[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Identification of Typhoon-Vulnerable Areas and Countermeasures in High-Density Coastal Cities: The Case of Macau
Previous Article in Journal
Generating Urban Road Networks with Conditional Diffusion Models
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJGI
Volume 13
Issue 6
10.3390/ijgi13060204
ijgi-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

Spatiotemporal Analysis of Ecological Security Based on Landscape Patterns

by
Huaidan Zhang
1,2,3,
Ke Nie
1,4 and
Xueling Wu
2,*
1
Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources, Shenzhen 518034, China
2
School of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, China
3
Social Application Node, Hubei, Natural Resources Satellite Application Technology Center, Wuhan 430050, China
4
Shenzhen Data Management Center of Planning and Natural Resources (Shenzhen Geospatial Information Center), Shenzhen 518040, China
*
Author to whom correspondence should be addressed.
ISPRS Int. J. Geo-Inf. 2024, 13(6), 204; https://doi.org/10.3390/ijgi13060204
Submission received: 25 March 2024 / Revised: 6 June 2024 / Accepted: 9 June 2024 / Published: 16 June 2024
Browse Figures
Figure 1
<p>Location of the study area in China. (<b>a</b>) Map of China. (<b>b</b>) Map of Henan Province. (<b>c</b>) Map of Song County. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 2
<p>A schematic flow chart of dynamic evaluation of ecological security. This figure was created using Visio (<a href="https://www.microsoft.com/" target="_blank">https://www.microsoft.com/</a>, accessed on 1 January 2019.).</p> "> Figure 3
<p>Evaluation index of ecological security evaluation in 2020. (<b>a</b>) Geological disaster susceptibility. (<b>b</b>) Population density. (<b>c</b>) Landscape fragmentation index. (<b>d</b>) Biological abundance index. (<b>e</b>) Water conservation index. (<b>f</b>) Vegetation coverage index. (<b>g</b>) Landscape disturbance index. (<b>h</b>) Landscape restoration index. (<b>i</b>) GDP density. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 3 Cont.
<p>Evaluation index of ecological security evaluation in 2020. (<b>a</b>) Geological disaster susceptibility. (<b>b</b>) Population density. (<b>c</b>) Landscape fragmentation index. (<b>d</b>) Biological abundance index. (<b>e</b>) Water conservation index. (<b>f</b>) Vegetation coverage index. (<b>g</b>) Landscape disturbance index. (<b>h</b>) Landscape restoration index. (<b>i</b>) GDP density. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 3 Cont.
<p>Evaluation index of ecological security evaluation in 2020. (<b>a</b>) Geological disaster susceptibility. (<b>b</b>) Population density. (<b>c</b>) Landscape fragmentation index. (<b>d</b>) Biological abundance index. (<b>e</b>) Water conservation index. (<b>f</b>) Vegetation coverage index. (<b>g</b>) Landscape disturbance index. (<b>h</b>) Landscape restoration index. (<b>i</b>) GDP density. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 4
<p>Evaluation indicators of geological disaster susceptibility in 2020. (<b>a</b>) Elevation. (<b>b</b>) Slope. (<b>c</b>) Aspect. (<b>d</b>) Relief of topography. (<b>e</b>) Engineering rock formation. (<b>f</b>) Distance from structure. (<b>g</b>) Landcover. (<b>h</b>)NDVI. (<b>i</b>)NDWI. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 4 Cont.
<p>Evaluation indicators of geological disaster susceptibility in 2020. (<b>a</b>) Elevation. (<b>b</b>) Slope. (<b>c</b>) Aspect. (<b>d</b>) Relief of topography. (<b>e</b>) Engineering rock formation. (<b>f</b>) Distance from structure. (<b>g</b>) Landcover. (<b>h</b>)NDVI. (<b>i</b>)NDWI. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 4 Cont.
<p>Evaluation indicators of geological disaster susceptibility in 2020. (<b>a</b>) Elevation. (<b>b</b>) Slope. (<b>c</b>) Aspect. (<b>d</b>) Relief of topography. (<b>e</b>) Engineering rock formation. (<b>f</b>) Distance from structure. (<b>g</b>) Landcover. (<b>h</b>)NDVI. (<b>i</b>)NDWI. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 5
<p>Classification results of landcover. (<b>a</b>) Landcover in 2005. (<b>b</b>) Landcover in 2010. (<b>c</b>) Landcover in 2015. (<b>d</b>) Landcover in 2020. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 6
<p>Change in patch type level index. (<b>a</b>) Number of patches (NP). (<b>b</b>) Edge density (ED). (<b>c</b>) Landscape type proportion (PLAND). (<b>d</b>) Maximum patch index (LPI).</p> "> Figure 6 Cont.
<p>Change in patch type level index. (<b>a</b>) Number of patches (NP). (<b>b</b>) Edge density (ED). (<b>c</b>) Landscape type proportion (PLAND). (<b>d</b>) Maximum patch index (LPI).</p> "> Figure 7
<p>Evaluation level of ecological security. (<b>a</b>) Ecological security level in 2005. (<b>b</b>) Ecological security level in 2010. (<b>c</b>) Ecological security level in 2015. (<b>d</b>) Ecological security level in 2020. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 8
<p>Moran’s I statistics of ecological security level in 2005, 2010, 2015, and 2020.</p> "> Figure 9
<p>LISA result of spatial autocorrelation. (<b>a</b>) 2005. (<b>b</b>) 2010. (<b>c</b>) 2015. (<b>d</b>) 2020. The figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> "> Figure 10
<p>Distribution map of cold/hotspots. (<b>a</b>) 2005. (<b>b</b>) 2010. (<b>c</b>) 2015. (<b>d</b>) 2020. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 Jun1 2021).</p> "> Figure 11
<p>Changing trend of ecological security. (<b>a</b>) Change slope. (<b>b</b>) Results of F significance test. This figure was created using ArcGIS ver. 10.4 (<a href="https://www.esri.com/" target="_blank">https://www.esri.com/</a>, accessed on 1 June 2021).</p> ">
Review Reports Versions Notes

Abstract

:
With rapid urbanization, environmental problems such as soil erosion and resource shortages have emerged. Ecological environmental quality is decreasing, and ecological security issues are becoming increasingly prominent; thus, relevant research is particularly urgent. The ecological security issue is complex due to many influencing factors. The transformation of landscape type is the most important factor affecting ecological security. Therefore, there is an urgent need to optimize and screen for the indicator factors that affect ecological security, carry out a dynamic evaluation of ecological security based on landscape pattern analysis, and analyze the driving forces behind ecological security changes. Song County is located in the ecological core area of the Funiu Mountains in western Henan, with complex topography and geomorphology; large changes in landscape patterns in recent years; frequent geological disasters, which have posed a greater threat to people’s life and property safety; and significant ecological security problems. This paper takes Song County as the research area, using the decision tree model to obtain the land use classification results of four periods in Song County in 2005, 2010, 2015, and 2020 based on remote sensing images. Landscape pattern analysis is conducted from two aspects: patch level and landscape level. On this basis, ecological security evaluation indicators are constructed from three levels: pressure, state, and response, and the comprehensive index model is used to obtain the results of four ecological security evaluations. Exploratory spatial data analysis (ESDA) is used to conduct research and prediction on spatiotemporal differentiation. Finally, the spatial heterogeneity relationship between the ecological security level and its driving factors in Song County is quantitatively analyzed using a geographic detector model. The results clearly show that the overall landscape form gradually tends to develop in the direction of complex irregularity. Due to frequent geological disasters and strong human engineering activities near the core areas of the Luhun Reservoir and Yi River basin, as well as Baihejie Village in Baihe Township and Che Village in Muzhijie Township, the landscape pattern is changing considerably. The self-restoration ability of the land’s ecosystem is gradually weakening, and the degree of ecological damage is gradually accelerating. The ecological security level is unsafe, the area of unsafe security is gradually increasing, and the ecological security index (ESI) will continue to decrease in the future. To improve ecological security, we recommend paying attention to land conservation and rational utilization while pursuing economic development.

1. Introduction

The term “ecological security” was originally proposed to be closely related to national security. Ecological security research has become one of the most important research fields in geography, resource and environmental sciences, and ecology [1,2]. The study of static and dynamic changes in ecological security is of practical significance, as both spatial and temporal changes have important effects on ecosystems. As a result of the improvement in living standards [3], the continual expansion of human needs, and the increasing intensity of development [4], the ecological environment quality in China is declining, and the intensification of its destruction has affected the stability of society [5,6].
Landscape ecology is a comprehensive discipline that studies the type composition, spatial pattern, and interaction with ecological processes of landscape units. It is aimed at understanding the stability and diversity of ecosystems, biodiversity protection, land use planning, resource management, and ecological restoration, and provides theoretical support and practical guidance for solving many ecological and environmental problems faced by humans. The idea was proposed by the German botanist Troll in the late 1940s, focusing on the study of the associations between different landscape units of nature and host organisms within a region [7]. Landscape genetics [8], urban landscape ecology [9], landscape modelling [10], landscape services [11], landscape sustainability, and landscape resilience [12] are hot topics in international landscape ecology research. The landscape pattern changes and driving mechanisms in special ecological function areas, as well as ecological security features, have become the focus of research.
Landscape patterns are a concrete manifestation of landscape heterogeneity and are one of the core contents of landscape ecology research. Many studies have shown that the formation and changes in aspect are not only closely related to ecological processes such as soil nutrients, soil erosion, biodiversity, and soil moisture [13,14] but also influenced by human activities. Landscape index analysis is one of the commonly used methods for landscape pattern analysis; “3S” technology has been applied in landscape pattern research by an increasing number of scholars, and it has become a common practice to use geospatial information technology such as GIS and RS to acquire spatial information on landscape metrics [15].
Ecological security evaluation refers to distinguishing the degree of ecological security using a series of evaluation indicators. Scholars at home and abroad have made certain achievements in ecological security evaluation. Taking the Dongliao River basin as the object of analysis, Zhang M. et al. explored delineation ecological security evaluation methods, providing a reference and basis for exploration of the ecological regulation of watersheds in an optimal way [16]. The fuzzy evaluation model was established to summarize the scores of the synthesis evaluation of the ecological security level of land resources. The evaluation results were verified by comparing ecological effect indicators such as the landscape pattern index and net primary productivity [17].
Building an effective scientific evaluation index system is the key to quantifying ecological security. The PSR model has been affirmed and applied by many scholars because of its clear ecological causality. It is used mainly to analyze and study significant indicators concerning the eco-environment [18]. The PSR model initially referred to the pressure response framework proposed by Canadian statisticians Frick and Schmidt in 1975, and later developed and applied by the United Nations Organization for Economic Cooperation and Development (OECD) and the United Nations Environment Program in the 1980s and 1990s for ecological evaluation [19]. It further developed into a universal pressure–state–response model. It is a dynamic model structure that includes three dimensions, pressure, state, and response, which can reflect the dynamic evolution process and internal logic of events. It provides important ideas for evaluating ecological health levels and sustainable development capabilities. Indicator systems for urban ecological security [20], watershed landscape ecology security [21], water environment ecological security [22], and other security assessments have successfully been constructed on the basis of the PSR model. Thus, PSR is a commonly used scientific model for the construction of an index system because of its wide range of applications and comprehensive nature.
The ecological environment of Song County is fragile and sensitive, so an ecological security strategy is especially necessary.

2. Study Area

Song County is located in western Henan Province, bordering Ruyang and Lushan in the east, Nanzhao and Neixiang in the south, Luanchuan and Luoning in the west, and Yiyang and Yichuan in the north. The site covers a region of approximately 3008.9 km2 (111°24′~112°22′ E, 33°35′~34°21′ N). The terrain of Song County is complex and gradually rises from northeast to southwest. There are mountains, hills, and other landforms in the territory. Luhun Reservoir, the second largest reservoir in Henan Province, is located near Luhun Town and the Yihe River, which is the secondary tributary of the Yellow River. The study area has complex geological environmental conditions, developed fault structures, strong human engineering activities, and serious damage sustained to the natural environment. It is a high-risk area for landslides, collapses, debris flows, and other geological disasters and has the characteristics of wide distribution, relative concentration, and poor stability. In addition, in recent years, changes in land use have led to changes in landscape patterns, which have seriously threatened the ecological security of this area (Figure 1).

3. Data and Methods

This paper takes the comprehensive landscape pattern index to study the diversity, fragmentation, and heterogeneity of the landscape pattern in Song County, taking the natural, social, and economic factors of regional ecological security evaluation into full consideration, to focus on the dynamic changes in ecological security in Song County during 2005–2020 on the basis of data collected in different periods. The results are used to provide constructive suggestions for the improvement of the ecological environment in a targeted manner. The technical roadmap of this paper is shown in Figure 2.

3.1. Data Sources

The data used in this paper are as follows: two scenes of Landsat8 OLI remote sensing images (path/row 125/36, 125/37) obtained in 2020 and 2015 have 9 bands, and two scenes of Landsat5 TM remote sensing images (path/row 128/40, 128/39) obtained in 2010 and 2005 have 7 bands, which were used to extract landcover information, the normalized difference water index (NDWI), and the normalized difference vegetation index (NDVI). Four 30-m-resolution digital elevation models (DEMs) from the Geospatial Data Cloud (www.gscloud.cn/, accessed on 30 July 2021) covering the study area were used to extract factors such as slope, aspect, and relief of topography. The strip number is from E111 to E112, and the line number is from N33 to N34. The interpolation data of population and economy were obtained from the Resource and Environment Science and Data Center (www.resdc.cn/, accessed on 1 April 2021). GDP density and population density are spatial grid data generated through spatial interpolation based on statistical data from districts and counties, taking into account factors closely related to them. Landcover information was from the third national land resource survey, which was the basis of landcover classification in various periods. A geological map was retrieved from the National Geological Archives and was used to obtain information such as faults and engineering rock formations (Table 1).

3.2. Selection of Landscape Pattern Index

Landscape pattern indices are used to describe landscape pattern information and reflect the spatial and temporal evolution characteristics and laws of landscape patterns in a research area [23,24]. In this article, the number of patches (NP), edge density (ED), percent of landscape (PLAND), and largest patch index (LPI) were selected to analyze the degree of landscape damage and dispersion at the type level. At the landscape level, the number of patches (NP), patch density (PD), contagion index (CONTAG), landscape shape index (LSI), interspersion and juxtaposition index (IJL), Shannon’s diversity index (SHDI), and Shannon’s evenness index (SHEI) were selected (Table 2).

3.3. Selection and Extraction of Evaluation Factors

This paper selected population density, the landscape fragmentation index, geological disaster susceptibility, the biological abundance index, the water conservation index, GDP density, vegetation coverage, the landscape disturbance index, and the landscape restoration index to construct an ecological security evaluation index system. Because different types of data have different dimensions and meanings, it was necessary to standardize the datasets. As the factors in different periods are treated in the same way, the following ecological security evaluation indicators in 2020 were taken as an example (Table 3, Figure 3).
(1) Pressure indicators
Population density refers to the total population per unit area in a region, and it is adversely correlated with the level of ecological security. In this paper, the population density distribution map with 30 m × 30 m resolution in the study area was obtained by converting and interpolating the obtained population density kilometer grid data.
Landscape fragmentation indicates the fragmentation degree of each ecological landscape, reflecting the impact of external disturbances on the landscape pattern such as human activities. In this paper, PD was selected as the landscape fragmentation index.
Geological hazards pose a serious threat to human life, wealth, ecological resources, and environmental safety [25,26,27,28]. Geological disaster susceptibility is affected mainly by topography, stratum lithology, geological structure, vegetation, and other factors, which are evaluated according to the activity rules and formation factors of geological hazards. On the basis of the reference of basic data and the related literature about the research area [29], this article selected the evaluation indicators and tested the correlation of influencing factors by multiple collinearity analysis, on which the evaluation index system of geological hazard susceptibility in the research area was constructed. The following geological disaster susceptibility evaluation indicators in 2020 were taken as an example (Table 4, Figure 4).
(2) State indicators
The biological abundance index is the variation in the number of biological species per unit area for different ecosystem types, and this value is an indirect reflection of the richness or lack of organisms in the study area (Equation (1)).
B I O = A b i o ∗ ( 0.35 ∗ A r e a   o f   f o r e s t + 0.21 ∗ A r e a   o f   g r a s s l a n d + 0.11 ∗ A r e a   o f   f a r m l a n d + 0.04 ∗   A r e a   o f   c o n s t r u c t i o n + 0.28 ∗ A r e a   o f   w a t e r + 0.01 ∗ A r e a   o f   u n u s e d ) / a r e a
where A b i o is equal to 400.02, which is the normalization coefficient of the biological abundance index.
The water conservation index is used to evaluate the water conservation function of regional ecosystems (Equation (2)).
C O N = A c o n ∗ ( 0.35 ∗ A r e a   o f   f o r e s t + 0.20 ∗ A r e a   o f   g r a s s l a n d + 0.45 ∗ A r e a   o f   w a t e r / a r e a )
where A c o n is equal to 222.2, which is the normalization coefficient of the water conservation index.
Vegetation coverage refers to the vertical projection of vegetation on a unit area, which is a comprehensive quantitative index reflecting information on vegetation growth and evolution under people’s activities and change in climate. The vegetation coverage of the study area was obtained by pixel dichotomy (Equations (3) and (4)).
N D V I = ρ N I R − ρ R e d ρ N I R + ρ R e d
V F C = N D V I − N D V I s o i l N D V I v e g − N D V I s o i l
where N D V I is the normalized difference vegetation index and V F C is the vegetation coverage index. N D V I s o i l is the bare soil or unvegetated area value and is regarded as the minimum N D V I value in the area, while N D V I v e g is the pixel value completely vegetated and is regarded as the maximum N D V I value in the area. To avoid the interference of image noise, this paper used the maximum and minimum N D V I values in the 90% confidence interval as N D V I v e g and N D V I s o i l according to experience.
(3) Response indicators
GDP density indicates the level of regional economic strength. Excessive economic growth will cause problems such as resource shortages and environmental deterioration, which will eventually constitute a risk to the ecological security of this area. In this paper, through the conversion and interpolation of the obtained grid data of GDP density kilometers, a distribution map of GDP density with a 30 m × 30 m resolution in the study area was obtained.
The landscape disturbance index can reflect the changes in the ecological environment of all kinds of landscape types under the action of human activities or natural stress to a certain extent, which is directly related to regional ecological security. In this paper, the LSI was selected to reflect the landscape disturbance degree, which depended on the disturbance factors and the landscape change trend.
The landscape restoration index is positively correlated with the variety of structures and functions of the landscape system. The more varied the landscape type and the more complex the structure, the more resistant to disturbance and the more resilient it will be. This paper used the patch richness density index (PRD), Shannon’s diversity index (SHDI), and Shannon’s evenness index (SHEI) to construct the landscape restoration index (LRI) (Equation (5)).
L R I = P R D ∗ S H D I ∗ S H E I

3.4. Methods

3.4.1. Decision Tree Model

A classification decision tree is a nonparametric hierarchical supervised classification method. It is a decision analysis method based on the known probability of various scenarios occurring. A decision tree is constructed to determine the probability of the expected value of the NPV being greater than a set threshold, and to judge its belonging category. The algorithm is based on information theory, which transforms complex and abstract information into understandable judgement [30].

3.4.2. CRITIC Weighting Method

CRITIC is a method of weighting that measures weight according to a criterion of interlayer correlation. This method combines the intensity of comparison within the index and the conflict between the indexes to objectively weight the indexes; this method is more reasonable than the subjective weighting method, coefficient of variation method, or correlation coefficient method, and the weighting result is more objective [31].

3.4.3. Exploratory Spatial Data Analysis (ESDA)

Spatial data analysis is widely defined as having three key parts: exploratory spatial data analysis (ESDA), visualization, and spatial modelling. ESDA focuses on the exploration of interesting “patterns” [32]. ESDA can be used for preprocessing spatial analysis data, including sorting data, searching the original hypothesis, determining distribution patterns, choosing variables, and selecting models. This paper has studied the spatiotemporal differentiation in ecological security evaluation grades. Spatiotemporal clustering is an exploratory analysis that is mainly divided into global and local levels [33].
(1) Global Moran’s I
Global spatial autocorrelation is often adopted to analyze the spatial correlation of a geographical object or a geographical attribute as a whole [34]. In this article, the global Moran’s I statistic index was used for quantifying the spatial correlation and difference level of the regional population (Equation (6)).
M o r a n ’ s   I = n ∑ i = 1 n ∑ j = 1 n ( x i − x ¯ ) ( x j − x ¯ ) ∑ i = 1 n ∑ j = 1 n W i j ∑ i = 1 n ( x i − x ¯ )
where n refers to the whole number of grids in the research area, x i refers to the ESI of evaluation unit i , x ¯ refers to the average value of the regional security index, and W i j is the binary adjacent space weight matrix. When evaluation units i and j are adjacent, W i j = 1 ; when they are not adjacent, W i j = 0 . The Moran’s I value is usually from −1 to 1. If Moran’s I is greater than 0, the ecological security pattern is clustered in spatial distribution, and the higher the value is, the higher the spatial correlation of its distribution and the smaller the difference. If Moran’s I is lower than 0, the pattern of ecological security differs in spatial distribution, and the lower the value is, the lower the spatial correlation of its distribution and the greater the difference.
(2) Anselin Local Moran’s I
Global Moran’s I cannot realize the patterns of visual spatial analysis. In addition, it usually analyzes only the overall spatial correlation of geographic elements or attributes from a global perspective. It cannot be used to estimate the difference degree of local spatial autocorrelation and cannot analyze whether there are agglomeration characteristics in local space. Therefore, it is often necessary to explore the local correlation of geographical elements or geographical attributes with the help of local autocorrelation analysis to analyze the spatial relationship features of ecological security in different regions. This paper adopted LISA analysis based on Anselin local Moran’s I (Equations (7) and (8)).
L o c a l   M o r a n ’ s   I = x i − x ¯ m ∑ i = 1 n W i j ( x j − x ¯ )
m = ( ∑ j = 1 , j ≠ i n x j 2 ) / ( n − 1 ) − x ¯ 2
LISA analysis refers to the calculation of Moran’s I value in local space and then cluster analysis. According to the LISA clustering results, the spatial distribution types of ecological security are classified into four types: high–high (H-H), (low–high) L-H, (low–low) L-L, and (high–low) H-L.
(3) Getis–Ord Gi* analysis
Hotspot analysis is used to reflect the features of spatial aggregation and distribution, which is used to identify the spatial clustering degree of hotspots (high-value regions) and coldspots (low-value regions) of specific variables in a region, and this method was proposed by J. Keith Ord and Arthur Getis. Getis–Ord Gi* statistics should be combined with Moran’s I statistics to reflect the characteristics of patterns that cannot be revealed by Moran’s I alone. Specifically, Getis–Ord Gi* statistics allow for us to test local dependence that may not appear while using global statistics [35] (Equation (9)).
G i ∗ d = ∑ j = 1 n W i j d X ∑ j = 1 n X i j
To intuitively reflect the characteristics of regional coldspots and hotspots, it is necessary to carry out retrograde standardization treatment on G i ∗ d (Equation (10)).
Z G i ∗ = G i ∗ − E G i ∗ v a r ( G i ∗ )
where W i j d refers to the spatial weighting matrix; E G i ∗ refers to the mathematical expectation of G i ∗ ; and v a r ( G i ∗ ) refers to the variable coefficient of G i . The aggregation is considered statistically significant when Z G i ∗ refers to positive and significant or negative and significant. Otherwise, the feature is considered to be a response to a random spatial distribution.

3.4.4. Change Slope Method

The trend of ecological security can be obtained by one-dimensional linear regression analysis, which uses a trend line to simulate the dynamic changes of each raster to obtain the changes in ecological security at different times (Equation (11)).
X = n × ∑ i = 1 n i × P i − ∑ i = 1 n i ∑ i = 1 n P i n × ∑ i = 1 n i 2 − ∑ i = 1 n i 2
where X refers to the change in slope, n refers to the amount of time in years, and P i refers to the ESI in the ith year. A positive slope of change indicates an increasing trend of ecological security in the region. On the contrary, the opposite is true. The slope of the change significance test is determined by the F test, and the statistics are calculated as follows (Equations (12)–(14)):
F = U ( n − 2 ) Q
Q = ∑ i = 1 n Y i − Y ^ i 2
U = ∑ i = 1 n Y ^ i − Y ¯ i 2
where Q refers to the sum of the squares of errors; U refers to the sum of regression squares; Y i refers to the actual value of ecological security in the ith year; Y ^ i refers to its regression value; and Y ¯ i refers to the annual average. According to the results of the change slope and significance test, the change slope can be divided into three categories: significant increase ( X > 0 ,     p < 0.05 ) , significant decrease ( X < 0 ,     p < 0.05 ) , and no significant change ( p > 0.05 ) .

3.4.5. Hurst Index

The Hurst index shows the autocorrelation of the time series, in particular, the hidden long-term trend in it, and is a useful method for quantitatively describing the information of the long-term dependence of time series. This index was first put forward by Hurst, who was a British hydrologist. It is used in a wide range of applications in hydrology, economics and geology [36].
Whether the time series of the Ecological security index is completely random or persistent is determined using the value of H . The time series is a persistent series when the value is from 0.5 to 1, that is, the future trend is in keeping with the past trend, and the closer the value is to 1 , the stronger the persistence. It represents a random time series when H = 0.5 ; that is, the future trend is not related to the past trend. Finally, the time series has inverse persistence when the value is from 0 to 0.5; that is, the trend is reversed from the past trend, and the closer H is to zero, the stronger the inverse persistence.

4. Results

4.1. Landscape Pattern Analysis

Landscape patterns refer mainly to the shape, proportion, and spatial configuration of the ecosystems or landcover types that constitute the landscape. The landscape pattern index provides a flexible reflection of pattern information, revealing its structure and spatial configuration [37]. In this paper, the classification results of four periods of landcover in 2005, 2010, 2015, and 2020 were obtained through the decision tree classification model (Figure 5). Some landscape pattern indexes were chosen to study the landscape pattern of the research area at the type level and at the landscape level.
Depending on the statistical analysis of the classification results, the landscape of Song County is dominated by forest, accounting for 60% to 70% of the whole area of the whole district. For the period of 2005–2020, the change in landcover in Song County mainly showed a decreasing trend in forest area, grassland area, and unused area, while farmland area and construction land area showed an increasing trend. Rapid urban expansion in recent years has led to massive areas of grasslands being occupied by construction land. In addition, continuous population growth leads to the continuous expansion of farmland area, and unused land is continuously developed and utilized.

4.1.1. Landscape Patch Type Level

According to the research purpose and the ecological significance of each index, four type level indexes were chosen to study the landscape pattern characteristics of various landscape types in Song County (Table 5, Figure 6).
NP refers to the number of landscape patches. The change in the number of patches can be indicative of changes in landscape patches in the study area, as well as the evolutionary dynamics of the landscape pattern. As shown in Figure 6a, the NP values of farmland, forest, construction land, and water showed an increasing trend, representing that the land use changes in recent years have intensified the fragmentation of these four landscape types. The NP values of grassland and unused land showed a decreasing trend, indicating that these landscape types were less fragmented.
The ED value of forest was the highest, followed by that of farmland and grassland, which showed the higher complexity of these three types of landscape patches. The ED of unused land and water was smaller, which indicates that their potential for internal exchange of material, energy, and other information is smaller. The ED values of grassland and unused land showed a declining trend in the last 15 years, indicating that their spatial layout was gradually dispersed. The ED values of farmland and construction land showed an increasing trend, indicating that their spatial layout is gradually aggregated.
PLAND reflects the proportion of landscape patches in the research area. As illustrated in Table 4 and Figure 6c, the main land use types in Song County were forest, farmland, and grassland, and the main landscape types were composed of them. In recent years, the PLAND value of grassland and unused land has declined, that of farmland and construction land has increased, and that of forest and water has remained basically unchanged.
The LPI reflects the dominance of landscape types. As shown in Figure 6d, the LPI value of forest was the largest, reaching 41.3 in 2005, followed by that of farmland, which showed that the landscape types in the research area were mainly forest and farmland. The LPI of grassland, forest, and unused land showed a declining trend in the last 15 years, reflecting that these landscape types have broken from larger patches into smaller patches, and the LPI of construction land and farmland showed an increasing trend. The weak change in water was mainly related to flood storage and water resource dispatching in various time periods in the Luhun Reservoir. Overall, the landscape connectivity of forest and grassland was good; fragmentation was low in the early period, and then urban expansion changed the original land use pattern, causing the patches to become gradually fragmented, while the patches of construction land gradually increased in size.

4.1.2. Landscape Level

To analyze the diversity, fragmentation, heterogeneity, and other characteristics of the landscape in Song County, seven landscape-level indexes were chosen to study the landscape pattern characteristics of the overall landscape in Song County (Table 6).
The overall landscape pattern features reflect the influence of natural and external activities on landscape pattern changes. Landscape fragmentation is the process of landscape transformation from uniformity to complexity. From 2005 to 2020, the NP value and PD value of the study area illustrated an increasing trend, indicating an overall increase in landscape fragmentation. CONTAG first decreased and then increased, with an overall decreasing trend, which indicated that the degree of concentration of different patch types has been decreasing; that is, the connectivity of the dominant landscape has tended to decrease. The LSI showed an increasing trend, from 67.93 in 2005 to 74.23 in 2020, which indicated that the overall landscape form gradually tended to develop in the direction of complex irregularity. There was a small change in IJL, with a slight overall increase, reflecting a decrease in the dispersion and juxtaposition between landscape types, with an increasingly uneven distribution of surrounding patch types.
The SHDI first increased and then decreased, with an overall increasing trend, indicating an increase in patch types or a balanced distribution of patch types across the landscape. The situation of land use is complicated. Similarly, the SHEI first increased and then decreased, with an overall increasing trend, indicating that the uniformity of the spatial disposition of different landscape types in the landscape pattern has increased, and the landscape dominance has gradually decreased. Both indexes showed an upwards trend during the study period, reflecting that the dominant landscape types gradually weakened their control over the study area. In addition, different landscape types have gradually changed from sporadic distribution to concentrated contiguity; thus, the stability and complexity of the landscape ecosystem has increased.

4.2. Dynamic Evaluation of Ecological Security

4.2.1. Weights of Evaluation Indicators

According to the evaluation index system constructed above, the weights of each index for each period were calculated by the CRITIC weighting method referring to Equations (15)–(21) (Table 7).

4.2.2. Determination of Ecological Security Grade

According to the comprehensive index method, each indicator was multiplied by the matching weight and summed to acquire the ESI of each period. To better reflect the change in ecological security of each administrative unit in the research area, the average of the ESI of each village was obtained by taking the village as a unit, and the ecological security was classified into five levels through the natural breakpoint method, namely, unsafe, less unsafe, moderately safe, less safe, and safe. Moreover, the proportion of each grade area was determined (Figure 7, Table 8).
In the last 15 years, the area of unsafe level has increased from 10.6% to 17.9%, and it is mostly distributed in the areas with Luhun Reservoir and Yihe River basin as the core, as well as near Baihe Street Village in Baihe Township and Che Village in Muzhi Street Township. Due to the distribution along rivers and the development of many geological disasters in this area, human engineering activities have been intense, resulting in great changes in landscape patterns; thus, the ecological security level of this area is high. However, the ecological security level in the central, northwest, and southwest areas of Song County is low, which is due to the high altitude, mostly forest landscape type, sparse population, and less change in landscape pattern.

4.3. Spatiotemporal Differentiation Analysis

4.3.1. Global Moran’s I

The statistical results of Moran’s I over the years are shown in Figure 8: Moran’s I > 0, Z score > +2.58, p value < 0.01. The main conclusions are as follows:
(1) Moran’s I indexes are all positive, which indicates that the calculated data are spatially positively correlated, and the calculated value of the dataset is directly proportional to the spatial aggregation degree; that is, the higher the index is, the higher the spatial aggregation level of the ecological security level.
(2) The Z score is higher than 2.58, and the p value is lower than 0.01, so the probability that the dataset is randomly generated is less than 0, which reflects a confidence level of approximately 100%, indicating that the data are scientifically significant.
In general, the Z score and p value support the credibility of the data source, and Moran’s I over time indicates that the dataset follows a clear pattern. The distribution patterns of spatial data include dispersed, random, and clustered. We know that the spatial disposition patterns of the ecological security levels in 2005, 2010, 2015, and 2020 were all clustered according to the results.

4.3.2. Anselin Local Moran’s I

The LISA aggregation map is mostly used to characterize the degree of association between specific attributes within a spatial unit and the surrounding neighboring units, and it can be noted from Figure 9 that the spatial characteristics of ecological security levels in Song County have evolved to different degrees over time. The “high–high” (H-H) type was mainly found in the south central and northwestern parts of Song County, indicating that the ecological security of the area and its surrounding areas is great. The regions with the H-H aggregation type in 2005 were primarily distributed in the northern parts of Muzhijie Township, Checun Township, Baihe Township, Deting Township, and Dazhang Township. From 2010 to 2015, this region mostly changed to insignificant, and then changed to the H-H aggregation type in 2020. The “low–low” (L-L) type was primarily distributed in the northern part of Song County, indicating that the ecological safety of the region itself and the surrounding areas was poor. The regions showing the L-L aggregation type 2005 were mainly distributed in the south central part of Jiuxian Township; the central part of Dazhang Township and Deting Township; the northwestern part of Zhifang Township; the southern part of Yanzhuang Township and Tianhu Township; and Fanpo Township and Huangzhuang Township. From 2010 to 2020, there were small changes in the degree of aggregation in the region, but most still showed L-L clustering. The regions were distributed near the Luhun Reservoir, with frequent geological hazards, obvious erosion by rivers, and strong human engineering activities, so the ecological security was poor. The “high–low” (H-L) type is the “mountainous area” in low-value areas, that is, the areas where the vulnerability of the land system in neighboring areas is weak but the vulnerability of its own system is strong; the “low–high” (L-H) type is a high-value area “valley”; that is, its own land system has a low degree of vulnerability, but its surrounding areas have a high degree of vulnerability. These two types of grid cells were less distributed, accounting for a relatively small proportion.

4.3.3. Getis–Ord Gi* Analysis

Based on Getis–Ord Gi* analysis tools in ArcGIS, the hotspots and sub-hotspots were mainly in south central and northwest Song County, while the coldspots and sub-coldspots were in north central Song County and distributed along the Luhun Reservoir (Figure 10). Wumasi Forest Farm in Baihe Township, the south of Dingbaoshi Village in Checun Township, the south of Shuimo Village, Maogou Village in Zhifang Township, Wangmangzhai Forest Farm, and Sanrenchang Village in Dazhang Township have always been fragile hotspots. However, Sishang Village in Jiuxian Town; Xue Village, Yangzhuang Village, and Zhaoling Village in Dazhang Township; Yang Village in Deting Township; the south of Chengguan Township; north of Zhifang Township; northeast of Fanpo Township; and south of Jiugao Township have all been fragile coldspots for a long time. Other hotspots and coldspots have also changed dynamically. Thus, the ecological security in Luhun has experienced a significant deterioration trend.

4.3.4. Change Slope

From 2005 to 2020, the trend in the ESI (Figure 11a) showed that the change slope of the regional ESI was positive, which accounted for 74.27% of the area; that is, the ESI showed an upwards trend; the change slope of the regional ESI was negative, which accounted for 25.73% of the area; that is, the ESI showed a decreasing trend. The results of the F test (Figure 11b) showed that the areas with significantly increased ESI accounted for 71.91% of the whole area, mainly spread out in Luhun Township, the north of Deting Township, the north of Dazhang Township, the south of Muzhijie Township, Checun Township, and Baihe Township. The significantly reduced area, 24.75% of the overall area, was primarily in Wan Village, Wangping Village, Xiaozhang Village, Mashigou Village, the north of Donggou Village in Dazhang Township, the middle of Jiuxian Township, Hecun Township, the middle and north of Yanzhuang Township, Fanpo Township, and Jiugao Township, which are distributed along Luhun Reservoir. Frequent geological disasters and intense human engineering activities have led to great changes in landscape patterns. In addition, the self-recovery ability of land ecosystems has gradually weakened. Therefore, consideration should be given to strengthening the conservation of the ecosystem in this area to prevent any further deterioration in ecological security. Areas with no significant increase accounted for 3.34%, primarily in the south of Muzhijie Township and the middle of Checun Township.

4.3.5. Hurst Index

The Hurst index of the ESI time series ranges from 0.5 to 0.86, with an average value of 0.74. The Hurst index of each evaluation unit was greater than 0.5, which demonstrates that the future trend of ecological security in the study area has strong positive sustainability.
In summary, the areas where the ecological security index has decreased are mainly located in the Luhun Reservoir area, and the Hurst index indicates a high probability of continuing the current trend of change in the future. Therefore, it is necessary to focus on the ecological security protection issues in this area.

5. Discussion

5.1. Analysis of the Causes of Landscape Pattern Change

The main landscape types of Song County are forest and farmland. Currently, Song County has seen a significant increase in economic standards. Rapid urban expansion, continuous population growth and the continuous development and utilization of unused land have led to decreases in forest, grassland, and unused land and increases in construction land and farmland. The research on the driving forces behind landscape pattern changes by domestic and foreign scholars usually focuses on two types: natural and socioeconomic. When the time span is relatively short, natural factors are more stable, so human activities are the most important factor, which has received increasing attention from scholars [38]. Scholars have used a combination of landscape pattern index and gradient change analysis to study and deeply explore the process of landscape change [39], pointing out that the change in landscape pattern is from agricultural to manmade. Moreover, some scholars also analyzed the relationship between human activities and landscape pattern changes and concluded that human activities are one of the main variables affecting landscape pattern change [40]. The government’s policy documents are important guides to the development of a region, such as the Land Use Master Plan, which states that construction land should be increased reasonably during the planning period to basically guarantee the land demand for the county’s economic and social development. The layout of urban and rural building land should be optimized, and priority should be given to protecting the demand for land in central cities and key towns, effectively controlling the scale of building land and fully reflecting the functions of all types of land. In the future development of Song County, reasonable land development and utilization should be carried out while ensuring that the landscape ecology is not damaged.

5.2. Analysis of the Causes of Ecological Security Changes

The ecological security of Song County as a whole showed a relatively unsafe area with the Luhun Reservoir and Yihe River basin as the core, and the ecological security level of the northern area was higher than that of the southern part. Landscape patterns and ecological processes are closely related to each other and are manifested as the landscape pattern being a carrier of ecological processes. Variations in landscape patterns lead to changes in related ecological processes. The serious destruction of landscape patterns will affect the habitats of animals, cause ecological problems such as soil erosion, and even lead to the collapse of ecological security systems. The quality of landscape ecological safety reflects the integrity of the ecosystem and the health status of the ecosystem. From 2005 to 2020, the increase in urban construction land, the construction of economic development zones, and other activities led to increasing traffic congestion, unabated environmental pollution, a substantial reduction in high-quality farmland, declines in wildlife habitats, and even serious destruction, bringing about a series of ecological security problems and lowering the ecological security level of the surrounding areas in towns. Furthermore, this level is related to the population, economy, geological disasters, etc. The rivers are dense near the Luhun Reservoir in Song County, and there is a large number of structural fault sections. In addition, in recent years, rainfall factors and human engineering activities have caused the region to constantly experience geological disasters, which have seriously damaged the natural environment and have seriously threatened ecological security.
It can be seen that in future development plans, special attention should be paid to the ecological status of areas with higher ecological security levels. Firstly, it is necessary to strengthen the public’s awareness of ecological environment protection and avoid a series of ecological security issues caused by excessive development and utilization; coordinate the promotion of economic development and ecological environment protection; strictly control various development and construction activities in ecologically unsafe areas; and prohibit the destruction of natural resources. It is not allowed to blindly cut down vegetation in pursuit of economic benefits, which causes damage to resources and the environment. Besides, Further deterioration from the influence of natural factors and human engineering activities should be prevented. At the same time, it is also necessary to conduct geological hazard risk investigations in advance, carry out early warning and prevention work, and minimize harm. The higher the level of ecological security, the more solid the resource foundation for social and economic development, and the more vibrant and dynamic it can be.

6. Conclusions

In the last 15 years, the cultivated land area in Song County has been continuously expanding, and unused land has also been continuously developed and utilized. In addition, the fragmentation of the landscape in the research area increased overall, and the degree of clustering of the different patch types decreased. The overall landscape form has gradually tended to develop in a complex and irregular direction. Frequent geological disasters in the areas with the Luhun Reservoir and Yihe River Basin as the core and intense human engineering activities have resulted in great changes in landscape patterns, the ecological security level of which is unsafe, and the affected area has been gradually increasing. The spatial characteristics of the ecological security degrees in Song County have evolved over time. Coldspots and sub-coldspots were in north central Song County and distributed along the Luhun Reservoir. Moreover, in recent years, the landscape pattern of the coldspots has varied considerably, resulting in a gradual weakening in the self-restoration capacity of the land ecosystem and gradual acceleration in the degree of damage to the ecological environment, thus predicting a continuing decrease in the ecological safety index in the forthcoming future.
A longer time series study is needed to explore the impact of landscape pattern variations on ecological security, and to accurately predict the future development trend to propose corresponding solutions.

Author Contributions

Conceptualization, Huaidan Zhang and Ke Nie; methodology, Huaidan Zhang; software, Huaidan Zhang; validation, Huaidan Zhang and Xueling Wu; formal analysis, Huaidan Zhang; investigation, Ke Nie; resources, Ke Nie; data curation, Huaidan Zhang and Xueling Wu; writing—original draft preparation, Huaidan Zhang and Xueling Wu; writing—review and editing, Huaidan Zhang; visualization, Xueling Wu; supervision, Ke Nie; project administration, Xueling Wu; funding acquisition, Xueling Wu and Ke Nie. All authors have read and agreed to the published version of the manuscript.”

Funding

This work was jointly supported by the National Natural Science Foundation of China [grant number 42071429] and the Open Fund of the Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources [grant number KF-2023-08-19].

Data Availability Statement

The data are not publicly available due to trade secrets and personal privacy restrictions.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Chen, J.; Wang, S.; Zou, Y. Construction of an ecological security pattern based on ecosystem sensitivity and the importance of ecological services: A case study of the Guanzhong Plain urban agglomeration, China. Ecol. Indic. 2022, 136, 108688. [Google Scholar] [CrossRef]
  2. Zhang, Y.; Zhao, Z.; Fu, B.; Ma, R.; Yang, Y.; Lü, Y.; Wu, X. Identifying ecological security patterns based on the supply, demand, and sensitivity of ecosystem service: A case study in the Yellow River Basin. China. J. Environ. Manag. 2022, 315, 115158. [Google Scholar] [CrossRef] [PubMed]
  3. Li, L.; Huang, X.; Wu, D.; Wang, Z.; Yang, H. Optimization of ecological security patterns considering both natural and social disturbances in China’s largest urban agglomeration. Ecol. Eng. 2022, 180, 106647. [Google Scholar] [CrossRef]
  4. Liu, C.; Wang, C.; Li, Y.; Wang, Y. Spatiotemporal differentiation and geographic detection mechanism of ecological security in Chongqing, China. Glob. Ecol. Conserv. 2022, 35, e02072. [Google Scholar] [CrossRef]
  5. Ning, X.; Zhu, N.; Liu, Y.; Wang, H. Quantifying impacts of climate and human activities on the grassland in the Three-River Headwater Region after two phases of Ecological Project. Geogr. Sustain. 2022, 3, 164–176. [Google Scholar] [CrossRef]
  6. Yuan, Y.; Bai, Z.; Zhang, J.; Xu, C. Increasing urban ecological resilience based on ecological security pattern: A case study in a resource-based city. Ecol. Eng. 2022, 175, 106486. [Google Scholar] [CrossRef]
  7. Troll, C. Aerial Photography and ecological studies of the earth. Z. Ges. Erdkd. Berl. 1939, 241–298. [Google Scholar]
  8. Ren, J.; Yan, D.; Ma, Y.; Liu, J.; Su, Z.; Ding, Y.; Wang, P.; Dang, Z.; Niu, J. Combining Phylogeography and Landscape Genetics Reveals Genetic Variation and Distribution Patterns of Stipa breviflora Populations. Flora 2022, 293, 152102. [Google Scholar] [CrossRef]
  9. Wu, Z.; Lei, S.; Yan, Q.; Bian, Z.; Lu, Q. Landscape ecological network construction controlling surface coal mining effect on landscape ecology: A case study of a mining city in semi-arid steppe. Ecol. Indic. 2021, 133, 108403. [Google Scholar] [CrossRef]
  10. Yu, Z.; Song, D.; Song, Y.; Lau, S.K.; Han, S. Research on a visual thermal landscape model of underground space based on the spatial interpolation method—A case study in Shanghai. Energy Rep. 2022, 8, 406–418. [Google Scholar] [CrossRef]
  11. Wang, J.; Cao, Y.; Fang, X.; Li, G.; Cao, Y. Identification of the trade-offs/synergies between rural landscape services in a spatially explicit way for sustainable rural development. J. Environ. Manag. 2021, 300, 113706. [Google Scholar] [CrossRef] [PubMed]
  12. Zhang, H.; Liang, X.; Chen, H.; Shi, Q. Spatio-temporal evolution of the social-ecological landscape resilience and management zoning in the loess hill and gully region of China. Environ. Dev. 2021, 39, 100616. [Google Scholar] [CrossRef]
  13. Stupariu, I.P.; Stupariu, M.S.; Stoicescu, I.; Peringer, A.; Buttler, A.; Fürst, C. Integrating geo-biodiversity features in the analysis of landscape patterns. Ecol. Indic. 2017, 80, 363–375. [Google Scholar] [CrossRef]
  14. Zhang, Y.W.; Shangguan, Z.P. The change of soil water storage in three land use types after 10 years on the Loess Plateau. Catena 2016, 147, 87–95. [Google Scholar] [CrossRef]
  15. Singh, J.S.; Roy, P.S.; Murthy, M.S.R.; Jha, C.S. Application of landscape ecology and remote sensing for assessment, monitoring and conservation of biodiversity. J. Indian Soc. Remote Sens. 2010, 38, 365–385. [Google Scholar] [CrossRef]
  16. Zhang, M.; Bao, Y.; Xu, J.; Han, A.; Liu, X.; Zhang, J.; Tong, Z. Ecological security evaluation and ecological regulation approach of East-Liao River basin based on ecological function area. Ecol. Indic. 2021, 132, 108255. [Google Scholar] [CrossRef]
  17. Cheng, H.; Zhu, L.; Meng, J. Fuzzy evaluation of the ecological security of land resources in mainland China based on the Pressure-State-Response framework. Sci. Total Environ. 2022, 804, 150053. [Google Scholar] [CrossRef] [PubMed]
  18. Mai, S.; Xu, J.; Xu, S.; Pan, Y. Application of the PSR model to evaluation of wetland ecosystem health. Trop. Geogr. 2005, 25, 317–321. [Google Scholar]
  19. Tu, J.; Wan, M.; Chen, Y.; Tan, L.; Wang, J. Biodiversity assessment in the near-shore waters of Tianjin city, China based on the Pressure-State-Response (PSR) method. Mar. Pollut. Bull. 2022, 184, 114123. [Google Scholar] [CrossRef]
  20. Wang, D.; Li, Y.; Yang, X.; Zhang, Z.; Gao, S.; Zhou, Q.; Zhuo, Y.; Wen, X.; Guo, Z. Evaluating urban ecological civilization and its obstacle factors based on integrated model of PSR-EVW-TOPSIS: A case study of 13 cities in Jiangsu Province, China. Ecol. Indic. 2021, 133, 108431. [Google Scholar] [CrossRef]
  21. Ma, L.B.; Bo, J.; Li, X.Y.; Fang, F.; Cheng, W.J. Identifying key landscape pattern indices influencing the ecological security of inland river basin: The middle and lower reaches of Shule River Basin as an example. Sci. Total Environ. 2019, 674, 424–438. [Google Scholar] [CrossRef] [PubMed]
  22. Zhang, J.; Gao, J. Lake ecological security assessment based on SSWSSC framework from 2005 to 2013 in an interior lake basin, China. Environ. Earth Sci. 2016, 75, 1–11. [Google Scholar] [CrossRef]
  23. Chen, A.; Zhao, X.; Yao, L.; Chen, L. Application of a new integrated landscape index to predict potential urban heat islands. Ecol. Ind. 2016, 69, 828–835. [Google Scholar] [CrossRef]
  24. McGarigal, K.; Compton, B.W.; Plunkett, E.B.; DeLuca, W.V.; Grand, J.; Ene, E.; Jackson, S.D. A landscape index of ecological integrity to inform landscape conservation. Landsc. Ecol. 2018, 33, 1029–1048. [Google Scholar] [CrossRef]
  25. Wang, X.D.; Zhang, C.B.; Wang, C.; Liu, G.W.; Wang, H.X. GIS-based for prediction and prevention of environmental geological disaster susceptibility: From a perspective of sustainable development. Ecotoxicol. Environ. Saf. 2021, 226, 112881. [Google Scholar] [CrossRef] [PubMed]
  26. Arabameri, A.; Yamani, M.; Pradhan, B. Novel ensembles of COPRAS multicriteria decision -making with logistic regression, boosted regression tree, and random forest for spatial prediction of gully erosion susceptibility. Sci. Total Environ. 2019, 688, 903–916. [Google Scholar] [CrossRef]
  27. Froude, M.J.; Petley, D.N. Global fatal landslide occurrence from 2004 to 2016. Nat. Hazards Earth Syst. Sci. 2018, 18, 2161–2181. [Google Scholar] [CrossRef]
  28. Wang, J.; Wei, X.; Guo, Q. A three-dimensional evaluation model for regional carrying capacity of ecological environment to social economic development: Model development and a case study in China. Ecol. Indic. 2018, 89, 348–355. [Google Scholar] [CrossRef]
  29. Qu, Y.; Jin, X. Geological hazards susceptibility evaluation based on multi-year spatial–temporal evolution of assessment factors in Luding area, Sichuan Province, China. Geol. J. 2024, 59, 1520–1538. [Google Scholar] [CrossRef]
  30. Yang, C.C.; Prasher, S.O.; Enright, P.; Madramootoo, C.; Burgess, M.; Goel, P.K.; Callum, L. Application of decision tree technology for image classification using remote sensing data. Agric. Syst. 2003, 76, 1101–1117. [Google Scholar] [CrossRef]
  31. Wang, X.A.; Zhou, Z.; Sun, L.C.; Xie, G.H.; Lou, Q.H. Research on the evaluation index system of “new energy cloud”operation mode based on CRITIC weighting method and AHP method. IOP Conf. Ser. Earth Environ. Sci. 2021, 831, 012017. [Google Scholar] [CrossRef]
  32. Anselin, L. How (not) to lie with spatial statistics. Am. J. Prev. Med. 2006, 30, S3–S6. [Google Scholar] [CrossRef]
  33. Mann, A.; Folchn, D.C.; Kauffman, R.J.; Anselin, L. Spatial and temporal trends in information technology outsourcing. Appl. Geogr. 2015, 63, 192–203. [Google Scholar] [CrossRef]
  34. Renato, M.; Assuncao Edna, A.R. A new proposal to adjust Moran’s I for population density. Stat. Med. 1999, 18, 2147–2162. [Google Scholar]
  35. Arthur, G.; Ord, J.K. The Analysis of Spatial Association by Use of Distance Statistics. Geogr. Anal. 1992, 24, 189–206. [Google Scholar]
  36. Xu, J.H. Mathematical Methods in Contemporary Geography; Higher Education Press: Beijing, China, 2022. [Google Scholar]
  37. Wang, G.; Wang, C.; Guo, Z.; Dai, L.; Wu, Y.; Liu, H.; Li, Y.; Chen, H.; Zhang, Y.; Zhao, Y.; et al. Integrating Maxent model and landscape ecology theory for studying spatiotemporal dynamics of habitat: Suggestions for conservation of endangered Red-crowned crane. Ecol. Indic. 2020, 116, 106472. [Google Scholar] [CrossRef]
  38. Turner, B.L.; Skole, D.L.; Sanderson, S.; Fischer, G.; Fresco, L.; Leemans, R. Land Cover Change Science/Research Plan; Global Change Report (Sweden); IIASA: Stockholm, Sweden, 1995. [Google Scholar]
  39. Soffianian, A.; Mokhtari, Z.; Khajeddin, S.J.; Ziaei, H.R. Gradient R analysis of urban landscape pattern (case study from Isfahan city). Hum. Geogr. Res. Q. 2013, 23, 23–39. [Google Scholar]
  40. Bürgi, M.; Straub, A.; Gimmi, U.; Salzmann, D. The recent landscape history of Limpach valley, Switzerland: Considering three empirical hypotheses on driving forces of landscape change. Landsc. Ecol. 2010, 25, 287–297. [Google Scholar] [CrossRef]
Ijgi 13 00204 g001
Figure 1. Location of the study area in China. (a) Map of China. (b) Map of Henan Province. (c) Map of Song County. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 1. Location of the study area in China. (a) Map of China. (b) Map of Henan Province. (c) Map of Song County. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g001
Ijgi 13 00204 g002
Figure 2. A schematic flow chart of dynamic evaluation of ecological security. This figure was created using Visio (https://www.microsoft.com/, accessed on 1 January 2019.).
Figure 2. A schematic flow chart of dynamic evaluation of ecological security. This figure was created using Visio (https://www.microsoft.com/, accessed on 1 January 2019.).
Ijgi 13 00204 g002
Ijgi 13 00204 g003aIjgi 13 00204 g003bIjgi 13 00204 g003c
Figure 3. Evaluation index of ecological security evaluation in 2020. (a) Geological disaster susceptibility. (b) Population density. (c) Landscape fragmentation index. (d) Biological abundance index. (e) Water conservation index. (f) Vegetation coverage index. (g) Landscape disturbance index. (h) Landscape restoration index. (i) GDP density. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 3. Evaluation index of ecological security evaluation in 2020. (a) Geological disaster susceptibility. (b) Population density. (c) Landscape fragmentation index. (d) Biological abundance index. (e) Water conservation index. (f) Vegetation coverage index. (g) Landscape disturbance index. (h) Landscape restoration index. (i) GDP density. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g003aIjgi 13 00204 g003bIjgi 13 00204 g003c
Ijgi 13 00204 g004aIjgi 13 00204 g004bIjgi 13 00204 g004c
Figure 4. Evaluation indicators of geological disaster susceptibility in 2020. (a) Elevation. (b) Slope. (c) Aspect. (d) Relief of topography. (e) Engineering rock formation. (f) Distance from structure. (g) Landcover. (h)NDVI. (i)NDWI. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 4. Evaluation indicators of geological disaster susceptibility in 2020. (a) Elevation. (b) Slope. (c) Aspect. (d) Relief of topography. (e) Engineering rock formation. (f) Distance from structure. (g) Landcover. (h)NDVI. (i)NDWI. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g004aIjgi 13 00204 g004bIjgi 13 00204 g004c
Ijgi 13 00204 g005
Figure 5. Classification results of landcover. (a) Landcover in 2005. (b) Landcover in 2010. (c) Landcover in 2015. (d) Landcover in 2020. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 5. Classification results of landcover. (a) Landcover in 2005. (b) Landcover in 2010. (c) Landcover in 2015. (d) Landcover in 2020. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g005
Ijgi 13 00204 g006aIjgi 13 00204 g006b
Figure 6. Change in patch type level index. (a) Number of patches (NP). (b) Edge density (ED). (c) Landscape type proportion (PLAND). (d) Maximum patch index (LPI).
Figure 6. Change in patch type level index. (a) Number of patches (NP). (b) Edge density (ED). (c) Landscape type proportion (PLAND). (d) Maximum patch index (LPI).
Ijgi 13 00204 g006aIjgi 13 00204 g006b
Ijgi 13 00204 g007
Figure 7. Evaluation level of ecological security. (a) Ecological security level in 2005. (b) Ecological security level in 2010. (c) Ecological security level in 2015. (d) Ecological security level in 2020. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 7. Evaluation level of ecological security. (a) Ecological security level in 2005. (b) Ecological security level in 2010. (c) Ecological security level in 2015. (d) Ecological security level in 2020. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g007
Ijgi 13 00204 g008
Figure 8. Moran’s I statistics of ecological security level in 2005, 2010, 2015, and 2020.
Figure 8. Moran’s I statistics of ecological security level in 2005, 2010, 2015, and 2020.
Ijgi 13 00204 g008
Ijgi 13 00204 g009
Figure 9. LISA result of spatial autocorrelation. (a) 2005. (b) 2010. (c) 2015. (d) 2020. The figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 9. LISA result of spatial autocorrelation. (a) 2005. (b) 2010. (c) 2015. (d) 2020. The figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g009
Ijgi 13 00204 g010
Figure 10. Distribution map of cold/hotspots. (a) 2005. (b) 2010. (c) 2015. (d) 2020. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 Jun1 2021).
Figure 10. Distribution map of cold/hotspots. (a) 2005. (b) 2010. (c) 2015. (d) 2020. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 Jun1 2021).
Ijgi 13 00204 g010
Ijgi 13 00204 g011
Figure 11. Changing trend of ecological security. (a) Change slope. (b) Results of F significance test. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Figure 11. Changing trend of ecological security. (a) Change slope. (b) Results of F significance test. This figure was created using ArcGIS ver. 10.4 (https://www.esri.com/, accessed on 1 June 2021).
Ijgi 13 00204 g011
Table 1. Data sources.
Table 1. Data sources.
DataYearFormatAccess Platform
Landsat8 OLI2015, 2020TIFGeospatial Data Cloud
Landsat5 TM2005, 2010TIFGeospatial Data Cloud
DEM2009TIFGeospatial Data Cloud
Boundary data2020SHPVectorize
Landcover data2019SHPThe third national land resource survey
Interpolation data2005, 2010, 2015, 2020TIFResource and Environment Science and Data Center
Geologic map2020SHPNational Geological Archives
Table 2. Landscape pattern indices.
Table 2. Landscape pattern indices.
IndexIndication
Type levelNPThis value is positively correlated with landscape fragmentation.
EDIt has high sensitivity to landscape type changes.
PLANDPart of the basis for determining the matrix or leading landscape elements in the landscape, and also is an important factor in determining ecosystem indicators such as biological diversity and the prevailing species in the landscape.
LPIThis value helps to reflect some ecological characteristics in the landscape, such as the dominant types and the abundance of internal species. In addition, it affects material migration in the ecosystem and the orientation and strength of people’s activities.
Landscape patternPDCan be used to make comparisons between different landscapes.
CONTAGRepresents the extent of aggregation or expansion trend of various patch types in the landscape.
LSIReflects the complexity and irregularity of the overall landscape shape.
IJLIndicates the overall spread and juxtaposition of patch types at the landscape level, which suggests that the spread characteristics of ecosystems are severely constrained by some natural factors.
SHDIReflects the heterogeneity of the landscape and is especially flexible to the unbalanced spread of each patch type in the landscape.
SHEIUsed to compare changes in diversity across landscapes or in the same landscape over time.
Table 3. Evaluation index system of ecological security.
Table 3. Evaluation index system of ecological security.
Criterion LayerFactor LayerUnitFactor Type
PressureGeological disaster susceptibility/Negative
Population densityPersons/km2Negative,
Landscape fragmentation index/Negative
StateBiological abundance index/Positive
Water conservation index/Positive
Vegetation coverage%Positive
ResponseLandscape disturbance index/Negative
Landscape restoration index/Positive
GDP densityWanyuan/km2Negative
Table 4. Evaluation index system of geological disaster susceptibility.
Table 4. Evaluation index system of geological disaster susceptibility.
Criterion LayerFactor LayerIndicator LayerUnitData Source
Geological disaster susceptibilityStatic factorElevationmDEM
Slope°DEM
Aspect°DEM
Relief of topographymDEM
Engineering rock formation/Geologic map
Distance from structuremGeologic map
Dynamic factor Landcover/Landsat images
Normalized vegetation index%Landsat images
Normalized water index%Landsat images
Table 5. Index of patch type level.
Table 5. Index of patch type level.
Landscape TypeTimePLANDNPLPIED
Grassland200511.663614681.233421.1442
20109.750913961.160919.2182
20157.008112900.197115.1222
20206.007811450.197112.9057
Forest200566.443086241.300135.5414
201065.521487940.753037.0594
201564.901192540.510437.2659
202064.6362103039.764037.4359
Construction20052.84888360.22126.7390
20104.167911250.24039.6562
20154.933313750.258311.4446
20205.862314790.526212.9336
Farmland200513.184714941.374321.8278
201015.007416601.381524.9227
201518.470019891.466331.3832
202019.439624761.448134.0384
Unused20053.27019010.09616.7835
20102.46758710.03125.8332
20151.51966640.03263.8909
20200.93774400.01982.4122
Water20052.58852241.24834.0238
20103.08563011.37625.1470
20153.10793081.24455.5046
20203.11623121.24455.5333
Table 6. Index of landscape level.
Table 6. Index of landscape level.
TIMENPPDLSICONTAGSHDISHEIIJL
200557851.930467.934960.1261.09720.612476.0032
201062322.079671.890358.99661.11970.624977.9305
201565512.186173.785859.45721.10.613974.75
202068822.296574.228959.86021.08790.60771.3802
Table 7. Weights of evaluation indicators.
Table 7. Weights of evaluation indicators.
Criterion LayerFactor LayerWeight (2005)Weight (2010)Weight (2015)Weight (2020)
PressureGeological disaster susceptibility0.248 0.265 0.269 0.272
Population density0.007 0.047 0.033 0.034
Landscape fragmentation index0.103 0.083 0.078 0.079
StateBiological abundance index0.138 0.106 0.127 0.126
Water conservation index0.144 0.117 0.129 0.128
Vegetation coverage0.146 0.136 0.133 0.115
ResponseLandscape disturbance index0.126 0.125 0.137 0.142
Landscape restoration index0.081 0.091 0.069 0.080
GDP density0.007 0.029 0.024 0.024
Table 8. Results of ecological security evaluation.
Table 8. Results of ecological security evaluation.
Ecological Security Level2005201020152020
Area (km2)Proportion (%)Area (km2)Proportion (%)Area (km2)Proportion (%)Area (km2)Proportion (%)
Unsafe318.270.106328.250.109374.070.124538.660.179
Less unsafe698.120.232792.970.264568.630.189596.840.198
Moderately safe646.710.215454.400.151601.410.200497.840.165
Less safe 745.910.248544.660.181588.080.195659.680.219
Safe599.890.199888.620.295876.710.291715.880.238
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

Zhang, H.; Nie, K.; Wu, X. Spatiotemporal Analysis of Ecological Security Based on Landscape Patterns. ISPRS Int. J. Geo-Inf. 2024, 13, 204. https://doi.org/10.3390/ijgi13060204

AMA Style

Zhang H, Nie K, Wu X. Spatiotemporal Analysis of Ecological Security Based on Landscape Patterns. ISPRS International Journal of Geo-Information. 2024; 13(6):204. https://doi.org/10.3390/ijgi13060204

Chicago/Turabian Style

Zhang, Huaidan, Ke Nie, and Xueling Wu. 2024. "Spatiotemporal Analysis of Ecological Security Based on Landscape Patterns" ISPRS International Journal of Geo-Information 13, no. 6: 204. https://doi.org/10.3390/ijgi13060204

APA Style

Zhang, H., Nie, K., & Wu, X. (2024). Spatiotemporal Analysis of Ecological Security Based on Landscape Patterns. ISPRS International Journal of Geo-Information, 13(6), 204. https://doi.org/10.3390/ijgi13060204

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