Geological, geophysical, and geotechnical approaches on the slip surface investigation on Cisumdawu Highway, West Java, Indonesia

This study used geological, geophysical, and geotechnical methods to investigate the causes of landslides in a newly established Cileunyi Sumedang Dawuan (Cisumdawu) Highway, Sumedang Regency, West Java Province. A combination of geological field observation, subsurface geoelectric resistivity-based survey, and borehole drilling were applied to obtain the subsurface data of a research area. A total of 10 boreholes were drilled to collect geoengineering data, including the soil material and N-SPT value for validation purposes. The soil hardness and resistivity were measured and compared to establish the relationship between resistivity and engineering properties. The result revealed that percolating water zone in the permeable loose soil was located above the impermeable layer, estimated as a slip surface. The analysed subsurface measurement and borehole data showed that the depth of the slip surface is between 26-27m. The study is important to determine the mitigation steps for the highway slope stability construction.


Introduction
Landslides occur due to a combination of triggers and causes, both by natural and artificial factors.It includes by the characteristics of the in-situ geology, human activities and heavy precipitation.The interferences from human for example, roads and settlements construction, which then cause the slope instability and vegetation disturbance in that area.The Cisumdawu Highway construction in Sumedang Regency is causing disruption to the slope stability in the hilly part of the highway.A soil creep, slowmoving landslide, occurred between 2016 and February 2021 at Section 2, Station 21 of the highway (Figure 1).This landslide study is using geological, geotechnical and geophysical approaches.Several studies applied these techniques for studying a specific type of landslide (e.g., creeping) at a certain site (e.g., highway) [1], [2], [3].
The assembly of Cileunyi-Sumedang-Dawuan Highway, abbreviated as Cisumdawu Highways started in 2013.The core reason of making this highway is for the efficiency of the trip to the Kertajati International Airport and the growth of Ciayu Majakuning (Cirebon, Indramayu, Majalengka, and Kuningan) economic area in West Java [4] gradually since 2016 have cut off the road at Section 2 and damaged 18 houses, paddy fields and some plantation land near the section (Figure 2).The affected area is located between Sirnamulya Village and Mulyasari Village.It is situated at about 500 -650 m asl, on a hilly to slope of 10° -30°.Based on the susceptibility to landslide zone map of Sumedang Regency [5], the area is in a zone of low to moderate susceptibility to landslide.Furthermore, through Lidar mapping provided by Ministry of Public Works and Housing (PUPR), shows that there are old landslide scars at this site.
There are several landslide studies in other section of this toll road [6], [7].However, those studies have not provided a clear understanding of the subsurface condition.The objective of this study is to investigate the slip surface along the slope of Section 2, both surface and subsurface condition, by correlating the geological, geophysical and geotechnical methods.The result is applicable to determine the mitigation steps for the highway slope stability construction.

Methods
This study combines three methods: geological, geophysical, and geotechnical.The primary data for geological analysis acquired from fieldwork by employing geological observation.The geological setting as a secondary data obtains from a geological map of the Bandung Sheet, Java at a scale of 1:100,000 [8].Based on the Geological Map [19], the research area consists of the Young undifferentiated Volcanic Rocks (Qyu) of Quaternary age, which is composed of tuffaceous sandstone, lapilli, volcanic breccia, and agglomerate lava.
The geophysical technique on this study was the electrical resistivity tomography (ERT) method measurement.The resistivity method has been confirmed to be functional for subsurface study related slope prone to landslide [9], [10].ERT depicted soil moisture; delineate landslide body, shape, slip surfaces and depth; and delineate slope instability [11], [12].The measurement was performed in a profile length of 470, using 4 electrodes linked in a straight line with 5 m electrode spacing, lines 1 -4 were outlined from north to south, intersecting lines 5-8 from west to east (Figure 3) [18].The processed data was completed by RES2DINV software and the Wenner-Schlumberger configuration [13], [14] [15].The model profiles presented in two-dimensional (2D) electrical resistivity section (resistivity subsurface image), showed the distribution of the electrical resistivity values.Then, the resistivity contrasts due to the lithological nature of the terrain and water content variation were identified.The geotechnical approach was carried out to determine the bearing capacity of the soil using a Standard Penetration Test (SPT) [18].The samples extracted from fourteen boreholes (Figure 4), near landslide sites, and spatially selected according to geological profile and geophysical assessment.Ten boreholes were drilled to determine the soil material description and N-SPT value, but only two boreholes, namely Station 600 (STA 21+600) and Station 550A (STA 21+550A), were chosen due to the significant data observed.

Result and Discussion
Geological field site observations of the research area consist of pyroclastic fall and lava flow deposits as shown in Figure 5.The pyroclastic fall is deposited over more than 50% of the area and is composed of scoria, lapilli, as well as ash.Scoria fall deposits are gray in fresh outcrop and light brown in weathered, but they are strongly weathered.The fragments are subangular to rounded grain sizes.Furthermore, clast size ranges from 1.5 to 5.5 cm in diameter, and basaltic andesite to andesite lithic has clasts of 0.2-3 cm in diameter.
The lava flow deposits showed grey to light brown basaltic andesite to andesite compositions, massive to sheeting jointing, porphyritic with plagioclase, and pyroxene phenocrysts, moderately to highly weathered.Organic soil was also at the top with 0.75-1 m thickness.Claystone found in the westnorthern part of the research area was grey to yellowish brown, strongly weathered, and has a soft consistency.The highly weathered rocks could lessen the hillslope stability and cause landslides during intense precipitation.
Geological structures, by means of joints, were discovered at the study site, mostly found in andesitic basaltic lava rocks from volcanic effusive eruptions, as a result of the activity of the Cileunyi-Tanjungsari fault.Joint bedding plane indicates weak zones which could generate slope failures.The outcomes of geophysical measurement are 2D images from lines 2A (STA 21+550A) (Figure 6) and 4A (STA 21+600) (Figure 7), with overview of the rock resistivity values (Figure 8).Based on the information, the layer with water-saturated and silt or clay has low resistivity range of at 1-100Ωm.This is validated by the discovery of outcrops of gray and weathered clay in the northeastern part of the research area.Below the water-saturated layer and also a harder rock in some places, a slightly higher resistivity value with range of 50 -200Ωm were found in which are presumed to be clay.The dividing line between these two layers, silt or clay layer and harder rock layer, with different resistivity values was reviewed as a slip surface of the landslides.Several soil gradations were spotted in a scattered position in Line 2A, while in Line 4A, the generality was located in the center, on the right with a shallower depth, and on the left with some deeper depths inside the resistivity images.The finding from the geophysical survey, the resistivity anomaly interpretation, have been confirmed by the geotechnical studies, the bore core data analysis of Station 600 and Station 550A.According to the Station 600, the subsurface profile material began as a soft layer of clay to sandy clay at a depth of 9 m and gradually changed to a hard layer at a deeper level.
The permeable soil percolated with water accumulation of Line 2A (Figure 6) was mostly located the left-hand side of the image, between elevation 575 -590 m asl (downslope and north of STA 21 500a mark), and at the right-hand side of the image (near the south), between elevation 570 -590 m asl, above slip surface line.The result for Line 4A (Figure 7), the result showed that the resistivity tomography for permeable soil percolated with water accumulation was significantly located at the center of the image (below the 21 600b mark), between elevation 560 -600 m asl, and at the right-hand side of the image, between elevation 565 -590 m asl, above slip surface line.Due to the presence of a percolated water zone, the materials low resistivity value was produced.This zone might have developed due to water infiltration from surface runoff and direct rainwater through porous subsurface materials in addition to the area preexisting groundwater zone.
Both geoelectrical images of Line 2A and Line 4A depicted that the resistivity subsided as the fine material and moisture content increased.It is also demonstrated that the thickness of the existing hard materials varied by approximately 10m and between 25m and 30m, respectively.
The results of laboratory tests and data from bore cores (Table 1 and Table 2) revealed that the subsurface soil profile was diverse, but mostly dominated by clay group (tuffaceous clay, sandy clay, and silty clay) and sand group (sand, sand pebble, and sand cobble).It also provides information on the soil hardness.For Station 550A and Station 600, the N-SPT values were higher at a depth of 27 m and above 26 m, respectively.Based on the 2D profiles of resistivity and the bore cores data, the study area significantly constructed of zone with resistivity value ranges from 1-1500 Ωm.The clay group layer has resistivity value range of 1-200Ωm were found at depth around 4-30 m, and the soil could be classified as stiff soil to hard material [17].The next layer consists of sand group with Ωm of > 200 Ωm, and it could be classified as very hard material [17].The resistivity also finds a fairly stiff granite readings, which ranged from 500 to 1,500Ωm.
Weak zones were identified in both selected samples, Line 2A and Line 4A, located near the downslope on the right-hand side of the images (Figure 6 and Figure 7).The condition might be affected by surface water discharge that reaches the valley or drainage path through the permeable material.As a result of its low resistivity levels, the localized weak zone, with 26-27m deep from north to south, can develop into a dangerous zone.This section, the border between high and low resistivity values, distinctly prone to soil erosion or slope failure, can be viewed as a weakness plane (slip surface).The N-SPT test results also showed that the slip surface of the deep-seated landslide at Station 550A and Station 600, located at a depth of 26-27m.This finding combined with the aforementioned fact of the occurrences of in-situ joints and the highly weathered rocks which also the indication of weak zones on a slope, is further proof that this region is prone to landslide disaster.

Conclusion
This study concluded that the combination of geophysical and geotechnical measurements, with an addition of geological field assessment, could discover the slip plane below the surface.The resistivity method validated by boreholes and laboratory examination of soil samples, and coincided with the result from field observation.The existence of landslide disaster in the study area, which damaged highways, paddy fields, and houses, were caused by the slope failures along the slip surface.The approaches used in the case study could be implemented in area with similar environment.For mitigation purposes, due to mostly high SPT values (N) in this area (> 15), the constructions on the highway slope suitable for medium to high load bearing structures.For further investigation, the determination of rocks or soil weathering and other geophysical methods is recommended to minimize the ambiguity of interpretation.

Figure 1 .
Figure 1.The research area and simplified geological map modified from Silitonga [19].

Figure 2 .
Figure 2. Landslide effects on roads and buildings in Cisumdawu Highway Station 21 and overview of Study area, Cisumdawu Highway, Sumedang Regency, West Java Province.

Figure 3 .
Figure 3.A map of the resistivity measurement area.The yellow line is a geo-electric data acquisition line.Red arrows show a general direction of groundwater.

Figure 4 .
Figure 4. Borehole locations along the Cisumdawu Highway of station 21.

Table 1 .
Material description and N-SPT Value of station 21+550A

Table 2 .
Material description and N-SPT Value of station 21+600