Bruce S. Hart

Comparison of linear regression and a probabilistic neural network to predict porosity from 3‐D seismic attributes in Lower Brushy Canyon channeled sandstones, southeast New Mexico

The Lower Brushy Canyon Formation of the Delaware Basin, New Mexico, consists of a series of overlying sand‐filled channels and associated fans separated by laterally extensive organic siltstone and carbonate interbeds. This laterally and vertically complex geology creates the need for precise interwell estimation of reservoir properties. In this paper we integrate wireline log and 3‐D seismic data to directly predict porosity in the area of an existing oil field in southeast New Mexico. The 3‐D seismic data were used to interpret the location of major stratigraphic markers between wells, and these seismic horizons were used to constrain a time window for a volume‐based attribute analysis. Stepwise regression and crossvalidation were used to combine seismic attributes to predict porosity in wells where the porosity was known from the well logs. The results of a linear regression porosity model showed good correlation (r2=0.74) between seven seismic attributes and the observed porosity logs at 11 wells in ...

3-D seismic horizon-based approaches to fracture-swarm sweet spot definition in tight-gas reservoirs

In this article we will discuss some horizon-attribute-based methods that we have been exploring to help identify productive fracture swarms in tight-gas reservoirs.

Channel detection in 3-D seismic data using sweetness

Sweetness is a seismic attribute that, especially when used in conjunction with coherency, can be very useful for channel detection in deep-water clastic and coastal-plain settings. Although the attribute is not new, previous documentation of its utility and derivation has been mostly lacking. In this article, I present images of channels from three-dimensional seismic volumes that were derived using sweetness, and discuss the physical basis of the attribute. Furthermore, the modeling presented here suggests that sweetness could be used in a semiquantitative way to predict net-to-gross ratio in channel systems. Sweetness is derived by dividing reflection strength by the square root of instantaneous frequency. This mathematical definition captures attribute relationships that seismic interpreters have been using qualitatively for many years: isolated sand bodies in shale successions tend to generate stronger, broader reflections than the surrounding shale. Sweetness becomes less useful for channel detection when acoustic impedance contrasts between sands and shales are low or when sands and shales are highly interbedded.

Tidal signatures in a shelf-margin delta

Based on its anomalous thickness (∼150 m) and stratigraphic position above continental-slope mudstone, an upward-coarsening succession consisting in part of tidal rhythmites in the Glenelg Field, offshore Nova Scotia, Canada, is interpreted to be a strongly tide-influenced shelf-margin-delta deposit. A large, funnel-shaped erosional shelf-edge invagination is observed where the paleoshelf edge is resolved in three-dimensional seismic data adjacent to Glenelg. We propose that the delta at Glenelg prograded into a similar shelf-edge invagination within which tidal currents were amplified and wave energy was attenuated. Given that funnel-shaped invaginations (e.g., slope canyon heads, slump scars, fluvially incised knickmarks) are relatively common along modern shelf edges, and that fluvio-deltaic systems should be focused into these topographic lows during regression across the shelf, it seems likely that shelf-edge invaginations play an important but underappreciated role in mediating terrigenous clastic sedimentation during sea-level lowstands.

Stratigraphy and sedimentology of shoreface and fluvial conglomerates: insights from the Cardium Formation in NW Alberta and adjacent British Columbia

ABSTRACT In this paper we describe little-studied exposures of the Upper Cretaceous Cardium Formation in the northern Foothills and Peace River Plains, located north of Twp. 58. We correlate sixteen principal outcrop sections along the Foothills and across the northern escarpment of the Cardium as far east as the Smoky River. Correlations are based on tracing bounding erosion surfaces, initially observed in well logs and cores, and later extended into outcrop. The Kakwa Member is conglomeratic in many of these sections. The member displays a spectrum of conglomerate types, of both shallow marine and fluvial origin. Depositional environments are differentiated on the basis of stratigraphic context, conglomerate texture, sedimentary structures, stratification style and paleogeographic setting. We describe in detail four localities exemplifying the main paleoenvironments. At Bay Tree, a 12 m thick section consists primarily of clast-supported conglomerates representing a wave-dominated shoreface and beachface. This environment is dominated by broadly horizontal stratification with horizontal facies heterogeneity on a metre scale. At Cutbank Lake, swaley-stratified shoreface sandstones contain pebble-lined scours that show an upward increase in thickness and lateral continuity and represent rip current deposits. At Mount Niles, conglomerate forms a 2-6 m thick, >200 m wide, channelized body consisting of heterolithic matrix-supported conglomerates and pebbly sandstones capped by gravel wave ripples. Deposition may have been in a river-mouth setting. At Horseshoe Mountain, the upper part of the Kakwa Member consists of metre-scale crossbedded pebbly sandstone and matrix-supported conglomerate deposited in a pebbly fluvial system lacking marine influence. For the entire study area, paleocurrent observations indicate a strong NW to SE alongshore transport of gravel, with storm waves approaching from the NE. Fluvial crossbedding is directed towards the NE. Some of the outcrop facies can also be recognized in core from the adjacent subsurface. Our descriptions of conglomerate facies in outcrop facilitate interpretation of the depositional environment of comparable units in subsurface. Unfortunately, the limited extent of exposures does not allow us to determine with precision the geometry of conglomeratic units. However, a broad conglomerate-rich zone >90 km in dip extent and 10-20 km in strike extent can be traced across the northern outcrop belt and probably records the position of a major and long-lasting pebbly fluvial system. Subsurface analogs would be exploration fairways. At the reservoir scale, conglomerate bodies showing greater influence of wave reworking would tend to be more elongate, and have greater facies continuity/permeability, in a shore-parallel direction. The outcrops display lateral textural heterogeneity at scales of metres to 10s of metres. This heterogeneity would affect reservoir performance, but could not be captured (for reservoir modeling) using typical development well spacing. End_Page 437------------------------

Seismic expression of fracture-swarm sweet spots, Upper Cretaceous tight-gas reservoirs, San Juan Basin

Fracture swarms associated with subtle structures greatly enhance production from tight-gas sandstones of the Mesaverde Group and Dakota Formation in the San Juan Basin. The structures include grabens, horsts, and normal faults, and they can be identified using curvature analyses of horizons mapped in three-dimensional seismic data. Their orientations and styles are consistent with the orientation of fractures that have been identified by other authors using outcrop, core, borehole imagery, and production analyses. Integration of production data (rate-versus-time plots) demonstrates the existence of a drainage interference for wells located on some of the structures. This observation further testifies to a positive correlation between the presence of subtle structures, high fracture intensity, and high fracture permeability. The methods and results described herein can be directly applied to other areas. However, because natural fractures can both enhance and retard hydrocarbon production depending on their character, calibration of seismic, engineering, and other data types will be needed to determine whether subtle structures should be considered as drilling targets, or whether they are to be avoided.

Stratigraphically significant attributes

Seismic attributes are derivatives of seismic data that are commonly used for two purposes, feature detection and to predict (usually quantitatively) physical properties of interest. Although there is a well-developed interest in using “physically significant” attributes (i.e., attributes thought to respond to or directly image changes in acoustic or elastic properties) to predict subsurface physical properties, in this article I make a case for identifying and using “stratigraphically significant” attributes. Stratigraphically significant attributes are seismic attributes that capture changes in waveform shape that are caused by changes in stratigraphy. As used here, the term “stratigraphy” refers to vertical and lateral changes in bed thickness and physical properties that are generally caused by changes in depositional processes. It should be recognized, however, that changes in physical properties caused by diagenesis can be important, especially in carbonates, and are here included in the definition ...

Validating seismic attribute studies: Beyond statistics

Broadly defined, seismic attribute studies (“attribute studies”) are attempts to use attributes derived from seismic data to predict the distribution of physical properties (e.g., porosity, lithology, bed thickness) of the strata being imaged seismically. Attribute studies may be either quantitative (i.e., the objective is to make numerical predictions of properties of interest throughout the seismic coverage) or qualitative (i.e., the objective is to find geobodies sharing similar physical properties). In either case, the interpreter must decide which attributes to derive, how to analyze the attributes, and (not least importantly) how to test the results of the correlation exercise before making exploration or development decisions.

A visual data-mining methodology for seismic facies analysis: Part 2 — Application to 3D seismic data

Avisual data-mining approach to unsupervised clustering analysis can be an effective tool for visualizing and understanding patterns inherent in seismic data i.e., seismic facies.The unsupervised clustering analysis is completely data-driven, requiring no external information e.g., well logs to guide the seismic-trace classification. We demonstrate the application of the visual data-mining approach to seismic faciesanalysisonareal3Dseismicdatavolume.Weselecttwo stratigraphic intervals, the first including a Devonian pinnacle reef system and the second containing a Jurassic siliciclastic channel system. Both analyses show major stratigraphic features that can be defined in horizon slices or other types of visualization. However, the visual data-mining approach creates seismic facies maps with improved visual detail,distinguishingseismictrace-shapevariabilityinthedata. We also compare the facies maps with those obtained from a commercial package for seismic facies classification. Both approaches created similar facies maps, but the visual strategy better depicts subtle stratigraphic changes in the bodies being imaged, offering insight into the nature of these features.

Three-dimensional seismic-based definition of fault-related porosity development: Trenton–Black River interval, Saybrook, Ohio

Oil and gas reservoirs of the Ordovician Trenton–Black River interval in the Appalachian Basin are commonly associated with fault-related hydrothermal dolomites. However, relationships between porosity development and fault geometry in these fields are poorly documented. In this article, we integrate three-dimensional (3-D) seismic and wire-line data from the Trenton–Black River interval at Saybrook field in northeastern Ohio to study relationships between faulting and porosity development there. Faults were mapped using a combination of amplitude and coherency versions of the seismic data, and a 3-D porosity volume was generated for the Trenton–Black River interval by integrating attributes derived from the seismic data with log-based measures of porosity. The productive trend in the Trenton–Black River interval at Saybrook is controlled by a 3.4-mi (5.5-km)-long, northwest-southeast–oriented basement fault that was probably reactivated during the Taconic orogeny (i.e., Late Ordovician). Strike-slip movement along the fault generated en echelon synthetic shear faults that branch at least 1350 ft (411.5 m) upward into the Trenton–Black River interval. The best porosity is developed in areas between overlapping synthetic shear faults. Antithetic shear faults probably formed at these locations and, when combined with minor dip-slip movement, created conduits for subsequent porosity-generating fluids. Circular collapse structures associated with localized extension between overlapping shear faults are the primary drilling targets, and horizontal wells running parallel to the strike of the fault would have the best chances of intercepting good porosity development.