For this study, we focused our attention on ODP Sites 974, 975, and 976 (Fig. 1). The cores were sampled by inserting plastic boxes of about 7 cm3 into the split-core sections. At first, we sampled the uppermost 25 m of sediments recovered from each of the three sites. After the preliminary studies, we decided that the existence of a continuous and expanded sedimentary record at Site 976 required additional samplings for Sites 976 and 975. The second sampling focused on obtaining a high resolution in a time interval that included the last full glacial-interglacial cycle (140 ka). In this view, more closely spaced samples were collected in the uppermost 10 m at Site 975 and down to a depth of 45 m at Site 976. In the Tyrrhenian Sea (Site 974), this time interval is very poorly preserved; therefore, we did not collect any further samples. We studied a total of 193 samples: 26 from Site 974, 63 from Site 975, and 104 from Site 976.
To characterize the magnetic mineralogy, detailed magnetic measurements were made to investigate the response of the sediments to a variety of applied magnetic fields. This response is mainly determined by the mineralogy, concentration, and grain-size distribution of the magnetic phases.
The procedure used for the magnetic measurements was as follows:
) at two different frequencies (0.47 and 4.7 KHz) were made by using a MS2 Bartington susceptibility meter. The difference between the two measurements was used to calculate the frequency dependence of susceptibility (fd). This parameter reflects the presence within the sediment of very fine (<0.03 µm for magnetite) ferrimagnetic grains in the superparamagnetic state (SP).Magnetic remanences are expressed in terms of mass by dividing the results by the weight of the samples. Considering that one of the purposes of our study was to investigate the presence of the sapropels in this part of the Mediterranean, at Sites 975 and 976 we measured typical geochemical parameters that characterize these layers, such as total organic carbon (TOC) and sulfur content (S) together with the total nitrogen (N) content. They were measured by using a Carlo Erba CHN analyzer. The values, expressed as percentages, are given in Table 1 and Table 2, along with the paleoclimatic values.
The samples for the microfaunal analyses were dried at 60°C, washed and sieved through 63- and 125-µm sieves to separate two size fractions: one >125 µm and the other between 63 and 125 µm. Both these sizes were used for qualitative analyses of the foraminifer microfauna. Quantitative analyses were conducted on the fraction above 125 µm for the samples collected at Sites 975 and 976, and above 63 µm at Hole 974B. For each sample, more than 300 planktonic specimens, which were separated with a microsplitter, were identified and counted. Some species, such as sinistral and dextral coiling of Globorotalia truncatulinoides and Neogloboquadrina pachyderma, were counted separately. We have also differentiated the morphotypes of Globigerinoides ruber (G. ruber var. rosea and alba), whereas Globigerinoides tenellus and Globigerina rubescens were counted all together.
The paleoclimatic curve was computed as the sum of the percentages of warm-water indicators (positive values) and the cold-water indicators (negative values) following the method proposed by Cita et al. (1977). Based on distribution patterns of living planktonic foraminifers in the Mediterranean sea (Cifelli, 1974; Thunnell, 1978; De Castro Coppa et al., 1980; Pujol and Vergnaud-Grazzini, 1995) we have considered warm-water species to be Globigerinoides ruber, G. gomitulus, G. elongatus, Globigerinoides tenellus, G. sacculifer, O. universa, Hastigerina pelagica, Hastigerina siphonifera, and Globigerina rubescens. Cold-water species are represented by Globigerina bulloides, T. quinqueloba, Globorotalia scitula, Neogloboquadrina pachyderma (d. and s.), and Globigerina glutinata.
