Precipitation of low-temperature dolomite from an anaerobic microbial consortium: the role of methanogenic Archaea P. A. KENWARD, R. H. GOLDSTEIN, L. A. GONZA ´ LEZ AND J. A. ROBERTS Department of Geology, University of Kansas, Lawrence, KS, USA ABSTRACT Here we report precipitation of dolomite at low temperature (30 °C) mediated by a mixed anaerobic microbial consortium composed of dissimilatory iron-reducing bacteria (DIRB), fermenters, and methanogens. Initial solu- tion geochemistry is controlled by DIRB, but after 90 days shifts to a system dominated by methanogens. In live experiments conditions are initially saturated with respect to dolomite (W dol = 19.40) and increase by two orders of magnitude (W dol = 2 330.77) only after the onset of methanogenesis, as judged by the increasing [CH 4 ] and the detection of methanogenic micro-organisms. We identify ordered dolomite in live microcosms after 90 days via powder X-ray diffraction, while sterile controls precipitate only calcite. Scanning electron microscopy and transmitted electron microscopy demonstrate that the precipitated dolomite is closely associated with cell walls and putative extra-cellular polysaccharides. Headspace gas measurements and denaturing gradient gel electro- phoresis confirm the presence of both autotrophic and acetoclastic methanogens and exclude the presence of DIRB and sulfate-reducing bacteria after dolomite begins forming. Furthermore, the absence of dolomite in the controls and prior to methanogenesis confirm that methanogenic Archaea are necessary for the low-temperature precipitation of dolomite under the experimental conditions tested. Received 27 February 2009; accepted 03 July 2009 Corresponding author: J. A. Roberts. Tel.: +1 785 864 1960; fax: +1 785 864 5276; e-mail: jenrob@ku.edu INTRODUCTION Despite its abundance in ancient rock modern dolomite is much less common, and at particularly low temperatures (<50 °C), a poorly constrained process. Modern dolomite is usually found in association with marine influenced and other saline environments such as coastal sabkhas (Mu ¨ller et al., 1990) or hemipelagic mud (Middleburg et al., 1990), hypers- aline evaporative lakes such as the Coorong region, Australia (Alderman & Skinner, 1957), and Deep Springs Lake, Cali- fornia (Jones, 1965). The formation of dolomite in non- marine, aqueous systems has also been documented but its occurrence is rare. Freshwater dolomite formation typically occurs when DIC (dissolved inorganic carbon)-rich meteoric water mixes with Mg-rich groundwaters. Magnesium can be sourced from brackish groundwaters (El-Sayed et al., 1991; Humphrey & Radjef, 1991), lake brines (Land & Hoops, 1973; Colson & Cojan, 1996) or as a result of the dissolution of Mg-bearing rocks such as basalt (e.g. Capo et al., 2000; Whipkey et al., 2002). Reproducing the nucleation and precipitation of dolomite at low temperature under laboratory conditions has been a notoriously difficult endeavor using traditional geochemical techniques and is the subject of much debate (McKenzie, 1991; Land, 1998). Limited precipitation of modern dolomite has been largely attributed to kinetic barriers at low temperature (<50 °C). While all minerals are subject to kinetic controls on mineral composition and distribution, dolomite is particularly susceptible. These constraints include saturation state, temperature, kinetics, pH, Mg:Ca ratio, concentration of carbonate, ion complexing, hydration spheres and sulfate (Goldsmith & Graf, 1958; Kitano, 1962; Folk, 1974; Folk & Land, 1975; Katz & Matthews, 1977; Baker & Kastner, 1981; Land, 1985; Hardie, 1987; Gonza ´lez & Lohmann, 1985; Zhong & Mucci, 1989; Slaughter & Hill, 1991; Arvid- son & Mackenzie, 1997; Wright & Wacey, 2004). As such, understanding the factors that control the precipitation of dolomite in both the laboratory and natural settings is an important step in understanding how dolomites formed in the past. Along with the above mentioned geochemical controls the role of live micro-organisms in overcoming the kinetic barriers to dolomite precipitation has been extensively studied (Vasconcelos et al., 1995; Warthmann et al., 2000; Van Lith et al., 2003; Moreira et al., 2004; Roberts et al., 2004; Wright & Wacey, 2004, 2005; Rivadeneyra et al., 2006; Sa ´nchez- Roma ´n et al., 2008). A number of metabolic pathways have been implicated in promoting dolomite precipitation under conditions that may 556 Ó 2009 Blackwell Publishing Ltd Geobiology (2009), 7, 556–565 DOI: 10.1111/j.1472-4669.2009.00210.x