Impact of Habitat Quality Changes on Regional Thermal Environment: A Case Study in Anhui Province, China

Abstract

Biodiversity degradation and loss represent critical global challenges, primarily driven by the urban heat island effect, which results from elevated surface temperatures. As urbanization and climate change continue to progress, these phenomena have a profound impact on both habitats and human residential environments. This study focuses on Anhui Province as a case study to systematically investigate the effects of changes in habitat quality (HQ) on the evolution of the regional thermal environment. The objective is to provide a scientific basis for addressing regional thermal environment issues and promoting biodiversity conservation. This paper employs the InVEST-HQ model to analyze HQ in Anhui Province from 2000 to 2020 and integrates surface temperature data to assess the response of HQ changes to variations in the regional thermal environment. The results show that: (1) From 2000 to 2020, the HQ index in Anhui Province exhibited a general decline, characterized by pronounced spatial heterogeneity, with lower values observed in the northern regions and higher values in the southwestern and southern areas. (2) Concurrently, the relative surface temperature in Anhui Province continued to rise, particularly in central urban areas such as Hefei, where the increase in impermeable surfaces has facilitated the expansion of high-temperature zones. (3) Different types of HQ had distinctly varying effects on regional thermal environments: habitats classified as poor HQ or worse HQ were associated with noticeable warming effects, while those categorized as good HQ or excellent HQ exhibited significant cooling effects. (4) The contribution index of varying HQ to relative surface temperature ranged from −0.2 to 0.3, indicating that poor HQ and worse HQ positively contributed to regional thermal environments, whereas good HQ and excellent HQ exerted a negative contribution. City-level analyses revealed that cities such as Suzhou, Chizhou, Wuhu, Anqing, Xuancheng, and Lu’an were associated with positive contributions to relative surface temperature, while cities including Bengbu, Fuyang, Chuzhou, Huaibei, Tongling, Ma’anshan, and Hefei demonstrated negative contributions. This study provides valuable insights for optimizing the spatial distribution of urban cold islands and promoting ecological sustainable development.

1. Introduction

Habitat quality (HQ) is increasingly recognized as a critical component of ecological civilization and plays a significant role in socioeconomic development. As an indicator of ecosystem health, HQ reflects, to a certain extent, the abundance of biodiversity and the integrity of ecosystem services [1,2]. Within the context of socioeconomic progress, HQ represents a vital point of interaction between humans and the environment [3,4]. Urban biodiversity is a driving force for sustainable urban development, contributing significantly to the establishment of stable and sustainable living environments. Since the onset of the reform and opening-up period, China has experienced rapid socioeconomic development and urbanization, resulting in the extensive encroachment of impermeable surfaces upon ecological land. This encroachment has altered the structure and functions of ecosystems, leading to ecological imbalances that adversely affect human well-being [5,6]. Key challenges during this urban development process include biodiversity loss, habitat degradation, and the urban heat island (UHI) effect, which have emerged as pressing issues [7]. Specifically, the UHI effect not only exacerbates air pollution [8] and affects species growth and development [9] but also increases greenhouse gas emissions [10]. These changes can lead to significant alterations in urban species richness, distribution patterns, and energy flow, potentially culminating in the extinction of certain species [11,12,13]. Consequently, addressing the inherent “contradiction” between urban development and ecological sustainability has become a focal point of scholarly inquiry [14].

As a fundamental ecological function within ecosystem services, HQ not only reflects regional biodiversity and human living conditions but also plays a crucial role in ensuring regional ecological security and enhancing human well-being [15,16]. The assessment of HQ has become a key determinant in evaluating regional ecological health and sustainability [17]. Currently, a variety of methods and models for assessing HQ are continually being developed. Notable examples include ARIES (Artificial Intelligence for Ecosystem Services) [18], the biodiversity assessment module in IDRISI, the Habitat Suitability Index (HSI) [19], the Maxent model [20], the SoLVES model [21], the Vitality-Organization-Resilience (VOR) model [22], and the InVEST model [23]. Among these, the InVEST model is particularly noteworthy due to its capacity to account for spatial heterogeneity and temporal variability, leading to its widespread application [24]. It has been effectively utilized in HQ studies across various scales, including national [25], provincial [26], municipal [27], and county levels [28]. Furthermore, numerous investigations have focused on existing land use, landscape patterns, and urban expansion.

The increasing frequency of extreme weather events has led to intensified phenomena such as summer heat waves, with urban heat events becoming more prevalent worldwide [29]. Land Surface Temperature (LST) serves as an indicator of UHI intensity, directly influencing the energy balance of the Earth’s land surface system and functioning as a crucial parameter in geophysical cycles [30,31]. In recent years, the rapid pace of global urbanization has significantly expanded the area of impervious surfaces, resulting in the accumulation of urban heat and a slower release of this heat. This trend has diminished the evaporation and transpiration of surface water, thereby weakening the latent heat effect in urban areas and exacerbating the UHI phenomenon [32]. Additionally, as urbanization progresses, natural landscapes are increasingly being replaced by artificial surfaces, leading to substantial changes in urban LST [33,34]. China, as the country experiencing the fastest urbanization growth globally, has seen its urbanization rate rise dramatically from 18% to 60% between 1978 and 2018 [35]. This trend is expected to continue over the next decade, suggesting further expansion of urban construction activities. Consequently, the distinctive characteristics of urban LST are likely to become even more pronounced. Changes in urban surface temperature not only adversely impact human well-being and health [36], but the ongoing rise in LST may also precipitate uncontrollable natural disasters [37]. The “14th Five-Year Plan” of China explicitly highlights the promotion of ecological civilization construction [38], underscoring the importance of LST in addressing urban ecological and environmental challenges.

The process of urbanization not only exacerbates the UHI effect but also significantly deteriorates HQ. Urban expansion encroaches upon ecological lands—such as forest, grassland, and aquatic ecosystems—thereby reducing urban green spaces, disrupting the urban cool island effect, and contributing to biodiversity loss and habitat degradation. Research on the interplay between HQ and LST remains inadequate. Anticipated rapid changes in population structure are likely to further accelerate urban expansion, thereby heightening the risk of ecological degradation. The decline of HQ due to urbanization has emerged as a critical concern for global urban ecosystems, with climate change expected to exacerbate this trend in the coming decades. Additionally, conflicts between human activities and resource utilization are anticipated to escalate [39]. Therefore, investigating the spatiotemporal variations in LST across diverse HQ conditions is essential for effectively optimizing and managing the quality of human living environments.

Anhui Province is characterized by a unique geographical location that exhibits significant variations in topography and climate among its southern (Wan South), central (Wan Central), and northern (Wan North) regions, effectively illustrating the diversity of northern and southern China. The province benefits from favorable hydrothermal conditions and rich ecosystem services, contributing to high species richness and playing a critical role in the ecological security of the Yangtze River Economic Belt. This ecological significance positions Anhui Province as a vital area within the national landscape. Furthermore, Anhui is strategically situated as an important link among several major economic zones in the country. However, in recent years, the acceleration of urbanization has led to a substantial increase in urban land area, with sprawling spatial expansion and extensive infrastructure development exerting considerable pressure on the ecological environment.

In the context of rapid urbanization and industrialization, habitats and biodiversity are increasingly under threat, and the UHI effect is becoming more pronounced. Consequently, it is imperative to understand the relationship between HQ and the urban thermal environment. This study focuses on Anhui Province and examines how changes in HQ influence the spatiotemporal evolution of the province’s thermal environment. The primary research objectives are as follows: (1) to assess HQ in Anhui Province utilizing the InVEST model, elucidating the spatiotemporal evolution of HQ at five distinct time points between 2000 and 2020; (2) to categorize LST using the mean–standard deviation method and analyze the spatiotemporal evolution of diurnal and nocturnal temperatures at these five time points; (3) to apply the Getis-Ord Gi* index to identify the spatial clustering characteristics of LST; and (4) to introduce a contribution index to determine the impact of varying levels of HQ and different urban regions on the thermal environment. This research aims to explore the spatial dynamics and changing patterns between HQ and LST, thereby providing a scientific foundation for future urban green space planning and contributing to the advancement of urban ecological civilization.

2. Study Area and Data Sources

2.1. Study Area

Anhui Province is located in the eastern part of China, defined by geographical coordinates from 114°54′ to 119°37′ E and from 29°41′ to 34°38′ N, encompassing a total area of 140,100 square kilometers. The province exhibits a terrain characterized by a southwest-high to northeast-low distribution pattern, resulting in distinct and varied landscapes between the northern and southern regions (Figure 1). Anhui lies within a transitional climate zone between the subtropical and warm temperate zones, with an annual average temperature ranging from 13 °C to 22 °C and average precipitation between 773 mm and 1670 mm. As of 2023, Anhui Province had a population of approximately 61.21 million, with a per capita GDP of CNY 76,800.

2.2. Data Sources

This study primarily incorporates land use data and surface temperature data. The land use data, sourced from the Resource and Environment Science Data Platform (https://www.resdc.cn/, accessed on 13 January 2024), are classified into 16 distinct types of land use, taking into account the terrain characteristics of the study area, and possess a spatial resolution of 30 m. Similarly, the surface temperature data, also obtained from the Resource and Environment Science Data Platform, have a spatial resolution of 1000 m.

3. Methodology

3.1. Research Framework

To evaluate the impact of HQ on the spatiotemporal evolution of the thermal environment in Anhui Province, this study employs a research framework encompassing three main aspects (Figure 2). First, the spatiotemporal evolution of land use in Anhui Province from 2000 to 2020 was analyzed, utilizing the InVEST model to investigate the effects of various land use types on HQ. Second, the study examines the spatiotemporal evolution of surface temperature in the province during the same period. The mean–standard deviation method was employed to classify surface temperatures, and hotspot analysis was conducted to elucidate their spatial distribution characteristics. Finally, the research explores the influence of HQ on surface temperature, introducing a contribution index to assess the impact of different types of HQ and urban areas on regional relative surface temperature.

3.2. Relative Land Surface Temperature (RLST)

The mean–standard deviation method is a statistical technique employed to describe the distribution and variability of values within a dataset [40]. This method has been extensively utilized in temperature grading, wherein the average surface temperature of a region is combined with varying multiples of the standard deviation to classify temperature levels. The standard deviation reflects the degree of deviation of temperature from the mean, while the interplay between the mean and standard deviation reveals temperature differences across various land cover types. A key advantage of this approach is its focus on the variation of land cover temperatures in relation to the mean temperature, allowing for the classification of temperature levels based on the relationship between multiples of the standard deviation and the mean temperature. Additionally, when comparing the evolution of UHI across different periods, this method mitigates the impact of temporal differences on research outcomes to some extent.

This study categorizes surface temperatures into five levels using the mean–standard deviation method: Extremely High Temperature (EHT), High Temperature (HT), Medium Temperature (MT), Low Temperature (LT), and Extremely Low Temperature (ELT). This approach effectively captures the concentration and variability of surface temperatures by integrating the mean with various multiples of the standard deviation [32]. Detailed classifications of temperature levels are presented in

Table 1. Classification standard of surface temperature class.

Temperature Grades	Basis	Temperature Grades	Basis
Extremely High Temperature	T > μ + 1.5 std	Low Temperature	μ − 1.5 std < T ≤ μ − 0.5 std
High Temperature	μ + 0.5 std < T ≤ μ + 1.5 std	Extremely Low Temperature	T < μ − 1.5 std
Medium Temperature	μ − 0.5 std < T ≤ μ + 0.5 std	/	/

.

3.3. Habitat Quality

The InVEST model evaluates HQ by analyzing the effects of land use changes and various land cover types on biodiversity within a given region. HQ encompasses the availability and usability of ecological resources, as well as the capacity of ecosystems, to deliver services that support individual and population survival, reproduction, and development [41]. In this study, the HQ model within the InVEST framework was employed, utilizing land use types, stressors, and sensitivity as model inputs to assess changes in HQ.

$$D_{xj}={\displaystyle {\sum }_{r=1}^R{\displaystyle {\sum }_{y=1}^{Y_r}(\frac{W_r}{{\displaystyle {\sum }_{r=1}^R{W}_r}})}}{r}_y{i}_{rxy}{\beta }_x{S}_{jr}$$

(1)

$$\begin{array}{l}{i}_{rxy}=1-({\displaystyle \frac{dxy}{dr\max }})\\ i_{rxy}=\exp ({\displaystyle \frac{-2.99dxy}{dr\max }})\end{array}$$

(2)

where R is the number of threat factors; W_r is the weight of threat source r. Y_r is the grid number of stress factor r. r_y is the stress intensity of grid y. i_rxy is the stress level of stress factor r on x in grid y. ${\beta }_x$ is the anti-interference level of the habitat; S_jr is the sensitivity of land type j to stress factor r. dxy is the linear distance between the lattice x and y; drmax is the maximum range (km) of stress factor r. In this study, cultivated land, industrial and mining land, rural residential sites, and construction land were selected as threat sources. For specific parameters, see Tables S1 and S2 in the annex.

3.4. Getis-Ord Gi* Index

The Getis-Ord Gi* analysis [42] identifies regions within a specific area that exhibit notably high service supply capabilities (hotspots) or relatively low capabilities (cold spots). By employing the Getis-Ord Gi* module in ArcGIS, this analysis provides a direct representation of the spatial clustering of high-value (hotspot) and low-value (cold spot) areas, which are categorized into seven distinct types based on the results. The calculation formula is as follows:

$$G*={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{Q}_{ij}-\overline{a}{\displaystyle \sum _{i=1}^n{Q}_{ij}}}}{\sqrt{\frac{{\displaystyle \sum _{i=1}^n{a}_i^2}}{n}}--a^{-2}\sqrt{\frac{\left[{\displaystyle \sum _{i=1}^n{Q}_{ij}^2-{({\displaystyle \sum _{i=1}^n{Q}_{ij}})}^2}\right]}{n-1}}}}$$

(3)

where ${{\displaystyle G}}_i^{*}$ is the aggregation index of grid i; a_i is the attribute value of grid i; Q_ij is the weight matrix; n is the total number of elements; and $\overline{a}$ is the average of all elements relative to the total surface temperature.

3.5. Quantification of Contribution Index

To investigate the contributions of different HQ types in various cities of Anhui Province to the overall regional relative LST over the past 20 years, we introduced the Contribution Index (CI). The formula for its calculation is as follows:

$$CI=(\overline{LS{T}_d}-\overline{LST})\times ({\displaystyle \frac{S_d}{S}})$$

(4)

where $\overline{LS{T}_d}$ and $\overline{LST}$ refer to the average relative surface temperature of region d and the average relative surface temperature of the whole region, where d refers to 5 HQ types and 16 cities, respectively. Sd and S are the area of region d and the area of the entire study area. If CI ≥ 0, it means that the LST of area d has a positive contribution to the LST of the whole study area, which is a warming effect; if CI < 0, it means that area d has a negative contribution to the LST elevation of the whole study area, which is a cooling effect.

4. Results

4.1. Characteristics of Spatial–Temporal Evolution of Land Use

From 2000 to 2020, the predominant land use types in Anhui Province were paddy fields (average proportion of 30.42%), drylands (average proportion of 26.60%), and wooded land (average proportion of 16.02%) (Figure 3a–e). Together, these land types comprised over 70% of the total area within the study region (Figure 4). During this period, both drylands and paddy fields experienced a gradual decline, decreasing by 1.44% and 1.38%, respectively (Figure 3f). Although there was an initial increase in these land types, as well as in forest areas, in 2005, subsequent reductions were observed. In contrast, areas designated for urban land and rural settlements have exhibited continuous growth, increasing by 1.32% and 0.54%, respectively.

In terms of spatial distribution, the study area was predominantly characterized by paddy field and dryland (Figure 3a–e). Dryland was primarily located in the northern region, whereas paddy field was concentrated in the central region. Wooded land was predominantly found in the southern and southwestern areas. Lake was mainly situated in the central and southwestern regions, while reservoir pond was distributed in scattered patches throughout the landscape. High-coverage grassland was primarily located in the southwestern part of the study area, with additional grassland appearing as isolated patches. Urban land was chiefly concentrated in the central and northern regions, with other areas exhibiting a more dispersed distribution.

4.2. Temporal and Spatial Evolution Characteristics of Habitat Quality

HQ was classified into five distinct levels using the equal interval method: poor (average proportion 36.33%), worse (average proportion 30.66%), medium (average proportion 1.02%), good (average proportion 15.92%), and excellent (average proportion 16.07%). This classification reveals pronounced spatial differentiation characteristics, as illustrated in Figure 5. The predominant distribution of HQ within the study area was categorized as poor and worse, resulting in contiguous zones, as shown in Figure 6. Conversely, good and excellent HQ levels were dispersed throughout the region, which hindered the formation of extensive continuous areas; these were primarily located in the southern, southwestern, and eastern sections of the study area. The medium HQ was predominantly concentrated near the Yangtze River Basin, although it occupied a relatively smaller area. Between 2000 and 2020, the area classified as poor HQ increased by 1.47%, while the area classified as worse HQ decreased by 1.41%, with negligible changes observed in the other HQ categories. The transfer characteristics of HQ are particularly notable, predominantly reflecting a transition from poor to good quality. Notably, the poor HQ in the northern region exhibited relative stability, with minimal evidence of transfer.

4.3. Temporal and Spatial Evolution Characteristics of Relative Surface Temperature Day and Night

This study examines the spatial pattern variations in the annual average RLST during both day and night in Anhui Province from 2000 to 2020. The overall RLST in Anhui demonstrated a consistent pattern of being higher during the day than at night, predominantly influenced by the MT and HT, which together comprised over 70% of the area (Figure 7). Notably, the spatial distribution of RLST revealed significant discrepancies: the ETL and LT were primarily concentrated in regions such as the Chaohu Lake water system, as well as the southwestern and eastern areas, while the MT was largely found in the central and northern regions. The HT and EHT were predominantly distributed in the northern region and urban areas. Between 2000 and 2020, the ETL, LT, and EHT exhibited an upward trend, whereas the MT and HT displayed a downward trajectory. By 2020, EHT was primarily concentrated in the northwest of the study area, forming a contiguous distribution, while the central urban region exhibited a more scattered pattern. In 2010, the MT reached its peak coverage, accounting for 45.5% of the area, followed by HT in 2015, which peaked at 30.87%. Over time, the ETL and LT in the southern and southwestern regions gradually expanded, with the ETL forming a substantial clustered distribution, whereas the eastern region experienced a gradual decline in ETL. Furthermore, the EHT in the urban land of Hefei has been steadily increasing, closely correlating with urbanization processes and the expansion of impervious surfaces.

4.4. Hot Spot Analysis of Day and Night Relative Surface Temperature

The spatial clustering of RLST in Anhui Province from 2000 to 2020 exhibited significant patterns. Between 2000 and 2010, the distribution revealed a predominance of hotspots in the northern region, with central areas showing negligible variation and cold spots prevalent in the south. By 2015, the distribution shifted predominantly towards cold spots and areas of insignificance. However, by 2020, the pattern of northern hotspots and central insignificance, alongside southern cold spots, became increasingly pronounced (Figure 8). Over the two-decade period, the total area classified as cold spots increased by 33.25%, while the area of insignificant regions decreased by 33.48%. In contrast, the total area of hotspots remained relatively stable, with only a slight increase of 0.23% (Figure 8). Notably, the expansion of cold spots primarily encroached upon previously established hotspot areas.

The spatial clustering characteristics of RLST during both day and night within the same year have exhibited considerable consistency. Since 2000, significant changes have occurred in the strong hotspot areas of Huaibei, Suzhou, and Bozhou, while other regions experienced only minor variations. Over the time series, both strong hotspots and strong cold spots have gradually expanded, whereas areas classified as insignificant have declined. From 2000 to 2010, strong and relatively strong hotspots were primarily concentrated in the northern regions, including Bozhou, Huaibei, Suzhou, and Bengbu, as well as in the southeastern region, particularly Xuancheng. In contrast, the central urban area of Hefei was characterized by the presence of weak and relatively strong hotspots. Strong cold spots were predominantly located in the southern regions, such as Chizhou and Huangshan, as well as in certain eastern areas, including Chuzhou.

From 2010 to 2015, the landscape of RLST was characterized by a predominance of weak cold spots and insignificant areas, accompanied by a marked decrease in the extent of strong hotspots. During this period, strong hotspots appeared sporadically in specific regions of Fuyang and Anqing, while strong cold spots became increasingly concentrated in Chizhou and Huangshan. By 2020, the spatial distribution of RLST underwent a significant transformation. Strong, relatively strong, and weak hotspots were primarily located in Fuyang, Bozhou, Bengbu, Huaibei, and Suzhou, whereas insignificant areas were predominantly found in parts of Lu’an, Anqing, Huainan, and Hefei. Strong and relatively strong cold spots were mainly identified in Hefei, Ma’anshan, Xuancheng, Anqing, Chizhou, and Huangshan. The spatial distribution of these cold and hot spots is closely associated with the land use characteristics of Anhui Province. Notably, the northwestern plain area is predominantly agricultural; the southern mountainous regions are largely covered by forest and shrub; and the central area, characterized by high levels of urbanization and extensive impervious surfaces, resulting in relatively high RLST.

4.5. Diurnal Variation Characteristics of Relative Land Surface Temperature in Different Habitat Quality Types

From 2000 to 2020, the variations in RLST across different HQ types in Anhui Province indicated that areas classified as poor HQ exhibited the highest average RLST (average 22.01 °C), followed closely by those designated as worse HQ (average 21.45 °C). This was followed by medium HQ (average 20.60 °C), medium-high HQ (average 20.23 °C), and high HQ (average 20.52 °C). These findings are illustrated in Figure 9.

There were significant differences in RLST between day and night across various HQ levels. Specifically, areas classified as having poor HQ exhibited the highest RLST, typically ranging from 20.95 °C to 23.28 °C, with their distribution area continuously expanding. In contrast, regions with good HQ demonstrated smaller fluctuations in RLST, predominantly remaining between 19.63 °C and 21.62 °C. The RLST in areas of worse HQ fluctuated between 20.60 °C and 22.62 °C. For medium and excellent HQ, the variation in RLST was primarily distributed between 19.82 °C and 22.01 °C, and between 19.78 °C and 21.93 °C, respectively. Over time, RLST across different HQ levels have generally shown a declining trend, although certain local areas have experienced relative increases. Notably, the disparity in RLST between poor and excellent HQ has been consistently increasing, with a change ratio of 68%.

4.6. Diurnal Contribution of Different Habitat Quality Types and Cities to Relative Land Surface Temperature

By introducing the CI, we analyzed the impact of different HQ types on RLST from 2000 to 2020. Overall, HQ types demonstrated consistent contributions to both diurnal and nocturnal RLST (Figure 10). Significant variations in HQ contributions to RLST were observed across different years. Notably, poor HQ and worse HQ substantially increased RLST due to their prevalence in agricultural and urban areas, resulting in pronounced warming effects. Specifically, the warming effect associated with poor HQ has steadily intensified since 2010. The CI for poor HQ fluctuated from 0 in 2000 to a peak of 0.061 in 2010, followed by a gradual decline. In contrast, moderate, good, and excellent HQs exhibited negative contributions, with good and excellent HQs displaying the most substantial negative impacts. The negative contribution of moderate HQ primarily ranged between −0.003 and −0.007. Since 2010, the negative contributions of good and excellent HQs have consistently increased, particularly in forests, contributing to a slight reduction in RLST.

In this study, we introduced the CI to analyze the impact of 16 cities in Anhui Province on regional thermal environments during both daytime and nighttime, revealing consistent differences in their contributions (Figure 11). From 2000 to 2020, significant variations were observed in how these cities influenced regional thermal conditions. Specifically, Bengbu, Fuyang, Chuzhou, Huaibei, and Tongling demonstrated negative contributions to RLST over the study period, while Suzhou, Chizhou, Wuhu, Anqing, Xuancheng, and Lu’an exhibited positive contributions. Ma’anshan, Huangshan, Hefei, Bozhou, and Huainan displayed varying degrees of contribution across different years.

During the daytime in 2010, both Ma’anshan and Huangshan recorded CIs of 0, whereas in other years, their contributions were negative. Bozhou and Huainan showed negative contributions in 2000 but positive contributions in other years, with the exception of 2015. Among these cities, Bengbu had the most substantial negative contribution, while Fuyang reached its lowest CI of −0.085 in 2020. In contrast, Lu’an recorded the highest positive contribution, achieving a CI of 0.232 in 2020. Over the period from 2000 to 2020, Bengbu exhibited the greatest variation in negative CI, with a change of −0.038, while Lu’an showed the largest variation in positive CI at 0.103. Huainan demonstrated the smallest variation in CI at 0.004. Nighttime trends mirrored those observed during the day; however, in 2010, Ma’anshan’s negative contribution was recorded at −0.001, compared to 0.000 during the day. Overall, cities in northern Anhui Province tended to exhibit negative contributions to regional thermal environments, whereas southern cities displayed positive contributions. These patterns closely correlate with levels of urbanization and land cover across the province.

5. Discussion

5.1. Implications of Habitat Quality for Regional Thermal Environment Management in Anhui Province

HQ, as a crucial indicator for assessing the health of regional ecosystems, is significantly influenced by alterations in industrial structure and land use [43]. This study employs LST data to investigate the relationship between HQ and the regional thermal environment. To promote ecological sustainability, it is essential to comprehensively understand the spatiotemporal dynamics of HQ and its effects on regional thermal conditions. Such insights will provide valuable scientific guidance for governmental and relevant agencies in the strategic planning of blue–green spaces and the advancement of ecological civilization.

From 2000 to 2020, the HQ index in Anhui Province exhibited an overall decline, revealing a spatial distribution marked by lower values in the northern region and higher values in the southwestern and southern areas. This pattern reflects significant spatial heterogeneity. The deterioration of HQ in the northern region can be primarily attributed to extensive agricultural land use, flat terrain, high population density, and urban expansion, which have collectively led to considerable ecological degradation. Moreover, prolonged and intensive cultivation practices associated with traditional agriculture have severely affected ecological land, including forests, grasslands, and water, resulting in a more fragile ecological environment [27,44]. In contrast, HQ in the southern and southwestern regions of Anhui Province has remained relatively stable. This stability is largely due to the extensive network of nature reserves, the presence of numerous forest and grassland ecosystems, and limited industrial and urban development. These factors have mitigated the adverse effects of human activities, thereby preserving a higher level of ecological quality.

The disparities in regional economic development significantly contributed to the spatial variations in the HQ. Between 2000 and 2020, the most pronounced degradation in HQ was observed in and around the provincial capital of Hefei, as well as in neighboring cities such as Chuzhou, Huainan, and Lu’an, and along the Yangtze River in cities including Ma’anshan, Wuhu, Tongling, Chizhou, and Anqing—all of which have experienced rapid economic growth [45]. These regions face intense pressure from population concentration and large-scale infrastructure development, resulting in a marked increase in demand for industrial and residential land. The rapid expansion of urban land has encroached upon significant portions of ecological land, including forests, grasslands, water, and even some croplands. Consequently, this encroachment has led to widespread degradation of HQ and placed considerable stress on the regional ecological environment.

LST and HQ exhibited a significant spatial correlation. Areas characterized by poor HQ were generally associated with elevated relative LST, whereas regions with excellent HQ were marked by lower RLST. This relationship indicates that RLST is influenced not only by topographic features but also by land cover types. For instance, the northern portion of the study area, dominated by plains and agricultural activities, exhibits a fragile ecological environment that is highly susceptible to human disturbance and monoculture cropping systems, both of which contribute to increased RLST. The ongoing processes of urbanization and industrialization exacerbate this issue, as urban expansion encroaches upon extensive green spaces. This urban sprawl generates significant heat and intensifies the UHI effect, particularly evident in Hefei. Conversely, the southwestern and southern regions of Anhui Province, which feature excellent HQ and abundant green cover—such as forest and grassland—demonstrate a pronounced cooling effect that effectively moderates RLST.

The UHI effect is intrinsically linked to the process of urbanization [46]. As urban areas expand rapidly, the proliferation of built environments leads to the degradation of urban green spaces and surrounding natural vegetation, thereby exacerbating the UHI effect [9,47]. While an increase in urban green space generally mitigates surface temperatures, the expansion of urban land typically results in elevated temperatures. Urban land, characterized by impermeable surfaces, high thermal capacity, and low evaporation rates, significantly impedes air circulation, facilitating the intensification of the UHI effect [48]. Empirical research indicates a positive correlation between the intensity of the UHI effect and the size of urban areas [49]; larger urban regions with a greater proportion of urban land absorb more heat, resulting in prolonged warming of these areas [50,51]. Furthermore, increased urbanization often correlates with economic development, which escalates energy consumption and further aggravates the UHI effect [52]. Therefore, it is imperative to regulate the expansion of impermeable surfaces and enhance the coverage of cooling surface types, such as green spaces and water, to effectively address thermal environmental challenges in urban and regional contexts.

The coefficient representing the influence of various HQ on RLST ranged from −0.2 to 0.3. Specifically, regions characterized by poor or deteriorating HQ, predominantly comprising dryland and paddy field, exhibited a positive effect on the regional thermal environment due to elevated temperatures resulting from direct solar radiation. In contrast, regions with good or excellent HQ, primarily forest, exerted a negative impact on the thermal environment, as tree canopies and evapotranspiration processes effectively mitigated RLST. The coefficient for urban contributions to RLST varied between −0.1 and 0.25. Cities such as Suzhou, Chizhou, Wuhu, Anqing, Xuancheng, and Lu’an demonstrated a positive influence on the regional thermal environment, while cities including Bengbu, Fuyang, Chuzhou, Huaibei, Tongling, Ma’anshan, and Hefei exhibited a negative impact. This variability is closely associated with regional topographical features and levels of economic development. Northern Anhui is characterized by plains and rapid urbanization, whereas southern Anhui is defined by mountainous terrain and a more balanced approach to urbanization. Therefore, strategic urban planning and economic restructuring are essential to mitigate regional thermal environmental challenges. Future urban residential development and biodiversity conservation efforts should adhere to the zoning recommendations presented (Figure 12).

5.2. Shortcomings and Prospects

This study aimed to evaluate the impact of spatiotemporal changes in HQ on the thermal environment of Anhui Province. However, the research was subject to several limitations. Although long-term sequence studies help mitigate uncertainties associated with short-term data fluctuations, this analysis relied on LST data with a resolution of 1000 m due to data availability constraints. While this resolution is appropriate for meso- and macro-scale investigations, higher-resolution data are essential for accurately characterizing the evolution of thermal environments in urban areas.

Furthermore, the analysis was based on annual average data, as limitations in data volume and availability precluded the use of daily data. This constraint may introduce uncertainties in differentiating between summer and winter LST. Future research should incorporate daily data to enable a more nuanced analysis. Additionally, establishing field meteorological stations to collect in situ LST measurements would facilitate the correction of errors in model inversions, thereby enhancing data accuracy.

Additionally, future studies should incorporate an analysis of socioeconomic factors influencing the thermal environment to strengthen the research framework. In addition to macro-scale provincial analyses, it is imperative to conduct urban thermal environment monitoring at meso- and micro-scales. Establishing a multi-scale and multidimensional research framework will provide a more comprehensive scientific basis for the investigation of regional thermal environments and HQ.

6. Conclusions

This study utilized the InVEST model’s HQ framework to examine the spatiotemporal evolution of HQ in Anhui Province from 2000 to 2020. By integrating LST data for the relevant periods, we analyzed the diurnal variation in RLST using the mean standard deviation method. The Getis-Ord Gi* hotspot analysis was employed to investigate the spatial heterogeneity of diurnal RLST distributions over time. Furthermore, the CI method was applied to assess the impact of various HQ types and municipal areas on the regional thermal environment.

The findings indicate that, between 2000 and 2020, land use in Anhui Province has been predominantly characterized by paddy field (average 30.42%), dryland (26.60%), and forest (16.02%), with these three land types collectively accounting for over 70% of the total study area. However, the areas of dryland and paddy field have experienced declines of 1.44% and 1.38%, respectively. In terms of HQ, Anhui Province was primarily comprised of regions classified as having poor HQ (36.33%) and worse HQ (30.66%), predominantly located in the northern and central parts of the province. In contrast, regions with good HQ and excellent HQ were concentrated in the southern, southwestern, and eastern areas.

In Anhui Province, the MT and HT zones of RLST were predominant, encompassing over 70% of the area. Overall, RLST exhibited a consistent upward trend, particularly pronounced in the northern and central regions, with the most significant increases noted in provincial capital cities. For instance, in Hefei City, the substantial rise in RLST corresponded with urban expansion. The effects of varying HQ on the regional thermal environment revealed distinct patterns: regions with worse HQ experienced notable temperature increases, whereas areas with excellent HQ demonstrated cooling effects. Furthermore, the impact of HQ on RLST varied significantly across different cities. Consequently, future spatial planning should prioritize the preservation of regions with excellent HQ and strive to enhance the ecological environment to achieve sustainable development and improve human well-being.