Academic Journal of Science and Technology 
ISSN: 2771-3032 | Vol. 8, No. 3, 2023 

 

177 

Study	on	the	Cause	Analysis	and	Trend	Prediction	of	
Typical	Landslide	Disasters	in	Pukou	District	
Wei Shi1, *, Fuzhao Feng1 

1 Jiangsu Nanjing Institute of Geological Engineering Survey, Nanjing 210041, China 

* Corresponding author: Wei Shi (Email: shitou2048@126.com) 

 

Abstract: This paper takes the typical soil-rock binary structure landslide on the east side of the middle section of Yanshan 
Avenue in Pukou District as the research object. Through field investigation, geological surveying and mapping, Lidar scanning, 
drilling, rainfall and deformation monitoring data analysis and other means, the basic characteristics of the landslide and the 
factors affecting the stability of the landslide are analyzed. It is found that the expansibility of kaolin and the floating support 
force of temporary confined water are one of the factors affecting the stability of the slope. Through real-time monitoring to 
explore the relationship between rainfall and surface horizontal displacement and vertical displacement, it is found that the 
surface deformation is greatly affected by the season. When the rainfall is large, the impact of temporary confined water has a 
large time lag. The faster the speed to reach the saturated state, the shorter the lag time, and vice versa. On this basis, the 
relationship model between landslide and rainfall is established, and the development trend of landslide is predicted. 

Keywords: Soil-rock binary structure; Landslide; Failure mode; Rainfall; Trend prediction. 
 

1. Introduction 
Landslide geological disaster is a common natural disaster 

that poses a huge threat to human life, property and the 
environment. In recent years, with the continuous growth of 
population and economy, human destruction of the natural 
environment has become more and more serious, resulting in 
the increasing frequency and scale of landslides. Between 
2021 and 2023, China has experienced multiple landslide 
disasters, causing significant economic losses. The causes of 
landslides are multifaceted, mainly including the influence of 
geological conditions, climate environment, hydrology and 
water conservancy, land use and other factors. Among them, 
human activities have a particularly obvious impact on 
landslide occurrence. For example, the acceleration of 
urbanization, large-scale land use change, engineering 
construction and other activities may lead to the change of the 
original geological environment, thereby increasing the 
probability of landslides. Therefore, how to control and 
prevent landslide disasters and reduce their impact on humans 
and the environment has become one of the important 
scientific research topics today.  

This paper will discuss and analyze the causes and 
development trend of landslides. Firstly, through the study of 
the causes of landslide disasters, the impact of different 
factors on landslide occurrence is discussed, and 
corresponding prevention and control strategies are proposed. 
Secondly, the trend of the current landslide development is 
predicted, and the scale and frequency of future landslides are 
discussed, and corresponding early warning and 
countermeasures are proposed. In recent years, the rainfall in 
Pukou District is not stable [1-3], increasing from 1091mm in 
2014 to 1768mm in 2015, and suddenly to 1807.7mm in 
2016.Moreover, summer rainstorms occur frequently, with 
monthly rainfall reaching 661.5mm in June 2015 and 
477.3mm in July 2016.Therefore, there are many rainfall-
induced landslides in Pukou and its surrounding areas, such 
as Fangshan, Niushoushan, and Youzishan in Jiangning [4-6]. 

In 2015, due to the heavy rain, a typical landslide in Pukou 

District developed cracks on the basis of existing anti-slide 
piles, with obvious shear outlets. Through real-time 
monitoring data, the entire gentle slope area in front of the 
mountain is still in an intermittent creep state. According to 
the analysis, the reason for the resurrection of the landslide is 
frequent rainfall. 

At present, the landslide has not caused casualties. For the 
overburden landslide with a gentle slope and poor stability in 
Pukou District, many scholars have conducted research. 
Zhong Yinqian [7] et al. believe that 90% of landslides are 
related to rainfall. Liu Guoen [8] et al. analyzed the 
deformation characteristics of the landslide body, the 
formation mechanism and external forces of the landslide in 
Pukou District, and proposed relative preventive measures. 
Yan Xiaoying [9] et al. concluded through numerical 
simulation and theoretical analysis that the landslide in Pukou 
District is a rainfall-type landslide, and the failure mode is the 
floating support force and seepage force generated 
synchronously with rainfall. Chang Jing [10] concluded 
through survey results that the sliding surface is mainly 
located in the kaolin layer that contacts the permeable layer 
and the impermeable layer, and the sliding surface is broken 
line type. Yu Liangchen [11] analyzed the formation process 
of temporary confined water in the landslide under the 
condition of continuous rainfall through numerical simulation, 
and analyzed the impact of the formation of temporary 
confined water. In this paper, through the field survey means 
of UAV aerial photography, crack measurement, slope 
displacement monitoring and other methods, combined with 
the previous survey results, the basic characteristics, 
boundary conditions, factors affecting the stability of the 
landslide and the sliding mode of the landslide are analyzed. 
Taking the landslide HP2 as the research object, real-time 
monitoring data processing is carried out. 

2. Analysis of Basic Landslides 
The typical landslide in the project area is located at the 

eastern end of Laoshan Mountain in Pukou District. 



 

178 

Influenced by the fault control in Pukou District, a set of 
Tertiary sand layers were deposited on the northwest slope of 
Pukou District. In the 1970s, there were large-scale mining 
activities in the area, and a large amount of artificial filling 
was filled, which increased the load of the slope and changed 
the circulation conditions of groundwater. After 2000, due to 
the shortage of land resources, the foot of the slope was 
gradually used. In the process of land use, unreasonable slope 
cutting caused the slope to slide many times, and many houses 
on the foot of the slope and the slope were destroyed, causing 
a large property loss and many people were threatened by 
landslides. The northern slope of Pukou District has long been 
a landslide disaster prone area. There are six large-scale 
landslides in the northern slope of Pukou District, forming a 
landslide group, which is numbered HP1-HP6. The specific 
distribution is shown in Figure 1. 

 

 
Figure 1. Landslide distribution map of the north slope of 

Pukou District 
 

In 2011, the local government carried out staged treatment, 
and the landslide has become stable.The geological disaster 
area of HP2 landslide is located at the northern foot of 
Laoshan Mountain.The aerial photograph of the landslide is 
shown in Figure 1.Due to the heavy rain in 2015, the HP2 
landslide has a resurrection trend, and the landslide has 
deformed. There are great potential geological disasters in 
this area. The landslide body is still in the creep stage, and the 
landslide body has cracks and obvious shear exits. The 
schematic diagram of the landslide danger is shown in Figure 
2. 

 

 
Figure 2 Schematic diagram of partial landslide hazards 

 

After preliminary investigation and demonstration, there 
are potential landslide geological disasters with a width of 
300m and a length of about 200m. It is estimated that the 

potential landslide volume is about 60,000 cubic meters, and 
the landslide scale belongs to small landslides. The landslide 
has caused obvious deformation in the local area of the built 
anti-slide piles, and the walls and slope protection at the shear 
exit have obvious cracks (as shown in Figure 3, Figure 4).The 
lower edge of the main sliding direction of the sliding body is 
threatened by about 60 households, houses and several nearby 
enterprises. Therefore, the HP2 landslide will pose a serious 
threat to people's life and property safety.  

 

 
Figure 3. Tension cracks at the trailing edge of landslides 

 

 
Figure 4. The fence damaged by the landslide shear exit 

 

The landslide section is mainly composed of the Xiashu 
formation silty clay. The stratification and physical properties 
of the soil in this section are shown in Table 1. Within the 
exploration depth range, the first layer is recently artificially 
filled soil with complex composition, containing more gravel, 
poor soil quality and large thickness, new time, soft structure, 
and poor stability; the second layer is residual slope 
accumulation soil with poor engineering geological 
properties, medium compressibility, medium and low strength, 
and extremely unstable distribution, which is not suitable for 
the supporting layer of retaining wall foundation; the third 
layer of soil has good engineering geological properties, 
medium and low compressibility, medium and high strength, 
and weak swelling potential, among which the strength of the 
3-2 layer of soil drops sharply when water is added, and the 



 

179 

water stability is poor; the fourth layer of gravel layer is 
locally distributed, medium and high strength; the fifth layer 
of soil is low compressibility, high strength soil layer. The 

stratigraphic structure of the landslide area is shown in Figure 
5. 

 
Table 1. Physical property indexes of soil (mean value) 

Number Name 
w γ e WL Wp Ip IL 

% KN/m3 — % % — — 

1 filling soil 27.5 18.6 0.828 38.3 22.0 16.3 0.50 

2-1 Powdery clay 26.8 19.0 0.796 39.4 22.0 17.4 0.26 

2-2 Powdery clay 27.9 19.5 0.728 34.7 18.4 16.3 0.47 

3-1 Clay 24.6 19.2 0.756 43.9 23.6 20.4 0.06 

3-2 kaolin clay 24.7 19.3 0.745 43.8 24.4 19.3 0.03 

4 gravel sand 14.4 19.1 0.594     

4—A Powdery clay 19.3 19.5 0.640 36.7 21.4 15.3 -0.14 

5-1 Clay 21.6 19.6 0.680 43.7 23.7 20.0 -0.12 
 

 
Figure 5. Schematic Diagram of Landslide Stratum Structur 

 

The three layers of soil and the 5-1 layer of soil in the 
survey area have certain swelling and shrinkage, among 
which the 3-2 layer is kaolin clay with a free swelling rate of 
46.7%, and the 5-1 layer of soil has a free swelling rate of 
44.3%, which is a weak swelling potential soil. The swelling 
deformation is large when dry and wet alternately, and the 
strength decreases greatly after long-term saturation, and the 
water stability is poor. In addition, according to the survey 
data of various expansive soil slopes, the natural stable slope 
foot of expansive soil is small, generally 3-20o, and the vast 
majority is less than 14o, which may increase the possibility 
of landslides. From the stratum situation revealed by this field 
investigation and survey, it can be seen that filling is the main 
constituent material of the landslide, and the sliding surface is 
basically located in the 3-2 layer of kaolin clay.The 4-layer 
gravel layer is sandwiched between the 3-layer and 5-1 layer 
of clay. When heavy rain falls frequently, the permeability 
coefficient of the 3-layer and 5-layer clay layer is small, 
which is quite different from that of the 4-layer gravel layer. 
The gravel layer will temporarily store a large amount of 
rainwater and cannot be excreted. The 3-layer and 5-layer 
clay layer can be used as a water barrier relative to the gravel 
layer, which will eventually cause the gravel layer to be 
saturated for a long time after the rainstorm. When the layer 
is saturated, the groundwater has a certain bearing capacity. 

Therefore, it is preliminarily estimated that there are two 
reasons for the landslide activity: first, the infiltration of 
rainwater significantly reduces the physical and mechanical 
parameters of the soil, thus reducing the stability of the slope. 
Under the action of external forces such as water seepage, the 
slope cannot maintain its stability and eventually 
slides.Second, the 3-layer and 5-layer clay layer swells due to 
water absorption, causing the stratum to rise, and the 4-layer 
groundwater with bearing water quality under saturated state 
floats and supports the overlying stratum, which destroys the 
stability of the slope and eventually leads to the sliding of the 
slope. In order to effectively avoid unnecessary damage 
caused by geological disasters in this area, the following 
measures have been taken:  

(1) Evacuate all residents in the residential buildings near 
the slope for three rows;set up a security cordon to prohibit 
personnel from entering.  

(2) Conduct real-time monitoring of the top of the built 

anti-slide piles and the landslide body, and report the 

displacement data of the landslide body and the anti-slide 

piles every day. 



 

180 

3. Analysis of Landslide Monitoring 
Point Data 

In order to better understand the specific situation of the 
typical landslide in Pukou District, eight monitoring points 
are set on the slope of Pukou District, named J1-J8. The 
setting position is shown in Figure 6.  

 

 
Figure 6. Landslide monitoring point layout 

 

The monitoring time of the eight monitoring points in 
Pukou District is from July 14, 2017 to February 1, 2021. In 
order to better represent the horizontal deformation of the 
monitoring body, two axes A and B are set. The setting 
position of axes A and B is shown in Figure 8. The direction 
of axis B is parallel to the long axis direction of the disaster-
hit object, and the direction close to the monitoring body is 
positive, and the direction away from the monitoring body is 
negative. The direction of axis A is perpendicular to the long 
axis direction of the disaster-hit object, and the direction away 
from the disaster-hit object is positive, and the direction close 
to the disaster-hit object is negative. 

Through statistics of the monitoring data, the cumulative 
vertical displacement map of the monitoring points is 
obtained (as shown in Figure 7). The monitoring vertical 
displacement is actually the settlement value of the 
monitoring body, with vertical downward as positive and 
vertical upward as negative. According to the monitoring data 
statistics, as of February 1, 2021, J1 and J2 are subsidence, 
with the cumulative settlement values of 54.2mm and 
159.7mm respectively.J3, J4, J5, J6, J7 and J8 are all uplifted, 
with the cumulative uplift values of 159.3mm, 94.0mm, 
112.9mm, 68.8mm, 115.4mm and 5.8mm respectively. Thus, 
it can be seen that most of the monitoring points are in a 
overall uplift state. The data fluctuate in June to August each 
year. Due to the large amount of rainfall, the rainwater cannot 
be discharged in time, which makes the expansion of the 
kaolin layer push the upper layer and the floating support 
force of the gravel layer reach the peak. At this time, the curve 
appears a trough, and the slope body has an overall uplift. 
With the passage of time, the groundwater gradually 
discharges, the support force gradually decreases, and the 
vertical displacement returns to the value before rainfall, and 
the curve rises. The vertical displacement tends to be stable in 
March 2020, without further expansion trend.  

 

 
Figure 7 Cumulative vertical displacement map of 

monitoring points 
 

In order to further discover the impact of rainfall on vertical 
displacement, the timeliness of the impact of kaolin 
expansion and temporary confined water on the vertical 
displacement of the slope after rainfall is explored. Draw the 
comparison chart of the uplift value and rainfall of J3-J8 (J1 
and J2 are excluded due to the final result of real-time 
monitoring for subsidence), as shown in Figure 10-Figure 
12.The span is three years from July 2017 to August 2020, 
and the rainfall is measured in units of months. To reduce the 
contingency and represent the overall situation of each month, 
the monthly uplift value is the average of the daily uplift value. 
To better highlight the relationship between the uplift value 
and rainfall, the Y axis is selected as being greater than 0 for 
uplift and less than 0 for subsidence. Since the vertical 
displacement of the monitoring points in each month is an 
independent sample, the law can be found by observing the 
time node when the uplift value is greater than 0. 

By analyzing Figure 8-Figure 10, it can be found that 
although the changes in the degree of rainfall received by each 
monitoring point are not the same, such as the sensitivity of 
monitoring points J7 and J8 to rainfall is higher than that of 
monitoring points J3-J6, the changes are basically the same. 
When the monthly rainfall is at a low level, the overall trend 
of the monitoring points is basically settlement. But when the 
monthly rainfall increases, it can be found that the influencing 
factors of rainfall on the slope have changed, and the 
dominant influence is the reduction of the mechanical 
properties of the soil layer caused by rainwater infiltration, 
which becomes the influence of the swelling of expansive soil 
and the floating support force of temporary confined water. 
The latter has a large time lag, with a lag time of half a month 
to one and a half months. The reason for this phenomenon is 
that rainwater infiltrates into the 3-2 layer of kaolin soil, 
which expands when absorbing water, resulting in a part of 
rainwater staying in the kaolin soil layer. With the increase of 
rainfall, rainwater enters the gravel layer. When the gravel 
layer reaches saturation, it will generate temporary confined 
water, which exerts floating support on the overlying soil 
layer. It can be seen that when the rainfall is large but does not 
reach a certain value, the thrust of expansive soil is dominant. 
When the rainfall is large and the fourth layer of gravel 
reaches saturation, the floating support of temporary confined 
water and the thrust of expansive soil work together. 
Therefore, the reasoning lag time depends on the speed of the 
gravel layer reaching saturation. The faster the speed of 
saturation, the shorter the lag time, and vice versa. When the 
rainfall is low for a long time, the temporary confined water 



 

181 

in the gravel layer is discharged through runoff, and the 3-2 
clay layer expands and gradually becomes constricted. As a 
result, the value of vertical displacement gradually decreases, 
and the amount of sedimentation gradually increases until the 
next rainy season. 

 
Figure 8. Comparison of vertical displacement and rainfall 

amount of monitoring points J3 and J4 
 

 
Figure 9. Comparison of vertical displacement map and 

rainfall of J5 and J6 monitoring points 
 

 
Figure 10. Comparison of vertical displacement map and 

rainfall of J7 and J8 monitoring points 

The horizontal displacement data of the monitoring points 
were statistically analyzed, and the cumulative displacement 
maps of Axis A and Axis B were generated.(Fig.11-Fig.12) 
The analysis of the cumulative displacement of Axis A 
revealed that J1-J8 moved towards the negative direction of 
Axis A, close to the disaster-affected objects, with a 
maximum displacement of -883.8mm;the analysis of the 
cumulative displacement of Axis B revealed that J1-J8 moved 
towards the positive direction of Axis B, away from the 
disaster-affected objects, with a maximum displacement of 
404mm.The overall landslide had a large displacement, but it 
has basically tended to be stable.Through the curve, it was 
found that the horizontal displacement was greatly affected 
by seasonality. In August 2018, the cumulative displacement 
increased, and the cumulative rate reached the maximum 
value. However, according to the statistics of the National 
Bureau of Statistics, in 2018 and 2019, the rainfall was 
1267mm and 721mm respectively, and in August 2018, the 
rainfall reached 273.6mm. Therefore, it was guessed that due 
to the large rainfall, the kaolin soil layer expanded, causing 
the horizontal displacement of the layer in the direction of A 
and B to increase, and the horizontal displacement value 
increased continuously. The data tended to be stable in early 
2020.There was a small fluctuation in June 2020, because the 
rainfall from June to July 2020 was above 270mm, similar to 
that in August 2018. The sudden increase in horizontal 
displacement caused by the rainfall in 2018 can demonstrate 
the effectiveness of the treatment, which effectively controls 
the horizontal displacement of the slope. 

 

  
Figure 11 Landslide vertical displacement-time curve  

 

  
Figure 12 Landslide horizontal displacement-time curve 



 

182 

4. Conclusion  
In the geological disaster investigation and treatment, this 

paper analyzes the basic characteristics of the HP2 landslide 
in Pukou District and the factors affecting the stability of the 
landslide, and explores the relationship between time and 
vertical and horizontal displacement through real-time 
monitoring. It also explores the relationship between rainfall 
and vertical displacement, and draws the following 
conclusions: 

(1) There are two reasons for the landslide activity: First, 
the rainwater infiltration causes the soil physical and 
mechanical parameters to be greatly reduced, which reduces 
the stability of the slope. Under the action of external forces 
such as water seepage, the slope cannot maintain its stability 
and eventually slides. Second, the 3-layer kaolin layer and 5-
layer clay layer swell due to water absorption, and the gravel 
layer in the saturated state, the groundwater with the property 
of bearing water has the floating support force on the 
overlying strata, which destroys the stability of the slope and 
eventually leads to the sliding of the slope. 

(2) The vertical displacement is greatly affected by the 
season. When the monthly rainfall increases, it can be found 
that the influencing factors of rainfall on the slope have 
changed. The dominant factor is the rainwater infiltration, 
which leads to the reduction of the mechanical properties of 
the soil layer, and becomes the influence of the expansion of 
the expansive soil and the floating support force of the 
temporary bearing water. The latter has a larger time lag, and 
the lag time depends on the speed at which the gravel layer 
reaches the saturated state. The faster the speed reaches the 
saturated state, the shorter the lag time, and vice versa. 

(3) The horizontal displacement is also greatly affected by 
the season. Due to the large rainfall, the kaolin layer expands, 
causing the layer to have a large horizontal displacement in A 
and B, and the horizontal displacement value increases 
continuously. 

References 
[1] National Bureau of Statistics. 2016 China Statistical Yearbook 

[M]. Beijing: China Statistics Press, 2015. 

[2] National Bureau of Statistics. 2016 China Statistical Yearbook 
[M]. Beijing: China Statistics Press, 2016. 

[3] National Bureau of Statistics. 2017 China Statistical Yearbook 
[M]. Beijing: China Statistics Press, 2017. 

[4] Yan Changhong, Xu Baotan, Wu Chengyu, et al. Stratigraphy 
and Stability Analysis of the Frontal Mild Slope in Ningzhen 
Area [J]. Journal of Engineering Geology, 2019, 27(1):48-54. 

[5] Yan X Y, Xu B T, Zhang L P, et al. Mechanism analysis of a 
landslide in highly weathered volcanic rocks of Niushoushan 
Hill in Nanjing[J]. Environmental Earth sciences, 2019, 78(24). 

[6] Yu Liangchen, Yan Changhong, Guo Shulan, et al. Analysis of 
the sliding mechanism of Youzishan in Nanjing affected by 
temporary confined water[J]. Hydrogeology and Engineering 
Geology, 2020,47(1):133-141. 

[7] Zhong Yinqian. Relationship between landslides and rainfall 
and their prediction[J]. Chinese Journal of Geological Hazards 
and Control, 1998(04):82-87.  

[8] Liu Guoen, Dai Bing, Sui Zhaoxian. Causes and prevention 
measures of landslides in Pukou District, Nanjing, Jiangsu 
Province[J]. Jiangsu Geology, 2007(04):319-322. 

[9] Yan Xiaoying, Yan Changhong, Tan Jinzhong, Wu Chengyu, 
Guo Shulan, Xu Chenghua, Wang Zhenxiang, Shi Wei. 
Reanalysis of the reactivation mechanism of a hillside gentle 
slope landslide in Pukou District, Nanjing[J]. Geological 
Review,2020,66(06):1626-1636. 
DOI:10.16509/j.georeview.2020.06.014.  

[10] Chang Jing. Analysis of influencing factors and stability 
evaluation of the northeastern slope landslide in Pukou District, 
Nanjing[D]. Shandong University of Science and Technology, 
2019. DOI:10.27275/d.cnki.gsdku.2019.001062. 

[11] Yu Liangchen. Simulation analysis of a hillside gentle slope 
landslide in Nanjing under the action of temporary confined 
water[D]. Nanjing University, 2020. 
DOI:10.27235/d.cnki.gnjiu.2020.000707.