Intensive exploitation of the reservoir at The Geysers geothermal area, California, induces myriads of small-magnitude earthquakes that are monitored by a dense, permanent seismometer network that covers most of the reservoir. However, majority of the seismic stations, which belong to the UNOCAL network are poorly calibrated. Station polarities, and sensor orientations for the 8 three-component stations of this network were determined by using accurate focal mechanism solutions from a temporary network and using a simple method of observing the waveform from known earthquake locations. Using data from the UNOCAL network, tomographic inversions were performed for the three-dimensional Vp and Vp/Vs-ratio structure of the reservoir for February 1993, October 1996 and August 1998, adding to the inversions for April 1991 and December 1994 that had already been performed by other investigators. The extensive low -Vp/Vs anomaly known to characterise the reservoir grew progressively in strength from a maximum of 9% to a maximum of 12.4% at sea level during the seven-year study period. The anomaly growth is attributed to the depletion of pore liquid water in the reservoir and its replacement with steam. This causes Vp to decrease by increasing compressibility, and Vs to increase because of the reduction in pore pressure and the drying of argillaceous minerals, e.g., illite, which increases the shear modulus. All these effects serendipitously combine to lower the Vp/Vs ratio, resulting in an exceptionally strong overall effect that provides a convenient tool for monitoring reservoir depletion in the seismogenic zone. Variations in the separate Vp and Vs fields indicate that water depletion was the most important process in the central part of the exploited reservoir, and that pressure reduction and mineral drying were the dominant effects more northwesterly and southeasterly. Relative relocation of micro earthquakes was also performed using the same network. Four regions were studied. Although most multiplets relocated into tighter clusters and the reduction in the RMS of the relative relocations was good, further work is needed to substantiate these initial findings.

Data used for the present study were recorded at the small aperture cross-linear array station which was installed at Kaptagat (in NW Kenya) by the University of Durham. The seismic array data from local earthquakes have been analysed by velocity/azimuth filtering technique. Apparent velocities and azimuths for first and later arrival phases were measured for local rift events from the immediate east, for local events from the south west and for more distant rift events to the north and south of Kaptagat. Data from local rift events originating from the immediate east of Kaptagat were used in the present analysis to study the structure of the lithosphere beneath the Gregory rift at about 0.5 N latitude. The first arrival data (apparent velocities and azimuths) were determined to a high degree of accuracy. The first and later arrival data have been interpreted in terms of a simple two layer model with a horizontal refracting interface at a depth of 13 + 5 km and having upper and lower layer uniform velocities of 5.8 + 0.2 km/ s and7.2 + 0,2 km/s respectively. The minimum lateral extent of the top surface of this refractor is estimated at about 30 km. A maximum dip of about 6 on the interface is allowed by the data. In the preferred three layer model, a 10 km thick top horizontal layer of velocity 5.8 km/s overlies a 10 km thick intermediate layer in which velocity increases uniformly from6.0 km/s at 10 km depth to 7.5 km/s at a depth of 20 km. The intermediate layer, in turn, overlies a 7.6 km/s refractor. The models derived from the present data are consistent with the theory that upward perturbation of the lithospher asthenosphere boundary giving rise to domal uplift, lithospheric tension and magmatic activity, is the primary causeof rifting.

The Geysers geothermal area, California is the world's largest and most intensively exploited steam field, providing about 6% of California's electrical power. The geothermal area is very active seismically, generating about 140 earthquakes per month with M(_D)≥l.2. Non-DC earthquakes have been routinely detected in other geothermal and volcanic areas such as the Hengill-Grensdalur volcanic complex, Iceland but previously went undetected at The Geysers. The steam field is, however, a likely source of non-DC earthquakes because large volumes of steam are extracted and condensate injected during the course of commercial exploitation which might cause cracks or fractures to open and close. Maps of seismic activity through time show conclusively that earthquakes initiate at the onset of production, continue through it and stop when production ceases. Furthermore the volume of steam extracted and/or condensate injected may directly control the rate of seismicity within the geothermal area. A temporary field experiment in April, 1991 recorded about 4000 high-quality earthquakes on three-component digital sensors. Three-dimensional tomographic models of V(_p) and, for the first time at The Geysers, V(_p)/v(_s) were determined using 3906 P-wave and 944 5-wave arrival times from 185 earthquakes. Variations m lithology, temperature and the pore-fluid phase probably produce the variations in V(_p). A strong low in the V(_p)/v(_s) model defines fluid-deficient areas in the steam reservoir and is surrounded by a "halo" of high-V(_p)/v(_s) anomalies. V(_s)/v(_s) can remotely monitor temporal depletion of liquid reserves in the steam reservoir. Well-constrained moment tensor solutions for 30 earthquakes were determined by inverting the polar' ices and amplitude ratios of P- and 5-wave arrivals. Strong evidence for the existence of non-DC earthquakes at The Geysers geothermal area was found. Explosive and implosive events occur in equal numbers and probably reflect source processes involving opening and closing cracks or cavities. The events form a symmetric pattern on source-type plots extending from the postive dipole-to-negative dipole loci, passing through the DC locus. The association with dipole loci rather than crack loci suggests the source process must also involve a compensating flow of fluids, liquid or steam.

This thesis describes the interpretation of the deep seismic structure from the wide-angle reflection/refraction Caledonian Suture Seismic experiment which extends from the Mid North Sea High across Northern England, the Irish Sea to southwest Ireland. A new method of displaying the large number of travel times in the form of contoured composite plots for a particular phase has been developed. The interpretation methods used include analysis of the wide-angle reflection travel times, time-term analysis, gravity modelling, and, ray tracing and synthetic seismogram modelling. A mid-crustal gradient which returns PcP occurs between depths of 15.0-18.5 km beneath the line and is overlain by an upper crust with an average velocity of 6.16-6.20 km/s excluding sediments. This mid-crustal gradient was not recognised previously in the Irish Sea. A lower-crustal boundary recognised from the PmP phase occurs at 25.0 km depth beneath the Irish Sea and at 30 km beneath the North Sea. The average crustal velocity is 6.49-6.54 km/s excluding sediments and for the lower crust is 6.75-6.77 km/s. The sub-Moho Pn velocity is estimated to be 8.19 km/s beneath the Irish Sea, 8.32 km/s beneath Northern England and 8.02 km/s beneath the North Sea. The PmP and Pn boundaries diverge beneath the Irish Sea and define a transitional Moho over a 5 km depth range with an average velocity of 7.8-7.9 km/s. The PcP and Pn boundaries correspond approximately with the top and base respectively of the reflective deep crust observed beneath the Irish Sea on BIRPS WINCH. The transitional Moho beneath the Irish Sea has a reflective character similar to the lower crust. The coincident PmP and Pn boundaries beneath the North Sea define an abrupt increase in velocity from 6.75 to 8.0 km/s which approximately corresponds to the prominent reflections beneath the non reflective lower crust observed on the BIRPS NEC line at 10.5 to 11.0 s .The upper crust of velocity 6.15-6.20 km/s appears to lie below the suture beneath Northern England and the North Sea along the line but may also occur above it further west. The lower crust and deeper structures lie below the suture and represent crust of the Southern Caledonides. The anomalous structure of the Moho and upper mantle beneath the Irish Sea found from CSSP and BIRPS may be related to the post-Caledonian formation of the Irish Sea basins. The reflective lower crust and transitional Moho may result from crustal thinning produced by ductile stretching.

Iceland is a sub-aerial part of the mid-Atlantic Ridge which has formed above an E migrating ridge centred hotspot. The Hengill area is a ridge-ridge-transform triple point that contains a central volcano-fissure swarm system and a large geothermal area. A seismological study of this triple point was conducted with the main emphasis on natural earthquake studies. The aims were to study the geothermal prospect and tectonic structure and to evaluate the passive seismic method as a geothermal prospecting tool. The area exhibits continuous small magnitude earthquake activity that correlates positively with surface geothermal displays, and negatively with surface faulting. The log (cumulative frequency) magnitude relationship is linear and indicates a b value of 0.74 ± 0.06. Focal mechanisms for 178 events indicated both shear and tensile crack type movements, the latter being confined to the high temperature geothermal area. Teleseismic and explosion data indicate a low velocity body beneath the central volcano in the depth range 0 - 10 km, flanked by higher velocity bodies to the W and E. Two volcanic systems occupy the Hengill area : the presently active Hengill system and the extinct Grensdalur system. The ongoing seismicity of the area is attributed to contraction cracking due to the action of cool groundwater fluids on hot rook, which, in a tensile stress regime, results in tensile crack formation. The high temperature area is fuelled by two heat sources associated with the two volcanic systems and may be divided into two separate fields that exhibit contrasting reservoir characteristics. Local seismioity studies may be applied to other Icelandic high temperature geothermal areas as a tool to map those volumes of rock that are fueling the geothermal reservoirs. The continuous formation of small tensile cracks on accretionary plate boundaries offers an explanation for the mechanism of dyke injection.

The structure of oceanic crust is surprisingly uniform, which suggests that crustal accretionary processes at mid-ocean ridges must be broadly similar, despite their different spreading rates and seafloor morphologies. Seismic studies have revealed the presence of sub-axial magma chambers at fast, slow and intermediate spreading ridges, but constraints on their shape and size are generally restricted to the fast spreading East Pacific Rise. The aim of this study is to compare the processes of crustal accretion at fast, slow and intermediate ridges by investigating the detailed crustal structure and magma chamber geometry of a magmatically active intermediate spreading ridge, the Valu Fa Ridge. A multidisciplinary geophysical experiment was conducted over the Central Valu Fa Ridge and its overlap with the Northern Valu Fa Ridge during R/V Maurice Swing Cruise EW9512, and wide-angle seismic data, recorded on a set of digital ocean bottom seismometers, were used to generate velocity-depth models on two across-axis, two along-axis and two axis-parallel profiles. These models were further constrained by modelling of the normal incidence seismic and gravity data and the resulting combined models of crustal structure were interpreted to reveal that a composite magma chamber exists beneath the Valu Fa Ridge crest. The magma chamber consists of a thin, narrow (1-1.5 km) melt lens, with an interconnected melt fraction, overlying a wider (-4 km) region of hot rock or low melt fraction. A reflection from the top of the melt lens is identified on both the normal incidence seismic and wide-angle seismic data and delay- time modelling indicates that velocities as high as 5.5 km s(^-1) are achieved -250 m below the top of the melt lens. The main body of the magma chamber corresponds to the region of hot rock below the melt lens and is delineated by anomalously low velocities, extending down through seismic layer 3 to within 1.5-2 km of the Moho. Moho reflections from beneath the overlapping spreading centre and a low on the mantle Bouguer anomaly map implies that this region is currently, or has recently been, the site of enhanced magmatism. This observation is contrary to popular models of ridge segmentation and melt delivery. The transition from pre-rift crust (both island arc and back-arc crust) to post-rift material, marked by considerable thinning of seismic layer 2, has also been uniquely identified in this study and describes the limit of VFR-generated crust. The size and temporal stability of magma chambers are largely dependent on their magma budget and the Valu Fa Ridge magma chamber model, developed as part of this study, may bridge the gap between the large, long-lived magma chambers identified at the East Pacific Rise and the more transitory magma chambers proposed at slow spreading ridges. Melt ascends as small isolated pockets through the main body of the magma chamber at the Valu Fa Ridge and resides in the melt lens until eruption. Seismic layer 2 is constructed solely from material 'erupted' from the melt lens, with the main body of the magma chamber cooling to form seismic layer 3. Convection currents, induced by large thermal gradients at the sides of the magma chamber, both accelerate the cooling process, thus limiting its size, and helps to generate the thick layered sequences as observed in ophiolite studies. The entire crust is emplaced within the axial region and a distinct Moho is formed at -0 Ma.

In November 1979, the Department of Geological Sciences, University of Durham, in conjunction with the Department of Geology, University of Glasgow, carried out a marine seismic refraction survey between Barra In the Outer Hebrides and Girvan on the Ayrshire coast. In August/September 1981, a shorter profile, between Mull and Kintyre, was undertaken. Temporary recording stations were set up on land and explosives and alrguns used as sources at sea. In addition, data were obtained from the permanent recording networks in Scotland. The application of digital filtering techniques to the alrgun lines Is presented and reviewed. The explosive shot data were interpreted using time-term analysis, the plus- minus method and ray-tracing. Gravity and magnetic studies were used as a control on the interpretation. The depth to the basin has been examined, indicating sedimentary cover of between 1 and 3 km along the line. The variation in basement velocity has been determined. A velocity model for the upper crust has been developed indicating large lateral changes. A steep gradient Is thought to exist between Golonsay and Jura and a more gentle gradient beneath Mull. There does not seem to be evidence for a mid-crustal refractor with a sharp velocity transition across the boundary. Large changes In Pn time-terms across the Minch and Inner Hebrides basin are thought to be largely due to the varying velocity structure of the basement. The basin formation and deformation of the lower crust are thought to Involve movement within the mantle.

In June 1987, during the BIRPS MOBIL normal-incidence seismic profiling programme, off the East coast of England, the University of Durham recorded simultaneously at several land based seismic stations in Northern England. The resulting wide-angle data, particularly from Line 1, have excellent resolution in both space and time due to the airgun source and 50 m shot spacing. The interpretation of the Line 1 wide-angle data at Durham used BEAM87, Cerveny's Gaussian beam modelling package. The main arrivals interpreted include the upper crustal refraction (Pg), the Moho wide-angle reflection (PmP), the upper mantle refraction (Pn), and a very high amplitude arrival (D) which merges into PmP. Modelling gave a crust about 30 km thick with a change in velocity gradient and a slight velocity contrast at about 20 km depth. There are several wide-angle reflections from interfaces at mid-crustal depths, between 10 and 20 km depth, and the bottom 2 km of the crust has a high velocity of about 7 kms(^-1). Two interesting results are that a lateral velocity change about 40 km offshore is required to fit the Pg travel times; also that arrival D is modelled best as the remnant of a step on the Moho at the same location. These appear to be borne out by the normal incidence data for line 1, which show a lateral decrease in the mid-crustal reflectivity above a set of strong, westerly-dipping reflections at Moho depths. These results suggest the presence of a major crustal fault about 40 km offshore. It is suggested that this fault may be the northward continuation of the Dowsing Fault Zone.

In recent years there have been significant advances in the acquisition and processing of crosshole seismic reflection data, and the method has been shown to be a high resolution imaging technique. However, the fidelity of the final images produced by this technique needs to be considered carefully to avoid incorrect interpretation. This thesis concerns the imaging capability of crosshole surveys, as well as advances made in processing techniques for application to crosshole seismic reflection surveys. In a migrated seismic section, a meaningful image is only obtained if a range of dips around the local structural dip is sampled at each image point. For crosshole seismic reflection surveys, the distribution of dips sampled at an image point is controlled principally by the survey geometry, including source and receiver array lengths and their element spacings. By considering the dips sampled, the imaging capability of crosshole reflection surveying is discussed, with suggestions as to how to ensure optimal imaging of the target zone. To overcome problems encountered in applying standard processing procedures, two new processing techniques are presented which enhance the imaging potential of crosshole reflection seismics. Generalised Berryhill migration has been developed as a full generalised Kirchhoff migration to include the near-field term, with the aim of improving image accuracy close to the source and receiver arrays. 3-D f-k-k filtering is an improved method of wavefield separation for crosshole seismic data. Finally, the results of processing three types of dataset are presented. One is from a site in the Groningen gas field, another was acquired through a model interrogated at ultrasonic frequencies in a water tank, and the third type was acquired using coal exploration boreholes in Yorkshire. The results demonstrate the imaging capability of the crosshole reflection method, and the success of the two new processing schemes.

A network of four independently-recording seismic stations was operated by the University of Durham in South-Central Afar during 1973 and 1974. Each station consisted of a three-component set of seismometers, whose signals were recorded on to magnetic tape. This study concerns local earthquakes recorded from February to September, 1974.250 earthquakes were located from relative arrival times of P and S phases using an optimized, laterally homogeneous, 4-layer structural model. Upper crustal P-wave velocities are found to be 4.4±0.2 km s(^-1) (0 to 4.5 km depth) and 6.2± 0.1 km s(^-1) (4.5 to 11 km). Deeper structure is poorly constrained. Anomalous upper mantle exists, with low seismic velocity (Vp about 7.4 km s(^-1)) and raised Poisson's ratio (0.31). S(_n) is transmitted, 8.0 km s(^-1) upper mantle cannot exist above about 43 km depth. Earthquake focal depths within Afar do not exceed 5 km. Epicentres correlate well with Recent axial volcanism. Spatial epicentral patterns reflect intense regional NW-SE extensional faulting. One line of epicentres shows the NNE-SSW trend of the Main Ethiopian rift. Focal mechanisms are very poorly constrained, but are consistent with NW-SE strike-slip or normal faulting, or with NE-SW dextral transcurrent faulting. Signal duration magnitude and Richter local magnitude scales are defined for Afar, Frequency-magnitude b-coefficient values are 0.87+0.05, The three-component records are polarization filtered, a technique previously applied only to teleseisms. The performance of the filters is discussed. Azimuths and apparent angles of incidence of events are determined from their first arrivals at a single recording station. Hypocentres are then obtained by ray tracing. Earthquake frequency spectra are computed through the fast Fourier transform. The spectra are dominated by the effects of the superficial crust below the receivers. Crustal transfer ratios are discussed. Increased attenuation is demonstrated below the Tendaho graben. Seismic source parameters are calculated using BRUNE's (1970) method. All results are consistent with diffuse NE-SW crustal extension. It is concluded that well-defined spreading axes do not yet exist.