Next Article in Journal
Interwoven Landscapes: Gender and Land in the Kafue Flats, Zambia
Next Article in Special Issue
Land Use Change and Landscape Ecological Risk Assessment Based on Terrain Gradients in Yuanmou Basin
Previous Article in Journal
The Land-Use and Land-Cover Changes in the Este District, South Gondar Zone, Northwestern Ethiopia, in the Last Four Decades (the 1980s to 2020s)
Previous Article in Special Issue
The Relationship between Rural Spatial Form and Carbon Emission—A Case Study of Suburban Integrated Villages in Hunan Province, China
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Land
Volume 12
Issue 9
10.3390/land12091656
land-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

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

Historical Spatial Radiation Range of the Yongding River Corridor in Beijing Plain Section: Implications for Landscape Patterns and Ecological Restoration

by
Ruiying Yang
,
Zhicheng Liu
,
Wenxin Kang
,
Junyi Su
,
Renfei Zhang
and
Xiaoming Xu
*
School of Landscape Architecture, Beijing Forestry University, Beijing 100083, China
*
Author to whom correspondence should be addressed.
Land 2023, 12(9), 1656; https://doi.org/10.3390/land12091656
Submission received: 17 July 2023 / Revised: 17 August 2023 / Accepted: 22 August 2023 / Published: 24 August 2023
(This article belongs to the Special Issue 2nd Edition: Land Use Change and Its Environmental Effects)
Browse Figures
Versions Notes

Abstract

:
The radiation range of the corridor effect holds great significance for the ecological restoration, planning, and sustainable development of river corridors. This study focuses on the Beijing plain section of the Yongding River, which has been cut off for half a century, and improves the research methodology. Utilizing land use data from 1967 and 1980, ArcGIS and Fragstats were employed to establish 5 km buffer zones on both sides of the Yongding River corridor. The buffer zone analysis method was then applied to investigate landscape pattern changes. Through SPSS correlation analysis and curve fitting, sensitive landscape indices were identified, and their change characteristics were analyzed to unveil the historical spatial radiation range and characteristics of the Yongding River corridor. The findings of this research are as follows: First, as the buffer width increases, the landscape pattern changes, showing a decrease in heterogeneity, an increase in aggregation and spread, and a good connection between dominant patches. Forest land exhibited higher levels of fragmentation and dispersion, cultivated land demonstrated improved dominance, and construction land became more regular and dispersed. Second, the spatial radiation range of the landscape level within the river corridor was approximately 4 km. The inflection point for the radiation range on forest land was found at 3.5–4 km, while for construction land, it occurred at 4.5 km. The outcomes of this study can be utilized to evaluate the impact of river corridors on landscape patterns in the period of good historical ecology. They also provide more targeted measures and scientific basis for landscape pattern protection and river ecological restoration planning after the restoration of water flow in the Yongding River plain.

1. Introduction

The river corridor, as a linear or zonal landscape element distinct from the surrounding matrix, refers to the vegetation zone along the river and its interlocking land and water areas on both sides, including the main channel, floodplain, and highland vegetation buffer zone [1,2]. The corridor system comprises corridors and geographic units within the corridor’s influence area. Given the unidirectional flow of the river, it is highly influenced by the surrounding geographic units. Therefore, when implementing ecological restoration, it is crucial to consider not only the river corridor itself but also the various ecological geographic units that interact with it, as they provide valuable insights and references [3]. The peripheral buffer zone of river corridors plays a significant role in influencing rivers. Once it deteriorates, landscape fragmentation also occurs. The term “landscape fragmentation” describes how once continuous landscape elements have been broken up into several erratic areas [4]. Landscape fragmentation is a significant factor contributing to the decline in regional biodiversity [5]. In order to avoid the fragmentation of landscape patterns, it is necessary to evaluate the radiation range of river corridor effects for the ecological restoration of river corridors. The corridor effect means the changes in the general landscape features and all the landscape classes with different distances away from the main watercourses, and it can reflect the radiative extent of river ecological functions [6]. Evaluating the spatial radiation range of the river corridor during its ecological restoration process is beneficial for understanding the impact of the river corridor on the landscape pattern. It also aids in planning processes to protect the landscape pattern from fragmentation.
Landscape pattern analysis methods have been widely applied in many fields [7,8,9,10]. Through landscape pattern analysis, studies of corridor effects have predominantly focused on green corridors, ventilation corridors, road traffic, and urban expansion [11,12,13,14,15,16,17], while studies on the river corridor effect mostly focus on temperature and humidity effects and thermal environment effects [18,19,20,21]. Zhou Huarong et al. contributed significantly to understanding the radiation range and characteristics of river corridors by analyzing changes in landscape indices within different buffer zones [6]. This work laid a foundation for studying the radiation range of river corridors. However, in terms of applying the landscape pattern analysis method, the spatial radiation range of the river corridor is often considered based on a single linear relationship [22]. Additionally, the selection of indices is typically based on the analysis of area, shape, and distribution of landscape patterns [23]. It should be noted that different widths and the quality of habitat in river corridors can alter environmental heterogeneity and species diversity [24]. Environmental heterogeneity refers to the changes in abiotic conditions between regions within a regional unit; that is, the environmental differences between regions [25], and the radiation range of the corridor effect can vary with changes in vegetation type and target species [26]. Some areas within the river corridor buffer zone show strong environmental heterogeneity, which suggests the landscape indices and river corridor width may have a nonlinear relationship. Buffer zones are places intended to improve the protection of a certain conservation area, usually on its outskirts [27]. Based on Liu Weiyi’s definition of the urban waterfront buffer zone, the definition of the river corridor buffer zone is summarized as follows: the river corridor buffer zone is the scope of influence of river corridors, which are zones where aquatic and terrestrial ecosystems intersect, functional transition zones connecting rivers and waterfront land, and a natural protective barrier for rivers [28]. Moreover, different landscape indices have different sensitivities to flow breaks, and the numerical changes of landscape indices before and after flow breaks may also differ. This greatly increases the uncertainty and complexity of the study. Therefore, the main question explored in this paper is how the characteristics of landscape pattern changes within the buffer zone exhibit sensitivity to stream damage and spatial scale. In order to carry out further research on the radiation range of the corridor effect, the method of landscape pattern analysis was used to study the landscape characteristics and the pattern of change of each land type with distance. It is closely related to the sustainable development goal of sustainable use of terrestrial ecosystems.
This study focused on the Beijing section of the Yongding River as the primary study area for the purpose of sustainable development of the river corridor. In order to provide a basis for the government’s work on ecological restoration of the Yongding River, the historical period representing the ecological integrity of the Yongding River was selected as the reference system [2]. The buffer zone was established by stratifying the land use data from 1967 (prior to the interruption of river flows) and 1980 (after the interruption of river flows). Through this approach, the evolution characteristics of the river corridor landscape pattern were analyzed at landscape level and class level. Furthermore, the study conducted an analysis of the correlation between landscape indices and river corridor buffers. After identifying the landscape indices that were sensitive to the interruption of river flow, the study quantitatively analyzed the variations of the landscape indices in response to different buffer widths. These findings hold significant value as they provide crucial insights for the delineation of the peripheral effect area of the corridor during the ongoing ecological restoration of the river. Additionally, the data obtained from this study can serve as a foundation for informing the planning of the landscape pattern structure within the corridor and its effect area.

2. Study Area

The study area is situated in the southwestern part of the central city of Beijing, specifically within the middle and lower reaches of the Yongding River [29]. The geographical scope extended from the Sanjiadian Water Conservancy Project in the north to the Beijing Boundary in the south (Figure 1). This region holds significant research significance due to its historical importance as a critical breakwater section. The Yongding River serves as an essential water-covered area, ecological barrier, and ecological corridor within the Beijing–Tianjin–Hebei region. However, it faces several prominent challenges, including the overexploitation of water resources, inadequate environmental carrying capacity, severe pollution, river disconnection, and ecosystem degradation [30]. Over the past half-century, the Yongding River segment stretching from Sanjiadian to Lugouqiao has experienced long-term disconnection, while the section from Lugouqiao to Qujiadian has essentially dried up year-round. To address the deteriorating ecological conditions of the Yongding River, the central government, Beijing municipal government, and relevant water conservancy departments have made the decision to implement ecological management measures. Among these initiatives, the key focus lies in restoring the plain section of Beijing, which suffers from severe disconnection and desertification [31]. The comprehensive treatment and ecological restoration of the Yongding River aim to create a green ecological river corridor. This effort holds significant importance as it represents a primary breakthrough in the ecological field and plays a vital role in improving the regional ecological environment, serving as a leading demonstration project in the pursuit of enhanced ecological conditions within the Beijing–Tianjin–Hebei region [32,33].

3. Methodology

To investigate the effects of the historical spatial radiation range of river corridors, the analysis employed the buffer zone analysis method to examine changes in landscape pattern. SPSS correlation analysis and curve fitting techniques were used to identify sensitive landscape indices and analyze their change characteristics. Subsequently, the obtained results were compared with actual data to analyze the overall spatial radiation range of the river corridor. Figure 2 illustrates the detailed workflow of the study, which can be broadly divided into three parts: (1) data processing, (2) buffer zone definition and analysis of landscape pattern changes in the river corridor, and (3) fitting regression analysis to uncover the spatial radiation range of the river corridor.

3.1. Data Processing

The remote sensing satellite images of the Yongding River corridor were selected from two time periods: 1967 and 1980. These time points serve as valuable references before and after a significant cutoff event. In the 1950s, the Yongding River faced traditional flood and debris flow disasters, which later transitioned into wind–sand disasters. Eventually, the river ceased to flow around the 1980s, a condition that persisted for approximately half a century [34]. Considering the historical context and data availability, 1967 was chosen as the earliest time point for obtaining effective satellite remote sensing data necessary for this research. During this period, the ecological environment of the Yongding River was not threatened by floods or disrupted river channels, and the overall ecological condition was relatively favorable. The selected remote sensing data from 1967 comprises black-and-white satellite images. It is worth noting that the peak growing season for plants in 1967 was from May to September, with September representing a period of abundant floodwater. Hence, satellite images from September 1967 were of high quality and suitable for analysis. To ensure consistency in data sources and account for the temporal aspect of the study, satellite images from September 1967 were chosen as the primary objects of analysis, while satellite images from the same period in 1980, after the river cutoff, were selected as control references.
The research scope of the river corridor was 10 km, extending outward based on the centerline of the river. Based on the setting standards of different domestic and foreign scholars on the width of river corridors [35], the historical maximum flood level of the Yongding River [34,36,37], the spatial structure of the corridor, and the demands for biodiversity, the research scope of each side of the river corridor was finally set at 5 times the width of the corridor (the width of the river corridor is approximately 1 km). This is because exceeding this range would involve successive large-scale urban environments, and a range of 5 km can ensure the ecological diversity of the buffer zone. Therefore, the study area of the buffer zone was taken as 5 km from the left and right banks.
The river corridor landscape was classified into eight categories: cultivated land, forest land, grassland, bottomland, islands, water, construction land, and unused land. When classifying the river corridor landscape types, many conditions were taken into account, including human characteristics of the natural environment and ecosystem of the Yongding River corridor, river morphology characteristics, river geomorphology conditions, national standards for classifying the current state of land use, land use status of the Yongding River, impacts of human activities, and the quality of remote sensing images [38,39,40]. After determining the landscape type, ENVI was applied to pre-correct Landsat 8 images. Subsequently, using the eight types of river corridor landscape types as a reference basis, visual interpretation was carried out using ArcGIS 10.4 software, so as to obtain a land cover map (Figure 3).

3.2. Definition of Buffer Zone of River Corridor

The corridor effect can be considered as the change in overall landscape characteristics and various landscape types at different distances from the main channel of the river, providing insights into the radiating range of river ecological functions [6]. To capture the spatial dynamics within the middle and lower reaches of the Yongding River and the specific characteristics of the selected image data, a buffer zone analysis was employed [41]. Utilizing the center of the Yongding River as a demarcation point, the average width of the Yongding River channel measures approximately 1 km, encompassing 0.5 km from each bank. Additionally, there are distinct embankments on either side of the river, varying in width. The span between the embankments on both sides ranges from 1 to 2 km, while the distance between embankments on each side typically falls between 0.5 and 1 km. Therefore, taking into account the special impacts of the Yongding River’s width and embankment width, the buffer zones were delineated at 0.5 km intervals within 5 km of each side of the river, as follows: 0.5 km, 1 km, 1.5 km, 2 km, 2.5 km, 3 km, 3.5 km, 4 km, 4.5 km, 5 km. In addition, 0.5 km is half the width of the river corridor, and the overall study area was taken as 10 km (10 times the corridor width) (Figure 4). Subsequently, the landscape type map was rasterized using Fragstats and ArcGIS, enabling the calculation of landscape indices. By analyzing the primary landscape indices, the spatial radiation range and characteristics of the river corridor were examined at landscape level and class level.

3.3. Analysis of Landscape Pattern Changes

Landscape pattern indices are crucial tools for analyzing pattern changes as they provide valuable insights into the spatial structure, temporal dynamics, and characteristics of patches, types, and landscape levels [42]. In this study, considering the structural characteristics of the Yongding River corridor, a comprehensive selection of landscape indices was made, which covered a wide range of patch characteristics, landscape heterogeneity, and spatial attributes at landscape level and class level. The selected landscape indices at landscape level included PD (patch density), ED (edge density), AI (aggregation index), LPI (largest patch index), CONTAG (contagion index), COHESION (patch cohesion index) and SHDI (Shannon’s diversity index). Selected landscape indices at class level consisted of the proportion of landscape area occupied by PLAND (percentage of landscape), PD (patch density), ED (edge density), AI (aggregation index), LPI (largest patch index) and COHESION (patch cohesion index) [6,43,44]. They were calculated by Fragstats, and their specific calculation formulas and interpretations can be found in Table 1.

3.4. Analysis of Spatial Radiation Range of River Corridor

In the realm of statistical analysis, it is imperative to note that the Pearson correlation coefficient necessitates the adherence of both variables to a normal distribution. However, when encountering scenarios where the two variables deviate from a normal distribution, an alternative avenue, the Spearman’s rank correlation coefficient, is introduced. This method deftly sidesteps the prerequisite for variable distribution conformity and resides within the domain of non-parametric testing [48]. In this study, Spearman correlation analysis by SPSS was used to identify landscape indices that are sensitive to river disconnection and landscape indices that were not significantly affected by river disconnection and buffer width changes were excluded from further analysis. In addition, the variables may not all exhibit a linear relationship with each other; therefore, multiple regression is used to fit the curves and build the regression model. The best-fit formula [49] was selected to establish the fitting function that describes the relationship between the landscape indices and buffer width. Bivariate regression curve fitting analysis was then conducted to elucidate the variations in the landscape indices across different buffer widths. This analysis aimed to uncover the directional impact of buffer width on the landscape pattern and determine the spatial radiation range of river corridors. The flowchart outlining the research method is presented in Figure 2.

4. Results

4.1. Changes of Landscape Pattern in Buffer Zone of River Corridor

4.1.1. Changes of Landscape Indices at Landscape Level

The changes in characteristics of the overall horizontal pattern indices of the landscape are presented in Figure 5. Patch density (PD) and edge density (ED) exhibited a decline as the buffer zone width increased. The rate of change in the buffer zone was highest between 0.5 and 1 km, gradually decreasing thereafter, and eventually stabilizing after reaching 3.5 km. This trend suggests a gradual reduction in the overall fragmentation degree and landscape complexity, followed by a period of stability. Largest patch index (LPI) displayed a gradual upward trend, with notable turning points at 1 km, 2.5 km, and 4 km. Attributable to the interplay of the Yongding River’s width and embankments, a transformational dynamic unfolds beyond the 1 km threshold. Within this expanse, patches of bottomland, forest land, cultivated land, and grassland within the Yongding River embankment were gradually replaced, and blocks of more scattered construction land and regularly planted cultivated land and forest land began to appear. The PD index at the landscape level demonstrates a progressive decrease. In contrast, the PD indices pertaining to cultivated land, forest land, and construction land experienced a consistent rise. The proportion of landscape types increased, but became more and more fragmented by anthropogenic influence. Therefore, the dominant species gradually decreased in the range of 1–2.5 km, and the LPI index gradually decreased. After 2.5 km, the area of cultivated land increased, and it was gradually linked together, and the dominant species were formed in the buffer zone, thus manifesting a subsequent ascending trajectory in the LPI index. At the same time, the Beijing section of the Yongding River has extensive artificial sand and wasteland [50], disrupting the spatial radiation range of the river corridor on the surrounding landscape pattern. As the distance from the river channel increased, the degree of artificial intervention became more pronounced. A gradual decline in SHDI index showed no clear inflection point, and the aggregation index (AI) showed an upward trend, reaching its maximum change rate between 0.5 to 1 km, and subsequently achieving a relatively stable state after 3.5 km. This indicates that, with increasing buffer zone width, the dispersed buffer patches gradually aggregated, forming larger, simpler patches. COHESION and CONTAG exhibited a gradual increase with the expansion of the buffer zone, albeit with a decreasing rate of change. These indices tended to stabilize at the 4 km buffer zone, implying that the landscape pattern within the buffer zone of the river corridor shifted from a dense pattern with multiple elements to a pattern dominated by a dominant patch type with good connectivity. Overall, when considering the overall horizontal change in the landscape pattern, it can be observed that the inflection point representing a significant shift in the landscape pattern within the buffer zone of the Yongding River corridor is situated within the range of 3.5 to 4 km.

4.1.2. Change of Landscape Indices at Class Level

Landscape indices at landscape level provide a comprehensive overview of the overall characteristics of the landscape pattern. However, to gain deeper insights into the distribution characteristics of each specific landscape type along the river corridor, it is essential to examine the changes in class-level indices associated with these landscape types. This approach enables a more detailed exploration of their spatial arrangement and relationships.
Based on the observations depicted in Figure 6 and Figure 7, it can be inferred that, in 1967, the proportions of cultivated land, forested land, and construction land underwent an augmentation as the buffer width expanded. Notably, the dominant categories within this progression were cultivated land and forested land. In 1980, discernible shifts in land allocation patterns were evident. The expanse of cultivated land experienced a notable increase, concomitant with a reduction in forest land. Simultaneously, there was an augmentation in the extent of grassland and bottomland. The construction land remained stable, and the percentage of water did not decrease significantly, but the river breakup was obvious.
Illustrated in Figure 8, the landscape dynamics of 1967 reveal a notable transition across diverse land cover types towards cultivated land, and part of them shifted to forest land and construction land, consequently elevating the prominence of cultivated land within the landscape. Through a comparative analysis of the proportional composition of distinct landscape types across the two years, the following trends emerge: cultivated land boasts a robust ecological dominance, closely trailed by forested land. The extent of construction land and water exhibits stability. Grassland, bottomland, islands, and water constitute a relatively minor proportion.
To delve deeper into this analysis, the study selectively scrutinized the transformation characteristics of forest land, cultivated land, and construction land, the three landscape categories exhibiting the most pronounced changes within the buffer zone.
The change characteristics of landscape indices at class level are presented in Figure 9. When examining PLAND and LPI, it can be observed that forest land demonstrated a stable trend after 2 km due to human disturbances, while the proportion of cultivated land continues to increase. As a result, cultivated land replaced forest land as the dominant land cover within the buffer zone after 2.5 km. Analyzing PD and ED reveals that forest land exhibited the highest PD index, indicating a trend of fragmentation as the distance from the river increases. This fragmentation can be attributed to the presence of artificial sand wastelands or orchards in the Yongding River area prior to the 1980s [50]. In contrast, construction land is primarily composed of fragmented rural construction land groups, with the exception of urban areas. These construction land groups are evenly distributed and exhibit more regular patch shapes. Examining AI and COHESION, it can be observed that the AI indices of forest land and construction land exhibited a turning point after 3 km, indicating a shift from high aggregation to dispersal. Additionally, the COHESION indices of all three landscape types showed an increasing trend, suggesting enhanced connectivity among patches. Specifically, the COHESION index of forest land tended to stabilize at 2 km, while the COHESION index of construction land tended to stabilize at 3 km.
In summary, the findings reveal that as the buffer width increased, forest land became more fragmented and dispersed, while cultivated land demonstrated increased connectivity and aggregation, accompanied by more complex patch shapes. Conversely, construction land exhibited a more regular pattern but remained relatively dispersed. The impact of the river corridor on the transformation of forest land, cultivated land, and construction land was most pronounced within a range of approximately 3 km.

4.2. Analysis of the Spatial Radiation Range of River Corridors

Upon analyzing the outcomes of the normality test conducted between the landscape indices and the buffer width, a noteworthy observation emerged: the p-value of the Shapiro–Wilk test registered below the critical threshold of 0.05, thereby signifying the data’s departure from normality (Table 2). Consequently, to best accommodate this non-parametric distribution, the Spearman correlation analysis method was judiciously employed.
The landscape indices of different buffer zones in 1967 and 1980 were subjected to analysis examining their relationship with the buffer zone width by Spearman correlation. The indices sensitive to the river break were identified and presented in Table 3. The analysis revealed a significant positive correlation between ED, AI, CONTAG, COHESION, and SHDI at the 0.01 level, while PD and LPI exhibited a significant negative correlation with buffer width at the 0.05 level. However, when compared to the 1980 correlation analysis, PD and ED no longer showed significant correlations with buffer width. Additionally, the significance level of AI changed from 0.01 to 0.05, suggesting that PD, ED, and AI were the landscape indices most sensitive to the river break response. The Spearman correlation coefficients for CONTAG, COHESION, and SHDI all yielded a value of 1, signifying a monotonic association between these landscape indices and buffer width. This relationship manifested as a monotonically increasing connection between CONTAG, COHESION, and buffer width, while SHDI showcased a monotonically decreasing trend concerning buffer width. In this context, the variables displayed a tendency to change simultaneously, albeit not necessarily at a uniform rate.
Based on the correlation analysis, the selected landscape indices were subjected to curve fitting regression analysis with the corridor buffer width in 1967. This process established a fitting function to determine the relationship between them and determine the radiation range of the river corridor. The fitting results are presented in Table 4. In the analysis, x represents the river corridor buffer width, y represents the landscape indices, and x exhibits a significant correlation with PD, ED, and AI. The curve estimation results indicate that the quadratic polynomial curve fitting yielded a significantly higher R2 value compared to the linear relationship fitting. This finding suggests that the influence of the river corridor on the landscape pattern is not solely governed by a linear relationship. Consequently, the influence function curve of the river corridor on the landscape indices at different scales was further fitted and presented in Figure 10.
According to the fitted curve, PD and ED exhibited a positive “U” shape relationship with the buffer width, while AI demonstrated an inverted “U” shape relationship. The inflection point for PD occurred at a buffer width of 3.5 km, while for ED, it was around 4 km. The maximum value of AI was observed at a buffer width of 4 km. The results of these three landscape indices aligned with each other. This finding suggests that the overall patch density within the 4 km buffer zone of the river corridor demonstrated a continuous decrease, accompanied by a shift towards more regular and clustered patch shapes. This observation is consistent with the trend of landscape pattern change and the location of the inflection point.

4.3. Analysis of the Spatial Radiation Range of Each Landscape Type of River Corridor

Forest land, cultivated land, and construction land, which exhibited the most significant changes within the buffer zone, were selected for further analysis. The Spearman test was conducted to examine the correlation of the landscape indices and the buffer width within different buffer zones in 1967 and 1980 for each of the three land types (Figure 11). Based on the sensitivity of the landscape indices to the response of river breakage, the PD, ED, AI, and COHESION indices were selected for curve fitting analysis with the river corridor buffer in 1967. The analysis aimed to establish a fitting function that describes the relationship, with x representing the river corridor buffer width and y representing the different types of landscape indices. The results of the curve fitting analysis are presented in Table 5.
The R2 value of the cultivated land PD curve fitting analysis in the table was low and not significant because the cultivated land PD index tended to flatten out as a whole, and the PD index did not change significantly with the increase of buffer width. Outside the Yongding River embankment, the vast majority of the cultivated land was in the regular form of anthropogenic planting, which did not have a greater impact with the buffer zone changes, and the spatial distribution of the patches did not produce significant differences, so the curve analysis was not significant.
The results of the quadratic polynomial fitting analysis for landscape pattern of forest land revealed distinct characteristics influenced by the river corridor. This finding suggests that the impact of the river corridor did not follow a simple linear trend, but rather exhibited spatial heterogeneity across different scales. Based on the fitting results presented in Figure 12, it was observed that the forest ED, AI, and COHESION indices exhibited an inverted “U” shaped relationship with the width of the river corridor buffer. The peaks observed in both the AI and COHESION indices of the forest land occurred at approximately 3.5 km, indicating an increasing level of aggregation and connectivity among patches within the buffer zone of the river corridor. Furthermore, the peaks observed in the ED index were within the range of 4–4.5 km, indicating a greater complexity in the shape of the forest land landscape patches within this buffer zone. In contrast, the PD index and its curve fitting results closely aligned with a linear relationship, indicating a consistent trend of increasing patch density as the buffer width expanded. This finding underscores the intensification of fragmentation within the forest land landscape patches, without any apparent moderation or control measures. The linear relationship observed highlights the persistent and unrestricted impact of buffer width on the patch density and subsequent fragmentation of forest land landscapes.
Based on the curve fitting results presented in Table 4, it was observed that there was no significant correlation between the river corridor buffer width and the PD index of cultivated land (p > 0.05). The landscape indices of cultivated land, namely ED, AI, and COHESION, were selected for regression analysis with the historical river corridor buffer width in 1967 (Figure 13). In the case of construction land, the landscape indices, comprising PD, ED, and COHESION, were selected for the corresponding calculations. These indices were used to analyze the relationship between construction land and the buffer width of the river corridor.
The results revealed a significant correlation between the landscape indices of cultivated land and construction land and the buffer width of the river corridor. Specifically, as the distance from the river increased, the impact on cultivated land and construction land weakened. Interestingly, an inverted “U” shape relationship was observed between the landscape indices of construction land and the buffer width of the river corridor, reaching a peak at 4–4.5 km. This suggests that before reaching the peak, the patch density and complexity of construction land increased, along with an enhancement in patch aggregation. Furthermore, the influence of the river corridor on cultivated land gradually diminished as the buffer width increased. As a result, it was observed that as the cultivated land moved farther away from the river, the complexity of the patches increased, resulting in a higher degree of patch aggregation.

5. Discussion

In 1967, the Yongding River witnessed the cessation of its historical flooding and mudslide disasters, resulting in a relatively undisturbed and ecologically thriving river corridor. During this period, the Yongding River exhibited robust hydrological connectivity, characterized by a high degree of aggregation and patch connectivity within its bottomland and grassland areas. Cultivated land and forest components displayed a relatively intricate structural arrangement. Analyzing the spatial evolution of landscape patterns in 1967, a trend of diminishing landscape fragmentation was observed as the buffer zone width increased. This observation aligns with previous investigations into the landscape pattern of river corridors [6]. However, the scenario transformed significantly by 1980, as illustrated in Figure 6, Figure 7 and Figure 8. The land cover composition within the study area indicated a notable reduction in the proportions of forest. A variety of land cover types were transferred to cultivated land, some to forest and construction land, and the dominant position of cultivated land was greatly enhanced. Much of this change can be attributed to the disruption of the river’s natural flow, resulting in discernible environmental degradation within the river corridor. Correlation analysis was used to screen out landscape indices that were more sensitive to the response to river breaks, and to analyze the spatial radiation range of the corridor effect of river corridors when the ecological conditions were good.
Evidently from the fitting outcomes portrayed in Figure 10, the influence intensity of the Yongding River corridor exhibited a gradual attenuation as the buffer zone’s width expanded. Notably, as the study region witnessed a progressive dominance of cultivated land and construction land, the extent of radiation range tended to stabilize, particularly around the 4 km mark. It is worth mentioning that the demographic expansion in this locale during 1967 was moderate, accompanied by an underdeveloped economy. Consequently, the potency of cultivated land and its associated transformations exerted substantial influence on the corridor effect of the Yongding River.
The landscape pattern was subjected to further analysis to examine the variations in different landscape types influenced by the radiation range of the river corridor. The study focused on the most significant changes observed in forest land, cultivated land, and construction land within the study area. Correlation analysis was performed between the buffer zone width and the landscape indices for the three landscape types, followed by curve fitting analysis to accurately determine the radiation range of the river corridor.
Although the inflection points of radiation range on forest land, cultivated land, and construction land were relatively distant from each other, they aligned closely with the inflection points observed at landscape level. Given that the study area is located within an urban setting, inflection points do not have a distinct spatial boundary but are primarily influenced by the peak impact of the river corridor on the surrounding area. Human activities and urban development are crucial factors influencing the radiation range.
During the 1960s, the downstream area of the Yongding River experienced perennially poor water quality due to increased water consumption in the upstream Guanting Reservoir and the implementation of downstream water diversion projects. A significant portion of the Yongding River’s water sources were allocated for industrial and agricultural purposes as well as residential use [51]. Consequently, the downstream section of the Yongding River remained in a perennial state of limited water availability throughout the 1960s. This alteration in water quantity directly influenced the landscape pattern of the river corridor and had an impact on the soil texture and vegetation types in the vicinity.
By conducting an analysis of the spatial radiation range of historical river corridors in 1967, it became possible to quantitatively measure the changes in the impact of river corridors on regional landscape pattern. These findings hold significant value for various applications. Firstly, they can aid in delineating areas that are likely to experience substantial impacts during future ecological restoration efforts of river corridors. This delineation helps prevent excessive artificial disturbance within these identified areas, ensuring the preservation and restoration of vegetation zones along the corridors. Moreover, the results of the study can facilitate an assessment of the impact of river corridors on distinct landscape types, such as forest, cultivated land, and construction land in the region. This assessment enables the development of precise and targeted conservation measures and enhancement strategies tailored to the specific characteristics of each landscape type.
In previous studies [6,52], the radiation range of the corridor effect is mainly reflected by comparing the changes of landscape indices in different buffer zones of the river corridor, and it is believed that the influence of river corridors on landscape pattern has been considered to diminish as the buffer width increases. A linear correlation between landscape indices and buffer width has often been assumed. However, solely analyzing the inflection points of landscape indices may overlook the methodological and statistical heterogeneity of river corridors. Moreover, attributing landscape indices that are insensitive to buffer changes to river corridor influence factors can lead to inaccuracies in assessing the spatial radiation range of river corridors. To address this issue, innovative research methodologies have been adopted in related fields. For instance, in the study of transportation corridors, Lingfei Weng et al. [47] utilized correlation analysis to identify landscape indices that are sensitive to railroad construction. This approach enabled the accurate assessment of the impact of the Sino-Lao railroad on the landscape pattern along its route. Similarly, Jie Li [52] employed correlation analysis to explore the relationship between urbanization levels and the evolution of river corridors in Zhengzhou. Similar studies on the radiation range of linear corridor effects and buffer changes have begun to consider the nonlinear relationship between landscape and buffers.
Building upon these advancements, this paper introduces innovative research methodologies. It fully considers the possibility of a nonlinear relationship between river corridors and landscape pattern, as well as the varying characteristics of river corridors that influence landscape indices. Through correlation analysis, landscape indices that are sensitive to river breaks and buffer width changes were identified. A binary curve fitting function was then established to analyze the impact on scale, thus enhancing the objectivity of the research findings. The identification of inflection points on the functional relationship is crucial for accurately assessing the spatial radiation range of river corridors. Existing studies on the Yongding River [53] and Longxi River [54] watersheds have consistently revealed a discernible trend: the progressive expansion of the buffer zone corresponding to a gradual reduction in landscape fragmentation. The outcomes of the landscape fragmentation analysis conducted in this study are similar to this conclusion. Building upon this foundation, the curve fitting outcomes concerning cultivated land, forest land, and construction land also further determined the degree of influence and radiation range of the Yongding River corridor on the landscape pattern (Figure 12 and Figure 13). In light of these findings, the methodology of employing correlation analysis and curve fitting to assess the nexus between the landscape indices and the river corridor’s buffer zone range holds considerable reliability. This approach serves as a potent means to gauge the spatial radiation range of the corridor effect. Furthermore, it is worth noting that this research methodology can be effectively extended to the exploration of other linear corridors akin to river corridors.
This study employs a novel research method to investigate the spatial radiation range of the historic river corridor of the Yongding River. Spearman correlation analysis is utilized to identify landscape indices that exhibit high sensitivity for functional analysis, ensuring the accuracy of the results. However, it is important to note that the spatial radiation range of the river corridor is influenced not only by landscape patterns but also by social factors. To enhance the research framework on the radiation range of river corridors, future studies should consider incorporating additional years of data and conducting correlation analysis with social drivers.

6. Conclusions

Based on land use data from the 10 km buffer zone along the Yongding River corridor in 1967, this study examined the characteristics of landscape pattern changes within the buffer zone, both at landscape level and class level. Additionally, by conducting correlation analysis with data from the 1980s, the study identified landscape indices that are sensitive to river disruptions and spatial scale. These indices were then curve-fitted with the buffer width in 1967 to investigate changes in landscape indices within the river corridor, considering both the overall landscape and specific landscape types.
(1)
After analyzing the characteristics of landscape pattern and the distribution of each landscape type within the river corridor in 1967, our findings reveal the following insights: at landscape level, the degree of landscape fragmentation and the complexity of the landscape shape experienced a gradual decrease, whereas landscape pattern aggregation and spreading experienced an increase, followed by a period of stabilization. The inflection point of the landscape pattern appeared at 3.5~4 km. The pivotal point signifies a transformation in the buffer zone’s landscape pattern, transitioning from a dense composition of diverse elements to a configuration predominantly dominated by well-connected cultivated land and forest post the inflection point. At class level, further analysis of the landscape indices revealed the inflection point of landscape pattern changed around 3 km. With the increase of buffer width, the most noticeable transformations were evident within cultivated land, forest land, and construction land. Notably, patch areas exhibited a gradual augmentation. It can be found that forest was more fragmented and dispersed, cultivated land increased in aggregation and consolidation, and construction land was more regular but relatively dispersed.
(2)
By conducting curve fitting of landscape indices within the study area, the investigation into the historical spatial radiation range revealed that, on the whole, the radiation range extended to approximately 4 km. This is because at landscape level, the peak of the river corridor’s influence on the landscape pattern index was at 4 km. Although the distribution of inflection points of the influence of upstream river corridor on the radiation range of forest, cultivated land, and construction land at class level was relatively scattered, it was not much different from that at 4 km. The peak influence of the landscape indices was around 3.5–4 km for forest and 4.5 km for construction land, while the landscape indices of cultivated land continued to weaken with the increase of the buffer zone, which was consistent with the trend of attenuation.
The results of the quadratic polynomial fit were higher than those of the linear fit, indicating that the landscape pattern showed different characteristics due to the influence of river corridors, and that the influence of river corridors did not change in a single linear fashion, but showed spatial heterogeneity in scale and type. Due to the attention to the methodological and statistical heterogeneity of river corridors, this paper provides more accurate results for the delineated historical spatial radiation range of river corridors, and provides data support for the implementation of ecological protection and restoration of river corridors in the urban section of the Yongding River plain.

Author Contributions

Conceptualization, R.Y.; methodology, R.Y.; software, R.Y.; data curation, J.S. and R.Z.; resources, R.Z. and J.S.; supervision, Z.L. and X.X.; visualization, R.Y. and W.K.; writing—original draft, R.Y.; writing—review and editing, R.Y. and W.K.; project administration, Z.L. and X.X. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Major Science and Technology Projects of China, grant number 2018ZX07101005.

Data Availability Statement

Restrictions apply to the availability of these data. Data were obtained from the China Institute of Water Resources and Hydropower Research and are available from Ruiying Yang with the permission of the China Institute of Water Resources and Hydropower Research.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Forman, R.T.T. Land Mosaics: The Ecology of Landscapes and Regions; Cambridge University Press: Cambridge, MA, USA, 1995; p. 246. [Google Scholar]
  2. Dong, Z. River Restoration; China Water & Power Press: Beijing, China, 2015; pp. 90–91. [Google Scholar]
  3. Xiao, D.N. Landscape Ecology of Wetlands in Bohai Delta; Science Press: Beijing, China, 2001; pp. 57–61. [Google Scholar]
  4. Miao, Y.; Dai, T.; Yang, X.; Song, J. Landscape fragmentation associated with the Qingzang Highway and its influencing factors—A comparison study on road sections and buffers. Geogr. Sustain. 2021, 2, 59–67. [Google Scholar] [CrossRef]
  5. Hao, S.N.; Dong, F.; Liu, X.B.; Guo, Y.G.; Su, L.B. Analysis on ecological risk of the evolution of land landscape pattern in the Nyang River basin. Res. Soil Water Conserv. 2023, 30, 378–383, 430. [Google Scholar]
  6. Zhou, H.; Xiao, D.; Zhou, K. Corridor effect of the spatial changes of landscape patterns in arid areas: A case study of the river corridor areas in the middle and lower reaches of Tarim River. Chin. Sci. Bull. 2006, 51, 82–91. [Google Scholar] [CrossRef]
  7. Peng, J.J.; Xu, L.P.; Cao, C. Spatio-temporal evolution of glacier landscape pattern in the Yarkant River basin from 1991 to 2017. Acta Ecol. Sin. 2022, 42, 8265–8275. [Google Scholar]
  8. Li, S.; He, W.; Wang, L.; Zhang, Z.; Chen, X.; Lei, T.; Wang, S.; Wang, Z. Optimization of landscape pattern in China Luojiang Xiaoxi basin based on landscape ecological risk assessment. Ecol. Indic. 2023, 146, 109887. [Google Scholar] [CrossRef]
  9. Jing, M.; Song, F.; Meng, K.; Su, F.; Wei, C. Optimization of landscape pattern in the main river basin of Liao River in China based on ecological network. Environ. Sci. Pollut. Res. 2023, 30, 65587–65601. [Google Scholar] [CrossRef]
  10. Liu, Z.; Gan, X.; Dai, W.; Huang, Y. Construction of an Ecological Security Pattern and the Evaluation of Corridor Priority Based on ESV and the “Importance–Connectivity” Index: A Case Study of Sichuan Province, China. Sustainability 2022, 14, 3985. [Google Scholar] [CrossRef]
  11. Yang, W.F.; Yu, K.Y.; Zhao, G.J.; Geng, J.W.; Zhao, Q.Y.; Yang, L.Q.; Liu, J. Optimization of greenways in Fuzhou based on heat island effect. J. Zhejiang A F Univ. 2022, 39, 876–883. [Google Scholar]
  12. Liu, H.N.; He, X.D.; Miao, S.G.; Yu, B.; Wei, L.H.; Wang, X.Y. A study on meteorological effect of the Hangzhou ventilation corridors based on high resolution numerical simulation. Clim. Environ. Res. 2019, 24, 22–36. [Google Scholar]
  13. He, L.; Liu, Y.; Wang, X.A. Research on the development of road integrated transportation corridor based on corridor effect. Highway 2021, 66, 270–276. [Google Scholar]
  14. Zhang, Y.H.; Zeng, Z.J. Corridor effect on urban land use in Panyu district, Guangzhou city. Remote Sens. Nat. Resour. 2014, 26, 157–162. [Google Scholar]
  15. Godfroy, J.; Lejot, J.; Demarchi, L.; Bizzi, S.; Michel, K.; Piégay, H. Combining Hyperspectral, LiDAR, and Forestry Data to Characterize Riparian Forests along Age and Hydrological Gradients. Remote Sens. 2022, 15, 17. [Google Scholar] [CrossRef]
  16. Wu, T.; Wang, L.; Liu, H. Spatiotemporal Differentiation of Land Surface Thermal Landscape in Yangtze River Delta Region, China. Sustainability 2021, 13, 8880. [Google Scholar] [CrossRef]
  17. Shi, Z.; Yang, J.; Zhang, Y.; Xiao, X.; Xia, J.C. Urban ventilation corridors and spatiotemporal divergence patterns of urban heat island intensity: A local climate zone perspective. Environ. Sci. Pollut. Res. 2022, 29, 74394–74406. [Google Scholar] [CrossRef]
  18. Ji, P.; Zhu, C.Y.; Gao, Y.F.; Li, S.H. Effects of different widths of greenbelt on the temperature and humidity in river corridor. Chin. Landsc. Archit. 2012, 28, 109–112. [Google Scholar]
  19. Li, H.F.; Li, Y.S.; Lu, Z.; Guo, K.; Peng, W.F. Analysis on the thermal environment effect of river corridor landscape. Geogr. Geo-Inf. Sci. 2015, 31, 51–54, 133. [Google Scholar]
  20. Jia, X.; Ran, Z.M.; Chen, S.J.; Yin, X.Y.; Chen, M.Y.; Huang, T.C. Analysis of the river corridor effect of the distribution of Apocheima cinerarius on Populus euphratica in the Yarkant River basin. Plant Prot. 2019, 45, 170–177. [Google Scholar]
  21. Lin, J.; Yang, W.; Yu, K.; Geng, J.; Liu, J. Construction of Water Corridors for Mitigation of Urban Heat Island Effect. Land 2023, 12, 308. [Google Scholar] [CrossRef]
  22. Yu, Q. Research on Riparian Green Space Planning of Mountainous City. Ph.D. Thesis, Chongqing University, Chongqing, China, 2019. [Google Scholar]
  23. Liu, K.X.; Wang, D.M.; Wei, Y.S.; Chang, G.L. Spatio-temporal evolution trend of multi-scale landscape ecological risk in miyun reservoir watershed. Acta Ecol. Sin. 2023, 43, 105–117. [Google Scholar]
  24. Cockburn, J.M.H.; Scott, A.; Villard, P.V. Evaluating Water and Carbon Retention in a Low-Order, Designed River Corridor. Land 2022, 11, 2256. [Google Scholar] [CrossRef]
  25. Anderson, M.J.; Ellingsen, K.E.; McArdle, B.H. Multivariate dispersion as a measure of beta diversity. Ecol. Lett. 2006, 9, 683–693. [Google Scholar] [CrossRef] [PubMed]
  26. Zhu, Q.; Yu, K.J.; Li, D.H. The width of ecological corridor in landscape planning. Acta Ecol. Sin. 2005, 25, 2406–2412. [Google Scholar]
  27. Buffer Zones Definition | Biodiversity A-Z. Available online: https://www.biodiversitya-z.org/content/buffer-zones (accessed on 24 December 2020).
  28. Liu, W. Research on the Recognition and Spatial Regulation Strategy of Urban Waterfront Buffer Zone: With the Case of Wuhan. Ph.D. Thesis, Huazhong University of Science and Technology, Wuhan, China, 2017. [Google Scholar]
  29. Li, H.; Xu, X.; Wu, M.; Liu, Z. Spatiotemporal Evolution Trajectory of Channel Morphology and Controlling Factors of Yongding River, Beijing, China. Water 2021, 13, 1489. [Google Scholar] [CrossRef]
  30. Hu, Z.L. Strategic implication of comprehensive governance and ecological restoration on the Yongding River. China Water Resour. 2019, 70, 16–18. [Google Scholar]
  31. Wu, M.; Wu, H.; Warner, A.T.; Li, H.; Liu, Z. Informing Environmental Flow Planning through Landscape Evolution Modeling in Heavily Modified Urban Rivers in China. Water 2021, 13, 3244. [Google Scholar] [CrossRef]
  32. Comprehensive Control and Ecological Restoration Plan of Yongding River. Available online: https://www.ndrc.gov.cn/fzggw/jgsj/njs/sjdt/201704/P020191101564948961022.pdf (accessed on 24 April 2017).
  33. Hu, Z.L. Comprehensive treatment and ecological restoration of the Yongding River: Practices and prospect. China Water Resour. 2022, 73, 14, 28–30. [Google Scholar]
  34. Budd, W.W.; Cohen, P.L.; Saunders, P.R.; Steiner, F.R. Stream corridor management in the Pacific Northwest: I. Determination of stream-corridor widths. Environ. Manag. 1987, 11, 587–597. [Google Scholar] [CrossRef]
  35. Zhang, L. On the Ecological Change of the Beijing Section of the Yongding River Basin: From the Republican Period to Present. Master’s Thesis, Beijing Forestry University, Beijing, China, 2015. [Google Scholar]
  36. Yin, J.K. Yongding River and Beijing in History; Beijing Yanshan Press: Beijing, China, 2008; p. 367. [Google Scholar]
  37. Zheng, Z.J. Water History of China; The Commercial Press: Beijing, China, 1993; p. 183. [Google Scholar]
  38. He, Y.H.; Xie, J.Q.; Sun, Y. FAO/UNEP-land cover classification system (LCCS) and use for reference. China Land Sci. 2005, 19, 45–49. [Google Scholar]
  39. Liu, Y.H.; Niu, Z.; Xu, Y.M.; Wang, C.Y.; Li, G.C. Design of land cover classification system for China and its application research based on MODIS data. Trans. Chin. Soc. Agric. Eng. 2006, 22, 99–104, 240. [Google Scholar]
  40. Ouyang, Z.Y.; Zhang, L.; Wu, B.F.; Li, X.S.; Xu, W.H.; Xiao, Y.; Zheng, H. An ecosystem classification system based on remote sensor information in China. Acta Ecol. Sin. 2015, 35, 219–226. [Google Scholar]
  41. Wu, L.Y.; He, D.J.; You, W.B.; Ji, Z.R.; Huang, X.Y. A gradient analysis of coastal landscape fragmentation change in Dongshan island, China. Acta Ecol. Sin. 2020, 40, 1055–1064. [Google Scholar]
  42. Wei, S.K.; Xu, W.W.; Zhang, Z.D.; Huang, X.R. Spatiotemporal scale effect of landscape pattern for natural vegetation in a forest-steppe zone of Hebei. Ecol. Sci. 2022, 41, 157–166. [Google Scholar]
  43. Bi, X.L.; Zhou, R.; Liu, L.J.; Hong, J.; Wang, X.Z.; Wang, H.; Suo, A.N.; Ge, J.P. Gradient variations in landscape pattern along the Jinghe River and their driving forces. Acta Ecol. Sin. 2005, 25, 1041–1047, 1238. [Google Scholar]
  44. Wu, J.G. Landscape Ecology: Pattern, Process, Scale and Hierarchy; Higher Education Press: Beijing, China, 2007; pp. 112–115. [Google Scholar]
  45. Zhou, H.R. Landscape Ecology of River Corridor in Arid Region: A Case Study of the Middle and Lower Reaches of Tarim River in Xinjiang; Science Press: Beijing, China, 2007; pp. 95–97. [Google Scholar]
  46. Liu, T.D.; Xu, D.W. Case Study of Landscape Ecology: Assessment of River Landscape Pattern and Ecological Vulnerability; Science Press: Beijing, China, 2015; pp. 83–88. [Google Scholar]
  47. Weng, L.; Bai, H.; Shen, L. Scaling impacts of China-Laos Railway construction on landscape patterns. Resour. Sci. 2021, 43, 2451–2464. [Google Scholar] [CrossRef]
  48. Xiong, Z.; Song, Q. Spss Data Analysis and Application; Heilongjiang Science & Technology Press: Harbin, China, 2022; p. 250. [Google Scholar]
  49. Jia, N. Study on Forest Color Quantification and Ornamental Effect. Ph.D. Thesis, Inner Mongolia Agricultural University, Hohhot, China, 2021. [Google Scholar]
  50. Zhang, K.B. Study on reconstruction and utilization of sandy land along Yongding River. J. Beijing For. Univ. 1996, 18, 15–21. [Google Scholar]
  51. Yu, B.P.; Guo, R.Q. Soil and water loss and conservation in Yongding River basin. Soil Water Conserv. China 1982, 3, 43–46. [Google Scholar]
  52. Li, J. Research on Temporal-Spatial Evolution and Ecological Restoration of River Landscape Corridor in Zhengzhou under the Background of Urbanization. Ph.D. Thesis, Henan Agricultural University, Zhengzhou, China, 2021. [Google Scholar]
  53. Cao, L.; Lang, Q.; Lei, K.; Wang, D.; Yang, k.; Yang, W. Analysis on landscape pattern dynamics and driving force in Yongding River basin from 1980 to 2020. J. Environ. Eng. Technol. 2023, 13, 143–153. [Google Scholar]
  54. Song, J. Study on the Relationship between Landscape Pattern and River Water Quality: A Case Study of the Longxi River Basin in Chongqing. Master’s Thesis, Southwest University, Chongqing, China, 2023. [Google Scholar]
Land 12 01656 g001
Figure 1. The study area is in the southwestern part of the central city of Beijing. (a) Location in China and (b) location in Beijing.
Figure 1. The study area is in the southwestern part of the central city of Beijing. (a) Location in China and (b) location in Beijing.
Land 12 01656 g001
Land 12 01656 g002
Figure 2. The detailed workflow of the study.
Figure 2. The detailed workflow of the study.
Land 12 01656 g002
Land 12 01656 g003
Figure 3. Landsat 8 image and visual interpretation. Schemes follow another format. (a) Landsat 8 image in 1967; (b) land cover map in 1967; (c) Landsat 8 image in 1980; (d) land cover map in 1980.
Figure 3. Landsat 8 image and visual interpretation. Schemes follow another format. (a) Landsat 8 image in 1967; (b) land cover map in 1967; (c) Landsat 8 image in 1980; (d) land cover map in 1980.
Land 12 01656 g003
Land 12 01656 g004
Figure 4. Buffer zone width delineation.
Figure 4. Buffer zone width delineation.
Land 12 01656 g004
Land 12 01656 g005
Figure 5. Changes of landscape indices at landscape level in river corridor buffer zone in 1967.
Figure 5. Changes of landscape indices at landscape level in river corridor buffer zone in 1967.
Land 12 01656 g005
Land 12 01656 g006
Figure 6. Changes of percentage of landscape (PLAND) in river corridor buffer zone in 1967.
Figure 6. Changes of percentage of landscape (PLAND) in river corridor buffer zone in 1967.
Land 12 01656 g006
Land 12 01656 g007
Figure 7. Changes of percentage of landscape (PLAND) in river corridor buffer zone in 1967 and 1980.
Figure 7. Changes of percentage of landscape (PLAND) in river corridor buffer zone in 1967 and 1980.
Land 12 01656 g007
Land 12 01656 g008
Figure 8. Land use transfers in 1967 and 1980.
Figure 8. Land use transfers in 1967 and 1980.
Land 12 01656 g008
Land 12 01656 g009
Figure 9. Changes of landscape indices at class level in river corridor buffer zone in 1967.
Figure 9. Changes of landscape indices at class level in river corridor buffer zone in 1967.
Land 12 01656 g009
Land 12 01656 g010
Figure 10. Relationship between landscape indices of river corridor and buffer zone width in 1967.
Figure 10. Relationship between landscape indices of river corridor and buffer zone width in 1967.
Land 12 01656 g010
Land 12 01656 g011
Figure 11. Correlation of landscape indices at landscape level and buffer width within the buffer zone in 1967 and 1980. (a) Comparison of the correlation results of forest land in 1967 and 1980; (b) comparison of the correlation results of cultivated land in 1967 and 1980; (c) comparison of the correlation results of construction land in 1967 and 1980.
Figure 11. Correlation of landscape indices at landscape level and buffer width within the buffer zone in 1967 and 1980. (a) Comparison of the correlation results of forest land in 1967 and 1980; (b) comparison of the correlation results of cultivated land in 1967 and 1980; (c) comparison of the correlation results of construction land in 1967 and 1980.
Land 12 01656 g011
Land 12 01656 g012
Figure 12. Relationship between landscape indices and buffer width for forest land in 1967.
Figure 12. Relationship between landscape indices and buffer width for forest land in 1967.
Land 12 01656 g012
Land 12 01656 g013
Figure 13. Relationship between landscape indices and buffer zone width for cultivated land and construction land in 1967.
Figure 13. Relationship between landscape indices and buffer zone width for cultivated land and construction land in 1967.
Land 12 01656 g013
Table 1. List of indices of landscape pattern.
Table 1. List of indices of landscape pattern.
Landscape Pattern IndexIndex Calculation FormulaSignification of VariablesDescription [2,45,46,47]
Patch DensityPD = n i / A n i is the number of patches in the landscape of patch type (class), A   is the total landscape area (m2).PD reflects the degree of landscape fragmentation and spatial distribution differences of patches.
Edge DensityED = ∑ k = 1 m e i k / A e i k is the total length (m) of edge in landscape involving patch type (class) i ; includes landscape boundary and background segments involving patch type i ; m is the number of patch types (classes) present in the landscape.ED reflects the complexity of landscape shape; the magnitude of edge density directly affects edge effects and species composition.
Aggregation IndexAI = g i i / max → g i i 100 g i i is the number of like adjacencies (joins) between pixels of patch type (class) i based on the single-count method, max → g i i is the maximum number of like adjacencies (joins) between pixels of patch type (class) i (see below) based on the single-count method.AI indicates the degree of spatial aggregation or complexity of landscape shapes, which can explain the possible maximum proximity of the landscape components.
Largest Patch IndexLPI = max a i j / A 100 a i j is the area (m2) of patch i j .LPI reflects ecological characteristics such as the abundance of dominant and internal species in the landscape, as well as the direction and strength of human activities.
Contagion IndexCONTAG = 1 + ∑ i = 1 m ∑ k = 1 m P i g i k ∑ k = 1 m g i k ln P i g i k ∑ k = 1 m g i k 2 ln m 100 P i is the proportion of the landscape occupied by patch type (class) I, g i k is the number of adjacencies (joins) between pixels of patch types (classes) i and k based on the double-count method.CONTAG describes the degree of aggregation or extension trend of different patch types in a landscape; this indicator contains spatial information and is one of the most important indicators for describing the landscape pattern.
Shannon’s Diversity IndexSHDI = − ∑ i = 1 m P i ln P i dittoSHDI reflects landscape heterogeneity, particularly sensitive to the uneven distribution of various patch types in the landscape.
Patch Cohesion IndexCOHESION = 1 − ∑ j = 1 n P i j ∑ j = 1 n P i j a i j 1 − 1 Z − 1 100 P i j is the perimeter of patch ij in terms of number of cell surfaces, Z is the total number of cells in the landscape.COHESION measures the physical connectedness of the corresponding patch type.
Percentage of LandscapePLAND = P i = ∑ j = 1 n a i j / A 100 n is the number of patches.PLAND is one of the bases for determining the matrix or dominant landscape elements in a landscape and is an important factor in determining ecosystem indicators such as biodiversity, dominant species, and quantity in the landscape.
Table 2. Normality test of landscape indices at landscape level in 1967 and 1980.
Table 2. Normality test of landscape indices at landscape level in 1967 and 1980.
Shapiro–Wilk in 1967Shapiro–Wilk in 1980
StatisticdfpStatisticdfp
PD0.690100.0010.720100.002
ED0.823100.0280.634100.000
LPI0.772100.0070.903100.236
CONTAG0.920100.3540.585100.349
COHESION0.930100.4520.919100.138
AI0.780100.0080.882100.000
SHDI0.921100.3640.918100.337
Table 3. Correlation of landscape indices at landscape level and buffer width.
Table 3. Correlation of landscape indices at landscape level and buffer width.
VariableYearSignificance IndexBuffer WidthPDLPIEDAICONTAGCOHESIONSHDI
Buffer width1967Spearman Correlation
Sig.		1−0.685 *
0.029		0.733 *
0.016		−0.915 **
0.000		0.915 **
0.000		1.000 **
0.000		1.000 **
0.000		−1.000 **
0.000
Buffer width1980Spearman Correlation
Sig.		1−0.200
0.580		0.939 **
0.000		−0.273
0.446		0.648 *
0.043		1.000 **
0.000		1.000 **
0.000		−1.000 **
0.000
* indicates significant correlation at 0.05 level, ** indicates significant correlation at 0.01 level.
Table 4. Results of fitting landscape indices to the buffer width in 1967.
Table 4. Results of fitting landscape indices to the buffer width in 1967.
Landscape Pattern IndexRegression ModelR2Fp
PDy = 16.881 − 2.21x + 0.312x20.89630.106<0.001
EDy = 131.148 − 7.366x + 0.872x20.971117.324<0.001
AIy = 96.543 + 0.283x − 0.036x20.96083.610<0.001
Table 5. Results of fitting landscape indices to the buffer width in 1967 for the three land types.
Table 5. Results of fitting landscape indices to the buffer width in 1967 for the three land types.
Dependent VariableLand TypeRegression ModelR2Fp
PDForesty = 5.823 + 0.243x + 0.002x20.977151.462<0.001
Cultivated landy = 2.548 + 0.302x − 0.043x20.5283.9100.072
Construction landy = 0.76 + 0.476x − 0.066x20.92241.266<0.001
EDForesty = 59.019 + 13.664x − 1.617x20.95778.359<0.001
Cultivated landy = 38.385 + 23.117x − 2.563x20.988282.808<0.001
Construction landy = 5.913 + 6.813x − 0.868x20.978156.999<0.001
AIForesty = 95.724 + 0.431x − 0.067x20.88526.912<0.001
Cultivated landy = 96.332 + 0.606x − 0.075x20.95472.877<0.001
Construction landy = 97.288 + 0.188x − 0.022x20.86121.7050.01
COHESIONForesty = 99.002 + 0.381x − 0.054x20.83317.4040.002
Cultivated landy = 99.240 + 0.270x − 0.026x20.9971384.910<0.001
Construction landy = 98.323 + 0.242x − 0.027x20.89931.225<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

Yang, R.; Liu, Z.; Kang, W.; Su, J.; Zhang, R.; Xu, X. Historical Spatial Radiation Range of the Yongding River Corridor in Beijing Plain Section: Implications for Landscape Patterns and Ecological Restoration. Land 2023, 12, 1656. https://doi.org/10.3390/land12091656

AMA Style

Yang R, Liu Z, Kang W, Su J, Zhang R, Xu X. Historical Spatial Radiation Range of the Yongding River Corridor in Beijing Plain Section: Implications for Landscape Patterns and Ecological Restoration. Land. 2023; 12(9):1656. https://doi.org/10.3390/land12091656

Chicago/Turabian Style

Yang, Ruiying, Zhicheng Liu, Wenxin Kang, Junyi Su, Renfei Zhang, and Xiaoming Xu. 2023. "Historical Spatial Radiation Range of the Yongding River Corridor in Beijing Plain Section: Implications for Landscape Patterns and Ecological Restoration" Land 12, no. 9: 1656. https://doi.org/10.3390/land12091656

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