A comparison of multiple proxies for export production in the equatorial Pacific

Abstract

[1] Several commonly used paleoproductivity proxies were compared to evaluate the validity of assumptions and limitations associated with each proxy. Export production fluxes (C export ) over glacial-interglacial timescales were calculated from previously developed, proxy-specific algorithms at TTN013-pc72 and TTN013-pc114, both located in the equatorial Pacific. Comparison of data from the same core intervals yields conflicting results despite calibrations based on the same core top samples. The periodicity of marine barite, excess Ba, and excess Al-based C export records is similar. The relative magnitude of C export when calculated using excess Ba and algorithms based on sediment trap data is significantly different than records based on core top calibrations, particularly during glacial intervals. At both sites, bulk sedimentary Al/Ti and Ba/Ti ratios covary; however, these ratios do not correspond with the downcore records of their respective excess concentrations (and their accumulation rates), or with contemporaneous records based on marine barite, excess Ba, and excess Al accumulation. Although downcore records based on sediment mass accumulation rates may be compromised by sediment focusing, this process cannot explain all the differences observed among the various data sets presented here. This implies that some or all of these proxies do not exclusively respond to changes in export production. The contradictions among these data highlight the importance of addressing inconsistencies among paleoproxies and re-examining assumptions imbedded in proxy fundamentals, prior to applying paleoproductivity proxies and interpreting paleoceanographic records.

Precipitation of barite by marine bacteria: A possible mechanism for marine barite formation

“…For example, the calculated equilibrium Sr content in barite is ~13 mol%, while observations indicate that the actual Sr content in marine barite is typically <0.5 mol% (Averyt and Paytan, 2004). Moreover, if bacterial activity that forms the phosphorous compounds that mediate Ba binding is sensitive to temperature, as expected for bacterial metabolism (Hall et al, 2008), the temperature dependence of distribution coeffi cients may deviate from that expected from theoretical considerations.…”

Geochemical characteristics in the Longmaxi Formation (Early Silurian) of South China: Implications for organic matter accumulation

“…Barium, however, is contained in other phases, some of which are also biogenically related (e.g., organic matter, biogenic silica and biogenic carbonate), and others that are not directly related to carbon export (e.g., terrigenous silicates, FeeMn oxides and hydroxides) (Eagle et al, 2003). The Ba bio content, which is estimated by correcting for Ba of detrital origin, is proven to be a good proxy to estimate carbon export (e.g., Eagle et al, 2003;Averyt and Paytan, 2004). Ba bio is determined from the total Ba concentration in the sediment minus the Ba associated with terrigenous material, which is calculated from total Al and normalization to a constant Ba/Al ratio: Ba bio ¼ Ba tot À ðAl tot  Ba=Al alu Þ ¼ Ba tot À ðAl tot  0:0075Þ…”

Problems with paleoproductivity proxies

“…Export fluxes derived by applying the algorithm of Dymond et al [1992] to down‐core records have a much greater peak‐to‐trough amplitude than do the records derived using other algorithms [ Averyt and Paytan , 2004, Figures 2 and 3]. For example, over the past 500 kyr in core TT013‐72PC the peak‐to‐trough range of export fluxes derived by applying the Dymond algorithm to excess Ba accumulation rates has an average amplitude of nearly a factor of 10, whereas the corresponding range for export production derived using barite accumulation rates has an average amplitude of less than a factor of 3 [ Averyt and Paytan , 2004, Table 1]. Export fluxes derived using the two algorithms are similar during interglacial periods (minimum export), whereas the Dymond algorithm produces much larger export fluxes during glacial periods (maximum export).…”

“…The absolute magnitude of export flux, the peak‐to‐trough amplitude of relative change in export flux, and the timing of maximum and minimum fluxes with respect to glacial cycles differed between the proxies. As emphasized by Averyt and Paytan [2004], paleoceanographers must be cautious in making quantitative interpretations of paleoproductivity records. Errors can be introduced if assumptions inherent in a method are violated, as well as through problems in the calibration of a method and in the derivation of the record itself.…”

“… Averyt and Paytan [2004] identified a number of factors that could have biased individual proxy records. Here we build upon their work by presenting evidence for two specific factors that must have contributed to the inconsistencies among their export production records.…”