GRANTEE: WOODS HOLE OCEANOGRAPHIC INSTITUTION
Department of Geology and Geophysics
Woods Hole, Masschusetts 02543
GRANT: DE-FG02-94ER14435
TITLE: Robust, Controlled Leverage Magnetotelluric Data Analysis
PERSON IN CHARGE: A.D. Chave (508-289-2833; Fax 508-457-2150; E-mail alan@faraday.whoi.edu)
Objectives: To develop an improved understanding of the causes of bias and variability in magnetotelluric response function estimates, particularly in the presence of source field problems and cultural noise, to develop new processing methods which will eliminate these problems, and to apply tensor decompositions for galvanic distortion to the response functions in an automatic fashion.
Project Description: Over the past decade, a collaboration with D.J. Thomson of AT&T Bell Laboratories has resulted in major improvements to Magnetotelluric data processing methodology, including robust remote reference algorithms which eliminate outliers in the electric field and extension of the jackknife to obtain nonparametric estimates of confidence limits. However, these methods do fail under some circumstances. This research will focus on further advances which will be driven by analysis of diverse types of data, including those of interest in industry.
The major problems to be addressed include:
1. Investigating the efficacy of multiple remote references.
2. Designing algorithms which control the influence of extreme data in the magnetic as well as the electric field (which conventional robust algorithms cannot do).
3. Tests of the hypothesis that statistically-based data sorting can remove many of the confounding influences which sometimes affect Magnetotelluric data.
4. Tests of the importance of magnetic field galvanic distortion in a variety of data.
Results: A robust, controlled-leverage algorithm has been developed which automatically removes the influence of outlying electric field data (as for conventional robust methods) and also controls the leverage effects of unusual magnetic field data. The latter is accomplished by weighting based on the size of the diagonal elements of the hat matrix, and has necessitated the derivation of the statistical distribution of this quantity for Gaussian data which has evidently not been done previously. The algorithm has been extended to allow multiple instead of single remote references. This involves computing the projection of the local magnetic field from all of the remote sites, and reduces to the conventional remote reference method when only a single reference site is available. In addition, the projection operation is performed robustly, opening up a new way of removing cultural noise in the local magnetic field if the remote site is not affected by it. The multiple remote reference, robust, controlled leverage algorithm with jackknife error estimates has been programmed and is being tested with a variety of data to refine the approach. The past year has been devoted to testing and debugging of this algrorithm, and to its application in a variety of places with a diversity of data. A manuscript describing these results is in the final stages of preparation.
An X-windows graphical user interface to the robust controlled leverage code has been completed, and is undergoing final testing. This interface greatly simplifies setting up and executing runs, and in particular facilitates viewing graphical output from a run. Without such an interface, it is very difficult to look at the statistical output from a run, and hence understand what is going on.
GRANTEE: WOODS HOLE OCEANOGRAPHIC INSTITUTION
Department of Marine Chemistry and Geochemistry
Woods Hole, Massachusetts 02543
GRANT: DE-FG02-92ER14232
TITLE: Geochemical Incorporation of Sulfur into Organic Matter: Role of Sulfur in the Formation and Diagenesis of Macromolecular Organic Matter in Sediments
PERSON IN CHARGE: Timothy I. Eglinton (508-289-2627; Fax 508-457-2164; E-mail teglinton@whoi.edu)
Objectives: The overall goal of this study is to improve our understanding of the role of sulfur in organic matter (OM) diagenesis. Two specific objectives are outlined: (1) development and evaluation of structural models quantitatively describing organically-bound sulfur (OBS) in sediments; (2) application of these models for (a) assessment of sulfur incorporation into organic matter as a preservation (kerogen formation) mechanism and (b) elucidation of diagenetic pathways for OBS.
Project Description: In marine sediments, sulfur is intimately involved in organic matter diagenesis. Several recent studies suggest an important role for sulfur in the formation of macromolecules - the dominant OM pool in sediments. The experimental and theoretical approach is based on the premise that OBS can be represented as model structures that differ in terms of the mode of sulfur attachment (i.e. inter-molecular linkages vs intra-molecular bonds), as well as the number of linkages per molecule and the number of S atoms in each linkage. These models are important for accurate prediction of organic carbon burial efficiency in sediments and delineation of temperature-time requirements for petroleum generation. X-ray absorption spectroscopy (XANES) and temperature programmed reduction (TPR) are used for determination of S-speciation; chemical degradation experiments provide information on linkage type, the sites of S-attachment, the number of linkages involved and the molecular structures of the S-containing molecules; analytical pyrolysis is used to derive structural information and to estimate organic S content; S-isotopic measurements yield information on the timing of diagenetic S incorporation. These analyses are performed on fractions isolated from sedimentary OM according to approximate molecular size and/or solubility. The Peru margin and Miocene Monterey Fm, CA are the primary study areas because of their high sedimentary organic carbon and sulfur contents, and equivalence in terms of (paleo)depositional environment.
Results: The primary research focus over the past year has been to undertake detailed sulfur isotopic studies of co-existing inorganic and organic sulfur pools within discrete sedimentary phases. The impetus for this aspect of our research is to better define the mode and timing of sulfur addition to sedimentary organic matter (and hence predict its distribution in sediments). Prior studies on crude sub-fractions of sedimentary sulfur have been limited by isotopic heterogeneities within sediment phases. Specifically, inadequate discrimination of "diagenetic" vs "biogenic" organic sulfur has frustrated attempts to pinpoint the timing of OBS formation with respect to other [inorganic] sulfur sinks. To address these issues, sediments have been subjected to three different separation schemes: (i) SPLITT fractionation; (ii) Gel Permeation Chromatography (GPC); (iii)
Preparative Capillary Gas Chromatography (PCGC). SPLITT (field flow) fractionation was used to separate particles from a surficial Peru sediment sample according to size/density. Sulfur isotopic measurements performed on pyrite-S, insoluble (kerogen) and soluble (bitumen) organic-S from ten different particle size fractions ranging from > 250 µm to < 1 µm revealed that pyrite and kerogen S each display a ca. 10 ‰ variation in 34S values, whereas bitumen S exhibits much narrower (ca. 2 ‰) spread in values. Pyrite S systematically yielded the most depleted sulfur isotopic ratios irrespective of particle size, suggesting the reactions of iron prior to organic matter with available sulfide. In both pyrite and kerogen S-pools, isotopic values generally became more enriched with increasing particle size, indicating that sulfur addition reactions preferentially occur in the smallest (highest surface area) size fractions. These results agree with the dominance of reduced organic sulfur forms (mainly organic polysulfides and sulfides) in the smaller size fractions. Interestingly, the particle sizes containing the most depleted isotopic values for the pyrite and kerogen S do not coincide, suggesting that biological or mineralogical control is superimposed on this relationship. Also of note was a distinct maximum of kerogen S isotopic values for a larger particle class (63-38 µm) which likely reflects a significant contribution of biogenic sulfur associated with partially intact diatom debris. Bitumen S shows a more uniform isotopic distribution implying that particle size has a minimal influence its timing of formation.
To further examine isotopic relationships within different organic matter pools, bitumen samples from a Peru and Monterey sediment were separated according to molecular size by GPC prior to isotopic analysis. For the Peru sample, the most depleted isotopic values were observed in intermediate molecular size (800-3000 daltons), with more enriched values observed for molecular sizes > 5000 daltons and < 500 daltons. These results are consistent with the premise that early diagenetic "vulcanization" of lipids and sugars forms cross-linked geopolymers. The more enriched values for larger and smaller molecular sizes likely reflects contributions for biogenic organosulfur compounds (peptides, sulfonated lipids etc.).
Finally, we have successfully developed and tested a method to determine the extent of sulfur isotopic heterogeneity at the molecular level. This approach is designed to provide a definitive answer concerning the sulfur isotopic composition of diagenetically derived OBS. A suite of C20 isoprenoid thiophenes, a steroidal thiophene (both diagenetic in origin) and synthetic standards were recovered from a Monterey shale extract by preparative capillary gas chromatography (PCGC). Using PCGC we isolated >100 µg of each target compound (equivalent to 0.2 µM S) in high purity. Microscale determinations on the standards spiked into, and recovered from the Monterey extract demonstrated that robust isotopic data can be derived from this approach. Isotopic values of individual sulfur compounds indigenous to the Monterey extract were similar to that of the total fraction. This approach is now being applied to immature Peru sediments where it is anticipated that these isotopic data will provide important constraints on diagenetic reactions involving sulfur.
GRANTEE: WOODS HOLE OCEANOGRAPHIC INSTITUTION
Department of Marine Chemistry and Geochemistry
Woods Hole, Massachusetts 02543
GRANT: DE-FG002-97ER14746
TITLE: Laboratory Constraints on the Stability of Petroleum at Elevated Temperatures: Implications for the Origin of Natural Gas
PERSON IN CHARGE: Jeffrey Seewald (508-289-2966; Fax 508-457-2164; E-mail jseewald@whoi.edu)
Objectives: The objectives are to constrain the stability of petroleum and the origin of natural gas in subsurface environments at elevated temperatures and pressures. In particular, the role of water, redox and sulfur in petroleum degradation reactions are being investigated.
Project Description: Presently, factors that regulate the generation and composition of natural gas during the thermal degradation of petroleum are poorly understood. Laboratory heating experiments are being conducted that will examine the stability of petroleum and N species at elevated temperatures and pressures. The experiments are unique in that they contain water and naturally occurring chemical buffers that fix ƒO2, aH2S, and pH at values consistent with natural systems. Specific topics being addressed are as follows:
1) The stability of oil at elevated temperatures and pressures as a function of redox, activity of aqueous sulfur species, and pH.
2) The relative influence of equilibrium and kinetic processes in regulating the abundance of organic species that constitute oil and natural gas.
3) The contribution of water derived hydrogen and oxygen to the production of methane and carbon dioxide during the degradation of oil.
The experiments conducted as part of this project allow interpretation of the data in terms of both a thermodynamic and kinetic framework. In this way, the experiments will develop and verify fundamental models to account for the generation of natural gas.
Results: This project has only recently been initiated.
GRANTEE: WOODS HOLE OCEANOGRAPHIC INSTITUTION
Department of Marine Chemistry and Geochemistry
Woods Hole, Massachusetts 02543
GRANT: DE-FG02-89ER13466
TITLE: Organic Geochemistry of Outer Continental Margins and Deep Water Sediments
PERSON IN CHARGE: J.K. Whelan (508-289-2819; Fax 508-457-2164; E-mail jwhelan@whoi.edu)
Objectives: The objective of this project is to develop a better understanding of processes of hydrocarbon generation and migration in coastal and offshore sedimentary basins as an aid in predicting favorable exploration areas for oil and gas.
Project Description: Our current research focuses on utilization of organic compounds in elucidating mechanisms, rates, and consequences of subsurface fluid flow processes. Our group at Woods Hole, in collaboration with the Geochemical and Environmental Research Group (GERG) at Texas A&M, has been the principal organic geochemical arm of the Global Basin Research Network (GBRN) since its formation in 1989. The GBRN is a distributed network of scientists working to understand the coupled physical and chemical processes that control fluid movement in sedimentary basins.
Results: Previous work provided evidence for on-going oil and gas injection (termed dynamic migration) into reservoirs of Eugene Island Block 330 (EI330) and areas to the south along the Louisiana Gulf Coast shelf edge and slope. A scenario for the subsurface plumbing system involved depends on gas derived from cracking of residual oil dispersed in sediments at depths of about 8-10 km in this actively subsiding basin. The collective geochemical, geological, and geophysical data are consistent with an on-going upward flow system where methane develops sufficient pressure to open a pervasive fracture network allowing gas to dissolve and entrain residual oil in a supercritical oil/gas phase. The entrained oil is the residue remaining after the initial generation and migration process was complete. The depths of gas production and oil dissolution are considerably below the present day EI330 producing reservoirs. This scenario is consistent with: 1) low concentrations of benzocarbazoles and phenols in EI330 oils suggesting a much higher degree of oil-rock contact for these Gulf Coast oils than is typical for North Sea oils which have migrated much greater distances; 2) C3 through C14 hydrocarbons are distributed throughout fractures in sandstones and in some shales in frozen core samples; 3) collective source and maturation data for EI330 oils and gases and basin modeling; and 4) estimation that the amount of residual oil in a source/reservoir system available for cracking to methane should be more than sufficient to drive this process over geologic time - globally, less than 1% of generated gas and oil worldwide ever make their way into commercial reservoirs with approximately 40% remaining dispersed in the source rock/migration system and about 60% leaking out of surface sediments. Thus, this dynamic migration mechanism may be very widespread and could be occurring in any actively subsiding basin.
The obvious next question is: what is the evidence for these processes affecting other petroleum reservoirs and surface sediments in the world's oceans? What oil and gas amounts and rates might be involved? The surprising answers are: 1) There is overwhelming evidence of massive upward flow of fluids, particularly gas, through the seafloor worldwide over long periods of time. 2) Rates of filling and emptying of petroleum reservoirs are poorly known. 3) Rates of seepage of oil and gas to the seafloor are virtually unknown. 4) There are no large scale surveys of the extent of oil and gas seepage from the ocean floor. Detection of seeps normally occurs only in suspected oil and gas productive areas or accidentally in other areas. Huge areas of the seafloor are unexplored.
During the past year, we were able to present these results very widely in a poster when our new Woods Hole Oceanographic Institution research vessel, the Atlantis, visited New York City, and Washington, DC and hosted open houses for trustees, corporation members, press, and government officials, including members of congress and funding agencies. An article will appear shortly in "Sea Technology" magazine where we have been able to present our case for dynamic hydrocarbon migration and its potential consequences more completely. We hope this article lead to development of new approaches and technologies for measuring the extent and rates of these processes.