Lower Cretaceous carbonate strata from the central DeSoto Canyon area, offshore Alabama and Florida were studied to determine the extent, intensity, and controlling factors on dissolution-collapse features within these strata. The collapsed zones across the study area were mapped using a tight 2-D seismic grid. The zones of dissolution-collapse form a crude rectilinear pattern in plan view with the average trend of more elongated sections of the dissolution features being subparallel to regional southwestward dip on the Lower Cretaceous platform. Larger collapse features developed near the modern erosional margin that defines the seaward limit of the Lower Cretaceous platform in the study area. Middle Cretaceous strata up to approximately the Coniacian-Santonian unconformity (middle Late Cretaceous age) have numerous compaction-related faults around sagged areas above apparent dissolution-collapse zones within Lower Cretaceous strata. Bi-directional stratal onlap into the collapsed and sagged zones is only found above the Coniacian-Santonian unconformity. These relationships suggest a regional confined freshwater aquifer system developed within the Lower Cretaceous interval at about Coniacian-Santonian time when meteoric groundwater likely flowed from recharge areas to the north in central Alabama and discharged along the western erosional escarpment of the Lower Cretaceous platform. This meteoric groundwater may have mixed either with seawater that infiltrated the platform from the escarpment edge or with hydrogen-sulfide-rich basinal fluids that migrated to structurally high areas (where the dissolution-collapse zones are found). Alternatively, some combination of these mixing processes may have been responsible for the intense and likely still ongoing dissolution in this area of the Lower Cretaceous carbonate platform.

---

Paleogene cooling (c. 50-30 Ma) started sometime in the early-middle Eocene. This was a time when high-latitude and deep-sea temperatures were significantly warmer than today. This cooling culminated during the earliest Oligocene marked by the sudden appearance of a major continental glacier on Antarctica. We examine this cooling trend by analyzing oxygen isotope variation within mollusc shells from the Gulf Coastal Plain of the southern U.S. Our records show a secular cooling trend of mean annual temperature (MAT) in the Mississippi Embayment from an early Eocene tropical climate (26-27 ºC), with a seasonal temperature range (seasonality) of ~6 °C, to an Oligocene paratropical climate (22-23ºC) with an seasonality of ~8 °C. These temperature records agree well with terrestrial climate proxies. This secular cooling trend, combined with sea-level change, was likely one of the major causes of molluscan turnover in the Mississippi Embayment to cool-tolerant taxa along the Paleogene cooling. Winter temperatures steadily decreased from the middle Eocene to early Oligocene. This contrasts with the sudden winter cooling at Eocene-Oligocene boundary proposed by Ivany et al. (2000).

We examined seasonal temperature distribution of the modern marine shelf of the present northern U.S. Gulf Coast. A deeper water temperature model fits well with isotopic temperature profiles derived from fossils shells of the Red Bluff and Yazoo Formations shells, consistent with the paleobathymetry estimates inferred from independent proxies. This reveals that depth effect is one of the major factors controlling seasonality recorded in mollusc shells, resulting in decreasing MAT estimates when temperature stratification exists as in the present ocean. Warm Eocene low-latitude temperatures derived from molluscan oxygen isotope data agree with computer modeling results incorporating higher greenhouse gas concentrations. This supports the contention that the major reason for warm earth climate is elevated concentration of the greenhouse gases, giving a new insight for future climate response to anthropogenic CO2 increase.

---

The Halley and Emperor fields, Winkler County, West Texas, are located along the western margin of the Central Basin Platform (CBP), a late Paleozoic fault-bounded structural high in the Permian Basin. Well data, regional 2D seismic lines, and a 3D seismic data set were used to develop a detailed structural and stratigraphic interpretation for the area. Variance volume attributes were derived from the 3D seismic data, which improved imaging of subsurface features.

The Halley and Emperor fields are situated over asymmetric anticlines with short steeper limbs that are faulted by steeply dipping reverse faults with a component of right-lateral strike-slip displacement. The orientation of the fold axes and faults is NNW-SSE, which is parallel to the overall trend of the CBP’s western margin. Deformation occurred during late Mississippian to early Leonardian (Early Permian) time.

The CBP can be subdivided into two major blocks or tectonic domains: the Andector Block to the north and the Fort Stockton Block to the south. These blocks were located between an inferred right-lateral mega-shear, which forced the Andector and Fort Stockton blocks to undergo clockwise rotation. A left-lateral shear zone must have existed along the E-W fault boundary between the Fort Stockton and Andector blocks, in order to accommodate clockwise rotation of the blocks. The Halley structure is situated at the southwestern corner of the Andector Block and shows evidence of younger middle Pennsylvanian to middle Leonardian left-lateral strike-slip deformation. In contrast, deformation along the Emperor structure to the north had ceased by late Pennsylvanian time, as indicated by the age of strata that onlap the structure and fault penetration patterns.

---

The "S5", "T" and "U1" sands, traditionally described as part of the lower section of the "Oficina" Formation, and the "U2" sand, as part of the upper interval of the "Merecure" Formation, contain the largest oil remaining reserves of the Leona Este Field, which is located in the southern portion of the Eastern Venezuela Basin.

Two or more of these reservoir sands, which are interbedded with shales, have been simultaneously produced pursuing an increase in the oil production rate, but an unexpected production performance was obtained: the accelerated and early increase of the water volume associated to the produced oil has caused the shut down of some wells in the Leona Este Field.

In order to understand this productive performance and to re-evaluate the hydrocarbon potential of the study interval, it is important to describe these reservoirs in terms of their depositional origin and trap formation. An integrated reservoir model was constructed using all the available geological, geophysical and production data.

The hydrocarbon trapping mechanism of each studied stratigraphic interval, traditionally known as the "S5", "TU", "TL", "U1U", "U1L", "U2U", "U2MA", "U2MB" and "U2L" sands, includes two components:

* Stratigraphic component: each stratigraphic interval presents one or more reservoir zones composed by sandy deposits that fill belts of stacked tidal-fluvial channels in a SSE-NNW trending tide-dominated estuarine system. In most intervals, these contemporaneous-in-deposition reservoir zones are not connected due to the lateral variation of facies present in the tide-dominated estuary.
* Structural component: northward dipping strata have been offset by a W-E trending major normal fault and secondary normal faults striking parallel to the major one. The major fault is the southern seal of the hydrocarbon traps.

The most important prospects of the study interval are the reservoir zones 1 and 2 of the "U1L" sand, the reservoir zone 3 of the "U2MB" sand, and the "U1U" sand, which show more than 15 feet of average net sand thickness, and contain the largest volume of recoverable oil per reservoir zone in the Leona Este Field.

---

The purpose of this study was to interpret and characterize hydrocarbon seeps using a 3D seismic data set. The information gained from this interpretation was then used to develop and understanding of the processes that resulted in the development of the seep features.

The study of the hydrocarbons seeps was made using a 3D seismic survey of the study area aided with interpretation software. Side-scan sonar data was also used to assist in the interpretation.

Three hydrocarbon seep features were identified and interpreted in Garden Banks lease blocks 424 and 425. The seep features consisted of two mud volcanoes and a mud in filled depression. The hydrocarbon seep features were characterized by their rates and
styles of seepage.

All of the seep features are centered over normal faults created in response to the movement of a local salt diapir. Faults in the study area were interpreted and divided into two groups: Series A and B. Series A faults formed from the movement of a salt diapir directly underlying the faults. Series B developed on the flanks of the salt diapir and was the response to increased gravitational forces from an increasing slope angle.

Seismic evidence showed the seep features developed from the upward movement of salt. Faults created from salt tectonics provided a pathway for fluid migration. Overpressured hydrocarbons and mud migrated along these faults to the seafloor. The seep features in Garden Banks 424 and 425 evolved into their present day appearance from continued venting episodes along these salt induced migration pathways.

Seafloor amplitude maps were created to interpret anomalies associated with the style and rate of seepage for each of the features. Vents, authigenic carbonates, and gas hydrates locations were inferred from this map. The abundance and nature of these types of anomalies were used to characterize the seeps.

The interpretation showed the southern mud volcano exhibited evidence of active seepage, while the northern mud volcano showed seismic evidence of being a less active seep. The depression is characteristic of a low flux seep, which has not developed into the higher flux features exhibited by the mud volcanoes.

---

High amplitude reflection packets (HARPs) refer to sheet-like sand deposits showing high-amplitude seismic-reflection character that are thought to be associated with constructional channel systems. Based on observations of Quaternary deposits from the Amazon Fan, HARPs are interpreted to be related to channel avulsion events. The depositional model from the Amazon Fan suggests when levees fail, sediment gravity flows move through the break into interchannel lows, where lack of confinement results in sheet like sand deposits. Based on 2D and 3D seismic data, HARPs from the Gulf of Mexico not only form from updip channel avulsions, but from additional geological processes. In salt withdrawal minibasins, sediment gravity flows encounter an underfilled depocenter, where lack of confinement results in sheet like deposits. After initial spill into an outboard basin and the development of a graded slope equilibrium profile, slope channel complexes develop as conduits for sediment transport. These depositional processes characterize the "fill and spill" model and can result in the creation of HARP-like deposits. At the toe-of-the Sigsbee Escarpment, turbidity currents flow through nickpoints at the seaward edge of the escarpment and become unconfined, resulting in HARP-like deposition on the abyssal floor. Seismically, these deposits are laterally continuous, have low relief, pinchout downdip, and have internal channelization. Erosive, low-relief, discontinuous channels and a low-relief, flat-bottomed trough cut into the TS-HARP deposits. On the outer fan, at the seaward limit of submarine channels, sediment gravity flows will also become unconfined and deposit HARPs. These HARPs typically consist of shingled sheet-like deposits that may form larger scale mounded features. These observations indicate that HARP-like deposits can form from a variety of depositional processes and in variable depositional settings along the Gulf of Mexico depositional profile.

---

The southwestern Midland Basin is located at the junction of several important late Paleozoic tectonic features within the Permian Basin, including the Central Basin Platform (CBP), Ozona Arch, and several basement-involved fault systems that partition the southwestern Midland Basin into smaller depocenters. These tectonic features formed far inboard of the Marathon fold-and-thrust belt. This study examined the late Paleozoic stratigraphic and structural characteristics of the southwestern Midland Basin and adjacent areas in order to constrain the timing of deformational events, the cause of the deformation, and document late Paleozoic tectonic history of the Permian Basin.

Three stages of deformation can be recognized based on significant changes in the lithofacies distribution, the style of deformation, and the area of active deformation through time. Before late Mississippian time, the study area was a tectonically stable region that was dominated by extensive shallow-water carbonate sedimentation. Minor en echelon folding reflected the initial, regionally distributed right-lateral strike-slip deformation in late Mississippian-middle Pennsylvanian time. Soon after deposition of Strawn carbonate ramp facies during a middle Pennsylvanian phase of relative tectonic quiescence, renewed and amplified right-lateral convergence enhanced structural relief of en echelon asymmetrical faulted anticlines. During late Pennsylvanian-Wolfcampian time, en echelon folding and faulting in the basin diminished, but continued right-lateral oblique-slip deformation was mostly accommodated along the eastern boundary of the CBP, which is characterized by steeply dipping reverse faults and asymmetrical flower structures. Major uplift of the CBP occurred during this last phase of intraforeland deformation and the CBP served as the source for the wedge-shaped upper Pennsylvanian through Wolfcampian synorogenic periplatform deposits. The whole area returned to tectonically stable conditions during development of Leonardian carbonate platforms, which built away from CBP that formed during previous phases of tectonic activity.

The tectonic relationships between the subtle structures within the Midland Basin and the CBP are an example of the sequential development of structures during progressive transpressional deformation across a foreland basin. This study may provide insight into the origins of similar intraforeland basement uplifts that developed elsewhere across the interior of the North America during late Paleozoic time.

---

This study reports an investigation of the potentialities of artificial neural networks in the field of reservoir characterization. A first step has been the review of theoretical principles involved in neural networks computations, in order to select among numerous networks which one could be the most efficient for approximating a relationship between seismic attributes and reservoir parameters (porosity, shaliness, water saturation) derived from well log data. The selected network is trained for learning a relationship between seismic attributes, interpolated at the well locations and reservoir parameters at the same locations. Then, the relationship learned along the well will be applied to the entire volume of a seismic cube to predict reservoir properties at the seismic scale.

It turns out that feed-forward networks, trained using a Levenberg-Marquardt algorithm are the most suited to the objective. A method to estimate the accuracy of the predictions has been developed based on the distributions of the values of the seismic attributes and those of the reservoir parameters in the training set, so that probability cubes can be associated with the prediction cubes.

Neural networks suited to the predictions of reservoir parameters have been designed and validated using a data set for which the predictions to obtained were known. They have demonstrated excellent prediction abilities. Then neural networks have been applied on a real data set from Edna (Jackson County, Texas). Predictions of porosity, shaliness and water saturation have been performed at the seismic scale using cubes of attributes.

The results have been associated with probability cubes that allow the quantification of the accuracy of the predictions and that give pertinent results. A simultaneous study of the predicted cubes of porosity, shaliness, and water saturation, along with their associated probability cubes helped characterizing an already producing reservoir. The three predicted parameters show an excellent correlation with well logs (SP, ILD, and DT).

---

Marine vertical cable acquisition is an emerging technology. It represents an alternative to surface seismic in areas congested by platforms or other obstacles. The vertical cable acquisition consists in recording pressure at several fixed vertical receiver arrays just like for VSP. A small vessel carries only the source and is very flexible. The vertical cable receivers are in a quiet environment that can explain the data quality.

One main concern with vertical cable data is to remove the multiples, which are mainly the free-surface multiples and the receiver ghosts of multiples, preserving the primaries. Furthermore the emerging imaging algorithms tend to use the receiver ghosts of primaries instead of the primaries themselves.

The classical approach to remove multiples in vertical cable data is to consider the primaries as up-going waves and the multiples as down-going waves. The current multiple attenuation methods used in E&P industry are actually up/down wavefield separation. Different techniques can be used to achieve the wavefield separation. The common technique is the F-K dip filtering. But this method suffers from the space sampling with the aliasing problem that controls the efficiency of the filtering. An interpolation method can be used to construct the needed traces to satisfy the sampling theorem and to remove the aliases. A wave equation based up/down separation can also be used for vertical cable surveys. This method has the advantage of being less influenced by the sampling problem but it is dependent on the availability of vertical particle velocity data. This particle velocity data can be calculated from the pressure gradient and a good cable design is required for using this method.

The up/down separation based multiple attenuation methods give only a partial answer to the multiple attenuation process. This technique ignores all the remaining free-surface multiples still contained in the up-going waves. So it fails in removing all the multiples.

Another multiple attenuation method based on inverse scattering theory has been proposed by Ikelle (Geophysics, May-June issue, 2001) to preserve the primaries or the receiver ghosts of primaries for vertical cable data. It is efficient, but is dependent on the availability of surface seismic data. The method consists in combining the streamer data with the vertical cable data to predict the multiples and then to subtract them from the original vertical cable data after applying a scaling factor. The application of this technique on a synthetic model with the combining streamer and vertical cable data gives promising results, but in some cases the streamer data is not commonly available. It is then necessary to construct it from the vertical cable data with an extrapolation technique.

---