2009 Annual Science Report

Pennsylvania State University Reporting  |  JUL 2008 – AUG 2009

Biosignatures in Ancient Rocks

Project Summary

The Earth’s Archean and Proterozoic eons offer the best opportunity for investigating a microbial world, such as might be found elsewhere in the cosmos. The ancient record on Earth provides an opportunity to see what geochemical signatures are produced by microbial life and how these signatures are preserved for geological time. Researchers have recognized a variety of mineralogical and geochemical characteristics in ancient rocks (sedimentary and igneous rocks; paleosols) that may be used as indicators of: (i) specific types of organisms that lived in the oceans, lakes and on land; and (ii) their environmental conditions (e.g., climate; atmospheric and oceanic chemistry). Our project addresses the following questions: Are some or all of these characteristics true or false signatures of organisms and/or indicators of specific environmental conditions? Do a “biosignature” in a specific geologic formation represent a local or global phenomenon? How are the biosignatures on Mars and other planets expected to be similar to (or different from) those in ancient terrestrial rocks?

4 Institutions
3 Teams
30 Publications
2 Field Sites
Field Sites

Project Progress

3.1. Field investigations of biosignatures in ancient rocks (Ohmoto, Watanabe (Research Associate), Ian Johnson (graduate student), Hiroshi Hamasaki (graduate student), and Jamie Brainard (undergraduate student)).
Collaborating researchers: Dr. Arthur Hickman (Geological Survey of Western Australia), Dr. Andrew Fitzpatrick (CSIRO, Perth, Australia), Mr. Masamichi Hoashi (Kyushu Univ., Japan), Dr. Kosei Yamaguchi (Toho Univ., Chiba, Japan), Dr. Yasuhiro Kato (Tokyo Univ., Japan), Dr. Clark Johnson (Univ. of Wisconsin, Madison), Dr. Brian Stewart (Univ. of Pittsburgh), and Dr. Paul Knauth (Arizona State Univ.).

* Publication of a discovery of primary hematite formation in an oxygenated sea 3.46 billion years ago

In Yrs 2003 and 2004, the Archean Biosphere Drilling Project (ABDP), under the sponsorship of the NASA Astrobiology Institute, the Japanese Ministry of Science, Education, Sports and Culture, and the Geological Survey of Western Australia, recovered deep cores from 10 drill holes that targeted specific geologic formations (~3.5 Ga to ~2.7 Ga in age) in the Eastern Pilbara Craton, Western Australia. Our group has been investigating the variety of biogeochemical signatures in these rocks since 2003. During this reporting period, we (Hoashi et al., 2009) published a paper in Nature Geoscience, reporting a discovery of primary hematite (Fe2O3) crystals in the 3.46 Ga old Marble Bar Chert/Jasper Formation in ABDP #1 drill hole. Previous researchers have reported the discovery of secondary hematite crystals in younger formations (e.g., ~2.6 Ga) that apparently formed by dehydration of ferrihydrite (Fe(OH)3) during diagnesis/metamorphism at temperatures >60°C. Such findings, however, cannot be linked to the presence or absence of O2 in the oceans, because ferrihydrite could have formed by UV reactions without molecular O2. Our mineralogical investigation of hematite crystals in the 3.46 Ga Marble Bar Jasper beds using electron microscopes and geological and geochemical investigation of their depositional setting have revealed that these hematite crystals most likely nucleated directly as hematite crystals at T >60°C in a deep (>200 m) and large (>30 km radius) sea during the mixing of Fe2+-rich submarine hydrothermal fluids and O2-rich seawater. This discovery is significant because it suggests that oxygenic photoautotrophs (e.g., cyanobacteria) had evolved and deep oceans (and the atmosphere) became oxygenated by ~3.5 Ga ago.

* Investigations on the behaviors of redox-sensitive elements in ~3.5 Ga submarine basalts

The dominant rock type in the Pilbara Craton is submarine basalt. Most of them were altered by contemporaneous seawater (i.e., submarine weathering) and/or by submarine hydrothermal fluids that evolved from seawater through water-rock interactions. Analyses of redox sensitive elements (e.g., U, Cr, V, Mo, Mn, Fe, Ce) in 3.43 Ga altered basalts in ABDP #1 drill hole and in other submarine basalts of ~3.5-3.2 Ga in age indicate that the behavior of the elements in these rocks were essentially the same as those in altered submarine basalts in modern oceans. This finding, to be reported by Ohmoto at the Annual Meeting of American Geophysical Union in December, 2009, is another line of evidence that the atmosphere and global oceans were already oxygenated ~3.5 Ga ago.

* Fieldwork in the Pilbara Craton, Australia

We (Watanabe, Johnson, Hamasaki, Brainard and Ohmoto from PSU and Dr. Fitzpatrick of CSIRO) carried out three-weeks of fieldwork in the Eastern Pilbara Craton, Western Australia during July and August of 2009. It was our fifth year of field investigations that have focused on the origin(s) of an aluminum-rich alteration zone underneath the world’s oldest (~3.43 Ga) unconformity (i.e., the oldest land surface) and of hematite-rich pods within the alteration zone. The primary objectives of this year’s fieldwork were to: (1) determine the spatial and temporal distribution of the alteration zone and hematite pods through geophysical surveys (magnetic susceptibility, resistivity and conductivity) and geological surveys of the alteration zone over a large area (>50×100 km); and (2) collect samples for petrological and geochemical investigations. Some previous researchers suggested that the alteration underneath the oldest unconformity was formed by hydrothermal processes and that the hematite-rich pods by modern weathering of sulfide bodies. However, our investigations have continued to produce various lines of evidence suggesting that the alteration zone and hematite-rich pods, respectively, represent a paleosol and laterites of ~3.43 Ga in age. Our suggestions would imply that the development of an oxygenated atmosphere and of microbial mats on land had occurred before ~3.4 Ga, because the formation of laterites require the abundance of organic acids to transport Fe2+ through soils and an O2-rich atmosphere to fix it as ferric (hydr)oxides. Investigations of the alteration zone and hematite pods in the Pilbara Craton have been carried out by Ian Johnson as a part of his MS thesis (completed in May, 2009), and have been continued by Johnson as a part of his Ph.D. thesis research.

3.2 Records of life and environment in the Proterozoic eon

* Earliest life on land (Kump, Schopf, Timothy White, PSU, and Lev Horodyskyj, Ph.D. student)

PSU Astrobiology graduate student Lev Horodyskyj investigated the earliest evidence (middle Cambrian) for an active role for fungi and bryophytes in soil weathering, using cores from Iowa and South Dakota. The geochemical results are compelling, but the most interesting result came through collaboration with Bill Schopf (UCLA). Below is a preliminary report from Schopf:

Elk Point Middle Cambrian or older, clay-rich paleosol (fossilized soil horizon) from Iowa.

Figure 1

Figure 1. ​Optical image showing Raman-imaged area (rectangle) with overlay of Raman kerogen image.

Figure 2

Figure 2. ​Optical photomicrograph of specimen

Above are figures of the “hockey stick” filament. The available data establish that: (1) the brownish-colored material of the specimen is carbonaceous kerogen (Raman and CLSM); (2) the bits and pieces of kerogen are “chucky” (Optical, Raman, and CLSM); (3) the specimen is “hollow” (CLSM) — i.e., its interior is quartz-filled (Raman); and (4) the encompassing matrix is quartz (Raman).

My best guess is that the specimen is the remnant of some sort of endolithic microorganism — as suggested/indicated by its compact filamentous form and kerogenous composition. The suggestion that it may be a fungal hypha is quite reasonable, but (1) such hyphae typically do not taper; (2) the filament is unlike fungal hyphae in that it is isolated, not in clumps of filaments (as fungal hyphae commonly are); (3) the filament shows no evidence of hypha-defining transverse cell walls or clamp connections to adjacent filaments (really important deficiencies, the presence of either of which would have “sealed the case”); and (4) the filament has a “chunkiness” of its organic components (which I have not previously seen in filamentous microbes, whether fungal, or bacterial, or cyanobacterial, or algal).

Given its composition and compact structure, I think the specimen is definitely biologic. Still, its overall organization and shape seem more microbial (cf. oscillatoriacean or rivulariacean cyanobacteria, filaments of both of which commonly taper) than fungal. Yet that notion, too, suspect, since I cannot discern defined cell segments.

The bottom line: It would be useful to analyze additional specimens to definitively sort this out.

We are attempting to locate additional material for analysis.

* Great Oxidation Event and Proterozoic “Euxinic” Oceans? (Kump, Arthur, Ph.D. student Chris Junium, visiting Ph.D. student Genming Luo, Undergraduate Ramiro Mata (Brown University; NSF REU)

Kump and Arthur and their students have focused their initial foray into Proterozoic environmental evolution on the FAR-DEEP drill cores collected from Fennoscandia in 2008 with funding from NAI, NSF, the International Continental Drilling Program, and the Norwegian and German governments. We have established coarse-resolution C and N isotope stratigraphies for the critical “Shunga” interval of exceedingly high organic carbon deposition, that paradoxically is though to occur after the “Lomagundi-Jatuli Carbon Isotope Anomaly”, a 200 million year large positive anomaly that was thought to herald the organic carbon burial event that caused the Great Oxidation Event (GOE). However, the excursion is too late to explain the GOE. Moreover, based on our preliminary analyses, it seems that the Shunga interval corresponds to (at least the final stages of) the C-isotope excursion. We were unable to extract chlorophyll derivatives for analysis because of the moderate thermal overprints on these strata. This work will mature when stratigraphic constraints are assembled for the cored intervals. So the paradox may be solved. Moreover, we have preliminary evidence that the sediments were deposited in an anoxic, sulfidic ocean basin, perhaps the first hint of the “Canfield Ocean”, the putative sustained euxinic Proterozoic oceanic state.

Kump, Arthur, and Guo will perform field work in China during summer 2010 focused on the deep-water equivalents of the very interesting Neoproterozoic Duoshanto formation, in an effort to establish the redox structure of this basin (i.e., was the deep basin euxinic). Time permitting we will also travel to North China to investigate older, Mesoproterozoic rocks for evidence of deepwater conditions at their time of deposition.

* Early Neoproterozoic Marine Environments: (PhD candidate Chris Junium, Michael Arthur)

The Chuar Group depositional history records an important record of the break-up of the super-continent of Rodinia, the rise of early eukaryotes and geochemical conditions leading up to the first of the Neoproterozoic Snowball Earth episodes.
The mudstones of the Kwagunt formation (~500 m thick) were deposited in an intracratonic, marine rift basin on the western edge of Laurentia. They display a wide range of sedimentary and geochemical characteristics that suggest that organic carbon burial in Kwagunt Formation was mediated by benthic microbial mats. Pseudo-cross laminated structures, carbonaceous lenses, crinkly, silty, anastomosing and discontinuous laminations are found in deeper intervals, and fenestral laminations, roll-up features and pustular surfaces, indicative of mat dessication, are more common in shallower regions. Even-over-odd preferences in n-alkane distributions and possible monomethyl-alkane series support previous findings of quaternary-branched-diethylalkanes and monomethyl-alkane series by other researchers that these geochemical features may be linked to sulfide oxidizing bacteria and benthic microbial mats.
The refined mudstone stratigraphy displays two modes in the relationship between total organic carbon (TOC) and silt content. The bulk of the stratigraphic section (~425 m) is TOC-poor (0-5 %TOC) and displays a positive relationship between silt content and TOC. Two TOC-rich (> 5 %TOC) intervals within the Walcott Member, the uppermost member of the Kwagunt formation, demonstrate an antithetic relationship where silt content is to proportional to 1/%TOC. These two relationships are attributed to variations in basin evolution, sediment distribution and dilution, increased primary productivity and more reducing conditions that enhance carbon burial. Greater TOC associated with increased silt delivery may be the result of increased riverine nutrient delivery or chemocline rise during wetter conditions. The two TOC-rich intervals within the Walcott Member are likely a result of basin deepening and expansion that traps silt nearshore which fosters the deposition TOC-rich black shales ( >5% TOC) by limiting siliclastic dilution.

* Research on Green Lake as Proterozoic analog environments (Kump, Freeman, Macalady, Arthur, Ph.D. student Katja Meyer, Undergraduate students Nathan Barber and Stamatina Hunter; Aubrey L. Zerkle, James Farquhar, University of Maryland)
Three lines of investigation are being pursued: Nitrogen cycling, how well the biomarkers preserved in the sediment represent the overlying microbial community, and the other in-situ experiments with mesocosm enclosures (limnocorrals) to assess the response of the structured microbial community to disturbances. The biomarker work has been completed and submitted for publication to Geobiology (Meyer et al., in review 2009). Two experiments are planned: 1) complete homogenization of the chemocline community, and 2) conversion of the water column (within the enclosure) from a sulfidic (euxinic) state to a ferruginous state, to mimic the proposed Neoproterozoic conversion. This summer we tested a new monitoring approach, itself quite novel, using the self-potential (in-situ voltage gradients) to monitor redox potential. This is a geophysical approach never attempted in a water body, and our initial results are promising. With this method we will be able to monitor the re-establishment of strong redox gradients in our homogenized enclosures without disturbing them by sampling.
We (Fulton et al., in review) also have characterized dissolved, particulate and sedimentary nitrogen for Green Lake in order to understand the relative roles of mixed layer primary production by cyanobacteria and diatoms using fixed nitrogen versus that occurring by nitrogen fixation associated with sulfide oxidation in the shallow chemocline. Nitrogen (and carbon) isotopes of bulk particulates and biomarkers (chlorophyll and its degradation products, bacteriochlorphylls, etc.) indicate a substantial contribution of nitrogen from chemocline-related fixation and variations in this proportion through the last several hundred years based on core samples.

* Mass Spectrometer Upgrade

Arthur is completing an upgrade of a Finnegan MAT 252 mass spectrometer to continuous flow, 5-collector array with online EA and TCEA for carbon, nitrogen and sulfur (multiple) isotope analyses. This will allow rapid analysis of bulk, organic extracts and isolates. The installation and testing is scheduled for November 9, 2009.

3.3. Mass-independent fractionation of sulfur

* Heterogeneous hydrothermal reactions as the source of MIF-S signature (Ohmoto, Watanabe)

The presence of MIF-S signatures in many (but not all) sedimentary rocks older than ~2.4 Ga and the virtual absence of MIF-S in younger rocks have been considered by many investigators as unequivocal evidence for a dramatic change from an anoxic to oxic atmosphere. This is because, until recently, the only known mechanism to cause MIF-S signatures in natural samples has been the UV photolysis of volcanic SO2 in an O2-poor atmosphere. However, we (Watanabe, Farquhar and Ohmoto) reported in a March issue of Science that reactions between amino acid powder and sulfate powder (+ H2O) at 150-200°C generated reduced S compounds (H2S and polysulfides) with significant MIF-S signatures. We have suggested that the MIF-S signatures in our experiments were created during the adsorption of sulfate on the surface of amino acid (i.e., chemisorption isotope effects), and that most (if not all) MIF-S signatures in natural rocks may have been created by chemisorption reactions involving immature organic matter (kerogen), clays, carbonates, iron oxides and sulfate-rich aqueous solutions under the influence of submarine hydrothermal activity during the early diagenetic stage of marine sediments. Our suggestion links the MIF-S record of sedimentary rocks to the evolutionary history of organisms and thermal history of Earth.

* Theoretical advances (Jim Lyons; Prof. Glenn Stark (Wellesley College), Prof. Juliet Pickering (Imperial College, UK) and Douglas Blackie (Imperial College, student)

To determine whether SO2 photolysis in the atmosphere can explain the S-MIF record observed in Archean and Paleoproterozoic rocks. This is done in two ways: 1) theoretical modeling of SO2 photolysis and subsequent isotope fractionation during conversion of photoproducts to ocean sediments, and 2) measurement of high-resolution so2 isotopologue cross section data. These are described as projects 1 and 2 below.

Project 1 was a theoretical evaluation of SO2 photolysis as the source of S-MIF seen in Archean rocks. This work used earlier modeling results on SO2 photolysis, and followed the photolysis products through ocean chemistry to their incorporation into sediments. The goal was to determine the range of mass-dependent isotope fractionation that could be expected, and to demonstrate that recent low-resolution SO2 isotopologue spectra yield incorrect S-MIF results. This work has just been published (Lyons, Chemical Geology, 2009).

Project 2 is the high-resolution measurement of SO2 isotopologue cross sections at Imperial College with the colleagues named above. The goal of the work is to measure isotopic SO2 spectra that can be used for accurate calculation of S-MIF effects during SO2 photolysis. This is necessary for Archean atmospheric models of sulfur isotpe processes. Results thus far show that the high-resolution spectra yield very different S-MIF results versus the the low-resolution spectra of Danielache et al. 2008. This was predicted in project 1 (Chem Geol 2009).

3.4 The genomic history of biosignatures

* Timetrees of life (Hedges, Ph.D. student Fabia Battistuzzi, S.B. Kumar, Arizona State)

The evolutionary context of biomarkers is critical for interpreting whether a molecule or compound is a biomarker at all, and its general usefulness. Our work involves generation of larger and better phylogenies scaled to geologic time (timetrees) to provide that context for interpreting and applying biomarkers for astrobiological questions. In January we published the results of a large study showing, suprisingly, that two-thirds of prokaryotes had a common ancestor that lived on land about 3 billion years ago. We named this group Terrabacteria and it includes cyanobacteria, Deinococcus-Thermus, gram positives, and Chloroflexi. All have adaptations related to terrestrial life. In March we published a book, Timetree of Life (Hedges and Kumar, Eds.; Oxford) with 81 chapters, 10 of which were authored or coauthored by Hedges, and most of those involved new analyses. The book has strong astrobiological context, including chapters on the origin of life and eukaryotes, and on major groups of prokaryotes. The entire book is available free online at http://www.timetree.org/. Our current and future work involves expanding on this same approach by adding thousands of species never analyzed, thus refining the global timetree of life.

* Microbial diversity and divergences (Shapiro, House, Erik Bloomquist (Ohio State), Ms. Sarah Rosengard (undergrad, Brown Univ.))

We developed a new statistical approach to investigate the timing of divergence between the major early lineages of life. Phylogenetic reconstructions of microbial divergence estimate that Archaea arose much earlier than Cyanobacteria, a conclusion contested by the geologic record. We developed a molecular clock- and tree-independent approach that estimates genetic diversity and incorporates alignment uncertainty.

Given the uncertainty in alignment when dealing with such divergent lineages, we adopt a statistical alignment approach, in which thousands of different alignments were created for each data set (datasets included 10-20 representative sequences per taxonomic group for five different genes) using a Bayesian approach. We then generated midpoint-rooted Neighbor-Joining trees for each alignment of alignment, and estimated pairwise distances
across each tree, generating distributions of genetic diversity for each data set.

Cross-gene comparisons revealed that Cyanobacteria and the archaeal group Crenarchaeota exhibited similar diversity levels, suggesting a possible discrepancy in the estimated age of the lineages relative to published studies. Our study affirmed the feasibility of using an alternative approach towards constructing an understanding of evolutionary history. These promising preliminary results will be followed up throughout the next year.