Overall, two groups of clay minerals display opposite behavior. Illite and chlorite mainly derive from the degradation of micas (biotite and muscovite) from igneous, metamorphic, or sedimentary rocks (Chamley, 1989). These two minerals are, therefore, considered as primary minerals, derived from physical erosion or weak chemical weathering. On the contrary, both smectite and kaolinite are formed by the hydrolysis of primary minerals in the soils. These soils are located in the lowermost parts of the Pearl River drainage basin, where soils can develop and produce secondary minerals. In response to the climatic changes described earlier, it is likely that the mineral composition will have changed because of the physical/chemical weathering balance. Since kaolinite is present only in low amounts, we focus our clay-mineral analysis on the smectite/(illite + chlorite) ratio. This ratio is compared with the L* index (Fig. F3), representing the sediment lightness (Blum, 1997; Balsam et al., 1999). This parameter typically increases with increasing calcium carbonate content and therefore could be a proxy for either elevated calcium carbonate production or reduced input of the detrital fraction. According to the preliminary age model (Shipboard Scientific Party, 2000), predominantly lighter intervals coincide with interglacial stages.
On a long-term basis, smectite exhibits a three-step behavior, with an increase between 1200 and 400 ka. During this interval, chlorite and illite decrease relatively, whereas the abundance of kaolinite appears to remain stable. At higher temporal frequency, smectite/(illite + chlorite) display large variations throughout the core, most of them being associated with changes in L*.