Deprem sırasında oluşan yer hareketinin genliğinde, frekans içeriğinde ve süresinde incelenen bölgeye bağlı meydana gelen değişimlere "yerel etkiler" denir. Yerel etkiler; başlıca, yüzeydeki zemin tabakalarıyla daha altta yer alan anakaya arasındaki özgül direnç farklılığından kaynaklanmakla birlikte, ova/vadi gibi derin oluşumlar ile bunların kenarlarında yanal jeolojik süreksizlikler tarafından meydana getirilen etkileri de kapsar. Yanal süreksizliklere sahip bu tür ortamlarda zemin tabakalarının yatay yönde sınırlı genişliğe sahip olması, vadi kenarlarında dalga hareketi dönüşümlerine sebep olmakta ve kuvvetli yer hareketinin süresi uzayabilmektedir. Bu tip durumlarda, yüzeydeki hareketin genliği ova/vadi üzerinde elde edildiği bölgeye göre değişim gösterecek ve frekans muhtevası aynı bölgede bir boyutlu analizle hesaplanana göre farklılaşabilecektir. Bu çalışmada, iki boyutlu kayma dalgası hızı profili geniş açıklıklı eş zamanlı mikrotremor ağ ve tekil mikrotremor ölçümleriyle belirlenmiş olan Düzce Ovası modelinde, kenardaki anakaya eğiminin değişmesi halinde farklı frekans içeriğine sahip deprem hareketleri karşısında yüzey hareketinin nasıl etkileneceği anlaşılmaya çalışılmıştır. Bu amaçla, modellerde farklı noktalar için bir (İD) ve iki boyutlu (2D) dinamik analizler yapılmış, ova kenarındaki zemin büyütmelerinin ve şiddet parametrelerinin geometriye bağlı değişimi incelenmiş, ova kenarındaki kısımlar için 2D/1D spektral ivme oranları (şiddet faktörleri) hesaplanmıştır. Farklı periyotlar için hesaplanan bu değerlerin istatistiksel açıdan değerlendirilmesiyle, farklı anakaya eğimine sahip modeller için ova kenarındaki anakaya mostrasından olan uzaklığa ve sismik anakaya derinliğine bağlı şiddet faktörü ilişkisi tanımlanmış ve bir boyutlu dinamik analizin yaklaşık geçerlilik sınırları belirlenmiştir.

Surface geology and the geotechnical properties of near surface soil layers have important effects on the earthquake ground motion. Site effects can be defined as the variation in the characteristics of incoming wavefield such as amplitude, frequency content and duration because of the geotechnical, dynamic and geometrical properties of soil layers and surface topography. The main source of site effects is the specific impedance contrast between near surface and deep soil layers. The calculation methods which were developed for the dynamic analysis of soil layers are defined as one, two and three dimensional. In one dimensional approach the soil medium is assumed as horizontally layered for simplicity and the analyses are based on the principal of body waves travelling up and down in the near surface layers. However, in fact sedimentary deposits form mediums which can only be defined by 2 or 3 dimensional methods. This hind of deposits with lateral geological discontinuities show trap behaviour. This trap affects the surface waves which develop during earthquakes and reverberate back and forth on the interface of sediment deposit and bedrock. As a result, the amplitude of surface ground motions may show variation dependent on the site where it occurs, also the frequency content of this surface motion will differ from site to site at the edge of deep deposits. In this paper, two dimensional basin edge models with four different slope values (H/D=I0, 5, 2, 1) were constituted to investigate the effects of basin edge on the variation of surface motion under earthquake excitations with different frequency content. Duzce basin shear wave velocity profile, which had been obtained from the analyses of microtremor array and single point measurements, was used to form the two dimensional geometry of soil layers and bedrock in the basin edge models. One (ID) and two dimensional (2D) dynamic analyses were performed for six different bedrock acceleration records by using these basin edge models. The results which would be obtained from the ID and 2 D analyses were aimed to reflect the seismotectonical structure of the faults in Turkey. Therefore four bedrock acceleration time histories were selected among the Turkey earthquakes. In order to eliminate the "box effects " which can occur during the two dimensional dynamic finite element analyses because of the model geometry and boundary conditions; viscous dashpots, which are calculated proportional to the shear and pressure waves of the relevant layers, were put at the vertical and horizontal layers. Modified equivalent linear method and equivalent linear methods were used in the ID and 2D analyses respectively. After performing ID and 2D dynamic analyses on the models, the calculated results were compared. The variations of the soil amplifications and acceleration spectrum intensity (ASI) values with the distance from the valley edges were investigated. To understand the difference between ID and 2D dynamic behaviours, the 2D/ID spectral acceleration ratios which are known as "aggravation factors", were calculated for different period values. Remarkable increase in soil amplification, ASI and aggravation factors was observed between the rock outcropping site and X/D=3 point at basin edges. The highest aggravation factor values were obtained at 0.2 sec-0.5 sec interval for all valley models, also especially for models with H/D= 2 and I these values reached to peak. Regardless of the periods considered, the aggravation factors generally converged to 1 after the points X/D=5, 4, 2, 1.5 for the basin models with H/D-10, 5, 2 and 1 respectively. The 2D effects were mostly becoming negligible after these points. After this, the acceleration time histories and absolute acceleration spectrums which had been evaluated for the different points on edge surface were considered statistically and the relation between the results of ID and 2D dynamic analyses was investigated by calculating aggravation factors for different period values. After evaluating the aggravation factors for different periods statistically, an aggravation factor relation was defined for the basin edge models with different edge slope values^ It was revealed that by using the 2D/1D aggravation factor relation, the limiting effect of the second dimension at basin edges can be reflected to the spectral accelerations evaluated from the results of one dimensional dynamic analysis, as a function of the distance from rock outcrop site and seismic bedrock depth. Also the X/D values, which the aggravation factors can be neglected after, were determined.