Search

Operations planning in smart grids is likely to become a more complex and demanding task in the next decades. In this paper we show how to formulate the problem of planning short-term load curtailment in a dense urban area, in the presence of uncertainty in electricity demand and in the state of the distribution grid, as a stochastic mixed-integer optimization problem. We propose three rolling-horizon look-ahead policies to approximately solve the optimization problem: a deterministic one and two based on approximate dynamic programming (ADP) techniques. We demonstrate through numerical experiments that the ADP-based policies yield curtailment plans that are more robust on average than the deterministic policy, but at the expense of the additional computational burden needed to calibrate the ADP-based policies. We also show how the worst case performance of the three approximation policies compares with a baseline policy where all curtailable loads are curtailed to the maximum amount possible.

Approximate dynamic programming (ADP) driven adaptive stochastic control (ASC) for the Smart Grid holds the promise of providing the autonomous intelligence required to elevate the electric grid to efficiency and self-healing capabilities more comparable to the internet. To that end, we demonstrate the load and source control necessary to optimize management of distributed generation and storage within the Smart Grid.

Vehicles, both personal and commercial, have become ubiquitous forms of transportation in the developed world. The auto industry is amidst a technological transformation in identifying alternative sources of energy to power vehicles due to two driving forces: environmental pollution prevention and depletion of fuel resources. This driver for developing “smarter” solutions to create a “smarter planet” is crucial to advancing the science behind electric vehicles (EVs). EVs have been in existence since the mid-19th century, and electric locomotion has been the commonplace in many other vehicle types such as trains. The focus of this chapter is to discuss the feasibility of EVs in smart cities. In particular, the chapter explores the types of EVs, advantages and challenges faced by EVs to penetrate the market, and to outline state-of-the-art research and technologies that are driving the creation of newer and better EVs for adoption in the smart cities of tomorrow.

Electrification for commercial vehicle fleets presents opportunity to cut emissions, reduce fuel costs, and improve operational metrics. However, infrastructure limitations in urban areas often inhibit the ability to charge a significant number of electric vehicles, especially under one roof. This paper highlights a novel controls approach developed at GE Global Research in conjunction with Columbia University to fulfill the stated needs for intelligent charging of a commercial fleet of electric vehicles. This novel approach combines traditional control techniques with machine learning algorithms to adapt to customer behavior over time. The stated controls system is designed to regulate the charging rate of multiple electric vehicle supply equipment devices (EVSEs) to facilitate cost-optimal charging subject to past and predicted building load, vehicle energy requirements, and current conditions. In this embodiment, the system is primarily designed to mitigate electric demand charges that may otherwise occur due to charging at inopportune times. The system will be deployed at a New York City FedEx Express delivery depot in partnership with the local utility, Consolidated Edison Company of New York.

Technical innovation in the E&P business is surprisingly difficult. It is a puzzling anachronism because the improved performance that the use of high technology brings to the industry is often counted on to improve our collective performance in the future. In any industry, it is easy to identify the financial benefits that accrue from an overriding philosophy supporting innovation, and it is fair to say that the present technology and information world outside oil and gas has been built on just such a foundation. It is my view that our difficulty arises not from a lack of infrastructure but from the inherent complexities of our “manufacturing” process. There are just simply so many variables that can go wrong during implementation of an invention in our business.

The presence of salt within a sedimentary basin can potentially modify its temperature distribution and history. In particular, the positive thermal anomaly associated with the top of salt domes has attracted considerable attention in the past. The role of the lesser appreciated negative thermal anomaly associated with the base of salt domes in modifying the maturation history of subsalt source rocks is explored. The finite element method was used to model the transient and steady-state conductive temperature perturbations induced by salt layers, domes and pillows. The results indicate that the modification of the thermal regime due to evolving salt domes may significantly affect the maturation level of source rocks in the vicinity of the domes. Modelling the temperature structure of various salt structures has shown that, in general, the refraction of heat flow induces a dipole-shaped temperature anomaly; a positive anomaly located towards the top of the salt structure and a negative anomaly located towards its base. These dipole anomalies can be strongly asymmetrical, the degree of asymmetry depending on the shape of the salt structure and the proximity of the top of the salt structure to the surface of the basin. However, when the salt structure reaches the surface, the dipole-shaped temperature anomaly collapses to a monopole. Below the salt structure, all sediments, independent of their depth and lithology, are colder relative to a section with no salt. Similarly, salt domes that reach the surface very efficiently drain the heat from below and from the side of the dome. These negative thermal anomalies may extend for a considerable depth beneath the base of the salt dome and may reach values of −85°C locally. Because of the large contrast in thermal conductivity between the highly porous sediments and salt at lower temperatures, the efficiency of a salt dome to channel heat increases the closer the salt dome is to the surface. These results indicate that deep sedimentary basins containing salt are more prospective than basins without salt and/or salt diapirism. In addition to the structural traps associated with salt tectonics, salt domes and tongues connected to their source dissipate heat more efficiently and thus keep deeper regions of the basin relatively colder and potentially within the oil window for a longer time. This cooling effect is maximized when the top of the salt dome remains close to the surface of the sedimentary basin for a significant period of time and may be especially important for continental margins such as Brazil and offshore West Africa, where most of the source rocks lie beneath extensive evaporite deposits. In contrast, it was found that for the Gulf of Mexico basin, pre- and Early Tertiary salt diapirism and sheet emplacement may have caused significant delays in the maturation of subsalt source rocks in the deeper regions of the Gulf basin, but the maturation is likely to be relatively insensitive to the Late Miocene-Pliocene stage of salt mobilization because the time interval has been too short (

A high percentage of newly-constructed commercial office buildings experience energy consumption that exceeds specifications and system failures after being put into use. This problem is even worse for older buildings. We present a new approach, ‘predictive building energy optimization’, which uses machine learning (ML) and automated online evaluation of historical and real-time building data to improve efficiency and reliability of building operations without requiring large amounts of additional capital investment. Our ML approach uses a predictive model to generate accurate energy demand forecasts and automated analyses that can guide optimization of building operations. In parallel, an automated online evaluation system monitors efficiency at multiple stages in the system workflow and provides building operators with continuous feedback. We implemented a prototype of this application in a large commercial building in Manhattan. Our predictive machine learning model applies Support Vector Regression (SVR) to the building's historical energy use and temperature and wet-bulb humidity data from the building's interior and exterior in order to model performance for each day. This predictive model closely approximates actual energy usage values, with some seasonal and occupant-specific variability, and the dependence of the data on day-of-the-week makes the model easily applicable to different types of buildings with minimal adjustment. In parallel, an automated online evaluator monitors the building's internal and external conditions, control actions and the results of those actions. Intelligent real-time data quality analysis components quickly detect anomalies and automatically transmit feedback to building management, who can then take necessary preventive or corrective actions. Our experiments show that this evaluator is responsive and effective in further ensuring reliable and energy-efficient operation of building systems.

A one-dimensional model of fluid pressure and porosity evolution is used to investigate the physical processes that control the development and maintenance of overpressure in a compacting sedimentary basin. We show that for shale-dominated sequences the variation of the hydraulic diffusivity in both space and time is such that it produces a minimum between 2 and 4 km depth, consistent with observations from the Gulf Coast basin. This minimum inhibits the upward flow of fluid by acting as a “bottleneck” and thus determines the shallowest position of the depth to the top of overpressure. Above this region of bottleneck, overpressure does not develop because the porosity is sufficiently large to maintain high values of hydraulic diffusivity that are conducive to the rapid dissipation of excess fluid pressure. Within the overpressured shales, compaction propagates downward through the section, releasing fluids from the upper part of the section while continuing to restrain the upward flow of fluids from deeper within the section. As such, overpressures are predicted to be maintained within the deeper regions of a basin for tens to hundreds of millions of years. Further, fluid viscosity plays an important role in defining the depth behavior of hydraulic diffusivity as a function of time. Assuming a temperature-dependent fluid viscosity guarantees that the hydraulic diffusivity minimum will always exist during the development of the basin. On the basis of our results, we find that the depth at which the porosity equals 14±4% correlates with the depth to the local hydraulic diffusivity minimum and thus the depth to the top of overpressure. Moreover, we interpret that the 14±4% represents the threshold porosity for which a shale actually begins to act as a seal. Within the Gulf Coast basin, the gross sediment facies consists of lower massive shales across which deltaic systems have prograded allowing the deposition of an alternating series of sandstones and shales that grade vertically into massive sandstones. The massive sandstones are highly permeable and are connected hydrologically to the surface. We conclude that these sandstones play little role in the development of overpressure because of their high permeability except to the extent that the base of the massive sandstones marks the minimum depth possible for the top of overpressure. In contrast, overpressuring is observed to develop within either the shale-dominated sequence or the region of interspersed/interfingering sands and clays. The clay-encompassed sands play only a passive role in the development and maintenance of overpressure because it is the low-permeability clays that control the movement of fluids into and out of the sands.

An important problem in reliability engineering is to predict the failure rate, that is, the frequency with which an engineered system or component fails. This paper presents a new method of estimating failure rate using a semiparametric model with Gaussian process smoothing. The method is able to provide accurate estimation based on historical data and it does not make strong a priori assumptions of failure rate pattern (e.g., constant or monotonic). Our experiments of applying this method in power system failure data compared with other models show its efficacy and accuracy. This method can be used in estimating reliability for many other systems, such as software systems or components.

Power companies can benefit from the use of knowledge discovery methods and statistical machine learning for preventive maintenance. We introduce a general process for transforming historical electrical grid data into models that aim to predict the risk of failures for components and systems. These models can be used directly by power companies to assist with prioritization of maintenance and repair work. Specialized versions of this process are used to produce (1) feeder failure rankings, (2) cable, joint, terminator, and transformer rankings, (3) feeder Mean Time Between Failure (MTBF) estimates, and (4) manhole events vulnerability rankings. The process in its most general form can handle diverse, noisy, sources that are historical (static), semi-real-time, or real-time, incorporates state-of-the-art machine learning algorithms for prioritization (supervised ranking or MTBF), and includes an evaluation of results via cross-validation and blind test. Above and beyond the ranked lists and MTBF estimates are business management interfaces that allow the prediction capability to be integrated directly into corporate planning and decision support; such interfaces rely on several important properties of our general modeling approach: that machine learning features are meaningful to domain experts, that the processing of data is transparent, and that prediction results are accurate enough to support sound decision making. We discuss the challenges in working with historical electrical grid data that were not designed for predictive purposes. The “rawness” of these data contrasts with the accuracy of the statistical models that can be obtained from the process; these models are sufficiently accurate to assist in maintaining New York City's electrical grid.

Ensuring reliability as the electrical grid morphs into the "smart grid" will require innovations in how we assess the state of the grid, for the purpose of proactive maintenance, rather than reactive maintenance; in the future, we will not only react to failures, but also try to anticipate and avoid them using predictive modeling (machine learning and data mining) techniques. To help in meeting this challenge, we present the Neutral Online Visualization-aided Autonomic evaluation framework (NOVA) for evaluating machine learning and data mining algorithms for preventive maintenance on the electrical grid. NOVA has three stages provided through a unified user interface: evaluation of input data quality, evaluation of machine learning and data mining results, and evaluation of the reliability improvement of the power grid. A prototype version of NOVA has been deployed for the power grid in New York City, and it is able to evaluate machine learning and data mining systems effectively and efficiently.

We have developed a high-throughput Si CVD epitaxy system for photovoltaic application. The system was designed with high throughput, low energy cost, and high utilization of precursors, thus enabling extremely low cost Si epitaxial film for photovoltaic application. The system will be capable of processing a large batch of >320 wafers in less than an hour. The novel system architecture effectively utilizes resources such as electricity and process gases, and the processing cost can be as low as $0.22 per wafer for 50 µm deposition. Initial results confirmed good power utilization and excellent film properties. As an example of application, solar cells have been fabricated with epitaxial films on wafers from ingots pulled from metallurgical Si in Czochralski method.

A sedimentary basin is a thermochemical reactor; i.e., a complex, dynamically interrelated system that is deformed and faulted by tectonic forces, modified by sedimentation and erosion, influenced by geopressuring and chemical reaction, and subjected to constantly changing temperature, pressure, and fluid composition and movement. We understand little about the coupling among these processes that comprise basin thermochemical reactors despite many recent advances in basin science and technology. However, we are certain that these processes are coupled so strongly that changes in one can completely alter the operation of all the others in a basin.

Heat Flow values, determined from temperature logs and estimates of thermal conductivity from geophysical logs range from 23 to 37 mW/m2 from 800 to 2500 m depth in the Toa Baja scientific drillhole on the north, central coast of Puerto Rico. Near the target seismic reflector at the base of the well, an active hydrothermal system was encountered in which heat flow of up to 90 mW/m2 was found in a mineralized zone beneath a volcanic sill or flow. The heat flow then dropped to 50 mW/m2 beneath this subhorizontal flow zone. The mining of heat from downdip is proposed to account for this thermal anomaly, as well as the scatter in the heat flow determined from the few other wells drilled into Puerto Rico. The time-temperature history of the well indicates that Eocene volcaniclastics of the lower 2 km were deposited into a geothermal gradient of 60°C/km north of an active arc (heat flow estimated to have been 120–180 mW/m2). Uplift, erosion and cooling occurred between 40 and 30 Ma. Reburial and deposition of Oligocene-Miocene Limestones produced the present-day geothermal gradient of 15 °C/km (heat flow of 30–50 mW/m2). Based upon comparisons with slab cooling models, the crustal thickness beneath Puerto Rico is estimated to be closer to continental than oceanic.

Milankovitch cycles in the upper part of the Toa Baja drillhole, detected by geophysical logs, are used to estimate sedimentation rates, and to localize the depths at which sedimentation rate changes occur. The log-derived age information is then combined with reference dates from palaeontological data and the “Vail” global sea-level curve to build a detailed record of the chronology of the upper stratigraphic sequence in the well.

Geophysical logs from the Toa Baja drillhole on the north central coast of Puerto Rico are used to distinguish the variation in physical properties with depth in this non-volcanic island arc massif. A marked unconformity separates shallow Oligocene to Miocene limestones with considerable permeability contrast from a deeper sequence of volcaniclastic rocks with decreasing shaliness. A volcanic sill or flow overlying a strongly mineralized zone is found to cause the impedance contrast responsible for the strong seismic reflection anomaly (bright spot) that was the exploration target.

Reflectors imaged in a multichannel seismic line at Toa Baja, Puerto Rico, were studied by comparing the positions of reflection events with changes in rock properties determined through analysis of logs and cuttings. Reflectors appear to be associated primarily with changes in rock properties across facies or formation boundaries, and more rarely unconformities and faults. Faults that seal versus faults that leak can be recognized on the basis of gas Chromatograph data. Several laterally discontinuous reflectors characterized by modest amplitude anomalies seem to correlate in depth with minor shows of methane gas, measured by onsite gas chromatography. Gas was not recognizeable using geophysical logs indicating small quantities. The gas is apparently associated with rock types and rock properties which affect reflector amplitude rather than directly enhancing reflector amplitude.

Abstract Tracking the flow of oil, gas, and water in real-time while reservoirs are being drained is now possible through the emergence of rugged downhole sensors, the Internet, high bandwidth satellite and fiber communications and rapid middleware wrapping of software applications. The question is what information delivers economic return on the investment required. The good news is that all along the value chain, prices are coming down because of commercial pressures from outside the energy business. For example, Direct PC will likely drop the cost and increase the bandwidth of satellite communications to remote locations by an order of magnitude over the next few years. Application The manufacturing process by which we produce energy is being reinvented like we have never before experienced in this business. The oil and gas field of the future will be part of a much larger Information Technology (IT) network. Each field will be a wired, internet-connected, real-time monitored, remotely controlled, electronic venture. Each well, pipeline, rig, production platform, compression facility, and even the pumpers themselves, will have a IP (Internet Protocol) address. Any browser on any laptop in the company (with proper password protection) can log onto the myoilcompany.com URL at any time of the day or night -- from anywhere in the world -- and visualize production, well test, logging, and other real-time measurements coming from the 4D monitoring of field performance. Results and Observations However, there are fundamental reasons why such an integrated IT system for the oil and gas field is a difficult mission for the energy industry to pull off successfully. Principal among them is that the "last mile," as it is called in the communications industry, (the field connectivity to the manufacturing facility itself) is about as difficult an IT environment as could be imagined. Other IT savvy manufacturers around the globe also deal with multiple vendors, millions of customers, and millions of parts as we do, but our SITE of manufacture of energy products is NOT usually in a metropolitan area with communications infrastructure in place. Instead, it is found in whatever water depth or remote geography (usually rural) the oil and gas happens to be discovered beneath. And then, of course, the "very last mile" to the resource itself is under the ground and cannot be visited, or even imaged clearly. Not only must energy companies deal with a constantly changing set of IT vendors and bandwidth constraints from field to field, but each must be specified from the start to operate continuously for 30+ years, as part of an internetconnected portfolio. In the future, oil and gas fields will form an information grid with other fields, pipelines, land, and seaborne traffic, refineries and storage facilities, etc. that form the complete enterprise of the company (Figure 1).

Abstract The developments in the deepwater triangle of the Gulf of Gulf of Mexico, Brazil and West Africa are setting new standards for the design of all aspects of the "oil field of the future". Such new projects could provide "green-field" opportunities to implement integrated design/build/support processimprovements from the aerospace industry. Boeing terms their new manufacturing paradigm an "advanced digital enterprise", of which visualization is a key component. Implementation of advanced tools and processes used by Boeing for the design and construction of offshore facilities might lead to revolutionary cost and cycle-time savings for the energy industry. Introduction The purpose of this paper is to look at how visualization is used in modern aerospace manufacturing, and at the prospect of enough commonality for it to be exportable to the energy business. Based on what we have seen of visualization technology in the energy industry, it is clear that the issue is not one of technology. The energy industry is clearly on the leading edge of visualization technology. For example, there have been significant advances in discovery and delineation of new oil and gas reservoirs through the use of 3D visualization to characterize subsurface geometries. This paper will look at the next step, or how visualization technology can be integrated into the total life cycle of upstream processes to increase the performance of the overall enterprise. Energy and Aerospace At first glance, the similarities between energy and aerospace seem few and far between. However, looking at the big picture, there are a several similar attributes:Both deal with large complex capital projectsBoth end products have long life spans and operate in hostile environmentsFailure at anytime can be catastrophic, both economically and in loss of life?Economic success is dependent on keeping to cost and time schedules Given this commonality, we hope that interaction of people and technologies between these different worlds can produce common benefit through the use of advanced tools and processes. The Advanced Digital Enterprise For many years there has been promise of the paperless manufacturing environment and the many productivity benefits such will bring. We have seen evolutionary benefit in many industries over the past ten years, but seldom do we see a truly revolutionary benefit, as has been discovered in aerospace. This is primarily because of the large scale and scope of the impact on the overall organization required to "going completely digital." To this end, Boeing implemented a program called the "Lean Engineering Thrust" specifically targeting the design, build and support of its large, complex military and commercial aircraft systems. The idea was not to take our paper world and digitize it, but to re-look at all our processes and determine what they would look like, and if they were even needed in a digital environment. Then, if they were, how could they be improved within the overall digital process? Out of this exercise came a proprietary product called the Digital Manufacturing and Production System (DMAPS).

Abstract The oil industry is still staggering from the recent price collapse, with management energy focused on cutting costs and improving return on capital employed (ROCE). The major fiscal problem of the energy business is that it is not competitive as an investment vehicle when compared to other growth industries such as computing, the internet, and biomedicine, because our ROCE is so poor. While new exploration hotspots like offshore west Africa and the ultra deepwater Gulf of Mexico offer the promise to return >30% ROCE, refining and marketing is a drag at Thus, reservoir development plans that deliver cash when it is needed for a company are required, and in all-important fields, 4D Reservoir Management becomes essential. The costs of repeated acquisition of 3D seismic surveys and continuous downhole instrumentation and monitoring become cost effective near term investments when considered in this long-term cash flow framework. There are several examples from key fields in prime offshore areas where it has already been demonstrated that 4D Reservoir Management is a key to economic success of the basin. We will review a Gulf of Mexico field where 4D seismic monitoring produced significantly different drainage patterns from those expected, and early on in the life of the field, as well. Extraction of as much of the discovered oil and gas in known reservoirs is a critical capability that will be required to balance supply/demand in the 21st century. 4D Reservoir Management returns substantial capital versus that invested, and therefore is an essential component of responsible Business Unit management in the modern age. Introduction to Portfolio Management (PM) Oil and gas production companies make money based upon on their skills in identifying a portfolio of properties and utilizing technologies to discover, produce, and sell oil and gas produced from those properties in an optimal manner. The key ingredient to improved business performance is portfolio management, which allows a company to present the performance of all its producing properties and exploration targets in a normalized way.

Abstract 4D seismic has become a widely accepted technique to interpret changes between successive 3D seismic surveys in terms of fluid substitution and pressure depletion in a producing reservoir. However, most time-lapse studies have been mostly qualitative and based on simplified reservoir representation in order to adjust to the time constraints of today's oil market. In this paper, we present how reservoir simulation constrained by stochastic characterization and non-linear optimization can be used in an integrated series of tools to refine 4D interpretation. Because of its direct relationship with pore fluid content and properties, we use seismic impedance rather than seismic amplitude as the primary data between the various steps of our 4D interpretation loop. Non-linear inversion of the 3D seismic data sets allows a preliminary interpretation. Stochastic simulation of the lithology and porosity, constrained by these "observed" impedance volumes and by well logs, provide the static reservoir characterization for the reservoir simulator. Once simulated production matches the recorded production history, empirical or Biot/Gassmann-type petrophysical models are used to calculate the "simulated" impedance volume from the fluid saturation and pressure distribution calculated by the reservoir simulator. Non-linear optimization is used iteratively to improve first the production history match and next the agreement between "simulated" and "observed" impedance volumes over time. This optimization is performed over a limited set of poorly constrained parameters in the permeability calculation and petrophysical models. In the case study presented here, the analysis of a turbidite reservoir in the South Timbalier 295 field, our results show how stochastic characterization helps reproducing the complexity of reservoir fluid dynamics while the results of the optimization underlines the robustness of the 4D interpretation despite the large amount of unknowns. Introduction Time-lapse, or 4-D, seismic monitoring is an integrated reservoir exploitation technique based on the analysis of successive 3D seismic surveys. Differences over time in seismic attributes are directly related to changes in pore fluids and pore pressure during the drainage of a reservoir under production. The detection of areas with significant changes or with unaltered hydrocarbon-indicative attributes, can be used to determine drilling targets where hydrocarbons remain trapped after several years of production. Making sure that seismic differences are related to fluid flows is critical for a complete time-lapse seismic study. Noise associated with differences in acquisition can generate seismic differences between surveys that are not related to the reservoir drainage pattern. In this paper, we describe how reservoir simulation can be used to generate independent impedance maps to validate or constrain 4-D impedance maps obtained from the inversion of successive legacy data sets. A complete 4D analysis is an iterative loop where the original interpretation can be refined along the later steps. First, we summarize the steps preceding the reservoir simulation, including the 3D seismic processing and inversion, and the preliminary time-lapse interpretation. We then describe the elastic models, the properties of reservoir fluids and the reservoir characterization that can be used to link the impedance volumes to fluid and lithology distributions in the reservoir.

4-D seismic reservoir simulation combines the analysis of 4-D (time-lapse) seismic changes measured in an oil and gas field with 3D elastic seismic modeling, reservoir characterization and fluid flow simulation to better understand drainage patterns of oil, gas and water into wells and to identify bypassed pay. We have solved this poorly constrained, inverse problem and arrived at a selfconsistent reservoir simulation that minimizes error and predicts 4-D seismic changes similar in space, time, and magnitude to those observed in a complex, intertwined, turbidite channel reservoir in the South Timbalier 295 field, Gulf of Mexico. Particle flow produced from the seismic reservoir simulation predicts that drainage was complex, with a pressure drop in the reservoir to below the bubble point producing a gas cap nearby the producing wells. Brightened seismic amplitudes both downdip and updip of the wells indicate that the gas coming out of solution filled the most permeable of the tubular turbidite channels. Poor oil migration downdip suggests the need for water injection, which the operator instituted in 1997 to recover this downdip oil and repressurize the reservoir. Additional production drilling must be placed precisely to target the tubular turbidite channels not filled with gas. The seismic reservoir simulation suggests that this enhanced recovery strategy should be successful. As we get better and better at 4-D seismic reservoir simulation, we should get more and more of the original oil-in-place from complex reservoirs.

Abstract We have imaged 4D seismic amplitude and impedance differences between 3D seismic surveys conducted in 1988, prior to production, and in 1994, when about half of the recoverable hydrocarbons had been produced, to attempt to quantify the variations in fluid saturations with time within turbidite reservoirs in a Gulf of Mexico oil field. The mission of this 4D seismic monitoring project was to develop the software tools and interpretation techniques 1) to isolate and interpret seismicchanges that have occurred in the field over time in terms of oil/gas/water changes in order to better understand the drainage of the field, and 2) to identify and locate bypassed pay. Two turbidite reservoirs of the South Timbalier 295 field, offshore Louisiana, Gulf of Mexico, provide contrasts in acoustic response that point to important lessons for 4D monitoring in other fields of the Gulf of Mexico. In the K-40reservoir, a "classic" updip water sweep pattern in the oil/water contact with time was deduced from the observation of the dimming of seismic amplitudes over time in this vigorous water-drive reservoir. A "finger" of dimmed seismic amplitudes that breaks updip of even the structurally highest perforations was observed between two of the producing wells. This finger may be related to sand quality and permeability pathways and/or to acoustic shadowing from brightening that occurred within the K-8/16 reservoir directly above. The K-40 sand is two-lobed, and our region growing analysis of 4D seismic changes suggests that water had migrated farther updip in the lower than in the upper lobe by 1994. In the K-8 reservoir sequence, complex depositional channeling appears to have isolated the sand from the regional aquifer system to produce a gas depletion drive reservoir. Volumetric region-growing about high seismic amplitude "seed" pointsproved very successful in mapping the turbidite fairways, but not in isolating the K-8 from the immediately underlying K-16 sand. Seismic differences between 1988 and 1994 indicate brightening over time occurred in the K-8 sand fairways, which is thought to be associated with increased exsolution of gas. A marked increase in GOR was observed between the times of the two seismic surveys in producing wells from the K-8 reservoir. Introduction The interpretation of drainage patterns and the tracking of the migration of oil, gas and water fronts in 4D, that is volumetrically with time, are valuable optimization tools for the economic development of producing oil and gas fields1. However, current production technology is predominantly 2D. Even when a 3D seismic survey is used, 2D map-based, interpretation "horizons" are derived and used by geologists, geophysicistsand, reservoir engineers, to plan and execute production decisions in virtually all currently active oil and gas fields. This process relies upon the accumulated experience-base of these integrated teams that use seismic, logging, pressure monitoring, and production histories to determine how oil, gas, and water move within reservoirs as production proceeds.

Twelve years ago, reading a passage from the submarine novel The Hunt for Red October by Tom Clancy was as dose as any exploration geophysicist got to gravity gradiometry. This early technique in Gulf Coast exploration, which faded from use with the development of modern gravity instrumentation in the 1930s, had been relegated to brief historical sections of introductory texts. In the 1970s, driven by both navigation and missile launching requirements, the U.S. Navy spent hundreds of millions of dollars developing a system to measure gravity gradients; this system was somewhat more complex than the fictional one Clancy installed on the Red October. The end of the Cold War triggered the introduction of classified military technology to exploration geophysics and other fields. Three years ago the U.S. Navy began to explore civilian applications for submarine gravity gradient technology. This article describes gravity gradients, the developing uses of gravity gradiometry in exploration, and future possibilities for the technique.

One of the most important features of time‐dependent reservoir characterization is the added additional constraints that can be incorporated to update the initial reservoir model. Uncertainty of the model may be gradually reduced as the iteration of time‐dependent reservoir characterization continues.

Gravity gradiometry has the potential to become a powerful exploration tool with enhanced resolution relative to classic marine gravity. The improved resolution of gravity gradiometry for exploration applications is equivalent to the improvement 3-D seismics provided over 2-D seismics. Over the last 2 years, gravity gradiometry acquired over exploration targets has been acquired with the Bell Aerospace gravity Gradiometry Survey System (GSS). The GSS is a formerly classified gravity sensing system that contains the world's only moving base gravity gradiometer. The system measures both gravitational acceleration and gravity gradients, yielding six measurements that define the local gravity field and its gradients in three dimension. This paper will present an overview of gravity gradiometry applications illustrated with actual data from the Gulf of Mexico. Exploration examples of high-resolution gravity gradiometry use include enhancement of seismic interpretations, identification of sub-salt sediment fairways and exact location of salt body boundaries.

The fifth Archie Conference, "Visualization Technology to Find and Develop More Oil and Gas," brought together 130 scientists and technologists to review current and future visualization technologies that are being developed and used in the petroleum and other industries. Visualization in the oil and gas industry can be considered a tool for characterizing and understanding surface and subsurface phenomena. In addition to allowing one to view and more easily understand large quantities of data, visualization is dramatically enhancing communications, and thus interaction, among members of integrated exploration and development teams. Current and potential end-users of visualization technology consider the most important aspects to include common formats for data interchang , greater availability to consultants and independents (perhaps through PC-based visualization hardware and software), bigger bandwidth capabilities to drive more powerful machines, and the application of sensitivity analysis to document uncertainty in visualizations at all scales. Visualization technology is in its infancy, but growing so rapidly that it promises to have major impact on many aspects of the petroleum industry, from improved day-to-day communications to better technology transfer and more powerful interpretive capabilities, all of which can ultimately lead to better economic decision making.

A new elemental oxide classification scheme for crystalline rocks is developed and applied to geochemical well logs from the Cajon Pass drill hole. This classification scheme takes advantage of measurements of elements taken by a geochemical logging tool string. It uses K_2O versus SiO_2/Al_2O_3 to distinguish between granites, granodiorites, tonalites, syenites, monzonites, diorites, and gabbros. Oxide measurements from cores are used to calibrate the elemental abundances determined from the well logs. From these logs, a detailed lithologic column of the core is generated. The lithologic column derived from the well log classification scheme is compared with a lithologic column constructed from core samples and well cuttings. In the upper 1295 m of the well, agreement between the two columns is good. Discrepancies occur from 1295 to 2073 m and are believed to be caused by the occurrence of rock types not distinguished by the classification scheme and/or the occurrence of secondary minerals. Despite these discrepancies, the well log-based classification scheme helps to distinguish changes in rock type and shows potential as an aid to the construction of lithologic columns in boreholes of crystalline rocks.

Geochemical well logging provides a continuous record of the variations in elemental abundances of the major rock-forming oxides of Si, Al, Ca, Fe, Ti, and K, as well as S, Gd, U, Th, and the H and Cl in the formation and pore fluid. Through the additional measurement of the photoelectric capture cross section of the rock, the sum of Mg + Na can also be estimated. Though not as accurate as laboratory analyses of recovered core samples, the log-derived abundances are precise enough to define the degree and extent of alteration, to identify igneous lithostratigraphy, and to calculate integrated chemical exchange between the oceanic crust and seawater. In this paper, the elemental yields from geochemical logging in basalts are calibrated against extensive XRF analyses of cutting samples from the Lamont 2 test well into the diabases of the Palisades Sill, New York. Accuracy and precision of the log-derived analyses are determined in the lower part of the well, and calibration equations are derived, which are then tested against core-derived “standards” from the upper part of the well. The calibrated, log-derived, elemental analyses are within one standard deviation of the core-derived results (except for the Mg + Na curve, which is somewhat noisier). These calibrations are then applied to geochemical logs from the oceanic crustal basalts of Ocean Drilling Program hole 504B, where core recovery was less than 20% of the section. The accuracy and precision of the calibrated, log-derived elemental abundances are tested against core-derived standards from seven dike and sill intervals. Then the corrected elemental analyses are used to derive a mineralogy model for hole 504B that shows the oceanic crust to contain secondary mineralization in the form of celadonites and smectites in the pillow basalts and chlorites in the dikes that are largely confined to fracture and breccia zones. Cyclicity in the Al and other elemental logs was found to vary with the abundances of these alteration products and with eruption and intrusion event boundaries. The geochemical logging data are then used to estimate the integrated chemical exchange resulting from hydrothermal alteration of the oceanic crust that has occurred over the last 5.9 m.y. in hole 504B. The primary change is from Ca loss and Mg gain caused by the reaction of basalt with seawater. A large Si increase found in the transition zone between the pillows and dikes is attributed to precipitation of quartz during mixing of hot, up welling hydrothermal fluids and cold, downwelling seawater at what was once a major permeability discontinuity. The present low-to-high permeability transition in hole 504B is found 500 m shallower. The K budget requires significant addition to the uppermost pillow basalts both from high-temperature depletion in the lower pillows and dikes and from low-temperature exchange with seawater. The geochemical logs further document that the total chemical exchange between the oceanic crust and seawater is as important to the long-term composition of the oceans as is the chemical input carried by rivers. Integrated “water/rock ratios” are then derived for the mass of seawater required to add enriched elements and the mass of hydrothermal fluid required to remove depleted elements in the oceanic crust of hole 504B. Whereas Ca, Mg, and K require relatively low water/rock ratios, high values for Si, Al, and Fe suggest that off-axis, ridge-flank exchange is as important to the total cation exchange budget as are ridge-axis processes.

Knowledge of the variation of thermal conductivity and heat production with depth is critical to understanding the thermal state of the crust. Continental and oceanic drill holes provide unique opportunities for the calibration of empirical and theoretical models relating crustal thermal properties to physical and chemical properties. These calibrations are based on more than 30,000 well log measurements of acoustic velocity (both compressional and shear), density, porosity, radioelement concentrations (K, U, Th), and major element chemistry (Al, Si, Ca, Fe, S, Mg + Na), in formations ranging in age from 5 to 620 Ma and in composition from granitic to gabbroic. The mineralogical data available from induced gamma ray spectroscopy logs, when combined with known mineral thermal conductivities in a model for the thermal conductivity of a crystalline aggregate, enable the in situ application of a laboratory estimation technique and yield continuous conductivity profiles. However, mineralogy-based conductivity models cannot account for the anisotropic conductivity observed in foliated metamorphic rocks. A phonon conduction model, which utilizes acoustic velocity and bulk density measurements, is accurate within ±15% in both isotropic and anisotropic formations of the upper continental crust and in oceanic crustal layers 2C and 3. In oceanic crustal layers 2A and 2B, however, the effects of fracturing on compressional and shear velocities lead to inaccurate results. The heat flow-heat production relationship, which provides one method of determining subsurface heat production, has well-documented limitations. Allis (1979) and Rybach and Buntebarth (1982, 1984), on the basis of laboratory measurements, have proposed quantitative relationships between heat production and rock compositional and physical properties. The well log K, U, and Th data indicate that heat production can be predicted from compositional data within a factor of 2 or 3. Predictions of heat production from acoustic velocity and bulk density measurements are only accurate to within an order of magnitude. Three factors account for these results: (1) the natural variability of velocity and density for a given rock composition, (2) the tendency of U and Th to associate with secondary rather than primary minerals, and (3) U and Th redistribution through fluid flow and metamorphism. Thus seismic velocity based heat production profiles can, at best, be taken as first-order estimates which may misrepresent the complexity of the continental crust. However, the correlation of formations of known heat production with subsurface structures, when combined with a suite of existing techniques for estimating heat production, may lead to more accurate results.

Hole 504B in the eastern equatorial Pacific is the deepest hole to penetrate oceanic basement, extending more than 1500 m beneath the seafloor. Two borehole televiewer (BHTV) logs were obtained in this hole, one through the upper 1000 m of basement during Deep Sea Drilling Project leg 83 and the other extending an additional 260 m deeper into the crust during Ocean Drilling Program leg 111. The two BHTV records have been combined and processed in terms of both acoustic amplitude and travel time in order to evaluate the extent and distribution of rock failure along the borehole wall. A histogram of borehole enlargements versus azimuth depicts a dominant breakout azimuth of N122.5°E which corresponds to the direction of minimum principal stress Sh and correlates well with stress orientations inferred from solutions to regional earthquake focal mechanisms. Furthermore, the bimodality of this histogram, with a secondary mode orthogonal to Sh, indicates that a significant number of enlargements are coalesced tensile fractures occurring along the orientation of SH, the maximum principal stress. The appearance of this orthogonal, bimodal distribution suggests that the regional horizontal stress field is highly anisotropic, a condition supported by seismic data which show all of the earthquakes in the area to be strike-slip. The frequency of borehole enlargements increases with increasing depth and depicts a systematic structural deterioration of the well bore. The tensile fractures along the SH azimuth contribute to this degradation and appear to be induced by thermal stresses due to the injection of cold water into hot rock. The frequency of these extensional features does not increase with depth. Rather, their appearances can be directly correlated with shipboard efforts at deliberately cooling the well and/or with the sudden resumption of drilling after the hole had been allowed to reequilibrate thermally for several days. These latter borehole enlargements are more pronounced then those commonly associated with hydraulic fracturing. The mechanisms for fracture initiation and growth, based upon temperature contrasts between the well bore fluid and the adjacent rock may enhance rock failure.

Summary. The East Pacific Rise at 12-15' S is topographically smooth with a crestal horst or linear volcanic peak marking the present axis of spreading. The Galapagos Rise at 14-17's is topographically rough with a possible central graben marking the extinct spreading axis. The seafloor spreading magnetic anomalies on the East Pacific Rise are of low amplitude, but fracture-zone anomalies at 13-14's have amplitudes of up to 1250 nT. Anomalies of this amplitude at the magnetic equator must be formed within the fracture zone by some combination of block reversal boundaries, anomalously-high magnetic intensities, and/or anomalously-large thicknesses of the magnetic layers within the fracture zone. Magnetization and majorelement chemical analyses of basalts dredged from four locales along the fracture zone indicate that the large magnetic-anomaly amplitudes are caused by the high iron and titanium content of these ferrobasalts. The magneticanomaly profdes from the Galapagos Rise and its fracture-zone system are of normal amplitude and are extremely difficult to correlate internally or with the geomagnetic timescale. Eighty-one heat-flow measurements indicate that the values measured are controlled by sediment thickness. Where the thickness of the sediment blanket is greater than loom, high heat flow is measured and possibly is representative of the total heat transfer at the seafloor. Where the sediment thickness is less than loom, seawater circulation in the oceanic crust is thought to remove most of the heat convectively; thus causing low conductive heat-flow values to be measured by the usual heat-flow apparatus. The heat loss by convective processes is probably a function also of topographic roughness and sediment permeability.

Two in situ measurements of principal stress directions have been made in DSDP Holes 504B, south of the Costa Rica Rift on the Nazca plate, and 597C, west of the East Pacific Rise on the Pacific plate. In both cases, the orientations of in situ principal stresses determined from borehole breakouts are consistent with the stress directions inferred from intraplate earthquakes located near the sites.

Closely spaced heat flow surveys at four sites on the flanks of the Central Indian Ridge and the Southeast Indian Ridge delineate a pattern of oscillatory heat flow which can only result from cellular convection of oceanic bottom water through the oceanic crust and overlying sediment. These cells have a wavelength of 5 to 10 kilometers and are presently active in sea floor 18 x 10(6), 25 x 10(6), and 45 x 10(6) years old of the Crozet Basin and in sea floor 55 x 10(6) years old of the Madagascar Basin. The precise measurement of nonlinear temperature profiles makes it possible to calculate the conductive and convective heat transfer components through the sea floor. Even in the oldest sites, geothermal convection is still a major component of heat transfer through both the crust and sedimentary layers. These observations coupled with the results of earlier oceanwide geothermal studies indicate that more than one-third of the entire surface area of the world's ocean floor contains presently active geothermal convection that is cellular in plan form.

Summary The spatial pattern of cooling near a spreading ridge crest was investigated with a suite of 71 precisely-navigated heat-flow stations on the Galapagos spreading centre, East Pacific, near 86"W longitude. Stations are on crust less than 1.0 My old on which bathymetry and sediment distribution are well known. Values vary from near zero to greater than 30 HFU (10-6 cal cm-2 s-1). The average over the entire region is significantly less than that predicted by theoretical conduction models of a cooling lithosphere. We observe a regular variation of the heat-flow pattern with a wavelength of 6 & 1 km approximately normal to the ridge crest. Heat-flow maxima are characteristically located near faults and local topographic highs. The locations of fields of small sediment mounds, apparently hydrothermal vents, are also restricted to these faulted, elevated areas of high heat flow. Near-axis, bottom water temperature anomalies of several hundredths "C were detected. The low average, the low minima in the heat-flow pattern, and the water temperature anomalies suggest that hydrothermal circulation accounts for approximately 80 per cent of the geothermal heat released near the ridge crest. We conclude that the hydrothermal circulation pattern is controlled by one or more of the following physical properties of the system : highly developed cellular convection, discrete zones of high permeability, variation in the strength of heat sources near the base of the crust, or bottom topography. Our results imply that heat-flow studies near active oceanic ridges will be of most value if they are sufficiently detailed and well navigated to define the systematic small-scale variations that appear to be caused by hydrothermal circulation.

Sixteen new heat flow stations plus a seismic reflection, bathymetric, and magnetic survey of the north flank of the Galapagos Spreading Center have provided additional data to further delineate a low heat flow zone approximately 50 km wide running from 85.5°W to 88°W. This anomalous, low heat flow zone appears to be related to a graben which is found as a continuous, fault-bounded trough, offset by fracture zones and trending parallel to the present spreading center. Thin sediment, symmetric magnetic anomalies, shoaling topography, and a topographic step between the graben and the present spreading center suggest that it is the central rift valley of a fossil spreading center which jumped to the south ∼4 m.y. B.P. The low heat flow measurements appear to be a manifestation of our inability to measure convective heat transfer in this area of thin sediment cover. The thin sediment cover appears to be caused by (1) the youth of the fossil ridge crest relative to the surrounding sea floor and (2) burial by extrusive activity which might have continued after the major components of sea floor spreading jumped to the presently active ridge crest.

We examine available marine geophysical and seismological data from the Caroline Sea region and conclude that a separate Caroline plate currently exists. The Caroline plate is moving relative to the Pacific plate on its northern and eastern boundaries, the Philippine plate on its western margin, and the Indian and smaller plates along its southern side in New Guinea and the Bismarck Sea area. The southern Yap Trench, the Palau Trench, and an accreting plate boundary within the Ayu Trough manifested as an axial rift valley comprise the Caroline-Philippine plate boundary. On the basis of sediment thickness and subsidence of basement away from the rift, we estimate that the Ayu Trough started to open during the Miocene. The northern section of Pacific-Caroline plate boundary coincides with the Sorol Trough which exhibits both strike-slip and extensional characteristics. The southeastern section of this boundary occurs along the Mussau Trench where Caroline plate underthrusts the Pacific plate. The section of plate boundary between the Sorol Trough and Mussau Trench is characterized by highly unusual deformational tectonics. Convergence between the Pacific and Caroline plates is apparently accommodated here by overthrusting of small slivers of sea floor towards the northeast. The intensity of deformation appears to increase southward towards the Mussau Trench. Our calculated instantaneous angular rotation vector for the Pacific-Caroline plates predicts that convergence rates increase uniformly south along the overthrust and underthrust sections of plate boundary. The transition in tectonic style from overthrusting to underthrusting occurs between 3° and 4°N.

Thirteen new and fourteen published earthquake fault plane solutions are used to examine the directions of the relative plate motion along the boundaries of the Nazca plate on the East Pacific Rise, Galapagos Spreading Center, and Chile Rise. In general, slip vectors agree well with one another and with the directions of plate motion given by Minster et al. A few solutions that disagree with these inferred directions of plate motion corroborate the independent evidence and suggestions of Forsyth and Herron that two small plates exist along the East Pacific Rise crest. Together, magnetic anomalies, bathymetry, seismicity and the fault plane solutions suggest that the northernmost of these two plates, the Easter plate, is bounded on the north by transform faults and on the south by spreading centers. The existence of the diamond-shaped Easter plate may be related to (1) its close proximity to Easter Island (a proposed hot spot) or, (2) the very high velocities of plate motion and, therefore, the thin lithosphere in this region.