Laboratory tests

The soil samples have been investigated by routine tests comprising classification, density, water content, liquid limit and undrained shear strength and sensitivity according to fall-cone tests. A large number of CRS oedometer tests have also been performed in order to determine the preconsolidation pressure and the permeability of the soil.

Supplementary determinations of the undrained shear strength have then been made by direct simple shear tests on specimen from selected levels at all sampling points. For two levels at Point 4, test series on specimens with different degrees of unloading have also been performed in order to study how an unloading affects the undrained shear strength according to this type of test.

A number of triaxial tests have also been performed in order to check the validity of the empirical relations for shear strength in clay for this particular soil and to determine the shear strength parameters in the upper sand and silt layer.

3.8 TEST RESULTS

3.8.1 Soil conditions – stratigraphy and variations over the area The results of the field and laboratory tests showed that the soil conditions within the investigated area at Strandbacken varied with depth and in certain respects also across the area, i.e. perpendicular to the river. The upper soil layers below the original ground surface have consisted in a thin layer of topsoil overlying sand and silt. At Point 6, about 25 metres behind the upper crest, these upper layers reach only about 2.5 metres below the ground surface. At the upper slope, this thickness has

increased to at least 4 metres and the entire upper slope consists of this type of soil.

The thickness has increased further at the centre of the excavated terrace where the upper 3 metres of the remaining soil is classified as silt. This means that the sand and silt layer here had a thickness of about 7 metres before the excavation. The disturbed samples taken at the crest of the lower slope indicate similar conditions at this point. However, it is not a straightforward task to determine the exact thickness of these layers since the transitions from sand to silt and silty clay with a decreasing silt content are gradual and thin layers of coarser material are found embedded in the clay, particularly in the uppermost part. The soil properties are also affected by the organic content, which is found in the form of embedded plant remains, thin layers of plant remains and more evenly distributed finer particles.

The upper soil layers are thereby designated as organic. The organic content is significant within a relatively large depth interval, and infusions of visible plant remains have been found down to 12 metres' depth below the original ground surface. The organic content varies between 2.2 and 3 % within the whole layer of clayey silt and silty clay down to this depth. The upper soil layers also contain various amounts of shells.

At about the same level, about 12 metres below the original ground surface, the character of the soil changes according to both field and laboratory tests. The generation of tip resistance and pore pressure in the CPT tests and the evaluated material index in the dilatometer test indicate that more homogeneous clay is encountered. According to the laboratory tests, it is a high-plastic sulphide spotted clay with some content of shells, which are absent in certain depth intervals and at certain levels form thin layers or lenses of shells. At greater depth, about 30 metres below the original ground surface, the soil becomes coarser and lenses and thin layers of coarse soil start to appear. At the level of about –27 metres, i.e. 37 metres below the original ground surface, an apparently continuous layer of coarser soil is found. At Points 1 and 2, this layer is found at a level of about –22 metres. The tests have penetrated further in all points and samples have also been taken at deeper levels. At those points where the sampling operations have caught the embedded layer, it has been found to consist of a number of alternating thin layers of fine sand and plant remnants.

Below this layer, the clay continues for a number of metres before stop in the soundings is obtained in sand or coarser soil. This clay at the bottom of the profile is coarser and contains single small channels, which have been stated to originate from small organisms that once lived in the bottom mud in the sea. In the deepest clay layers, there is a transition from grey, sulphide spotted clay to brown-grey

varved clay with an increasing content of thin silt and sand layers. The results of the CPT tests indicate that the thickness of the clay layers is smaller below the river than in the area behind the riverbank. This could have been a result of the limited penetration force from the unanchored drill rig on the raft, but it is verified by results from earlier experience, where the stop levels for primarily the "machine sounding tests" showed the same bottom configuration. The established borders for different layers and clay types in the profiles also support these conditions, Fig. 93.

Certain anomalies with embedded layers of coarser soil were found in the upper metres of the profile at Point 2. This is assumed to be a result of the previously continuous process of erosion and superficial slides where the superficial soil layers at the toe of the slope consists of slide debris originating from higher levels in the soil profile.

The liquid limit is a measure of the composition and character of the soil, and a compilation of the measured values shows that the soil is very similar throughout the area, Fig. 94, apart from the fact that the thickness of the clay layer decreases below the riverbank and the river and the thickness of the sand and silt layer decreases between Points 4 and 6, i.e. with distance away from the river. The measured values correspond well to those measured in earlier investigations in the area, see Fig. 70.

A corresponding compilation of the measured natural water content in the test points shows a very good correlation between the results at Points 1–4 when the varying thickness of the layers is considered, Fig. 95. However, a considerable difference was obtained at Point 6, where the water contents are considerably higher than in the other points, particularly in the upper soil layers. This is also reflected in the measured bulk densities, which are lower at Point 6, Fig. 96.

The relation between the water content and the liquid limit in a soil is a measure of its state of consolidation and also affects properties such as remoulded shear strength and sensitivity. The quasi liquidity index that is represented by the quotient w_N/w_L shows clearly that the upper part of the soil profile at Point 6 differs from the soil at the other points in this respect, Fig. 97.

That the undrained shear strength was lower at Point 6 had been shown already in the CPT tests. A closer study of the measured sensitivities in the laboratory showed that these were significantly higher in the upper part of the soil profile at Point 6, Fig. 98.

Fig. 93. Variation in stratigraphy at Strandbacken with distance from the river.

(The different layers are described in more detail in the text. Different colours have been used in the upper layers of topsoil, sand and silt to illustrate the extent of the excavation.)

-45

Fig. 94. Compilation of measured liquid limits in the new test points at Strandbacken.

Fig. 95. Compilation of measured water contents at the new test points at Strandbacken.

Fig. 96. Compilation of measured bulk densities at the new test points at Strandbacken.

Fig. 97. Quasi liquidity index at the new test points at Strandbacken.

-40

Higher quasi liquidity indices and sensitivities were also measured in the bottom layers at Point 1, but this can be assumed to be related to the fact that these determinations have been made in the coarser varved and layered soil just above the underlying coarse bottom layers.

The relation between quasi liquidity index and sensitivity in principle follows the relation that was found in the Göta-älv investigation, Fig. 99. The sensitivity is thereby high in the upper soil layers at Point 6. The maximum quasi liquidity index was 1.1 and there was no quick clay within the investigated area. However, the previous investigations have shown that the clay becomes quick at further distance from the river.

-45 -40 -35 -30 -25 -20 -15 -10 -5 0 5 10

0 5 10 15 20 25 30 35 40

Sensitivity

Level, m

Point 1 Point 2 Point 4 Point 6

Fig. 98. Compilation of measured sensitivities at the new test points at Strandbacken.