1 ATS Building Damage Assessment Next to Karaj Subway Station Eshagh Namazi 1, Mohsen Hajihassani 2, Mohammad khosrotash 3, Hisham Mohammad 4, Masoumeh Karimi Shahrbabaki 5 1 Research Assistant, Steel Technology Centre, Universiti Teknologi Malaysia. 2 Research Assistant, Department of Geotechnics and Transportation, Faculty of Civil Engineering, Universiti Teknologi Malaysia. 3 Managing Director, Tunnel Rod Construction Consulting Engineering Inc. Iran. 4 Senior Lecturer, Faculty of Civil Engineering, Universiti Teknologi Malaysia. 5 Department of Mining Engineering, Islamic Azad University, Tehran, Iran. ABSTRACT Tunnel excavation in the urban areas can induce ground movements, which distort and, in serve cases, damage overlying buildings and services. To predict damage in the buildings an analytical framework approach based on the concept of limiting tensile strain is used world-widely. In developing the approach clearly, there is the conspicuous shortage of well-documented case histories of measured building response to ground movement. In this paper, effects of construction of the Station E in Karaj Subway System on the adjacent 2-storey commercial buildings are presented. The Station was constructed by enlargement of the NATM tunnels. Damage to the building is assessed in two ways. First, the analytical assessment of building damage is made by calculating tensile strains due to settlement. Second, an external visual inspection was made of cracking or damage to verify the analytical prediction. The results provide an important frame of reference for interpreting the measured responses of the nearby buildings. KEYWORDS Karaj Subway; Surface Settlement; Building Damage Assessment; Subway Station; Cracks.
2 1. INTRODUCTION Having 2.6 million inhabitants, Karaj is a large city located around 40 km to the west of Tehran, the capital of Iran. With ever-increasing population and the number of vehicles, Karaj is currently under development and construction of a new subway system to address the transportation problems. Line 2 of Karaj Urban Railway (KUR), distancing for 24 km, is being constructed between Kamal-Shahr, in north-western of Karaj, to Malard, placed in south of the city. The KUR includes a tunnel with 7.80 m height and 8.40m width at the depths of m below the ground level. The full length of the KUR is shown in the simple map of Fig. 1. In all, there are 23 stations constructed in open cut and enlarged tunnels by NATM tunneling techniques. One particular interest of KUR is a part next to station E, where the station is bored next to businesses, Fig1. Being confronted with increasing population and the numerous existing vehicles, Karaj is currently under development and construction of a new subway system to overcome the transportation problems. The aim of the study is to evaluate the ground movement influences on the existing surrounding buildings (buildings A, B, C, D, E and F in Fig. 1). To minimize damage to adjacent buildings, it is essential to have a comprehensive documentation of the responses of these buildings to station construction. This includes a complete record of ground conditions and the response during and after construction. This paper summarizes the conditions at the site and presents the correlation among construction activities, measured displacements, distortions and consequent damages to buildings. Kamal Shahr A B D C B Station E C Ghehem Street D m Scale E E F Almehdi Street G F H I O Y U W J Q S K N M Malard Fig. 1. Line 2 of Karaj Urban Railway and the location of case study buildings related to station E.  2. GROUND CONDITIONS In general, 40 boreholes and 18 shafts were bored to obtain continuous samples for visual description . The samples collected from boreholes were tested in the form of laboratory uniaxial compressive and direct shear tests. The in-situ plate loading test, in situ direct shear test and Standard Penetration Testing (SPT) also were performed to provide more information on the properties of the soil. Between Stations E and F, seven boreholes (BH201-BH207) with the distance of 100 m and one shaft (TP-239) were excavated. Borehole 201 is located about 100 m south-east of the reference point. Fig. 2 is a summary log of this borehole to show the associated sub-division of the soil in the site. Visual inspection reveals that the soil contains inorganic clay with clayey sand at the top, followed by clayey sand and seldom silty sand overlying clayey and silty gravel. Furthermore, Fig. 2 summarizes the properties of the mentioned soil layers obtained from geotechnical tests. The ground water table was not observed in any boreholes. V X T P R
3 Fig. 3. Typical main stages for construction of station E.  4. BACKGROUND OF BUILDING DAMAGE ASSESSMENT 5 4 Fig. 2. Summary log and geotechnical properties of the soil close to station E.  3. SEQUENCE OF THE STATION CONSTRUCTION The station E was constructed as enlarged tunnels with approximately a length of 160 m and 16 m width. Fig. 1 illustrates the position of the station from the buildings. All buildings are a two-storey, steel-framed building founded on a reinforced concrete raft. On the KUR, NATM technique has been used for construction of station E. In this case, the large tunnel was excavated in several parts in order to help the ground to stand until lining completion. The sequence construction is presented in Fig. 3. The shape of the tunnel is different from conventional circular tunnels. As the figure presents, the process is started with removing and the enlargement of sides drifts. Subsequently, the excavation and compensation grouting of the piles are performed. Removing and grouting of arch ribs is a next stage in the sequence of construction followed by the excavation of the whole tunnel. Finally, subsequent to the installation of waterproof membrane, the invert is closed by the concreting. Supposing the whole length of the station divided into three blocks (block 1, 2 and 3 in Fig. 4), block 1 is excavated first. Prior to the construction termination of the block 1, the exaction of the second block is started and so on. The first and the last excavated ribs in block 1 were Rib 05 and Rib 10, whereas there was Rib 31 and Rib 42 in blocks 2 and Rib 59 and Rib 56 in block 3, respectively (Fig. 4). After process completion, the excavation height and width were around 15 m and 18.8 m at the depth of 8.4 m below the ground surface. Damage to buildings due to excavation-induced ground movement depends on conditions of the building before excavation begins, the magnitude of the ground displacement cause by excavation, and the structural system of the effected building . Because the potential sources of damage are various, simple criteria which correlate the damage categories to a component of displacement are not universally applicable. The earliest method based on empirical models used numerous case studies to establish correlations between distortion parameters and the corresponding damage limits (Skempton and MacDonald, 1956; Polshin and Tolkar, 1957 and Bjerrum, 1963). These methods usually deal with the risk of buildings owing to displacement under their own weight and not necessarily applicable to the movements of buildings due to adjacent excavation. Building's criteria based on two most critical components of displacement, called angular distortion and deflection ratio, are shown in Table 1. Deflection ratio /L, is the relative deflection divided by the distance between the two reference points which relative deflection is the maximum displacement of the settlement profile of a structure relative to the straight line connecting two settlement reference points. Angular distortion β is defined as the differential settlement between two points divided by the distance between them, less the tilt. Currently, the analytical method has been widely used in the engineering practice to predict building damage. This method was first proposed by Burland & Wroth (1974) and Burland et al. (1977). They assumed the onset of crack is associated with average tensile strain within the building. To obtain the maximum tensile strain in the buildings, they represented individual walls of the buildings as a linear-elastic deep beam subjected to a point load at the centre. The authors used the structural engineering principles to derive the relationships between the deflection ratio and the limiting tensile strains in the beam. They consider the effects of different deformed shape called hogging for concave downward deflection profiles and sagging for concave upward deflection profiles. They had shown excavation-displacement can result in sagging, hogging or
4 both in the buildings, whereas the hogging is far more damaging than sagging. Table. 1. Empirical damage criteria caused by Building selfweight Reference Building Type Severity Criterion Skempton and Frame Onset of Β=1/300 McDonald building cracking (1956) Polshin and Load Onset of 1/2000< /L<1/1400 Tolkar (1957) bearing cracking For L/H>5 brick wall 1/3300< /L<1/2500 For L/H<3 Bjerrum (1963) Frame Negligible β=1/500 Structure Structural β=1/150 Damage Afterwards, Boscardin and Cording (1989) completed Burland and Worth (1974) model by including horizontal tensile strain using simple superposition to consider the role of horizontal displacement induced by adjacent excavation and tunneling. They defined categories of damage based on angular distortion and horizontal strain and compared the damage predicted by their suggested method with recorded cases of damage using Table 2. Recently, Mair et al. (1996), Burland (1995) and Finno et al. (2005) extended the previous studies by defining the new criteria for damage based on deflection ratio and horizontal strain. Even the approach presented in the previous section was originally developed for assessing the risk of subsidence damage for the London Underground Jubilee Line Extension project , it is now widely used internationally with minor variation. But a problem is a limited amount of comprehensive case studies were reported for urban tunneling project to verify the current method. 5. FIELD INSTRUMENTATION Fig. 4 shows the excavation site along with monitoring sections. There are six monitoring sections, which are shown with their position related to a local co-ordinate (named S-). Surface settlements in these sections were monitored using optical survey points established on the centre, left and right sides of the tunnel axis with the distance of 5 m (Fig. 4b). The monitoring points consisting of steel rods grouted into the ground about 100 cm below the ground surface to isolate the rods from asphalt movement. Table. 2. Category of damage (modified after Burland et al. 1977). Extensometer measurements also made across several spans within the tunnel in some sections (named C-). The spans were across the horizontal diameter (axis level, less often at the shoulder or knee level) and from the crown to the same points at the end of the horizontal spans. Only section S-3385, S-3400, S and S-3460 are presented in this study to obtain the surface settlement trough. Degree of severity Crack width (C.W) Ease of repair Negligible (0) <1mm - Very slight(1) 1mm<C.W<5mm Fine cracks which are
5 Other monitoring methods such as monitoring the displacements of buildings and floor across existing cracks or other discontinuity were made during construction. The widths of cracks were measured using a matt steel ruler or gauge. Fig. 4a demonstrates the position of old and new cracks which are developed in response to ground movements induced by station construction. During construction, the monitoring team obtained the location of these cracks from simple visual inspection and then measured the cracks and joints frequently.
6 C2 C16 C21 Rib 01 C19 C20 C26 C3 C8 C9 Rib 05 Rib 10 C14 Block 1 Rib 26 Rib 31 Block 2 Rib 42 Block 3 S C Rib 56 S S C S C Rib 59 Rib 64 S C S (a) (b) Fig. 4. Location of Instrumentation (a) Plane and (b) Section, 
7 Surface Settlement (mm) 6. OBSERVED SURFACE DISPLACEMENT The vertical transverse displacements at the location of monitoring points in the sections are shown in Fig 5. The displacements are given after completing construction of the station when the movements had reached to the steady state's conditions. The figure illustrates the transverse displacement against the distance from tunnel centre line for four sections. The displacement in section S-3400 is the highest one with a maximum around 75 mm. The surface settlement trough in sections S-3385 and S-3460 are identical excluding the monitoring point S1. The maximum displacement in these sections was 30 mm, while on the contrary, in section S-3440, that shows the shallowest surface settlement trough, was just 11 mm. The volume loss incorporated with maximum surface settlement and soil properties were obtained in the site. Volume loss is defined as the volume of ground loss as a proportion of the final tunnel volume . At the site, the volume losses in sections S-3385, S-3400, S-3440 and S-3460 were estimated 0.4%, 0.9%, 0.1% and 4%, respectively. These volume losses in comparing to the previous case studies reported by Lake et al. (1992) in the greenfield conditions with the same excavation geometry and geological conditions, are lower. It is important to note that the stiffness of the buildings decrease the displacement and volume loss induced by excavation . The Gaussian error function of the settlement profile suggested by Peck (1969) is chosen to obtain a graphical best fit to the observed movements in the sections. Based on the suggested displacement, the ground movement under the buildings was represented. These ground movement models close to the buildings are shown in Fig. 6. The shapes of settlement profiles show that all the buildings experienced hogging mode of deformation. The boundaries between the two zones of hogging and sagging were defined by the inflection points of the settlement profiles obtained in a 10 m distance from the tunnel centre pint. Given the data from Canadian Geotechnical Society (1992), the cracks are onset at lower angular distortions in the region of hogging than sagging. Fig. 6 clearly shows the difference in the settlement trough under the existing buildings. The deflection ratio from these settlements can be obtained from the relative deflection dividing by the length. In the absence of other information, the horizontal displacements adjacent to the excavation were assumed equal to vertical displacement . The horizontal strain, ε h on the surface in every distance from tunnel centre line can be obtained by differentiating horizontal displacement . From these analyses, each building was evaluated and categorized for potential damage using the method suggested by Burland (1995). Table 3 summarizes the assessed displacement's parameters and predicted damage of the buildings. Refer to table 2 as one basis for judging damage severity S-33 S-34 S-34 S Transverse Dispalacement from Tunnel Axis (m) Fig. 5. Measured ground settlement, 
8 Surface Settlement (mm) Surface Settlement (mm) Surface Settlement (mm) Distance from Tunnel Centre line (m) A S-3385 S Distance from Tunnel Centre line (m) B, C, D S Distance from Tunnel Centre line (m) E, F S-3440 Fig. 6. Profile of measurements and predicted ground settlement adjacent to the buildings in the site
9 Table. 3. Summary of displacement parameters and damage category estimation Buildings L H (m) Δ/L(%) ε h (%) ε br (%) ** ε dr (%)** Damage Severity A Negligible B, C, D Moderate E, F Slight *length (L) and height (H) of buildings **Resultant bending tensile strain ***Resultant diagonal tensile strain (in every case, only resultant bending or diagonal strain which is critical is given in the table) 7. CRACK OBSERVATIONS The visible cracks on the interior and exterior walls in the existing buildings were surveyed by consulting groups during and after construction of the stations. In Fig. 4a, just cracks possessing the width of greater than 1-2 mm were represented. Only a few cracks were opened up more than 15 mm. These cracks were observed in buildings D and E extended in the whole height of the interior wall from floor to roof (C8, C19 and C26 in Fig. 4a). According to these cracks, the observed damage in these buildings is characterized as severe and very severe (table 2). In building B, the damage was mainly consisted of hairline cracks in the walls except crack C21 with the width of 5 mm and the length of more than 1 m. Moreover, in this building a crack of around 8 mm wide is reported in the floor from which the building can be classified as suffering Slight damage. The damages recorded within buildings A and E were incorporated with cracks C2 and C14 less than 5 mm. The existing cracks within building A and F did not exceed more than 5 mm. The recorded damages within these buildings were very slight and slight." 8. CONCLUSION From the results of these case studies, a number of conclusions can be drawn. In general, the damage to the buildings consisted mainly of hair cracks except a few cracks with the width more than 1-2 mm. These few cracks have an important role in categorizing the damage based on the observation. The method proposed by Burland and Wroth (1974) underestimated the damage in most of the cases except building B and F. This method proposes a number of assumptions that may not be appropriate for these buildings. For example, in this method the effects of the old cracks in building damage induced by excavation are not considered. The method also considered a same boundary of damage severity for hogging and sagging mode of deformation. In this study, it is illustrated that when the buildings are occurred in the hogging zone the limiting value of deflection ratio and horizontal strain were less than the limiting values proposed by Burland (1995). It can be the reason of under predicting of the damage in some of these buildings. References  Bjerrum L. (1963). Allowable settlement of structures. Proc. European Conf. on Soil Mech. and Found. Engr., Wiesbaden, Germany, No. II, pp  Boscardin, M. D. and Cording, J. C. (1989). Building response to excavation-induced settlement. ASCE Journal of Geotechnical Engineering, 115, No. 1,  Burland J. B. (1995). Assessment of risk of damage to buildings due to tunnelling and excavation. Invited Special Lecture. 1 st Int. Conf. on Earthquake Geotech. Engineering, Tokyo, pp  Burland J. B., Broms B. B., and de Mello V. F. B. (1977). Behaviour of foundations and structures. State-of-the-Art Report. Proc. 9th Int. Conf. on Soil Mech. and Found. Engr., Tokyo, Japan, pp  Burland, J.B., Standing, J.R., Jardine, F.M. (2001).Building response to tunnelling: case studies from construction of the Jubilee Line Extension, London, CIRIA and Thomas Telford, ISBN:  Burland, J. B. and Wroth, C. P. (1974). Settlement of buildings and associated damage. Settlement of Structures, Proc. Conf. Organized by Geotechnical Engineering Society. Pentech Press, London, pp  Canadian Foundation Engineering Manual. (1992). Canadian Geotechnical society. 3 rd Edition, BiTech Publishers, Vancouver, BC.  Darya Khak Pay Consulting Engineering Company. (2005). Geotechnical reports of Karaj urban railway organization. In Persian.  Dimmock, P. S., Mair, R. J. (2007). Estimating volume loss for open-face tunnels in London clay. Geotechnical engineering, l 60, PP:  Finno, R.J. and Bryson, L.S. (2002). Response of Building Adjacent to Stiff Excavation Support System in Soft Clay, Journal of Performance of Constructed Facilities, ASCE, Vol. 16, No. 1,  Finno R. J., Voss F. T., Rossow, E., and Tanner Blackburn J. (2005). Evaluating damage potential in buildings affected by
10 excavations. Journal of Geotechnical and Geoenvironmental Engineering. 131, No. 10,  Khosrotash, M., Taskindoost, M., and Kashfi, M. (2005). Circumstance monitoring report during construction of Karaj Subway tunnels. Tunnel Rod Construction Consulting Engineering Inc.  Khosrotash, M., Behbody, A., and Yazarloo, H. (2008). Monitoring in urban area, case study: construction of Karaj Subway tunnels. Iranian tunneling association magazine.  Lake, L. M., Rankin, W. J and Hawley, J. (1992). Prediction and effects of ground movements caused by tunnelling in soft ground beneath urban areas. CIRIA Project report 30, Construction Industry research and Information Association, London.  Mair R. J., Taylor R. N., and Burland J. B. (1996). Prediction of ground movements and assessment of risk of building damage due to bored tunnelling. Proc. Int. Symp. Geotechnical Aspects of Underground Construction in Soft Ground. Balkema, Rotterdam. 1996, pp  O Rourk, T. D., Cording, E. J., and Boscarding, M. D. (1976). The ground movement related to braced excavations and their influence on adjacent structures. Univ. Of Illinois Rep for the U. S Dept. of Transportation, Rep. No. DOT-TST-76t-22, Washington D.C.  O Reilly, M. P., & New, B. M. (1982). Settlements above tunnels in the united kingdom- their magnitude and prediction. The institution of mining and metallurgy, London. Pages:  Peck, R. B. (1969). Deep excavations and tunnelling in soft ground. proc.7 th int. Conference on soil mechanics and foundation engineering. Pages  Polshin D. E. and Tokar R. A. (1957). Maximum allowable nonuniform settlement of structures. Proc. 4 th Int. Conf. on Soil Mech. and Found. Engr., London, England, pp  Potts, D. M., & Addenbrooke, T. I. (1997). A structure's influence on tunnelling-induced ground movements. Proc. Instn. Civ. Engrs. Geotech. Engineering, 125,  Skempton A. W. and MacDonald D. H. (1956). The allowable settlement of buildings. Proc. Inst of Civ. Engrs., Part III, 5, pp  Tunnel Rod Construction Consulting Engineering Inc. (2008). Monitoring reports of line 2 of Karaj subway tunnels. In Persian.