İstanbul ve Erzincan şehirlerinde zemin büyütme etkilerine göre mikrobölgeleme

thumbnail.default.alt
Tarih
1994
Yazarlar
Lav, M. Ayşen
Süreli Yayın başlığı
Süreli Yayın ISSN
Cilt Başlığı
Yayınevi
Fen Bilimleri Enstitüsü
Institute of Science and Technology
Özet
Depremler esnasında yapxsal hasarlara neden olan en önemli faktörlerden biri, yerel şartların deprem dalgaları üzerindeki büyütme etkisidir. Farklı büyütme potansiyeline sahip bölgelerin belirlenmesi ile depreme dayanıklı yapı tasarımı veya takviyesi için, farklı tasarım kriterleri gereken bölgeler ayırılabilir ve böylece yapılaşmaya ekonomik çözümler de getirilebilir. Eski İstanbul ve Erzincan, şehir bölgesinde, deprem dalgaları üzerinde farklı büyütme etkilerine neden olabilecek yerel şartları belirlemek amacıyla çalışmalar yapılmıştır. Zemin tabakalarının büyütme özelliklerinin belirlenmesinde esas olarak, çok tabakalı, yatay, viskoelastik ortamlarda, düşey doğrultuda kayma dalgaları yayılması prensibine dayanan bir program kullanılmıştır. Büyütme özellikleri benzer olan bölgeler gruplandırılarak, sonuçlar, depremlerde gözlenmiş hasarlarla karşılaştırılmıştır. Farklı özelliklere sahip depremlerde, zemin tabakalarının büyütme özelliklerinin değişimi incelenmiş ve bu değişim de dikkate alınarak mikrobölgeler oluşturulmuştur. Eski İstanbul'da yerel zemin koşulları ve özellikleri, geçmiş dönemlerde yapılmış jeolojik ve geoteknik çalışmalardan elde edilmiş bilgilere dayanarak belirlenmiştir. Eski İstanbul'da yeralan zeminlerin tabakalaşma ve özelliklerinin çeşitliliği, farklı büyütme etkilerinin açıkça gözlenmesini sağlamış ve büyütme potansiyeli en yüksek olan bölgelerin, genellikle, kalın, yüksek plastisiteli kil tabakaları bulunan bölgeler olduğu belirlenmiş ve bunun nedenleri araştırılmıştır. Geçmiş depremler ve özellikle son, büyük 1894 depreminin Eski İstanbul'da sebep olduğu hasarların dağılımı, ve ağır hasarın genellikle belirli bazı bölgelerde yoğunlaşması ile zemin tabakalarının büyütme özellikleri arasındaki ilişkiler araştırılmıştır. Erzincan durumunda ise, zemin özelliklerinin bir bölgeden diğerine değişimi, İstanbul'daki kadar belirgin farklılıklar göstermemektedir. 13 Mart 1992 depreminde gözlenen hasarlarla zemin büyütme özellikleri ve diğer zemin özellikleri arasında ilişkiler araştırılmıştır. Yüzey kayıt özelliklerinin derinlikle değişimi incelenmiştir.
Microzonation is a process that purposes proper land-use planning to decrease earthquake effects. In order to carry out proper land-use planning, microzonation involves incorporation of geologic, seismological and geotechnical factors into economically, sociologically and politically justifiable and defensible zonation for earthquake effects, so that engineers can site and design structures that will be less susceptible to damage during major earthquakes. Microzonation involves not only planning land-use for new structures but also taking protective measures for existing structures. Therefore microzonation should provide information on the relative damage potential for existing structures in a region." In this inter-disciplinary process local site conditions play an important role as a transition media for earthquake waves between rock formations and structures, that cause a change in properties of earthquake waves as well as being influenced by them in many ways. The earthquake damage due to local site conditions alone may occur in four different ways; as an amplification effect of local site conditions on earthquake waves; as loss of strength of foundation soil due to liquefaction; as dislocation and displacement of foundation soil due to land slides and, as vertical displacements in foundation soil due to densif ication. The phenomena of liquefaction, landslide and densif ication occur under special local soil and topographic conditions, but the site amplification effect on earthquake ground motion is the most common and the major cause of damage in most earthquakes. Different combinations of ground motion parameters, incident wave fields and site characteristics reveal a rich variety of site effects. Classification of similar site effects in an area form the microzonation map of this area according to the site effects. There are two alternative approaches to the meaningful microzonation according to site effects. The first one is to measure site-specific amplification factors empirically using the data from large and small earthquakes. The second XVII is to predict the frequency dependent amplification effects as close to reality as possible using theoretical methods and representative site properties. The first alternative requires large number of earthquake records taken simultaneously during different earthquakes with different magnitudes. The registration of ground motion accelerations simultaneously at sufficient locations in an area for different earthquakes has been achieved in only a small number of research arrays in the world for particular research purposes. To achieve a microzonation map based on real earthquake records requires a large number of expensive equipment, and waiting for earthquakes to happen is another problem. On the other hand, the second alternative requires well established theoretical models and convergent solution techniques as well as detailed information on the different properties of the site and local soil conditions. In the recent decades, research on the second alternative has reached to a level that, if ground motion parameters of probable earthquakes and local site properties in a region are known in detail, the site effect on ground motion for many realistic situations may be predicted. Some of these predictive methods have been shown to exhibit successful comparisons with the observations for a number of cases in the literature. This study purposes to use one of these predictive methods to carry out microzonation maps in terms of soil amplification effects for two selected region. The first one is the old sector of the city of Istanbul and the other is the central part of the city of Erzincan. The requirements for selection these regions for this type of study stem from the facts that Istanbul is the most risky city in its region due to high concentration of both population and important buildings and most of the monuments and historical mosques are located inside the old city walls, and Erzincan is located in a seismically very active region so that a strong earthquake in every fifty years causes important losses. Prediction of site amplification properties for different locations during different earthquakes in these regions would show the changes in the effect of soil factors from one place to another and it would be possible to compare and to discuss these effects with the observed damage and to show the areas that require different criteria for earthquake resistant design. XVIII The old part of the city of Istanbul, bounded by the historic city walls in the northwest, the Golden Horn in the northeast and the Marmara Sea in the south was selected to conduct a microzonation investigation as a first stage in this study. The major emphasis was to correlate observed damage patterns in earthquakes with the amplification properties of site conditions and to find out the range of changes in amplification properties due to different earthquake effects. In this study, to predict soil amplification properties and total earthquake-site-structure response, a widely accepted solution technique, which depends on the theory of one dimensional wave propagation in horizontally layered, linear viscoelastic medium, called Shake has been used. This solution technique also takes into account nonlinear material properties in the form of equivalent linear solution. In predicting amplification properties of soil layers and site response during earthquakes, detailed information on the properties of these earthquakes and local soil conditions are required because: 1- The site amplification properties depend on many physical parameters, including: level of dynamic shear strain induced by earthquake loadings, and shear wave velocity, density, material damping, thickness of soil layers. 2- The level of dynamic shear strain and its effects on soil properties are the most important aspects in site response. Depending upon soil type and position, the shear strain level induced in the soil layers by the earthquake ground motion, increases as the acceleration levels (magnitude) increases and decreases as the distance from energy release increases. 3- The response of the soil layers depends strongly on the strain dependent properties of the soil. Depending on the level of dynamic shear strain and th' contrast in the physical properties of the soil rock, the soil acts either as an energy amplifier energy transformer. XIX 4- Site amplification increases the surface ground motion in a narrow period band, and the magnitude of site amplification and the place of this period band are related to the properties of the soil layers and shear strain levels induced during earthquakes. Istanbul as a large city has been subjected to many different types of civil engineering projects. To fulfill the above mentioned data requirements, efforts were made to collect detailed information concerning borehole and observed information on different properties of soil layers. The overwhelming features of the soil conditions in the area were a thick layer of miscellaneous fill that covers most of the area with different thicknesses, thick green silty clay layers that underlay the fill in most parts of the area and, growacke and limestone layers that underlay fill layers at some places in the area. Later the soil profiles reduced to a manageable number of representative soil profiles for each location. Soil properties were converted mostly from SPT-N values. The strain- dependent changes in dynamic soil properties were represented by the average of the curves for soil classes encountered in the area. The prediction of amplification properties of soil layers, by means of a dynamic analysis, requires an earthquake record. If results are compared with damage occurring during a particular event then it would be best to use that particular record. The last strong earthquake in the area experienced on 12 July 1894 but the recording of earthquake motion was only possible after the 1900; therefore, there was no earthquake record for this last event and others. Besides, the estimated intensity for the area have had the influence of local soil conditions. Therefore, artificial earthquake records produced by several experts for the city of Istanbul have been used in this study by scaling the maximum acceleration value 0.25-0.30-0.35 g; because this was the most probable range of the maximum acceleration on rock affecting the area during this event. Site amplification functions were assigned to each location as the average of the amplification functions obtained in order to capture 'the probable range for both physical properties and the dynamic shear strain level induced by this particular earthquake. XX extensive damage surveys and site investigations carried out within the city limits in the months following the 13 March 1992 earthquake. The main purpose of this second case in the study was again the same as with the Old Istanbul case. In addition to this, a number of parameters were available for to seek correlations among local geotechnical factors, earthquake and damage distribution characteristics. The earthquake ground motion was registered at a station in the city center. But, this record was influenced by soil conditions underlying the station. A base rock motion or an outcrop rock motion is required to calculate free field motions at any site. The outcrop rock motions were obtained from the free field record by a deconvolution process on the soil profile obtained from a bore conducted near the station. The free field ground motions and amplification functions as well as response spectra were calculated at every borehole location using the outcrop rock motion. The damage ratios which were used in the correlations were calculated within the circular area with a 250 m radius around every borehole. The resultant correlations between observed damage and the predicted parameters were discussed and used to separate zones with similar parameter values. During the process of deconvolution, the variation in the characteristics of the surface ground motion with depth has also been examined. The results have shown that the amplification factors at certain frequencies were influenced by certain soil layers. The surface layers situated on the thick gravelly layers (dense, partly cemented) has much influence at short period range. But this influence remains highly limited due to the nonliear soil behavior in the surface layers during the strong earthquake. While the deeper layers has much influence at intermediate period range. In summary, geographic distribution of the amplification properties of the soil layers within the two city center was predicted taking into account the effect of different earthquakes. Damage distribution during strong earthquakes was compared with the amplification properties and overwhelming results of these comparisons were highlighted. Microzones were established in terms of amplification properties. XXIII The site amplification functions in the area showed some certain characteristics for different locations. The clearest difference was between those for fill layers underlined by growacke (or limestone for some locations) and fill layers underlined by thick, green, highly plastic, medium to stiff silty clay layers. The functions for the former site showed slight peaks around smaller periods that were the predominant soil periods, but the functions for the latter sites generally had sharp peaks around the medium period band. The microzones were separated according to similar amplification and predominant soil periods, but these results need to be compared to observed damage patterns in the area. The information on the damage which was experienced during earthquakes in the area, was collected from different sources. Emphasize has been given to those which occurred during the 1894 earthquake and damage zones separated in terms of damage levels. Furthermore, the predominant periods of structures needed to be taken into account in comparisons, and these periods were approximately estimated by empirical relations. The predominant period of structures is important, because the response of a structure can also be increased or decreased depending on its natural period of vibration. The clearest comparison between the predicted amplification functions and observed damage was between the sites with fill layers underlined by growacke or limestone layers, and the sites which contain thick clay layers. No recorded damage to structures has been found in the former sites as it is evident from the amplification functions that the predominant soil periods appeared to be low and no significant amplification occurred within the period range of structures. Most of the damage is concentrated on the latter sites and the amplification functions also supported this fact, having significantly higher values within the period range of structures. The reasons of high amplification concentration particularly in some regions have been discussed and detailed comparisons between amplification properties and earthquake damage for the other sites are presented in the text. In order to investigate the effects on soil amplification properties due to different earthquakes, a number of available strong motion acceleration records have been used scaling the peak acceleration levels. XXI The site amplification functions in the area showed some certain characteristics for different locations. The clearest difference was between those for fill layers underlined by growacke (or limestone for some locations) and fill layers underlined by thick, green, highly plastic, medium to stiff silty clay layers. The functions for the former site showed slight peaks around smaller periods that were the predominant soil periods, but the functions for the latter sites generally had sharp peaks around the medium period band. The microzones were separated according to similar amplification and predominant soil periods, but these results need to be compared to observed damage patterns in the area. The information on the damage which was experienced during earthquakes in the area, was collected from different sources. Emphasize has been given to those which occurred during the 1894 earthquake and damage zones separated in terms of damage levels. Furthermore, the predominant periods of structures needed to be taken into account in comparisons, and these periods were approximately estimated by empirical relations. The predominant period of structures is important, because the response of a structure can also be increased or decreased depending on its natural period of vibration. The clearest comparison between the predicted amplification functions and observed damage was between the sites with fill layers underlined by growacke or limestone layers, and the sites which contain thick clay layers. No recorded damage to structures has been found in the former sites as it is evident from the amplification functions that the predominant soil periods appeared to be low and no significant amplification occurred within the period range of structures. Most of the damage is concentrated on the latter sites and the amplification functions also supported this fact, having significantly higher values within the period range of structures. The reasons of high amplification concentration particularly in some regions have been discussed and detailed comparisons between amplification properties and earthquake damage for the other sites are presented in the text. In order to investigate the effects on soil amplification properties due to different earthquakes, a number of available strong motion acceleration records have been used scaling the peak acceleration levels.
Açıklama
Tez (Doktora) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 1994
Thesis (Ph.D.) -- İstanbul Technical University, Institute of Science and Technology, 1994
Anahtar kelimeler
Deprem, Deprem davranışı, Erzincan, İstanbul, Earthquake, Earthquake behavior, Erzincan, Istanbul
Alıntı