Technical Comments

Response to Comment on “Calcareous Nannoplankton Response to Surface-Water Acidification Around Oceanic Anoxic Event 1a”

See allHide authors and affiliations

Science  08 Apr 2011:
Vol. 332, Issue 6026, pp. 175
DOI: 10.1126/science.1199608


Gibbs et al. question our reconstruction of surface- and deepwater acidification around Oceanic Anoxic Event 1a. We answer their criticisms to better substantiate our arguments and original conclusions. Contrary to their suggestion, preservation cannot explain the nannofossil changes we documented, which trace perturbations in the photic zone, including a substantial increase in partial pressure of CO2 (pCO2) and an inferred decreased pH as derived from geochemical proxies.

We documented micropaleontological, sedimentological, and geochemical data suggestive of ocean surface- and deep-water acidification during the Early Cretaceous Oceanic Anoxic Event 1a (OAE1a) (1). Gibbs et al. (2) challenge our conclusions, and we welcome the opportunity to provide further details supporting our reconstructions and to respond to their concerns.

Gibbs et al. (2) recognize the nannoplankton response to environmental changes during OAE1a but raise three major concerns. First, they doubt that nannofossil paleofluxes trace carbonate productivity, due to modifications by dissolution during transport and burial, sediment dilution, and diagenesis. We excluded dilution by terrigenous input because of persisting low sedimentation rates and confirm that homogeneous moderate preservation suggests a comparable dissolution-induced decrease of carbonates. In disputing our use of paleofluxes for estimating paleo-CO2, Gibbs et al. mistakenly refer to our assumption that calcification changes within a modern species or strain is an identical process to assemblage-wide carbonate production. Actually, we inferred our paleo-CO2 estimates (3) from experimental data for total nannoplankton calcification (4) and isotope data (5).

Our nannofossil assemblages do have “less than half the diversity of global estimates of this age,” but we emphasize that dissolution is not “a major control” and that our assemblages are oceanic, normally characterized by low species richness. Biodiversity is not a measure per se of nannoplankton calcite production (3), and their fluxes must reflect the minimum production and accumulation of nannoplankton carbonate.

We concur with Gibbs et al. (2) that various ecologic factors control abundance, composition, and morphology of calcareous nannoplankton. As far as temperature is concerned, warm peaks were reconstructed in the early phase of OAE1a (6, 7), but nannofossil changes are not restricted to these intervals. Undeniably, smaller-sized coccoliths might result from higher fertility (1) that is fundamental for taxa showing dwarfism during the OAE1a–carbon isotopic event (CIE), because they are Cretaceous mesotrophic forms and OAE1a is associated with increased primary productivity (8). However, in the studied sections, dwarfing is not systematically associated with increased abundance of mesotrophic taxa but rather is probably related to sinking and accumulation within fecal pellets that are particularly abundant under high primary productivity, providing means of rapid transport and dissolution protection. High primary productivity is recorded during OAE1a also above the CIE and in several mid-Cretaceous intervals: They contain common mesotrophic taxa, but coccolith dwarfism is not recorded.

Gibbs et al. (2) underscore that “a change in the degree of calcification of the same-sized coccoliths would be a better indicator of a carbonate chemistry response.” Indeed, thinning of several Watznaueria barnesiae coccoliths is considered a possible response to surface-water acidification (1). The nannoconid decline and crisis have not been attributed solely to surface-water acidification (3, 9), yet temperature and fertility changes do not satisfactorily explain their failure. The most substantial reduction in nannoconid abundance did not occur 1 million years before the OAE1a, as erroneously stated by Gibbs et al., but corresponds to the crisis preceding the onset of OAE1a by ~30,000.

The CO2 pulses were primarily based on geochemical data, including C-isotope composition of biomarkers, with a total estimated release of 9600 Gt of mantle carbon (5). Such independent proxies were used to infer intervals of presumed ocean acidification. Gibbs et al. argue that there is “no convincing evidence that surface water carbonate chemistry changes had a major impact on the calcifying plankton” during the Paleocene Eocene Thermal Maximum. However, they do not take into account the decrease in nannoplankton productivity and carbonate at the CIE onset (10) and the occurrence of “excursion taxa” (11) interpreted as calcification failure and malformation under excess CO2-induced acidification similar to results of laboratory experiments (4, 12, 13).

The second major concern of Gibbs et al. pertains to the reliability of our data collected in diagenetically modified rocks. Preservation is not detailed in (5) because the evaluation of dissolution on nannofossil assemblages was previously published (3, 14). The nannoconid decline and crisis, the abundance increase and size change of Biscutum constans, Zeugrhabdotus erectus, and Discorhabdus rotatorius (B-Z-D), and the malformation of W. barnesiae specimens cannot be ascribed to dissolution or differential diagenesis for the following reasons: (i) changes in abundance, size, and structure are similar and synchronous at sites in different oceans; (ii) changes occur within the same lithology, yielding identical preservation; (iii) nannoconids are most resistant to dissolution, whereas B-Z-D coccoliths are prone to dissolution (their abundance patterns are exactly opposite to what dissolution would produce); (iv) dwarfing was observed in complete specimens with no corrosion in the outline; and (v) malformed W. barnesiae specimens showing strong ellipticity, asymmetry, and rim thinning co-occur with standard-sized specimens presenting normal ellipticity and symmetry.

Gibbs et al. (2) also suggest a diagenetic control for the short-term changes in our O isotopic data. As for nannofossils, the δ18O fluctuations are similar and coeval at very distant sites, consistent with other δ18O records (6, 7) and biomarker data (15). Diagenesis had an impact on the oxygen isotope composition, but we exclude that observed high-amplitude changes are caused by diagenesis.

The third criticism by Gibbs et al. (2) centers on modeling of ocean acidification to dispute our inferred effects of lowered pH on calcareous nannoplankton and the diachroneity between surface- and deepwater acidification. Although sub-Milankovitch resolution of OAE1a is still problematic, magnitude and rate of CO2 pulses have been estimated (5). Injections of 3200 Gt of mantle C during individual pulses make ocean acidification and lowered pH plausible. Limestone pseudonodularity implies accelerated dissolution at the sediment/water interface, under very low sedimentation rate and intense bioturbation (16). The pseudonodular limestone, occurring ~25,000 to 30,000 after the nannoconid crisis and just before the first OAE1a black shale, correlates with an interval of reduced sedimentation rate (17) and is taken as evidence of bottom-ocean acidification when the lysocline shoaled to Cismon paleo-water depth (>1000 m).

Our geochemical and micropaleontological records collected from long-distance settings indicate global synchronous changes in ocean carbonate chemistry, fertility, and temperature. We reiterate (1) that diagenesis was not of relevance and that surface- and deepwater acidification probably was concurrent with the ecosystem perturbation. Estimates of rate and magnitude of CO2 pulses are available to attempt quantification and modeling of the potential biotic response to inferred ocean acidification around OAE1a.


View Abstract

Navigate This Article