Rapid Resurgence of Marine Productivity After the Cretaceous-Paleogene Mass Extinction

See allHide authors and affiliations

Science  02 Oct 2009:
Vol. 326, Issue 5949, pp. 129-132
DOI: 10.1126/science.1176233

Algal Rebound

The extinction at the Cretaceous-Paleogene boundary 65.5 million years ago represented a sudden and dramatic disruption of global ecosystems. Sepúlveda et al. (p. 129) now show, however, that algae recovered rapidly and that photosynthesis and primary production thus also recovered. The authors tracked algal productivity in the thick boundary layer in Denmark through a series of diagnostic biomarkers and isotopes. Algal productivity dropped abruptly during the extinction event but then recovered within the boundary layer, perhaps as quickly as within 50 years of the impact.


The course of the biotic recovery after the impact-related disruption of photosynthesis and mass extinction event at the Cretaceous-Paleogene boundary has been intensely debated. The resurgence of marine primary production in the aftermath remains poorly constrained because of the paucity of fossil records tracing primary producers that lack skeletons. Here we present a high-resolution record of geochemical variation in the remarkably thick Fiskeler (also known as the Fish Clay) boundary layer at Kulstirenden, Denmark. Converging evidence from the stable isotopes of carbon and nitrogen and abundances of algal steranes and bacterial hopanes indicates that algal primary productivity was strongly reduced for only a brief period of possibly less than a century after the impact, followed by a rapid resurgence of carbon fixation and ecological reorganization.

At the Cretaceous-Paleogene boundary (KPB) 65 million years ago, an asteroid impact was associated with mass extinction and widespread disruption of photosynthesis (13), possibly because of a reduction in incoming solar radiation caused by atmospheric debris and sulfate aerosols (3, 4). Scenarios proposed for the immediate aftermath include a “Strangelove ocean,” where primary productivity was suppressed for a substantial interval of time (5), or an alternative “living ocean,” where there was little interruption of productivity and, instead, the flux of organic matter from the surface to the deep ocean ceased in the postextinction ocean (2, 6). Model simulations suggest that the physical effects of reduced sunlight may have lasted no longer than a decade (4). It is assumed that pelagic ecosystems required up to ~3 million years to fully recuperate (2, 6), although continental margin ecosystems seemed to have recovered rather quickly (7, 8), possibly because opportunistic neritic species with benthic cysts or resting stages may have selectively survived (911), particularly at high latitudes (12). However, processes taking place in the immediate aftermath remain poorly understood. Fossils are scarce and interpretations further compromised by poor preservation and reworking in the cosmopolitan boundary clay layer (13, 14), which presumably was deposited in ~10,000 ± 2000 years (15). This interval is hypothesized to represent the duration required for the restoration of food chains and ecosystems after the impact (15). However, many planktonic organisms are devoid of preservable hard parts, including microorganisms at the basis of the marine food chain, such as most microalgae and cyanobacteria.

Here we present high-resolution geochemical and biomarker-based records (16) of a 37-cm-thick section of the Fiskeler clay layer [also known as the Fish Clay (FC)] at Kulstirenden, Stevns Klint, Denmark, an expanded counterpart of the renowned section at Højerup (13, 17) (fig. S1). Hydrocarbon biomarkers obtained from the Fiskeler show evidence of low-to-moderate biodegradation (fig. S3). Using a sensitive mass spectrometry technique, we focused on biomarkers that are resistant to biodegradation, such as C27-29 steranes and C27-35 hopanes [supporting online material (SOM)]. Well-preserved and thermally immature bitumens found within the bottom 16 cm of the FC (gray area in Fig. 1; figs. S4 and S5) contrast strongly with previously fossilized and thermally mature organic matter before the boundary, in the upper 20 cm of the Fiskeler, and in the Cerithium Limestone that had entered the system through weathering of local sedimentary rocks. This stark maturity contrast in a continuously deposited sedimentary sequence is prima facie evidence for mixtures of organic matter of differing ages and thermal histories (18). Bitumen isolated from the lower 16 cm of the Fiskeler contains a higher proportion of autochthonous, thermally immature organic matter comprising the preserved remains of organisms present in the postextinction water column, probably as a result of increased organic matter preservation under oxygen-depleted conditions and an enhanced clay content in the sediments. The preservation of autochthonous organic matter in late Maastrichtian chalk was probably poor in this richly oxygenated and well-mixed, shallow-water depositional environment (13, 14). The proportion of allochthonous biomarkers increased abruptly ~17 cm above the boundary and reaches pre-boundary values at ~27 cm (Fig. 1). This trend is interpreted as the return to a depositional environment where organic matter is vigorously remineralized, and it occurs before the full recuperation of carbonate sedimentation (fig. S2). Therefore, the bottom 16 cm of the Fiskeler provides an exceptional opportunity for studying the ecological recovery after the KPB using biomarkers.

Fig. 1

Stratigraphy and biomarker-based thermal maturity indexes of the Kulstirenden section. (A) C31 S/(S+R) hopane ratio; (B) C30 αβ/(αβ+ββ) hopane ratio; (C) C29 βα/(βα+αβ) hopane ratio (inverted scale); (D) C29 ααα S/(S+R) sterane ratio. The gray area denotes the predominance of stereoisomers with the original biologically inherited stereochemistry, and thus the presence of thermally immature autochthonous organic matter within the lower 16 cm of the Fiskeler suitable for interpretation of biomarkers sources. Shaded areas represent the red clay (dark gray circles) and the black clay (vertical lines).

The carbon isotopic composition of bulk carbonate (δ13Ccarb) from below the boundary is lower than that seen in earlier observations on fine-fraction carbonate (13). At the boundary, δ13Ccarb shifts by about +0.5 per mil (‰) and then remains rather uniform, with a slight and steady shift toward lower values (fig. S2). The stable isotopic compositions of C and N in kerogen, δ13Corg and δ15Norg, respectively, are lowest in the black clay and the lowest part of the Fiskeler (Fig. 2). We suggest that these coupled negative excursions reflect the post-impact shutdown and early stages of recovery of primary productivity. Accordingly, diminished availability and biological assimilation of nitrate resulted in minimal N-isotopic fractionation (19), and the associated minimal algal growth rate led to maximum C-isotopic fractionation (20). As at other boundary sections (7, 21), this early period of the deposition of the clay layer was probably associated with oxygen-depleted conditions, at least in bottom waters. This is supported by the presence of noticeable laminations (fig. S1), a relatively high total organic carbon (TOC) content (Fig. 2), and pyrite with stable S isotope values indicative of intense microbially mediated sulfate reduction (22). Thus, nitrate depletion due to denitrification under oxygen-depleted waters is likely to have occurred. The steady increase of δ13Corg and δ15Norg above the black clay is consistent with a progressive recovery of the ecosystem’s productivity on a centennial-to-millennial time scale if we assume that the sedimentation rate was constant over the entire ~10,000 years of clay deposition (table S2 and SOM).

Fig. 2

Bulk geochemistry and biomarker-based source indexes. (A) TOC content. (B) C isotopic composition of kerogen. (C) N isotopic composition of kerogen. (D) C27-29 steranes/(C27-29 steranes + C30-35 hopanes) ratio, indicative of relative changes among eukaryotic and bacterial sources. (E) C28 relative to C29 steranes, indicating the contribution of chlorophyll c algae relative to chlorophyll a and -b algae (28); (F) 2-MeHI, indicative of the relative contribution of cyanobacteria (29). Black circles in (D), (E), and (F) designate samples with predominantly autochthonous biomarkers suitable for interpretation in an ecological context. Gray circles, in contrast, represent samples with strong allochthonous contributions to the biomarker record (compare with Fig. 1); therefore, the resulting values are not directly comparable to samples indicated by black circles. Error bars represent SD, and gray and shaded areas correspond to those in Fig. 1.

The presence of C27-29 algal steranes, C30 4-methyl steranes and dinosteranes, and bacterial triterpenoids indicates the presence of “normal” marine organic matter (fig. S4) and suggests that photosynthesis, although diminished, persisted. The contribution of terrestrial organic matter appears to be minor (SOM). The low concentration of steranes in the black clay (fig. S6) and a notably low steranes/(steranes+hopanes) ratio [S/(S+H)] (Fig. 2) is probably the result of reduced algal production while reduced sunlight transmission and nitrate-limiting conditions prevailed (SOM). Although a low [S/S+H] could also be related to increased prokaryotic input, such as observed in the Permian-Triassic boundary (PTB) (23), the concentration profiles of both hopane and steranes strongly suggest that this is not the case (fig. S6). An immediate increase of [S/(S+H)] in the following millimeters indicates a rapid resurgence of algal production, as suggested to have occurred once optimal incoming solar radiation levels returned (1, 2). However, δ13Corg and δ15Norg data suggest that initially, primary production remained subdued (Fig. 2). Algal communities may have included photosynthetic microbiota capable of producing resting structures (9, 10) and/or opportunistic organisms capable of adapting quickly to environmental changes by, for example, switching between autotrophic and heterotrophic metabolism [mixotrophy (24)]. Although an early recovery of primary production has been suggested on the basis of microfossil (11, 25), isotopic (26), and geochemical records (7, 8, 25), estimates of its timing vary enormously. In view of the rapid sedimentation of the entire clay layer [~10,000 years (table S2) (8, 15, 27)], and considering several scenarios of sedimentation, the deposition of the 2-mm-thick black clay layer at Kulstirenden could have occurred on a decadal-to-centennial time scale (SOM), which is the period corresponding to maximal disruption of primary production after the deposition event of the fallout lamina.

Low values of the C28/(C28+C29) sterane ratio of less than 0.6 at the base of the Fiskeler (Fig. 2) reveal a diminished contribution of chlorophyll c–containing eukaryotic phytoplankton, such as diatoms, dinoflagellates, and coccolithophorids (C28) relative to green algae (C29) (28). These low values reflect the extinction of groups of phytoplankton and an overall diminished productivity at the boundary because they strongly differ from those in well-studied petroleum samples from the Cenozoic (>0.9) and the second half of the Mesozoic (>0.8), when the geological record shows that the major expansion of chlorophyll c–containing algae took place (28). The presence of cyanobacterial 2-methyl hopanes throughout the section (fig. S5), combined with the low variability of the 2-methyl hopane index (2-MeHI) (29) (Fig. 2), suggest that cyanobacteria were not affected in the same manner as eukaryotic algae and, by inference, indicates that variations in cyanobacterial productivity did not substantially influence the main shift observed in [S/(S+H)]. Values of the 2-MeHI between 2 and 5 at the base of the FC are typical for mid- to high-latitude Phanerozoic sedimentary rocks (28, 30), implying that productivity levels of 2-methylhopane–producing cyanobacteria were normal. However, these values are low compared to those reported for the PTB (23) and oceanic anoxic events (30), when N2-fixing cyanobacteria flourished, implying that it was not enhanced N2 fixation but decreased eukaryotic productivity that was responsible for low δ15N values at the boundary (Fig. 2).

Our study shows that primary algal production in a neritic high-latitude ecosystem rapidly recovered after the impact event at the KPB, evidencing a rapid biological turnover consistent with models suggesting a rapid resurgence of photosynthesis after solar radiation levels were restored (2, 4). Yet the global implications of our findings from a single locality remain to be verified at other boundary-clay layers worldwide.

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 to S6

Tables S1 and S2


  • Present address: Department of Earth, Atmospheric and Planetary Sciences, MIT, 45 Carleton Street E25-623, Cambridge, MA 02139, USA.

  • Present address: Institute of Earth Sciences, Faculty of Chemistry and Earth Sciences, Friedrich-Schiller-Universität Jena, Burgweg 11, 07749 Jena, Germany.

References and Notes

  1. Materials and methods are available as supporting material on Science Online.
  2. This research was funded by the Deutsche Forschungsgemeinschaft through the European Graduate College EUROPROX and MARUM, and NASA (Astrobiology Institute, Exobiology Program). We thank X. Prieto, C. Colonero, L. Sherman, J. Salacup, H. Buschhoff, M. Segl, U. Beckert, P. Simundic, C. Gonzalez, A. Kelly, B. Schmincke, M. Milling-Goldbach, and A. Skorokhod for their laboratory and/or logistic assistance. R.E.S. acknowledges the Hanse-Wissenschaftskolleg and the Alexander von Humboldt-Stiftung. J.S. acknowledges EUROPROX for supporting his doctoral studies and a research visit to MIT.

Stay Connected to Science

Navigate This Article