T.; resources, J.W. and Y.L.; software, J.W.; supervision, H.Y. and Y.L.; validation, C.L.; visualization, J.W.; writing—original draft, J.W.; writing—review and editing, J.W., Y.T. and Y.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported by the Postdoctoral Research Project of Shanghai Investigation, Design & Research Institute Co., Ltd. (Grant No. 2023HJ(83)-022), the National Natural Science Foundation of China (Grants No. 42371200, 42201290, and 42101284), the University Natural Science Project of Jiangsu Province (Grant No. Z231687), and the Yunnan Fundamental Research Projects (No. 202301AT070335). Special thanks to the referees and editors for their valuable comments, suggestions, and assistance in editing the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

Authors Junqing Wei and Hongzhou Yuan were employed by the company Shanghai Investigation, Design & Research Institute Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. De la Fuente, B.; Mateo-Sánchez, M.C.; Rodríguez, G.; Gastón, A.; Pérez de Ayala, R.; Colomina-Pérez, D.; Melero, M.; Saura, S. Natura 2000 sites, public forests and riparian corridors: The connectivity backbone of forest green infrastructure. Land Use Policy 2018, 75, 429–441. [Google Scholar] [CrossRef]
  2. Peng, J.; Yang, Y.; Liu, Y.; Hu, Y.; Du, Y.; Meersmans, J.; Qiu, S. Linking ecosystem services and circuit theory to identify ecological security patterns. Sci. Total Environ. 2018, 644, 781–790. [Google Scholar] [CrossRef] [PubMed]
  3. Kong, X.; Zhou, Z.; Jiao, L. Hotspots of land-use change in global biodiversity hotspots. Resour. Conserv. Recycl. 2021, 174, 105770. [Google Scholar] [CrossRef]
  4. Cohen, E.B.; Horton, K.G.; Marra, P.P.; Clipp, H.L.; Farnsworth, A.; Smolinsky, J.A.; Sheldon, D.; Buler, J.J. A place to land: Spatiotemporal drivers of stopover habitat use by migrating birds. Ecol. Lett. 2021, 24, 38–49. [Google Scholar] [CrossRef] [PubMed]
  5. Haddad, N.M.; Brudvig, L.A.; Clobert, J.; Davies, K.F.; Gonzalez, A.; Holt, R.D.; Lovejoy, T.E.; Sexton, J.O.; Austin, M.P.; Collins, C.D.; et al. Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci. Adv. 2015, 1, e1500052. [Google Scholar] [CrossRef]
  6. Van Moorter, B.; Kivimaki, I.; Panzacchi, M.; Saura, S.; Brandao Niebuhr, B.; Strand, O.; Saerens, M. Habitat functionality: Integrating environmental and geographic space in niche modeling for conservation planning. Ecology 2023, 104, e4105. [Google Scholar] [CrossRef]
  7. Fluet-Chouinard, E.; Stocker, B.D.; Zhang, Z.; Malhotra, A.; Melton, J.R.; Poulter, B.; Kaplan, J.O.; Goldewijk, K.K.; Siebert, S.; Minayeva, T.; et al. Extensive global wetland loss over the past three centuries. Nature 2023, 614, 281–286. [Google Scholar] [CrossRef]
  8. Luedtke, J.A.; Chanson, J.; Neam, K.; Hobin, L.; Maciel, A.O.; Catenazzi, A.; Borzée, A.; Hamidy, A.; Aowphol, A.; Jean, A.; et al. Ongoing declines for the world’s amphibians in the face of emerging threats. Nature 2023, 622, 308–314. [Google Scholar] [CrossRef]
  9. Schuerch, M.; Spencer, T.; Temmerman, S.; Kirwan, M.L.; Wolff, C.; Lincke, D.; McOwen, C.J.; Pickering, M.D.; Reef, R.; Vafeidis, A.T.; et al. Future response of global coastal wetlands to sea-level rise. Nature 2018, 561, 231–234. [Google Scholar] [CrossRef]
  10. Wang, G.; Wu, H.; Dai, J.; Xiong, Y.; Long, Y.; Cai, X.; Mo, S.; Yang, R.; Liu, Y. Priorities identification of habitat restoration for migratory birds under the increased water level during the middle of dry season: A case study of Poyang Lake and Dongting Lake wetlands, China. Ecol. Indic. 2023, 151, 110322. [Google Scholar] [CrossRef]
  11. Kong, X.; Fu, M.; Zhao, X.; Wang, J.; Jiang, P. Ecological effects of land-use change on two sides of the Hu Huanyong Line in China. Land Use Policy 2022, 113, 105895. [Google Scholar] [CrossRef]
  12. Sallustio, L.; De Toni, A.; Strollo, A.; Di Febbraro, M.; Gissi, E.; Casella, L.; Geneletti, D.; Munafò, M.; Vizzarri, M.; Marchetti, M. Assessing habitat quality in relation to the spatial distribution of protected areas in Italy. J. Environ. Manag. 2017, 201, 129–137. [Google Scholar] [CrossRef] [PubMed]
  13. Ulloa, A.M.; Jax, K.; Karlsson-Vinkhuyzen, S.I. Enhancing implementation of the Convention on Biological Diversity: A novel peer-review mechanism aims to promote accountability and mutual learning. Biol. Conserv. 2018, 217, 371–376. [Google Scholar] [CrossRef]
  14. Courouble, M.; Davidson, N.; Dinesen, L.; Fennessy, S.; Galewski, T.; Guelmami, A.; Kumar, R.; McInnes, R.; Perennou, C.; Rebelo, L.-M.; et al. Convention on Wetlands. In Global Wetland Outlook: Special Edition 2021; Secretariat of the Convention on Wetlands: Gland, Switzerland, 2021. [Google Scholar]
  15. Sala, E.; Mayorga, J.; Bradley, D.; Cabral, R.B.; Atwood, T.B.; Auber, A.; Cheung, W.; Costello, C.; Ferretti, F.; Friedlander, A.M.; et al. Protecting the global ocean for biodiversity, food and climate. Nature 2021, 592, 397–402. [Google Scholar] [CrossRef]
  16. Allan, J.R.; Possingham, H.P.; Atkinson, S.C.; Waldron, A.; Di Marco, M.; Butchart, S.H.M.; Adams, V.M.; Kissling, W.D.; Worsdell, T.; Sandbrook, C.; et al. The minimum land area requiring conservation attention to safeguard biodiversity. Science 2022, 376, 1094–1101. [Google Scholar] [CrossRef]
  17. Li, G.; Fang, C.; Watson, J.E.M.; Sun, S.; Qi, W.; Wang, Z.; Liu, J. Mixed effectiveness of global protected areas in resisting habitat loss. Nat. Commun. 2024, 15, 8389. [Google Scholar] [CrossRef]
  18. Xia, S.; Yu, X.; Millington, S.; Liu, Y.; Jia, Y.; Wang, L.; Hou, X.; Jiang, L. Identifying priority sites and gaps for the conservation of migratory waterbirds in China’s coastal wetlands. Biol. Conserv. 2017, 210, 72–82. [Google Scholar] [CrossRef]
  19. Li, X.; Hou, X.; Shan, K.; Liu, Y.; Song, Y.; Wang, X.; Du, P.; Fan, C. Identifying shorebird conservation hotspots and restoration gaps in stopover sites: A perspective of ‘ecologically linked’ habitats. Glob. Ecol. Conserv. 2023, 48, e02725. [Google Scholar] [CrossRef]
  20. Barredo, J.I.; Caudullo, G.; Dosio, A. Mediterranean habitat loss under future climate conditions: Assessing impacts on the Natura 2000 protected area network. Appl. Geogr. 2016, 75, 83–92. [Google Scholar] [CrossRef]
  21. Xie, Z.; Zhang, B.; Shi, Y.; Zhang, X.; Sun, Z. Changes and protections of urban habitat quality in Shanghai of China. Sci. Rep. 2023, 13, 10976. [Google Scholar] [CrossRef]
  22. Xu, Z.; Dong, B.; Wei, Z.; Lu, Z.; Liu, X.; Xu, H. Study on habitat suitability change and habitat network of rare wintering cranes in important international wetlands. Ecol. Indic. 2023, 154, 110692. [Google Scholar] [CrossRef]
  23. Hu, J.; Zhang, J.; Li, Y. Exploring the spatial and temporal driving mechanisms of landscape patterns on habitat quality in a city undergoing rapid urbanization based on GTWR and MGWR: The case of Nanjing, China. Ecol. Indic. 2022, 143, 109333. [Google Scholar] [CrossRef]
  24. Nematollahi, S.; Fakheran, S.; Kienast, F.; Jafari, A. Application of InVEST habitat quality module in spatially vulnerability assessment of natural habitats (case study: Chaharmahal and Bakhtiari province, Iran). Environ. Monit Assess 2020, 192, 487. [Google Scholar] [CrossRef] [PubMed]
  25. He, J.; Huang, J.; Li, C. The evaluation for the impact of land use change on habitat quality: A joint contribution of cellular automata scenario simulation and habitat quality assessment model. Ecol. Model. 2017, 366, 58–67. [Google Scholar] [CrossRef]
  26. Wang, B.; Oguchi, T.; Liang, X. Evaluating future habitat quality responding to land use change under different city compaction scenarios in Southern China. Cities 2023, 140, 104410. [Google Scholar] [CrossRef]
  27. McKinnon, E.A.; Rotenberg, J.A.; Stutchbury, B.J.M. Seasonal change in tropical habitat quality and body condition for a declining migratory songbird. Oecologia 2015, 179, 363–375. [Google Scholar] [CrossRef]
  28. Both, C.; Van Turnhout, C.A.M.; Bijlsma, R.G.; Siepel, H.; Van Strien, A.J.; Foppen, R.P.B. Avian population consequences of climate change are most severe for long-distance migrants in seasonal habitats. Proc. R. Soc. B Biol. Sci. 2009, 277, 1259–1266. [Google Scholar] [CrossRef]
  29. Papageorgiou, D.; Rozen-Rechels, D.; Nyaguthii, B.; Farine, D.R. Seasonality impacts collective movements in a wild group-living bird. Mov. Ecol. 2021, 9, 38. [Google Scholar] [CrossRef]
  30. Schmitt, T.; Ulrich, W.; Delic, A.; Teucher, M.; Habel, J.C. Seasonality and landscape characteristics impact species community structure and temporal dynamics of East African butterflies. Sci. Rep. 2021, 11, 15103. [Google Scholar] [CrossRef]
  31. Bełcik, M.; Lenda, M.; Amano, T.; Skórka, P. Different response of the taxonomic, phylogenetic and functional diversity of birds to forest fragmentation. Sci. Rep. 2020, 10, 20320. [Google Scholar] [CrossRef]
  32. Leveau, L.M.; Leveau, C.M. Street design in suburban areas and its impact on bird communities: Considering different diversity facets over the year. Urban For. Urban Green. 2020, 48, 126578. [Google Scholar] [CrossRef]
  33. Walsh, J.C.; Gibson, M.R.; Simmonds, J.S.; Mayfield, H.J.; Bracey, C.; Melton, C.B.; Reside, A.E.; Maron, M. Effectiveness of conservation interventions for Australian woodland birds: A systematic review. Biol. Conserv. 2023, 282, 110030. [Google Scholar] [CrossRef]
  34. Kremen, C.; Cameron, A.; Moilanen, A.; Phillips, S.J.; Thomas, C.D.; Beentje, H.; Dransfield, J.; Fisher, B.L.; Glaw, F.; Good, T.C.; et al. Aligning Conservation Priorities Across Taxa in Madagascar with High-Resolution Planning Tools. Science 2008, 320, 222–226. [Google Scholar] [CrossRef] [PubMed]
  35. Fern, R.R.; Morrison, M.L. Mapping critical areas for migratory songbirds using a fusion of remote sensing and distributional modeling techniques. Ecol. Inform. 2017, 42, 55–60. [Google Scholar] [CrossRef]
  36. Zeng, W.; Tang, H.; Liang, X.; Hu, Z.; Yang, Z.; Guan, Q. Using ecological security pattern to identify priority protected areas: A case study in the Wuhan Metropolitan Area, China. Ecol. Indic. 2023, 148, 110121. [Google Scholar] [CrossRef]
  37. Zhang, Y.; Hu, W.; Min, M.; Zhao, K.; Zhang, S.; Liu, T. Optimization of ecological connectivity and construction of supply-demand network in Wuhan Metropolitan Area, China. Ecol. Indic. 2023, 146, 109799. [Google Scholar] [CrossRef]
  38. Lin, J.; He, S.; Liu, X.; Huang, Z.; Li, M.; Chen, B.; Hu, W. Identifying conservation and restoration priorities for degraded coastal wetland vegetations: Integrating species distribution model and GeoDetector. Sci. Total Environ. 2024, 906, 167491. [Google Scholar] [CrossRef]
  39. Liu, Z.; Huang, Q.; Tang, G. Identification of urban flight corridors for migratory birds in the coastal regions of Shenzhen city based on three-dimensional landscapes. Landsc. Ecol. 2021, 36, 2043–2057. [Google Scholar] [CrossRef]
  40. Zhao, H.; Liu, D.; Li, F.; Liu, X.; Niu, J.; He, J.; Liu, Y. Incorporating spatio-temporal connectivity for prioritized conservation of individual habitat patches in a dynamic landscape. Ecol. Indic. 2021, 124, 107414. [Google Scholar] [CrossRef]
  41. Yang, L.; Xu, H.; Pan, S.; Chen, W.; Zeng, J. Identifying the impact of global human activities expansion on natural habitats. J. Clean. Prod. 2024, 434, 140247. [Google Scholar] [CrossRef]
  42. Duan, H.; Yu, X. Linking landscape characteristics to shorebird habitat quality changes in a key stopover site along the East Asian–Australasian Flyway migratory route. Ecol. Indic. 2022, 144, 109490. [Google Scholar] [CrossRef]
  43. Avon, C.; Bergès, L. Prioritization of habitat patches for landscape connectivity conservation differs between least-cost and resistance distances. Landsc. Ecol. 2016, 31, 1551–1565. [Google Scholar] [CrossRef]
  44. Duflot, R.; Avon, C.; Roche, P.; Bergès, L. Combining habitat suitability models and spatial graphs for more effective landscape conservation planning: An applied methodological framework and a species case study. J. Nat. Conserv. 2018, 46, 38–47. [Google Scholar] [CrossRef]
  45. Peng, J.; Pan, Y.; Liu, Y.; Zhao, H.; Wang, Y. Linking ecological degradation risk to identify ecological security patterns in a rapidly urbanizing landscape. Habitat Int. 2018, 71, 110–124. [Google Scholar] [CrossRef]
  46. Hubei Provincial Department of Natural Resources. Main Data Bulletin of the Third National Land Survey of Hubei Province; Hubei Provincial Department of Natural Resources: Wuhan, China, 2021. (In Chinese) [Google Scholar]
  47. Shi, D.; Liao, S.-Y.; Zhu, L.; Li, B.V. Review on bird-building collisions and the case study of a systematic survey in China. Biodiv. Sci. 2022, 30, 21321. [Google Scholar] [CrossRef]
  48. Zhang, Q.; Schaaf, C.; Seto, K.C. The Vegetation Adjusted NTL Urban Index: A new approach to reduce saturation and increase variation in nighttime luminosity. Remote Sens. Environ. 2013, 129, 32–41. [Google Scholar] [CrossRef]
  49. Ma, Q.; He, C.; Wu, J. Behind the rapid expansion of urban impervious surfaces in China: Major influencing factors revealed by a hierarchical multiscale analysis. Land Use Policy 2016, 59, 434–445. [Google Scholar] [CrossRef]
  50. Tian, Y.; Qian, J. Suburban identification based on multi-source data and landscape analysis of its construction land: A case study of Jiangsu Province, China. Habitat Int. 2021, 118, 102459. [Google Scholar] [CrossRef]
  51. Ma, Q.; Wu, J.; He, C. A hierarchical analysis of the relationship between urban impervious surfaces and land surface temperatures: Spatial scale dependence, temporal variations, and bioclimatic modulation. Landsc. Ecol. 2016, 31, 1139–1153. [Google Scholar] [CrossRef]
  52. Jaynes, E.T. Information Theory and Statistical Mechanics. Phys. Rev. 1957, 106, 620–630. [Google Scholar] [CrossRef]
  53. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef]
  54. Phillips, S.J.; Dudík, M. Modeling of species distributions with Maxent: New extensions and a comprehensive evaluation. Ecography 2008, 31, 161–175. [Google Scholar] [CrossRef]
  55. Elith, J.; Graham, C.H.; Valavi, R.; Abegg, M.; Bruce, C.; Ferrier, S.; Ford, A.J.; Guisan, A.; Hijmans, R.J.; Huettmann, F.; et al. Presence-only and Presence-absence Data for Comparing Species Distribution Modeling Methods. Biodivers. Inform. 2020, 15, 69–80. [Google Scholar] [CrossRef]
  56. Valavi, R.; Guillera-Arroita, G.; Lahoz-Monfort, J.J.; Elith, J. Predictive performance of presence-only species distribution models: A benchmark study with reproducible code. Ecol. Monogr. 2021, 92, e1486. [Google Scholar] [CrossRef]
  57. Brown, J.L.; Anderson, B. SDMtoolbox: A python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Methods Ecol. Evol. 2014, 5, 694–700. [Google Scholar] [CrossRef]
  58. Brown, J.L.; Bennett, J.R.; French, C.M. SDMtoolbox 2.0, the next generation Python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. PeerJ 2017, 5, e4095. [Google Scholar] [CrossRef]
  59. Hernandez, P.A.; Graham, C.H.; Master, L.L.; Albert, D.L. The effect of sample size and species characteristics on performance of different species distribution modeling methods. Ecography 2006, 29, 773–785. [Google Scholar] [CrossRef]
  60. Chin, A.; Ruppert, J.; Shrestha, N.; Fortin, M.-J. Urban Avian Conservation Planning Using Species Functional Traits and Habitat Suitability Mapping. Land 2022, 11, 1831. [Google Scholar] [CrossRef]
  61. Dias, R.A.; Marcon, A.P.; Kappes, B.B.; Azpiroz, A.B.; Barbosa, F.G.; Bencke, G.A.; Clay, R.; Di Giacomo, A.S.; Fontana, C.S.; Repenning, M.; et al. A new analytical framework for Maxent species distribution models unveils complex spatiotemporal suitability patterns for two migratory seedeaters (Aves: Sporophila) of conservation concern. Ecol. Inform. 2023, 77, 102189. [Google Scholar] [CrossRef]
  62. Wang, F.; Yuan, X.; Sun, Y.; Liu, Y. Species distribution modeling based on MaxEnt to inform biodiversity conservation in the Central Urban Area of Chongqing Municipality. Ecol. Indic. 2024, 158, 111491. [Google Scholar] [CrossRef]
  63. Elith, J.; Phillips, S.J.; Hastie, T.; Dudík, M.; Chee, Y.E.; Yates, C.J. A statistical explanation of MaxEnt for ecologists. Divers. Distrib. 2011, 17, 43–57. [Google Scholar] [CrossRef]
  64. Merow, C.; Smith, M.J.; Silander, J.A. A practical guide to MaxEnt for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography 2013, 36, 1058–1069. [Google Scholar] [CrossRef]
  65. Elith, J. Quantitative Methods for Modeling Species Habitat: Comparative Performance and an Application to Australian Plants. In Quantitative Methods for Conservation Biology; Ferson, S., Burgman, M., Eds.; Springer: New York, NY, USA, 2000; pp. 39–58. [Google Scholar]
  66. Swets, J.A. Measuring the Accuracy of Diagnostic Systems. Science 1988, 240, 1285–1293. [Google Scholar] [CrossRef]
  67. Sharp, R.; Chaplin-Kramer, R.; Wood, S.; Guerry, A.; Tallis, H.; Ricketts, T.; Nelson, E.; Ennaanay, D.; Wolny, S.; Olwero, N.; et al. InVEST User’s Guide; The Natural Capital Project, Stanford University, University of Minnesota, The Nature Conservancy, and World Wildlife Fund: Gland, Switzerland, 2018. [Google Scholar]
  68. Arkema, K.K.; Verutes, G.; Bernhardt, J.R.; Clarke, C.; Rosado, S.; Canto, M.; Wood, S.A.; Ruckelshaus, M.; Rosenthal, A.; McField, M.; et al. Assessing habitat risk from human activities to inform coastal and marine spatial planning: A demonstration in Belize. Environ. Res. Lett. 2014, 9, 114016. [Google Scholar] [CrossRef]
  69. Duggan, J.M.; Eichelberger, B.A.; Ma, S.; Lawler, J.J.; Ziv, G. Informing management of rare species with an approach combining scenario modeling and spatially explicit risk assessment. Ecosyst. Health Sustain. 2015, 1, 1–18. [Google Scholar] [CrossRef]
  70. Hyrenbach, D.; Wyatt, K.H.; Griffin, R.; Guerry, A.D.; Ruckelshaus, M.; Fogarty, M.; Arkema, K.K. Habitat risk assessment for regional ocean planning in the U.S. Northeast and Mid-Atlantic. PLoS ONE 2017, 12, e0188776. [Google Scholar]
  71. Niz, W.C.; Laurino, I.R.A.; de Freitas, D.M.; Rolim, F.A.; Motta, F.S.; Pereira-Filho, G.H. Modeling risks in marine protected areas: Mapping of habitats, biodiversity, and cultural ecosystem services in the southernmost atlantic coral reef. J. Environ. Manag. 2023, 345, 118855. [Google Scholar] [CrossRef]
  72. Samhouri, J.F.; Levin, P.S. Linking land- and sea-based activities to risk in coastal ecosystems. Biol. Conserv. 2012, 145, 118–129. [Google Scholar] [CrossRef]
  73. Caro, C.; Marques, J.C.; Cunha, P.P.; Teixeira, Z. Ecosystem services as a resilience descriptor in habitat risk assessment using the InVEST model. Ecol. Indic. 2020, 115, 106426. [Google Scholar] [CrossRef]
  74. Bastos, M.I.; Roebeling, P.C.; Alves, F.L.; Villasante, S.; Magalhães Filho, L. High risk water pollution hazards affecting Aveiro coastal lagoon (Portugal)–A habitat risk assessment using InVEST. Ecol. Inform. 2023, 76, 102144. [Google Scholar] [CrossRef]
  75. GB3838-2002; State Environmental Protection Administration. Environmental Quality Standards for Surface Water. China Environmental Science Press: Beijing, China, 2002.
  76. Li, J.; Wang, Z.; Lai, C.; Wu, X.; Zeng, Z.; Chen, X.; Lian, Y. Response of net primary production to land use and land cover change in mainland China since the late 1980s. Sci. Total Environ. 2018, 639, 237–247. [Google Scholar] [CrossRef] [PubMed]
  77. Chen, S.; Jiang, H.; Chen, Y.; Cai, Z. Spatial-temporal patterns of net primary production in Anji (China) between 1984 and 2014. Ecol. Indic. 2020, 110, 105954. [Google Scholar] [CrossRef]
  78. Zhai, T.; Wang, J.; Fang, Y.; Qin, Y.; Huang, L.; Chen, Y. Assessing ecological risks caused by human activities in rapid urbanization coastal areas: Towards an integrated approach to determining key areas of terrestrial-oceanic ecosystems preservation and restoration. Sci. Total Environ. 2020, 708, 135153. [Google Scholar] [CrossRef] [PubMed]
  79. Studwell, A.; Hines, E.; Nur, N.; Jahncke, J. Using habitat risk assessment to assess disturbance from maritime activities to inform seabird conservation in a coastal marine ecosystem. Ocean Coast. Manag. 2021, 199, 105431. [Google Scholar] [CrossRef]
  80. De Kort, H.; Baguette, M.; Lenoir, J.; Stevens, V.M. Toward reliable habitat suitability and accessibility models in an era of multiple environmental stressors. Ecol. Evol. 2020, 10, 10937–10952. [Google Scholar] [CrossRef]
  81. Hartigan, J.A.; Wong, M.A. Algorithm AS 136, A K-Means Clustering Algorithm. J. R. Stat. Soc. Ser. C (Appl. Stat.) 1979, 28, 100–108. [Google Scholar] [CrossRef]
  82. Jigu, S.; Jie, L.; Lianyu, Z. Clustering Algorithms Research. J. Softw. 2008, 19, 48–61. [Google Scholar]
  83. Jain, A.K. Data clustering: 50 years beyond K-means. Pattern Recognit. Lett. 2010, 31, 651–666. [Google Scholar] [CrossRef]
  84. Dai, J.; Fang, J.; Guo, Z.; Hou, J. Research on ecological restoration assessment and eco-economic development of sea area by introducing the K-means clustering algorithm. Environ. Sci. Pollut. Res. Int. 2023, 30, 118318–118331. [Google Scholar] [CrossRef]
  85. Ikotun, A.M.; Ezugwu, A.E.; Abualigah, L.; Abuhaija, B.; Heming, J. K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data. Inf. Sci. 2023, 622, 178–210. [Google Scholar] [CrossRef]
  86. Thorndike, R.L. Who belongs in the family? Psychometrika 1953, 18, 267–276. [Google Scholar] [CrossRef]
  87. Syakur, M.A.; Khotimah, B.K.; Rochman, E.M.S.; Satoto, B.D. Integration K-Means Clustering Method and Elbow Method For Identification of The Best Customer Profile Cluster. IOP Conf. Ser. Mater. Sci. Eng. 2018, 336, 012017. [Google Scholar] [CrossRef]
  88. Jenkins, C.N.; Van Houtan, K.S.; Pimm, S.L.; Sexton, J.O. US protected lands mismatch biodiversity priorities. Proc. Natl. Acad. Sci. USA 2015, 112, 5081–5086. [Google Scholar] [CrossRef]
  89. Ghoddousi, A.; Loos, J.; Kuemmerle, T. An Outcome-Oriented, Social–Ecological Framework for Assessing Protected Area Effectiveness. BioScience 2021, 72, 201–212. [Google Scholar] [CrossRef] [PubMed]
  90. Fan, F.; Wen, X.; Feng, Z.; Gao, Y.; Li, W. Optimizing urban ecological space based on the scenario of ecological security patterns: The case of central Wuhan, China. Appl. Geogr. 2022, 138, 102619. [Google Scholar] [CrossRef]
  91. Jokimäki, J.; Suhonen, J. Distribution and habitat selection of wintering birds in urban environments. Landsc. Urban Plan. 1998, 39, 253–263. [Google Scholar] [CrossRef]
  92. McKinney, M.L. Urbanization, biodiversity, and conservation. BioScience 2002, 52, 883–890. [Google Scholar] [CrossRef]
  93. Duan, H.; Yu, X.; Shan, K.; Zhang, C.; Liu, H. Effects of habitat loss on migratory shorebird community structure at stopover sites: A case study in the Yellow River Delta, China. Front. Mar. Sci. 2022, 9, 1049765. [Google Scholar] [CrossRef]
  94. Yan, Z.; Teng, M.; He, W.; Wang, Y.; Yang, J.; Wang, P. Improving conservation effectiveness of nature reserve for golden snub-nosed monkey, a niche-based approach. Ecol. Evol. 2018, 8, 9315–9325. [Google Scholar] [CrossRef]
  95. Wuhan Water Resources Bulletin. Wuhan Water Affairs Bureau. 2020. Available online: https://swj.wuhan.gov.cn/szy/202104/t20210413_1666703.html (accessed on 15 July 2024).
  96. Wang, M.; Yao, J.; Chang, H.; Liu, R.; Xu, N.; Liu, Z.; Gong, H.; Zheng, H.; Wang, J.; Guo, X.; et al. Underground well water level observation grid dataset from 2005 to 2022. Sci. Data 2025, 12, 728. [Google Scholar] [CrossRef]
  97. Terrado, M.; Sabater, S.; Chaplin-Kramer, B.; Mandle, L.; Ziv, G.; Acuna, V. Model development for the assessment of terrestrial and aquatic habitat quality in conservation planning. Sci. Total Environ. 2016, 540, 63–70. [Google Scholar] [CrossRef] [PubMed]
  98. Li, Z. Bird Diversity and Interannual, Seasonal Variation Dynamics in Wuhan City. Hubei For. Sci. Technol. 2021, 50, 39–42,88. (In Chinese) [Google Scholar]
  99. Hu, H.; Tang, R.; Tang, Z. Studies on the Birds of Wuhan (II): Seasonal Variation in Species Composition and Population Numbers. J. Wuhan Univ. (Nat. Sci. Ed.) 1981, 4, 94–106. (In Chinese) [Google Scholar]
  100. Yuan, L.; Liu, D.; Tian, B.; Yuan, X.; Bo, S.; Ma, Q.; Wu, W.; Zhao, Z.; Zhang, L.; Keesing, J.K. A solution for restoration of critical wetlands and waterbird habitats in coastal deltaic systems. J. Environ. Manag. 2022, 302, 113996. [Google Scholar] [CrossRef]
  101. Moores, N.; Jung, H.; Kim, H.-J.; Hwang, B.-Y.; Hur, W.-H.; Borzée, A. The Hwaseong Wetlands Reclamation Area and Tidal Flats, Republic of Korea: A Case of Waterbird Conservation in the Yellow Sea. Conservation 2022, 2, 526–549. [Google Scholar] [CrossRef]
  102. Lee, J.-H.; Kim, I.-C.; Lee, S.-W.; Son, J.-J.; Jang, J.-U.; Sung, H.-C. International importance of tidal flats in the Republic of Korea as shorebird stopover sites in the East Asian–Australasian flyway. Avian Conserv. Ecol. 2023, 18, 23. [Google Scholar] [CrossRef]
  103. Shi, X.; Hu, C.; Soderholm, J.; Chapman, J.; Mao, H.; Cui, K.; Ma, Z.; Wu, D.; Fuller, R.A.; Lecours, V.; et al. Prospects for monitoring bird migration along the East Asian-Australasian Flyway using weather radar. Remote Sens. Ecol. Conserv. 2022, 9, 169–181. [Google Scholar] [CrossRef]
  104. Xia, S.; Yu, X.; Lei, J.; Hearn, R.; Smith, B.; Lei, G.; Xie, P. Priority sites and conservation gaps of wintering waterbirds in the Yangtze River floodplain. J. Geogr. Sci. 2021, 30, 1617–1632. [Google Scholar] [CrossRef]
  105. Matutini, F.; Baudry, J.; Fortin, M.-J.; Pain, G.; Pithon, J. Conservation networks do not match the ecological requirements of amphibians. Peer Community J. 2023, 3, e58. [Google Scholar] [CrossRef]
  106. Ye, Y.; Lin, W.; Song, J.; Wei, Q.; Zheng, Y. Spatial mismatches and multi-dimensional conservation priorities for urban birds in China. J. Environ. Manag. 2025, 379, 124905. [Google Scholar] [CrossRef]
  107. Sun, W.; Zhao, Y.; Chen, W.; Bai, Y. Current national nature reserves are insufficient to safeguard the long-term survival of birds and mammals in China. Commun. Earth Environ. 2024, 5, 304. [Google Scholar] [CrossRef]
  108. Bai, Y.; Wong, C.P.; Jiang, B.; Hughes, A.C.; Wang, M.; Wang, Q. Developing China’s Ecological Redline Policy using ecosystem services assessments for land use planning. Nat. Commun. 2018, 9, 3034. [Google Scholar] [CrossRef] [PubMed]
  109. Xu, X.; Yang, G.; Tan, Y. Identifying ecological red lines in China’s Yangtze River Economic Belt: A regional approach. Ecol. Indic. 2019, 96, 635–646. [Google Scholar] [CrossRef]
  110. Wei, J.; Tian, Y.; Li, C.; Yuan, H.; Liu, Y. The Coordinative Evaluation of Suburban Construction Land from Spatial, Socio-Economic, and Ecological Dimensions: A Case Study of Suburban Wuhan, Central China. Land 2025, 14, 900. [Google Scholar] [CrossRef]
  111. Hua, Y.; Cui, B.; He, W.; Cai, Y. Identifying potential restoration areas of freshwater wetlands in a river delta. Ecol. Indic. 2016, 71, 438–448. [Google Scholar] [CrossRef]
  112. Liang, Q.; Zhai, J.; Li, C. From separation to incorporation: Development of a unifying framework that integrated bird habitats with public recreation spaces within the wetland park system. J. Clean. Prod. 2023, 430, 139647. [Google Scholar] [CrossRef]
  113. Wei, J.; Zhang, Y.; Liu, Y.; Li, C.; Tian, Y.; Qian, J.; Gao, Y.; Hong, Y.; Liu, Y. The impact of different road grades on ecological networks in a mega-city Wuhan City, China. Ecol. Indic. 2022, 137, 108784. [Google Scholar] [CrossRef]
  114. Xie, S.; Wang, X.; Ren, Y.; Su, Z.; Su, Y.; Wang, S.; Zhou, W.; Lu, F.; Qian, Y.; Gong, C.; et al. Factors responsible for forest and water bird distributions in rivers and lakes along an urban gradient in Beijing. Sci. Total Environ. 2020, 735, 139308. [Google Scholar] [CrossRef]
  115. González-Sargas, E.; Meehan, T.D.; Hinojosa-Huerta, O.; Villagomez-Palma, S.; Calvo-Fonseca, A.; Dodge, C.; Gómez-Sapiens, M.; Shafroth, P.B. Bird community response to one decade of riparian restoration along the Colorado River delta in Mexico. Ecol. Eng. 2024, 205, 107291. [Google Scholar] [CrossRef]
  116. Sun, X.; Shen, J.; Xiao, Y.; Li, S.; Cao, M. Habitat suitability and potential biological corridors for waterbirds in Yancheng coastal wetland of China. Ecol. Indic. 2023, 148, 110090. [Google Scholar] [CrossRef]
  117. Tarjuelo, R.; Benítez-López, A.; Casas, F.; Martín, C.A.; García, J.T.; Viñuela, J.; Mougeot, F. Living in seasonally dynamic farmland: The role of natural and semi-natural habitats in the movements and habitat selection of a declining bird. Biol. Conserv. 2020, 251, 108794. [Google Scholar] [CrossRef]
  118. U.S. Fish and Wildlife Service. Urban Bird Treaty Program Guidebook V.3, Making Cities Healthier Places for Birds and People; U.S. Fish and Wildlife Service: Hadley, MA, USA, 2020.
  119. Lin, H.-Y.; Schuster, R.; Wilson, S.; Cooke, S.J.; Rodewald, A.D.; Bennett, J.R. Integrating season-specific needs of migratory and resident birds in conservation planning. Biol. Conserv. 2020, 252, 108826. [Google Scholar] [CrossRef]
  120. Soares, R.M.V.; Lira, P.K.; Manes, S.; Vale, M.M. A methodological framework for prioritizing habitat patches in urban ecosystems based on landscape functional connectivity. Urban Ecosyst. 2023, 27, 147–157. [Google Scholar] [CrossRef]
  121. Convention on Biological Diversity (CBD). Decision Adopted by the Conference of the Parties to the Convention on Biological Diversity: 14/8. In Protected Areas and Other Effective Area-Based Conservation Measures; CBD Secretariat: Sharm El-Sheikh, Egypt, 2018. [Google Scholar]
  122. Cook, C.N. Progress developing the concept of other effective area-based conservation measures. Conserv. Biol. 2024, 38, e14106. [Google Scholar] [CrossRef]
  123. Bleyhl, B.; Baumann, M.; Griffiths, P.; Heidelberg, A.; Manvelyan, K.; Radeloff, V.C.; Zazanashvili, N.; Kuemmerle, T. Assessing landscape connectivity for large mammals in the Caucasus using Landsat 8 seasonal image composites. Remote Sens. Environ. 2017, 193, 193–203. [Google Scholar] [CrossRef]
  124. He, J.; Huang, J.; Liu, D.; Wang, H.; Li, C. Updating the habitat conservation institution by prioritizing important connectivity and resilience providers outside. Ecol. Indic. 2018, 88, 219–231. [Google Scholar] [CrossRef]
  125. Huang, J.; Andrello, M.; Martensen, A.C.; Saura, S.; Liu, D.; He, J.; Fortin, M.J. Importance of spatio–temporal connectivity to maintain species experiencing range shifts. Ecography 2020, 43, 591–603. [Google Scholar] [CrossRef]
  126. Churko, G.; Kienast, F.; Bolliger, J. A Multispecies Assessment to Identify the Functional Connectivity of Amphibians in a Human-Dominated Landscape. ISPRS Int. J. Geo-Inform. 2020, 9, 287. [Google Scholar] [CrossRef]
  127. Donaldson, L.; Bennie, J.J.; Wilson, R.J.; Maclean, I.M.D. Designing effective protected area networks for multiple species. Biol. Conserv. 2021, 258, 109125. [Google Scholar] [CrossRef]
  128. Bellard, C.; Bertelsmeier, C.; Leadley, P.; Thuiller, W.; Courchamp, F. Impacts of climate change on the future of biodiversity. Ecol. Lett. 2012, 15, 365–377. [Google Scholar] [CrossRef]
Figure 1. The study area in Wuhan, Hubei, China.
Figure 1. The study area in Wuhan, Hubei, China.
Sustainability 17 06078 g001
Figure 2. The framework for identifying priority bird habitats through an integrated habitat suitability–risk–quality assessment.
Figure 2. The framework for identifying priority bird habitats through an integrated habitat suitability–risk–quality assessment.
Sustainability 17 06078 g002
Figure 3. Spatial distribution of bird habitat suitability levels in Wuhan. (ad) Seasonal distributions of habitat suitability levels. (eh) Detailed maps of areas with significant seasonal variation in habitat suitability.
Figure 3. Spatial distribution of bird habitat suitability levels in Wuhan. (ad) Seasonal distributions of habitat suitability levels. (eh) Detailed maps of areas with significant seasonal variation in habitat suitability.
Sustainability 17 06078 g003
Figure 4. Bird habitat risk values and spatial distribution of habitat risk levels in Wuhan. (a) Heatmaps of habitat risk values at the city level. (b) Heatmaps of habitat risk values at the district level. (c) Spatial distribution of habitat risk levels. Note: MT refers to multiple threats.
Figure 4. Bird habitat risk values and spatial distribution of habitat risk levels in Wuhan. (a) Heatmaps of habitat risk values at the city level. (b) Heatmaps of habitat risk values at the district level. (c) Spatial distribution of habitat risk levels. Note: MT refers to multiple threats.
Sustainability 17 06078 g004
Figure 5. Bird habitat quality maps for each season in Wuhan. (ad) Seasonal habitat quality distributions. (eh) Detailed maps of areas with significant seasonal variation in habitat quality.
Figure 5. Bird habitat quality maps for each season in Wuhan. (ad) Seasonal habitat quality distributions. (eh) Detailed maps of areas with significant seasonal variation in habitat quality.
Sustainability 17 06078 g005
Figure 6. The four clusters of bird habitats identified using K-means based on seasonal quality. (a) Spatial distribution of four clusters. (b) Seasonal quality characteristics of each cluster, with longer segments indicating higher quality.
Figure 6. The four clusters of bird habitats identified using K-means based on seasonal quality. (a) Spatial distribution of four clusters. (b) Seasonal quality characteristics of each cluster, with longer segments indicating higher quality.
Sustainability 17 06078 g006
Table 1. Data sources and contents.
Table 1. Data sources and contents.
Data NameTypeContentFormatYearSources
Bird Observation DatasetSeasonalSpecies name, taxonomy, time, coordinates, and count.txt2016–2020Global Biodiversity Information Facility (GBIF), https://www.gbif.org/ (accessed on 11 September 2024).
Land Use/Land Cover
(LULC)
AnnualLand-use type: cropland, forest, grassland, wetlands, urban land, and bare land.tif2020GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery, https://data.casearth.cn/ (accessed on 8 June 2024).
Digital Elevation Model (DEM)AnnualSlope, aspect, and elevation.hdf NASA Shuttle Radar Topography Mission Global 1 arc second V003 (SRTMGL1) datasets with a spatial resolution of 30 m, https://www.earthdata.nasa.gov/data/catalog/lpcloud-srtmgl1-003 (accessed on 10 June 2024).
Normalized Difference Vegetation Index (NDVI)SeasonalSurface vegetation coverage; seasonal NDVI derived using MVC method.hdf2020MODIS/Terra Vegetation Indices 16-Day L3 Global 250 m SIN Grid V061, https://www.earthdata.nasa.gov/data/catalog/lpcloud-mod13q1-061 (accessed on 15 July 2024).
Net Primary Productivity
(NPP)
AnnualAnnual Gross and Net Primary Production.hdf2020MODIS/Terra Net Primary Production Gap-Filled Yearly L4 Global 500 m SIN Grid V061, https://www.earthdata.nasa.gov/data/catalog/lpcloud-mod17a3hgf-061 (accessed on 15 July 2024).
TemperatureSeasonalMonthly temperature data, aggregated to seasonal values using AVC method.nc20201 km monthly temperature dataset for China,
National Earth System Science Data Center, National Science and Technology Infrastructure of China, http://www.geodata.cn (accessed on 18 July 2024).
PrecipitationSeasonalMonthly precipitation data, aggregated to seasonal values using AVC method.nc20201 km monthly precipitation dataset for China,
National Earth System Science Data Center, National Science and Technology Infrastructure of China, http://www.geodata.cn (accessed on 18 July 2024).
RoadsAnnualRailways, highways, and urban expressways classified as high-level roads; primary and secondary roads classified as low-level roads.osm2020OpenStreetMap, https://www.openstreetmap.org (accessed on 15 July 2024).
Nighttime LightSeasonalHuman activity range; seasonal nighttime light extracted using MVC; urban boundaries identified using the VANUI index, integrating NDVI and nighttime light data.tif2020500 m Nighttime Light map by EOG, https://eogdata.mines.edu/products/vnl/ (accessed on 1 September 2024).
GDPAnnualGross domestic product for a 1 km grid, in million yuan per square kilometer, corrected using district-level GDP data from the Wuhan Statistical Yearbook.tif2020China GDP Spatial Distribution Kilometer Grid Dataset, https://www.resdc.cn (accessed on 1 September 2024).
Population DensityAnnualPopulation density per grid cell, in the number of people per square kilometer, corrected using district-level population data from the Wuhan Statistical Yearbook.tif2020Global High Resolution Population Denominators Project, https://dx.doi.org/10.5258/SOTON/WP00675 (accessed on 10 September 2024).
Building HeightAnnualUnits in meters; classified into four categories based on avian collision risk studies.tif2020Urban building height data with a spatial resolution of 1 m, The Department of Natural Resources of Hubei Province
Nature ReserveAnnualLocation, size, and time of approval, defined as inaccessible areas.shp2020The Department of Natural Resources of Hubei Province
Table 2. Environmental variables considered for modeling seasonal habitat suitability.
Table 2. Environmental variables considered for modeling seasonal habitat suitability.
Variable CategoryEnvironmental Variable (unit)Abbr.
TopographyElevation (m)DEM
Slope (°)Slope
Aspect (°)Aspect
Land coverLand-use typeLUCC
ClimateTemperature (℃)TEMP
Precipitation (mm)PRE
VegetationNormalized Difference Vegetation IndexNDVI
Net Primary Production (kgC/m2/a)NPP
Distance effectDistance to wetland (m)DtoW
Distance to forest (m)DtoF
Distance to bare land (m)DtoB
Distance to urban areas (m)DtoU
Distance to cropland (m)DtoC
Distance to high-level roads (m)DtoHR
Distance to low-level roads (m)DtoLR
Distance to buildings (m)DtoBL
Table 3. Scoring criteria for exposure and consequence dimensions.
Table 3. Scoring criteria for exposure and consequence dimensions.
CriteriaLow (1)Medium (2)High (3)Reference(s)
ExposureSpatial Overlap Area Rating<10%10–30%≥30%Samhouri [68,72]
Temporal Overlap Rating0–4 months per year4–8 months per year8–12 months per year[67,73]
Intensity RatingLow intensityMedium intensityHigh intensity[67,73]
Consequence—SensitivityChange in AreaLow loss (<20%)Medium loss (20–50%)High loss (>50%)[67,73]
Consequence—ResilienceConnectivityHigh (>100 km)Medium (10–100 km)Low (<10 km)[67,73]
Water Quality>70% of the area with a water quality of better than Class III50–60% of the area with a water quality of better than Class III<50% of the area with a water quality of better than Class III[75]
NPP (kgC/m2/a)High (>0.5)Moderate (0.2–0.5)Low (<0.2)[76,77]
Table 4. Habitat suitability of land-use types.
Table 4. Habitat suitability of land-use types.
Land-Use TypeHabitat Suitability Classification from MaxEntAdjusted Habitat Suitability in HQ Model
Wetland/Forest/Grassland<0.0490.5
0.049–0.1260.6
0.126–0.2260.7
0.226–0.3540.8
0.354–0.5160.9
≥0.5161
Cropland/Urban/Bare Land00
Table 5. Habitat sensitivity to threats.
Table 5. Habitat sensitivity to threats.
Land-Use TypeCroplandUrbanBare LandHigh-Level RoadsLow-Level RoadsBuildings
Wetland0.79700.87370.63000.75000.60000.8500
Forest0.63640.91890.56000.80000.70000.8500
Grassland0.57810.56630.49070.60000.40000.8500
Cropland000000
Urban000000
Bare Land000000
Table 6. The maximum impact distances and weights of the threats.
Table 6. The maximum impact distances and weights of the threats.
ThreatMaximum Impact Distance (km)WeightDistance-Decay Function
Cropland0.50.3Linear
Urban20.9Exponential
Bare Land0.50.1Linear
High-level Roads11Linear
Low-level Roads0.80.6Linear
Buildings10.9Exponential
Table 7. Seasonal means and standard deviations of habitat suitability.
Table 7. Seasonal means and standard deviations of habitat suitability.
Habitat TypeSpringSummerAutumnWinter
MeanSDMeanSDMeanSDMeanSD
All habitats0.0745 0.10 0.0504 0.08 0.1393 0.16 0.1123 0.11
Wetland 0.0775 0.10 0.0561 0.08 0.1854 0.16 0.1405 0.12
Forest0.0665 0.10 0.0385 0.07 0.0444 0.09 0.0542 0.07
Grassland0.0939 0.11 0.0453 0.07 0.0922 0.12 0.0855 0.09
Note: The first row represents the overall habitat values; the following rows correspond to specific habitat types (wetland, forest, and grassland).
Table 8. Seasonal means and standard deviations of habitat quality.
Table 8. Seasonal means and standard deviations of habitat quality.
Habitat TypeSpringSummerAutumnWinter
MeanSDMeanSDMeanSDMeanSD
All habitats0.5418 0.09 0.5208 0.08 0.6063 0.12 0.5938 0.12
Wetland 0.5400 0.09 0.5224 0.08 0.6475 0.12 0.6228 0.13
Forest0.5443 0.09 0.5180 0.07 0.5223 0.07 0.5345 0.08
Grassland0.5598 0.10 0.5108 0.07 0.5524 0.09 0.5590 0.10
Note: The first row represents the overall habitat values; the following rows correspond to specific habitat types (wetland, forest, and grassland).
Table 9. Protection status of habitat clusters and results of the chi-squared test.
Table 9. Protection status of habitat clusters and results of the chi-squared test.
ClusterProtected Area (km2)Unprotected Area (km2)Total Area (km2)Protection Rate (%)
Cluster 146.22 53.16 99.38 46.51
Cluster 2187.47 655.68 843.15 22.23
Cluster 3179.44 208.16 387.60 46.30
Cluster 470.21 152.68 222.89 31.50
Total483.341069.681553.0231.22
Note: Chi-squared test comparing protection status across clusters yielded χ2 = 64.72, df = 3, p < 0.001.
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

Wei, J.; Tian, Y.; Li, C.; Zhang, Y.; Yuan, H.; Liu, Y. Identifying Priority Bird Habitats Through Seasonal Dynamics: An Integrated Habitat Suitability–Risk–Quality Framework. Sustainability 2025, 17, 6078. https://doi.org/10.3390/su17136078

AMA Style

Wei J, Tian Y, Li C, Zhang Y, Yuan H, Liu Y. Identifying Priority Bird Habitats Through Seasonal Dynamics: An Integrated Habitat Suitability–Risk–Quality Framework. Sustainability. 2025; 17(13):6078. https://doi.org/10.3390/su17136078

Chicago/Turabian Style

Wei, Junqing, Yasi Tian, Chun Li, Yan Zhang, Hongzhou Yuan, and Yanfang Liu. 2025. "Identifying Priority Bird Habitats Through Seasonal Dynamics: An Integrated Habitat Suitability–Risk–Quality Framework" Sustainability 17, no. 13: 6078. https://doi.org/10.3390/su17136078

APA Style

Wei, J., Tian, Y., Li, C., Zhang, Y., Yuan, H., & Liu, Y. (2025). Identifying Priority Bird Habitats Through Seasonal Dynamics: An Integrated Habitat Suitability–Risk–Quality Framework. Sustainability, 17(13), 6078. https://doi.org/10.3390/su17136078

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