During 1979, an oil prospecting project was initiated in the south of Sweden, outside the coast of the municipality of Skåne. No drilling program was initiated, and the project was put on ice. However, with the oil prices of today and the interest in carbon dioxide storage, the area could once again be interesting. In this thesis, 3 seismic lines have been reprocessed with new software, Claritas, in an attempt to obtain information of the area. The thesis is also thought of as an easy to read tutorial as to how oil prospecting works. The result of the reprocessing showed that the new software could reduce multiples and high- and low frequency noise with an fx-decon filter. / Under 1979 inleddes ett projekt med oljeprospektering utanför Skånes kust i södra Sverige. Inget borrprogram initierades och projektet lades ned. Med dagens oljebrist och intresset för koldioxidförvaring så skulle området än en gång kunna vara intressant. I detta examensarbete har därför 3 seismiklinjer omprocesserats med ny programvara, Claritas, i ett försök att förbättra vår bild av detta område. Examensarbetet är även tänkt att fungera som en inledande litteratur om hur oljeprospektering fungerar. Resultatet av omprocesseringen av seismikdatat visade att man med den nya programvaran kunde reducera multipler och hög- och lågfrekvent brus med ett fx-decon filter.

Reflection seismic and seismic tomography methods can be used to image the upper kilometer of hard bedrock and the loose unconsolidated sediments covering it. Developments of these two methods and their application, as well as identifying issues concerning their usage, are the main focus of the thesis. Data used for this development were acquired at three different sites in Sweden, in Forsmark 140 km north of Stockholm, in the Oskarshamn area in southern Sweden, and in the northern part of the Siljan Ring impact crater area. The reflection seismic data were acquired with long source-receiver offsets relative to some of the targeted depths to be imaged. In the initial processing standard steps were applied, but the uppermost part of the sections were not always clear. The longer offsets imply that pre-stack migration is necessary in order to image the uppermost bedrock as clearly as possible. Careful choice of filters and velocity functions improve the pre-stack migrated image, allowing better correlation with near-surface geological information. The seismic tomography method has been enhanced to calculate, simultaneously with the velocity inversion, optimal corrections to the picked first break travel times in order to compensate for the delays due to the seismic waves passing through the loose sediments covering the bedrock. The reflection seismic processing used in this thesis has produced high-quality images of the upper kilometers, and in one example from the Forsmark site, the image of the uppermost 250 meters of the bedrock has been improved. The three-dimensional orientation of reflections has been determined at the Oskarshamn site. Correlation with borehole data shows that many of these reflections originate from fracture zones. The developed seismic tomography method produces high-detail velocity models for the site in the Siljan impact area and for the Forsmark site. In Forsmark, detailed estimates of the bedrock topography were calculated with the use of the developed tomography method.

Controlled source seismic methods were employed in this study to investigate the reflectivity and velocity structure of two Hercynian orogens – the Uralides and Variscides. Conventional common depth point (CDP) sections from five reflection seismic campaigns and a velocity model obtained from tomographic inversion of wide-angle observations were the main datasets studied from the Middle Urals. These were complemented with the near-vertical seismic sections and velocity models from the Southern Urals. In the Variscides, conventional CDP processing, along with non-standard processing and synthetic data modeling, were used to obtain and interpret reflection seismic images of the Southwestern Iberian crust. Although, the Uralian and Variscan belts were formed in Late Paleozoic time in apparently similar plate collisional settings, a comparison of the seismic results show that the crust of these two orogens looks quite different at depth. In the Urals, collision of Baltica with Asian terranes (Siberia and Kazakhstan) resulted in a highly diversely reflective crust of 40-45 km thickness. The axial zone of the orogen is characterized by a high velocity crustal root of diffuse reflectivity and an imbricated Moho, with a crustal thickness reaching 55-60 km. The Moho discontinuity is marked by a sharp decrease in reflectivity and is well imaged in most locations except in the crustal root zone. The Southwestern Iberian Variscan crust is 30-35 km thick and is characterized by a highly reflective two-layered structure that resulted from collision of Luarussia and Gondwana, including terranes in-between them. This type of crustal structure is very similar to those imaged in other regions of the Variscan belt in the Europe. The Moho discontinuity is flat and appears to be the deepest reflection. This thesis compares the deep structure of the two orogens and interprets mountain building processes related to late Paleozoic plate movements.

<p>Reflection seismic and seismic tomography methods can be used to image the upper kilometer of hard bedrock and the loose unconsolidated sediments covering it. Developments of these two methods and their application, as well as identifying issues concerning their usage, are the main focus of the thesis. Data used for this development were acquired at three different sites in Sweden, in Forsmark 140 km north of Stockholm, in the Oskarshamn area in southern Sweden, and in the northern part of the Siljan Ring impact crater area.</p><p>The reflection seismic data were acquired with long source-receiver offsets relative to some of the targeted depths to be imaged. In the initial processing standard steps were applied, but the uppermost part of the sections were not always clear. The longer offsets imply that pre-stack migration is necessary in order to image the uppermost bedrock as clearly as possible. Careful choice of filters and velocity functions improve the pre-stack migrated image, allowing better correlation with near-surface geological information.</p><p>The seismic tomography method has been enhanced to calculate, simultaneously with the velocity inversion, optimal corrections to the picked first break travel times in order to compensate for the delays due to the seismic waves passing through the loose sediments covering the bedrock.</p><p>The reflection seismic processing used in this thesis has produced high-quality images of the upper kilometers, and in one example from the Forsmark site, the image of the uppermost 250 meters of the bedrock has been improved. The three-dimensional orientation of reflections has been determined at the Oskarshamn site. Correlation with borehole data shows that many of these reflections originate from fracture zones. </p><p>The developed seismic tomography method produces high-detail velocity models for the site in the Siljan impact area and for the Forsmark site. In Forsmark, detailed estimates of the bedrock topography were calculated with the use of the developed tomography method.</p>

Seismic methods have been used in sedimentary environment for almost 80 years. During that time, exploration geophysicists have developed a number of techniques to handle specific aspects of working in sedimentary areas. This is not the case for studies in the hardrock environment, where significantly less time and money have been invested on seismic investigations. Therefore, there is still a need to develop the right techniques appropriate for working in hardrock environments. The research presented here, covers aspects of acquisition, processing and interpretation in hardrock environments. A cost-effective alternative for two-dimensional data acquisition is presented. Acquisition parameters are also discussed and recommendations for future work are given. The main effort of this thesis, however, was to find appropriate processing methods to address some of the different problems present in datasets acquired in the hardrock environment. Comparison of two computer programs for first arrival seismic tomography was performed in order to find the most suitable one for processing crooked line geometries. Three-dimensional pre-stack depth migration was also tested to find a detailed near-surface image. A processing method geared to enhance the signal-to-noise ratio was applied to the dataset with the lowest signal amplitudes to improve the quality of the stack. Finally, cross-dip analysis and corrections were performed on two of the three datasets included in this thesis. Cross-dip analysis was also applied as an interpretation tool to provide the information needed for estimation of the true dip of some of the reflectors related to geological structures. The results presented in this thesis indicate that cross-dip analysis and corrections are one of the most powerful tools for processing and interpretation in the presence of complex geology. Therefore, it is recommended to include this method as a standard step in the processing and interpretation sequence of data acquired in hardrock environments.

<p>Seismic methods have been used in sedimentary environment for almost 80 years. During that time, exploration geophysicists have developed a number of techniques to handle specific aspects of working in sedimentary areas. This is not the case for studies in the hardrock environment, where significantly less time and money have been invested on seismic investigations. Therefore, there is still a need to develop the right techniques appropriate for working in hardrock environments. The research presented here, covers aspects of acquisition, processing and interpretation in hardrock environments. A cost-effective alternative for two-dimensional data acquisition is presented. Acquisition parameters are also discussed and recommendations for future work are given. The main effort of this thesis, however, was to find appropriate processing methods to address some of the different problems present in datasets acquired in the hardrock environment. Comparison of two computer programs for first arrival seismic tomography was performed in order to find the most suitable one for processing crooked line geometries. Three-dimensional pre-stack depth migration was also tested to find a detailed near-surface image. A processing method geared to enhance the signal-to-noise ratio was applied to the dataset with the lowest signal amplitudes to improve the quality of the stack. Finally, cross-dip analysis and corrections were performed on two of the three datasets included in this thesis. Cross-dip analysis was also applied as an interpretation tool to provide the information needed for estimation of the true dip of some of the reflectors related to geological structures. The results presented in this thesis indicate that cross-dip analysis and corrections are one of the most powerful tools for processing and interpretation in the presence of complex geology. Therefore, it is recommended to include this method as a standard step in the processing and interpretation sequence of data acquired in hardrock environments. </p>