Staff Scientist: Adam Klaus
Chief Scientist: J. Casey Moore
Abstract
Deformation of accretionary prisms changes the physical properties of sediments, thereby producing fluid, controlling fluid flow, altering rheologic properties, and affecting seismic arrival times and reflection characteristics. Consolidation and chemical diagenesis change the specific physical properties of porosity, density, and sonic velocity. These changes are both distributed (due to the loss of fluids in response to accumulating stresses; Bray and Karig, 1985; Bangs et al., 1990) and localized along discrete structures (such as faults) in response to overpressuring, fluid migration, or fault collapse (Shipley et al., 1994; Tobin et al., 1994). Because consolidation and fluid overpressuring affect seismic arrival times and seismic reflections, seismic data provide direct clues to physical property evolution and to physical property changes coupled to deformation.
Physical property evolution in sedimentary sequences cannot be comprehensively evaluated with recovered cores. Elastic rebound and microcracking of coherent sedimentary samples degrade shipboard physical property measurements. Fault gouge and other incoherent lithologies are either not recovered or cannot be measured after recovery; therefore, transient properties (e.g. overpressuring) must be measured in situ (Fisher et al., in press).
Sediments in tectonically active areas experience rapid changes in physical properties. Accretionary prisms are exceptional, natural laboratories to study these changes because of this rapid deformation and the shallow burial depth of the deformed features, which can therefore be drilled and imaged seismically. The information discerned at convergent margins about fault geology, overall sedimentary consolidation, in addition to seismic imaging of these processes, will be applicable to other less active sedimentary environments, and therefore impact our understanding of hydrocarbons, groundwater, and aspects of earthquake systems. To better understand the interrelationships of deformation, fluid flow, seismic imaging, and changes in physical properties, we propose a Logging While Drilling (LWD) transect of a setting dramatically influenced by pore fluids: the Barbados accretionary prism.
Logging While Drilling is the most effective tool for measurement of physical properties in poorly consolidated sediments. LWD acquires a continuous log of physical properties directly above the drill bit where hole conditions are optimal for logging. It is an "off the shelf" industry technology already used by the Ocean Drilling Program (ODP) during Leg 156 that can be directly applied to ODP operations without development costs. LWD will yield important results from accretionary prisms where wireline logging has failed.
LWD acquires data from sensors integrated into the drill string immediately above the drill bit, and records data minutes after cutting the hole when it most closely approximates in situ conditions. This technology provides high quality logging information in environments where standard wireline systems previously acquired either no data or poor quality data (Fig. 1). Specifically, LWD provides excellent quality results in the shallowest sediment sections and in holes with marginally stable conditions that preclude wireline log runs. Logging While Drilling compares to standard wireline logging as the Advanced Piston Coring System compares to standard rotary drilling.
A comparison of density data from cores, LWD, and wireline logs for the Barbados accretionary prism highlights the value of data acquisition during drilling (Fig. 1). The LWD data closely reproduce the individual core measurements but provide much more detail. The density data from LWD apparently "see" hydrofractures in the Barbados décollement zone (Fig. 2), which would either not be resolved by wireline data or would have closed due to fluid loss to the borehole. In the Barbados example (Site 948) the wireline density data are not usable, because density measurements require consistent pad contact and are very sensitive to changes in hole diameter (in comparison to resistivity or velocity measurements). However, for this study density is the most important measurement for analysis of consolidation.
Wireline tools are more sophisticated than LWD tools and in principle should yield more accurate measurement of physical properties. However, the difficult hole conditions encountered by drilling, especially at active margins, destroy the inherent advantage of wireline tools. In this type of environment, the excellent correlation of LWD density data to core sample density data clearly shows the superiority of LWD to wireline-measurements (Fig. 1).
Off-the-shelf LWD tools provide neutron porosity, resistivity, density, and gamma-ray data. An LWD sonic tool currently exists that can measure formation velocities greater than about 3 km/s. An improved version may be available for this cruise. Such a sonic velocity tool would be necessary to achieve the objectives of this cruise. Lacking this sonic tool, the velocity data would have to be acquired using wireline techniques in the LWD hole. Reentry of the LWD hole would be achieved by dropping a minicone and proceeding with a wireline velocity measurement, using the side-entry-sub if necessary. Because density and velocity are strongly correlated, synthetic seismograms can be created with knowledge of either; hence most of the objectives to determine the polarity and shape of the seismic waveform from fault zones could be met, even in the absence of velocity data.
The absence or failure of wireline logging operations means that hundreds of previously drilled Deep Sea Drilling Project (DSDP) and ODP holes provide scientifically exciting locales for LWD. Barbados is especially attractive for focused LWD investigation because:
1) Overall Prism Consolidation and Velocity-Porosity Relationships
Porosity is the foundation for a variety of studies about the large-scale, long-term fluid budget of accretionary prisms. Logs can be used to determine a continuous record of density and porosity as a function of depth as was done on Leg 156 (Figs. 1 and 2). Between-site variation in the porosity-depth relationship provides an estimate of the amount of fluid expulsion (and therefore volumetric strain). Unfortunately, measurements of volume change are usually impossible with standard logs as they frequently fail due to bad hole conditions in this setting (Fig. 1). Even under ideal conditions wireline logs do not sense the top 60-120 m because the drill pipe extends below the seafloor, nor do they sense the bottom 60-120 m of the hole because of fill (Fig. 1). The shallowest hundred meters, where porosity reduction is the greatest, is of particular interest in this study. Only LWD can obtain reliable porosity logs from the entire depth range, including the critical top 100 meters.
Profiles of porosity versus depth provide a tantalizing but incomplete view of the fluid expulsion pattern of an accretionary prism. Velocity data, either from multichannel seismic data (Bray and Karig, 1985; Bangs et al., 1990; Cochrane et al., 1994) or ocean bottom seismograph (OBS) studies (Ye et al., submitted), is a powerful tool for studying prism porosity structure. The fundamental limitation in determining porosity from velocity is the conversion between these two parameters. This relationship is well known for normally consolidated, low-porosity sediments (e.g., Gardner et al., 1974), but it is much less certain for high-porosity sediments, where changes in terms of fluid production and volumetric strain are more important. Furthermore, our analysis of logs from the Cascadia accretionary prism indicates that prism deformation dramatically changes the porosity-velocity relationship (Jarrard et al., 1995). In contrast to pelagic sediments, accretionary prism sediments of the same porosity can exhibit a wide range of elastic moduli and, therefore, velocities; this complexity results from variability in cementation, compression-induced modification of intergrain contacts, and fracturing. Theoretical relationships of porosity to velocity (e.g., Gassman, 1951) are of little utility in this environment; we must determine the velocity-porosity relationship for each prism empirically, and we must investigate the possibility that this relationship changes laterally within a prism. In situ velocity and porosity logs that sample the section completely are the only means of reaching this objective.
The overall fluid budget of the Barbados prism requires analysis to evaluate the fluid loss and geochemical budgets (e.g., Bekins et al., in press). The series of LWD holes planned here, plus existing penetrations, will help constrain this problem.
2) Correlation of Physical Properties of Faults with Displacement and Fluid Flow
An LWD transect across the Barbadian décollement can address the following issues (1) do fault collapse and strain harden with displacement (e.g., Karig, 1986) and (2) does active fluid flow retard this process and are collapsed faults inactive with respect to fluid flow (e.g., Brown et al., 1994). Structural, biostratigraphic, and seismic reflection criteria identify faults. Anomalies in pore-water geochemistry (e.g., Kastner et al., 1991) and thermal anomalies (Fisher and Hounslow, 1990) indicate fluid flow. With the positive identification of faults, LWD can measure their physical properties. These properties then can be correlated to variations in displacement and fluid activity.
3) Consolidation State of Sediments in and Around Faults
At Site 948 in the Barbados prism, high quality density measurements demonstrated under-consolidation around faults, indicating the faults had recently loaded subjacent sediments. The consolidation state can also be interpreted in terms of effective stress and fluid pressure. Clearly consolidation varies around faults and should be defined in order to develop any tectonic-hydrologic model of the fluid expulsion system.
4) Polarity and Shape of the Seismic Waveform from Fault Zones
Seismic reflections are created by changes in physical properties that can in turn be measured in boreholes. In principle, the seismic data provide a proxy for these larger scale changes in physical properties. The polarity and shape of the seismic waveform were mapped and various models formulated for the waveform across décollement zones beneath accretionary prisms (Bangs and Westbrook, 1991; Moore and Shipley, 1993). Negative polarity reflections have been interpreted as resulting from either (1) overthrusting of higher impedance sediment over lower impedance sediment in Costa Rica (Shipley et al., 1990) or (2) the reduction of fault zone impedance through dilation at Barbados (Figs. 3 and 4; Bangs and Westbrook, 1991; Shipley et al., 1994; Bangs et al., 1996). The modeling, however, is incomplete without ground-truthing through the in situ measurement of physical properties across fault zones in areas with good seismic data.
To date, logging data has only been acquired at one décollement locality (Leg 156 Shipboard Scientific Party, Site 948, 1995). This LWD data from Barbados is in an area of positive reflection polarity and shows impedance increases that reproduce the positive polarity in synthetic seismograms (Leg 156 Shipboard Scientific Party, Site 948, 1995). The LWD results also suggest thin (0.5-1.5 m) hydrofractures within the interval of positive polarity in the décollement zone. The hydrofractures apparently are too thin to be resolved seismically. A major question is whether negative polarities elsewhere in the Barbados décollement consist of thicker zones of hydrofractures.
LWD investigations of the Barbados prism will build on existing LWD measurements at Site 948 that penetrated the décollement where it is of positive polarity (Figs. 1 and 2). At Site 947 LWD penetrated a locality never cored because of the great depth to the décollement and unstable hole conditions discovered during the LWD penetration. Proposed LWD sites will focus on determining the characteristics of the negative polarity reflections at Barbados, measuring the physical properties of faults, and determining the physical properties of the incoming sedimentary sequence. The sum of all penetrations will provide an overview of prism consolidation and velocity-porosity relationships. In prior drilling through the North Barbados Ridge accretionary prism during Leg 156, 1,152 m of logs in Hole 947A and 948A were obtained using LWD technology. This proposal is specifically designed to acquire more LWD data in four additional holes in the Barbados accretionary prism. LWD tools are mounted in the rotating bottom-hole assembly, allowing estimates of porosity, fluid pressure, and seismic properties to be measured through the prism minutes after cutting the hole, closely approximating in situ conditions.
These additional LWD investigations of the Barbados prism will focus on determining the characteristics of the negative polarity reflections at two sites, building on existing data at Site 948 that penetrated the décollement where it is of positive polarity, and at Site 947 where unstable hole conditions limited penetration. Thin, low-density layers in the décollement zone, possibly hydrofractures at Site 948, were observed during Leg 156. One site as a reference section east of the deformation front and one site coincident with the CORK site (NBR-9) will also be drilled.
The proposed tools are the same as those used during Leg 156, directly measuring in situ resistivity, porosity, density, and natural gamma-ray. An LWD sonic tool is currently available from Anadrill for velocities >2,000 m/s, which may not be low enough for these sediments in the accretionary prism. The sonic tool will yield data on the higher velocity underthrust section and may be improved to measure the velocity of the prism by 1997. In the proposed program, the total logged interval is 2,335 m, ~2-2.5 times more section than drilled during Leg 156, generating a total operations time estimate of 10-12 days. Several days of transit and port calls associated with the tool logistics will also be required.
There will be no coring in this leg. In order of priority the sites are:
1) NBR-8: NBR-8 will establish the physical properties of the negative polarity reflections in the Barbados prism. It is located in an area of negative polarity about 2500 m west of the deformation front (Figs. 3 and 4). Shipley et al. (1994) predict that the negative polarities are dilatant zones. Accordingly they may be characterized by "hydrofractures" or zones of fluidized sediment more numerous than those encountered at Site 948. Because the depth of the décollement is 400 m as opposed to the more than 600 m at Site 947, and the negative amplitude is less than at Site 947, NBR-8 can be successfully completed. The site has never been cored; however, safety problems are not anticipated because nearby penetrations show negligible hydrocarbons. Correlations from nearby holes and the 3-D seismic data should provide basic lithologic information.
2) NBR-9: NBR-9, located at Cork Site 949 1800 m west of the deformation front, will establish the physical properties of a décollement zone with intermediate reflection polarity characteristics, and determine the physical property profile at this bore hole seal site. This site is also cut by an imbricate thrust fault that is actively deforming the accretionary prism and will provide information on the physical properties of thrusts.
3) NBR-10: NBR-10, located at Site 676, will determine the character of the initial deformation of the accretionary prism. This site is located about 800 m inboard of the deformation front and penetrates the incipiently developed décollement zone and several thrusts in the offscraped section.
4) NBR-11: NBR-11 is located at the oceanic "reference" Site 672, 6000 m east of the deformation front. This site showed incipient deformation and a geochemical anomaly at the stratigraphic level of the projected décollement zone. LWD here will provide information on the inception of deformation and fluid flow in the incoming sedimentary section as well as a general overview of physical properties of the oceanic sedimentary section.
To Leg 171B Proposed Site Information