Seal integrity and feasibility of CO<Subscript>2</Subscript> sequestration in the Teapot Dome EOR pilot: geomechanical site characterization

This paper reports a preliminary investigation of CO₂ sequestration and seal integrity at Teapot Dome oil field, Wyoming, USA, with the objective of predicting the potential risk of CO₂ leakage along reservoir-bounding faults. CO₂ injection into reservoirs creates anomalously high pore pressure at the top of the reservoir that could potentially hydraulically fracture the caprock or trigger slip on reservoir-bounding faults. The Tensleep Formation, a Pennsylvanian age eolian sandstone is evaluated as the target horizon for a pilot CO₂ EOR-carbon storage experiment, in a three-way closure trap against a bounding fault, termed the S1 fault. A preliminary geomechanical model of the Tensleep Formation has been developed to evaluate the potential for CO₂ injection inducing slip on the S1 fault and thus threatening seal integrity. Uncertainties in the stress tensor and fault geometry have been incorporated into the analysis using Monte Carlo simulation. The authors find that even the most pessimistic risk scenario would require ∼10 MPa of excess pressure to cause the S1 fault to reactivate and provide a potential leakage pathway. This would correspond to a CO₂ column height of ∼1,500 m, whereas the structural closure of the Tensleep Formation in the pilot injection area does not exceed 100 m. It is therefore apparent that CO₂ injection is not likely to compromise the S1 fault stability. Better constraint of the least principal stress is needed to establish a more reliable estimate of the maximum reservoir pressure required to hydrofracture the caprock.     

A field-based population model for the sediment toxicity test organism Leptocheirus plumulosus: I. Model development

We describe the development of a stage-structured population matrix model for the estuarine amphipod Leptocheirus plumulosus, a sediment toxicity test organism, based on regular sampling of a natural population for approximately 2 years. Annual population dynamics were characterized by abundance peaks in the spring and fall, and low densities in the summer and winter. We use a sequence of seasonal projection matrices to capture these qualitative patterns in population dynamics. We compare two methods for estimating transition probabilities: multiple regression and truncated singular value decomposition (SVD). Overall, SVD gave more realistic parameter estimates than regression, although fecundity estimation was problematic. There were predictable trends in growth, survival and fecundity within years. The SVD-derived model provides a framework for assessing the population-level effects of contaminants on Leptocheirus.     

Tidally incised valleys on tropical carbonate shelves: An example from the northern Great Barrier Reef, Australia

The formation of incised valleys on continental shelves is generally attributed to fluvial erosion under low sea level conditions. However, there are exceptions. A multibeam sonar survey at the northern end of Australia's Great Barrier Reef, adjacent to the southern edge of the Gulf of Papua, mapped a shelf valley system up to 220 m deep that extends for more than 90 km across the continental shelf. This is the deepest shelf valley yet found in the Great Barrier Reef and is well below the maximum depth of fluvial incision that could have occurred under a − 120 m, eustatic sea level low-stand, as what occurred on this margin during the last ice age. These valleys appear to have formed by a combination of reef growth and tidal current scour, probably in relation to a sea level at around 30–50 m below its present position.

Tidally incised depressions in the valley floor exhibit closed bathymetric contours at both ends. Valley floor sediments are mainly calcareous muddy, gravelly sand on the middle shelf, giving way to well-sorted, gravely sand containing a large relict fraction on the outer shelf. The valley extends between broad platform reefs and framework coral growth, which accumulated through the late Quaternary, coincides with tidal current scour to produce steep-sided (locally vertical) valley walls. The deepest segments of the valley were probably the sites of lakes during the last ice age, when Torres Strait formed an emergent land-bridge between Australia and Papua New Guinea. Numerical modeling predicts that the strongest tidal currents occur over the deepest, outer-shelf segment of the valley when sea level is about 40–50 m below its present position. These results are consistent with a Pleistocene age and relict origin of the valley.

Based on these observations, we propose a new conceptual model for the formation of tidally incised shelf valleys. Tidal erosion on meso- to macro-tidal, rimmed carbonate shelves is enhanced during sea level rise and fall when a tidal, hydraulic pressure gradient is established between the shelf-lagoon and the adjacent ocean basin. Tidal flows attain a maximum, and channel incision is greatest, when a large hydraulic pressure gradient coincides with small channel cross sections. Our tidal-incision model may explain the observation of other workers, that sediment is exported from the Great Barrier Reef shelf to the adjacent ocean basins during intermediate (rather than last glacial maximum) low-stand, sea level positions. The model may apply to other rimmed shelves, both modern and ancient.     

Large-scale mass wasting in ancient volcanic materials

Giant landslides are significant hazards associated with many active volcanic edifices. We describe a similar feature on ancient (>4 Ma) volcanic deposits subject to active tectonism. The landslide is approximately 3 km long by 1 km wide, with an estimated depth of 400 m. Side margins are straight and parallel, mimicking regional structure; narrow valleys incised down these margins provide low-strength side-release surfaces. Between these is a giant slump consisting of at least four, largely intact, discrete blocks that have moved down-dip a distance of >500 m. A series of flows with areal extents ranging from 0.01 to 0.5 km² extends from the front of the failure. The materials represent an eroded sequence of andesite flows on the flanks of a stratovolcano. These have undergone two phases of hydrothermal alteration, and are deeply weathered to low-density (1040±80 kg m⁻³) silt (59%) and clay (35%) materials with strength properties typical of weathered silts (c=26±3 kN m⁻²; φ=42±8°). The size and location of this landslide preclude detailed geotechnical investigation of the failure. The worth of numerical stability analysis as an alternative technique in assessing the nature of the failure and hence the risk it poses to nearby communities is investigated. Sensitivity analysis identified likely conditions under which initial failure may have occurred: analyses for sensitivity to strength and earthquake acceleration needed conversion to critical combinations (F=1.0) of water table and strength/acceleration to remove the overriding influence of water table fluctuations. Failure was likely initiated either by a high water table level (83-84%), or some combination of intensity VII-IX earthquake waves together with water table heights of 40-80%. A general hazard assessment indicates that the risk associated with creep and catastrophic failure of the main mass is small, whereas the risk from flow failures near the toe of the landslide may be high. Important parameters (hydrological regime, flow failure morphology, age of initiation, and rates of movement) requiring closer investigation are identified. Development of a model is crucial to assessing the hazard associated with a feature such as that described here. With limited resources, a detailed stability analysis is a powerful tool as an initial stage in hazard analysis.     

Geological controls on rock mass classification of coal from Huntly East Mine, New Zealand

Rock Mass Rating (RMR) measurements from 65 sites within Huntly East underground coal mine are presented. All measurements are in coal, for which the dominant discontinuities are vertical cleat. Basic RMR values using two discontinuity spacings are presented: overall RMR based on the average spacing of all individual discontinuities; and cleat zone RMR based on the average spacing between zones of cleat. Cleat orientations are highly variable, but on average approximately parallel horizontal stress axes (face cleat follows maximum horizontal stress axis, butt cleat follows minimum horizontal stress axis).Contours of RMR variations throughout the mine are used to compare rock mass conditions with geological structure. It is apparent that: (1) RMR is least within downthrown fault blocks, and particularly immediately on the downthrown sides of faults, and greatest in upthrown fault blocks; and (2) RMR contours parallel horizontal stress axes within fault-bounded blocks, and bend to parallel faults at block boundaries. From similar contours for parameters contributing to RMR, the Rock Quality Designation (RQD), groundwater rating, and discontinuity condition rating create most of the observed variations in RMR. RQD is determined from the measured discontinuity frequency and hence is a measure of the degree of fracturing of the rock mass. This is interpreted as influencing the groundwater and condition parameters directly by allowing greater water ingress. Discontinuity frequency is greatest (least spacing) in the immediate vicinity of faults, and in downthrown fault blocks, generating low RMR values. Within fault blocks RQD varies little, so RMR contours align with cleat orientations.As RMR contours, faults, stress field and cleat orientation are clearly interrelated, there is unequivocally a connection between RMR and structural geology; this allows some predictive capacity in terms of ground conditions. If geological features can be accurately defined through either drilling programs or seismic surveys, then ground conditions may be predicted before panels are driven.     

Tidal river sediments in the Washington,D.C. area. III. Biological effects associated with sediment contamination

Sediment toxicity and benthic macroinvertebrate community structure were measured as one component of a study cohceived to determine the distribution and effect of sediment contamination in tidal freshwater portions of the Potomac and Anacostia rivers in the Washington, D.C., area. Samples were collected at 15 sites. Analyses included a partial life cycle (28 d) whole sediment test using the amphipod Hyalella azteca (Talitridae) and an assessment of benthic community structure. Survival and growth (as estimated by amphipod length) were experimental endopoints for the toxicity test. Significant mortality was observed in 5 of 10 sites in the lower Anacostia River basin and at the main channel Potomac River site. Sublethal toxicity, as measured by inhibition of amphipod growth, was not observed. Toxicity test results were in general agreement with synoptically measured sediment contaminant concentrations. Porewater total ammonia (NH₃+NH₄ ⁺) appears to be responsible for the toxicity of sediments from the Potomac River, while correlation analysis and simultaneously extracted metals: acid volatile sulfide (SEM∶AVS) results suggest that the toxicity associated with Anacostia River sediments was due to organic compounds. Twenty-eight macroinvertebrate taxa were identified among all sites, with richness varying from 5 to 17 taxa per site. Groups of benthic assemblages identified by group-average cluster analysis exhibited variable agreement with sediment chemical and sediment toxicity results. Integration of toxicological, chemical, and ecological components suggests that adverse environmental effects manifest in the lower Anacostia River benthos result from chemical contamination of sediment.     

Assessment of episodic acidification in Sierra Nevada,California

Monte-Carlo simulations were used to assess the extent of shortterm alkalinity depressions occuring in Sierra Nevada lakes due to acidic deposition events. The Episodic Event Model (EEM) was used to simulate spring snowmelt events. Snow course data, precipitation data and lake acidification surveys were used to derive values for the EEM parameters. Spring snowmelt events were shown to have great impacts on the water quality of Sierran lakes. Lakes are likely to be most affected by the early-spring snowmelt event because the epilimnion depth is at a minimum, which indicates minimum dilution. Under annual average loading conditions, no Sierran lake has been reported as acidic although 29% of the lakes have alkalinities less than 40 µeq/L indicating a sensitivity to acidification. In simulations of early-spring snowmelt events, using present-day acidic loading conditions, it was estimated 79% ± 9% of the lakes would experience shortterm lake alkalinity depressions to levels less than 40 µeq/L. The results provided by the model simulations are valuable in establishing upper and lower limits on the extent of possible episodic acidification to lake-resources-at-risk. The most critical parameters controlling the magnitude of lake alkalinity depressions during snowmelt episodic events are a) the lake area to watershed area ratio — a measure of input loading, and b) the epilimnion volume — a measure of dilution and mixing.     

Anomalous hydrogen emissions from the San Andreas fault observed at the Cienega Winery,central California

We began continuous monitoring of H₂ concentration in soil along the San Andreas and Calaveras faults in central California in December 1980, using small H₂/O₂ fuel-cell sensors. Ten monitoring stations deployed to date have shown that anomalous H₂ emissions take place occasionally in addition to diurnal changes. Among the ten sites, the Cienega Winery site has produced data that are characterized by very small diurnal changes, a stable baseline, and remarkably distinct spike-like H₂ anomalies since its installation in July 1982. A major peak appeared on 1–10 November 1982, and another on 3 April 1983, and a medium peak on 1 November 1983. The occurrences of these peaks coincided with periods of very low seismicity within a radius of 50 km from the site. In order to methodically assess how these peaks are related to earthquakes, three H₂ degassing models were examined. A plausible correlational pattern was obtained by using a model that (1) adopts a hemicircular spreading pattern of H₂ along an incipient fracture plane from the hypocenter of an earthquake, (2) relies on the FeO–H₂O reaction for H₂ generation, and (3) relates the accumulated amount of H₂ to the mass of serpentinization of underlying ophiolitic rocks; the mass was tentatively assumed to be proportional to the seismic energy of the earthquake.     

Characteristics of high-resolution marine reflection profiling sources

Midwater signatures of eight common types of marine seismic sources have been recorded and characteristics of importance for high-resolution reflection profiling have been analyzed. Analyzed characteristics include wavelet shape, peak frequency, bandwidth, repeatability and directivity. Digitization rates required to describe the signatures before any processing and after whitening deconvolution were determined.It was intended that most of the broad range of available source types be represented. Toward that end, both resonant and impulsive types were studied; the impulsive types being both electric-discharge and pneumatic. They included a conventional sonar transducer, a chirped sonar transducer, a boomer, a plasmagun, a multi tip sparker, a watergun, an airgun and three sizes of sleevegun.It was observed that the peak frequency of each conventional sonar is similar to its nominal value and that of the chirped sonar is about the midpoint of its nominal sweep bandwidth. The dominant frequencies of the electric-discharge impulsive sources are about an order of magnitude higher than those of the pneumatic impulsive sources. Among impulsive sources, the boomer has the highest peak frequency and the watergun has the broadest bandwidth. The sleevegun and airgun were found to be comparable, both having low frequencies, poor repeatability and weak directivity. The conventional sonar and the boomer are the most repeatable sources. The boomer shows the strongest directivity with the chip sonar and the watergun being the most nondirectional.Sample rates required to describe the unprocessed signatures vary from 2 to 88 times the peak frequency and from 1 to 4 times the bandwidth 40dB below peak power. The required rate depends largely on source type but can also be affected by energy level. Bandwidth seems to be a more stable indicator than peak frequency. Rates required to describe the results of applying whitening deconvolution to signatures were about 1.5–3 times higher than rates calculated for the unprocessed signatures. This leads to the conclusion that the digitization rate required in a specific situation can depend as much on the intended processing scheme as it does on source type.