Abstract
In recent years, there have been many studies on regional geological hazard risk assessment. However, most previous research focused on natural geological disasters such as landslides and debris flows. There were few special regional risk studies aimed at highway slopes influenced deeply by artificial factors. According to the characteristics of highway slope, based on the evaluation of regional natural influencing factors, the artificial influencing factors and geological engineering information of an individual slope are introduced in this paper. The information evaluation model, combined with the analytic hierarchy process (AHP) and ArcGIS spatial analysis, was used to evaluate the risk of disaster in the study area. Then, the risk assessment results and risk zoning map along the highway were obtained. By comparing and analyzing the data of the management unit, the research results were in good agreement with the actual slope. The results of this paper can effectively guide the maintenance and management of highway slopes and provide a reference for site selection of subsequent highway reconstruction and expansion projects. In general, this study can improve the level of geological risk management in highway engineering.
1. Introduction
Slope engineering is the most typical geological hazard in highway construction and operation, which seriously affects the safety of passing vehicles and pedestrians. Located on the southeast coast, a mountainous, and rainy area, the geological environment of Zhejiang province, China, is very complex. Under the dual action of physical and chemical weathering and rainfall erosion, the geotechnical properties of highway slopes continue to weaken; furthermore, highway protection facilities are damaged to varying degrees, or even completely lose their protection performance [1]. As a result, a series of highway disruptions, such as collapse and landslide, occur, which pose a great threat to safe highway operations.
In recent years, the evaluation and maintenance management of national and provincial highway slopes in Zhejiang province have had many achievements, which have laid a solid foundation for safe operations. However, the scattered risk assessment of single highway slopes cannot fully reflect the current safety of the whole area along and surrounding national and provincial highways. Based on this, it is particularly necessary to carry out a regional risk assessment of national and provincial highways [2]. In the past, the information quantity method was used to evaluate regional slope risk. The object of the evaluation was the natural slope, and the regional natural factors were selected accordingly as evaluation factors. At present, there are few studies on the regional risk assessment of manually excavated highway slopes. Compared with previous research results, this paper focused on the characteristics of highway slopes. The evaluation factors were selected by considering natural and artificial factors and combining the geological engineering conditions and historical conservation data of every slope. The evaluation result was more in line with reality.
This paper is based on a collection of geological engineering information on 850 cutting slopes of five national and provincial highways in Quzhou City, Zhejiang province. Taking the 400 m bandwidth range along both sides of national and provincial highways as the research area, the regional disaster risk of national and provincial highways was studied through the information evaluation model combined with the analytic hierarchy process (AHP) and ArcGIS. The research provides important scientific reference and guiding significance for Quzhou City to carry out slope maintenance and management of national and provincial highways as well as reconstruction and expansion in later periods [3].
2. Brief Introduction of Information Method
Currently, risk assessment on a single slope, no matter domestic or abroad, has formed a relatively mature and complete theory and method. However, the regional slope hazard assessment is still in the stage of incomplete development. There is still a lack of relevant research, especially for the geological hazard assessment of highway belt region. In the previous achievements, many experts and scholars evaluated the regional geological hazard of the whole county or even the whole province. Based on the information quantity model and ArcGIS spatial overlay analysis function, regional geological disaster susceptibility zoning map was obtained. For example, Li et al. [4] analyzed the effect of geological hazard susceptibility evaluation in Urumqi city. Chen et al. [5] evaluated the landslide susceptibility of the distribution area of clastic rocks in Guangxi, and the landslide susceptibility zoning and prevention zoning of the clastic rocks in Guangxi were obtained. Wang et al. [6] evaluated the susceptibility of geological disasters in Chizhou, Anhui Province. Xu et al. [7] carried out the landslide susceptibility regionalization and landslide susceptibility evaluation in the granite distribution area of Guangxi. Sun et al. [8] established a set of regional evaluation methods for high slope stability through field geological investigation and analysis based on fuzzy theoretical model. Yin and Zhu [9] conducted a systematic and in-depth study on the theoretical system of spatial regionalization of landslide disaster and slope instability, and carried out a study on the application of spatial regionalization of disaster risk in the worst-hit area of landslide such as Hanjiang River basin, and achieved relatively mature results. Eldeen [10] from Sweden introduced the research model of flood risk into the study of debris flow risk and studied the risk of landslide and debris flow disaster with the risk area map in 1980. In 1981, Hollingsworth and Kovacs [11] in the United States adopted the scoring method to construct the basic framework for the risk assessment of landslide and debris flow and completed the landslide risk assessment by means of factor stacking method, providing methodological guidance for the risk assessment of most landslides and debris flow nowadays.
The rapid development of computer technology and GIS technology provides new ideas and powerful tools for the risk assessment of landslide and debris flow. In 1991, Carrara et al. [12] from the Department of Earth Sciences, University of Roorkee, India, applied ArcGIS method to regionalization of landslide hazard in Ramganga Catchment area of Himalayan foothills. Through multisource data processing, storage, management, and spatial superposition analysis of ArcGIS, the comprehensive partition map of landslide hazard was drawn. Carrara et al. [13] combined GIS technology with statistical model to evaluate landslide hazard. Up to now, a variety of evaluation models have been formed for regional slope safety evaluation, including artificial neural network, weight of evidence method [14], information method model, multivariate statistical model, expert scoring model, grey system model, logistic regression model, nonlinear model, and fuzzy theory model [15–22]. Due to the advantages of simple calculation, wide applicability, easy modeling, and high reliability, information method model has been widely used in scientific research.
The theoretical basis of information quantity method comes from the information theory proposed by the American engineer Shannon in 1948 [23]. In the application of landslide prediction, it is considered that the occurrence of landslide disaster is related to the quantity and quality of information acquired in the process of landslide prediction, and it is measured by the amount of information. The greater the amount of information, the higher the possibility of landslide disaster. Principle of information law is transforming the measured values of various factors of regional stability into information quantity. In other words, the information value of each factor evaluating regional stability is used to characterize the influence degree of each factor on regional stability [23]. Based on the single information quantity of the influence factor, superposition analysis can get all the factors combination the total quantity of information, which can establish an evaluation model of susceptibility for regional geological hazards. The greater the information magnitude is, the higher the possibility of geological disaster and the higher the risk degree will be [24]. Generally, the information magnitude can be calculated according to the following formula:where is the information quantity to geological hazards by the factor combination ; is the occurrence probability of geological disaster under the factor combination ; is the prior probability of geological disaster.
In practical application, due to the limitations of many aspects, such as the complex combination of factors and the large number of sample statistics, the application of formula (1) to calculate the amount of information is cumbersome. Therefore, the simplified single factor information quantity is usually calculated first, and then, the total factor combination information quantity is obtained by superposition analysis. In practical application, the value of single factor information can be calculated according to the following simplified formula:where Ni is the number of geological disaster units of a specific category distributed in factor xi; N is the number of units in the study area containing evaluation factor xi; Si is the total number of units containing geological hazards in the study area; S is the total number of evaluation units in the study area.
3. Risk Assessment of Highway Slope Based on ArcGIS Information Quantity Method
3.1. Evaluation Method and Unit Division
In the past, a large number of natural factors, DEM (digital elevation model), topographic slope, topographic aspect, topographic roughness, etc., were used as evaluation factors for the assessment of regional geological hazard susceptibility. However, due to subjective factors, such as artificial cutting excavation and the application of drainage facilities, the level of some natural factors has changed, such as slope gradient, height, deformation, and slope stability. Therefore, artificial excavation protection and other factors should also be considered in the regional risk assessment of highway slopes. In this paper, the combined action of natural and artificial factors is considered. Firstly, using natural factors, information on the 400 m regional highway bandwidth is calculated with the help of the information method and superposition of ArcGIS. Then, the influencing factors of the artificially excavated slope’s stability were selected as the artificial evaluation factors. The combined information values of the artificial factors in all excavated slope regions were obtained using the same method. Finally, the spatial analysis function of ArcGIS was used to superimpose and classify them, and the risk assessment results of geological disasters were obtained for the whole region.
The division of evaluation units would directly affect the accuracy of evaluation results and work efficiency. Therefore, an evaluation unit with higher accuracy and computational efficiency should be selected, if possible, for the ArcGIS evaluation model. ArcGIS’s raster unit had incomparable advantages over other evaluation units (grid unit and slope unit), such as simple calculation, less memory occupation, and higher accuracy. Therefore, the 30 m × 30 m grid unit was selected as the evaluation unit in highway slope hazard evaluation.
3.2. Data Sources of Evaluation Factors
For 850 slopes along national and provincial highways in Quzhou City, the geological hazard collapse and landslide information were obtained by collecting historical slope maintenance and geological engineering information. In the collapse and landslide risk assessment of a 400 m bandwidth area on both sides of the national and provincial highways, the slopes with potential safety hazards and those that produced adverse geological disasters are the research objects. The DEM (digital elevation model) data used in the study area were derived from the geographical spatial data cloud platform (https://www.gscloud.cn/home) at a resolution of 30 × 30 m. Slope gradient, slope direction, and degree of surface relief data were generated with the help of the ArcGIS 10.2 software analysis tools by DEM data. Quzhou City administrative divisions, road networks, and water system data were obtained from the National Earth System Science Data Center (https://www.geodata.cn/). Rainfall data were obtained from 68 monitoring stations in Zhejiang province from 1981 to 2010 by the National Meteorological Information Center (https://data.cma.cn/), and an annual average rainfall distribution map was generated using the ArcGIS 10.2 Kriging interpolation function. Regional fault and rock mass type data were obtained from the OSGeo China central website (https://www.osgeo.cn/).
3.3. Selection of Evaluation Factors and Computational Analysis of Information Quantity
Most of the national and provincial roads in Quzhou City are mountainous, and the stability of the natural slopes is largely affected by topography and landform. Furthermore, the annual average rainfall in Quzhou City is more than 1700 mm, which causes serious erosion and scouring effect on slopes. However, for the regional geological hazard assessment, randomness and uncertainty of collapse and landslide disasters are often controlled by various conditions regarding the complexity of the geological environment. Therefore, in this article, we combined the natural environmental characteristics of the study area and the present slope and gave full consideration to the related data availability, the elevation, topographic slope, distance from fault, rock and soil type, relief degree, distance from the river, average annual rainfall, terrain slope, and so on. Eight factors were selected as the natural evaluation factors, as shown in Table 1. For the manually excavated slope, based on maintenance history survey and engineering geological information collection, ten factors (slope excavation angle, slope excavation height, slope deformation history and present situation, soil slope compactness, soil slope moisture, rock hardness, degree of rock slope, development degree of structure, surface of rock slope, surface and underground drainage, degree of slope protection) were comprehensively selected as evaluation factors. All of the above evaluation factors were generated in ArcGIS with a resolution of 30 m × 30 m by using the relevant data obtained, and the Reclass tool was used to reclassify according to the factor classification in Tables 1 and 2. Finally, the raster calculation layers of 16 evaluation factors were obtained (due to the small soil slope, the compactness and moisture state of the soil slope were not considered in the figure), as shown in Figures 1 and 2.
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
ArcGIS was used for spatial superposition analysis on each evaluation factor grid cell layer and adverse geological hazard point distribution layer so that the distribution of adverse geological hazard points in each evaluation factor could be obtained. Thus, the value of each evaluation factor of natural and artificially excavated slopes was calculated by formula (2), as shown in Tables 1 and 2.
However, from Tables 1 and 2, for part of the evaluation factors, the value was negative because of less numerous geological hazard points with the factor classification. This was inconsistent with the true situation, such as the severe classification of slope deformation history. It was beneficial to the occurrence of geological disasters, but its information value was negative. Therefore, the values of all similar factors were analyzed and corrected one by one. It is appropriate for slope risk assessment to modify the value that is beneficial to disaster occurrence to be positive [25].
4. Risk Zoning of Highway Slope Based on ArcGIS Information Quantity Method
4.1. Analytic Hierarchy Process (AHP) Weight Determination
After obtaining the values of all evaluation factors, the natural slope and artificial excavated slope evaluation factors were combined and superimposed. For multifactor combination analysis, the weight of all factors should be determined first, and analytic hierarchy process (AHP) is a common method. By establishing the hierarchical structure model (Figure 3), constructing the judgment matrix, and consistency testing, the 1–9 scale method was used to compare and score the investigated factors in pairs to calculate the weight of all factors [26]. In this paper, the expert evaluation method was used to score the impact of different levels on slope risk. Thus, the weight values of the evaluation factor category, criterion layer, and indicator layer were obtained. Finally, the final weight values of all evaluation factors were obtained through multiplication, as shown in Table 3. As can be seen from Table 3, for the natural study area, the weight proportion of topographic relief factors was the highest, followed by topographic slope, rock, and soil mass type. For the artificially excavated slope area, the history and current status of slope deformation factors accounted for the highest proportion, followed by the density degree of soil slope, the development degree of rock slope structural plane, slope angle, and so on.
(a)
(b)
4.2. Geological Hazard Regionalization
According to the calculation results of single-factor information quantity in Section 3, the grid calculator in ArcGIS was used to assign values to the grid images of each factor in Figures 1 and 2. Then, the weighted information maps of the natural research area and artificial excavation area could be obtained by weighted superposition calculation in ArcGIS. Finally, the grid calculator was used to superposition the two, and the comprehensive information quantity map of the whole research area could be obtained, as shown in Figure 4. The information quantity value was [−0.868412, 0.777231]. Each grid in the figure represented an information attribute value, which is the influence degree of all factors combined in the grid region on slope geological hazard. The larger the information value, the more dangerous the region, and the higher the probability of geological disasters such as collapse and landslide.
The ArcGIS spatial analysis of Reclass tool and the natural discontinuous method were used to divide the comprehensive information quantity map of the study area into five categories. The default values of the discontinuity point were −0.642539, −0.255329, 0.125427, and 0.344846. Finally, the regional slope risk map was obtained (Figure 5). They were divided into basically risk-free area, mild-risk area, moderate-risk area, high-risk area, and extremely high-risk area [27]. As shown in Figure 5 and Table 4, the areas of the above five regions accounted for 48.35%, 22.57%, 13.10%, 10.21%, and 5.78%. Among them, the high-risk area and above accounted for 16.0%, mainly distributed in the northern and southern end of the Quzhou section of the G205 national highway and the junction of Changshan County and Jiangshan City of G205 national highway and S221 provincial highway. High-risk and extremely high-risk areas were mainly distributed in the high-altitude mountainous area. Geological disasters such as collapse, slump, landslide, and debris flow commonly occur in complex geological environments such as rivers, geological tectonic movement, terrain changes, and overlying loose stratum distribution, combined with abundant rainfall, which is also relatively concentrated.
5. Comparative Analysis of Slope Classification and Evaluation
5.1. Slope Classification and Evaluation Method
To further improve the highway maintenance management level and industrial governance capacity of Zhejiang Province, conduct detailed maintenance, and further strengthen the slope maintenance management of common national and provincial roads, Zhejiang Provincial Highway and Transportation Management Center proposed the “Implementation Opinions on the Establishment of long-term Mechanism for Slope Maintenance and Classification Control of Common National and Provincial Roads” (Zhejiang Gongyun (2020) No. 55) based on the pilot experience of Jinhua City.
Combined with the practice of slope management of common national and provincial roads in Zhejiang Province, any slope angles >30°\ were classified and evaluated according to the requirements of Table 5.
After obtaining the basic slope information, slope dynamic information, and slope protection information, a comprehensive evaluation was carried out. The specific classification evaluation indexes included slope deformation history and current situation (A), geometric characteristics (B), geological engineering conditions (C), hydrogeological conditions (D), and protection engineering status (E). The slope stability index (SH) of each slope was obtained through a weighted calculation based on the comprehensive analysis and evaluation of the investigation status. The calculation formula was determined as follows:(1)Simple slope protection project:(2)With important retaining works:where A, B, C, D, E are scoring values of each indicator.
Finally, the influence degree and influence range of slope disease on highway traffic and surrounding structures were comprehensively considered to obtain the highway slope evaluation composite index (SRI), which was calculated by the following formula (slope disease influence coefficient (SV) value is shown in Table 6):
According to the highway slope evaluation composite index (SRI), the slopes were divided into four levels, according to Table 7.
5.2. Results of Slope Classification and Evaluation
After collecting historical data, the geological engineering survey, and the analysis from 2020 to 2021, classification and evaluation of 850 slopes on five national and provincial highways (including G205, G320, G351, G528, and S221) in Quzhou City were completed. The specific classification and its proportion are shown in Figures 6. Figure 7 shows the distribution of various slopes along national and provincial roads in Quzhou City.
5.3. Comparative Analysis of Classification Evaluation and Regional Risk Regionalization
Through the ArcGIS spatial analysis and extraction tool, the distribution of various types of highway slopes in the regional hazard zoning map was obtained statistically, as shown in Figure 5 and Table 4 in Section 4.2.
As can be seen from Table 4, 99.1% of level 1 slopes were distributed in basically risk-free areas, 78.3% of level 2 slopes were distributed in moderate or lower risk areas, and 98.08% of level 3 slopes were distributed in high or above risk areas. The above conclusions indicate that the results of highway slope regional risk zoning were consistent with the slope classification evaluation.
6. Conclusion
In order to further improve the maintenance and management of highway slopes, this study took Quzhou City’s national and provincial roads as an example to conduct regional risk assessment research on highway slopes. Combined with the characteristics of highway slopes and considering the natural factors, artificial factors, and geological engineering information of individual slopes, the regional risk zoning map of highway slopes was obtained. The prone areas of geological disasters were defined, which provides a basis for the decision of highway slope maintenance management.(1)Compared with previous research methods, this study combined artificial influencing factors and individual geological slope engineering information on the basis of considering natural influencing factors. The evaluation results were more accurate and reliable.(2)Combined with the analytic hierarchy process (AHP) and ArcGIS spatial analysis model evaluation, the risk zoning map of banded areas along the highway was obtained. It not only points out the direction for highway slope maintenance management but also further improves management efficiency and reduces maintenance costs.(3)The regional risk division obtained in this paper based on the research of Quzhou City national and provincial road slopes was more consistent with the slope classification evaluation results and the actual slopes in Zhejiang Province, China. It can be used as the basic research method for slope management.
In the future, during regional risk assessments, evaluation factors can be selected specifically based on the research objects. Meanwhile, the field collection of research objects should be strengthened. This can greatly improve the accuracy of the research results and has more practical significance for project management.
Data Availability
The data supporting the current study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The research was performed as part of the employment of the authors at Zhejiang Institute of Communications Co., Ltd., Zhejiang Communications Construction Group Co., Ltd., and Chongqing Jiaotong University.