96 during the Cenozoic. Blue and magenta colours indicate reconstructions at 20 Ma

96 during the Cenozoic. Blue and magenta colours indicate reconstructions at 20 Ma and 50 Ma, respectively. The orange band shows the present-day subtropical highpressure belt. The sizes of the yellow arrows represent the inferred change in dust flux from the Australian Brefeldin A solubility continent associated with plate motion. This map was created by using the Generic Mapping Tools software (https://www.soest.hawaii.edu/gmt/), Version 4.5.8 59, and GPlates software30,31 (http://www.gplates.org), Version 1.2.0.red clay at the site. During the Palaeogene, the Australian continent was located farther south that its present position (Fig. 5f), and its climate was substantially more humid49,50. Therefore, the supply of aeolian dust from the Australian continent to the South Pacific could have been much smaller than the present level8. This, combined with the extremely low surface productivity in the ultra-oligotrophic South Pacific Gyre, should have resulted in an extremely low sedimentation rate in the area during the Palaeogene. Subsequently, the IC1 scores at Site 596 abruptly decreased at 23 Ma, or the Palaeogene/Neogene boundary, and remained low during the Neogene. This feature is attributed to the increased fraction of detrital materials including Stattic web terrigenous aeolian dust and the andesitic component from convergent-margin magmatic arcs since the Paleogene/Neogene boundary48 (Fig. 5e). The increased fraction of detrital and volcanic components enhanced the sedimentation rate, which in turn suppressed the REY enrichment in the bulk sediments. The increased flux of terrigenous materials during the Neogene can be attributed to the increased aridity on the Australian continent since the middle Miocene8,48?1. The northward and northwestward motions of the Indo ustralian and Pacific plates (Figs 3 and 5f), respectively, moved the Australian continent into the subtropical high-pressure belt, or the dry climatic zone, and pushed the site closer to the continental dust source. This may have resulted in the significantly increased supply of terrigenous materials to the site (Fig. 5f). Combined with the previous works showing the important linkage between global environmental changes and the influence of aeolian dust on deep-sea sediments8,48, the behaviour of REY-enrichment IC at this site suggests that Earth system dynamics such as climate change, geochemical cycles, and plate tectonics have intricately affected the sedimentation rate in the central South Pacific Ocean and may have contributed to the formation of high-grade REY-rich mud in the area over geologic time. Kato et al.2 reported a huge geochemical dataset of 2,037 bulk sediment samples collected from 51 drill cores obtained by DSDP/ODP and 27 piston cores obtained by the Ocean Research Institute, University of Tokyo. In the present work, in order to completely characterise the bulk sediment composition, we remeasured the major element content by implementing X-ray fluorescence analysis, as well as loss on ignition, for all the samples of which major elements had been analysed using inductively coupled plasma mass spectrometry, as discussed in the Supplementary Information. By adding 268 samples from the studied sites in ref. 2 and four new sites in the Pacific Ocean (Supplementary Data S1) and by combining new data from 1,663 samples from the DSDP/ODP sites in the Indian Ocean3,4 (Supplementary Data S2), we constructed a comprehensive compositional dataset of 3,968 bulk sediment samples from 82 sites.96 during the Cenozoic. Blue and magenta colours indicate reconstructions at 20 Ma and 50 Ma, respectively. The orange band shows the present-day subtropical highpressure belt. The sizes of the yellow arrows represent the inferred change in dust flux from the Australian continent associated with plate motion. This map was created by using the Generic Mapping Tools software (https://www.soest.hawaii.edu/gmt/), Version 4.5.8 59, and GPlates software30,31 (http://www.gplates.org), Version 1.2.0.red clay at the site. During the Palaeogene, the Australian continent was located farther south that its present position (Fig. 5f), and its climate was substantially more humid49,50. Therefore, the supply of aeolian dust from the Australian continent to the South Pacific could have been much smaller than the present level8. This, combined with the extremely low surface productivity in the ultra-oligotrophic South Pacific Gyre, should have resulted in an extremely low sedimentation rate in the area during the Palaeogene. Subsequently, the IC1 scores at Site 596 abruptly decreased at 23 Ma, or the Palaeogene/Neogene boundary, and remained low during the Neogene. This feature is attributed to the increased fraction of detrital materials including terrigenous aeolian dust and the andesitic component from convergent-margin magmatic arcs since the Paleogene/Neogene boundary48 (Fig. 5e). The increased fraction of detrital and volcanic components enhanced the sedimentation rate, which in turn suppressed the REY enrichment in the bulk sediments. The increased flux of terrigenous materials during the Neogene can be attributed to the increased aridity on the Australian continent since the middle Miocene8,48?1. The northward and northwestward motions of the Indo ustralian and Pacific plates (Figs 3 and 5f), respectively, moved the Australian continent into the subtropical high-pressure belt, or the dry climatic zone, and pushed the site closer to the continental dust source. This may have resulted in the significantly increased supply of terrigenous materials to the site (Fig. 5f). Combined with the previous works showing the important linkage between global environmental changes and the influence of aeolian dust on deep-sea sediments8,48, the behaviour of REY-enrichment IC at this site suggests that Earth system dynamics such as climate change, geochemical cycles, and plate tectonics have intricately affected the sedimentation rate in the central South Pacific Ocean and may have contributed to the formation of high-grade REY-rich mud in the area over geologic time. Kato et al.2 reported a huge geochemical dataset of 2,037 bulk sediment samples collected from 51 drill cores obtained by DSDP/ODP and 27 piston cores obtained by the Ocean Research Institute, University of Tokyo. In the present work, in order to completely characterise the bulk sediment composition, we remeasured the major element content by implementing X-ray fluorescence analysis, as well as loss on ignition, for all the samples of which major elements had been analysed using inductively coupled plasma mass spectrometry, as discussed in the Supplementary Information. By adding 268 samples from the studied sites in ref. 2 and four new sites in the Pacific Ocean (Supplementary Data S1) and by combining new data from 1,663 samples from the DSDP/ODP sites in the Indian Ocean3,4 (Supplementary Data S2), we constructed a comprehensive compositional dataset of 3,968 bulk sediment samples from 82 sites.