[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Structural Equation Models to Determine the Relationship Between Startup Incubation Stages and the Graduation Rate of Incubators in Spain
Previous Article in Journal
A Sustainability-Oriented Spatial Multi-Criteria Decision Analysis Framework for Optimizing Recreational Ecological Park Development
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
Sustainability
Volume 17
Issue 2
10.3390/su17020732
sustainability-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
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Disaster-Pregnant Environment Stability Evaluation of Geohazards in the Yellow River–Huangshui River Valley, China

by
Tengyue Zhang
1,
Qiang Zhou
1,2,3,*,
Weidong Ma
1,2,3,
Yuan Gao
1,
Hanmei Li
1 and
Qiuyang Zhang
1
1
School of Geography, Qinghai Normal University, Xining 810008, China
2
Academy of Plateau Science and Sustainability, Xining 810008, China
3
School of National Safety and Emergency Management, Qinghai Normal University, Xining 810008, China
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(2), 732; https://doi.org/10.3390/su17020732
Submission received: 29 November 2024 / Revised: 12 January 2025 / Accepted: 13 January 2025 / Published: 17 January 2025
Browse Figures
Figure 1
<p>Overview of the study area.</p> "> Figure 2
<p>Density map of geohazard sites in the YHV.</p> "> Figure 3
<p>Random forest schematic.</p> "> Figure 4
<p>Spatial characterization of the static disaster-pregnant environment in the study area. (<b>a</b>) Lithology, (<b>b</b>) fracture zone density, (<b>c</b>) topographic elevation, (<b>d</b>) slope, (<b>e</b>) river network density, (<b>f</b>) vegetation cover.</p> "> Figure 5
<p>Spatial characterization of the dynamic disaster-pregnant environment in the study area, 2003–2022. (<b>a</b>) R99p, (<b>b</b>) CWDs, (<b>c</b>) average annual rainfall, (<b>d</b>) monthly maximum temperature, (<b>e</b>) maximum monthly temperature difference, (<b>f</b>) average annual temperature, (<b>g</b>) human activity.</p> "> Figure 6
<p>Importance of the geohazard dynamic disaster-pregnant environment based on random forest.</p> "> Figure 7
<p>Inter-annual variation in the dynamic disaster-pregnant environment. (<b>a</b>) Average annual temperature, (<b>b</b>) maximum monthly temperature difference, (<b>c</b>) monthly maximum temperature, (<b>d</b>) human activity, (<b>e</b>) average annual rainfall, (<b>f</b>) R99p, (<b>g</b>) CWDs.</p> "> Figure 8
<p>Trend features and significance tests of the dynamic disaster-pregnant environment. (<b>a</b>) Average annual temperature, (<b>b</b>) average annual rainfall, (<b>c</b>) maximum monthly temperature difference, (<b>d</b>) monthly maximum temperature, (<b>e</b>) R99p, (<b>f</b>) CWDs, (<b>g</b>) human activity.</p> "> Figure 9
<p>Distribution of the stability of the static disaster-pregnant environments.</p> "> Figure 10
<p>Stability distribution of the dynamic disaster-pregnant environments.</p> "> Figure 11
<p>Distribution of the stability of the comprehensive disaster-pregnant environments.</p> "> Figure 12
<p>Comprehensive disaster-pregnant environment stability trend features.</p> ">
Versions Notes

Abstract

:
This study aims to identify the key factors contributing to the destabilization of the geohazard disaster-pregnant environment in the Yellow River–Huangshui River Valley and provide a robust scientific basis for proactive disaster prevention, management of disaster chains, and mitigation of multi-hazard clusters in unstable regions. The research focuses on the Yellow River–Huangshui River Valley, evaluating the stability of its geohazard disaster-pregnant environment. The disaster-pregnant environment is classified into static and dynamic categories. The static disaster-pregnant environment encompasses factors such as lithology, fracture density, topography, slope, river network density, and vegetation cover. The dynamic disaster-pregnant environment incorporates variables such as extreme rainfall, consecutive rainy days, annual rainfall averages, monthly high temperatures, monthly maximum temperature variations, average annual air temperatures, and human activities. A random forest model was employed to quantitatively assess the stability of the geohazard disaster-pregnant environment in the Yellow River–Huangshui River Valley. The findings indicated that (1) extreme indicators were the primary contributors to the destabilization of the disaster-pregnant environment, with very heavy rainfall contributing 28% and consecutive rainy days contributing 27%. Human activities ranked next, accounting for 15%. (2) Unstable regions for static, dynamic, and integrated disaster-pregnant environments accounted for 44%, 45%, and 44% of the study area, respectively, with all unstable areas concentrated in river valley regions. (3) The overall trend of stability in the disaster-pregnant environment was characterized by widespread instability. Extremely unstable areas were predominantly located in river valley regions, largely influenced by human activities. Conversely, only 0.1% of the region exhibited signs of stability, and 2.1% showed a tendency toward extreme stability.

1. Introduction

The concept of “disaster-pregnant environment stability” originates from the regional disaster system theory, which identifies disaster-causing factors, disaster-pregnant environments, and disaster-bearing bodies as essential components in the process of disaster formation [1,2]. The process of regional disaster formation is understood through regional disaster dynamics, encompassing the risk of disaster-causing factors, the stability of the disaster-pregnant environments, and the vulnerability of disaster-bearing bodies, which collectively form the functional framework of the regional disaster system. In the functional framework of disasters, all three components play equally vital roles, and disaster risk is collectively influenced by their interactions [1,2]. Although the disaster-causing factors, disaster-pregnant environments, and disaster-bearing bodies hold equally important roles within the functional framework of the regional disaster system, their fundamental characteristics remain distinct. In the process of disaster formation, disaster-causing factors emerge as a result of interactions with the disaster-bearing bodies [3,4,5,6,7,8,9,10], while the disaster-bearing bodies serve as the targets of these factors [11,12]. The disaster-pregnant environment reflects changes in regional and global environmental conditions [13]. The assessment of disaster-causing factors’ risk and disaster-bearing bodies’ vulnerability primarily relies on post-disaster analysis, which reflects the severity of disasters. In contrast, evaluating the stability of disaster-pregnant environments focuses on identifying unstable areas and preparing for potential disaster types and formation patterns in these regions ahead of time.
The stability of the disaster-pregnant environment is reflected in the fluctuating characteristics of its indicators and its capacity to resist disasters. This stability is further characterized by its spatial distribution, which exhibits both zonal and non-zonal patterns [14]. Its impact on disasters lies not in their severity but in the dynamic changes of the disaster-pregnant environment. These changes—such as fluctuations, gradual shifts, sudden alterations, and other dynamic transformations—can trigger a series of chain reactions that influence the disaster incubation process [14], for example, variability in precipitation and temperature. This results in the overall instability of the regional disaster incubation environment and the destabilization of the entire disaster system. It also triggers the emergence of new disasters and novel disaster combinations, disrupting existing disaster preparedness strategies and leading to greater disaster losses and increased risks. Therefore, the stability of the disaster-pregnant environment serves as a fundamental prerequisite and foundation for understanding changes in disaster dynamics.
The evaluation of disaster-pregnant environment stability involves delineating the extent of dynamic changes within the environment [15]. Researchers supporting the theory of the disaster-pregnant environment argue that the increased frequency of disasters in recent years is closely linked to regional and global environmental changes, particularly shifts in climate and land cover [13,16,17,18,19]. Therefore, this study categorizes the disaster-pregnant environment into static and dynamic types to evaluate its stability. The static disaster-pregnant environment is defined by the background conditions associated with geological disasters, which remain relatively unchanged over short periods. In contrast, the dynamic disaster-pregnant environment is determined by meteorological factors and human activities. With global warming, dynamic changes in precipitation, temperature, and other triggering factors will inevitably alter the risk and susceptibility of geological hazards. Therefore, the dynamic disaster-pregnant environment is used to identify the triggering factors that influence the stability of the disaster-pregnant environment. Additionally, the dynamic disaster-pregnant environment drives the evolution of the static disaster-pregnant environment, ultimately affecting the overall stability of the disaster-pregnant environment.
From a theoretical perspective, the current evaluation of disaster-pregnant environment stability has yet to establish a comprehensive and complete theoretical framework. Research on the stability of disaster-pregnant environments primarily relies on the regional disaster system theory, which assesses disaster risk by integrating the geologic disasters of disaster-causing factors with the vulnerability of disaster-bearing bodies [20,21,22,23]. While significant progress has been made in evaluating the risks posed by disaster-causing factors and the vulnerability of disaster-bearing bodies, standalone assessments of the stability of disaster-pregnant environments remain insufficiently explored. From a methodological perspective, the evaluation of disaster-pregnant environment stability is predominantly confined to the realm of mathematical modeling. For example, Li Ziwei [24] developed a flood risk evaluation model based on three dimensions: the risk of disaster-causing factors, the sensitivity of the disaster-pregnant environment, and the vulnerability of the disaster-bearing body, utilizing the entropy weighting and hierarchical analysis methods. Slope direction and topographic relief were selected to represent the sensitivity of the disaster-pregnant environment. Based on the scientific theory of disaster risk, a risk assessment model and index system for coal mine flood disasters were constructed by Sun Zuo, and a risk assessment method was proposed based on the projection pursuit and fuzzy cluster analysis. The evaluation of the stability of the disaster-pregnant environment mainly considers the terrain and river around the mine, the hydrogeology type of the mine, the number of wellheads, the current situation of flood prevention, and drainage capacity [20]. Drawing on hazard system theory and a comprehensive analysis of hazard risk indices, including disaster-causing factors and disaster-pregnant environments, as well as the vulnerability of the disaster-bearing bodies, Wang Xiujie selected eight risk assessment indices to construct an ice-jam flood hazard risk assessment model that uses the random forest (RF) algorithm. The three disaster-pregnant environment indices were elevation, terrain slope, and the distance to the river channel [25]. However, the above evaluations relied solely on subsurface data, overlooking the dynamic impacts of climate change, such as increased temperatures and rainfall, which introduce new characteristics and trends to the disaster-pregnant environment. Additionally, the continuous expansion of human activities further exacerbates the instability of the disaster-pregnant environment. Therefore, the evaluation of disaster-pregnant environment stability must incorporate meteorological and human activity factors to adapt to evolving environmental conditions. Furthermore, traditional mathematical models for calculating weights are susceptible to human bias. In contrast, the random forest model, recognized for its superior performance among machine learning models, is widely applicable across various domains [26]. As a result, significant opportunities remain to improve both the selection of indicators and the evaluation methods for assessing the stability of disaster-pregnant environments.
In recent years, under the combined influence of climatic and geological activities, the Yellow River–Huangshui River Valley (hereinafter referred to as “YHV”) has shown an increasing trend in the frequency and intensity of geologic disasters, including debris flows, collapses, and landslides. According to statistics from the People’s Government of Qinghai Province, the 8.17 Datong flash floods in 2022 caused direct economic losses of CNY 690 million. In 2023, several geologic disasters occurred in the YHV, affecting 3158 people and resulting in direct economic losses of CNY 436 million. Furthermore, on 3–4 September 2024, exceptionally heavy rainfall triggered landslides in multiple counties and cities in eastern Qinghai Province. These data indicate that the YHV experiences a high frequency of geologic disasters in Qinghai Province, with most of these events triggered by sudden geologic hazards caused by exceptionally heavy rainfall. This type of sudden disaster, triggered by exceptionally heavy rainfall, is often met with a lack of response preparedness compared to routine disasters. Additionally, this series of frequent geological disasters suggests that the disaster-pregnant environment in this region has experienced significant changes, disrupting its original stability and leading to the frequent and sudden occurrence of geological disasters. Therefore, this study uses the YHV as a case study, selecting both static and dynamic disaster disaster-pregnant environments as evaluation indices. It replaces the traditional subjective weighting method with the random forest model to calculate index weights, thereby completing the evaluation of the stability of the disaster-pregnant environment. Identifying the factors inducing instability in the geological disaster environment of the YHV and providing a scientific basis for early warning and response to potential disasters, disaster chains, or multi-disaster clusters in unstable areas can effectively mitigate losses caused by geological disasters. Additionally, it offers a theoretical foundation for the development of an evaluation system to assess the stability of the disaster-pregnant environment.

2. Research Methods and Data Sources

2.1. Study Areas

The YHV is situated on the northeastern edge of the Tibetan Plateau, spanning latitudes 35.01° to 37.87° N and longitudes 100.38° to 103.07° E (Figure 1), with elevations ranging from 1605 to 5232 m. The region primarily consists of Xining City, Datong County, Huangzhong District, Huangyuan County, Ledu District, Pingan District, Minhe County, Huzhu County, Hualong County, Xunhua County, Menyuan County, Tongren County, Jianzha County, Guide County, and 14 other districts, spanning an area of 35,789.80 km2. The region is situated in the transition zone between the Tibetan Plateau and the Loess Plateau, as well as the interface between the eastern monsoon region and the northwestern arid zone. The geological conditions in the area are complex, with bedrock primarily composed of Cretaceous siltstone, mudstone, thin sandstone gravel, and other unstable rock layers. Additionally, Cenozoic Tertiary laterite and Quaternary loess are prominently developed [27]. The region is shaped by crustal tectonics and plate movements [28]. Since the Late Cenozoic, tectonic deformation has been intense, with the area characterized by retrograde folded fractures, strike-slip faults, sustained crustal shortening, and steep shearing. It has become the forefront of the Tibetan Plateau’s extension [29,30,31]. The plates surrounding the region include the Longxi Plate, West Qinling Plate, Alashan Plate, Ordos Plate, and other sub-plates [30]. In the current tectonic and geomorphic framework, the southern margin of the Alxa Massif has emerged as the latest addition to the zone. In the current tectonic and geomorphologic framework, the southern edge of the Alxa Massif has emerged as the newest component of the region, intensifying fault activity and creating favorable conditions for the development of geologic disasters [32].
The YHV is shaped by the alluvial valleys of the Huangshui River and the Yellow River, representing a typical river basin geomorphology [33]. Annual precipitation ranges from 166.4 to 646.5 mm, with the majority occurring between May and October [34]. The fragmented rock bodies along rivers and channels are highly vulnerable to geologic disasters, particularly following erosion caused by heavy rains [35]. Based on collected information, 1663 geohazard sites were identified in the district, categorized into six major types, including avalanches, subsidence, mudslides, ground cracks, landslides, and slopes. The data were compiled up to approximately 2019. Among these, there are 186 collapses, 2 ground fissures, 11 ground collapses, 513 landslides, 525 mudflows, and 426 slopes. The majority of these geologic disasters, including avalanches and landslides (Figure 2), are primarily concentrated in the central valley area. Administratively, they are mainly distributed across Ping’an District, Ledu District, Xining City, Huzhu County, Datong County, Huangyuan County, and surrounding areas. These geohazards exhibit characteristics such as wide distribution, multi-faceted manifestations, uneven spatial distribution, and localized clustering.

2.2. Data Sources

①
Geohazard site data
The geohazard data were sourced from the Resource and Environmental Science and Data Center, covering six major categories: collapses, subsidence, mudslides, geocracks, landslides, and slopes. These data were compiled up to approximately 2019. The kernel density of geohazard points was calculated using ArcGIS (10.8) to evaluate the stability of the disaster-pregnant environment.
②
Subsurface data
The DEM data were obtained from the Geospatial Data Cloud with a spatial resolution of 30 m × 30 m. Using ArcGIS (10.8), slope, elevation, and river network density were extracted from the DEM data.
The lithology data were sourced from the National Seismic Science Data Center, with a resolution of 250 m × 250 m. Using the ArcGIS (10.8) reclassification tool, the data were categorized into hard, medium-soft and hard, medium-hard and soft, and soft and loose rock groups.
The rupture data were sourced from the National Tibetan Plateau Science Data Center at a resolution of 1:4 million. Rupture density was calculated using ArcGIS (10.8) to represent the stability of the disaster-pregnant environment in relation to crustal motion activity.
Using the monthly NDVI raster data from the MOD13A3 dataset under the MODIS dataset, the maximum synthesis method was applied to generate annual NDVI raster data for the years 2003 to 2022. The spatial distribution of the multi-year average NDVI in the YHV was quantified using ArcGIS (10.8) at a resolution of 1 km × 1 km, with the data sourced from EARTHDATA.
③
Meteorological data
The spatial distribution of multi-year average temperature and rainfall in the YHV was quantified using ArcGIS (10.8), based on annual average temperature and rainfall data (2003–2022) with a resolution of 1 km × 1 km. The data were sourced from the Earth Resources Data Cloud.
The spatial distribution of multi-year monthly maximum temperature and maximum temperature difference in the YHV was quantified using ArcGIS (10.8), based on monthly-scale data (2003–2022) with a resolution of 1 km × 1 km, sourced from the National Tibetan Plateau Science Data Center.
According to the extreme climate indicators recommended by the WMO, continuous rainy days (CWDs) and extreme rainfall (R99p) were calculated using meteorological station data. Continuous rainy days (CWDs) represent the longest stretch of consecutive rainy days, while extreme rainfall (R99p) refers to the total precipitation for days with daily rainfall exceeding the 99th percentile of the interquartile range. These calculations were performed using the RClimDex module in R language (R4.4.2). Subsequently, the CWD and R99p data were interpolated to 30 m × 30 m resolution raster data using the inverse distance weighting interpolation method in ArcGIS (10.8).
④
Human activity data
The human footprint for the years 2000–2020 was selected as the human activity data, which was obtained from the UEMM team and reflects eight variables such as built environment, population density, nighttime lighting, farmland, pastureland, roads, railroads, and navigable waterways for different aspects of human pressure. The data are in the form of a raster with image element values ranging from (0, 50), where [0, 1) means not affected by human activities, [1, 4) means slightly affected by human activities, and [4, 50] means more affected by human activities. Multi-year average human activity data for the YHV were quantified using ArcGIS (10.8).
The above data were processed at a resolution of 30 m × 30 m and normalized to evaluate the stability of the geohazard-bearing environment, ensuring the elimination of magnitude discrepancies between the indicators.

2.3. Research Methods

2.3.1. The Random Forest Model

The analysis reveals that support vector machines and random forests outperform neural networks, with the random forest model achieving the highest accuracy at 96.3%. Consequently, the random forest model is selected for this study (Figure 3) [26,36].
Random forest, introduced by Leo Breiman, is a data mining technique used for classification and regression using decision trees. The model is a nonparametric supervised machine learning classifier that utilizes multiple aggregated random decision trees to classify datasets based on the prediction patterns of all decision trees [37]. By combining the modeling capabilities of decision trees with the benefits of ensemble learning, the model offers high flexibility and robustness [36]. During the training process of a random forest, samples are randomly selected from the original dataset to form subsets. Each decision tree is trained on a different set of samples and feature sets. A hierarchical decision tree is constructed by recursively partitioning the feature space into smaller subspaces. Increasing the number of decision trees enhances the accuracy of the random forest model’s predictions. Moreover, a higher number of trees reduces the likelihood of overfitting [38]. In random forest classifiers, the final result is determined through majority voting. Each decision tree outputs a classification, and the category with the highest occurrence is selected, ensuring the minority aligns with the majority to reduce generalization error [39]. As a result, random forests have a significant advantage in addressing nonlinear problems [40].

2.3.2. Sen’s Slope Estimation

The Theil–Sen Median method, also referred to as Sen’s slope estimation, is a robust and reliable approach in nonparametric statistical analysis [41]. The method is computationally efficient and well-suited for trend analysis of long time series data. It has been widely applied in trend detection analyses of hydrometeorological variables, including flow, temperature, and precipitation [42,43,44,45,46], and is calculated as follows:
β = M e d i a n x j − x i j − i ∀ j > i
where median() represents the calculation of the median value. If β is greater than 0, it signifies a growing trend; otherwise, a declining trend is indicated.

2.3.3. Mann–Kendall Trend Test

The Mann–Kendall (MK) test is a nonparametric method for detecting trends in time series data. It does not require the data to follow a specific distribution, is robust against missing values and outliers, and is well suited for trend significance testing in long time series datasets [47,48].
The procedure is as follows. For the sequence Xt = x1, x2, x3, …, xn, first determine the magnitude relationship between xi and xj (set to S) among all pairs of values (xi, xj, j > i). Make the following assumptions, H0, that the data in the series are randomly ordered, i.e., there is no significant trend, and H1, that there is an upward or downward trend in the series. The test statistic S is calculated as follows:
S = ∑ i = 1 n − 1     ∑ j = i + 1 n x j − x i
where Sgn() is the sign function and is calculated as
Sgn ⁡ x j − x i = + 1   x j − x i   >   0 0   x j − x i = 0 − 1   x j − x i   <   0
A trend test is performed using the test statistic Z. The Z value is calculated as follows:
Z = S − 1 V a r ( S )   S > 0 0         S = 0 S + 1 V a r ( S )         S < 0
where Var(S) is calculated by the formula
V a r ( S ) = n n − 1 2 n + 5 − ∑ i = 1 n t i i − 1 2 i + 5 18
where n is the number of data in the sequence and ti is the number of repetitions in the ith repetitive dataset.
A bilateral trend test is used, and the critical value Z1−α/2 is checked in the normal distribution table at the given significance level. When |Z| ≤ Z1−α/2, the original hypothesis is accepted, i.e., the trend is not significant; if |Z| > Z1−α/2, the original hypothesis is rejected, i.e., the trend is considered to be significant. In this paper, given the significance level α = 0.05, the critical value Z1−α/2 = ±1.96, when the absolute value of Z is greater than 1.65, 1.96, and 2.58, means that the trend passes the significance test with a confidence level of 90%, 95%, and 99%, respectively. The method of determining the significance of the trend is shown in the table (Table 1).

3. Analysis of the Results

3.1. Spatial Characteristics of the Disaster-Pregnant Environment

3.1.1. Spatial Characteristics of the Static Disaster-Pregnant Environment

Rocks exhibit varying degrees of hardness, which directly influence their resistance to shear forces and weathering processes. From the perspective of geologic disaster mechanisms, the hardness or softness of rock significantly influences its shear resistance and weathering capacity. Harder rocks exhibit greater resistance to shear forces and weathering, making them less susceptible to geologic disasters under identical external conditions [49]. In the southeastern region of the YHV and scattered central areas, the distribution of soft rock and loose rock groups indicates weaker shear and weathering resistance. The capacity of the region to withstand geologic disasters is influenced by rock fragmentation and slope instability, leading to geologic disasters of varying scales. The lithology of the YHV is primarily composed of hard rock groups and medium-hard soft rock groups. However, soft rock and loose rock groups are mainly distributed in the southeastern YHV and scattered central areas. The distribution ratios of hard rock groups, medium-soft hard rock groups, medium-hard soft rock groups, soft rock groups, and loose rock groups are 26%, 18%, 38%, 3%, and 15%, respectively (Figure 4a).
In most parts of the YHV, the fault density exceeds 1.5 km, reaching up to 2.56 km. Only in the northeastern area of the YHV does the fault density exceed 2.56 km, while in small portions of the southeastern and southwestern regions, it is less than 1.5 km. Rupture zones exert significant control over the regional distribution and development of geohazards [50]. The fractured structure of the fault zone can extend for several kilometers to tens of kilometers. Along the fault zone, weak tectonic surfaces develop, leading to rock fragmentation and the formation of dynamic metamorphic rocks such as cataclasite, fractured rock, and breccia. This process accelerates weathering, resulting in the formation of deep weathering zones. The fractured rock structure provides a material basis for geohazard incubation [51]. Therefore, this study uses fault density as a metric to assess the influence on geologic disaster formation in the study area. Higher fault density correlates with a greater likelihood of geologic disaster occurrence (Figure 4b).
High-slope areas (>20°) are primarily concentrated in the river valley region of the YHV (Figure 4d). Slope stability is influenced by the steepness of the slope, which in turn affects the stability of loose solid materials. Steeper slopes result in increased gravitational and shear forces, making them more susceptible to geologic disasters [52]. In the valley regions of the study area, loose accumulations and falling broken rocks experience significantly higher gravitational effects compared to mountainous areas, resulting in a higher frequency of geohazards in the valley regions.
Topography and geomorphology are fundamental to the formation of geologic disasters, serving as key factors that significantly influence and control their occurrence [53]. The study area has an elevation range of 1605–5232 m, with most of the region’s elevation falling between 3000 and 4000 m, accounting for 48% of the study area. The YHV’s topography is characterized by lower elevations in the center and higher elevations at both ends, with areas exceeding 5000 m primarily located on the northern and southern edges, accounting for 3% of the study area (Figure 4c). From the perspective of geologic disaster mechanisms, certain topographic and geomorphological conditions are necessary for geologic disasters to occur. The YHV features significant terrain undulation, large elevation differences, intense gully erosion, dense gully development, and deep incisions, forming a canyon-like landscape that provides an ideal disaster-pregnant environment for geologic disasters.
Most of the rivers in the study area have a density ranging from 0 to 5 km, covering 57% of the study area. The maximum river density reaches 32 km, accounting for 3% of the total area (Figure 4e). From the perspective of geologic disaster formation, river erosion is a primary driving force contributing to the destabilization of mountain slopes in river valley regions, resulting in geohazards [54]. Therefore, this study uses proximity to river density as an indicator to reflect the degree of river erosion’s influence on the study area. Higher river density indicates stronger river erosion. The YHV is characterized by significant river drops and typical river valley geomorphology, with steep and unstable slopes on both sides of the river, making it prone to geologic disasters.
The spatial distribution of vegetation cover is influenced by various factors, including precipitation, temperature, soil, slope, and elevation, among others [49]. Influenced by various factors, the spatial distribution of vegetation cover in the YHV is characterized by higher values in the south and lower values in the north, with a regional average of 0.61 (Figure 4f). The impact of vegetation on geologic disasters requires comprehensive analysis, considering factors such as precipitation, slope gradient, and slope stability. Vegetation cover influences precipitation infiltration, while the root system affects the structural and mechanical properties of the surface soil. In areas with small slope gradients and stable slope surfaces, dense native vegetation can reinforce the soil, enhancing its shear strength resistance. Roots anchor the soil by embedding it into the bedrock or the lower layers of old loess, while the foliage reduces precipitation descent rates, providing a protective effect. Conversely, on steep slopes with unstable surfaces, the self-weight of vegetation increases the slope’s load and transfers wind-induced loads to the slope body. Additionally, root system extensions may destabilize the mechanical stability of the slope, exacerbating geohazard formation [49,55]. Therefore, areas in the central YHV with low vegetation cover and unstable slopes at gradients of 30–40° significantly contribute to geohazard formation.

3.1.2. Spatial Characterization and Importance of the Dynamic Disaster-Pregnant Environment

Climate change, characterized by rising temperatures and increased rainfall, has introduced new characteristics and trends to the disaster-pregnant environment. Simultaneously, the expansion of human activities has further exacerbated the instability of these environments. Consequently, meteorological factors and human activities were chosen to analyze the spatial distribution patterns and temporal trends of the dynamic disaster-pregnant environment in the YHV. Furthermore, this study evaluates the relative influence of each factor on the occurrence of geologic disasters.
Rainfall is the primary factor triggering geologic disasters, with extreme rainfall events (R99p) being particularly significant. Influenced by topography, geomorphology, and geotechnical properties, R99p can induce large-scale landslides, collapses, and other geohazards [56]. From 2003 to 2022, the R99p in the YHV ranged from 48 to 68 mm, with an average value of 55 mm. The highest values were recorded in the northeastern and southeastern mountainous areas, while the lowest values were found in most of the western region, showing a gradual decrease from east to west (Figure 5a). Similarly, the average annual rainfall in the YHV from 2003 to 2022 ranged from 353 to 722 mm, with an average of 528 mm. Overall, rainfall was higher in the northern and southern regions and lower in the central area (Figure 5c).
CWDs in the YHV from 2003 to 2022 ranged from 8 to 12 days, with an average value of 9 days (Figure 5b). CWDs are a key indicator of extreme climate events. Prolonged rainfall can rapidly increase the soil moisture content in mountainous areas, softening the soil and reducing its shear strength. Simultaneously, rainwater infiltrates downward through rock fractures and soil pores, forming underground water flow that further destabilizes the mountain. When the mountain body is subjected to forces exceeding its shear strength, geologic disasters, such as landslides and mudslides, occur.
Global warming increases the risk of geologic disasters, with temperature changes serving as a direct factor influencing their occurrence. The warming trend results in more frequent flood outbreaks in the Yellow River Basin [57]. In the YHV, the monthly maximum temperature difference ranges from 21 to 30 °C (Figure 5e), with an average of 25 °C, gradually decreasing from the mountainous areas to the valley regions. Similarly, the monthly maximum temperature ranges from 6 to 30 °C (Figure 5d), with an average of 20 °C, gradually decreasing from the valley areas to the northern and southern mountainous regions, showing an inverse trend to the temperature difference. From 2003 to 2022, the annual average air temperature in the YHV ranged from 9.35 °C to 12.84 °C, with an average of 1.15 °C (Figure 5f), and it was higher in the central areas and lower on both sides. Changes in precipitation and air temperature are interdependent and influence each other, leading to changes in the stability of the slope surface layer and geotechnical structure. When external forces cause the slope to reach a critical state, various types of landslides and other geohazards may occur.
Human activities are a key contributing factor to geologic disasters, with the intensity of these activities influencing the likelihood of geologic disaster occurrences to varying degrees. High-intensity human activities cause greater damage to the land surface, increasing susceptibility to geologic disasters. Conversely, areas with lower human activity experience less impact, leaving the land surface relatively stable. In the YHV, regions most affected by human activities are primarily located in the river valley areas (Figure 5g).
Using random forest modeling, the YHV geohazard dynamic disaster-pregnant environment stability model was developed. Each indicator factor was weighted as the dependent variable, while individual indicators were treated as independent variables. The final dataset was input into the random forest model to determine the relative importance of each influencing factor. As shown in Figure 6, the primary factors influencing geologic disasters include R99p, CWDs, human activities, maximum monthly temperature difference, average annual temperature, average annual precipitation, and maximum monthly temperature. Extreme meteorological events play a significant role in destabilizing the disaster-pregnant environment, making it more susceptible to geologic disasters.

3.1.3. Inter-Annual Variation in the Dynamic Disaster-Pregnant Environment

From 2003 to 2022, the mean annual temperature exhibited a distinct upward trend (Figure 7a), with an extremely significant increase dominating 43% of the study area (Figure 8a). Regions showing no significant increase, significant increase, and slightly significant increase were predominantly located in the river valley regions and sporadically scattered throughout the area, accounting for 22%, 16%, and 7% of the study area, respectively. Conversely, areas with no significant decrease and no changes were primarily confined to the mountainous regions on the northern and southern sides, comprising 11% and 1% of the study area, respectively. The observed temperature increase in the valleys, contrasted with the stable or declining temperatures in mountainous regions, can be attributed to the differential effects of global warming. Specifically, low-lying terrains exhibit a greater propensity for heat accumulation, amplifying the temperature rise in these regions. In comparison to valley regions, mountainous areas are characterized by higher altitudes, a thinner atmospheric layer, and reduced insulation capacity. Moreover, mountainous regions experience dynamic air currents that induce cooling through adiabatic processes, often resulting in precipitation, such as rainfall or snowfall, which further suppresses temperatures in these areas. Finally, anthropogenic activities represent a critical factor influencing temperature variations between valley and mountainous regions. The YHV, serving as a pivotal hub for agricultural production and the economic and cultural development of Qinghai Province, experiences intensified temperature increases driven by human activities and the heat island effect. In contrast, the sparse population density in mountainous regions minimizes the impact of anthropogenic activities on temperature, thereby leading to stable or even declining temperature trends.
The average annual rainfall from 2003 to 2022 demonstrated an overall declining trend (Figure 7e), consistent with previous studies indicating that the average annual precipitation in the YHV during the 1990s was 17.2 mm less than that of the 1960s [58] (Figure 8b). Spatially, the trend was dominated by a non-significant decrease, primarily occurring in the mountainous regions, which accounted for 60% of the study area. In contrast, the remaining 39% of the study area, located in the valley regions, exhibited a non-significant increase. This phenomenon is closely linked to the region’s distinct topography and climatic conditions. The topography and climatic conditions of the region play a critical role in shaping precipitation patterns. The YHV is characterized by its “trumpet-shaped” topography, with higher elevations in the northwest and lower elevations in the southeast, which facilitate the uplift of airflow within the valley and the subsequent formation of orographic (terrain-induced) rainfall. When humid airflow encounters mountainous terrain, it ascends and cools, leading to precipitation on the windward slope. Conversely, the leeward slope remains relatively dry, resulting in reduced precipitation in mountainous areas [59]. Areas showing no change, marginally significant increase, or highly significant increase were sporadically scattered throughout the study region, collectively accounting for less than 1% of the total area.
The monthly maximum temperature difference exhibited an overall decreasing trend from 2003 to 2022 (Figure 7b), characterized by a spatially insignificant decline predominantly concentrated in the southwestern mountainous region, which accounts for 48% of the study area (Figure 8c). In the northeastern region, 47% of the area experienced no changes, maintaining a stable temperature difference. Marginal decreases and increases were primarily located in the southeastern and northeastern parts of the study area, each accounting for 2% of the total area. The remaining 1% of the study area, located in the north-central region, exhibited a minor, non-significant upward trend in the monthly maximum temperature difference. The decline in the maximum monthly temperature difference can be attributed to several factors. First, climate change trends have led to an increase in seasonal average temperatures in the YHV, with temperatures in the 1990s being 0.4 to 0.8 °C higher than those in the 1960s. This warming is particularly pronounced in winter, significantly contributing to the reduction in temperature difference [58]. Second, the increase in vegetation cover has contributed to lowering surface temperatures, thereby narrowing the temperature difference [58]. Finally, human activities, including urbanization and industrialization, have further influenced the decline in the temperature difference.
From 2003 to 2022, the monthly maximum temperature exhibited an overall increasing trend (Figure 7c), with a spatially insignificant rise observed across 26% of the study area (Figure 8d), primarily concentrated in the northern and southern mountainous regions. Insignificant decreases were scattered throughout the study area, accounting for 25% of the total area, while more substantial concentrations of slightly significant decreases were identified in the southern and northeastern regions, covering 18% of the study area. Areas with no change in the monthly maximum temperature were primarily distributed along the river valleys and sporadically scattered across the northern and southern sides, accounting for 16% of the study area. Slightly significant increases were predominantly located in the northern part of the study area, with minor distributions in the southern regions, collectively covering 14% of the study area. Two key factors contribute to the overall increase in the monthly maximum temperature in the region. First, the overarching trend of global warming has driven the rise in monthly maximum temperatures [58]. Second, urbanization and industrialization have further amplified this effect. Spatially, variations in elevation have caused localized decreases in the monthly maximum temperature in certain areas, resulting in heterogeneous patterns of temperature change.
R99p showed an overall increasing trend from 2003 to 2022 (Figure 7f), and through the mutation test of R99p from 2003 to 2022, the UF curve of R99p was between 0 and 1.96 after 2006, indicating that the trend and mutation of the change curve were not obvious and the series showed an upward trend. At the confidence level of 0.05, there were three mutations of R99p in the YHV, which occurred around 2007, 2011, and 2015, respectively (Figure 8e). The overall increase in extreme rainfall in the region can be attributed to two primary factors. First, there has been a marked rise in the frequency and intensity of extreme weather events, particularly an increase in summer extreme precipitation over the Tibetan Plateau, a region highly sensitive to global climate change. Second, large-scale atmospheric circulation patterns play a crucial role, with anomalous convergence of water vapor over the plateau and intensified water vapor transport during years of high-intensity precipitation significantly influencing the occurrence of extreme rainfall events [60].
The overall trend of CWDs from 2003 to 2022 shows a decreasing trend (Figure 7g); through the mutation test of CWDs from 2003 to 2022, the UF value of CWDs is between 0 and 1.96 before 2012, which indicates that the trend and mutation of the change curve are not obvious and the series is decreasing. After 2012, the UF curve is over the critical value, indicating a very clear downward trend for CWDs. At a confidence level of 0.05, there were two sudden changes in the number of CWDs in the YHV, which occurred around 2005 and 2007, respectively (Figure 8f). The declining trend in consecutive rainy days within the district can be attributed to two key factors. First, rising annual average temperatures, coupled with a reduction in annual average precipitation and increased surface evaporation, have exacerbated drought conditions, thereby contributing to a decline in the number of consecutive rainy days [58]. Second, the prevalence of extreme precipitation events, characterized by a short duration but intense rainfall, further impacts the reduction in consecutive rainy days.
From 2000 to 2020, human activities exhibited an overall upward trend (Figure 7d), with spatial patterns predominantly characterized by a non-significant rise (Figure 8g), covering 49% of the study area and distributed across various parts of the region. Similarly, a non-significant decline was observed across 27% of the area, which was also widely distributed. Both non-significant rises and declines were dispersed throughout the study area. Minimally significant rises were concentrated in the central region, encompassing 13% of the study area, while minimally significant declines were sporadically distributed, accounting for only 1%. Areas showing no change were primarily located in the mountainous regions to the north and south, comprising 10% of the study area. Both insignificant increases and decreases were widely distributed across the study area. Insignificant increases were primarily concentrated in the central region, accounting for 13% of the total area, while insignificant decreases were sporadically scattered, covering only 1%. Areas with no observable changes were primarily located in the northern and southern mountainous regions, representing 10% of the study area. The notable spatial variations in human activities across the region can be attributed to the unique characteristics of the YHV, which serves as a transitional zone between agricultural and pastoral activities. This transitional nature leads to heightened mobility in human activities.

3.2. Spatial Characterization of the Stability Disaster-Pregnant Environment

3.2.1. Spatial Characterization of the Static Disaster-Pregnant Environment Stability

Using the natural breakpoint method, the stability levels of the YHV geologic disaster static disaster-pregnant environment were classified into four grades: high, medium-high, medium-low, and low. As shown in Figure 9, the static disaster-pregnant environment in the region is predominantly medium-high stability, covering an area of 11,277.10 km2, which accounts for 32% of the study area. The high stability area covers 8591.34 km2, or 24% of the study area. The medium-low stability area spans 9446.52 km2, making up 27% of the study area, while the low stability area covers 6011.10 km2, representing 17% of the study area.

3.2.2. Spatial Characterization of the Dynamic Disaster-Pregnant Environment Stability

Using the natural breakpoint method, the YHV geologic disaster dynamic disaster-pregnant environment stability levels were classified into four grades: high, medium-high, medium-low, and low. As shown in Figure 10, the dynamic disaster-pregnant environment in the region is predominantly of medium-low stability, covering an area of 12,936.11 km2, which accounts for 37% of the study area. The high stability area covers 9018.08 km2, or 26% of the study area. The medium-high stability area spans 10,530.23 km2, representing 30% of the study area, while the low stability area covers 2841.63 km2, accounting for 8% of the study area.

3.2.3. Spatial Characterization and Trend Features in the Comprehensive Disaster-Pregnant Environment Stability

Using the natural breakpoint method, the comprehensive stability of the YHV disaster-pregnant environment was classified into four grades: high, medium-high, medium-low, and low. As shown in Figure 11, the comprehensive stability of the disaster-pregnant environment is predominantly medium-high, covering 12,546.97 km2, which accounts for 36% of the study area. The high-stability area spans 7140.84 km2, representing 20% of the study area. The medium-low stability area encompasses 11,057.66 km2, accounting for 31% of the study area, while the low-stability area covers 4487.88 km2, comprising 13% of the study area.
As shown in Figure 12, the changing trend of the disaster-pregnant environment in the region is predominantly unstable, accounting for 71.8% of the study area and evenly distributed throughout the region. This instability is attributed to the complex natural profile of the YHV, whose unique geographic location places it in the transitional zones between the Qinghai–Tibetan Plateau and the Loess Plateau, as well as between the eastern monsoon and the northwestern arid zones. Additionally, the region’s complex geological conditions further contribute to the instability of the geohazard-prone environment. In the river valley area, 16.6% of the region is classified as extremely unstable, which is due to the central valley’s loose rock group distribution, high fracture density, river density, steep slopes, and elevations exceeding 3000 m, all of which are sensitive to climate change. Coupled with the presence of gullies and intense human activity, these factors exacerbate instability. In contrast, 9.4% of the northern region remained stable from 2003 to 2022. Only 0.1% of the area is classified as stable, and 2.1% is extremely stable, mainly due to minimal human activity and the positive influence of vegetation cover on slope stability.

4. Discussion and Conclusions

4.1. Discussion

4.1.1. Classification of Stability and Instability in Disaster-Inducing Environments

Disaster-inducing factors originate from distinct environmental systems. This study categorizes the stability and instability of disaster-pregnant environments into two cases. The stability of these environments is defined as follows. (1) A regional disaster-pregnant environment is considered stable if no disaster-causing factors are present, and no disasters occur. (2) When natural disaster-causing factors result in a single disaster, the primary source originates from natural environmental conditions (disaster-pregnant environment). For instance, the persistent drought in Northwest China reflects the stability of the disaster-pregnant environment. The instability of disaster-pregnant environments is characterized as follows. (1) In regions lacking pre-existing disaster-causing factors, environmental changes can trigger sudden disasters, indicating instability in the disaster-pregnant environment. (2) Environmental changes associated with a single natural disaster-causing factor may give rise to new types of disasters or trigger disaster chains and clusters, signifying instability in the disaster-pregnant environment. For example, the “7.20” heavy rainfall event in Zhengzhou, Henan Province, resulted in severe urban flooding, flash floods, landslides, and other concurrent disasters.

4.1.2. Selection of Indicators for Evaluating the Stability of Geologic Disaster-Prone Environments

Against the backdrop of global climate change, the rising frequency and intensity of extreme weather and climate events have garnered significant attention for their impact on the stability of disaster-pregnant environments. This study adopts relatively traditional evaluation indicators for assessing the stability of geologic disaster-pregnant environments, focusing primarily on extreme precipitation and other meteorological extremes as key factors contributing to instability. Although earthquakes are critical triggers of landslides and other geologic disasters, particularly in mountainous regions, their contribution as a disaster-causing factor in the study area is relatively minor compared to the frequency of other geologic disasters. Furthermore, earthquakes can trigger complex disaster chains, including co-seismic and post-earthquake events. This complexity makes it challenging to incorporate earthquakes into a straightforward evaluation of the stability of disaster-pregnant environments. The interaction between earthquakes and extreme weather events creates geohazard chains that pose significant challenges. Further research is needed to clarify the triggering mechanisms of geohazards and to refine the evaluation of the stability of disaster-pregnant environments. Future studies will integrate triggering factors, such as earthquakes and extreme rainfall, to develop a more comprehensive framework for evaluating the stability of disaster-pregnant environments, ultimately providing a robust theoretical basis for local disaster prevention and mitigation efforts. Future research aims to conduct a comprehensive evaluation of the stability of geologic disaster-pregnant environments by integrating triggering factors, such as earthquakes and extreme rainfall, thereby offering a more detailed theoretical foundation for disaster prevention and mitigation strategies.

4.1.3. Trends in Evaluating the Stability of Geologic Hazard-Prone Environments

In evaluating the environmental stability of geologic disasters, this study considers six major categories of hazards—collapse, subsidence, mudslides, geologic fissures, landslides, and slope failures—as an integrated system. However, the triggering mechanisms of different geologic disasters vary significantly, directly influencing the selection of evaluation indicators and the classification of stability levels in disaster-pregnant environments. With the increasing frequency of extreme meteorological events, geologic disasters have evolved from isolated occurrences into complex disaster chains. This increasing complexity poses significant challenges to evaluating the stability of disaster-pregnant environments. To address these challenges, future research will focus on identifying the thresholds of instability in disaster-pregnant environments, using key instability factors as a foundation. This approach will be integrated with local early warning systems to provide a more accurate theoretical basis for disaster prevention and mitigation strategies.

4.2. Conclusions

(1) To identify the key drivers of instability in disaster-pregnant environments, this study highlights meteorological extremes as the primary contributors to destabilization, with extremely heavy rainfall contributing 28% and consecutive rainy days contributing 27%. Human activities follow as secondary contributors with a 15% contribution, while conventional meteorological indicators play a lesser role, as determined by weight calculations using the random forest model.
(2) This study defines areas with grades below medium-low (including medium-low) as unstable. Static unstable environments are primarily concentrated in river valleys and north–south mountainous areas, accounting for 44% of the study area. Similarly, dynamic unstable environments are predominantly located along rivers, comprising 45% of the area. Integrated unstable environments share similar spatial distributions with static unstable environments, both of which are concentrated in river valleys and north–south mountainous regions, also accounting for 44% of the study area.
(3) The stability trends of the integrated disaster-pregnant environment indicate widespread instability, with extremely unstable areas predominantly located in river valleys due to the influence of human activities. Conversely, only 0.1% of the area shows a tendency toward stability, while 2.1% tends to be extremely stable.

Author Contributions

T.Z. conducted the research, analyzed the data, and wrote the paper; W.M., Y.G., H.L. and Q.Z. (Qiuyang Zhang) made suggestions for this paper; Q.Z. (Qiang Zhou) conceived the research and provided support for the project. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China (NSFC): Dynamic Changes and Disaster Response of the Frost Weathering Zone in the Tibetan Plateau (42271127), and the Research Project on the Second Comprehensive Scientific Expedition to the Tibetan Plateau: Integrated Disaster Risk Evaluation and Defense (2019QZKK0906).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available upon request from the corresponding author. The data are not publicly available due to privacy.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Burton, I.; Kates, R.W.; White, G.F. The Environment as Hazard, 2nd ed.; The Builford Press: New York, NY, USA, 1993. [Google Scholar]
  2. Blaikie, P.M.; Cannon, T.; Davis, I.; Wisner, B. At Risk: Natural Hazards, People’s Vulnerability, and Disasters; Routledge: London, UK, 1994. [Google Scholar]
  3. EL-Sabh, M.I.; Murty, T.S.; Venkalesh, S.; Siccardi, F.; Andah, K. Recent Studies in Geophysical Hazards; Kluwer Academic Publisher: Dordrecht, The Netherlands; Boston, MA, USA; London, UK, 1994. [Google Scholar]
  4. Bolt, B.A.; Horn, W.L.; Macedonald, G.A.; Scott, R.F. Geological Hazards Revised, 2nd ed.; Springer: New York, NY, USA, 1977. [Google Scholar]
  5. Tucker, B.E.; Erdik, M.; Hwang, C.N. Issues in Urban Earthquake Risk; Kluwer Academic Publishers: Dordrecht, The Netherlands; Boston, MA, USA; London, UK, 1994. [Google Scholar]
  6. Busoni, E.; Sanchis, P.S.; Calzolari, C.; Romagnoli, A. Mass Movement and Erosion Hazard Patterns by Multivariate Analysis of Landscape Integrated Data: The Upper Orcia River Valley (Siena, Italy) case. Catena 1995, 25, 169–185. [Google Scholar] [CrossRef]
  7. Varnes, D.J. Landslide Hazard Zonation: A review of Principles and Practice; UNESCO: Paris, France, 1984. [Google Scholar]
  8. EL-Sabh, M.I.; Murty, T.S. (Eds.) Natural and Man-Made Hazards; D. Reidel Publishing Company: Dordrecht, The Netherlands; Holland, The Netherlands, 1988. [Google Scholar]
  9. Mccall, G.J.H.; Laming, D.J.C.; Scott, S.C. Geoharzards—Natural and Man-Made; Chapman Hall: London, UK; New York, NY, USA; Tokyo, Japan; Milbourne, UK; Madras, India, 1992. [Google Scholar]
  10. Rossi, G.; Harmancioglu, N.; Yevjevich, V. Coping with Floods; Kluwer Academic Publishers: Dordrecht, The Netherlands; Boston, MA, USA; London, UK, 1994. [Google Scholar]
  11. Chung, R.M. (Ed.) Natural Disaster Studies an Investigative Series of the Committee on Natural Disasters Volume Six, Hurricane Hugo; National Academy Press: Washington, DC, USA, 1994. [Google Scholar]
  12. Turner, B.L., II; Meyer, W.B. Land Use and Land Cover in Global Environmental Change: Consideration for Study. Int. Soc. Sci. J. 1991, 43, 668–679. [Google Scholar]
  13. Parsons, M.L. Global Warming—The Truth Behind the Myth; Plenum Press: New York, NY, USA; London, UK, 1995. [Google Scholar]
  14. Zwerrer, S.; Rompaey, R.S.A.R.; Kok, M.J.; Berk, M.M. (Eds.) Climate Change Research—Evaluation and Policy Implications; Elsevier: Amsterdam, The Netherlands, 1995. [Google Scholar]
  15. Kasperson, J.X.; Kasperson, R.E.; Turner, B.L., II. Regions at Risk: Comparisons of Threatened Environments; United Nations University Press: Tokyo, Japan; New York, NY, USA; Paris, France, 1995. [Google Scholar]
  16. Eddy, J.A.; Baker, D.J.; Broecker, W.S.; Drake, C.; Eagleson, P.; Fyfe, W.S.; Gates, W.L.; Imbrie, J.; Loomis, R.; McCarthy, J.J.; et al. Global Change in the Greosphere Biosphere; National Academy Press: Washington, DC, USA, 1986. [Google Scholar]
  17. Committee on Environmental and Natural Resources Research of the National Science and Technology Council. Our Changing Planet—The Fiscal Year 1995; US Global Chanqu Research Program: Washington, DC, USA, 1995. [Google Scholar]
  18. Philander, S.G. EL Nino, La Nina, and the Southern Oscillation; Academic Press, Inc.: San Diego, CA, USA; New York, NY, USA, 1990. [Google Scholar]
  19. Park, C. Environment Issues. Prog. Phys. Geogr. 1995, 19, 379–390. [Google Scholar] [CrossRef]
  20. Sun, Z.; Liu, Y.; Qi, Q.; Liu, W.; Li, D.; Chai, J. Risk Assessment of Coal Mine Flood Disasters Based on Projection Pursuit Clustering Model. Sustainability 2022, 14, 11131. [Google Scholar] [CrossRef]
  21. Qin, X.; Wu, Y.; Lin, T.; Gao, L. Urban Flood Dynamic Risk Assessment Based on Typhoon Rainfall Process: A Case Study of Typhoon “Lupit” (2109) in Fuzhou, China. Remote Sens. 2023, 15, 3116. [Google Scholar] [CrossRef]
  22. Liu, L.; Xiao, H.; Wang, C.; Yao, T. Sensitivity analysis of disaster-causing factors for landslide geohazards in the red layer area of Hunan province. Min. Metall. Eng. 2024, 44, 169–174. [Google Scholar]
  23. Chu, C.; Wang, W.; Shi, Y.; Zhang, H.; Huang, L. Vulnerability assessment of composite hazards in Guangdong-Hong Kong-Macao Greater Bay Area. J. Three Gorges Univ. (Nat. Sci. Ed.) 2024, 46, 60–67. [Google Scholar]
  24. Li, Z.; Tang, X.; Li, L.; Chu, Y.; Wang, X.; Yang, D. GIS-based risk assessment of flood disaster in the Lijiang River Basin. Sci. Rep. 2023, 13, 6160. [Google Scholar]
  25. Wang, X.; Qu, Z.; Tian, F.; Wang, Y.; Yuan, X.; Xu, K. Ice-jam flood hazard risk assessment under simulated levee breaches using the random forest algorithm. Nat. Hazards 2023, 115, 331–355. [Google Scholar] [CrossRef]
  26. Ma, J.; Lei, D.; Ren, Z.; Tan, C.; Xia, D.; Guo, H. Automated Machine Learning-Based Landslide Susceptibility Mapping for the Three Gorges Reservoir Area, China. Math. Geosci. 2024, 56, 975–1010. [Google Scholar] [CrossRef]
  27. Tang, X.; Liu, S.; Zhan, D.; Wang, G.; Luo, Y.; Hu, S.; Xu, Q. Geothermal Accumulation Constrained by the Tectonic Transformation in the Gonghe Basin, Northeastern Tibetan Plateau. Lithosphere 2022, 2021, 3936881. [Google Scholar] [CrossRef]
  28. Tan, S.; Tian, X.; Zeng, X.; Nie, F.; Qu, C.; Yu, C. Crustal structure beneath the northern part of the southeastern Tibetan Plateau revealed by a seismic dense nodal array. J. Asian Earth Sci. 2023, 258, 105593. [Google Scholar] [CrossRef]
  29. Burchfiel, B.; Zhang, P.; Wang, Y.; Zhang, W.; Song, F.; Deng, Q.; Molnar, P.; Royden, L. Geology of the Haiyuan fault zone, Ningxia-Hui Autonomous Region, China, and its relation to the evolution of the northeastern margin of the Tibetan Plateau. Tectonics 1991, 10, 1091–1110. [Google Scholar] [CrossRef]
  30. Meyer, B.; Tapponnier, P.; Bourjot, L.; Metivier, F.; Gaudemer, Y.; Peltzer, G.; Shunmin, G.; Zhitai, C. Crustal thickening in Gansu-Qinghai, lithospheric mantle subduction, and oblique, strike-slip controlled growth of the Tibet plateau. Geophys. J. Int. 1998, 135, 1–47. [Google Scholar] [CrossRef]
  31. Tapponnier, P.; Zhiqin, X.; Roger, F.; Meyer, B.; Arnaud, N.; Wittlinger, G.; Jingsui, Y. Oblique stepwise rise and growth of the Tibet Plateau. Science 2001, 294, 1671–1677. [Google Scholar] [CrossRef]
  32. Zheng, W.; Zhang, P.; Ge, W.; Molnar, P.; Zhang, H.; Yuan, D.; Liu, J. Late Quaternary Slip Rate of the South Heli Shan Fault (Northern Hexi Corridor, NW China) and Its Implications for Northeastward Growth of the Tibetan Plateau. Tectonics 2013, 32, 271–293. [Google Scholar] [CrossRef]
  33. Dong, Z.; Yao, Z.; Lei, F. Active faults and modern tectonic stress field in eastern Qinghai. Crustal Deform. Earthq. 1992, 12, 64–71. [Google Scholar]
  34. Li, H.; Yang, Y.; Zhang, F.; Guo, X.; Li, Y.; Lin, L.; Li, Y.; Cao, G.; Zhou, H. Rainfall Determines Shallow Soil Seepage in a Piedmont Summer Pasture of Alpine Meadow on the Northeastern Qinghai–Tibetan Plateau. Water 2021, 13, 2815. [Google Scholar] [CrossRef]
  35. Zhao, D.; Lancuo, Z.; Hou, G.; Xu, C.; Li, W. Evaluation of geologic hazard susceptibility in the Yellow River-Huangshui River Valley of Qinghai Province. J. Geomech. 2021, 27, 83–95. [Google Scholar]
  36. Chen, T.; Zhu, L.; Niu, R.-Q.; Trinder, C.J.; Peng, L.; Lei, T. Mapping landslide susceptibility at the Three Gorges Reservoir, China, using gradient boosting decision tree, random forest and information value models. J. Mt. Sci. 2020, 17, 670–685. [Google Scholar] [CrossRef]
  37. Li, H.; Shi, Y.; Shen, A.; Ma, Y.; Liang, Y.; Li, X.; Dong, J.; Zhao, N.; Yu, J.; Wang, T.; et al. A study on microtopographic classification of the floodplain fan in the east foothills of Helan Mountains based on random forest algorithm. Quat. Res. 2024, 1–13. [Google Scholar]
  38. Asgari, M.; Yang, W.; Farnaghi, M. Spatiotemporal data partitioning for distributed random forest algorithm: Air quality prediction using imbalanced big spatiotemporal data on spark distributed framework. Environ. Technol. Innov. 2022, 27, 102776. [Google Scholar] [CrossRef]
  39. Qi, L.; Wang, X.; Wang, C.; Wang, H.; Li, X. Random forest-based screening of environmental geohazard probability factors in Panshi city, China. Adv. Space Res. 2025, 75, 414–431. [Google Scholar] [CrossRef]
  40. Wang, C.; Wang, K.; Liu, D.; Zhang, L.; Li, M.; Khan, M.; Li, T.; Cui, S. Development and application of a comprehensive assessment method of regional flood disaster risk based on a refined random forest model using beluga whale optimization. J. Hydrol. 2024, 633, 130963. [Google Scholar] [CrossRef]
  41. Alimujiang, K.; Zhang, X.; Liang, H. Analysis of the spatial relationship between seasonal surface temperature and landscape pattern in the urban agglomeration on the north slope of Tianshan Mountain. Geogr. Res. 2024, 43, 1267–1287. [Google Scholar]
  42. Alkama, R.; Marchand, L.; Ribes, A.; Decharme, B. Detection of global runoff changes: Results from observations and CMIP5 experiments. Hydrol. Earth Syst. Sci. 2013, 17, 2967–2979. [Google Scholar] [CrossRef]
  43. Bayazit, M. Nonstationarity of hydrological records and recent trends in trend analysis: A state-of-the-art review. Environ. Process. 2015, 2, 527–542. [Google Scholar] [CrossRef]
  44. Lettenmaier, D.P.; Wood, E.F.; Wallis, J.R. Hydro-climatological trends in the continental United States, 1948–1988. J. Clim. 1994, 7, 586–607. [Google Scholar] [CrossRef]
  45. Madsen, H.; Lawrence, D.; Lang, M.; Martinkova, M.; Kjeldsen, T. Review of trend analysis and climate change projections of extreme precipitation and floods in Europe. J. Hydrol. 2014, 519, 3634–3650. [Google Scholar] [CrossRef]
  46. Vogel, R.M.; Yaindl, C.; Walter, M. Nonstationarity: Flood magnification and recurrence reduction factors in the United States 1. J. Am. Water Resour. Assoc. 2011, 47, 464–474. [Google Scholar] [CrossRef]
  47. Kendall, M.G. Rank Correlation Methods, 4th ed.; Charles Grifin: London, UK, 1975. [Google Scholar]
  48. Mann, H.B. Non-parametric tests against trend. Econometrica 1945, 13, 245. [Google Scholar] [CrossRef]
  49. Shen, Z.; Zhang, Q.; Wu, W.; Song, C. Spatial pattern and driving factors of geohazard-prone areas in the Tibetan Plateau and Hengduan Mountains. J. Geogr. 2022, 77, 1211–1224. [Google Scholar]
  50. Qin, Y.; Zhang, D.; Zheng, W.; Liu, B.; Gong, Z. Spatial distribution of near-fault landslides along Litang fault zones, eastern Tibetan Plateau. Geomorphology 2024, 455, 109189. [Google Scholar] [CrossRef]
  51. Shaller, P.J.; Doroudian, M.; Hart, M.W. The Eureka Valley Landslide: Evidence of a Dual Failure Mechanism for a Long-Runout Landslide. Lithosphere 2020, 2020, 8860819. [Google Scholar] [CrossRef]
  52. Liu, Q.; Su, L.; Zhang, C.; Hu, B.; Xiao, S. Dynamic variations of interception loss-infiltration-runoff in three land-use types and their influence on slope stability: An example from the eastern margin of the Tibetan Plateau. J. Hydrol. 2022, 612 Pt B, 128218. [Google Scholar] [CrossRef]
  53. Zhang, X.; Jiang, L.; Deng, W.; Shu, Z.; Gao, M.; Liu, G. Risk Assessment of Geological Hazards in the Alpine Gorge Region and Its Influencing Factors: A Case Study of Jiulong County, China. Sustainability 2024, 16, 1949. [Google Scholar] [CrossRef]
  54. Yao, X.; Deng, J.; Liu, X.; Zhou, Z.; Yao, J.; Dai, F.; Ren, K.; Li, L. Primary recognition of active landslides and development rule analysis for pan three-river-parallel. Adv. Eng. Sci. 2020, 52, 16–37. [Google Scholar]
  55. Niyomukiza, J.; Eisazadeh, A.; Tangtermsirikul, S. Recent advances in slope stabilization using porous vegetation concrete in landslide-prone regions: A review. J. Build. Eng. 2023, 76, 107129. [Google Scholar] [CrossRef]
  56. Shi, P.; Yang, W. Impacts of earthquakes and extreme weather and climate on geohazards in a mountainous disaster-bearing environment. Prog. Clim. Chang. Res. 2020, 16, 405–414. [Google Scholar]
  57. Gao, Y.; Li, B.; Feng, Z.; Zuo, X. Analysis of global climate change and geohazard response. J. Geomech. 2017, 23, 65–77. [Google Scholar]
  58. Jiao, K.; Gao, J.; Liu, Z. Precipitation Drives the NDVI Distribution on the Tibetan Plateau While High Warming Rates May Intensify Its Ecological Droughts. Remote Sens. 2021, 13, 1305. [Google Scholar] [CrossRef]
  59. Li, Z.; Sun, W.; Yu, R. Analysis of the three-dimensional circulation structure that produced the “22-8” extreme precipitation in Qinghai. J. Meteorol. 2024, 82, 37–54. [Google Scholar]
  60. Yu, C.; Qing, Y.; Zi, C. Relationship between intense summer precipitation and large-scale circulation over the central-eastern Tibetan Plateau, 1961–2019. Glacial Permafr. 2021, 43, 1290–1300. [Google Scholar]
Sustainability 17 00732 g001
Figure 1. Overview of the study area.
Figure 1. Overview of the study area.
Sustainability 17 00732 g001
Sustainability 17 00732 g002
Figure 2. Density map of geohazard sites in the YHV.
Figure 2. Density map of geohazard sites in the YHV.
Sustainability 17 00732 g002
Sustainability 17 00732 g003
Figure 3. Random forest schematic.
Figure 3. Random forest schematic.
Sustainability 17 00732 g003
Sustainability 17 00732 g004
Figure 4. Spatial characterization of the static disaster-pregnant environment in the study area. (a) Lithology, (b) fracture zone density, (c) topographic elevation, (d) slope, (e) river network density, (f) vegetation cover.
Figure 4. Spatial characterization of the static disaster-pregnant environment in the study area. (a) Lithology, (b) fracture zone density, (c) topographic elevation, (d) slope, (e) river network density, (f) vegetation cover.
Sustainability 17 00732 g004
Sustainability 17 00732 g005
Figure 5. Spatial characterization of the dynamic disaster-pregnant environment in the study area, 2003–2022. (a) R99p, (b) CWDs, (c) average annual rainfall, (d) monthly maximum temperature, (e) maximum monthly temperature difference, (f) average annual temperature, (g) human activity.
Figure 5. Spatial characterization of the dynamic disaster-pregnant environment in the study area, 2003–2022. (a) R99p, (b) CWDs, (c) average annual rainfall, (d) monthly maximum temperature, (e) maximum monthly temperature difference, (f) average annual temperature, (g) human activity.
Sustainability 17 00732 g005
Sustainability 17 00732 g006
Figure 6. Importance of the geohazard dynamic disaster-pregnant environment based on random forest.
Figure 6. Importance of the geohazard dynamic disaster-pregnant environment based on random forest.
Sustainability 17 00732 g006
Sustainability 17 00732 g007
Figure 7. Inter-annual variation in the dynamic disaster-pregnant environment. (a) Average annual temperature, (b) maximum monthly temperature difference, (c) monthly maximum temperature, (d) human activity, (e) average annual rainfall, (f) R99p, (g) CWDs.
Figure 7. Inter-annual variation in the dynamic disaster-pregnant environment. (a) Average annual temperature, (b) maximum monthly temperature difference, (c) monthly maximum temperature, (d) human activity, (e) average annual rainfall, (f) R99p, (g) CWDs.
Sustainability 17 00732 g007
Sustainability 17 00732 g008
Figure 8. Trend features and significance tests of the dynamic disaster-pregnant environment. (a) Average annual temperature, (b) average annual rainfall, (c) maximum monthly temperature difference, (d) monthly maximum temperature, (e) R99p, (f) CWDs, (g) human activity.
Figure 8. Trend features and significance tests of the dynamic disaster-pregnant environment. (a) Average annual temperature, (b) average annual rainfall, (c) maximum monthly temperature difference, (d) monthly maximum temperature, (e) R99p, (f) CWDs, (g) human activity.
Sustainability 17 00732 g008
Sustainability 17 00732 g009
Figure 9. Distribution of the stability of the static disaster-pregnant environments.
Figure 9. Distribution of the stability of the static disaster-pregnant environments.
Sustainability 17 00732 g009
Sustainability 17 00732 g010
Figure 10. Stability distribution of the dynamic disaster-pregnant environments.
Figure 10. Stability distribution of the dynamic disaster-pregnant environments.
Sustainability 17 00732 g010
Sustainability 17 00732 g011
Figure 11. Distribution of the stability of the comprehensive disaster-pregnant environments.
Figure 11. Distribution of the stability of the comprehensive disaster-pregnant environments.
Sustainability 17 00732 g011
Sustainability 17 00732 g012
Figure 12. Comprehensive disaster-pregnant environment stability trend features.
Figure 12. Comprehensive disaster-pregnant environment stability trend features.
Sustainability 17 00732 g012
Table 1. Trend significance judgment table.
Table 1. Trend significance judgment table.
βZTrend TypeTrend Features
β > 02.58 < Z4Extremely significant increase
1.96 < Z ≤ 2.583Significant increase
1.65 < Z ≤ 1.962Slightly significant increase
Z ≤ 1.651No significant increase
β = 0Z0No changes
β < 0Z ≤ 1.65−1No significant decrease
1.65 < Z ≤ 1.96−2Slightly significant decrease
1.96 < Z ≤ 2.58−3Significant decrease
2.58 < Z−4Extremely significant decrease
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, T.; Zhou, Q.; Ma, W.; Gao, Y.; Li, H.; Zhang, Q. Disaster-Pregnant Environment Stability Evaluation of Geohazards in the Yellow River–Huangshui River Valley, China. Sustainability 2025, 17, 732. https://doi.org/10.3390/su17020732

AMA Style

Zhang T, Zhou Q, Ma W, Gao Y, Li H, Zhang Q. Disaster-Pregnant Environment Stability Evaluation of Geohazards in the Yellow River–Huangshui River Valley, China. Sustainability. 2025; 17(2):732. https://doi.org/10.3390/su17020732

Chicago/Turabian Style

Zhang, Tengyue, Qiang Zhou, Weidong Ma, Yuan Gao, Hanmei Li, and Qiuyang Zhang. 2025. "Disaster-Pregnant Environment Stability Evaluation of Geohazards in the Yellow River–Huangshui River Valley, China" Sustainability 17, no. 2: 732. https://doi.org/10.3390/su17020732

APA Style

Zhang, T., Zhou, Q., Ma, W., Gao, Y., Li, H., & Zhang, Q. (2025). Disaster-Pregnant Environment Stability Evaluation of Geohazards in the Yellow River–Huangshui River Valley, China. Sustainability, 17(2), 732. https://doi.org/10.3390/su17020732

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