NS22A-01 INVITED
Massive Bedforms and their Movement Mapped at the Mouth of San Francisco Bay using Multibeam Sonar
Seafloor mapping was completed at the mouth of San Francisco Bay over 44 days in Fall 2004 and 2005 using a Reson 8101 multibeam sonar system. The survey, which consisted of over 1.1 billion soundings, revealed a field of giant sand waves, among the largest in the world (n= 40, max wavelength= 220 m, max amplitude= 5 m, max depth= 106 m), immediately outside the Bay entrance. Seven subsequent surveys over the heart of this bedform field reveal daily and weekly meter-scale shifts in sand wave shape and position that allow for calculations of bed sediment transport rates. The analysis and interpretation of this data set is important for a number of applied problems. For example, identifying pathways of sediment transport is essential for coastal managers in this region who direct dredging operations and seek the most efficient means of mitigating erosion on adjacent Ocean Beach. Also, determining volumetric changes of sediment at the mouth of San Francisco Bay over the last century is an essential step in assessing a sediment budget for the San Francisco Bay system, and the long- term anthropogenic influence of sand mining inside the bay, hydraulic mining in the Sierra Foothills, Bay development causing tidal prism reduction, and dredging practices inside and outside the bay. Recent advancements in the use of multibeam sonar combined with Differential Global Positioning Systems (DGPS) allow mapping of the seafloor with very high resolution, (1 million depth points/ square kilometer), speed (3000 soundings/sec), and accuracy (decimeter precision in the horizontal and vertical planes). These advances greatly broaden the application of bathymetric surveys to include substrate identification (especially with the collection of backscatter data), detailed bedform characteristics, relative wave and hydrodynamic influences, interpretation of 2-D and 3-D flow structure, and hydrodynamic and sediment transport modeling.
http://walrus.wr.usgs.gov/coastal_processes/
NS22A-02
Combining Refraction Seismic Tomography and Analysis of Wide-Angle Reflections in Geothermal Exploration
I-GET is an European Union funded project to develop integrated exploration methods for fractured and/or fluid bearing geothermal reservoirs. The methods are tested at exemplary sites in volcanic, metamorphic and sedimentary environments. We present results from the geothermal site at Gross Schoenebeck which is located within the Northeast German basin. The target horizons for a middle enthalpy geothermal system consist of Rotliegend sandstones and volcanic rocks at depths between 3.8 and 4.3 km. Seismic and magnetotelluric methods are used to explore the sedimentary and tectonic setting, and to identify regions of high porosity and potential fluid pathways. We show results from a 40 km long seismic profile centered at the well site at Gross Schoenebeck. Refraction seismic tomography provides new insights into the relationship between the stratigraphy and the deeper volcanic structures. This information can improve our understanding of the host formation genesis. Reflection seismic imaging reveals a zone of bright reflectivity within the geothermal target area which is associated with strong wide angle reflections. Based on the results from the tomography and the wide-angle reflection offset-dependent effects, we test models in order to constrain the lithology and pore space properties of the reflective zone.
NS22A-03
Imaging the Shallow Subsurface for a New Underground Subway Line in Barcelona
A detailed characterization of the shallow subsurface by seismic techniques has been used to aid horizontal drilling of tunnels for a new subway line in Barcelona (Spain). Seismic data acquisition in densely populated cities is very difficult. The street layout determines the geometry of the seismic data acquisition experiments. The instrumentation (source and receivers) can not always be located on the surface projection of the tunnel trace, therefore, pseudo 3D acquisition is required deploying the instrumentation were it is possible. Furthermore, the shallow subsurface features extremely heterogeneous �weathered� layer of variable thickness (building foundations, sewage system, water supply conductions, etc), and the background noise is very high (car traffic, electricity lines, etc). Relatively old cities also lack a detailed geological control of the subsurface. The seismic data was acquired using as source a 8 s long Vibroseis sweep which provided relatively good S/N ratio. The shallow subsurface featured an extremely heterogeneous weathered layer characterized by very low seismic velocities (1000-1300 m/s) of variable thickness that made difficult to interpret the seismic reflection images at target depth. The first-arrival seismic tomography revealed as a key tool to clearly differentiate the different rock volumes characterized by different physical properties, especially the contact between the weathered layer and the more competent rock. Several high velocity anomalies (up to 5500 m/s) were observed at tunnel depth that were correlated with sub vertical porphyric dykes. This were surrounded by low velocity anomalies that correspond to fault systems that cut and displace dykes. The tomographic velocity models also provide a new image of an important Miocene fault, quite different to the previous geological section derived from the core interpretation.
NS22A-04
Geological Model of Potential Rockslide Based on Structural Mapping, Surface Geophysical Data and Borehole Logging at Aaknes, Western Norway
Unstable rock slopes possess a threat to the inhabitants along Norwegian fjords, where pre-historic and historic rock avalanches have caused tsunamis, some causing severe casualties (Blikra et al. 2005). The presented site, Aaknes, is a large potential landslide of minimum 30 - 45 million m3 rock mass. Continuous creep of the rock mass and the fact that Aaknes is situated in the vicinity of one of Norway's most visited tourist attractions � the Geiranger fjord, listed on the UNESCO`s World heritage list, have triggered a comprehensive mapping program. The overall aim is to assess the likelihood that the landslide accelerates into a rock avalanche. The potential landslide area at �knes has been mapped by structurally mapped in detailed, whereas subsurface data come from 2D resistivity, Ground Penetrating Radar (GRP), refraction seismics, core drillings and geophysical logging of the boreholes. In symphony, these data give a detailed 3D geological model of the area, in which the depth to � and the geometry of the basal slide surface(s) can be identified. A grid of 2D resistivity profiles indicate an undulating slide surfaces that can be followed from the large tension fracture in the back to the foot of the mapped slide area. Geophysical borehole logging including resistivity, water conductivity, gamma ray of bedrock, and sonic log are consistent with the properties of the bedrock found in the 2D resistivity profiles and in the drill cores. When correlated with drill cores, the sliding surface coincide well with intensely fractured bedrock and layers of fault rocks, such as gouge and breccia. Fracturing along the foliation of the host rock in combination with reactivation of fracture sets controls sub-block sizes as well as the pattern of movement, the latter consistent with a wedge failure model. Trends of fractures follow major trends of lineaments in the area, of which the most pronounced lineaments coincide with major fjords. The importance of fracturing along the foliation in the bedrock can be seen by the geometry of the back fracture. This composite structure is steep to sub vertical, but changes along strike as the foliation is folded. Farther down-slope, the foliations dips moderately towards the fjord and has an undulating geometry. There, the developed subsurface sliding plane shows a similar geometry, basically being sub parallel to the topographic slope. In a regional context, it has been documented that rockslides are more common where the foliation is sub parallel to the slope (Henderson et al. 2006). Through the locating of the sliding surface of the rockslide, a precise estimate of the volume is possibly. Further, it opens for a better understanding of the sliding mechanism(s) of the area. New work, such as borehole monitoring, will contribute to locating the sliding surface more precisely and yield additional quantified data regarding spatial and temporal sliding velocities. The aim of the project is to achieve a state of readiness with direct monitoring of movement, so that the local communities are able to evacuate in time.
http://www.ngu.no
NS22A-05
A Geophysical and Archaeological Analysis of Early Formative Physical Environments at Los Naranjos, Honduras
During the summer of 2003, ground-penetrating radar (GPR) and magnetometry data were acquired at the archaeological site of Los Naranjos, Honduras, one of the earliest archaeological sites in Mesoamerica with monumental architecture and sculpture and one of the easternmost sites in which Olmec artifacts have been recovered. The site was occupied continuously from the Early Formative to the Early Postclassic Periods, with large-scale monumental earthen platform construction occurring during much of its occupation. Geophysical surveys were undertaken in the open plaza areas in the Principal Group, where some of the largest platforms were constructed early in the history of the site. GPR data were collected with 50, 100, 200, and 250 MHz antennae in both 2-dimensional grids and linear profiles, with penetration in excess of 5 meters in some locations. Magnetometry data collection provided a much greater areal coverage of the site, but both methods covered both apparent archaeological and geological features. In some cases, correlation between the geophysical methods is strong, especially when integrated with archaeological information. Interpretation of the geophysical data, combined with analysis of cultural material from archaeological excavations, are helping to understand how the construction and significant expansion of large earthen platform mounds altered the Early Formative physical environment in dramatic ways, both reflecting and intensifying social differentiation and concentration of political power and generating both intended and unintended consequences.
NS22A-06 INVITED
On the use of cross-borehole GPR in integrated geophysical-hydrological investigations of the unsaturated zone
Cross-borehole GPR data are used in integrated geophysical-hydrogeological studies of the upper ca. 10 m of the unsaturated zone in sandy environments. Tomographic algorithms are used for estimating the radar wave velocities between the boreholes, and the estimated velocity values are converted to values of water saturation. The results obtained from the inversion of picked radar wave travel times are strongly influenced by the assumptions that are made regarding model and data error correlation. We analyse and quantify key characteristics of model and data error correlation using different independent sources of information, and we account for these characteristics during inversion. We use normal-incidence reflected radar data sections acquired along profile lines on the surface to constrain dip and spatial correlation lengths of the geological structures. Functions describing the correlation properties of the radar wave velocity fluctuations of the subsurface are estimated based on these observations. These properties are used as a priori information in the tomographic inversion process. Thereby, models capturing realistic heterogeneity of the subsurface are estimated. Good knowledge of fine-scale heterogeneity is critical when estimating water content and flow characteristics. Different sources of correlated data errors exist: Incorrect positioning of the receiver and/or the transmitter antenna during data acquisition; cavities around the borehole walls; unknown anomalies close to the borehole walls; time jumps due to mis-calibration of the transmitted pulse; accidental picking of undesired refracted arrivals which have not followed a straight ray path between the source and the receiver. If not accounted for, such data errors may give rise to significant artefacts in the tomographic images. The correlated data errors are accounted for by specification of data error covariance matrices which are included in the inverse operator used for obtaining the velocity distribution from the picked travel times. We demonstrate that accounting for the correlated data errors results in more optimal inverse estimates and more realistic model resolution estimates.
NS22A-07
Use of 2D and 3D Resistivity Methods to Monitor Dilution of a Conductive Plume in Fractured Basalt
2D cross-borehole and 3D surface electrical resistance tomography (ERT) methods have been shown to be useful in delineating conductive plume migration in porous media. However, their application in fractured basalt, and to monitoring in situ dilution of conductive plumes has been largely uninvestigated. The objective of this study was to monitor the dilution of a conductive plume by more resistive water to delineate the spatial distribution of resistivity changes over time. Eight wells were drilled for the hydrogeophysical experiments. A KCl solution was injected into the partially saturated, fractured basalt via a centrally located injection well for 76 days prior to this dilution experiment. Tap water was then injected into the injection well for 34 days. ERT was used to monitor the dilution and displacement of the KCl plume during tap water injection, and during a subsequent 62-day monitoring period. Data were collected between the wells and at land surface. The ERT data collected during the investigation show the spatial distribution of resistivity changes caused by the influx of diluting water. 3D images of surface ERT results delineate broad areas of increased resistivity due to dilution/displacement of the KCl plume. Cross-borehole ERT data delineate specific locations of water influx. Injection-well resistivities delineate specific locations where tap water seeped from the injection well via preferential flow paths determined by time-dependent resistivity increases at different elevations. Monitoring-well resistivities delineate specific fracture locations and clustered areas of resistivity changes due to the dilution and displacement of the KCl solution. The experimental results presented herein illustrate the application of combined ERT methods to delineate spatially distributed dilution in fractured rock.
NS22A-08
Effects of Strike on Automatic Depth Estimation for 2D Magnetic Structures
Many analysis methods have been developed to process densely sampled magnetic and/or gravity data to estimate source parameters. Werner deconvolution (Werner, 1953), analytic signal (Nabighian, 1972) and Euler deconvolution(Thompson, 1983) are among the most popular methods. They work either on profile data (Bastani and Pedersen, 2001) or on a regular grid (Thurston et al., 2002). All methods developed to estimate source parameters of the 2D magnetic structures work in the strike co-ordinate system. Werner deconvolution makes use of profile data to locate and compute the depth to the top and dip of thin sheets (dikes) with infinite strike and depth extent. Nabighian (1972) introduced the analytic signal to calculate the dip and depth to a set of 2D magnetic sources. The strike angle is assumed to be the same for all the magnetic anomalies along the profile. These methods use the horizontal and vertical derivatives of the total magnetic field to estimate the source parameters. While the vertical derivative is independent of strike direction the horizontal derivative is proportional to the sine of the angle between the profile and strike directions: the profile angle. Bastani and Pedersen (2001) used the analytic signal of the total magnetic field anomaly along a profile to estimate the dip, depth, width and strike of dikes. They introduced a method to estimate the strike of various anomalies at selected points along profiles by searching for coherent signals in neighboring profiles. Here we have used the same method to estimate strike of 2D anomalies. In order to illustrate the importance of strike angle on the estimated source parameters we have constructed synthetic data from a model that comprises a set of thin dikes with the same physical characteristics but with different strikes. We then applied 2D Werner deconvolution, 2D analytic signal (by Bastani and Pedersen), 2D and 3D Euler deconvolution to the data set. As expected the depth estimates are highly biased and varies with the inverse of the sine of the profile angle. Especially for small profile angles this effect becomes significant. With 3D Euler deconvolution depths estimates were not affected by strike variations, but large deviations were observed at the grid points between the profiles and anomalies crossings. Therefore it is necessary to use the strike information prior to gridding. We do this by estimating artificial data points by linear interpolation along lines connecting adjacent profiles. In this way the so-called bull's-eyes effects are reduced and the quality of the interpolation is improved. As a result the error introduced on the 3D depth estimates between adjacent profiles was considerably reduced. References: Bastani, M., and Pedersen, L. B., 2001. Automatic interpretation of magnetic dike parameters using the analytical signal technique. Geophysics, 66, No. 2, P. 551�-561. Nabighian, M. N., 1972. The Analytic Signal of Two dimensional Magnetic Bodies with Polygonal Cross-section: Its Properties and Use For Automated Anomaly Interpretation. Geophysics,37, NO. 3, P. 507--517. Thompson, D. T., 1982, EULDPH: A new technique for making computer-assisted depth estimates from magnetic data, Geophysics, 47, 31-37. Thurston, J. B., Smith, R. S., and Guillonz, J.-C., 2002. A multimodel method for depth estimation from magnetic data. Geophysics, 67, No. 2, P. 555�-561. Werner, S., 1953, Interpretation of magnetic anomalies at sheet like bodies: SGU. Serv C. $\AA$rsbok 43, no. 6.