Next Article in Journal
Fuel–Water Emulsions as an Alternative Fuel for Gas Turbines: A Project Summary
Previous Article in Journal
Valorization of Grass Clipping Waste: A Sustainable Approach to Cellulose Extraction and Paper Manufacturing
Previous Article in Special Issue
Three-Dimensional Model-Based Line-of-Sight Analysis for Optimal Installation of IoT Monitoring Devices in Underground Mines
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 14
Issue 15
10.3390/app14156685
applsci-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

Correlation between Soil Moisture Change and Geological Disasters in E’bian Area (Sichuan, China)

by
Hongyi Guo
and
Antonio Miguel Martínez-Graña
*
Dpto. Geología, Faculty of Sciences, University of Salamanca, Plaza de la Caidos s/n, 37008 Salamanca, Spain
*
Author to whom correspondence should be addressed.
Appl. Sci. 2024, 14(15), 6685; https://doi.org/10.3390/app14156685
Submission received: 19 May 2024 / Revised: 19 July 2024 / Accepted: 28 July 2024 / Published: 31 July 2024
(This article belongs to the Special Issue Geographic Visualization: Evaluation and Monitoring of Geohazards—2nd Edition)
Browse Figures
Versions Notes

Abstract

:
E’bian Yi Autonomous County is a mineral-rich area located in a complex geological structure zone. The region experiences frequent geological disasters due to concentrated rainfall, steep terrain, and uneven vegetation cover. In particular, during the rainy season, large amounts of rainwater rapidly accumulate, increasing soil moisture and slope pressure, making landslides and debris flows more likely. Additionally, human activities such as mining, road construction, and building can alter the original geological structure, exacerbating the risk of geological disasters. According to publicly available data from the Leshan government, various types of geological disasters occurred in 2019, 2020, 2022, and 2023, resulting in economic losses and casualties. Although some studies have focused on geological disaster issues in E’bian, these studies are often limited to specific areas or types of disasters and lack comprehensive spatial and temporal analysis. Furthermore, due to constraints in technology, funding, and manpower, geophysical exploration, field geological exploration, and environmental ecological investigations have been challenging to carry out comprehensively, leading to insufficient and unsystematic data collection. To provide data support and monitoring for regional territorial spatial planning and geological disaster prevention and control, this paper proposes a new method to study the correlation between soil moisture changes and geological disasters. Six high-resolution Landsat remote sensing images were used as the main data sources to process the image band data, and terrain factors were extracted and classified using a digital elevation model (DEM). Meanwhile, a Normalized Difference Vegetation Index–Land Surface Temperature (NDVI-LST) feature space was constructed. The Temperature Vegetation Drought Index (TVDI) was calculated to analyze the variation trend and influencing factors of soil moisture in the study area. The research results showed that the variation in soil moisture in the study area was relatively stable, and the overall soil moisture content was high (0.18 < TVDI < 0.33). However, due to the large variation in topographic relief, it could provide power and be a source basis for geological disasters such as landslide and collapse, so the inversion value of TVDI was small. The minimum and maximum values of the correlation coefficient (R2) were 0.60 and 0.72, respectively, indicating that the surface water content was relatively large, which was in good agreement with the calculated results of vegetation coverage and conducive to the restoration of ecological stability. In general, based on the characteristics of remote sensing technology and the division of soil moisture critical values, the promoting and hindering effects of soil moisture on geological hazards can be accurately described, and the research results can provide effective guidance for the prevention and control of geological hazards in this region.

1. Introduction

Sichuan Province, located in the southwestern region of China, features complex geological structures and significant variations in terrain, making it prone to frequent geological hazards. Particularly in the mountainous areas surrounding the Sichuan Basin, the combination of complex topography, geological conditions, and climatic factors contributes to a diverse range of geological hazards. The main geological hazards in this region include landslides, debris flows, ground subsidence, and fissures. Due to the rugged terrain and abundant precipitation, landslides often occur, especially during the rainy season, posing risks such as road blockages. Loose materials on slopes can be washed down, forming debris flows that threaten human life and economic development. Under certain geological conditions, human activities and tectonic movements can induce ground subsidence, potentially causing building collapses. Additionally, crustal movements or changes in groundwater levels can cause fissures to appear on the ground, further damaging buildings and infrastructure. Overall, the geological hazards in the Sichuan region, particularly in the mountainous areas surrounding the Sichuan Basin, present a variety of risks to human life and economic activities. Mitigating these hazards requires comprehensive monitoring, prevention, and response measures to ensure the safety and stability of the region’s inhabitants and infrastructure.
E’bian County is located in the Xiaoliang Mountains, which serve as a transitional zone between the Sichuan Basin and the Western Sichuan Plateau. The geological conditions are complex, with strong neotectonic movements and well-developed geological hazards. According to historical data published by the Sichuan Provincial Natural Resources Library, the main geological hazards in E’bian County include landslides, collapses, debris flows, and unstable slopes. In recent years, due to changes in soil moisture and the combined effects of various natural and human factors, E’bian Yi Autonomous County in Leshan City, Sichuan Province, has faced frequent geological disasters. However, most current research by many scholars focuses on predicting geological disasters over a large area of Leshan City, with relatively few studies specifically addressing geological hazards in the E’bian region [1,2,3,4,5,6].
The study of the correlation between soil moisture changes and geological disasters is an important field in geology and environmental science. An increase in soil moisture can lead to an increase in soil weight, which weakens its shear strength and thus increases the likelihood of landslides. In addition, the wetting and swelling effects of water change the physical properties of the soil, further increasing the risk of landslides. Especially in seasons with abundant rainfall, loose material on the slopes may be washed down, forming a mudslide with extremely fast speed and huge destructive power, posing a threat to human life and economic development. And under certain geological conditions, the decrease in soil moisture can lead to the formation of underground cavities, which then cause ground subsidence. Under the influence of crustal movement or changes in groundwater level, cracks may occur on the ground. These cracks can also cause damage to buildings and infrastructure.
With the development of remote sensing technology, multisource remote sensing techniques have been widely applied in the prediction and monitoring of geological disasters [7,8,9,10,11,12]. Among them, FVC (Fraction of Vegetation Cover) and TVDI (Temperature Vegetation Dryness Index) are two commonly used remote sensing indicators that can provide important information about land surface conditions. This paper proposes a geological disaster prediction method based on remote sensing technology, integrating FVC and TVDI. Satellite imagery data of the target area are acquired through satellite platforms, and the obtained remote sensing images are processed and analyzed to extract FVC and TVDI data. Specifically, FVC reflects the vegetation coverage of the land surface, while TVDI indicates the degree of surface dryness. These data are used to establish a statistical model for predicting geological disasters and obtaining predictions of geological disaster occurrence [13,14,15].
Therefore, this study takes E’bian County as the target area, utilizing Landsat high-resolution remote sensing imagery and a digital elevation model (DEM), combined with Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST), to calculate the Temperature Vegetation Dryness Index (TVDI). This analysis aims to deeply investigate the soil moisture variation trends in E’bian County and their impact on geological disasters. The research findings of this study can partly fill this research gap and provide a scientific basis and technical support for local geological disaster prevention and mitigation efforts.

2. Materials and Methods

2.1. Study Area

The western Sichuan Basin, located in the western part of Sichuan Province and on the eastern edge of the Qinghai–Tibet Plateau, is characterized by complex terrain and diverse geological conditions, making it one of the regions in China where geological disasters frequently occur. The topographical features of E’bian Yi Autonomous County are mainly manifested as folds, faults, and ductile shear zones. These structural features have a significant impact on the occurrence and development of geological disasters, making this a frequent area of geological disasters in the western Sichuan Basin.
E’bian is located in Leshan City, upstream of the Yangtze River. The terrain of this area is mainly composed of mountains, hills, and plains. Mountains account for most of the total area of the county, mainly distributed in the western and northern parts of the county. The hilly area is mainly distributed in the middle of the county, while the plain is mainly distributed in the eastern part of the county. Due to the influence of the E’meishan Fault Zone, the E’bian region is one of the areas with frequent seismic activity. The most prominent fault in this area is the E’bian Fault in Leshan, which is one of the main faults causing seismic activity. In addition to the E’bian Fault, there are also several minor geological faults distributed in the region, which also have an impact on the geological structure (Figure 1).
The stratigraphic geological development of the E’bian region reflects a complex geological evolution spanning multiple periods of geological history, primarily including the Paleozoic, Mesozoic, and Cenozoic eras. During the Paleozoic era, the E’bian region was submerged under marine environments, accumulating abundant marine sedimentary rocks such as limestone and shale. Granite is widely distributed in the E’bian region, particularly in mountainous and hilly areas. These granites formed under high-temperature and high-pressure conditions deep within the Earth’s crust, undergoing prolonged cooling and crystallization, resulting in their hard and durable characteristics, commonly observed in outcrops and mountainous terrain. During the Mesozoic era, the E’bian region experienced tectonic movements and magmatic activity, giving rise to various igneous and metamorphic rocks. These include granites, gneisses, schists, and quartzites. In sedimentary basins and river valleys of the E’bian region, sandstones and shales are commonly found, formed through prolonged sedimentation and compaction, recording the region’s geological history and environmental changes. Since the Cenozoic era, the E’bian region has undergone tectonic uplift and erosion, shaping the present-day landscape. Sedimentary processes have continued during this period, resulting in the formation of new sedimentary rock layers in environments such as rivers and lakes. These rocks contain rich mineral resources, such as coal, iron ore, and copper ore [16,17,18]. The E’bian area belongs to a subtropical humid climate, with four distinct seasons, abundant rainfall, and sufficient sunlight.
At the same time, the water system in this area is developed, with many rivers passing through, the most important of which is the Minjiang River, an important tributary of the Yangtze River. These rivers provide abundant water resources for the ecological environment of this area, so the vegetation in this area is rich and diverse. However, with the development of urbanization and the increasing demand for mineral resources, human activities have played a leading role in the destruction of the ecological environment. Excessive deforestation may lead to a decline in forest coverage, a reduction in biodiversity, an intensification of soil erosion, and a decrease in water source protection capabilities. Industrial pollution, agricultural pollution, and domestic pollution lead to a decline in water quality, soil quality, and air quality, posing threats to human health and the ecological environment. Excessive reclamation and unreasonable construction, etc., lead to land degradation and a decline in ecosystem service functions. These activities have uncovered hidden dangers for the development of geological disasters.
The E’bian Yi Autonomous County has an altitude ranging from 1000 to 4000 m, with an average altitude of 1200 m. The terrain is higher in the west and south and lower in the east and north. Influenced by the Longmen Mountain Fault Zone, the current terrain undulations have been formed. The 10 m digital elevation model map, shown in Figure 2, drawn based on satellite image data can clearly reflect the range of elevation changes.
Fraction of Vegetation Cover (FVC) refers to the size of the vertical projected area of plants in a specific area, which usually fluctuates according to different factors such as elevation and slope, and is also a key factor affecting soil moisture [19,20,21,22,23]. The total length of roads along the E‘bian area is 1545.2 km, the river system in this area is developed (Figure 3), and the average annual rainfall is as high as 799 mm.
In the range of high altitude (1600–4000 m), the distribution of water and heat conditions is not uniform due to the large difference in terrain height, which results in a fluctuation trend in vegetation coverage. In the flat terrain area, the soil moisture content is higher, so there is an increasing trend in vegetation coverage. In the range of elevation below 600 m, the intensification of human activities such as engineering construction and mineral exploitation will lead to a decreasing trend in soil water content (Table 1). The change in ecological environment (temperature, humidity, etc.) in the study area will also have a great impact on vegetation coverage. Therefore, the characteristics of various influencing factors should be combined in the process of geological hazard analysis [24,25,26,27,28].
With the rapid development of remote sensing technology, one of its important advantages is that it can provide continuous spatiotemporal data. Higher resolution image data can provide technical means for extracting spatiotemporal information of soil moisture and vegetation coverage in a large range. Soil moisture is an important factor affecting vegetation growth and climate change, and vegetation coverage is an important indicator reflecting the growth status of surface vegetation. Due to the sensitivity of the infrared and microwave bands in remote sensing images to soil moisture and vegetation coverage, through remote sensing technology, we can obtain information such as soil humidity and temperature and analyze the data of these bands to obtain the distribution of soil moisture and vegetation coverage [29,30,31,32].
Based on the acquisition of terrain and landform image data of the study area using remote sensing technology, DEM data are utilized for terrain analysis to extract parameters such as slope and aspect. In this article, slope, slope direction, and surface relief were selected as the constraint factors to study the surface micro-environment and vegetation growth change (Figure 4).
Slope can represent the steepness of local surfaces, which is an important factor affecting surface material flow (especially surface water and underground water) and also an important factor affecting surface temperature. Larger gradients can cause some areas to be more exposed to direct sunlight, making them relatively warm and able to change soil texture. At the same time, the distribution of vegetation on different slopes may also affect the TVDI value. In this paper, the slope of the study area is divided into <10.6°, 10.6°, 10.6°~17.22°, 17.22°, 17.22°~23.18°, 23.18°, 23.18°~28.81°, 28.81°, 28.81°~34.11°, 34.11°, 34.11°~39.07°, 39.07°, 39.07°~44.37°, 44.37°, 44.37°~50.66°, 50.66°, 50.66°~58.94°, 58.94°, 58.94°~94.42°. The relationship between vegetation cover and slope was analyzed by superposition (Figure 5).
In the field of geological disaster prevention and monitoring, slope orientation has important early warning significance for surface sunlight exposure. The orientation of a slope determines its exposure to sunlight. Southern slopes, which receive more direct sunlight, tend to have higher surface temperatures. Conversely, northern slopes, which receive less direct sunlight, are typically cooler [33]. The temperature difference caused by slope orientation can directly impact the occurrence and evolution of geological hazards. Higher surface temperatures on southern slopes can increase the risk of localized debris flows. On the other hand, lower temperatures on northern slopes can make them more susceptible to freeze–thaw action, increasing the likelihood of landslides [34,35]. At the same time, slope aspect plays a crucial role in the field of geological hazards and is a key factor in the study of ecosystems and climate change. For instance, slope aspect can influence vegetation patterns, which can in turn affect soil stability and susceptibility to erosion.
In this paper, the slope direction is divided into flat (−1°), north slope (0°~22.5°, 337.5°~360°), northeast slope (22.5°~67.5°), east slope (67.5°~112.5°), southeast slope (112.5° and 157.5°), south slope (157.5° and 202.5°), southwest slope (202.5° and 247.5°), west slope (247.5°~292.5°), and northwest slope (292.5°~337.5°) for superposition analysis (Figure 6).
The complex interaction between slope aspect and surface relief creates a topographic effect, which can influence the occurrence and evolution of geological disasters. For instance, the degree of surface relief can affect the stability and mobility of landslides. This topographic effect can lead to significant differences in surface temperatures for different slope gradients in the geographical environment, thereby affecting the Temperature Vegetation Dryness Index (TVDI). TVDI is calculated based on the empirical relationship between surface temperature and the Normalized Difference Vegetation Index (NDVI). Surface relief can influence both surface temperature and NDVI, and changes in local weather conditions can also affect surface temperature, thereby impacting the quality of TVDI. Moreover, TVDI is one of the effective means for assessing the risk of geological disasters, such as monitoring and predicting the development of landslides and debris flows. The ups and downs of terrain directly affect the sunlight exposure and the distribution of surface temperature. Therefore, the influence of the terrain effect must be taken into account when studying slope direction. Changes in surface form fluctuations may cause local temperature differences, which is closely related to the terrain effect. On steeper slopes, direct sunlight is more concentrated, resulting in relatively high surface temperatures, while on gentler slopes, sunlight is relatively uniform and surface temperatures may be stable. Therefore, in the context of surface fluctuation, slope orientation is not only a simple direction of surface inclination but also involves the regulation of temperature distribution by topographic effects, which also has an impact on the surface temperature reflection of TVDI (Figure 7).
Terrain undulation analysis is a method that uses digital elevation models (DEMs) to gain more information about terrain features, which helps to understand the impact of various surface factors such as geological structure and geomorphological shaping processes. In addition, it can automatically detect terrain features (such as ridges, noses, cliffs, and peaks) by controlling the height and orientation of the light source and use an edge detection filter based on azimuth on the light relief to identify specific terrain features [36,37]. In the process of terrain undulation analysis, this article adopts a more accurate analysis method based on the traditional method of terrain undulation analysis, that is, moving window analysis technology. Moving window analysis is a method of sliding window statistics on multidimensional data in a specified dimension. This method can calculate different statistics according to different window sizes and directions, divide the terrain surface into a small area, and combine the scale of terrain features in the study area to achieve high-resolution data selection through different algorithms and parameters. In the process of dealing with terrain smoothness and noise reduction, one can reduce noise and sharp transitions by analyzing each small area, making the terrain look more natural and real (Equation (1)). The calculation of the window size in terrain undulation analysis is based on statistical information within the region. Moving window analysis has many advantages in the field of Geographic Information Systems (GISs) and terrain analysis. It can analyze geographic data within local areas, which helps to capture the features of local terrain, landforms, and other geographic phenomena, making the analysis more detailed and meticulous [38,39]. In terrain undulation analysis, the choice of window size is crucial. According to research findings [40,41], for drainage basins, a window of 7 × 7 units (217 × 217 m) is most suitable. For mountainous areas, a window of 9 × 9 units (90 × 90 m) works best. The more diverse the terrain in the study area, and in combination with the terrain and landform features, adjusting the spatial autocorrelation degree of different areas and selecting a smaller calculation window can truly reflect the geological features of the area. By sliding the window on the geographic data, spatial changes and patterns can be detected, which are helpful to understand the spatial distribution and change trend in surface features so that surface features and geographical phenomena can be better understood.
Mean ( i , j ) = 1 n 2 ∑ p = 1 n ∑ q = 1 n Z ( i − p + n 2 , j − q + n 2 )
where Z is the matrix of elevation values (window size n × n), Mean(i,j) is the mean of the window at (i,j), and n2 is the total number of elements of the window.
NDVI, TVDI, and LST are currently widely used in the field of environmental research. Among them, the Temperature Vegetation Dryness Index (TVDI) has a certain limitation, that is, it is not applicable to areas with high vegetation coverage (FVC). Vegetation can affect surface temperature by changing the physical characteristics of the surface. In addition, vegetation can induce geological disasters by affecting the stability and erosion resistance of the soil. On the other hand, vegetation can reduce the risk of geological disasters by enhancing the structural stability of the soil through its root system and reducing soil erosion. Surface temperature can affect the development of geological disasters by affecting soil moisture and the evaporation rate. For example, high surface temperatures can accelerate the evaporation of soil moisture, cause soil drying, and increase the risk of geological disasters. Generally speaking, the higher the vegetation coverage, the lower the surface temperature. Therefore, this article uses remote sensing technology as a research method; combines the vegetation coverage, surface temperature, and Temperature Vegetation Dryness Index; constructs a virtual triangle model formed by surface temperature and normalized vegetation index (NDVI), where TVDI is used as an indicator of drought changes and reflects the relationship and change law between NDVI and surface temperature (LST); studies the development law of geological disasters in the research area from an environmental perspective; and opens up a new research direction. In this article, the spectral feature space of normalized vegetation index (NDVI) (Equation (2)) and surface temperature (LST) was constructed by using Landsat high-spatial resolution images, and the Temperature Vegetation Drought Index (TVDI) of the study area was extracted to compare the correlation between vegetation coverage and soil moisture change from a mathematical perspective. Combined with the classification of soil moisture critical values [42,43], the promoting and hindering effects of soil moisture on geological disasters were analyzed, which provided a basis for the restoration of ecological environment in the later period.

2.2. Data and Preprocessing

Current remote sensing has formed a multi-level, multi-perspective, and multi-domain observation system that extends from the ground to the air and even to space. It encompasses information data collection, processing, interpretation, analysis, and application, enabling global detection and monitoring. Remote sensing has become an important means of obtaining information about Earth’s resources and the environment [44]. Landsat imagery provides a way for repeated and comprehensive observation of the Earth, representing the world’s longest continuously acquired space-based medium-resolution land remote sensing data. At the same time, Landsat data can play a good role in decision-making and management in many interdisciplinary fields. Generally speaking, Landsat sensors use a spatial resolution of 30 m, which is the ideal scale for observing surface changes. This article downloads image data through platforms such as Earth Explorer, Landsat Look, and Glo Vis, and in order to improve the accuracy of data processing, the enhanced Landsat instrument used has eight spectral bands, of which bands 1 to 5 and 7 have a spatial resolution of 30 m, and band 6 (thermal infrared) has a spatial resolution of 60 m. The spectral characteristics of Landsat imagery can reflect the biomass of vegetation, while DEM data can reflect the complexity and changes in terrain. Therefore, Landsat imagery and DEM data are used as independent variables of the regression model to construct an environmental model. The advantages of DEM generated based on Landsat image data for monitoring and predicting geological disasters are more obvious, mainly including that Landsat data can accelerate calculations to enhance the real-time monitoring capability of geological disasters. Landsat data have the ability to process large amounts of data and can improve the stability and fault tolerance of the system. Since Landsat data retain basic advantages such as linear computational complexity, unsupervised learning, and non-parametric methods, they can be used to predict the time and scale of possible geological disasters, as well as to evaluate the effectiveness of different disaster prevention measures.
In this paper, 6 high-resolution Landsat image types are selected as basic data, and the band information used are shown in Table 2. Based on DEM data and related geological disaster survey results, it could be seen that the highest elevation in the region is 4288 m, and the lowest is 496 m, with a large slope difference between regions, indicating that topographic relief varies greatly in different elevation ranges. Combined with the change in cutting degree and topographic relief, it could be seen that the landform types in the study area are mainly high-altitude hills and small rolling mountains.

2.3. Research Method

The binary pixel model for remote sensing data is a method for processing remote sensing data. In the binary pixel model, each pixel value is usually encoded into two categories, namely, changed pixels and unchanged pixels. Therefore, this model can be used to process multispectral remote sensing data and detect changes in vegetation coverage. The binary pixel model assumes that the pixel information received by the satellite sensor is composed of vegetation and soil, and the vegetation coverage (FVC) represents the percentage of pixels occupied by vegetation. This model not only has a simple and easy-to-understand theoretical basis, but also is easy to implement and apply. It is also the most widely used method for estimating vegetation coverage at present and can be applied to various types of geographical environments and vegetation types, with high versatility. Therefore, the binary pixel model is combined with remote sensing technology to use satellite image data for large-scale and efficient calculation of vegetation coverage, which can overcome the time and labor restrictions of traditional manual measurement methods, and improve the efficiency and accuracy of vegetation coverage calculation; it is also a new means and effective method for monitoring and predicting geological disasters.
In this paper, five remote sensing image data in the study area were preprocessed, the pixel binary model was used to calculate the vegetation coverage (FVC), and the normalized vegetation index (NDVI) was used to study the changes in FVC in the study area.
When analyzing the proportion of pixel vegetation coverage, the confidence degree is taken as the basis for discriminating samples, that is, a 4–5% confidence degree is selected to calculate the NDVI value in a certain period (Equation (2)). When the cumulative percentage is 5%, the NDVI value is NDVIsoil, indicating bare soil or no vegetation coverage; when the cumulative percentage is greater than or equal to 95%, the NDVI value is NDVIveg. At this time, NDVI value represents complete vegetation cover [45,46].
According to the change rule of vegetation cover area (Table 3) that obtained by calculation and analysis (Equation (3)), the vegetation cover degree of the study area was divided into four levels: lower cover, low cover, medium cover, and high cover.
NDVI = NIR − R NIR + R
where NIR is the NIR reflectance in the near-infrared band and R is the reflectance in the visible red band.
FVC = NDVI − NDVI soil NDVI veg − NDVI soil
Based on the analysis of dynamic monitoring results of vegetation growth in the study area, a correlation analysis system of surface temperature and vegetation cover index was established. This method is called Temperature Vegetation Drought Index analysis (TVDI), which is an important means for retrieving surface soil moisture [47,48,49]. The principle is to obtain the dynamic development law of vegetation growth caused by topography, geomorphology, and climate conditions through radiometric calibration, atmospheric correction, and banding of remote sensing images by using full remote sensing data. With the increase in vegetation coverage, surface vegetation will convert part of the absorbed radiant energy into heat energy, while the conversion effect of sensible heat is relatively weakened. Which causes the surface soil temperature to drop.
The feature space of NDVI-TS took NDVI as the horizontal coordinate and LST as the vertical coordinate, and the scatter plot drawn presented a triangular form (Equation (4)). The negative correlation between the slope of LST and NDVI and soil water is an important statistical feature in the feature space (triangular space). The TVDI expression was as follows:
TVDI = T S − T Smin T Smax − T Smin
where TS is the surface temperature data (LST), TSmin is the lowest LST corresponding to NDVI pixel (soil moisture sufficiency), that is, the wet edge, and TSmax is the highest LST corresponding to NDVI pixel (soil drought), that is, the dry edge. a1 and b1 are the coefficients of the dry-side linear fitting equation, and a2 and b2 are the coefficients of the wet-side linear fitting equation.
T Smax = a 1 + b 1 × NDVI
T Smin = a 2 + b 2 × NDVI
Based on 5 high-resolution Landsat remote sensing images and other data from 2018 to 2023, several sets of NDVI and LST remote sensing images were selected to fit the highest and lowest land surface values by the least square method. The slope corresponding to the dry side (Equation (5)) and the slope corresponding to the wet side (Equation (6)) of the NDVI−LST feature space were obtained. As can be seen from Figure 8, the variation amplitude of the wet edge was smaller than that of the dry edge and did not reach a parallel state with the X-axis. With the increasing NDVI value, the gap between TSmin and TSmax tended to minimize and finally presented a triangular feature space. The closer the TVDI value is to 1, the higher the degree of soil drought or the more obvious the trend is about to be dry; on the contrary, the smaller the TVDI value is, the higher the soil moisture or the more obvious the trend is about to be wet.
The calculated dry edge correlation coefficients (R2) are 0.71, 0.72, 0.67, 0.69, 0.60, and 0.71. Compared with the range of correlation coefficients (R2) calculated from traditional data (0.47–0.55), TVDI generated by high-resolution Landsat remote sensing image data can more accurately reflect the characteristics of soil moisture change.

3. Results

The binary pixel model inversion method was used to study the change rule of vegetation coverage (FVC), and a trend chart of FVC change was drawn (Figure 9). The results showed that the southern region was mainly a lower and low vegetation coverage area, but the distribution area gradually decreased, and the vegetation coverage decreased slightly or significantly. At this time, NDVI could only reflect slight changes and could not accurately indicate the specific vegetation growth amount. In the western region, the four types of coverage were distributed, and the change in vegetation coverage increased significantly. NDVI showed a significant positive correlation with the increase trend in vegetation coverage, but when the vegetation coverage increased to a certain range, the sensitivity of DVI to the change in vegetation cover decreased. In the northern region, the four kinds of coverage were also distributed, showing a significant improvement in general, and the late treatment effect was better. The eastern region showed alternating phenomena of increasing and decreasing coverage, which belonged to the state of ecological restoration. The increasing area of vegetation cover was mainly in the low sea wave and slow terrain, which was easy for ecological restoration.
In order to more directly observe the spatial distribution characteristics of soil moisture in the study area, a spatial distribution map of soil moisture levels in the study area was drawn based on the analysis results of the dry and wet boundary scatter plots of NDVI-LST feature spaces from 2018 to 2023, which were mainly divided into five levels: wet, normal, slightly arid, arid, and severely arid (Figure 10).
By comparing the data of TVDI and measured soil moisture content, the effectiveness and accuracy of TVDI as a soil moisture monitoring indicator can be verified. Therefore, this paper plots the relationship between TVDI and soil moisture content and quantifies the strength of their linear relationship by calculating the correlation coefficient (R2) (Figure 11). This provides a quantitative basis for using TVDI for soil moisture retrieval. The experimental results show that when soil moisture content is high, vegetation growth is good and the surface temperature is relatively low, resulting in lower calculated TVDI values. Conversely, when soil moisture content is low, vegetation growth is restricted and the surface temperature is relatively high, resulting in higher calculated TVDI values. The R2 value is 0.6934, reflecting high consistency between the two, thus proving the authenticity and accuracy of the experimental results.
The maximum and minimum values of vegetation coverage change in the study area were statistically analyzed, and a trend chart of mean value change was obtained (Figure 12a). The results showed that the mean value of FVC showed an increasing trend from 2019 to 2020, from 2021 to 2022, and from 2022 to 2023. On the contrary, from 2018 to 2019 and from 2020 to 2021, the average FVC showed a downward trend. The comparison between the trends in vegetation cover and soil moisture variation (Figure 12b) shows a high degree of consistency. The increase in soil moisture is primarily attributed to the increase in vegetation cover and sufficient rainfall.
However, this conclusion only suggests that the surface water content has not exceeded the critical threshold of negative correlation. When the surface water content is below the critical threshold, increasing it will have positive effects, such as increasing soil moisture and enhancing vegetation growth. Although vegetation and rainfall have a positive impact on soil moisture, studies indicate that the surface water content has not surpassed the critical threshold of negative correlation. This implies that after a certain point, the increase in moisture may no longer have a positive effect on vegetation growth and could even trigger waterlogging disasters and soil erosion.
Another innovative aspect of this paper is the proposal of the correlation between the negative correlation threshold of surface water content and the development of geological disasters, which is also a key factor in the subsequent geological disaster assessment. The negative correlation threshold of surface water content is not a fixed value but is influenced by multiple factors such as soil type, vegetation cover, and climatic conditions. When the surface water content exceeds a specific negative correlation threshold, human engineering activities may accelerate the expansion of surface cracks, further promoting the deeper infiltration of surface water. However, this infiltration phenomenon may have the opposite effect, leading to a decrease in the moisture content of the soil surface layer, directly threatening the healthy growth environment of vegetation.
Once the surface water content surpasses this critical threshold, excessive rainfall makes it difficult for the soil to rapidly absorb or effectively discharge into the water system, leading to a rapid rise in water levels and surface water accumulation, which could ultimately trigger flood disasters. Simultaneously, the stability of the soil structure decreases, creating a hidden danger for the occurrence of geological disasters such as debris flows. Under specific conditions, rainfall or other geological activities can easily trigger large-scale soil and rock slides, forming highly destructive debris flows that pose severe threats to surrounding areas. Additionally, the weakening of soil shear strength increases the risk of landslides and ground collapses, further complicating the prediction and accuracy of geological disasters. Given these potential risks, continuous monitoring of surface vegetation water content becomes particularly important. It not only allows for the assessment of vegetation growth and soil moisture levels but also serves as a critical basis for predicting geological disasters and formulating geological disaster prevention and control strategies.

4. Discussion

The E’bian region is located at the eastern edge of the Sichuan Basin, characterized by complex geological structures with diverse rock types, structural formations, and geological systems. The terrain is rugged, and transportation is inconvenient, posing challenges to both traditional geological exploration and geophysical surveys. Being remote, the area incurs higher costs for resource development, thereby limiting investments in geological exploration activities. Additionally, insufficient funding further constrains geological exploration efforts in this remote region. Advanced technological support is essential for geological exploration, including geophysical surveys, geochemical exploration, and remote sensing techniques. However, the dissemination and application of such technologies may face limitations in remote areas, leading to gaps in geological exploration data. Consequently, traditional methods dominate geological exploration in this region, often relying on on-site sampling for geological hazard monitoring, which has limited effectiveness. Both traditional geological exploration and geophysical surveys typically require extended durations to yield results, hindering real-time data acquisition and precise disaster prediction. Moreover, enhancing data accuracy demands increased resource inputs, exacerbating the constraints. These combined factors contribute to the scarcity of fundamental geological exploration data in the study area.
Currently, there are few studies on geological disasters in the E’bian area. The main methods include susceptibility assessment based on Geographic Information Systems (GISs) and slope units [50,51,52,53,54], using remote sensing technology to identify geological disasters (such as landslides and debris flows) [55,56,57,58], early identification of geological disasters through integrated sky–ground monitoring [59,60,61], and using Landsat technology to identify landslides [62,63,64]. Additionally, many experts have conducted extensive research using remote sensing technology combined with the Temperature Vegetation Drought Index (TVDI) in forestry, agriculture, and other fields [65,66,67], but there are fewer studies focused on geological disasters. Traditional soil moisture measurement methods include the gravimetric method, tensiometer method, time domain reflectometry (TDR), conductivity method, and neutron scattering method. These methods have several drawbacks in monitoring and predicting geological disasters [68,69], such as the need for on-site sampling. Since sampling points are typically sparse, they cannot comprehensively cover the entire study area. Due to significant spatial variations in soil moisture, the soil moisture at a single sampling point is difficult to use to represent the soil moisture of the entire region. These methods usually require on-site sampling, so their monitoring range is limited and time-consuming, making it difficult to achieve dynamic monitoring of the monitoring area. Compared to traditional methods, soil moisture measurement methods based on remote sensing technology have several advantages [70,71,72]. Firstly, remote sensing technology can cover large areas and provide soil moisture information globally, which is very effective for large-scale geological disaster prediction and monitoring. Remote sensing technology can provide continuous time series data, allowing us to monitor changes in soil moisture. Additionally, remote sensing technology is non-destructive for soil monitoring. Secondly, remote sensing technology can provide multispectral soil moisture information, which is useful for understanding the spatial variability and temporal dynamics of soil moisture. Furthermore, remote sensing technology can quickly obtain large amounts of soil moisture data, which is crucial for real-time geological disaster prediction and monitoring. Lastly, remote sensing technology can provide long-term soil moisture monitoring data, which is useful for understanding and predicting long-term trends in geological disasters.
This study is based on high-resolution remote sensing images from Landsat and captures subtle surface changes through detailed band data processing, providing high-precision foundational data for geological disaster monitoring and prediction. Additionally, terrain factors are extracted and classified using a digital elevation model (DEM), achieving the integration of multi-source data and providing more comprehensive and accurate foundational data support for geological disaster monitoring and prediction [73,74,75,76]. On this basis, the NDVI-LST feature space was constructed, and the Temperature Vegetation Drought Index (TVDI) was calculated and analyzed for its trends, effectively reflecting the dynamic changes in soil moisture in the study area. Based on the evaluation of geological hazard susceptibility in this study area by some scholars using Geographic Information Systems and terrain factors, this paper added the classification criteria of soil moisture critical value, which could more accurately describe the promoting and hindering effects of soil moisture on geological hazards and put forward new and effective guiding ideas for the prevention and control of geological hazards in this region. This method not only considers the impact of vegetation coverage on soil moisture but also incorporates surface temperature as a key factor, making the assessment of soil moisture more comprehensive and accurate. This study reveals the changing trends and influencing factors of soil moisture in the study area. Through the analysis of the correlation coefficient (R2), it quantifies the high consistency between soil moisture and vegetation coverage. More importantly, by classifying the critical values of soil moisture, this study accurately describes the promoting and inhibiting effects of soil moisture on geological disasters, providing a scientific basis for the prevention and control of geological disasters. The research results showed that the variation in soil moisture in the study area is relatively stable, and the overall soil moisture content is high (0.18 < TVDI < 0.33). However, due to the large variation in topographic relief, it can provide power and be a source basis for geological disasters such as landslide and collapse, so the inversion value of TVDI is small. The minimum and maximum values of correlation coefficient (R2) are 0.60 and 0.72, respectively, indicating that the surface water content is relatively large, which is in good agreement with the calculated results of vegetation coverage, and is conducive to the restoration of ecological stability.
Meanwhile, this method also can make up for the lack of geological engineering exploration data in the study area, resulting in a decrease in the accuracy of geological hazard assessment. The research results of this paper show that the special topographic characteristics of E’bian area are the basic conditions for the occurrence of geological disasters in this region. Rainfall and human activities affect the changes in soil moisture and vegetation coverage in this region. The lower the soil moisture is, the smaller the vegetation coverage is, and the surface is prone to wind and hydraulic denudation at this time, and the smaller the organic matter content is. This is not conducive to the formation of mineral rich areas with economic value. On the contrary, the higher the soil moisture, the greater the vegetation coverage, which makes it easier to form soil with a high decomposition of organic matter.
Therefore, the innovation of this paper lies in the comprehensive application of high-resolution remote sensing images, DEM, NDVI-LST feature space, and TVDI to thoroughly analyze the relationship between soil moisture and geological disasters in the E’bian area. Through detailed soil moisture change analysis and high correlation verification, it provides new methods and perspectives for geological disaster monitoring and prediction, significantly enhancing the scientific capacity and effectiveness of geological disaster prevention and control in the study area. Compared to existing studies, this paper not only expands the application scope of remote sensing technology in geological disaster research but also provides important references for practical disaster prevention and mitigation efforts.
Despite the achievements of this study, there are still some limitations. Although Landsat images have relatively high spatial resolution, higher resolution data are needed in certain situations, such as monitoring subtle geological disasters, to achieve more accurate results. Although this study used NDVI-LST feature space and the TVDI to analyze soil moisture, the occurrence of geological disasters is influenced by multiple factors, including rainfall, earthquakes, topography, vegetation, and human activities. Therefore, in practical applications, more factors need to be considered to improve the accuracy of predictions. This study focuses on E’bian Yi Autonomous County, so the results and conclusions may have some regional specificity. When extending the research results to other areas, regional differences and applicability need to be considered. Therefore, to address the limitations of this study, the following methods need to be adopted in future research for improvement and optimization:
  • Increase sample size and diversity: Expand the study area and increase the sample size to cover more types of geological disasters and soil types, which will help improve the generalizability and reliability of the research results. Additionally, consider the impacts of different seasons, climatic conditions, and human activities on geological disasters. By increasing the diversity of samples, the adaptability of the model can be enhanced.
  • Integrate more data sources: in addition to remote sensing data, it is necessary to integrate multi-source data such as meteorological data, geological survey data, topographic data, and socio-economic data.
  • Improve models and methods: Combine complex machine learning and deep learning algorithms to construct geological disaster prediction models. Additionally, incorporate spatio-temporal analysis techniques, such as spatio-temporal autocorrelation analysis and spatio-temporal clustering analysis, to more accurately capture the temporal and spatial patterns of geological disasters.
  • Consider socio-economic factors: In geological disaster monitoring and prediction, it is essential to consider the impact of socio-economic factors on disaster risk, in addition to natural factors. For instance, population density, building distribution, and infrastructure conditions may all affect the severity and extent of losses caused by geological disasters.
In summary, through increasing sample size and diversity, integrating more data sources, improving models and methods, and considering socio-economic factors, geological disaster monitoring and prediction research can be further improved and optimized.

5. Conclusions

Landsat satellite imagery, with its long coverage period and continuous data, is well-suited for long-term monitoring and change detection, enabling detailed capturing of surface changes. Additionally, it can capture temporal variations in the development of geological hazards. Due to the multiple bands covered by Landsat imagery, it provides rich spectral information that aids in extracting different surface features such as vegetation, water bodies, and soil and supports the calculation of various remote sensing indices. This study focuses on E’bian Yi Autonomous County as the research target area. Based on Landsat remote sensing imagery from 2018 to 2023, combined with in-depth analysis of digital elevation models (DEMs), a pixel binary model was used to retrieve the vegetation cover change pattern (FVC) in the study area, followed by mean statistical analysis. The key topographic factors influencing the occurrence of geological hazards were successfully extracted and finely classified. On this basis, an NDVI-LST feature space was constructed, where each pixel corresponds to an NDVI value and an LST value, visually presenting the relationship between vegetation cover and surface temperature, and further analyzing the spatial distribution of soil moisture. By fitting the obtained dry and wet edge equations, TVDI values were calculated, and ultimately, these were integrated with the development patterns of geological hazards for comprehensive study.
The innovative methods adopted in this study and their integration with the development patterns of geological hazards are mainly reflected in the following two aspects:
  • Geological Hazard Identification: By combining TVDI data and DEM data, potential geological hazard areas were identified. High TVDI values (indicating low soil moisture) and unfavorable topographic conditions (such as steep slopes and gullies) are often high-risk areas for geological hazards.
  • Development Process Analysis: Using long-term series of Landsat imagery data, the development process of geological hazards at different time points was analyzed. By comparing TVDI values and DEM characteristics at different time points, the evolution patterns and triggering mechanisms of geological hazards can be revealed.
This innovative method greatly enhances the monitoring accuracy and analytical capability of soil moisture dynamic changes in the research target area. Through a series of scientifically rigorous data processing and analysis methods, the following main conclusions were drawn:
(1)
Compared with traditional data, remote sensing images of Landsat high-resolution data can better reflect soil moisture variation rules in the study area and are highly consistent with vegetation coverage variation rules.
(2)
FVC in the study area showed an overall growth trend from 2018 to 2022, and the composition structure of FVC significantly improved, with the improvement area of higher-grade FVC (45% < FVC < 65%) being 1501 km2 and that of higher-grade FVC (65% < FVC < 100%) being 1715 km2. In general, the area increased by FVC is greater than the area decreased by FVC.
(3)
The soil moisture variation in E’bian Yi Autonomous County is relatively stable, with the overall soil moisture content at a high level (TVDI values ranging between 0.18 and 0.33). This indicates that the region maintains certain moisture conditions, which are conducive to vegetation growth and ecological stability.
(4)
The calculated correlation coefficient (R2) ranges from 0.60 to 0.72, strongly demonstrating a significant positive correlation between surface water content and vegetation cover. This finding not only validates the accuracy of remote sensing in monitoring soil moisture but also reveals the important role of vegetation in soil and water conservation and climate regulation.
(5)
TVDI can accurately reflect the spatial distribution of soil moisture in the study area. In the process of TVDI calculation, although the annual change in surface vegetation cover has an impact on the stability of the TVDI reference system, in the follow-up research, besides considering the effects of air pressure and radiation, it is also necessary to conduct detailed analysis of the development degree of ground fractures in order to increase the accuracy of the study of soil moisture change.

Author Contributions

Conceptualization, H.G. and A.M.M.-G.; methodology, H.G.; software, H.G.; validation, H.G. and A.M.M.-G.; formal analysis, H.G.; investigation, H.G. and A.M.M.-G.; resources, H.G.; data curation, H.G.; writing—original draft preparation, H.G. and A.M.M.-G.; writing—review and editing, H.G. and A.M.M.-G.; visualization, H.G.; supervision H.G. and A.M.M.-G.; project administration A.M.M.-G.; funding acquisition, A.M.M.-G. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by MCIN/AEI/10.13039/501100011033 and the GEAPAGE research group (Environmental Geomorphology and Geological Heritage) of the University of Salamanca.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Acknowledgments

GEAPAGE research group (Environmental Geomorphology and Geological Heritage) of the University of Salamanca.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Wang, X.; Hu, X.; Zhang, X. Study on zoning evaluation and management system of urban geological hazards in Leshan. Geol. Hazards Environ. Prot. 2009, 20, 25–30. [Google Scholar]
  2. Futamori, T. Lithostratigraphy and Sr isotope stratigraphy of Leshan City Middle-Late Permian carbonates in the E’bian area, Sichuan, South China. Keynote Speech Geol. Soc. Jpn. 2013, 2012, 90. [Google Scholar]
  3. Zhang, D.; Zhong, J. Development characteristics of debris flow geological hazards along Dadu River in Jinkou District, Leshan City. Geol. Hazard. Environ. Prot. 2017, 28, 17–21. (In Chinese) [Google Scholar]
  4. Lei, Q. Study on the cause mechanism and environmental effects of collapse and slide disasters in Hanyuan and Tongjie Section of Dadu River. Ph.D. Thesis, Chengdu University of Technology, Chengdu, China, 2017. [Google Scholar]
  5. Huang, W.; Chen, Y.; Zhou, W. Evaluation and prevention measures of geological hazards in Shizhong District of Leshan City, Sichuan Province. Metall. Manag. 2020, 1, 245–246. [Google Scholar]
  6. Ma, Y.; Li, C. Construction of geological hazard risk management and disaster reduction system in mountain tourism area: A case study of Leshan Mountain Scenic Area in Sichuan Province. Sci. Technol. Eng. 2019, 23, 9719–9727. [Google Scholar]
  7. Zhu, C.; Luo, J.; Shen, Z.; Huang, C. Wetland mapping in the Balqash Lake Basin Using Multi-source Remote Sensing Data and Topographic features Synergic Retrieval. In Proceedings of the 2011 3rd International Conference on Environmental Science and Information Application Technology (ESIAT 2011 Part 3–2), Beijing, China, 18–19 June 2011. [Google Scholar]
  8. Liu, J.; Bai, L.G.; Yin, S.; Jianyin, G. Suitability Evaluation of Ecological Control Measures for Typical Small Watershed Soft Rock Area Based on TVDI and NDVI changes. In 9th Annual Meeting of Risk Analysis Council of China Association for Disaster Prevention (RAC 2020); Atlantis Press: Amsterdam, The Netherlands, 2021. [Google Scholar]
  9. Han, Z.; Zhao, W. Comparison of Spatiotemporal Fusion Models for Producing High Spatiotemporal Resolution Normalized Difference Vegetation Index Time Series Data Sets. J. Comput. Commun. 2019, 7, 65–71. [Google Scholar] [CrossRef]
  10. Wang, F.; Liu, X.; Liu, X.; Li, Y.; Wang, T. Impacts of Land Use Change on NDVI in Shaanxi Province of China. In Proceedings of the 2020 6th International Conference on Energy Materials and Environment Engineering (ICEMEE 2020) (VOL.2), Tianjin, China, 24–26 April 2020. [Google Scholar]
  11. Zhou, Q. Application of Remote Sensing Technologies in Environmental Monitoring and Geological Surveys. In Proceedings of the 3rd International Conference on Materials Chemistry and Environmental Engineering (part1), Stanford, CA, USA, 18 March 2023. [Google Scholar]
  12. Chen, H. Application of remote sensing techniques in lithology identification in Almeria. In Proceedings of the 3rd International Conference on Materials Chemistry and Environmental Engineering (part2), Stanford, CA, USA, 18 March 2023. [Google Scholar]
  13. Li, Z.; Cheng, P.; Zheng, J. Prediction of time to slope failure based on a new model. Bull. Eng. Geol. Environ. 2021, 80, 5279–5291. [Google Scholar] [CrossRef]
  14. Zhang, C.; Li, Z.; Yu, C.; Chen, B.; Ding, M.; Zhu, W.; Peng, J. An integrated framework for wide-area active landslide detection with InSAR observations and SAR pixel offsets. Landslides 2022, 19, 2905–2923. [Google Scholar] [CrossRef]
  15. Niu, W.; Hu, X.; Lin, B.; Meng, F.; Zhang, Y.; Zhao, J. Detection and Monitoring of Potential Geological Disaster Using SBAS-InSAR Technology. KSCE J. Civ. Eng. 2023, 27, 4884–4896. [Google Scholar] [CrossRef]
  16. Tian, S.; Liu, X. E’Bian yi nationality autonomous county of Sichuan province geological disaster characteristics and genetic analysis. J. Jilin Water Conserv. 2015, 1, 36–40. [Google Scholar]
  17. Liu, R.; Luo, S.; Xu, Y.; Li, G.; Gou, X. Geological hazard susceptibility evaluation in E’bian County based on GIS and slope unit. Sci. Technol. Eng. 2019, 23, 7678–7685. [Google Scholar]
  18. Zheng, L.; Li, X.; Xu, R. Evaluation of regional landslide sensitivity based on slope units: A case study of Xiaojiang Basin, Yunnan Province. Sci. Technol. Eng. 2019, 21, 12322–12329. [Google Scholar]
  19. Zhou, Z. Spatial-temporal pattern change of vegetation cover in Weichang County based on Landsat remote sensing images. Hydrogeol. Eng. Geol. 2020, 47, 1–90. [Google Scholar]
  20. Liu, Z.; Chen, J.P. Relationship between temporal and spatial characteristics of vegetation coverage and topographic factors in Beijing based on Landsat-8 image data. J. Chengdu Univ. Technol. 2022, 49, 19–128. [Google Scholar]
  21. Jia, F.; Chen, M. Correlation between vegetation coverage and topographic factors in Weichang County based on Landsat remote sensing images. Li Plan. Des. 2023, 3, 11–116. [Google Scholar]
  22. Valencia Ortiz, J.A.; Martínez-Graña, A. Calculation of precipitation and seismicity thresholds as triggers for mass movements in the region of Bucaramanga, Colombia. Ecol. Indic. 2023, 152, 110355. [Google Scholar] [CrossRef]
  23. Merchán, L.; Martínez-Graña, A.; Nieto, C.; Criado, M.; Cabero, T. Geospatial Characterization of Gravitational and Erosion Risks to Establish Conservation Practices in Vineyards in the Arribes del Duero Natural Park (Spain). Agronomy 2023, 13, 2102. [Google Scholar] [CrossRef]
  24. Gillies, R.R. A method to make use of thermal infrared temperature and NDVI measurements to infer surface soil water content and fractional vegetation cover. Remote Sens. Rev. 1994, 9, 161–173. [Google Scholar]
  25. Cando Jácome, M.; Martínez-Graña, A. Numerical modeling of flow patterns applied to the analysis of the susceptibility to movements of the ground. Geosciences 2018, 8, 340. [Google Scholar] [CrossRef]
  26. Xu, X.; Sun, W.B.; Wang, Z. Analysis of soil moisture variation in Burtai mine based on TVDI. J. Min. Sci. 2019, 4, 285–291. [Google Scholar]
  27. Criado, M.; Martínez-Graña, A.M.; Santos-Francés, F.; Merchán, L. Improving the Management of a Semi-Arid Agricultural Ecosystem through Digital Mapping of Soil Properties: The Case of Salamanca (Spain). Agronomy 2021, 11, 1189. [Google Scholar] [CrossRef]
  28. Xu, X.; Wang, P.; Zhao, Y.J.; Wang, L.; Zhang, C.X. TVDI-based inversion of soil moisture in mining areas and analysis of the influence of topographic factors. Water Informatiz. 2023, 1, 40–45. [Google Scholar]
  29. Moran, M.S.; Clarke, T.R.; Inoue, Y. Estimating crop water deficit using the relation between surface-air temperature and spectral vegetation index. Remote Sens. Environ. 1994, 49, 246–263. [Google Scholar] [CrossRef]
  30. Marcelo, C.-J.; Antonio, M.-G.; Kervin, C.; Eduardo, O.-H. Satellite radar interferometry for assessing Coseismic liquefaction in Portoviejo city, induced by the Mw 7.8 2016 Pedernales, Ecuador earthquake. Environ. Earth Sci. 2020, 79, 467. [Google Scholar]
  31. Hussain, S.; Raza, A.; Abdo, H.G.; Mubeen, M.; Tariq, A.; Nasim, W.; Al Dughairi, A. ARelation of land surface temperature with different vegetation indices using multi-temporal remote sensing data in Sahiwal region, Pakistan. Geosci. Lett. 2023, 10, 56–68. [Google Scholar] [CrossRef]
  32. Kassim, K.; Ahmed, O.M.; Ching, T.D.L. Spatial-temporal mapping of forest vegetation cover changes along highways in Brunei using deep learning techniques and Sentinel-2 images. Ecol. Inform. 2023, 77, 102193. [Google Scholar]
  33. Oorthuis, R.; Vaunat, J.; Hürlimann, M.; Lloret, A.; Moya, J.; Puig-Polo, C.; Fraccica, A. Slope Orientation and Vegetation Effects on Soil Thermo-Hydraulic Behavior. An Experimental Study. Sustainability 2020, 13, 14. [Google Scholar] [CrossRef]
  34. Xiurong, Y.; Annan, J.; Shuai, Z. Analysis of the effect of freeze-thaw cycles and creep characteristics on slope stability. Arab. J. Geosci. 2021, 14, 1033. [Google Scholar]
  35. Singeisen, C.; Massey, C.; Wolter, A.; Kellett, R.; Bloom, C.; Stahl, T.; Jones, K. Mechanisms of rock slope failures triggered by the 2016 Mw 7.8 Kaikōura earthquake and implications for landslide susceptibility. Geomorphology 2022, 415, 108386. [Google Scholar] [CrossRef]
  36. Syzdykbayev, M.; Karimi, B.; Karimi, H.A. A Method for Extracting Some Key Terrain Features from Shaded Relief of Digital Terrain Models. Remote Sens. 2020, 12, 2809. [Google Scholar] [CrossRef]
  37. Xiong, L.; Tang, G.; Yang, X.; Li, F. Geomorphology-oriented digital terrain analysis: Progress and perspectives. J. Geogr. Sci. 2021, 31, 456–476. [Google Scholar] [CrossRef]
  38. Jordan, G.; CSILLAC, G.; Szucs, A.; Qvarfort, U. Application of digital terrain modelling and GIS methods for the for the morphotectonic investigation of the Kali Basin, Hungary. Z. Für Geomorphol. 2003, 47, 145–169. [Google Scholar] [CrossRef]
  39. Shirani, K.; Solhi, S.; Pasandi, M. Automatic Landform Recognition, Extraction, and Classification using Kernel Pattern Modeling. J. Geovisualization Spat. Anal. 2023, 7, 2. [Google Scholar] [CrossRef]
  40. Kim, Y.-P. Analysis of Distribution Characteristics According to TPI(Topographic Position Index) In Gayasan National Park Using GIS. J. Korean Inst. For. Recreat. 2011, 15, 73–82. [Google Scholar]
  41. Muddarisna, N.; Yuniwati, E.D.; Masruroh, H.; Oktaviansyah, A.R. An Automated Approach Using Topographic Position Index (TPI) for Landform Mapping (Case Study: Gede Watershed, Malang Regency, East Java, Indonesia). IOP Conf. Ser. Earth Environ. Sci. 2020, 412, 012027. [Google Scholar] [CrossRef]
  42. Tang, F.Q.; Li, W.W.; Gu, J. Study on the characteristics of soil moisture changes caused by mining subsidence in loess mining areas. Soil Bull. 2019, 50, 1139–1144. [Google Scholar]
  43. Criado, M.; Santos-Francés, F.; Martínez-Graña, A.M.; Sánchez-Sánchez, Y.; Merchán, L. Multitemporal Analysis of Soil Sealing and Land Use Changes Linked to Urban Expansion of Salamanca (Spain) Using Landsat Images and Soil Carbon Management as a Mitigating Tool for Climate Change. Remote Sens. 2020, 12, 1131. [Google Scholar] [CrossRef]
  44. Guo, H.; Martínez-Graña, A.M. Susceptibility of Landslide Debris Flow in Yanghe Township Based on Multi-Source Remote Sensing Information Extraction Technology (Sichuan, China). Land 2024, 13, 206. [Google Scholar] [CrossRef]
  45. Cando-Jácome, M.; Martínez-Graña, A.M.; Valdés, V. Detection of terrain deformations by InSAR techniques and geophysical methods (Zaruma City, Ecuador). Remote Sens. 2020, 12, 1598. [Google Scholar]
  46. Sandholt, I.; Rasmussen, K.; Andersen, J. A simple interpretation of the surface temperature/vegetation index space for assessment of surface moisture statues. Remote Sense Environ. 2002, 79, 213–224. [Google Scholar] [CrossRef]
  47. Zhang, D.H.; Ren, Z.Y.; Wang, X.F. Change characteristics and driving factors of vegetation coverage in Shaanxi Loess Plateau based on MODIS. J. Ecol. Rural. Environ. 2013, 29, 29–35. [Google Scholar]
  48. Wang, Y.T.; Kong, J.L.; Yang, L.Y. SVR-based remote sensing inversion of soil moisture under sparse vegetation cover in early zone. J. Geoinf. Sci. 2019, 21, 1275–1283. [Google Scholar]
  49. Yan, N.; Zeyue, M.; Profeng, Z.; Lei, Y.; Jing, Y. Research on soil moisture inversion and monitoring in the Aksu River basin. J. Ecol. 2019, 39, 5138–5148. [Google Scholar]
  50. Nefros, C.; Loupasakis, C.; Alatza, S.; Kontoes, C. Creating a Comprehensive Landslides Inventory Using Remote Sensing Techniques and Open Access Datasup †/sup. Environ. Sci. Proc. 2024, 29, 60. [Google Scholar]
  51. Lu, H.; Li, W.; Xu, Q.; Yu, W.; Zhou, S.; Li, Z.; Zhan, W.; Li, W.; Xu, S.; Zhang, P.; et al. Active landslide detection using integrated remote sensing technologies for a wide region and multiple stages: A case study in southwestern China. Sci. Total Environ. 2024, 931, 172709. [Google Scholar] [CrossRef] [PubMed]
  52. Galone, L.; Villani, F.; Colica, E.; Pistillo, D.; Baccheschi, P.; Panzera, F.; D’Amico, S. Integrating near-surface geophysical methods and remote sensing techniques for reconstructing fault-bounded valleys (Mellieha valley, Malta). Tectonophysics 2024, 875, 230263. [Google Scholar] [CrossRef]
  53. Rash AJ, H.; Khodakarami, L.; Muhedin, D.A.; Hamakareem, M.I.; Ali, H.F.H. Spatial modeling of geotechnical soil parameters: Integrating ground-based data, RS technique, spatial statistics and GWR model. J. Eng. Res. 2024, 12, 75–85. [Google Scholar] [CrossRef]
  54. Selvam, G.; Anbarasu, S. Identification of groundwater potential zones using multi-influencing factor method, GIS and remote sensing techniques in the hard rock terrain of Madurai district, southern India. Sustain. Water Resour. Manag. 2024, 10, 54. [Google Scholar]
  55. Zhou, B.; Zhang, S.; Xue, R.; Li, J.; Wang, S. A review of Space-Air-Ground integrated remote sensing techniques for atmospheric monitoring. J. Environ. Sci. 2021, 123, 13–14. [Google Scholar] [CrossRef]
  56. He, P.; Guo, Z.; Chen, H.; Shi, P.; Zhou, X.; Wang, G. Research and Application of Early Identification of Geological Hazards Technology in Railway Disaster Prevention and Control: A Case Study of Southeastern Gansu, China. Sustainability 2023, 15, 16705. [Google Scholar] [CrossRef]
  57. Wu, Z.; Ye, R.; Yang, S.; Wen, T.; Huang, J.; Chen, Y. Study on Early Identification of Rainfall-Induced Accumulation Landslide Hazards in the Three Gorges Reservoir Area. Remote Sens. 2024, 16, 1669. [Google Scholar] [CrossRef]
  58. Wang, H.; Zhang, J.; Gong, Y.; Wang, G.; Hu, Z. Early Identification of Geological Disasters in Loess Hilly Area of Ningnan Based on InSAR Technology: A case study of Yuanzhou District, China. J. Phys. Conf. Ser. 2024, 2706, 012095. [Google Scholar] [CrossRef]
  59. Yan, Y.; Zhuang, Q.; Zan, C.; Ren, J.; Yang, L.; Wen, Y.; Kong, L. Using the Google Earth Engine to rapidly monitor impacts of geohazards on ecological quality in highly susceptible areas. Ecol. Indic. 2021, 132, 108258. [Google Scholar] [CrossRef]
  60. Xichao, H.; Meng, W.; Bing, H.; Tianbin, Y.; Yu, J. Study on Early Identification of Landslide Hazard in Mountain Valley Area based on InSAR and Optical Remote Sensing Technology. IOP Conf. Ser. Earth Environ. Sci. 2020, 570, 062047. [Google Scholar] [CrossRef]
  61. El-Arafy, R.A.; Abd El Salam, H.F.; Shaheen, M.A.; Gawad, A.E.A. Integrative mapping of radioactive and alteration zones in Um Had plutons, Egypt: Using Landsat 9, ASTER imagery, and airborne geophysical data. J. Afr. Earth Sci. 2024, 216, 105313. [Google Scholar] [CrossRef]
  62. Zormand, S.; Jafari, R.; Koupaei, S.S. Assessment of PDI, MPDI and TVDI drought indices derived from MODIS Aqua/Terra Level 1B data in natural lands. Nat. Hazard. 2017, 86, 757–777. [Google Scholar] [CrossRef]
  63. Yuan, L.; Li, L.; Zhang, T.; Chen, L.; Zhao, J.; Hu, S.; Liu, W. Soil Moisture Estimation for the Chinese Loess Plateau using MODIS-derived ATI and TVDI. Remote Sens. 2020, 12, 3040. [Google Scholar] [CrossRef]
  64. Lu, X.J.; Zhou, B.; Yan, H.B.; Luo, L.; Huang, Y.H.; Wu, C.L. Remote Sensing Retrieval Of Soil Moisture In Guangxi Based On Ati And Tvdi Models. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2020, XLII-3/W10, 895–902. [Google Scholar] [CrossRef]
  65. Singla, J.G.; Trivedi, S.; Pandya, M.R. Two-Dimensional and 3D Change Detection in Urban Area Using Very High-Resolution Satellite Data and Impact of Urbanization over LST and NDVI. J. Indian Soc. Remote Sens. 2023, 51, 1955–1970. [Google Scholar] [CrossRef]
  66. Demisse, M.N.; Solomon, H.; Kefelegn, G. Urban land use, land cover change and urban microclimate dynamics in Addis Ababa, Ethiopia. Discov. Environ. 2024, 2, 71. [Google Scholar]
  67. Al-Harbi, H.F.; Alhuqail, A.A.; Islam, Z.; Ghrefat, H. Vegetation trends and dynamics in Shada Mountain, Saudi Arabia, (1984–2023): Insights from Google Earth Engine and R analysis. Front. Environ. Sci. 2024, 12, 1397825. [Google Scholar] [CrossRef]
  68. Valencia Ortiz, J.A.; Martínez-Graña, A.; Mejía Mendez, L. Evaluation of Susceptibility by Mass Movements through Stochastic and Statistical Methods for a Region of Bucaramanga, Colombia. Remote Sens. 2023, 15, 4567. [Google Scholar] [CrossRef]
  69. Han, T.; Pan, X.; Wang, X. Evaluating and improving the sand storm numerical simulation performance in Northwestern China using WRF-Chem and remote sensing soil moisture data. Atmos. Res. 2021, 251, 105411. [Google Scholar] [CrossRef]
  70. Chen, S.; She, D.; Zhang, L.; Guo, M.; Liu, X. Spatial Downscaling Methods of Soil Moisture Based on Multisource Remote Sensing Data and Its Application. Water 2019, 11, 1401. [Google Scholar] [CrossRef]
  71. Liu, H.; Shi, J.; Yu, F. Study on the Method of Retrieving Surface Soil Moisture in Western Semi-arid Area by Remote Sensing. Int. J. Front. Eng. Technol. 2022, 4, 19–24. [Google Scholar]
  72. Basak, A.; Schmidt, K.M.; Mengshoel, O.J. From data to interpretable models: Machine learning for soil moisture forecasting. Int. J. Data Sci. Anal. 2022, 15, 21–24. [Google Scholar] [CrossRef] [PubMed]
  73. Selamat, S.N.; Abd Majid, N.; Taha, M.R.; Osman, A. Application of geographical information system (GIS) using artificial neural networks (ANN) for landslide study in Langat Basin, Selangor. IOP Conf. Ser. Earth Environ. Sci. 2022, 1064, 012052. [Google Scholar] [CrossRef]
  74. Hariyanto, T.; Nurwatik, N.; Ramandany, V. Analysis of Flood-Vulnerable Areas Using Scoring and Weighting Methods Based on Geographic Information Systems (Case Study: Sidoarjo Regency). IOP Conf. Ser. Earth Environ. Sci. 2023, 1276, 012074. [Google Scholar] [CrossRef]
  75. Yang, Z.; Lu, H.; Zhang, Z.; Liu, C.; Nie, R.; Zhang, W.; Fan, G.; Chen, C.; Ma, L.; Dai, X.; et al. Visualization analysis of rainfall-induced landslides hazards based on remote sensing and geographic information system-an overview. Int. J. Digit. Earth 2023, 16, 2374–2402. [Google Scholar] [CrossRef]
  76. Yulianto, W.Y.; Cholil, M. Landslide Susceptibility Levels Analysis Based on Geographic Information System (GIS) in Simo District, Boyolali Regency. IOP Conf. Ser. Earth Environ. Sci. 2024, 1357, 012044. [Google Scholar] [CrossRef]
Applsci 14 06685 g001
Figure 1. Geological map of the study area.
Figure 1. Geological map of the study area.
Applsci 14 06685 g001
Applsci 14 06685 g002
Figure 2. Digital elevation model of the study area.
Figure 2. Digital elevation model of the study area.
Applsci 14 06685 g002
Applsci 14 06685 g003
Figure 3. Satellite map of the study area.
Figure 3. Satellite map of the study area.
Applsci 14 06685 g003
Applsci 14 06685 g004
Figure 4. Flow chart.
Figure 4. Flow chart.
Applsci 14 06685 g004
Applsci 14 06685 g005
Figure 5. Slope map of study region from 2018 to 2023.
Figure 5. Slope map of study region from 2018 to 2023.
Applsci 14 06685 g005
Applsci 14 06685 g006
Figure 6. Aspect map of study region from 2018 to 2023.
Figure 6. Aspect map of study region from 2018 to 2023.
Applsci 14 06685 g006
Applsci 14 06685 g007
Figure 7. Fluctuation map of study region from 2018 to 2023.
Figure 7. Fluctuation map of study region from 2018 to 2023.
Applsci 14 06685 g007
Applsci 14 06685 g008
Figure 8. NDVI−Ts Feature Space Conceptual Model.
Figure 8. NDVI−Ts Feature Space Conceptual Model.
Applsci 14 06685 g008
Applsci 14 06685 g009
Figure 9. Classification chart of changes from 2018 to 2023.
Figure 9. Classification chart of changes from 2018 to 2023.
Applsci 14 06685 g009
Applsci 14 06685 g010
Figure 10. Distribution of soil moisture levels in the study area during 2018–2023. (a–f) are the changes in the study area from 2018 to 2023.
Figure 10. Distribution of soil moisture levels in the study area during 2018–2023. (a–f) are the changes in the study area from 2018 to 2023.
Applsci 14 06685 g010
Applsci 14 06685 g011
Figure 11. The correlation between TVDI and soil moisture content.
Figure 11. The correlation between TVDI and soil moisture content.
Applsci 14 06685 g011
Applsci 14 06685 g012
Figure 12. (a) Trend diagram of the mean value of FVC from 2018 to 2023; (b) trend diagram of the mean value of TVDI from 2018 to 2023.
Figure 12. (a) Trend diagram of the mean value of FVC from 2018 to 2023; (b) trend diagram of the mean value of TVDI from 2018 to 2023.
Applsci 14 06685 g012
Table 1. Geological disaster statistics (topographic factors).
Table 1. Geological disaster statistics (topographic factors).
Geological Disaster Distribution Statistics (by Elevation)
Type<600 m600–1500 m1500–2400 m2400–3300 m>3300 m
Landslide492716No survey data
Collapse24172
Debris flow391
Dangerous rock532
Terrain subsidence031
Geomorphology unit area280.72 km2518.43 km21017.87 km2
Disaster point density0.3780.1470.0320
Geological Disaster Distribution Statistics (by Slope)
Type<15°15°–30°30°–45°45°–60°>60°
Landslide388240
Collapse225134
Debris flow43682
Terrain subsidence031100
Slope area681.75 km2765.31 km2441.72 km2288.65 km264.09 km2
Disaster point density0.0120.0210.3120.0790.083
Table 2. Information of each band.
Table 2. Information of each band.
BandBand ColorWave Length/umResolution/m
Band 1Blue band0.45–0.5230
Band 2Green band0.52–0.6030
Band 3Red band0.63–0.6930
Band 4Near-infrared band0.76–0.9030
Table 3. Proportion of vegetation cover (%) area from 2018 to 2022.
Table 3. Proportion of vegetation cover (%) area from 2018 to 2022.
GradeLower
Coverage Area		Low
Coverage Area		Medium
Coverage Area		High
Coverage Area
201854.339.15.291.32
201984.715.20.070.54
2020-53.3117.921.230.8
2020-60.667.9527.834.4
202115.927.27.9525.8
202233.749.15.979.27
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

Guo, H.; Martínez-Graña, A.M. Correlation between Soil Moisture Change and Geological Disasters in E’bian Area (Sichuan, China). Appl. Sci. 2024, 14, 6685. https://doi.org/10.3390/app14156685

AMA Style

Guo H, Martínez-Graña AM. Correlation between Soil Moisture Change and Geological Disasters in E’bian Area (Sichuan, China). Applied Sciences. 2024; 14(15):6685. https://doi.org/10.3390/app14156685

Chicago/Turabian Style

Guo, Hongyi, and Antonio Miguel Martínez-Graña. 2024. "Correlation between Soil Moisture Change and Geological Disasters in E’bian Area (Sichuan, China)" Applied Sciences 14, no. 15: 6685. https://doi.org/10.3390/app14156685

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