Core Log and Cone Penetration Test Approach for Bearing Capacity Analysis of Quaternary Deposit and its Correlation to Facies Distribution in Southern Bali

Area development deals with optimal land use and the reduction of the risk of geological disasters. The coastal area of South Bali is prone to land settlement hazards. In order to mitigate the risk, it is important to understand the depositional environment of the area related to its bearing capacity and geological hazard risks. The aim of this research is to understand the subsurface depositional environment and quantifying its bearing capacity. Quantitative modeling was carried out to obtain the sediment-bearing capacity of the Pendungan area in Bali, Indonesia. The methods used in this research were the observation of borehole cores, the identification of the cone penetration test (CPTu) curves pattern, the sediment index property test, the soil strength laboratory, and bearing capacity analysis. Based on lithologic association, the CPTu curve pattern, and grain size analysis, there are three facies developing in the study area with different bearing capacity values. Generally, beach ridge sand has a higher bearing capacity (N-SPT value of 8 – 52) for shallow foundation than fluvial clay. Meanwhile, floodplain facies has the lowest bearing capacity (N-SPT value of 2 – 20).


Introduction
The Southern Bali area includes Sanur, Serangan, and Pendungan, which are in a coastal zone, generally formed by the Holocene deposit of coastal plain. It's located in the Tanjung Benoa Basin and bordered by the Limestone South Formation in the southern part (Hadiwidjojo et al., 1998). This area is also included in a high earthquake vulnerability zone because it is located in ± 150 km south part of the active subduction zone. Shallow seismicity is centered in the plain of Bali caused by NW-SW or W-E fault activity (McCaffrey & Nabalek, 1987).
Area land capability and foundation design information are the most important factors which are required in area development. It is easily managed if accompanied by the foundation planning by adjusting the condition of the soil (Burgart et al., 2015). Furthermore, the aim of this study is to describe the subsurface condition in order to find the soil strength and whether or not it can resist load construction without the occurrence of shear failure and settlement state. Therefore, information on engineering geology is needed to support the plan. An engineering geological model such as vertical subsurface, engineering geology, and spatial condition can be the basic site or geological hazard impact study (Keche-bour, 2015; Rosye et al., 2009;URL, 2017). It can also be used for identifying engineering geology problems and estimating material properties realistically (Delgado et al., 2003). Moreover, sediment bearing capacity and facies condition information of the development area are required in order to produce optimal construction. Grain contact has a connection with bearing capacity because of physical characteristics which are contained by the facies. Sediment bearing capacity values that are obtained in every CPTu point are different due to geological factors influenced by the rock conditions, such as compactness, grain size distribution (as shown in Figure  1), grain contact, and water content (groundwater table). Furthermore, Eslami et al. (2011) andPariseau (2007) explain that lithology physical characteristics have different height, size, shape, and sorting of the grain so it influences the internal friction angle and shear strength between the grain. Also, Craig (2004) and Das (1997) added that the bearing capacity is controlled by the factors such as grain size, permeability, and consolidation. It is explained that the value of bearing capacity in each facies will be different based on the lithology. Generally, bearing capacity is only used to recognize the sediment strength for surface loading pressure without any settlement occurring (Das & Sivakugan, 2007). It is necessary to obtain further information related to the bearing capacity of shallow and deep foundation as a reference of physical development before the construction started.
Rudarsko-geološko-naftni zbornik i autori (The Mining-Geology-Petroleum Engineering Bulletin and the authors) ©, 2020, pp. 1-14, DOI: 10.17794/rgn.2020.4.1 (3.142 m) and Batur Mountain (1.717 m). Some other inactive volcanoes reach heights between 1,000 and 2,000 m. The central Bali mountain chain divides North Bali, which consists of narrow lower land from the foothill and mountain, from South Bali consisting of wide and plain lower land (see Figure 2).
Mainly analyzed from the elevation slope, Bali consists of a land of 0 -2% to 15 -40% elevation, the rest is land with an elevation of more than 40% (URL 1, 2018). The research area belongs to the southern physiography of Bali consisting of sedimentary and volcanic rocks. The characteristics of existing soils are more affected by volcanic weathering and alluvial deposits caused by an overflow of river transporting sediment. When flooding occurs, it will cause plated deposit structures in alluvial plains in low elevations.
Southern Bali coastal areas such as Sanur, Serangan, and Pendungan are areas formed by Holocene coastal plain deposits consisting of loose materials in the form of gravel, sand, silt and clay with a narrow to wide coastal plain (see Figure 3). The geology of this area is specified by dominant sandstone recurrence with fine to coarse grain size along with an intercalation of silt and clay. The depth of this Quaternary sediment reaches to a depth of around ± 20 m (Soebowo et al., 2010). The sediment of the coastal plain is a seismic zone path with high and active seismicity categorized into the 3 rd , 4 th , and 5 th seismic zones (URL 2, 2011; SNI 1726, 2012). Therefore, the Bali coastal area is prone to geological hazards.
The geological setting of the Quaternary Bali Basin consists of alluvium deposits and early sediment, proven by shallow groundwater levels. The distribution of earthquake epicenters spread out to the front and back part of the subduction zone, while other parts are concentrated in the south arc of the Java, Bali and Nusa Tenggara islands. Seismic activity located in the surrounding oceanic trough is an earthquake produced by  (Satriyo, 2017) Nevertheless, grain size and curve pattern analysis can show further information regarding the correlation between physical characteristics, facies development, and sediment bearing capacity. Based on research that has been conducted, the focus study on sediment bearing capacity can be related to the type of facies developed in a certain area.

Research location and geology
Most parts of Bali provinces are mountainous and hilly areas. Bali's island relief is west to east with elongated mountain chains. Among its mountainous areas, there are still active volcanoes, such as Agung Mountain intercontinental subduction. Shallow earthquake activity centered in the island of Bali is often caused by active faults which commonly shift from NW-SE or W-E (Mc-Caffrey & Nabalek, 1987).

Previous Research
The South Bali coast is a part of Indonesia which has high earthquake hazard vulnerability because this area is located ± 100 -150 km in the south part of an active subduction zone. In earthquake zonation maps, besides high earthquake hazard vulnerability, the South Bali coast is also located near active fault zones which cause shallow earthquakes, commonly in the SW-SE or W-E directions (URL 2, 2011).
Based on maps of Indonesian earthquake sources and hazards (URL 3, 2017), Bali is located in an area that has a peak bed acceleration of 0.4 -0.5 g (see Figure 4). This means that the south coastal area has a lower vulnerability to earthquake hazards. The earthquake history of this site has recorded high earthquake occurrences such as in 1862: MMI VII, in 1890: MMI VII, in 1917: MMI VII, in 1938: MMI VII, in 1961: MMI VII in 1976: MMI VIII, in 1979: MMI VII -VIII, in 1985: 6.2 SR, in 1987: 5.7 SR, in 2004.1 SR, 6.2 SR, 5.5 SR in southern Bali and the last time was on October 13, 2011 with 6.8 SR (URL 4, 2017).
A description of groundwater distribution in Pendungan -Sanur -Serangan, South Bali is generally reflected from river flow, habitat well and water drilling. Generally, the groundwater level is very shallow to shallow with a depth around 0.45 -3.5 m and in other areas more than 3.5 m (Seobowo et al., 2010). This condition is influenced by the lithology factor which underlies the narrow South Bali plain on the beach part which consists of alluvial deposits, marsh deposits, and beach ridges. Quaternary sediment deposits in this area with narrow groundwater levels can lead to the possible occurrence of liquefaction in several locations (Seobowo et al., 2010).

Methodology
Geological area development is a concept associated with bearing capacity and geological aspects such as lithology, physical characteristics, and facies. In this research, several methods are used to prove if there is a correlation between all the aspects, such as 1) literature review, 2) engineering geology investigation, 3) data processing, 4) modeling and integrating data. Geological engineering investigation including 2 bore holes and 11 CPTu points (Cone penetration test with pore water measurement) has been done in a research study (see Figure 3). The locations of two bore holes are near the Pendungan area, where the first (BH-1) is in a private company's residence close to Ngurah Rai By Pass Street (with a maximum depth of 25 m) and the other (BH-2) is in the Department of Transportation of Denpasar City (with a maximum depth of 26 m) (see Figure 3). The laboratory analysis was carried out by the grain size method approach using a sieve shaker with a sample taken from a depth of 2 m and 10 m for clay and a depth of 1 m, 13 m, and 22 m for sand. At these depths, the samples are considered to represent lithological changes in the sediment layer of the research area.
Furthermore, N-SPT data taken from a drilling point that correlated to a depth and a suppression energy was used to interpret the engineering subsurface condition (Roger, 2006); for cohesive soil, cohesion and consistency estimation was implemented (see Table 1) and for non-cohesive soil, internal friction angle and the level of density estimation were used (see Table 2).

Standard Penetration Test (SPT)
A SPT test is a test used to recognize in-situ sediment properties (SNI 4135, 2008). This test is done by hitting a standard tube to the bottom of the drill hole as deep as 45 cm by using a 63.5 kg hammer that falls freely from a height of 76 cm. The number of strokes required for penetration every 15 cm is recorded, but for the initial 15 cm, penetration is ignored because its sediment properties may be impaired at the time of drilling. Meyerhof (1956) and Roger (2006) stated that the amount of penetration in the last 30 cm was recorded as the N (N-value) which often correlated with sedimentary properties such as density, internal shear angle, and q c value (see Tables 1 and 2).
Conus pressure value (q c ) corrected to pore pressure (u), the result is a correction conus pressure value (q t ) as a: Where: qt -corrected conus resistance (MPa), u -pore pressure (MPa), a -ratio of conus area Stratigraphy interpretation method from CPTu data using the SBT index curve (Equation 1 and 2) from Lunne et al. (1997) that has a good correlation to soil classification based on grain size ASTM (2017) and Amorosi & Marchi (1999). The plotting process of conus resistance (q t ) to friction ratio is an expression of grain size, texture, and consistency, so it can be used to gain a subsurface stratigraphy profile (Lunne et al., 1997).

Cone Penetration Test with Pore Pressure (CPTu)
The CPTu test generates a subsurface stratification and engineering feature continuously. Furthermore, the primary application of the CPTu is to measure soil resistance, providing a classification of the soil texture with depth for geotechnical purposes (Begemann, 1965). The obtained soil response parameter is a profile of conus resistance (q c ), friction sleeve (f s ), pore pressure (u 2 ). The conus resistance value (q c ) indicates soil density and consistency, while the friction sleeve (f s ) is represented in the friction ratio (FR) parameter which is a transformation of grain size and texture (see Figure 5).
Engineering geological analysis was performed in a way to compile subsurface stratigraphy and engineering properties based on lithology, CPTu and N-SPT value interpretation, relative density parameters and laboratory results. Then, the correlation process is done between the subsurface section which describes the engineering properties and spatial interpolation, depth variation and layer thickness so a horizontal and spatial description of engineering properties can be obtained. The correlation and interpolation process also describes the bearing capacity zone of the land capability of cohesive and noncohesive soil so that it points out which areas have a soil capacity that is suitable for construction.
The N-Value is dominated by the amount of energy resulted by the SPT in the bore hole (Kovacs et al., 1981) and (Schertmann, 1978). Douglas & Olsen (1981) explain that the type of hammer will result in a significant q c /N ratio rather than soil density. Furthermore, a study conducted by Kovacs et al. (1981) has shown that the q c /N relies on the fine content of sandy soil. From the studies, major factors that can influence the q c /N ratio are permeability and compressibility modification of the sand. Sanglerat (1972) reviewed and presented a number of studies implemented in several countries. A variety of these studies have assigned a constant ratio of q c /N for each soil type; for instance, sandy soils have mostly q c /N ratios larger than 4, while clays have ratios less than 4 (Jarushi et al., 2015). Furthermore, these studies have drawn important conclusions that an increase in soil compactness and relative density would decrease the q c /N ratio (Jarushi et al., 2015).
The determination of the developed depositional environment in the study area used the comparison between gamma ray, borehole sample and CPTu. A borehole sample is used for cross-checking the grain size distribution and description. In its description, gammaray logs are one of the logs in the petroleum industry commonly used for lithologic interpretation, and the principle is the evaluation of the shale content (V-shale) or fine fraction and records grain size changes of a sediment through a certain curve pattern. CPTu is a commonly used geotechnical tool for determining the geotechnical sediment properties and describes the sediment stratigraphy. The CPTu principle is to calculate the end/ strength and adhesive resistance which will be used for lithologic interpretation through fine fraction content, in addition, CPTu also records large grain changes through certain curve patterns. Looking at the information, both tools can be compared to the interpretation of geological lithology and facies (see Figure 6) (Slatt et al., 1992). Several studies regarding the use of CPTu for bearing capacity in some purposes have been done. For example, predicting the bearing capacity of a bored pile (Iyad et al., 2015) and a piled foundation (Nicolay et al., 2017).
This method consists of pressing a pile to examine the penetration or shear resistance. The pile can either be a round pole or a closed round pipe with a conical end and/or a soil sampling tube, so it can estimate the physical properties of a layer and the locations with variations of resistance at the time of the erection of the tool (Lunne et al., 1997). This method serves for exploration and in-situ testing.
One way to calculate the shallow strength of the shallow foundation commonly used in Indonesia by the field testing method is the cone penetration test (CPTu). The shallow bearing capacity of the foundation is found by using simple conversion formulas. Below, some formulas of shallow ground capacity based on CPTu data will be shown. The measurement of allowed and ultimate bearing capacity (q all and q ult ) of cohesive and non-cohesive soil can be obtained from: Sandy soil (non-cohesive): • Sanglerat (1972) • Schmertmann (1978) Clayey soil (cohesive): • Schmertmann (1978) lane foundation (6) square foundation (7) Sandy or clayey soil: where: q all = allowed bearing capacity (q ult /3, score 3 is a safety factor) q ult = ultimate bearing capacity q c = average conus penetration arithmetic q c from base foundation to 1.5B under foundation P ult = conus tip resistance F ult = conus sleeve resistance B = foundation width (assumed of 1 meter) D f = foundation depth γ = effective soil density around foundation (1000 kg/m 3 ) N q and N γ = non-dimensional bearing capacity factor In the formulas above, the allowed (q all ) and ultimate (q ult ) bearing capacity score for non-cohesive soil conditions can be calculated using the parameters approach of conus resistance (q c ), foundation width (B), and foundation depth (D f ) (Equations 3, 4, and 5) (Meyerhof, 1956;Sanglerat, 1972;Schmertmann, 1978). However, for cohesive soil such as clayey and silty sand, Schmertmann (1978) explain that for ultimate bearing capacity (q ult ) should add the type of foundation, for example lane or square foundation (see Equations 6 and 7). Furthermore, Meyerhof (1956) added, the calculation of ultimate bearing capacity (q ult ) in a mixture of soil between sandy and clayey soil should add up several parameters such as conus tip resistance (P ult ), conus sleeve resistance (F ult ), and conus diameter (A(q c )) (see Equations 8 and 9).

Develop Facies
Qualitative interpretation was performed on three drilling cores and three CPTu curves residing in the research area (see Figure 7) to determine whether or not they had developed facies, which would then become the object of the study. The four CPTus' (CPTu 01, CPTu 03, CPTu 4, and CPTu 10) were paired with two boreholes (BH 01 -BH 02) to assist the facies analysis based on observing the CPTu curve pattern. The CPTu and bores were carried out to a depth of less than 30 m. From the observation of lithologic association with the observation of the CPTu curve pattern, there were three facies that developed in the research area. Several factors such as thickness, sediment type, and the physical properties of sediments also distinguish the three facies.

Facies of Flood Plains
Based on the lithology characteristics, the floodplain clay facies are dominantly composed of clay and silt. The clay is colored dark-gray-green, and is soft, with a mixture of shells, while the silt is gray in color with a slightly soft to moderate density, with a thickness of 1.5 -10 m, and both are located at a depth of 0.5 -10.5 m (BH-01). In general, these facies occupy 50% of the re-Rudarsko-geološko-naftni zbornik i autori (The Mining-Geology-Petroleum Engineering Bulletin and the authors) ©, 2020, pp. 1-14, DOI: 10.17794/rgn.2020.4.1 search area, dominantly in the north, east, and west of the study area (see Figure 8).
Based on the observation of the CPTu curve, the facies are characterized by a crescentic curve (Walker and James, 1992) with a value of 0.5 -5 MPa (see Figure 9). The floodplain clay facies have varying N-SPT, ranging from 2 -20 (very soft to slightly dense) as shown in Figure 9.

Clay and Sand Fluvial Facies
Based on its lithologic features, the facies are composed of clay and sand. The clay is dark gray to brown, of medium density, contains gravel, and is located at a depth of 0 -1.5 m (see Figure 10). Meanwhile, the sand is white-brown to gray white, medium to coarse sand, poorly sorted, rounded to sub rounded, very dense, does not contain shells of sea fauna, with a thickness of 1 -2 m, and is located at a depth of 10.5 -12 m (see Figure  11). In general, this facies occupies 10% of the study area (see Figure 12).
Based on the observation of the CPTu curve pattern, the facies is characterized by an uneven, symmetrical, and fining upward (FU) (Walker and James, 1992) with a q c value of 6 -17 MPa (see Figure 10 and Figure  11). Physically, clay and sand fluvial facies have varia-

Beach Ridge Sand Facies
Beach ridge sand facies are generally composed by the lithology of fine to coarse sand, feculent white, well sorted, sub angular to sub rounded with the degree of density from very loose to medium dense. This facies has a thickness of 2 m (see Figure 13) and it's located at a depth of 0 -2 m (BH-02). Beach ridge sand facies generally occupy 40% covering the center of the study area (see Figure 14).
The typical pattern of the CPTu curve that characterizes the beach ridge sand facies is coarsening upward (CU) (Walker and James, 1992) (see Figure 13). This facies has a range of q c values between 2 -13 MPa (see Figure 13). The type of developed sediment and N-SPT possessed by beach ridge sand is also more varied than fluvial sand facies.

Shallow and Deep Bearing Capacity Score of Research Location
Scoring analysis has been done using data regarding grain size, depth, sediment type, CPTu, and N-SPT. This analysis uses 1 meter length (L) and width (B) of foundation with a depth (D f ) of also 1 meter (SNI 1727, 2013). For the measurement of deep bearing capacity, 2 types of foundations were used (SNI 1727, 2013), which are: 1. a circle-shaped foundation with a diameter (D) of 1 meter, a cross-sectional area (Ap) of 0.785 m 2 , and 2. a square-shaped foundation with a length side of 1 meter. The results of calculating the bearing capacity score of shallow sediment foundations in the research location resulted in a fairly variable range of values (see Table   Figure (Satriyo, 2017) 3). The value is influenced by physical characteristics and lithology that developed in the study area.
From the values of bearing capacity of shallow foundation sediments obtained from CPTu, further determination of the minimum and maximum value of the total value of CPTu was achieved by using statistical analysis. After the value of shallow foundation bearing capacity is obtained for each CPTu, the next step is to group facies based on the CPTu curve pattern and the q c value. The aim is to discover which areas have the lowest to the highest bearing capacity and their relationship with the distribution of facies.  The analysis calculation result was obtained by grouping the value of each CPTu and subsequently classifying the soil classification based on the British Standard (BS) number 8004 (BS 8004, 2015) so that the type of land classification developed at each point on CPTu in the research area can be known (see Table 3). From these results, the classification of soils at each of these CPTu points indicates that the type of sediments developing in the study area is dominated by dense sand and soft clays, but does not indicate if the lithology is only clay, silt, or sand because a mixture of mud and gravel was found in the area, as shown in the physical description.
The value of sediment deep foundation bearing capacity in the research area is done by calculating N-SPT in log drill BH-01 and BH-02 (see Table 4) by using a circle pile with a diameter of 1 m and a square shape with a side length of 1 m (SNI 1727, 2013).

Floodplain Clay Facies, Physical Properties, and The Bearing Capacity Score
Floodplain clay facies are generally composed of clay and silt. Physically, this facies has an N-SPT ranging from 2 -20 (soft to slightly dense). Based on the calculation analysis of the bearing capacity, this facies has the highest shallow foundation bearing capacity score of    Table 4). This condition shows compatibility with the values generated from CPTu data (see Tables 5 and 6).

Fluvial Sand Facies, Physical Properties, and
the Bearing Capacity Score Based on its lithologic features, the fluvial sand facies are generally composed of medium sand to coarse sand  Table 4). This illustrates that this facies has a bearing capacity higher than the bearing capacity found in the floodplain facies. The density level of the facies of a floodplain is smaller than the density in fluvial facies  because it has a low N-SPT score. The mean score of q c and N-SPT of these facies are 11.5 and 42, thus showing that the physical and engineering characteristics possessed by this fluvial sand reflect its good sediment bearing capacity when used as the foundation at a depth of 10.5 -12.5 m, which is in correlation with the bearing capacity value found in CPTu data (see Tables 7 and 8).

Beach Ridge Sand Facies, Physical Properties, and
The Bearing Capacity Score Based on its lithologic features, the beach ridge sand facies are generally composed of medium to coarse sand. This facies has N-SPT values ranging from 8 -52 (loose to very hard). The highest shallow bearing capacity for this facies is 721.16 kN/m 2 and the deep bearing capacity scores of the foundation are 98.39 kN/m 2 (circle foundation) and 170.67 kN/m 2 (square foundation) (see Table 4). This state also has correlation with CPTu measurement, as seen in Tables 9 and 10.
There are 2 divisions of territory including shallow and deep bearing capacity. Generally, the area with good shallow bearing capacity is the beach ridge, especially at depths less than 4 m. However, areas that have good foundation carrying capacity values, particularly those with a recommended depth of more than 10 m, are areas that can be found throughout the entire research site, as seen in Figure 15.

Conclusions
Relying on data analysis from two boreholes and CPTu identification, there are 3 developed facies in the research area, which are floodplain clay facies, fluvial clay and sand facies, and beach ridge sand facies. Vertical facies changes can be identified from 2 existing boreholes, i.e.: • In BH-01, the fluvial clay facies are at a depth of 0 -0.5 m, from a depth of 0.5 -10.5 m of the facies transformed into a clay floodplain facies. • At BH-02, the fluvial sand facies are at a depth of 11.5 -12.5 m, from a depth of 12 -15.25 m the facies turn into facies of floodplain clay. Based upon physical characteristics and bearing capacity analysis, each facies has different values; the floodplain facies has the lowest value and the beach ridge facies has the highest shallow foundation bearing capacity value. Nevertheless, in cases of deep foundation bearing capacity, floodplain facies have the highest bearing capacity value since they have the deepest depth and the beach ridge has the lowest value because it has the shallowest surface. All this proves that bearing capacity of the facies is controlled by depth, natural compaction, grain size distribution, and the lithology of each sediment layer. Based on the value of bearing capacity, this research recommends the use of square foundations, which have a higher value than circle foundations. Moreover, this circular foundation has a wider base area in comparison to a square foundation in the same width of 1 meter, which is something to keep in mind when working in residential areas. Besides, in terms of geotechnics, a square foundation can support development because it can be used in the shallow layer with a maximum of 3 meters, thus, it can be more efficient in the use of development funds. However, if the construction is intended to build a two-story building or more, it is better to add to the depth of fluvial sand facies, which is at a depth of 11.5 meters.