Chapter 4
Site Investigation

4.1 SITE INVESTIGATION AT BORROW AREAS
4.2 SITE INVESTIGATION PRIOR TO RECLAMATION AND SOIL IMPROVEMENT
4.3 SOIL INVESTIGATION DURING RECLAMATION
4.4 POST-IMPROVEMENT SOIL INVESTIGATION

Site investigations are usually carried out at both the reclamation area and the borrow areas. Investigation works should be carried out for every stage of the projects, such as prior to reclamation, after general filling, during soil improvement and after soil improvement. The purposes of site investigation vary depending upon the type and time of investigation. Some are carried out to explore and quantify the volume of borrow materials whereas others may be done just to profile the underlying soil or to characterize the geotechnical properties of underlying formations. Some are done to assess how the soil may be improved whereas others are carried out for quality control on soil improvement works. This section describes the site investigation works carried out at various stages of reclamation and soil improvement work. Details of site investigation practice in land reclamation projects are also given in Bo and Choa (2000).

4.1 SITE INVESTIGATION AT BORROW AREAS

The purpose of site investigation at borrow areas is to assess the quality of and quantify the available fill material. A preliminary check on the quantity of sand deposit can be made after running a geophysical survey. A seismic survey can indicate the thickness and extent of the sand mine. Figure 4.1 shows a geophysical survey in progress, and Figure 4.2 shows a typical seismic reflection survey which indicates the thickness of the sand deposit. The quality of sand can be verified either by an in-situ test or test results from a sample taken. Vibrocoring is one of the best methods. Alternatively, a sample can be grabbed if the thickness of the sand layer is less than 5 meters. Figure 4.3 shows vibrocore sampling in progress. Samples obtained are tested in a geotechnical laboratory to find out the type of sand and grain size distribution. Figure 3.1 shows the grain size distribution of sand from two borrow sources. Generally, sand with less than 10% of fine is considered good for land reclamation. Sand deposits with fine content of greater than

10% and up to 20 to 25% are still usable since a certain amount of fine can be flushed out during the dredging, loading, and unloading processes.

Another useful method of sourcing for sand deposits is the Cone Penetration Test (CPT). Since CPT can classify the type of soil, both quantification and quality verification of the sand mine is possible with this equipment.

4.2 SITE INVESTIGATION PRIOR TO RECLAMATION AND SOIL IMPROVEMENT

4.2.1 Seismic reflection survey

Site investigations are carried out prior to reclamation in order to profile the underlying geological formation and also to characterize the geotechnical properties of the underlying soils and rocks. The first site investigation to be carried out prior to reclamation is the seismic reflection survey. By interpreting the seismic reflection survey results the thickness of the soft layer can be identified. Figure 4.4 shows the isoline of the thickness of the soft clay layer interpreted from the seismic reflection survey in the Changi reclamation project.

4.2.2 Boring and sampling

Offshore site investigations using boreholes are carried out to profile the formation of the soil and also to collect samples for laboratory tests for geotechnical characterization. Boreholes are usually drilled using the rotary mud flush drilling technique. Figure 4.5 shows rotary drilling which is commonly used in site investigation. Site investigations at a foreshore location are usually carried out from a jack-up pontoon (Figure 4.6). Site investigations at deep sea locations are usually carried out from a vessel on which a boring rig is mounted. An on-board laboratory is generally built on the ship (Figure 4.7).

In most site investigations continuous undisturbed sampling with either a piston sampler or a thin wall sampler is carried out. Samples of soft clay are taken with either a 75mm or 100mm diameter piston sampler while

firm clay samples are taken with a 100mm diameter Shelby tube thin wall sampler. Samples of stiff to hard clay are taken with a thick wall drive sampler. Figure 4.8 shows a piston sampler. The collected sample tubes are sealed straight away with a mixture of wax and velcelin on site and sent to the on-site laboratory with great care during transportation. If the clay is soft or firm, field vane shear tests are carried out adjacent to the borehole to measure the in-situ undrained shear strength of the clay.

4.2.3 Field vane shear test (FVT)

Field vane tests (FVT) are carried out with a Geonor Vane with two types of vane blades, measuring 55mm x 110mm and 65mm x 130mm, depending upon the types of clay encountered. A certain waiting time—generally 5 minutes after insertion of blade—and a suitable rotation rate of 12 revolutions per minute are used, as advised by Chandler (1988). A field vane test basically measures the torque required to rotate the blade until failure. Field vane equipment needs to be calibrated from time to time. Field vane tests are usually terminated when the field vane shear strength reaches 90 – 100 kPa. Remolded strength tests are also carried out after undisturbed tests. Figure 4.9 shows field vane blades and field vane equipment. Figure 4.10 shows typical calibration results of field vane equipment, and Figure 4.11 shows typical field vane shear strength results. Standard penetration tests are carried out either in the sand formation or alluvial cemented sand rock, usually by collecting disturbed samples. Boreholes are terminated after three consecutive blows of SPT 50/300 mm have been obtained. Figure 4.12 shows SPT testing in progress.

4.2.4 Cone penetration test (CPT)

Cone penetration tests (CPTs) are becoming more and more popular because of their simplicity and immediate results. There are several types of cones with different functions and capacities manufactured by various manufacturers. Table 4.1 shows the types of cones with various capacities available in the market, produced by Geomail. Figure 4.13 shows the geometry and design of the Gouda cone. The basic testing procedure involves a continuous penetration of the cone into the sub-soils, at a standard penetration rate of 20mm per second, and recording the cone resistance (q_c), the sleeve friction (f_s), and the penetration pore pressure. CPT can provide several measurements, such as cone resistance (q_c), friction (f_s), pore pressure (u_bt) and inclination. CPT can classify the types of soil by applying the Robertson

and Campanella (1983) chart (Figure 4.14). Figure 4.15 shows a comparison of the soil profile interpreted from CPT and observed from the borehole. It can be seen that CPT can accurately classify the types of soil. Some rigs are mounted on trolleys whereas others are mounted on a crawler or truck, as shown in Figure 4.16.

There are some CPTs which can carry out tests in foreshore conditions. Such CPTs are usually done on the seabed and controlled from the vessel or barges by remote control (Figure 4.17).

The interpretation of undrained shear strength from in-situ tests has been comprehensively discussed in Bo et al. (2000a), and the use of CPT in land reclamation project has been described by Bo and Choa (2001).

4.2.4.1—Undrained shear strength from CPT

The undrained shear strength (S_u) can be determined from the cone resistance using the following equation:

where δ_V is overburden stress, N_k is the cone factor.

N_k is reported to be between 11 to 19, based on a correction of the field shear strength (Lunne and Kleven, 1981), and 17, based on a triaxial compression test on non-fissured over-consolidated clay. Kjekstad et al. (1978) and Battaglio et al. (1986) have reported N_c = 14 for soft homogenous highly structured CaCO₃ Cemented Fucino Lacustrine Clay, based on a field vane and triaxial test.

where q_t is the corrected cone resistance, and it can be calculated from the cone resistance (q _c) using the following equation:

where a is an unequal area ratio and u_bt is the pore pressure at the cone base.

N_kt values are reported to be 10 – 15 for normally consolidated clay, and 15 – 20 for over-consolidated clay (De Ruiter 1982). Dobie (1988) has reported N_k values of between 15 and 21 for on-land Singapore marine clay. La Rochelle et al. (1988), Rad and Lunne (1988), and Powell and Quarterman (1988) has reported N_kt values of 8 – 29, depending upon I_p based on the triaxal compression test. Aas et al. (1986) proposed a

relationship between N_kt and the plasticity index (I_p) of clay as follows:

Bo et al. (2000a) has reported the relationship between N_kt and Singapore marine clay as follows:

By using the above correlation, the undrained shear strength of clay can be estimated from q_c values. A comparison of the estimated and measured field vane shear strength is shown in Figure 4.18 for a Singapore marine clay.

Several others have reported various N_kt values for different types of clay from all over the world either based on triaxial tests or field vane test data. Those based on triaxial tests are N_kt of 13.7 for Newcastle clay in Australia (Jones 1995), N_kt of 13.5 to 15.5 for Sarapui soft clay in Brazil (Rocha-Filho and Alencar 1985), and 10.3 to 15 for Recife soft clay in Brazil (Coutinho et al. 1993). N_kt ranging from 12 to 20 have been reported for normally consolidated clay in southern Nigeria by George and Ajayi (1995).

N_kt values based on field vane tests are N_kt = 14.5 for Jacarepaqua clay in Brazil (Rocha-Filho, 1987), N_kt = 15 for Porto Alegre soft clay (Soares et al. 1986), and Quilombo soft clay (Arabe 1995), N_kt = 10 for different deposits of clay in Denmark (Denver 1988; Kammer Mortensen et al. 1991; Jorgensen and Denver 1992), and N_kt = 9 – 14 for Japanese

marine clay (Tanaka 1994). Tanaka also reported N_kt values of between 8 and 16 based on laboratory Unconfined Compression Test.

In Germany, the deduction of overburden pressure is not taken into account and the cone factor is also not used. The direct relationship between cone resistance (q_c) and undrained shear strength is proposed as:

where N varies between 10 to 20.

N = 12 for soft clay and 20 for OC clay was reported by EAU (1990). Sanglerat (1972) has reported N = 10 for q_c 0.5 MPa, and N = 18 for q_c< 0.5 MPa. A similar direct correlation was also used in Vietnam, which reported values of 20 for soft silty clay (Nhuan et al. 1985).

4.2.4.2—Over-consolidation ratio from CPT

In addition to estimating undrained shear strength, CPT can be used to predict the over-consolidation ratio (OCR) of soft clay. OCR can be estimated from q_t using the following equation:

where is effective overburden stress, and αis constant, ranging from 0.2 to 0.5. A value of 0.33 was reported based on the CPT pore pressure measured on the shoulder of a cone (Kulhawy and Mayne 1990), and a = 0.81 based on mid-face element (Chen and Mayne 1996). Sonneset et al. (1982), and Konrad and Law (1987) reported αvalues of 0.49 based on pore pressure measurement on the shoulder. For Singapore marine clay, Bo et al. (1998a) has proposed a k value of 0.32.

Figure 4.19 shows a comparison of OCR, interpreted from various in-situ tests with that interpreted from laboratory oedometer tests.

4.2.4.3—Coefficient of consolidation due to horizontal flow (C_h) from CPT

Since CPT equipment has a pore pressure transducer, it is also possible to carry out the pore pressure dissipation test in the clay. When CPT penetrates into the clay, dynamic pore pressure occurs. However, if the cone is held at the same position for a longer duration, the dynamic pore pressure will dissipate with time. This pore pressure dissipation curve can be analyzed by applying Baligh and Levadoux's (1980) strain path method. The coefficient of consolidation due to horizontal flow C_h values can be calculated from a relevant time factor T using the following equation:

where R is the radius of the pushing cone in meters, T is a dimensionless time factor, and t is the time lapse needed to reach a given degree of consolidation in years.

The resultant C_h values need to be corrected to a normally consolidated (NC) condition using recompression ratio. Figure 4.20 shows some pore pressure dissipation curves measured by a CPTU test, and Figure 4.21 shows a comparison of C_h values measured by various types of laboratory and field in-situ tests.

Apart from the above types of simple in-situ tests, there are several specialist in-situ tests which are carried out to characterize the soft marine clay prior to reclamation.

4.2.5 Flat dilatometer test (DMT)

A Marchetti flat dilatometer blade (Marchetti, 1980) with a steel membrane on one side (Figure 4.22) is used in a flat dilatometer test. The test involves driving the flat dilatometer into the seabed with a 20 ton static rig at a standard penetration rate of 20 mm per second. When the driving is temporarily stopped at selected depth intervals, two pressure readings, corresponding to two prefixed states of expansion of the membrane, are recorded. The first reading corresponds to the membrane lift-off pressure and the second reading records the pressure required for the center of the membrane to deflect by a preset distance of 1mm into the soil. These readings are called P₀ and P₁, respectively, after allowing for the effects of the membrane stiffness. The testing procedure follows the instructions of the dilatometer operation manual prepared by Marchetti and Crapps (1981). Figure 4.23 shows the photographic features of a flat dilatometer blade, and a flat dilatometer test in progress.

The flat dilatometer measures two pressure values, called P₀ and P₁. For these two values, three indices, such as the material index (I_D), the

horizontal stress index (K_D), and the dilatometer modulus (E_D) can be obtained using the following equations:

where u₀ is the pre-insertion pore water pressure.

Marchetti (1980) has proposed the classification of soil using material index values. Figure 4.24 shows measured and calculated indices from DMT tests for a test area. It was found that a dilatometer could classify the soil type closely. Like CPT, s_u can also be estimated from K_D values obtained from a DMT test. Marchetti (1980) has proposed the undrained shear strength s_u with lateral stress index (K_D) as follows:

Bo et al. (2000a) has proposed a power function of 1.0 for upper and intermediate Singapore marine clay, and 0.7 for lower Singapore marine clay, instead of 1.25. Figure 4.18 shows the s_u values estimated from the DMT.

From the lateral stress index K_D, the OCR of clay can be estimated, as proposed by Marchetti (1980), as follows:

Bo et al. (1998a) proposed the power function 1.0 for lower and upper Singapore marine clay and 0.8 for intermediate Singapore marine clay instead of 1.56. Figure 4.20 also shows the OCR estimated from a DMT test.

A DMTA test measures the total stress of the soil and from the dissipation of the total lateral stress C_h values can again be calculated using an equation proposed by Marchetti and Totani (1989) for A reading dissipation tests.

where T_flex is the dimensionless time factor. Figure 4.25a shows the dissipation curve from DMT tests in marine clay. From the C reading, dissipation test C_h is given by:

4.2.6 Self-boring pressuremeter test (SBPT)

A Cambridge-type self-boring pressuremeter (Worth 1984) with strain-measuring arms located at the mid-level, as shown in Figure 4.26, is one of the most useful equipment for in-situ tests to characterize soft clay. The instrument has strain-gauge type transducers attached to the center core or pressuremeter body, which is covered with a rubber membrane, for direct recording of the radial displacement and the applied pressure.

A self-boring pressuremeter is equipped with a rotary bit at the base. The SBPT involves firstly insertion of the pressuremeter to the selected depth in the ground using a self-boring technique. Following the insertion of the apparatus, a rubber membrane is inflated by the injection of gas pressure and both the applied pressure and the corresponding displacement of borehole (cavity) wall are measured. The test procedure generally follows Mair and Wood (1987) and Hawkins et al. (1990). The test results are usually presented in a plot of applied pressure versus (radial) cavity strain, which can be interpreted by the cavity expansion theory. Figure 4.27 shows typical results from a self-boring pressuremeter test. Figure 4.26 shows the geometry and dimension of the self-boring pressuremeter, while Figure 4.28 shows a self-boring pressuremeter test in progress.

Tests can be carried out either on stress control or strain control. From the test, the following basic measurements can be obtained:

i). Lift-off pressure

ii). Stress vs strain curve

iii). Several unload, reload loops

iv). Limit pressure (PL)

v). Pore pressure dissipation curve from a dissipation test.

From the lift-off pressure, lateral earth pressure can be obtained. From the stress-strain curves, various types of modules, such as initial tangent modules, secant modules, and unload-reload modules can be obtained. Undrained shear strength can also be estimated from limit pressure using the following equation:

Marsland and Randolf (1977) adopted an N_p ranging between 5.5 and 6.8. It can also be suggested that the N_p values for a specific type of clay should be locally obtained by empirical correlation. Bo et al. (2000a) has suggested that N_p values for Singapore marine clay at Changi are 6.0, 6.4, and 7.2 for upper, intermediate, and lower marine clay. Figure 4.18 also shows a comparison between field vane shear strength and that interpreted from SBPT.

Since a self-boring pressuremeter can measure the total horizontal

stress, it is possible to determine the K_o values, and hence the OCR can be estimated. Figure 4.19 shows the OCR obtained from a self-boring pressuremeter compared with laboratory results.

From the pore pressure dissipation test, the coefficient of consolidation due to horizontal flow (C_h) can be estimated using the following equation:

where t₅₀ is time taken for the excess pore pressure to fall half of its maximum value, T₅₀ is the time factor, and ρ is the radius of cavity.

Figure 4.21 shows C_h values measured from various in-situ tests. k_h can be calculated from C_h values. k_h interpreted from various in-situ tests are shown in Figure 4.29. An interpretation of k_h from various in-situ tests can be found in Bo et al. (1998e).

4.3 SOIL INVESTIGATION DURING RECLAMATION

Reclamation is usually carried out by filling in stages. Especially when soil improvement works are involved, reclamation is done in two stages. In the first stage, filling is usually done to a level slightly above the high tide and prefabricated vertical drains are installed at that level before raising the fill level to the required surcharge level. At such site investigation, boreholes can be drilled where profiling and characterization of boreholes have not yet been done. Site investigations are carried out soon after the filling to just above the high tide level. No prefabricated vertical drain has been installed at this stage since the change or improvement of the soil parameters is minimal.

Another useful site investigation equipment at this stage is the CPT. By using CPT, profiling and contouring of the soft clay layer are possible, and the exact penetration depth required for a prefabricated vertical drain can be determined. Figure 4.30 shows the profile and contour of a soft clay layer determined from the CPT carried out after filling to a level above the high tide level. In addition to the profiling of soil, CPT can indicate the quality of sand as well as any mud traps or mud waves that occur during the first stage of filling. Figure 4.31 shows mud traps detected by CPT equipment, and Figure 4.32 shows mud wave detected by CPT equipment.

Generally, the payment for reclamation is made based on a pre- and post-survey of the area. However, during the general filling stage, some changes can occur. One is the settlement of the seabed, another is the heaving up of the seabed due to mud-waves, and still another is a mud trap within the sand fill. Thus, the sand volume measured by a pre- and post-survey may not be the actual volume deposited. The actual volume can be measured by carrying out a CPT with certain grid patterns. A deduction of the volume of mud trapped inside the sand fill can also be made. For this exercise, it is important to achieve the verticality of the CPT test carried out.

An alternative in-situ testing equipment which can detect changes on the seabed, mud wave, and mud trap, is the Auto-ram sounding equipment. Figure 4.33 shows typical Auto-ram sounding results which indicate the mud trap.

4.4 POST-IMPROVEMENT SOIL INVESTIGATION

In the post-improvement stage almost all the site investigation equipment used prior to reclamation can be utilized. However, there are some differences in the usage and careful interpretation is required. Borehole and sampling are carried out from the surcharge level. Generally, only wash

boring is carried out within the sand fill area since soil parameters for filling sand are normally not required in the post-improvement site investigation stage. When drilling reaches the clay layer, it is important to maintain the drilling fluid inside the borehole. Without this, disturbance to the clay could be encountered since formation clay has a high level of surcharge. Post-improvement site investigations are carried out in order to assess the improvement of the soil. This is done when geotechnical instruments

monitoring results indicate that the improvements are close to the required degree. All the different types of in-situ tests done prior to reclamation and to prefabricated vertical drain installation are repeated. However, special attention needs to be given to some of the special tests and this will be discussed later. The types of site investigations carried out in the post-improvement stage include boring, sampling, field vane shear test, CPT and CPTU test, DMT, SBPT, BAT, CPMT, and seismic cone tests.

Site investigation borehole, sampling, and field vane tests are usually carried out at the same locations as the tests prior to reclamation in order to be able to compare the pre- and post-improvement geotechnical parameters. However, this may not be strictly necessary just to assess whether the required degree of consolidation has been achieved. The tests can be done at any location and then compared with the required degree of consolidation, specified strength and effective stress stated in the technical specifications. Figure 4.34 shows a comparison of pre- and post-improvement borehole and field vane tests together with geotechnical parameters obtained from laboratory results.

4.4.1 Cone penetration tests (CPT and CPTU)

Cone penetration tests are carried out in the same manner as those prior to reclamation. As shown in Figure 4.35(a), a significant increase in cone resistance can be seen. As explained in Section 2, the undrained shear strength can be estimated from the CPT test. Figure 4.35(b) shows a comparison of undrained shear strength measured by CPT prior to and after improvement. However, OCR cannot be estimated from the soil when consolidation is in progress unless the effective stress is known. If the effective stress is known, there is no reason to estimate OCR from the CPT because it can be calculated directly. Therefore, the pore pressure method is usually applied to estimate the OCR from CPT tests.

Dissipation tests can also be carried out in the same way as those prior to reclamation. However, pore pressure should be normalized with equilibrium pore pressure obtained from CPTU measurement, rather than using static pore pressure.

where u_i is the initial pore pressure, u_t is the pore pressure at time t, and u_e is the equilibrium pore pressure measured from CPTU test. Figure 4.36 shows a comparison of C_h values measured before and after improvement.

An additional test can be carried out to check the improvement of soil, and that is a long-term holding test. If the CPT cone is held at a certain elevation for a long time, the pore pressure will dissipate to equilibrium. This equilibrium pore pressure will be the same as the pore pressure in the

soil at the time of measurement, and will also be the same as the pore pressure measured with the piezometer. In this case, the degree of consolidation and effective stress can be estimated from a CPT long-term holding test. Figure 4.37 shows a comparison of the equilibrium pore pressure measured by a CPT long-term holding test with that measured by a piezometer. It can be seen that the CPT long-term holding test measures the equilibrium pore pressure quite accurately.

4.4.2 Dilatometer test (DMT)

The dilatometer test can be repeated for a post-improvement in-situ testing. However, the interpretation of most parameters from a dilatometer test requires knowledge of the effective stress. As such, assessing improvement independently without knowing the effective stress is not possible with a dilatometer test. Only an increase in modulus can be detected with a dilatometer since E_D does not include pore pressure and effective stress parameters. A dilatometer dissipation test can also be used to estimate the C_h values and k_h values after improvement. Figure 4.36 also shows a comparison of C_h values prior to and post-improvement determined from DMT tests.

4.4.3 Self-boring pressuremeter test (SBPT)

A self-boring pressuremeter test can be carried out in the same way as that prior to reclamation, and undrained shear strength and OCR can be estimated using Equation 4.16. A pore pressure dissipation test can also be carried out in the same way as described earlier. However, the equilibrium pore pressure should be used to normalize to obtain the degree of dissipation as in the post-improvement CPTU test.

Figure 4.38 shows a comparison of the parameters measured by the SBPT prior to reclamation and after improvement. A dissipation test can also be carried out after improvement to interpret the C_h and k_h values. Bo et al. (1997b) have compared C_h and k_h values from pre- and post-improvement tests using the in-situ dissipation test data.

4.4.4 Cone pressuremeter test (CPMT)

A cone pressuremeter is a combination of a cone penetrometer and a pressuremeter. Therefore, it can measure the same parameters as the CPT and pressuremeter. However, the CPT cone is usually bigger than the conventional cone, with a cone base area of 15 cm². The CPT test is carried out in the same manner as the standard CPT and measured by the same parameters, and hence can obtain the same soil parameters as the standard

CPT tests. The pressuremeter is attached above the cone and its diameter is 43.7 mm and 2 meters in length.

The test can be carried out in the same manner as the SBPT test and hence the same sets of geotechnical parameters can be obtained. Figure 4.39 shows the geometry and dimension of a cone pressuremeter. The advantage of the CPMT test is that pre-boring or self-boring is not required. However, soil disturbance in the clay or contraction in the granular soil can occur due to the penetration. This type of CPMT test is suitable for granular soil where maintaining a regular size borehole is difficult. Figure 4.40 shows some geotechnical parameters measured by a CPMT test and pre- and post-modulus cone resistance from compaction quality control.

4.4. Seismic cone test

A seismic cone is a combination of a CPT and seismic geophone receiver. It can be used to carry out a conventional CPT test and to collect and interpret similar sets of geotechnical parameters. Figure 4.41 shows the geometry and dimension of a seismic cone. At a certain interval, the penetration of the cone can be stopped to take a seismic measurement. Normally, seismic force is provided by applying a hammer to the wooden plate or a certain static load, and the seismic wave is detected by a receiver near the cone tip. Figure 4.42 shows seismic cone testing in progress. From the data, compression wave (ρ) and shear wave (S) can be calculated. In turn, compression and shear wave velocity (v_p) & (v_s) can be obtained. Figure 4.43 shows shear velocity, shear modulus, and cone resistance measured by a seismic cone test. Small strain shear modulus (υ₀) and constrained

modulus (M) can be estimated from v_s and v_p by using the following formulae:

4.4.6 Use of CPT for compaction quality control

Densification requirement is usually specified with a certain cone resistance value which relates to relative density. Pre- and post-CPT tests are carried out to assess the achievement of densification. Figure 4.44 shows a comparison of pre- and post-CPT at a deep compaction area.

4.4.7 Auto-ram sounding

Swedish ram sounding is dynamic probing with a solid cone. There are several types of ram sounding such as light, medium and heavy duty. The hammers used for various categories are shown in Table 4.2. The drop height is usually 50 cm, and the number of blows is counted for every 20 cm penetration. Ram sounding can detect the density of granular fill, trapped mud, and an interface between granular soil and clay. Figure 4.45 shows the geometry and dimension of auto-ram sounding equipment, while Figure 4.46 shows auto-ram sounding in progress. A comparison of pre- and post-compaction ram sounding results is displayed in Figure 4.47.