Published in the Geochemical News
Issue #108: 16-19, July 2001
G. Bebout, J. Blichert-Toft, R. Carlson, K. Cooper, M. Evans, K. Kelley, J. Konter, C. Reif, O. Rouxel, A. Shaw, G. Shields, H. Staudigel, and D. Stegman
Introduction
The third Geochemical Earth Reference Model (GERM) workshop was held in La Jolla, California, on March 6-9, 2001. The GERM Initiative is supported by the National Science Foundation as a grass-roots effort working towards a better understanding of Earth as a global, chemical and dynamic system. GERM views the Earth as a planetary body in the solar system and includes all of the Earth's major geochemical reservoirs, from the core to the oceans and the atmosphere. GERM works to bridge the gaps between geochemical disciplines that often focus separately on the study of distinct reservoirs. GERM also emphasizes the relationships between geophysics and geochemistry, in particular for the study of deep Earth reservoirs that are not directly accessible to sampling. GERM organizes workshops, supports publications of state-of-the-art research on the Earth's dynamic system, provides technical help in establishing data bases, and provides technical tools for geochemical research. GERM maintains a website with information on the composition of earth reservoirs, partition coefficients, modeling tools, and a range of information on GERM and its workshop activities at http://earthref.org/GERM/.
GERM 3 was openly advertised for participation by any interested member of the Earth Science community. 69 participants registered, mostly from institutions in North America and Europe, including 10 graduate students. GERM 3 included three types of sessions, keynotes with extended discussion periods, poster presentations, and break-out sessions that were focused on a series of more practical problems in geochemistry. Among the latter, GERM 3 discussed aspects of GERM approaches to geochemistry (arc systems and subcontintenal mantle) and geochemistry infrastructure topics such as editorial guidelines for data publication and the need for establishing a geochemical sample repository accessible to all members of the community (read more about the geochemical sample repository initiative elsewhere in this issue). In this report we will provide brief summaries of keynote talks and break-out sessions. Other information about the meeting can be found on http://earthref.org/events/GERM/2001/.
GERM 3 was opened with welcoming remarks by SIO Director C. Kennel and IGPP Director J. Orcutt. Invited speakers to the opening session included Bill White, editor of Geochernistry, Geophysics and Geosysterns (G-cubed), Al Hofmann, and John Helly. Bill White spoke on the development of G-cubed which is jointly published by the GS and the AGU. The scientific focus of G-cubed is on global geochemistry and geophysics, and it offers a range of conventional and new publication types. Conventional publication types include letters, articles and reviews and the more novel types of publications include data briefs, technical briefs, and characterizations. The latter three publication types have particular relevance for data bases, such as the GERM initiative. Data briefs allow for the publication of all data that are important contributions to science even though they may not necessarily support any major new conclusions. Technical briefs are used to publish innovative chemical or computational techniques, and characterizations may be used to publish chemical or physical properties of large scale features on earth (the mantle, lithosphere etc.), without the necessity of major new interpretations. The copyright policy of G-cubed allows free access to all papers that have a relevance to databases (like GERM). Al Hofmann talked about some problems in geochemical data publications and suggested that much improvement is needed for scholarly data publications. . Particular problems include `bad habits' such as publication of data in diagrams only, or the poor documentation of analytical data or samples analyzed. He further suggested that conventional publication procedures appear unlikely to be able to keep up with the dramatically increasing volume of high-quality geochemical data. Electronic data publication and data bases will be increasingly needed to keep up with the increasing data flow, but it is essential to structure data bases such that they contain a scholarly documentation of data. John Helly expanded on the theme of controlled data publication and provided an Information Technology perspective on machinery and methods available today for the publication of data. These technologies are extremely diverse and adjustable to the needs of a particular application and discipline. The range of option includes system architectures where data are published in centralized, major data bases, a network of a large number of nodes that carry subsets of the available data, or peer-to-peer (NAPSTER) type data bases. Data review methods may include a strict peer review with editorial control to an author-controlled `caveat-emptor' data publication. G-cubed as a peer-reviewed electronic journal was mentioned as an important can step towards fully using information technologies in geochemical data publication. Helly also mentioned EarthRef.org, a data base system being developed at SIO and SDSC that explores a variety of data archiving and publication methods, ranging from the archiving of data supporting GERM to the publication of seamount bathymetry maps and geochemical modeling tools.
Keynotes
Most of the conference focused on a series of 45 minute keynotes followed by a typically quite vigorous 45 minute discussion period. Keynotes covered a wide range of topics from the Earths' core to the hydrosphere, from geochemistry and geophysics.
Claude Allegre on the silicate Earth and the core: Many existing models for determining the composition of bulk earth are overly complicated. Developing a more simplistic model that fits data and observations is therefore a worthwhile endeavor. Bulk earth can be determined based on carbonaceous chondrites where the different elements are distributed based on their different volatilities. It is assumed that bulk earth is a simple mass balance of its component reservoirs: continental crust, depleted mantle, primitive mantle, and core. Assuming further that continental crust and depleted mantle were derived from the primitive mantle, the equation for the bulk earth can be reduced to a sum of the primitive mantle and the core. Since certain refractory elements (Mg, Ca, Al, etc.) do not partition into the Earth's core, it can also be assumed that their bulk earth ratios are equivalent to their ratios in the primitive mantle. Because the primitive mantle is a mixture of continental crust and depleted mantle, the partitioning behavior of different kinds of elements between these two reservoirs allows them to be proportionately mixed back together to compute the composition of the primitive mantle. The core, then, is calculated as a simple difference between the bulk earth and primitive mantle compositions. [ read complete summary ... or download PDF ]
Marc Hirschmann on the thermodynamics of mantle melting: The recent increase in available experimental data on a range of peridotite compositions allows for a more comprehensive evaluation of the role of chemical composition on the temperature of the solidus. The effect of compositional variations on the volatile-free peridotite solidus depends in large part on the competing effects between increased abundance of melt-enhancing components (alkalies, lower Mg#) and the increased compatibility of those components (clinopyroxene mode or CaO%). Based on theoretical, experimental, and field studies it seems plausible to infer that a veined peridotite/pyroxenite assemblage is a good candidate to represent the lithologies of the upper mantle. Studies of the thermodynamics of adiabatic decompression melting of veined mantle and the consequences for MORB petrogenesis and geochemistry were reviewed. Partial melting of pyroxenite during mantle upwelling can be particularly productive in the depth interval where pyroxenite is melting but peridotite is not. This results from a lower solidus for pyroxenite and melt enhancement by conductive heat flow from neighboring non-melting material. The proportion of melt derived from pyroxenite may therefore be greater than the abundance of pyroxenite in a mantle source region. Further investigations in progress (such as melt inclusion studies) should lead to a more comprehensive picture of the scale of mantle heterogeneities and the way they are produced and preserved in the convecting mantle. [ read complete summary ... or download PDF ]
Bernie Wood on element partitioning and Earth's core composition: Determining the composition of the Earth's Core is essential for understanding the internal structure, evolution, and present dynamics of the Earth. Studying how and to what degree elements are partitioned into either metal (siderophile elements) or silicate (lithophile elements) is key for understanding the processes in which the Earth differentiated and what the composition of the core might be. Several interesting observations have been made by examining the abundance of siderophile elements in the silicate portion of the Earth relative to a CI chondrite. First, very siderophile elements with a high temperature of condensation are not depleted enough in the silicate portion. Next, elements that are seen at low pressure to be only weakly siderophile are too depleted. Another interesting observation is that the core's density cannot be explained with a pure iron composition so there must be a sufficient amount (about 10 weight percent) of some light element or combination of light elements in the core that can account for the lower density. Therefore, any scenario of core formation needs to be able to explain both the abundances of siderophile elements in the silicate Earth as well as the presence of a light element in the core. There have been a number of efforts to estimate both the composition of the Earth as well as the composition of the core. One recent effort (Allegre, 2001, EPSL, 185, 49 or keynote above) hypothesizes that the major component of the light element of the core is silicon (roughly 7%). Therefore, the behavior of silicon under various conditions has been studied to test this hypothesis. In general it is observed that high pressure can dramatically change whether an element will partition into metal or silicate and to what degree it will partition. Experiments done at high pressure (25 GPa) and high temperatures (from 2500 to 2800 °C) show that silicon does become siderophile. By extrapolating just beyond the experimental data, the conditions required to dissolve roughly 7% silicon in the core can be estimated and suggest that a magma ocean much hotter than 3000 °C would be needed. In conclusion, it is possible to have silicon be the light element in the core, but due to large differences in the models used for extrapolation, it is difficult to say how plausible. It seems that it is difficult to put as much as 7% silicon in the core, but certainly a few percent is likely. In any case, if silicon were the major component of the light elements, then as a refractory lithophile element, it would have to become more siderophile than all of the other refractory lithophile elements. [ read complete summary ... or download PDF ]
Jerome Gaillardet on global chemical fluxes in rivers: This presentation covered potamochemistry from 1) river discharge into the oceans, with boron as a case study, to 2) silicate weathering and its influence on the carbon cycle. In addition, some implications were discussed for the composition of the continental crust. Based on chemical analyses of both major and minor rivers worldwide, it was outlined how river chemistry can be used to constrain the effects of carbonate versus silicate weathering on, for example, C02 consumption, while emphasizing that refined riverine budgets are an essential first step to understanding ocean chemistry. [ read complete summary ... or download PDF ]
Billy Moore on dynamics and fluxes of fresh water input into the oceans: Coastal aquifers have relatively high nutrient and carbon concentrations and therefore any significant flux of water from these aquifers to the ocean will strongly affect coastal water nutrient, trace metal, and carbon budgets. Groundwater inputs may be identified using chemical tracers, in this case short-lived Ra isotopes. If the system is in balance, the rate at which the tracer is lost by mixing offshore must equal the rate at which it is being added near the coast. If we can measure the terms in this mass balance, we can establish the groundwater flux. Along the South Carolina coast, this groundwater flux has been estimated at 500 m3/s, about half of the river input to the oceans. Based on these results, groundwater flux should be considered when constructing coastal ocean solute and nutrient budgets. [ read complete summary ... or download PDF ]
Tim Elliott on mass balances of subduction: A first order estimate of the global flux of sediment and altered ocean crust into and out of the mantle via subduction zones is essential to understanding large scale Earth processes. Constraining this flux of material requires reliable input (both sediments and altered oceanic crust) and output (fluid, volcanic front and back-arc) data (compositions and volumes). The Mariana arc system has been used as a case study to quantitatively estimate fluxes through the subduction zone for various incompatible elements. The results show that for the islands studied, the subducted continental input flux lost to the arc front varies from 10 to 40%. Furthermore, if the back-arc basin volcanics are included in the mass balance, between 80 and 90% of the highly incompatible elements are returned to the crust through the arc and back-arc related volcanism. This seems to suggest that very little of the highly incompatible elements actually make it down into the deep mantle and thus cannot account for the recycled component in OIBs. [ read complete summary ... or download PDF ]
Dan Schrag on D/H constraints on mantle-ocean hydrogen fluxes: It has previously been noted that there is a trend of increasing Ca and decreasing Mg and 180/160 with depth in pore fluid from deep-sea sediment cores. This pattern has been interpreted as a signature of high-temperature alteration of basalt, where the Ca, Mg, and 180/160 profiles reflect diffusion upward through the pore fluid in the sediment column. More recently, it has also been noted that D/H decreases in sediment pore fluid with depth in virtually all sediment cores, regardless of other variables such as organic content, sedimentation rate, and chemistry of sediment. Several explanations for the data have been proposed. Alteration of oceanic crust and diffusive exchange with pore fluid, adsorption of interlayer water in clays, and the presence of paleoseawater are all unsatisfactory explanations for the decrease in D/H in pore fluid. The data appear to be best explained by a flux of low D/H water from the oceanic crust or upper mantle into the base of the sediment pile. The importance of this interpretation in terms of fluxes between GERM reservoirs is that if the deuterium depletion in pore fluid is due to flux of mantle water from the crust, then this process could represent an important exchange between ocean water and mantle water. [ read complete summary ... or download PDF ]
Guy Masters on what, if anything, tomography says about geochemistry: To interpret what mantle tomography can tell us about earth properties, we must first understand the strengths and weaknesses of the data used in the inversions. Then we can evaluate what aspects of the models are robust. Of recent interest is the possibility of a dense layer in the lower mantle and the current status of possible compositional jumps in the upper mantle and transition zone. The 1D and 3D inversions use a combination of body waves, surface waves, and normal modes to constrain the seismic velocity models. Using resolution analysis, it is possible to choose a combination of data that is sensitive to one parameter at one depth. The resulting density models are well constrained, and raising the density by just 1% in a few hundred kilometer thick layer near the base of the mantle would wreak havoc on the mode data. Thus, it is unlikely that a < stealth» layer exists, though the complicated trade-offs involved in modeling free oscillation frequencies means that we cannot rigorously preclude the existence of such a model. Mineral physics has now provided reliable equations of state for most of the possible constituents of the mantle. Recent work indicates that apyrolitic composition can fit the seismic data without any change in composition at the 660 km or 410 km discontinuity though, again, trade-offs with temperature preclude us from saying the mantle is isochemical with absolute certainty. Tomography results show that Vc (bulk sound speed) and Vs (shear wave velocity) are anti-correlated in the lower mantle. Most mantle anomalies can be explained thermally, but this anti-correlation cannot be explained by thermal effects alone. Therefore, we are searching for a combination of partial melt and chemistry changes in parts of the lower mantle to explain the anticorrelation. A new high-pressure polymorph of perovskite which may appear in regions of elevated temperature in the lower mantle could be an explanation. [ read complete summary ... or download PDF ]
Break-Out Sessions
Two break-out sessions focused on scientific issues, namely arcs and the subcontinental mantle, and two sessions on technical issues including the publication of geochemical data and geochemical data base activities.
Subduction-Zone It was agreed upon that the existing conceptual (straw man) model for fluxing of seafloor materials into the mantle is sufficiently accepted by the community to al low for a first-order attempt at convergent-margin chemical mass balancing. Although each has its own embedded complexities (which we should explore), the major flux categories presently used on the GERM Subduction Filter pages remain appropriate. Those include in particular fluxes in forearcs, the subarc slab, mantle wedge, backarcs and the deeper mantle. All present at the break-out session agreed that the theme of the subduction flux section of GERM should be the quantitative mass-balance of materials across individual convergent margins, in a representative set of relatively well known chemical and physical endmembers (e.g., thermal structure, nature and thickness of incoming sediment section, accretionary vs. non-accretionary, island-arc vs. continental arc). Because of the obvious overlap in our goals with those of the MARGINS Subduction Factory initiative, we will include in our subset of systems (8 or 9 systems), and initially concentrate on, the Izu-Bonin-Marianas (IBM) and Central American (Centam) convergent margins. We will attempt to coordinate our efforts with MARGINS to the greatest extent possible, beginning with the establishment of links between our web pages. One extremely significant development greatly enhancing our pursuits is the recent production of chemical databases for arc volcanic rocks (including Carr's CENTAM database and arc volcanics in GEOROC/Mainz). In addition to the many minor but necessary web page tweaks, we will on our web pages begin our system-specific efforts by first concentrating on the IBM and Centam arc-trench systems. This will initially involve the construction of individual web pages for each of these systems, with lists of web links selected to allow one to quickly/efficiently access background physical and chemical information for that system. Our goal here is not to simply produce attractive web pages, but to provide clean and mean ability to synthesize the broad diversity of observations regarding the geometry, mechanical state, thermal structure, and geochemistry of each setting. We will solicit papers on issues key to our understanding of chemical fluxing at convergent margins, beginning in a Special Session at the Upcoming AGU Meeting in San Francisco and potentially leading to themes published in G-cubed. Another category of papers we will seek will be those directly attempting to mass-balance material fluxes across individual convergent margins. We hope to exploit the interfacing possible between our web pages and G-cubed-published data and models, and we will attempt to fully exploit the extremely broad database capabilities of Earthref.org in our interfacing of chemical data with other diverse observations.
Subcontinental Lithospheric Mantle (SCL) Discussions centered on 3 questions: 1) Since most SCL rocks have suffered incompatible element enrichment through metasomatism, is the SCL a significant reservoir of incompatible elements?, 2) What role does the SCL play in continental magmatism? Source? Impediment? Modifier of sublithospheric melts?, and 3) What factors contribute most to the long-term stability of the SCL and what can cause its destruction? Attention also was focused on the presence of pyroxene-dominated, as opposed to peridotitic lithologies in the SCL and how such materials could influence the melting behavior of the SCL and the composition of SCL melts. The GERM data base already contains a significant number of analyses of SCL xenoliths, but a large number of new analyses have become available in the last few years. These new analyses often include both abundances of the platinum group elements and Re-Os isotope systematics that can provide information on the initial melt depletion events experience by the SCL as opposed to later metasomatic events that dominate the lithophile element budget of these materials. Improvements to the GERM data base on the SCL were proposed to take the form of a review of current understanding of the geochemistry, mineralogy and melting behavior of the various lithologies found in the SCL to be submitted to G-cubed, accompanied by updating of the GERM data base on SCL xenolith analyses. An additional contribution would take the form of making available on the GERM web page programs to calculate pressure and temperature from mineral chemistry of SCL xenoliths; a data base of elastic parameters for minerals that would allow calculation of density and seismic velocity for mantle rocks; and programs to calculate geotherms given user input distributions of heat producing elements.
Geochemical Data Publication The discussions covered several concerns in geochemical data publication activity, including the attribution of data to original references, editorial guide lines, methods of data archiving and the longevity of data archives, access to electronic data supplements, and the copyright policy issues. The increasing tendency to use geochemical data from data bases marginalizes the original producer of data. It was emphasized that it is essential to any scholarly data base that every query in a data base should have the option of recovering all contributing data references. However, it was considered impractical and counter-productive to cite all of the original references in a paper that used a data base query. Editorial guidelines in geochemical journals do not contain any specific guidelines for what comprises a scholarly documentation of geochemical data. Too many papers are published with poor analytical documentation and insufficient sample descriptions. Too many data are published only in figures or element ratios and remain effectively inaccessible to the research community. A plan was made to begin a dialogue on what comprises scholarly publishing in geochemistry and to hold a conference for editors in geochemistry to establish a new and consistent set of editorial guidelines in Geochemistry. Most participants supported publication of data in electronic supplements and the publishers present agreed to free access to these data supplements. In fact the publishers agreed to work with the community to give most free and unfettered access to all data published, including the use of scanned original data tables. Overall, this session ended with a very positive outlook on the future of geochemical data publication by society and commercial publishers.
Databases The breakout session on databases began with presentations on the major recent developments for databases in geochemistry, including in particular the PetDB Ridge petrological data base at Lamont, GEOROC at Mainz/ Germany and Earthref.org at Scripps. Even though there are important differences between these data bases, all of them were viewed as very positive and important contributions to geochemistry, even though such data bases appear to be a constant `work in progress'. Several issues were named as much needed developments, in particular the transfer of databases into more robust database applications (Oracle), versioning of the databases, as well as open access to all data and the interoperability of data bases. In particular the desire was expressed to develop a common geochemical metadata standard that will allow open exchange of data between data bases and between databases and users.
Conclusion
GERM 3, like previous GERM meetings, demonstrated the need to blend scientific vision in geochemistry with its publication and database infrastructure needs. GERM 3 will publish a theme volume in G-cubed including scientific contributions, data base contributions, reviews, and technical papers. This volume is likely to be completed by summer 2002. GERM also set the date and venue for the Next GERM Meeting to be held in Lyon, 20-23 May, 2003, thus returning the meeting to where the initiative originally started with GERM 1 in 1996.