Jerry L. Jensen

Natural fractures in the Spraberry Formation, Midland basin, Texas: The effects of mechanical stratigraphy on fracture variability and reservoir behavior

Horizontal cores from sandstone-siltstone reservoirs in the Spraberry Formation (Midland basin, west Texas) have documented two systems of dramatically different yet dynamically compatible natural fractures, in reservoirs separated vertically by only 145 ft (44 m). Each system is capable of producing a different degree of the northeast-trending permeability anisotropy recognized in Spraberry reservoirs. One fracture system consists of two vertical fracture sets with an apparent conjugate geometry (striking north-northeast and east-northeast). The other system consists of evenly spaced, northeast-striking vertical fractures, nearly bisecting the acute angle of the first system. Although lithologically similar, differences in quartz-overgrowth and clay content in the layers resulted in a yield strength of the lower bed that is only half of that of the upper layer, producing different fracture systems in the two reservoirs despite their proximity. Such differences in the mechanical properties, due to variations in diagenetic and depositional histories of the strata, are probably widespread within the formation. They have the potential to cause significant vertical and lateral variation in the Spraberry fracture system across the basin. Low present-day in-situ stresses in the reservoirs allow the fractures to open, to become more conductive, and even to propagate, under very low injection pressures.

Using sequential indicator Simulation as a tool in reservoir description : Issues and uncertainties

Flow simulation studies require an accurate model of the reservoir in terms of its sedimentological architecture. Pixel-based reservoir modeling techniques are often used to model this architecture. There are, however, two problem areas with such techniques. First, several statistical parameters have to be provided whose influence on the resulting model is not readily inferable. Second, conditioning the models to relevant geological data that carry great uncertainty on their own adds to the difficulty of obtaining reliable models and assessing model reliability. The Sequential Indicator Simulation (SIS) method has been used to examine the impact of such uncertainties on the final reservoir model. The effects of varying variogram types, frequencies of lithology occurrence, and the gridblock model orientation with respect to the sedimentological trends are illustrated using different reservoir modeling studies. Results indicate, for example, that the choice of variogram type can have a significant impact on the facies model. Also, reproduction of sedimentological trends and large geometries requires careful parameter selection. By choosing the appropriate modeling strategy, sedimentological principles can be translated into the numerical model. Solutions for dealing with such issues and the geological uncertainties are presented. In conclusion, each reservoir modeling study should begin by developing a thorough quantitative sedimentological understanding of the reservoir under study, followed by detailed sensitivity analyses of relevant statistical and geological parameters.

Object and Pixel-Based Reservoir Modeling of a Braided Fluvial Reservoir

To assess differences between object and pixel-based reservoir modeling techniques, ten realizations of a UK Continental Shelf braided fluvial reservoir were produced using Boolean Simulation (BS) and Sequential Indicator Simulation (SIS). Various sensitivities associated with geological input data as well as with technique-specific modeling parameters were analyzed for both techniques. The resulting realizations from the object-based and pixel-based modeling efforts were assessed by visual inspection and by evaluation of the values and ranges of the single-phase effective permeability tensors, obtained through upscaling. The BS method performed well for the modeling of two types of fluvial channels, yielding well-confined channels, but failed to represent the complex interaction of these with sheetflood and other deposits present in the reservoir. SIS gave less confined channels and had great difficulty in representing the large-scale geometries of one type of channel while maintaining its appropriate proportions. Adding an SIS background to the Boolean channels, as opposed to a Boolean background, resulted in an improved distribution of sheetflood bodies. The permeability results indicated that the SIS method yielded models with much higher horizontal permeability values (20–100%) and lower horizontal anisotropy than the BS versions. By widening the channel distribution and increasing the range of azimuths, however, the BS-produced models gave results approaching the SIS behavior. For this reservoir, we chose to combine the two methods by using object-based channels and a pixel-based heterogeneous background, resulting in moderate permeability and anisotropy levels.

Detection of cyclic patterns using wavelets: An example study in the Ormskirk Sandstone, Irish sea

This study shows how wavelet analysis can be used on well log and drill core data to identify cyclicity in sedimentary sequences. Three possible methods for determining wavelength were investigated: the Morlet wavelet, the Fourier transform, and the semivariogram. When applied to several hypothetical signals similar to those observed in petrophysical measurements in hydrocarbon reservoirs, all three methods could identify the presence of cyclicity. Only the wavelet scalogram, however, gave a clear indication of when the cyclic element was present and where frequency changes occurred in the signal. To illustrate the wavelet analysis, we processed well log and core data from a well in the Ormskirk Sandstone and determined the wavelet coefficients for each zone and the wavelengths of the strongest cyclicities. The cyclicities observed corresponded well with sedimentary features of the formation (e.g., channels and channel sets). Also, ratios of the cyclicity wavelengths corresponded with ratios of the Milankovitch precession, obliquity, and eccentricity periods. This result is in agreement with other investigators, who have proposed that Milankovitch-driven climate changes exercised an important control on Ormskirk Sandstone deposition.

Character, relative age and implications of fractures and other mesoscopic structures associated with detachment folds: an example from the Lisburne Group of the northeastern Brooks Range, Alaska

ABSTRACT Fractures and other mesoscopic structures formed at different times during the evolution of individual detachment folds in Lisburne Group carbonates of the northeastern Brooks Range. These structures provide clues to the mechanism of folding, the conditions under which folds evolved and the paragenesis of fractures in the fold-and-thrust belt as a whole. The earliest fractures strike NNW and probably represent orogen-normal extension fractures that developed in the foreland basin in advance of the fold-and-thrust belt. These rocks and fractures were later incorporated into the thrust belt, where they were thrust-faulted and folded. Later fractures, strained markers and dissolution cleavage developed during detachment folding as a result of flexural slip and homogeneous flattening. Fracturing associated with flexural slip occurred early in the development of folds. These early fractures were commonly overprinted or destroyed by ductile strain as later homogeneous flattening accommodated additional shortening. This penetrative strain was in turn overprinted by late extension fractures that formed during flexural slip in the waning phases of folding or after folding due to unroofing of the orogenic wedge. Early fracturing, overprinting by ductile structures and subsequent later fracturing in detachment-folded Lisburne Group emphasizes the importance of understanding the unique character and history of each fold-and-thrust belt in a successful hydrocarbon exploration effort. In particular, the mechanical stratigraphy and conditions of deformation play an important role in the type of fold that develops, the fold mechanisms that are active and the subsequent distribution and character of fractures and other mesoscopic structures. 1 Current address: National Park Service, Denali National Park, Alaska USA End_Page 121------------------------

Lithological and Zonal Porosity-Permeability Distributions in the Arab-D Reservoir, Uthmaniyah Field, Saudi Arabia: Discussion

Saner and Sahin (1999) considered the porosity and permeability distributions for the Arab-D reservoir in the Uthmaniyah field, Saudi Arabia, and presented summaries of the data by (1) reservoir zone and (2) lithofacies type. They identified three lithofacies types, granular, muddy-granular, and muddy, and described the permeability and porosity distributions by layer and by lithofacies. This approach, breaking down the lithofacies, is broadly following that promoted by Lucia (1999) because Saner and Sahin assumed that the pore types are lithofacies-controlled. This approach is suggested by the representative photomicrographs shown in Saner and Sahin (1999). Although they considered the variability, expressed by the coefficient of variation ( Cυ ), of both porosity and permeability and showed how this increases with depth in the reservoir (Figure 1), they did not appear to appreciate the wider implications of their analysis in their interpretation of the permeability data. Figure 1 Plot of coefficient of variation ( Cυ ) and tolerance (see equation 2 in text) …

Neural Network Modeling with Sparse Datasets

Abstract Neural networks are important tools for the analysis and modeling of many types of petroleum data. Small datasets limit their utility, however, because of the need to provide separate training and testing datasets. We train neural networks so that all the data are used for training, model development, and testing. We use the training procedure on a network to analyze fracture spacing in the Lisburne Formation, northern Alaska. Analyzing the effect of bed thickness on the spacing, we find only a weak influence, with closer fractures in thick beds. This result agrees with statistical analysis of the Lisburne data, but is contrary to relationships reported elsewhere.

A Virtual Conversation on the Impact of Geostatistics in Petroleum, Production, and Reservoir Engineering

Throughout the years, geostatistics has greatly influenced the broad spectrum of petroleum and earth sciences. In a directed conversation, five experts from respected university petroleum engineering departments discuss the impact of geostatistics on petroleum, production, and reservoir engineering and debate the current state of the art of geostatistical methodology in these disciplines.

Methodology for integrating analog geologic data in 3-D variogram modeling: Discussion

We commend Kupfersberger and Deutsch (1999) for tackling an important and difficult topic in geostatistical modeling in reservoir studies. They cover several important issues, including how the semivariogram reflects sedimentary organization and the relationship between vertical and lateral structure. In this discussion, we would like to add our views and experience regarding these issues by making four points regarding cyclicity. First, the appearance of cyclicity in the semivariogram (e.g., their figure 5) is an important indication that affects the modeling in both the vertical and lateral directions. Repetition of petrophysical properties may suggest a natural length scale in the vertical direction. This scale can then be used …

CO2 SEQUESTRATION POTENTIAL OF TEXAS LOW-RANK COALS

The objectives of this project are to evaluate the feasibility of carbon dioxide (CO{sub 2}) sequestration in Texas low-rank coals and to determine the potential for enhanced coalbed methane (ECBM) recovery as an added benefit of sequestration. The main objectives for this reporting period were to perform reservoir simulation and economic sensitivity studies to (1) determine the effects of injection gas composition, (2) determine the effects of injection rate, and (3) determine the effects of coal dewatering prior to CO{sub 2} injection on CO{sub 2} sequestration in the Lower Calvert Bluff Formation (LCB) of the Wilcox Group coals in east-central Texas. To predict CO{sub 2} sequestration and ECBM in LCB coal beds for these three sensitivity studies, we constructed a 5-spot pattern reservoir simulation model and selected reservoir parameters representative of a typical depth, approximately 6,200-ft, of potential LCB coalbed reservoirs in the focus area of East-Central Texas. Simulation results of flue gas injection (13% CO{sub 2} - 87% N{sub 2}) in an 80-acre 5-spot pattern (40-ac well spacing) indicate that LCB coals with average net thickness of 20 ft can store a median value of 0.46 Bcf of CO{sub 2} at depths of 6,200 ft, with a median ECBM recovery of 0.94 Bcf and median CO{sub 2} breakthrough time of 4,270 days (11.7 years). Simulation of 100% CO{sub 2} injection in an 80-acre 5-spot pattern indicated that these same coals with average net thickness of 20 ft can store a median value of 1.75 Bcf of CO{sub 2} at depths of 6,200 ft with a median ECBM recovery of 0.67 Bcf and median CO{sub 2} breakthrough time of 1,650 days (4.5 years). Breakthrough was defined as the point when CO{sub 2} comprised 5% of the production stream for all cases. The injection rate sensitivity study for pure CO{sub 2} injection in an 80-acre 5-spot pattern at 6,200-ft depth shows that total volumes of CO{sub 2} sequestered and methane produced do not have significant sensitivity to injection rate. The main difference is in timing, with longer breakthrough times resulting as injection rate decreases. Breakthrough times for 80-acre patterns (40-acre well spacing) ranged from 670 days (1.8 years) to 7,240 days (19.8 years) for the reservoir parameters and well operating conditions investigated. The dewatering sensitivity study for pure CO{sub 2} injection in an 80-acre 5-spot pattern at 6,200-ft depth shows that total volumes of CO{sub 2} sequestered and methane produced do not have significant sensitivity to dewatering prior to CO{sub 2} injection. As time to start CO{sub 2} injection increases, the time to reach breakthrough also increases. Breakthrough times for 80-acre patterns (40-acre well spacing) ranged from 850 days (2.3 years) to 5,380 days (14.7 years) for the reservoir parameters and well injection/production schedules investigated. Preliminary economic modeling results using a gas price of