Investigation of Natural and Human-Induced Landslides in Red Basaltic Soils

1. Introduction

Landslides are mass movements of rock, soil, or debris driven by gravitational forces, particularly in highland areas. They can be classified on the basis of movements (fall, topple, slide, spread, flow, or slope deformation), materials involved (rock, earth, debris, or mud), and the velocity of movement [1,2,3]. Some landslides occur gradually, such as solifluction, soil creep, or slumping, while others happen rapidly, including mudflows, debris flows, and rockfalls. The speed and nature of movement are influenced by factors such as water content, slope angle, and grain size distribution [4,5,6,7]. Landslides can lead to severe consequences, posing significant threats to human lives and property [8,9].

Landslide is normally caused by natural factors such as geological structure, moisture, rainfall or vegetation cover [10,11,12,13]. Nowadays, landslide occurs more frequent due to climate change (heavy rainfall) and human activities such as deforestation, excavation, construction, or mining [14,15]. These activities alter the land cover and slope equilibrium leading to instability state [16,17]. Geological survey, seismic ultrasonic or soil properties analysis are used to elucidate the root cause of mass movement event in specific area at a certain moment [18,19,20]. Various traditional solutions have been applied and obtained a good results such as vegetation restoration, reinforcement, and land use management...[21,22]. Frequency and consequence of the landslide is dependent on the event location, soil composition and natural conditions [23,24,25,26].

Basaltic soil is a weatherd soil which is a weathering product of the basalt rock, a kind of fine-grained extruded rock [27]. Basalt rock contributes approximately 5% earth’s land cover, weathering and bahaviour of them are a crucial in global biogeochemical cycle [27,28]. Basaltic soil, also known as “red basaltic soil or laterite”, is nutrient-rich and fertile, making it ideal for numerous crops. Red basaltic soil consits of gibbsite (45-50%), goethite, kaolinite minerals [29]. Recent studies revealed that basaltic soil exposed to high precipitation may reduce the ferric (Fe³⁺) and cation (Na⁺, Ca²⁺), capturing the CO₂ via siliate weathering [27,30,31]. In addition the collidal properties of basaltic soil is sensitive to high rainfall that may reduce the surface charging of clay materials in the soil [29].

This study utilizes a comprehensive methodology to elucidate mechanisms of landslide events occurring in the red basalt soil due to high precipitation. A field observation has been conducted to examine the geological structure of the soil [10,22,27]. Undisturbed core sample were collected and analysed in the laboratory to examine the physical properties of the soils. In addition, numerical model has been used to analyse the stability of the slope under various rainy conditions. Findings from this study elucidate the root causes of the mass movement events which provides an in-sight of the rainfall-induced events in the red basalt soil area.

2. Materials and Methods

2.1. Field Investigation

Dalat city is located on the Langbian Plateau in the Central Highlands of Vietnam, at an elevation of approximately 1,500 meters above sea level. The topography is characterized by mountainous terrain in the northeast, plains in the southwest, and intervening valleys in the central region. Dalat experiences a subtropical highland climate with year-round temperate weather, and average temperatures ranging from 14 to 23 °C (Figure 2). The city has two distinct seasons: a rainy season from May to October and a dry season from November to April. The average annual rainfall of 2022, 2023 and 2024 were 2,165; 2,230, and 2,050 mm respectively. Dalat city has dense network of river and stream including Dong Nai mainstream river and its tributaries. There are two soil groups: red feralite soil and humus alisols, both distributed at altitudes between 1,000 and 2,000 meters [32]. The geological structure of Dalat comprises Precambrian basement rocks, Jurassic sedimentary formations, Late Mesozoic igneous rocks, and Cenozoic basalt formations [33,34,35].

Figure 1 illustrates the boundary of the survey area, where a field investigation was conducted from April 29^th to May 22^th, 2017, to study soil characteristics and geological structures. A subsequent observation was carried out in July 2019 to examine hydrological parameters and perform a pumping test. There was total 06 boreholes namely HK01 - HK06 were drilled up to 15 - 20 m to explore the geological structure. The surface layer consists of 1–2 meters of fill material, followed by a second layer dominated by lean clay with a reddish-brown to dark brown-white color. This layer is homogeneous and exhibits a hard plastic consistency. The deeper layer comprises silty clay rich in silt content, with colors varying from dark yellow to reddish brown. During the investigation period, cummulative precipitation reached 268 mm, with a peak at 65 mm. Undisturbed core samples were collected for laboratory analysis to assess the physical properties and mechanical behavior of the soils [36].

2.2. Physical Properties Analysis

The sieve experiments were conducted in the laboratory to determine the contribution of various grain sizes contained in a core sample. The mechenical sieve was used to determine the coarse particle size (> 75 µm) and the hydrometer analysis was utilized for the finer size [37,38]. The whole nest of sieves is given a horizontal shaking for 10 min in sieve shaker till the soil retained on each reaches a constant value. The contribution of different grain sizes regulates the physical properties of soil and its behaviour under external loads. Meanwhile, permeability, shear strength and moisture are reported to affect the landslide susceptibility negatively or positively [39,40].

Figure 2. Natural condition at the study site. The top panel indicates the rainfall (red column) and the average temperature the lower panel indicates the relationship between humidity (dash black line) and humidity in the red dash line.

Figure 2. Natural condition at the study site. The top panel indicates the rainfall (red column) and the average temperature the lower panel indicates the relationship between humidity (dash black line) and humidity in the red dash line.

2.3. Real-Time Observations System

A tilt sensor and strain gauge system were installed to monitor the groundwater table and horizontal displacement of the soil layers [10,41,42]. There total 15 sections of strain gauge were installed in a drill hole at every 1 meter to observe the deformation behavior inside the cut slope. Figure 3 illustrates the schematic of the early warning system with the tilt sensor, pipe strain gauge and water level gauge to detect any displacement of the slope. The gauge system was installed in a 60 mm diameter borehole drill to observe the horizon displacement up to 15 m at every meter. It is defined that, positive or negative values are registered for unstable condition of the soil layer [10]. Meanwhile, monitoring data is recorded every 10 minutes by the real-time data logger name as NetLG-301NE® (Osasi Co. Technos Inc, Japan). A wireless transmitter is used to transfer the monitoring data to the satellite orbit which could be received by the portable laptop computer or mobile devices. Signals from titl sensors reveals the horizontal displacement of the soil structure [10].

2.4. Slope Stability Analysis

There are several factors that govern stability during rainfall including slope geometry, soil properties, rainfall intensity, soi moisture, and initial water table [43,44]. The objective of numerical model is to highlight the relative importance of some of these factors on the stability of predominately unsaturated slopes. In his research, Fredlund et al. (1978) proposed a linear relationship between soil shearstrength and effective cohesion, air pore-air pressure and friction angle which is:

$$\tau =c^{\prime }+\left({\sigma }_n-{\sigma }_u\right)tan{\varphi }^{\prime }+\left(u_a-u_w\right)tan{\varphi }^b$$

(1)

where $\tau$ is the shear strength, $c^{\prime }$ is the effective cohesion, $\left({\sigma }_n-{\sigma }_u\right)$ is the net normal stress on the failure plane, ${\sigma }_n$ is the total normal stress, $u_a$ is the air pore-air pressure, $u_w$ is the pore-water pressure, $\left(u_a-u_w\right)$ is the matric suction, ${\varphi }^{\prime }$ is the friction angle, ${\varphi }^b$ is the angle linking the rate of increase in shear strength with increasing matric suction.

To determine the factor of safety, groundwater level, rainfall intensities, and soil properties were input to numerical model. Daily rainfall intensities in May 24^th to June 1^st 2022 was used to simulate the slope stability model. Observations groundwater level obtained from the tilt sensor system was also utilized for the model. Physical properties of the soil layers obtained from the sieve analysis was adopted for Mohr–Coulomb’s theory (Figure 4).

3. Results

3.1. Subsection

Figure 5 presents the average rainfall recorded in the calendar year of 2023, with cummulative total of 2,230 mm. In the observation period, the number of rainy day is 188 days, which are primarily between May to November, with the highest daily reaching 82.5 mm. During this period, three major mass movement events were recorded. The first event occured on June 17^th 2023 on the plantation farm, burying approximately 50,000 m² coffee crops. Within the following six hours, the situation escalated, resulting in the complete destruction of an additional 200,000 m² of coffee farms. On June 29^th, 2023, a 30-meter-high backfill slope failed, leading to two fatalities, three injuries, and damage to 12 houses. The most catastrophic event took place on July 30^th 2023, resulting in four fatalities and compeleted destroyed the section of National Road No. 20, a main route to Dalat City. This sudden landslide involved the failure of a 50-meter hill slope of plantation farm after two months of continuous rainfall. The high intense and frequency of rainfall altered the soil saturation condition, increased the groundwater level and elevated the risk of slope instability.

3.2. Physical Characteristic of the Soil Structure

Table 1 indicates the physical characteristic in various depth layers of the soil structure obtained during the field survey. The water content in various layers dropped down in range of 41.47 to 33.01 % vertically. In addition, the void ratio were reduced 10% from the top layer whereas the cohesion increased from 14.7 to 20.7 kPa. Figure 6a illustrates the fine grain size composition (< 75 µm) of soil structures with a contribution of > 70% in the core samples. On the other hand, the moisture is vertically dropped down in range of 45 – 30% (Figure 6b). The fine grain component with negative charged and high water retention, facilitates the soils structure [45]. Physical characteristic of the soil structure regulates the bahaviour of the slope to external factor.

3.3. Seasonal Water Table Variations

Figure 7 indicates the time series of groundwater table observed from the May 1^st 2022 to Jan 1^st 2023. During the rainy season, the average rainfall varied from 20 – 50 mm, hit a peak at 80 mm, whereas the rainfall was below 20 mm in the dry season. The number days of rainfall was four times to sunny days in this observed period. Intense rainfall has resulted in high accumulation of the precipitation (> 2000 mm) leading to high amplitude of groundwater table fluctuation. Rainfalls mediates the groundwater level, soil moisture and slope stability such as erosion, mass flow movements [46,47,48,49]. In addition, intense precipitation facilitates the variation of groundwater and pore pressure, leading to slope stability and landslide occurrence [44,50,51,52].

Figure 8 indicates the relationship between water content and shear strength parameters of the soil. It was seen that, the increase of water content reduced the cohesion (c) and friction angle ($\varphi$). These correlation enhances the instability of the soil structure which is in good agreement with previous studies [53,54].

3.4. Rainfall-Mediated Slope Stability Analysis

Figure 9 indicated results of nine-rainy day simulations using Plaxis^® 2D software. The accummulative precipitation is 403 mm with the highest rate of 81 mm. Results shown that, factor of safety dropped down from 1.135 to 1.09. It is noted that, groundwater level varied in -4.31 to -4.65 m in this period. The increase in groundwater table may result in soil moisture which affects the shear strength of the soil and slope stability [44,49].

3.5. Observation of Horizontal Displacement of the Soil Structure

Figure 10 indicated horizontal displacement of the soil structure in various depth layer observed from the tilt system in the period of 00:00:00 May 1^st 2022 to 23:00:00 Dec 31^st 2022. The 3-wired strain gaue system identified a signal in channel No. of 5, 6, 13 and 15. In channels 5 and 6 (in -5 – -6 m depth), the strain sensors responded a signal of ~8,000 microstrains whereas a microstrains of 40 – 120 were recored at the channel No. 13 and 16 respectively. Other layers were identified stable during the observation period. Strains gauge signals provides a an entropy value that can be used for detection and early warning of landslide events [55].

4. Discussion

4.1. Rainfall - Induced Weathering Reduces the Shear Strength and Slope Stability

Physicochemical properties of basalt soil is predominant with calcium, magnesium-rich silicate minerals, fenspate, and low total nitrogen, organic carbon [30,56]. Basaltic weathering plays a crucial role in global weathering patterns and biogeochemical cycling [28,57,58]. High precipitaiton enhances the weathering of the basaltic soils through exchanges of cation. Removing of the sodium ion results in the reducing of shear strength of the soil structures [30]. In addition, intense precipitation reduce Rainfall and rainfall frequency mediates the landslide whereas anthropogenic enhances the rate of the hazards [59].

Previous studies reveals that, water content, porosity, and clay content are crucial in the soil stability. Red basaltic soil typically comprises with Ca²⁺, Mg²⁺, Na⁺, Si⁴⁺, Al³⁺ and Fe³⁺ [60]. The higher cation exchange capacity (CEC), the higher Ca²⁺ and OH^- ions to be consumed for C–S–H cycle, resulting in the lower of soil strength. In addition, the cation release in structure may lead to increase the void volume that are available room for water entrapment, enhancing the void pressure of soil structures.

Figure 11. Conceivable mechanisms of slope failure due to rainfall-induced weathering process [61].

Figure 11. Conceivable mechanisms of slope failure due to rainfall-induced weathering process [61].

4.2. Groundwater Level Facilitates the Slope Stability

Groundwater table is a crucial factor on the slope stability [62,63]. Figure 7 illustrates the close relationship of ground water and rainfall. Water table increased in companion with the accumulative precipitation which facilitates the soil moisture, resulting in the drop down of cohesion (c) and friction angle ($\varphi$) that may result in decreasing of safety factor (See Figure 12). The higher of shear strength parameter in soil strucutre would enhance the factor of safety of the slope [64]. Figure 12c further indicates the highly correlated relationship of the variable safety factor and seasonal water table.

Fluctuation of groundwater level mediates the soil moisture which play a crucial roles in slope stability [65]. The rise of groundwater results in requent slope failure. In addition, higher of soil moisture reduces the shear strength of the fine grain slope due to reducing of cohesion force of the silt-clay particle [64]. Therefore, dewatering is one of the suitable solution to control the hydrological regime, reduces the slope unstability and slope mass movement problems [52].

4.3. Strain Gauge Sensor to Provide an Early Detect the Mass Movement

Early warning system (EWS) is now become popular in landslide prevention measure. The integrated system including internet of thing (IoT) on-site sensors automatically monitor and record parameters that cause landslides such as: slope displacement, hydrological and physical properties in soil, as well as precipitation using various tools [55]. Improvements in sensor technology and information transformation open a new horizon in multi-hazard mitigation and prevention. Signal recorded from strain auge sensor offers a valuable information to early detect the mass movement. The strain gause sensor continuously collects the data, transform to the users through wireless system and provide an alert on the susceptibility of the soil structure and slope stability.

5. Conclusions

This paper elucidates the infuence of rainfall and human activities on landslide disaster in the red basaltic soil in the highland terrain, highlight the following key findings are:

• High precipitation rates accelerate the weathering of basaltic soils by leaching essential cations, such as sodium, which leads to a reduction in the shear strength of soil structures. Moreover, frequent and intense rainfall events significantly increase the likelihood of landslides;
• Fluctuations in groundwater levels contribute to increased soil moisture, which, in turn, reduces shear strength parameters and destabilizes slopes;
• Human-induced activities exacerbate the frequency of mass movement events. Common triggers include slope cutting for construction and cultivation on steep, vulnerable slopes. These activities increase external loading and promote water entrapment within the soil, thereby reducing slope stability;
• Tilt sensor systems provide real-time data on groundwater levels and horizontal displacement at various depths. This technology forms a crucial component of early warning systems for multi-hazard risk management. In addition, integrating smart technologies and Internet of Things (IoT) solutions is highly beneficial for effective natural disaster management.

Last but not least, climate change—with increased precipitation and extreme weather events—poses an additional stressor to slope stability and hazard prevention efforts. Urbanization and agricultural activities further compound these risks. Therefore, the deployment of real-time monitoring systems is increasingly important. The early warning data they provide are vital for disaster prevention, mitigation, and safeguarding communities at risk.