ABSTRACT: Planktonic foraminiferal morphotypes differ in their susceptibilities to post mortem dissolution. Tropical morphotypes are readily assigned to groups that are dissolution-resistant (primarily non-spinose, r) and dissolution-susceptible (primarily spinose walled, s). Morphotypes in the uppermost Quaternary (marine isotope stages MIS 7-2) at Ceara Rise ODP Hole 926A were assigned to these r and s groups, and the stratigraphic distributions of both the morphotypes and the groups examined. Total recovery was co-dominated by s morphotypes (Globigerinoides ruber Morphotype B, Trilobatus sacculifer s.s.) with lesser, but nevertheless abundant, r morphotypes (Menardella menardiisinistral, Truncorotalia excelsadextral). Dissolution intensity, quantified using the dissolution index Res (%) = 100r / (r + s), showed an overall decrease over time, values of Res(%) being negatively correlated with sample depth below the seafloor. It differed more or less markedly between adjacent samples. These fluctuations were greater towards the bottom of the studied section. Between-sample assemblage turnover was quantified using an assemblage turnover index ATIs (SIGMA p i2 - p i1 , in which pi1 and pi2 are the proportional abundances of the ith morphotype in consecutive samples). Values of ATIs for the total assemblage (ATI tot), when assigned to glacial and interglacial MISs, suggested that the mean value of glacial ATItot did not differ significantly from the mean interglacial ATI tot. More variable and greater dissolution in the lower part of the section is reflected in the significant positive correlation between the sample depth and values of ATItot. Assemblage turnover indices were calculated for the dissolution-resistant r and dissolution-susceptible s groups, the latter being calculated in two ways. For ATI sus1, Trilobatus sacculifer was split into several distinguishable morphotypes based on test form. For ATI sus2, these morphotypes were grouped as T. sacculifer s.s. The correlations between sample depth and the ATIres andATIsus2 were not significant. The values of ATIsus2 and ATI res were positively correlated, though mean ATIsus2 was less than mean ATI res. This implies that, despite dissolution, the s group was more stable overall than the r group. Proportional abundances of the s group members T. sacculifer s.s. and G. ruber gr. (including G. ruber Morphotype B) increased through the section, being positively correlated with each other but negatively correlated with depth below the seafloor. In contrast, percentages of the total recovery as M. menardii and T. excelsa were negatively correlated. Menardella menardii was proportionally abundant throughout the interval below mid MIS 5, and T. excelsa was abundant above that depth. Ceara Rise currently lays beneath the North Equatorial Counter Current, which flows seasonally and induces annual phytoplankton blooms. Dextrally coiled T. truncatulinoides, which is also the main coiling direction of our T. excelsa, characterize environments with periodic phytoplankton blooms, while M. menardii prefers stable environments.We speculate that the replacement of M. menardii by T. excelsa might reflect a change in the seasonality of phytodetrital production at ODP Site 926.

Files