Introduction

Before planning of the new settlement areas, detailed geological, geophysical and geotechnical studies should be performed, focusing on seismicity. The study area has been seriously affected by the devastating Gerede (01.02.1944), Marmara (17.09.1999) and Duzce (12.11.1999) earthquakes. Bolu plain is a pull-apart type of basin (Gokten et al. 1998). Liquefaction mostly occurs in the saturated and non-cohesive sandy soils with shear waves, affecting the structures beneath and above the ground level. Geological, geophysical and geotechnical properties of the soil groups in Bolu settlement area have been investigated and land use map was prepared (Varol et al. 2000). It was indicated by (Aydan et al. 2000) that, no liquefaction risk exists in the study area, since the effects could not reach the surface due to the thick fine-grained soil cover. Kilic et al. (2001) studied geotechnical properties of the geological units in Bolu settlement area and noted that no liquefaction was observed after the Marmara and Düzce earthquakes. Recent advance has been encountered in microzonation studies after these earthquakes focusing in liquefaction (Topal et al. 2003; Yilmaz and Yavuzer 2005).

In this paper, index and physical properties of soils in Bolu are presented (Fig. 1). Besides groundwater level, P and S wave velocities (V p, V s) were measured. Safety factor for the liquefaction potential of soils in the study area was determined.

Fig. 1
figure 1

Location map of the study area

Full size image

Geological setting

Quaternary aged alluvium (Qal), alluvium fan (Qaly) and alluvium terraces (T-Qtr1, Qtr-2) of the Neotectonic period (Pliocene-recent) exist in and around Bolu settlement. The terrace type deposits with uncemented gravel and sand, have been transported by the drainage system, flowing through the lower elevations. The plain dips gently towards South. The clayey and silty flood plain sediments and the gravely, sandy material of the Buyuksu creek are mostly in transition within each other. Quaternary alluvium overlies Pliocene and older units unconformably (Fig. 2). Maximum thickness of the Quaternary aged alluvium is 230 m in the Bolu plain (DSI 1962). In the South of the study area, Büyüksu creek (Fig. 3) flows from East to West. It was observed that groundwater level changed between 0.45 and 10.50 m during March–April 2001.

Fig. 2
figure 2

Engineering geological map of the study area

Full size image
Fig. 3
figure 3

A view from Büyüksu creek

Full size image

Tectonics and seismicity

Bolu plain is a pull-apart type basin, that is, controlled by the lateral strike fault system in the North Anatolian Fault Zone—NAFZ (Gokten et al. 1998). Paleo and Neo-tectonic period active faults exist at the North and South of the plain. There exist several faults, parallel and oblique to this major fault. Surface rupture of the 12 November 1999 earthquake aligns through the South of Gölyaka towards Kaynaşlı on the East, ending in the Asarsuyu valley and Bolu tunnel. Bolu centrum is settled in the middle of the plain on a pressure ridge type hill, that is, probably tectonically controlled. Major earthquake records are given in Table 1. Historical earthquakes have been recorded on the Abant-Bayramören segment at South. There are 12 earthquakes between 967 and 1890. The earthquake of 17.08.1668 (M s  = 8) caused a disaster in Anatolia (Demirtas 2000) whereas aftershocks have continued for 6 months (Ambrasseys et al. 1988). The Bolu-Gerede earthquake (M w = 7.3) on 01.02.1944 is a major one through the instrumental period (Fig. 4). It was noted that 2,381 people died and 50,000 houses were damaged (Tasman 1944; Ketin 1948). Although 17 August 1999 Marmara and 12 November 1999 Düzce earthquakes occurred on the Western segments of the North Anatolian Fault, Bolu has been strongly affected. The measured average value of the horizontal ground acceleration was 0.48 g in Bolu (Ugras et al. 2000).

Table 1 Magnitude and damage records of the earthquakes around Bolu (GDDA 2005)
Full size table
Fig. 4
figure 4

A view from study area and the alignment of 1944 surface rupture

Full size image

Field studies

Geotechnical boreholes at 88 locations with a total depth of 1,760 m have been drilled in order to determine the consistency of fine soils, stiffness of the coarse soils, to obtain undisturbed and disturbed samples, and to measure the groundwater level. The depths of boreholes ranged between 6.0 and 22.0 m. Standart penetration test—SPT (ASTM D1586-99) was performed and SPT N blow counts were obtained in the boreholes. P and S wave velocities have been measured by seismic refraction method down to 22 m depth for six x-sections (Fig. 5). Engineering geological map (Fig. 2) cross-sections and groundwater level map of the study area were prepared (Figs. 6, 7).

Fig. 5
figure 5

S wave velocity model (a) and soil classes (b) of GF3 profile (Kilic et al. 2001)

Full size image
Fig. 6
figure 6

a–c A–A′, B–B′ and C–C′ engineering geological cross-sections

Full size image
Fig. 7
figure 7

Groundwater level map of the study area

Full size image

Geotechnical evaluation

Index and physical properties of the soil samples have been investigated in Ankara University Engineering Geology laboratory (ASTM 1994). Quaternary alluvium deposits in the study area were classified as clay with low and high plasticity (CL and CH), silt with low and high plasticity (ML and MH), and clayey sand (SC) and silty sand (SM) based on Unified Soil Classification System. Liquid limit of the clayey soils changes between 19 and 66%, plastic limit changes between 11 and 31% and plasticity index ranges between 3 and 40%. Natural moisture content is between 11 and 42%, while the unit weight ranges between 17.05 and 18.42 kN/m3 (Table 2).

Table 2 Statistical distribution of geotechnical properties of clay with low and high plasticity
Full size table

Liquid limit of the silty soils range between 27 and 59%, plastic limit changes between 23 and 33% and plasticity index changes between 4 and 26%. Natural moisture content is between 30 and 42%, and the unit weight ranges between 17.90 and 18.50 kN/m3 (Table 3).

Table 3 Statistical distribution of geotechnical properties of silt with low and high plasticity
Full size table

Liquid limit of the sandy soils range between 20 and 68%, plastic limit changes between 10 and 29% and plasticity index ranges between 6 and 39%. Natural moisture content is between 7 and 33%. Unit weight of the sandy soils are limited due to the difficulty in undisturbed sampling, values ranging between 18.23 and 19.70 kN/m3 (Table 4). The fine-grained soils in the study area are classified as “plastic” group according to consistency, plasticity and plasticity index. The coarse soils have been classified as compact-dense based on SPT N blow count.

Table 4 Statistical distribution of geotechnical properties of clayey sand and silty sand
Full size table

Liquefaction potential

The process of seismically induced liquefaction of saturated granular sediments has been studied extensively and is reasonably well understood (Obermeier 1996; Seed 1979; National Research Council 1985; Castro 1987; Dobry 1989). The combination of elevated pore pressure with large loss of strength is sometimes referred to as “complete liquefaction” (Obermeier 1996; Youd 1973). Liquefaction potential of the soils in the study area was investigated by the methods proposed by Youd and Idriss (2001) and Cetin et al. (2004) (Fig. 8, Tables 5, 6). Besides, liquefaction analysis was also performed according to the shear wave velocity (V s) derived from seismic refraction studies. Liquefaction potential evaluation based on V s is shown in Table 7. Moment magnitude (M w ) and maximum horizontal ground acceleration (a max) have been chosen as 7.5 and 0.48 g, accordingly (GDDA 1996). Liquefaction potential of the silty and clayey soils were studied according to the procedure of Seed et al. (2003). The majority of the soils took place in the “B” region (Fig. 9). However, these soils do not provide the ω > 0.85 × LL criteria. Thus, no liquefaction is expected for the fine soils. Probable surface damage in liquefiable soils has been investigated as proposed by Ishihara (1985) and liquefaction induced probable ground damage has been determined in some boreholes (Fig. 10). Main shock of the 12 November 1999 earthquake was determined by 20 accelerometers located on different distances and soil conditions by the Earthquake Research Department of General Directorate of Disaster Affairs of Turkey. The maximum acceleration value has been measured as 0.80 g (E–W) at Public Works and Settlement Management of Bolu station during the Düzce Earthquake 12 November 1999. However this value seems to be high due to the instrumental reasons and this peak value does not have an effect on structures. The maximum amplitude of S wave is 0.48 g in 4-s period in Fig. 11, overall the whole waveform, reported by General Directorate of Disaster Affairs (Ceken 1999). Hence, the acceleration value of 0.48 g was taken into consideration for the liquefaction analysis in this study.

Fig. 8
figure 8

Cyclic stress ratio and (SPT N)1,60 plot based on Youd and Idriss (2001)

Full size image
Table 5 Depth of the layers, groundwater level and FS values byYoud and Idriss (2001) method
Full size table
Table 6 Depth of sandy layers, groundwater, factor of safety by Cetin et al. (2004) method
Full size table
Table 7 Liquefaction potential based on V s by Andrus and Stokoe (2000) method
Full size table
Fig. 9
figure 9

Liquefaction potential of silty soils (Seed et al. 2003)

Full size image
Fig. 10
figure 10

Probable surface damage in liquefiable soils (Ishihara 1985)

Full size image
Fig. 11
figure 11

Bolu station record and the abrupt peak amplitude of the waveform (Ceken 1999)

Full size image

The results of the methods proposed by Youd and Idriss (2001) and Cetin et al. (2004) are in accordance by means of factor of safety values. FS < 1.0 has been chosen as a threshold value for liquefaction. Numbers of the boreholes with liquefaction potential based on the methods above, are as follows: BH5-13-28-41-42-43-46-59-60-63-64-65-72-90-92-96-97-98-99-100-104-107-114-124-129-130 (Fig. 12). During the construction of the microzonation map, the liquefaction potential of the soils, which have been derived from the methods above have been taken into account and the LZ boundaries have been drawn accordingly. Aydan et al. (2000) and Kilic et al. (2001) indicated that, no liquefaction risk exists in the study area since the effects could not reach the surface due to the thick fine-grained cap soil. However, if an earthquake of M w  = 7.5 and a max = 0.48 g occurs, liquefaction may be a problem based on those assessments.

Fig. 12
figure 12

Microzonation map based on liquefaction potential and FS

Full size image

Environmental impacts

Liquefaction induced volumetric settlements or lateral movements have not been observed on ground level because of the existence of thick cap soil cover. Besides, structural failures or damage on buried structures such as pipelines were also not observed due to liquefaction. The damage on the structures seem to be mostly the result of construction problems (Fig. 13). According to the results of liquefaction potential study of the soils, liquefiable layers were determined at different levels. The study area takes place between two major strike slip faults with secondary oblique ones. These cracks may lead the effects of liquefaction reach up to the ground level. In this case, the study area may suffer from the environmental affects of liquefaction by means of settlement, sand cones and agriculture. Moreover, structural failures and the possibility of life loss may be considered.

Fig. 13
figure 13

Structural damage of some buildings

Full size image

Results and recommendations

Geological, geotechnical and geophysical investigations around Bolu settlement have been carried out and liquefaction potential of the alluvium has been determined by different methods using a probable scenario earthquake. The possible environmental impacts of liquefaction were discussed. The result and recommendations are given as follows:

Quaternary aged alluvium is the main formation of the area, made up of clay with low-high plasticity, silt with low-high plasticity, clayey and sily sand with lateral and horizontal transitions. Groundwater level changed between 0.45 and 10.50 m during March–April 2001.

Liquefaction potential of the study area has been evaluated according to Youd and Idriss (2001) and Cetin et al. (2004) methods based on SPT data. For the liquefaction potential assessment, maximum horizontal ground acceleration (a max) and moment magnitude (M w ) have been taken into consideration as 0.48 and 7.5 g, respectively. Besides, liquefaction potential has been determined by the procedure using V s (Andrus and Stokoe 2000).

In the southern section of Bolu settlement area liquefaction under dynamic loading is expected locally in boreholes BH5-13-28-41-42-43-46-59-60-63-64-65-72-90-92-96-97-98-99-100-104-107-114-124-129-130 with depths between 2.0 and 15.0 m according to the two proposed methods.

Since liquefaction seems to be significant in local boreholes, threshold value of safety factor for liquefaction (FS < 1.0) has been taken into consideration. Because of the existence of thick cap soil, liquefaction was not observed during the 12 November 1999 Düzce earthquake in Bolu province.

Considering foundation design of buildings and the other engineering structures, geotechnical properties of construction sites must be investigated in details.