LETTERS Increased terrestrial methane cycling at the PalaeoceneEocene thermal maximum Richard D. Pancost 1 , David S. Steart 2 , Luke Handley 1 , Margaret E. Collinson 2 , Jerry J. Hooker 3 , Andrew C. Scott 2 , Nathalie V. Grassineau 2 & Ian J. Glasspool 4 The Palaeocene–Eocene thermal maximum (PETM), a period of intense, global warming about 55 million years ago 1 , has been attributed to a rapid rise in greenhouse gas levels, with dissoci- ation of methane hydrates being the most commonly invoked explanation 2 . It has been suggested previously that high-latitude methane emissions from terrestrial environments could have enhanced the warming effect 3,4 , but direct evidence for an increased methane flux from wetlands is lacking. The Cobham Lignite, a recently characterized expanded lacustrine/mire deposit in England, spans the onset of the PETM 5 and therefore provides an opportunity to examine the biogeochemical response of wet- land-type ecosystems at that time. Here we report the occurrence of hopanoids, biomarkers derived from bacteria, in the mire sedi- ments from Cobham. We measure a decrease in the carbon isotope values of the hopanoids at the onset of the PETM interval, which suggests an increase in the methanotroph population. We propose that this reflects an increase in methane production potentially driven by changes to a warmer 1,6 and wetter climate 7,8 . Our data suggest that the release of methane from the terrestrial biosphere increased and possibly acted as a positive feedback mechanism to global warming. Although the PETM is not a direct analogue for future global warming because it occurs during a time when global temperatures were significantly higher than now 9 , it still allows the investigation of rapid warming comparable to that occurring today. A significant challenge in the investigation of terrestrial methane cycling at the PETM is the lack of appropriate sedimentary sequences. However, we have recently reported the identification of the PETM interval in lake and mire sediments from Cobham, southeast England 5 . The section, represented primarily by lignite through the PETM interval, provides a unique opportunity to study wetland microbial processes in a stratigraphically expanded and relatively complete continental section. The PETM is stratigraphically constrained (Methods) and characterized by a negative carbon isotope excursion (CIE) recorded by plant organic matter 5 (Fig. 1). The pronounced CIE is one of the defining characteristics of the PETM 6,9 and is typically attributed to the catastrophic release of 13 C-depleted methane from marine gas hydrates. Critically, at Cobham, the 2-m sequence between the top of the Upnor Formation and the base of the WSB (Woolwich Shell Beds) is entirely freshwater. The black slickensided clay at the base of the sand and mud unit contains leech cocoons, termite coprolites, pollen and spores, but lacks dinoflagellates. The overlying sand (sand and mud unit) contains continental and freshwater pollen and spores and lacks the Ophiomorpha burrows indicative of saline conditions that are present in the underlying Upnor Formation. The Cobham Lignite Bed (lignites and included thin clay bands) contains an abundance of continental and freshwater aquatic pollen and spores and algal cysts, and lacks dinoflagellates 5,10 ; it is in this bed, with organic carbon contents ranging from 15 to 36 wt%, that the negative CIE has been recorded 5 (Fig. 1). A similar sequence with a freshwater lignite (although a thinner one than at Cobham) followed by brackish shelly clays, has been described from the north French coast at Varengeville; like Cobham, the CIE is recorded in the lignitic beds 11 , testifying to the extensive nature of the low-lying area of the southwest North Sea Basin at the time. Further to the east, in north Belgium, the Doel and Kallo boreholes, which are down dip from Kent in the North Sea Basin, show that the CIE occurs in a mixed freshwater and brackish sequence 12 . Thus, the nearest occurrence of contemporaneous saline conditions is perhaps hundreds of kilometres from the Cobham Lignite studied here. Such a setting provides an ideal, perhaps even unique, opportunity to examine the biogeochemical response of a continental lacustrine/mire system to increased global temperature at the PETM. Examination of microbially mediated biogeochemical processes in ancient sediments can be challenging, but biomarkers (organic com- pounds that can be structurally related to biological precursors) have proved to be useful in studying the ancient microbiology of, for example, the Permo-Triassic boundary 13 . In the Cobham Lignite, the hydrocarbon distribution is dominated by n-alkanes with an odd-over-even carbon number predominance indicative of a higher-plant origin 14 and exceptional abundances of hopanes and hopenes (Fig. 2) derived from bacteria 15 . Triterpenes and sterenes of inferred higher-plant origin are also present but in subordinate abundances. In the polar fractions, which are discussed here only briefly, n-alkanols and sterols, also of inferred higher-plant origin, are predominant. The hopanes are dominated by the C 29 ,C 30 and C 31 homologues (21-norhopane, hopane and homohopane, respectively), with the 17a,21b(H), 17b,21b(H) and 17b,21a(H) isomers all present. The presence of abundant hopanes with the biological bb configuration and the lack of hopanes with 22S stereochemistry, combined with high abundances of hopenes, indicate that the Cobham Lignite is relatively immature with regard to petroleum generation, which is consistent with vitrinite reflectance values (R o 5 0.39 6 0.1 (mean 6 s.d.)) 10 . Traditionally, hopanes and their precursor hopanoids have been attributed to aerobic bacteria, but recent work has revealed an anae- robic origin for hopanoids as well 16 ; consequently, the interpretation of hopanoid distributions and depth profiles is complex. In the Cobham Lignite, further complexity is imposed by marked variations in hopa- noid distributions and stereochemistry, the latter possibly being due to changing environmental conditions during sediment deposition 17 . 1 Organic Geochemistry Unit, Bristol Biogeochemistry Research Centre, School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK. 2 Department of Geology, Royal Holloway University of London, Egham, Surrey TW20 0EX, UK. 3 Palaeontology Department, Natural History Museum, Cromwell Road, London SW7 5BD, UK. 4 Department of Geology, The Field Museum, 1400 S. Lake Shore Drive, Chicago, Illinois 60605, USA. Vol 449 | 20 September 2007 | doi:10.1038/nature06012 332 Nature ©2007 Publishing Group