Yer Radarı İle Karstik Boşluk Araştırmaları

Abstract

Bu çalışmada karbonatlı kayaçların hakim olduğu Konya ilinin Ilgın ilçesinde karstik boşlukların araştırılmasına yönelik yer radarı ölçümleri yapılmıştır. Yer radarı yöntemi, yüksek frekanslı elektromanyetik prensip ile çalışan sığ jeofizik araştırma yöntemlerinden biridir. Teknolojik gelişmelerle birlikte kullanım alanı da genişleyen yer radarı, jeolojik, arkeolojik ve diğer birçok araştırma alanında kullanılmaya başlanmıştır. Karstik bölgelerde insan sağlığı ve jeoteknik açıdan tehlike barındıran bu boşlukların tespiti ile söz konusu alanlara yapılacak mühendislik yapılarının planlaması daha doğru olarak yapılabilmektedir. Bu alanlarda oluşturulacak yapılar yer altında bulunan boşluklarda gerçekleşebilecek çökmeler neticesinde ciddi hasar alma ihtimaline sahiptir. Bu nedenle karstik boşlukların oluşum şekillerinin ve yer altındaki uzanımlarının dikkatle araştırılması gerekmektedir. Yer radarının kolay uygulanabilen bir yöntem olması ve kısa zamanda geniş alanlar tarayabilmesi, bu gibi araştırmalarda yoğun olarak kullanılmasını sağlamaktadır. Çalışma sahasında 100 MHz'lik merkez frekansa sahip anten kullanarak toplamda uzunluğu 7.3 km'yi derinliği 5 m'yi bulan profil verisi elde edilmiştir. Daha önceki çalışmalar dikkate alınarak, çeşitli aşamalardan oluşan veri işlem sonuçları karşılaştırılmalı olarak değerlendirilerek en uygun veri işlem yöntemi belirlenmiştir. Elde edilen radargramlardan toplam uzunluğu 350 metreyi bulan dört tanesi seçilerek yorumlanmış ve boşluklu, çatlaklı ve deformasyonlu bölgeler belirlenmiştir. Yorumlanan verilerde detaylı olarak polarizasyon analizi ve genlik analizi yapılmıştır. Boşluk olan yapılar belirlenerek empedans değerleri üzerinden birbirleriyle olan ilişkileri ortaya konmuştur. Elde edilen sonuçlar bölgede mevcut sondaj ve jeoloji verilerinin ışığı altında değerlendirilerek karstik yapıların modellemesi yapılmıştır. Radargramlarda yorumlanan anomalilerin parametreleri hesaplanarak önce basit modelleme ardından da karmaşık modelleme yapılmıştır. Modelleme sonuçları da radargramlarda yapılan analizleri desteklemektedir. Bu tez çalışması ile karstik bölgelerde yer radarı verilerinin işlenmesinde izlenilecek yaklaşımların belirlenmesi, elde edilen radargramların yorumlanması ve modellenmesi konusunda bilgiler sunan bir kaynak ortaya konulmuştur.

In this study, Ground Penetrating Radar measurements has been made to investigate the karstic cavities in Konya, Ilgın which has limestone geological structure densely. The aim of this thesis is to find optimum ways to process and model ground penetrating radar which is widely used in finding, locating and characterizing karstic structures. Karstic structures are important features in perspective of geology and buildings. The chosen survey location has a potential of being a foundation of a new building. Ground Penetrating Radar (GPR) is one of geophysical methods used in near surface investigations which is based on electromagnetic principles. Radar technology has existed for many decades, but only for last years it has been used for near surface ground investigations. GPR is one of the main near surface investigation methods for subjects such as geology, archeology and civil engineering. Since GPR can be easily applied and can cover large areas quick and efficiently, it is a method used widespread. GPR generally consists of transmitter and receiver antenna, control unit and a recorder. With the help of latest technologies, various designs from different companies are available. There are models with a measurement wheel, special designs for railroads and walls and which the antenna doesn't need to be in contact with the ground. Karst is a term used for specific type of landscape which develops from the dissolving action of water on soluble bedrock. Generally this kind of structures develop in limestone, marble, dolostone, gypsum and halite. They are characterized by cavities and waterways in different sizes and shapes. This landscapes develop in long time scales which can be up to thousands of years. Primarily, this three dimensional distinctive type of landscape develops as a result of surface water traveling into the underground. As a result of this movement, soluble bedrock becomes dissolved and karstic landscapes are formed. Most used geophysical methods for karstic investigations are gravity method, electric resistivity method, seismic method and GPR. With the help of these detecting methods for karstic cavities, it is possible to plan and avoid any hazards which are highly possible in such areas. Karstic cavities develop in such a way that it may not be possible to see any surface indications. However, even a small change in the surface mass can cause a destructive hazard since the caves and openings beneath the surface are highly possible to collapse. Not only it can play a vital role in constructional subjects, it is also very important to understand these underground drainage systems in scope of geothermal and environmental points. Because of these reasons it is very important to investigate the development of karstic landscapes and it's effects under the ground. Taking into consideration that karstic cavities can be fast and efficiently be detected by GPR, measurements were made in Ilgın, Konya, TURKEY. As a first step, the method and the depth have been decided. Investigation was made by a 100 MHz central frequency shielded GPR antenna. Since the topography was not smooth, it has been decided to use a GPS system in order to get coordinates and make needed corrections to the data. Measurements took 4 days and as a result 32 measurements were made. A total of 7290 meters of GPR measurements were recorded. In order to investigate the cavities, it was planned to take the profiles in parallel and perpendicular geometries. However, in several places the measurement geometry had to be changed due to the topographic conditions. In addition, a building was present in measurement site. In order to avoid any noise caused by a construction, measurements near building was not taken into account. Also, profiles which had low quality GPS data were eliminated. After filtering all data available, it was decided to use 3, 4, 23, 27 numbered profiles for interpretation. As a first step in data processing, the data was checked in scope of GPS coordinates and smoothed especially in elevation. After this point the data was prepared for filtering operations. Data processing consists of moving start time, subtracting mean (dewow), gain recovery, band pass filter, running average, background removal, enveloping and topographic correction. As a result, processed radargrams were created and prepared for interpretation. Karstic cavities are explained. In order to make a proper interpretation it is important to understand how to characterize the karstic landscapes. The geology and drilling logs of the field were examined. It has been seen that there are typically 2 layers beneath the surface which can vary in depth. With the help of drilling logs it has been seen that there could be cavities and deformation zones under the 0,3 m level. As a next step filtered radargrams are interpreted and correlated with their enveloped forms. Enveloping gave good results due to the fact that there should be high scattering values since the subsurface is karstic. Profile 3 and profile 23 has anomalies interpreted as collapsed limestone. This collapsing movement is starting from 1 meter and going deep until 3 as a total of 2 meter. Since this collapse is right on the limestone it is interpreted that this anomaly can be a form of solution in the karstic section of the subsurface. Also in profile 3 and 4 air filled cavities have been detected. This analysis was made by the anomalies' polarization. It has been showed that if the anomaly has the same polarization as the direct wave from the air, then it can be said that the target has a lower dielectric value than the media. On the other hand a water filled cavity has been interpreted on profile 23. This analysis was made by a reverse polarized anomaly. After this stage, several modeling attempts have been made. First, 2-3 layered simple models have been calculated to understand the subsurface. Then, more complex structured models were calculated to support the interpretations of the radargrams. In this stage profile 3, 4 and 23 chosen since they have the most obvious anomalies to model.In the modeled Profile 3 the collision zone at the 25 m distance was successfully modeled. For Profile 4 the modeling process consists of a cavity structure. In the comparison of modeled and collected data it has been seen that the polarization approach has been justified and an air filled cavity created a normal polarized anomaly. Profile 23 has been modeled in two sections. In the first section, a similar approach as used in Profile 3 was implemented and the deformation zone has been successfully modeled. In the second section, a water filled cavity has been modeled with the dielectric value of 81. This value created a reverse polarized anomaly.