No abstract

Facies distributions, stratal geometry and regional erosional bevelling surfaces in Upper Cretaceous (Cenomanian‐Santonian) strata of the Alberta foreland basin are interpreted in terms of high‐frequency (probably eustatic) relative changes in sea level, superimposed on longer‐term basin‐floor warping, related to episodic tectonic loading. Thick marine shales correspond to periods of rapid subsidence whereas thin but extensive strandplain sandstones record rapid progradation during slow subsidence. Westward‐thickening wedges of coastal plain strata were deposited during initial downwarping of a near‐horizontal strandplain, prior to marine transgression. Surfaces of erosional bevelling beneath which between 40 and >160m of strata have been removed extend at least 300 km from the present deformation front and are interpreted to reflect forebulge uplift in the east. Uplift appears to have lagged about 0.25‐0.5 Myr behind the onset of accelerated loading. Thin marine sandstones which grade westward into mudstone are interpreted as material winnowed from the crest of the rising forebulge. Subsidence and/or westward migration of the forebulge allowed the sea to flood westward across the eastern flank of the eroded forebulge. The transgressive shoreface cut asymmetric notches which were later blanketed by marine shales which lap out from east to west. The two unconformities which embody the largest erosional vacuity are veneered locally with oolitic ironstone which accumulated in a shallow, sediment‐starved setting on the crest of the forebulge. The consistent pattern of erosional bevelling and lap‐out of transgressive shales might be interpreted as evidence that the forebulge migrated towards the thrust load over a period of <1 Myr.

No abstract

No abstract

The morphology and dynamics of modern gravel shorefaces are poorly documented. This hinders the interpretation of possible ancient counterparts. A comparative study of a modern (Chesil Beach, England) and an ancient (Baytree Member of the Cardium Formation, Alberta) gravel shoreface shows that the two systems are very similar close to and above sea‐level, with a high (about 1 m) gravel plunge step lying below plane‐bedded sands and gravels of the beachface. The shoreface at Chesil Beach is dominated by asymmetrical gravel wave ripples. These are oriented offshore near the toe of the shoreface, and onshore in shallower depths. This may reflect offshore movement during storms and landward reworking during fair weather. The Baytree Member is over 12 m thick and comprises over 80% conglomerate. Conglomerate is decimetre‐bedded, massive or cross‐bedded, with sets over 60 cm thick produced by gravel bedforms migrating alongshore. It is interbedded with discontinuous cm‐ to dm‐bedded sandstones which may be cross‐bedded. Pebble fabric and cross‐bed orientation both indicate strong alongshore sediment transport. Near the base of the section, pebble orientations suggest that gravel wave‐ripples developed below the zone of strong longshore flows. Differences between these two examples may be attributed to different directions of wave approach.

No abstract

Much industry interest is centered on how to integrate well data and attributes derived from 3-D seismic data sets in the hope of defining reservoir properties in interwell areas. Unfortunately, the statistical underpinnings of the methods become less robust in areas where only a few wells are available, as might be the case in a new or small field. Especially in areas of limited well availability, we suggest that the physical basis of the attributes selected during the correlation procedure be validated by generating synthetic seismic sections from geologic models, then deriving attributes from the sections. We demonstrate this approach with a case study from Appleton field of southwestern Alabama. In this small field, dolomites of the Jurassic Smackover Formation produce from an anticlinal feature about 3800 m deep. We used available geologic information to generate synthetic seismic sections that showed the expected seismic response of the target formation; then we picked the relevant horizons in a 3-D seismic data volume that spanned the study area. Using multiple regression, we derived an empirical relationship between three seismic attributes of this 3-D volume and a log‐derived porosity indicator. Our choice of attributes was validated by deriving complex trace attributes from our seismic modeling results and confirming that the relationships between well properties and real‐data attributes were physically valid. Additionally, the porosity distribution predicted by the 3-D seismic data was reasonable within the context of the depositional model used for the area. Results from a new well drilled after our study validated our porosity prediction, although our structural prediction for the top of the porosity zone was erroneous. These results remind us that seismic interpretations should be viewed as works in progress which need to be updated when new data become available.

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 [Formula: see text] between seven seismic attributes and the observed porosity logs at 11 wells in the study area, but the porosity volume created from the regression model did not display the known geologic features. A probabilistic neural network was then trained to look for a nonlinear relationship between the input data (the seven attributes) and the observed porosity at the 11 wells. The correlation was better [Formula: see text], but the biggest improvement over the linear regression model came in the more geologically realistic predicted porosity distribution.