Fish Teeth Sr Isotope Stratigraphy and Nd Isotope Variations: New Insights on REY Enrichments in Deep-Sea Sediments in the Paciﬁc

: Rare earth elements and yttrium (REY) are widely recognized as strategic materials for advanced technological applications. Deep-sea sediments from the eastern South Paciﬁc and central North Paciﬁc were ﬁrst reported as potential resources containing signiﬁcant amounts of REY that are comparable to, or greater than, those of land-based deposits. Despite nearly a decade of research, quantitative abundances and spatial distributions of these deposits remain insufﬁcient. Age controls are generally absent due to the lack of biostratigraphic constraints. Thus, the factors controlling the formation of REY-rich sediments are still controversial. In this study, the REY contents of surface sediments (<2 m depth) in 14 piston cores from the Middle and Western Paciﬁc were investigated. The results show that deep-sea sediments with high REY contents (>1000 µ g/g) were mainly concentrated around seamounts (e.g., the Marshall Islands). The REY contents of surface sediments generally decreased with distance from the seamounts. Biostratigraphic and ﬁsh teeth debris (apatite) Sr isotopic stratigraphy of one piston cores (P10) from the Middle Paciﬁc indicate that deep-sea sediments with high REY contents were aged from early Oligocene to early Miocene. Since the opening of the Drake Passage during the early Oligocene, the northward-ﬂowing Antarctic Bottom Water (AABW) would have led to an upwelling of nutrients around seamounts with topographic barriers, and at the same time, AABW would delay the rate of sediment burial to try for enough time for REY entering and enriching in the apatite (ﬁsh teeth debris). Understanding the spatial distribution of fertile regions for REY-rich sediments provides guidance for searching for other REY resources in the Paciﬁc and in other oceans.


Introduction
Rare earth elements and yttrium (REY) are widely recognized as strategic materials for advanced technological applications [1,2]. Global demands for REY are increasing rapidly [3]. At present, deposits associated with intrusive carbonate complexes and ionadsorbed deposits are the world's most important source of REY [4,5]. Considering the supply risk, many countries have begun to explore REY resources beyond traditional terrestrial mines. Deep-sea sediments from the eastern South Pacific and central North Pacific were first reported as potential resources containing significant amounts of REY, comparable to, or greater than, those of land-based deposits [6]. Subsequently, REY-rich deep-sea sediments have also been found in the Indian Ocean [7]. The mineralogical, as well as major and trace elemental compositions, of deep-sea sediments have been investigated in the Pacific and India Ocean, with the goal of analyzing the host phase of REY elements and their provenance. It has been suggested that hydrothermal activity at mid-ocean ridges

Study Area and Samples
The study area is located in the northwest and northeast Marshall Islands (Figure 1), which are potential areas for REY-rich sediments [6]. The study area included the Pigafetta Basin, Eastern Mariana Basin, and parts of the Middle Pacific Basin. The sediments were mainly pelagic clays, with minor radiolarian and calcareous oozes [25] at water depths of 4824-6251 m (Figure 2).
Sample information is provided in Table 1. Samples were taken for REY analysis at a sampling interval of 15 cm continuously from the top to the bottom of the cores. The average REY contents of the surface sediments (<2 m depth) of 14 piston cores were calculated ( Figure 2). In addition, a 7.2 m core (P10) from the Middle Pacific was selected for further in situ Sr and Nd isotopic analyses of apatite (fish teeth detritus) to constrain the age and the source of the high REY content sediment. The upper (0-210 cm) and lower (600-720 cm) sections of P10 mainly contained pelagic clay, whereas the middle layer (210-600 cm) was composed of zeolite clay with a higher REY content (REY > 1000 µg) than the other layers. Two magnetic shields are adopted to summarize the magnetic field into the mid space between them for improved transfer efficiency [6]. Based on the typical circle a square type coils, many new coil shapes have been proposed to achieve better trans efficiency, such as the double D-type (DD) [7], the double D-type quadrature (DDQ) the bipolar pad (BP) [9], the tripolar pad (TP) [10], the quadruple pad (QP) [11], taichi-type [12] and some tridimensional-type coils [13,14]. Cost advantage and a low leaked magnetic field were achieved with these new-type coils, as well as better stabi and an anti-offset of the charging process.
The shape of the shield needs to be adjusted according to the coil shape, such as cylindrical or pad shield for the circle coil [15], the quadrangular layer for the square c [16,17] and the strip-type layers for DD and DDQ coils [18,19]. These shapes are co monly used due to their low costs and simple structures. Excepting these conventio structures of shields, some novel structures are presented to improve efficiency. aluminum plate, ring and ferrite bars are combined as a shield in [20,21] and opera with a circle coil and a DD coil, respectively. Aluminum and ferrite are the commo used materials for shields, and double-layer and multi-layer shields with multiple ma rials are used to improve shielding efficiency of magnet fields with different frequen characteristics [22]. Some new-type and composite materials with larger relative perm ability also can be applied, such as the nanocrystalline material in [23] and the hyb high-temperature superconductor/ferromagnetic material in [24].
If a soft magnetic material is adopted to make the secondary side shield, the mag field strength at the edge of the shield will be increased according to the results of nite-element analysis and experiments. Since the chassis is always made from a ha magnetic material such as steel or aluminum alloy, the shield can be seen as a dou layer. This phenomenon becomes more serious if the adopted material has larger relat permeability and the shield has double or multiple layers; furthermore, the transfer e ciency is affected due to the increased strength. This phenomenon has been named fringe effect of shields, and it is necessary to improve shield structure to solve this pr lem.
For the secondary shield of the soft magnet material, the relationships between increment of the magnet field strength by the fringe effect and some parameters inclu  [25], AABW path after [26] (600-720 cm) sections of P10 mainly contained pelagic clay, whereas the middle layer (210-600 cm) was composed of zeolite clay with a higher REY content (REY > 1000 μg) than the other layers.  [25], AABW path after [26]).

Major and Trace Elements
Analyses of major and trace elements were performed at the MNR Key Laboratory of Marine Mineral Resources, Guangzhou Marine Geological Survey. Major elements were analyzed using X-ray fluorescence (XRF; Axios), with a detection limit of 0.01-0.1% and a relative standard deviation (RSD) of <2%.
Trace elements were analyzed using inductively coupled plasma mass spectrometry (ICP-MS). Each sediment sample (0.1 g) was first placed in a beaker, to which 4 mL of 1:1 HCl, 10 mL of HF, and 1.5 mL of HClO 3 were added. The mixture was heated until it formed a dry paste. Then, 4 mL of 1:1 HCl was added to make a 25 mL solution; 1 mL of which was pipetted and diluted with 2% HNO 3 to a certain volume for analysis. Marine standard sediments GBW07313, GBW07315, and GBW07316 were used for quality control. The instrumental detection limit was 0.01-0.1 µg/mL and the RSD was <2%.

In Situ Sr and Nd Isotopes
The fish teeth debris were selected after sieving the wet sediment samples through a 63 µm mesh and then cleaned in an ultrasonic bath to remove any adhering clay particles and organic materials. After drying, the clean fish teeth were selected using a microscope and pasted on the annular target for in situ Sr and Nd isotopic analyses. In order to eliminate the influences of clay or other substances adhering to or infilling the fish teeth as much as possible, only triangular fish teeth or clear fish debris were selected for analysis ( Figure 3). In addition, only fish teeth with high Sr and Nd contents (Sr > 400 µg/g; Nd >1 000 µg/g) were analyzed. Multiple fish teeth were analyzed in order to characterize the reproducibility in the same interval.
In situ Sr isotopic analyses were performed using an NWR 193 laser ablation system attached to a Thermo Fisher Scientific Neptune Plus multi-collector (MC) ICP-MS at the Beijing Createch Testing Technology Co., Ltd. A spot size of 38 µm was used, with a 6-8 Hz repetition rate and an energy density of 10 J/cm 2 , depending on the Sr concentration of the sample. Sr isotopic data were acquired using static multi-collection in low-resolution mode with nine Faraday collectors. Prior to laser analysis, the Neptune MC-ICP-MS was tuned using a standard to obtain the maximum sensitivity. A typical data acquisition cycle consisted of a 40 s measurement of the Ar gas blank with the laser switched off, followed by 60 s of measurement with laser ablation. The Durango1 apatite reference material was analyzed every 10 samples for external calibration.
Data reduction was conducted offline, and potential isobaric interferences were accounted for in the following order: Kr, Yb 2+ , Er 2+ , and Rb. Initially, the interferences of 84 Kr and 86 Kr on 84 Sr and 86 Sr, respectively, were removed using the 40 s Kr gas baseline measurement. The presence of 167 Er 2+ , 171 Yb 2+ , and 173 Yb 2+ at masses of 83.5, 85.5, and 86.5, respectively, were then monitored. Subsequently, the natural ratio of 85 Rb/ 87 Rb (2.5926) was used to correct for the isobaric interference of 87 Rb on 87 Sr using the exponential law, assuming that Rb has the same mass discrimination behavior as Sr. In addition, interferences from Ca argides or dimers and Ca-P-O were not considered for high Sr contents (>400 µg/g). The 87 Sr/ 86 Sr ratios were then calculated and normalized from the interference-corrected 86 Sr/ 88 Sr ratios using the exponential law. The data reduction procedure was performed using an Excel Visual Basic for Applications macro program [27,28]. In situ Sr isotopic analyses were performed using an NWR 193 laser ablation system attached to a Thermo Fisher Scientific Neptune Plus multi-collector (MC) ICP-MS at the Beijing Createch Testing Technology Co., Ltd. A spot size of 38 μm was used, with a 6-8 Hz repetition rate and an energy density of 10 J/cm 2 , depending on the Sr concentration of the sample. Sr isotopic data were acquired using static multi-collection in low-resolution mode with nine Faraday collectors. Prior to laser analysis, the Neptune MC-ICP-MS was tuned using a standard to obtain the maximum sensitivity. A typical data acquisition cycle consisted of a 40 s measurement of the Ar gas blank with the laser switched off, followed by 60 s of measurement with laser ablation. The Durango1 apatite reference material was analyzed every 10 samples for external calibration.
Data reduction was conducted offline, and potential isobaric interferences were accounted for in the following order: Kr, Yb 2+ , Er 2+ , and Rb. Initially, the interferences of 84 Kr and 86 Kr on 84 Sr and 86 Sr, respectively, were removed using the 40 s Kr gas baseline measurement. The presence of 167 Er 2+ , 171 Yb 2+ , and 173 Yb 2+ at masses of 83.5, 85.5, and 86.5, respectively, were then monitored. Subsequently, the natural ratio of 85 Rb/ 87 Rb (2.5926) was used to correct for the isobaric interference of 87 Rb on 87 Sr using the exponential law, assuming that Rb has the same mass discrimination behavior as Sr. In addition, interferences from Ca argides or dimers and Ca-P-O were not considered for high Sr contents (>400 μg/g). The 87 Sr/ 86 Sr ratios were then calculated and normalized from the interference-cor- All of the in situ Nd isotopic analyses were performed using a Neptune Plus MC-ICP-MS (Thermo Scientific, Waltham, MA, USA), coupled with a RESOlution M-50 193 nm laser ablation system (Resonetics, Nashua, NH, USA), hosted at the State Key Laboratory of Isotope Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences. A detailed description of these instruments can be found in [29]. A small N 2 (2 mL L −1 ) flow and X skimmer cone at the interface were used to improve the instrumental sensitivity. All isotope signals were detected using Faraday cups under static mode. Laser parameters were set as follows: a beam diameter of 82-112 µm; a repetition rate of 6 Hz; and an energy density of~4 J cm −2 . Helium was chosen as the carrier gas (800 mL min −1 ). Each analysis consisted of 250 cycles, with an integration time of 0.262 s per cycle. The first 29 s were used to detect the gas blank with the laser beam off, followed by 30 s of laser ablation for sample signal collection with the laser beam. During the measurements, gas blanks of 143 Nd were less than 0.2 mv. The interferences of 144 Sm on 144 Nd were derived from the 147 Sm intensities, with a natural 143 Sm/ 147 Sm ratio of 0.20504 [30]. The mass bias factor of Sm was calculated from the measured isotopic ratio of 147 Sm/ 149 Sm and its accepted value (1.08507; [30]). The mass bias of 143 Nd/ 144 Nd was normalized to 146 Nd/ 144 Nd = 0.7129 using an exponential law. A detailed description of the data reduc-tion procedure can be found in [29]. In total, 40 analyses of McClure apatite and 25 analyses of Durango apatite yielded weighted means of 143 Nd/ 144 Nd = 0.512280 ± 0.000055 (2 SD) and 143 Nd/ 144 Nd = 0.512470 ± 0.000060 (2 SD), respectively, which are consistent (within error) with the values reported by [31].

REY Characteristics of Surface Sediments
The REY contents of the surface sediments from the Central and North Pacific are listed in Table 1. The data indicate that the average REY contents of the surface sediments at different sites in the Pigafetta Basin (northwest of the Marshall Islands) varied from 332 to 1221 µg/g (Figure 2). Heavy REEs (HREEs), including Y, varied from 106 to 572 µg/g, while light REEs (LREEs) varied from 226 to 649 µg/g.
The average REY contents of the surface sediments at different sites in the mid-Pacific basin (east of the Marshall Islands) varied from 418 to 1070 µg/g (Figure 2). HREE (including Y) contents varied from 143 to 514 µg/g, while LREE contents varied from 275 to 573 µg/g.

Vertical Distributions of REY
Vertical REY contents of sediments from the Central and Western Pacific varied from core to core. All of the analyzed piston cores contained multiple REY-rich sediment layers at different depths and with different thicknesses (Figure 4). For example, Core P7 contained three layers with REY contents of >1000 µg/g and two layers with REY contents ranging from 700 to 1000 µg/g. However, core P10 contained one layer with a REY content of >1000 µg/g, two layers with REY contents ranging from 700 to 1000 µg/g, and two layers with REY contents ranging from 400 to 700 µg/g.
In North American shale composite (NASC)-normalized REE patterns ( Figure 5), all of the sediments of P10 exhibited significant negative Ce anomalies and positive Y anomalies.

Fish Teeth Sr Isotopic Compositions
In situ fish teeth Sr values, at different depths in core P10, are listed in Table 2 and Figure 6. The initial 87 Sr/ 86 Sr ratios varied, ranging from 0.707960 to 0.709288, and multiple fish teeth from the same interval exhibited different isotopic characteristics.

Fish Teeth Sr Isotopic Compositions
In situ fish teeth Sr values, at different depths in core P10, are listed in Table 2 and Figure 6. The initial 87 Sr/ 86 Sr ratios varied, ranging from 0.707960 to 0.709288, and multiple fish teeth from the same interval exhibited different isotopic characteristics.  The initial 87 Sr/ 86 Sr ratios ranged from 0.708961 to 0.709288 in the upper 210 cm of the cores. Some fish teeth exhibited lower 87 Sr/ 86 Sr values (such as 0.708961) than the 87 Sr/ 86 Sr value of modern seawater (0.70924, [19]), indicating that they were old fish teeth that had been re-deposited [36]. The apatite 87 Sr/ 86 Sr values of the upper 210 cm in core P10 were generally similar to the modern seawater value, except for the unreasonable data described above.
Fish teeth 87 Sr/ 86 Sr ratios in the 210-600 cm layer and the lower 600-720 cm layer exhibited similar characteristics, ranging from 0.708005 to 0.708426 and from 0.707960 to 0.708481, respectively, indicating that these two layers were continuously deposited.
Theoretically, the fish teeth 87 Sr/ 86 Sr values should exhibit a decreasing trend in the vertical direction. However, the values fluctuated throughout the core length (Figure 6b). For example, some fish teeth 87 Sr/ 86 Sr values of lower layers (600-720 cm) were greater than those of the 210-600 cm layer (Table 2, Figure 6b); moreover, different fish teeth particles at the same interval showed different 87 Sr/ 86 Sr values. These fluctuating 87 Sr/ 86 Sr valuesmay have been raised by the clay materials adhering to the fish teeth [37] or reflect the overturning of the old and new sediments. Nevertheless, the fish teeth 87 Sr/ 86 Sr values in the upper sediment with lower REY contents, and in the middle layer sediment with higher REY contents (Figure 6a), showed significant difference, indicating that fish teeth 87 Sr/ 86 Sr values were of significance for the determination of sediment chronology. than those of the 210-600 cm layer (Table 2, Figure 6b); moreover, different fish teeth particles at the same interval showed different 87 Sr/ 86 Sr values. These fluctuating 87 Sr/ 86 Sr valuesmay have been raised by the clay materials adhering to the fish teeth [37] or reflect the overturning of the old and new sediments. Nevertheless, the fish teeth 87 Sr/ 86 Sr values in the upper sediment with lower REY contents, and in the middle layer sediment with higher REY contents (Figure 6a), showed significant difference, indicating that fish teeth 87 Sr/ 86 Sr values were of significance for the determination of sediment chronology.

Fish Teeth Nd Isotopic Compositions
In situ fish teeth Nd data from core P10 are listed in Table 3. The 143 Nd/ 144 Nd varied from 0.512110 to 0.512533, corresponding to an εNd range from −10.29 to −2.04 for the whole core. However, there were clear vertical variations in the Nd isotopes of apatite ( Figure   Figure 6. The REY content of sediment (a) and in situ fish teeth 87 Sr/ 86 Sr values (b) in core P10.

Fish Teeth Nd Isotopic Compositions
In situ fish teeth Nd data from core P10 are listed in Table 3. The 143 Nd/ 144 Nd varied from 0.512110 to 0.512533, corresponding to an ε Nd range from −10.29 to −2.04 for the whole core. However, there were clear vertical variations in the Nd isotopes of apatite ( Figure 7). Apatite ε Nd values varied from −6.42 to −2.04, concentrated between −5 and −3, with an average value of −4.48 in the upper 210 cm. In the 210-600 cm layer, apatite ε Nd values varied widely from −10.29 to −2.87 (i.e., significantly lower than in the upper layer). Apatite ε Nd values in the middle layer were divided into two ranges: from −10.29 to −7 and −6 to −2. At 600-705 cm, the values ranged from −7.69 to −4.27, with an average of −5.97. Apatite ε Nd is related to the REY content of the sediment. The REY contents of the 0-210 cm and 600-705 cm layers were lower than that of the 210-600 cm layer.

Distribution of REY-Rich Sediments in the Central and Western Pacific
As mentioned above, deep-sea sediments contain high concentrations of REY at numerous sites throughout the eastern South and central North Pacific [6]. However, the specific distributions of REY-rich sediments in the study area are still unknown.
In this study, REY-rich sediments in the surface layer (<2 m) of the Middle and Western Pacific Ocean exhibited a regular distribution (Figure 2). Sediments with high REY contents (>1000 µg/g) were mainly concentrated around seamounts (e.g., the Marshall Seamounts and Middle Pacific Seamounts). The REY contents of the surface sediments decreased with increasing distance from the seamounts. In the northwestern Pacific (northwestern Marshall Islands), the REY contents of the surface sediments gradually decreased from southeast to northwest. Surface sediments in the southeast had the highest REY contents (1000-1500 µg/g); whereas, those of the northwestern surface sediments were generally low (<400 µg/g). In the Middle Pacific, the REY contents of the surface sediments also exhibited regular patterns, decreasing gradually with increasing distance from seamounts.
The REY content in the deeper core sediments were showed in Figure 4. Similar to those of the surface sediments, core sediments with high REY contents (e.g., >700 µg/g) also usually occurred around the Marshall Islands within a~10 m depth range (i.e., cores P7, P8, and P10).

Dating REY-Rich Sediments
Core P10 was collected adjacent to Deep Sea Drilling Project (DSDP, the first of three international scientific ocean drilling programs that have operated over more than 40 years) site 170 in the Middle Pacific. The DSDP initial reports (Volume 17, Site 170) indicate that the uppermost 16 m of zeolitic brown clay were late Oligocene-Quaternary in age; whereas, the 16-36 m layer was mainly Eocene-Oligocene cherts. Therefore, the deep-sea sediment in this area is not older than the Oligocene.
In this study, for each 87 Sr/ 86 Sr value, an age was assigned. Therefore, we were able to produce an age-depth curve for the clay (Table 2, Figure 8). Fish teeth Sr isotopic analyses from the upper 210 cm of the P10 yielded 87 Sr/ 86 Sr ratios of 0.7109143-0.709288 (with outliers of 0.709005 and 0.708961), which were identical (within error) to that of modern seawater (0.7092, [19]), except for the unreasonable data (mentioned in Section 3.4). In addition, common radiolarian fossils from the Quaternary period (e.g., Druppatractus testudo, Euchitonia trianglulum, and Stylodictya validispina) were observed in the upper 30 cm (internal report, unpublished data). According to the biostratigraphic and fish teeth Sr isotopic analyses, the age of the upper 210 cm of the core is Quaternary. We calculated a linear sedimentation rate (LSR) of~1.0 mm/ky. Fish teeth 87 Sr/ 86 Sr values in the middle (225-600 cm) layer, which had high REY contents (>1000 µg/g), ranged from 0.708005 to 0.708426, indicating a clear hiatus in deposition between the upper (210 cm) and middle layers. The fish teeth 87 Sr/ 86 Sr values of the lower (600-720 cm) layer were continuous with those of the middle layer (0.707960-0.708481), indicating that these two layers were continuously deposited. Fish teeth Sr ages indicate that the REY-rich sediments were Oligocene-early Miocene (30.13-20.08 Ma) in age. Stichocorys delmontensis radiolarian, which occurred in the early to late Miocene (internal report, unpublished data), were observed at 600 cm depth. Therefore, the radiolarian biostratigraphy is in agreement with the ages obtained from the fish teeth. We calculated an LSR of~0.51 mm/ky, which was much slower than the upper layer.
From 84 Ma, the Drake Passage began to open [38]. Until the Oligocene and Early Miocene (~29 Ma) [39,40], the opening of the Drake Channel led to the formation of deep circumpolar currents [41]. The evolution of the Miocene paleo-ocean was characterized by the opening and closing of channels, changes in ocean circulation, and the development of glaciers. Affected by underflows, extensive sedimentary discontinuities exist in Miocene sediments throughout the ocean basins [36,42]. In core P10, the Miocene deep-sea hiatuses are coincident with the global hiatuses. Therefore, the ages of the fish teeth obtained using in situ Sr isotopes are reliable and reasonable.

Antarctic Bottom Water (AABW) and REY Enrichment
Clear vertical variations were observed in the Nd isotopes of fish teeth in core P10 (Figure 7), indicating that the sediments were affected by different water masses at different depths. The seawater εNd value in the North and Middle Pacific ranges from −5 to −3 at depths of 1000-5000 m [43]. In the upper 210 cm and lower 120 cm of core P10, the fish teeth εNd values were consistent with the Pacific deep water εNd, indicating that the REEs in these sediments were mainly derived from seawater.
However, the fish teeth in the middle layer (210-600 cm) were less radiogenic, exhibiting significantly reduced εNd values with amplitudes exceeding 2 units, which is much lower than that of the seawater εNd. This indicates that the REY in these sediments was affected by other factors. In general, εNd values are mainly affected by continental and mantle materials. Overall, volcanic material from the seabed is more radiogenic and characterized by positive values; whereas, terrestrial materials are less radiogenic and characterized by negative values. For example, mid-ocean ridge basalts in the Pacific have an average εNd value of +10; whereas, terrestrial silicates from the northern Central Pacific have an average εNd value of −10.2 [44]. Therefore, significant inputs of terrestrial material will lower the εNd value. However, Figure 9 shows that the Al2O3 and SiO2 contents of the sediments in the 210-600 cm layer were slightly lower than those in the upper-and lowermost layers, indicating that the low apatite εNd was not caused by an input of terrestrial materials. The changes in Nd isotopic compositions were therefore mainly caused by the mixing of ocean water masses with different Nd isotopic compositions under the condition that external inputs remained stable [45].
As aforementioned, during the early Oligocene, the opening of Drake Passage resulted in an established deep circumpolar current and an increase in AABW activity. The AABW forms two branches after inflowing into the Middle Pacific Basin through the Samoa channel, of which, one branch flows to the Northwestern Pacific Basin through the Marshall Islands (Figure 1, [26,46]). The sediments in the study area are located in the area where the AABW flows. The AABW is a low-temperature, high salinity, high density, and oxygen-rich water mass. It controls deep ocean circulation and provides a strong oxidizing environment for crust formation at seamounts [47]. The environmental redox indicators of the Ce anomaly have been verified [48,49]. The NASC-normalized REE pattern ( Figure 5) exhibited a significant negative Ce anomaly in the sediments of core P10, indicating that they were formed under strongly oxidizing conditions. In contrast, the fish teeth εNd values of the 210-600 cm layer of core P10 ranged from −10.29 to −2.87, with some values similar to the εNd value of the AABW (≈−9, [24]), implying that the REY-rich sediments were affected by the AABW. However, where the AABW is strong, sediment is

Antarctic Bottom Water (AABW) and REY Enrichment
Clear vertical variations were observed in the Nd isotopes of fish teeth in core P10 (Figure 7), indicating that the sediments were affected by different water masses at different depths. The seawater ε Nd value in the North and Middle Pacific ranges from −5 to −3 at depths of 1000-5000 m [43]. In the upper 210 cm and lower 120 cm of core P10, the fish teeth ε Nd values were consistent with the Pacific deep water ε Nd , indicating that the REEs in these sediments were mainly derived from seawater.
However, the fish teeth in the middle layer (210-600 cm) were less radiogenic, exhibiting significantly reduced ε Nd values with amplitudes exceeding 2 units, which is much lower than that of the seawater ε Nd . This indicates that the REY in these sediments was affected by other factors. In general, ε Nd values are mainly affected by continental and mantle materials. Overall, volcanic material from the seabed is more radiogenic and characterized by positive values; whereas, terrestrial materials are less radiogenic and characterized by negative values. For example, mid-ocean ridge basalts in the Pacific have an average ε Nd value of +10; whereas, terrestrial silicates from the northern Central Pacific have an average ε Nd value of −10.2 [44]. Therefore, significant inputs of terrestrial material will lower the ε Nd value. However, Figure 9 shows that the Al 2 O 3 and SiO 2 contents of the sediments in the 210-600 cm layer were slightly lower than those in the upperand lowermost layers, indicating that the low apatite ε Nd was not caused by an input of terrestrial materials. The changes in Nd isotopic compositions were therefore mainly caused by the mixing of ocean water masses with different Nd isotopic compositions under the condition that external inputs remained stable [45].
As aforementioned, during the early Oligocene, the opening of Drake Passage resulted in an established deep circumpolar current and an increase in AABW activity. The AABW forms two branches after inflowing into the Middle Pacific Basin through the Samoa channel, of which, one branch flows to the Northwestern Pacific Basin through the Marshall Islands ( Figure 1, [26,46]). The sediments in the study area are located in the area where the AABW flows. The AABW is a low-temperature, high salinity, high density, and oxygen-rich water mass. It controls deep ocean circulation and provides a strong oxidizing environment for crust formation at seamounts [47]. The environmental redox indicators of the Ce anomaly have been verified [48,49]. The NASC-normalized REE pattern ( Figure 5) exhibited a significant negative Ce anomaly in the sediments of core P10, indicating that they were formed under strongly oxidizing conditions. In contrast, the fish teeth ε Nd values of the 210-600 cm layer of core P10 ranged from −10.29 to −2.87, with some values similar to the εNd value of the AABW (≈−9, [24]), implying that the REY-rich sediments were affected by the AABW. However, where the AABW is strong, sediment is eroded or transported, which is not conducive for deposition. The fish teeth ε Nd values of REY-rich sediment had a relatively broad range, indicating that they were partly affected by the AABW. As the thermal gradient between the polar region and the equator increased, the activity of the AABW strengthened during the Miocene. Correspondingly, sediments were eroded and transported away by the strong AABW, explaining the depositional hiatus observed in core P10 during this period.
In the modern ocean, enhanced primary productivity around seamounts is a result of upwelling generated by seamount-current interactions [50,51]. The intensified northward AABW would have led to the upwelling of nutrients in regions with topographic barriers that were steep and large enough to allow upwelling [52,53]. This supply of nutrients in oligotrophic pelagic regions may have been sufficient to increase local fish proliferation [53]. The phenomenon, whereby REY-rich sediments with high contents of fish teeth were observed mainly around the seamounts in this study, also supports the influence of the AABW on the enrichment of REY in the sediment. eroded or transported, which is not conducive for deposition. The fish teeth εNd values of REY-rich sediment had a relatively broad range, indicating that they were partly affected by the AABW. As the thermal gradient between the polar region and the equator increased, the activity of the AABW strengthened during the Miocene. Correspondingly, sediments were eroded and transported away by the strong AABW, explaining the depositional hiatus observed in core P10 during this period.
In the modern ocean, enhanced primary productivity around seamounts is a result of upwelling generated by seamount-current interactions [50,51]. The intensified northward AABW would have led to the upwelling of nutrients in regions with topographic barriers that were steep and large enough to allow upwelling [52,53]. This supply of nutrients in oligotrophic pelagic regions may have been sufficient to increase local fish proliferation [53]. The phenomenon, whereby REY-rich sediments with high contents of fish teeth were observed mainly around the seamounts in this study, also supports the influence of the AABW on the enrichment of REY in the sediment.

Conclusions
This study indicates that the REY-rich sediments (>1000 μg/g) in the Central and Western Pacific have a clear spatial distribution and are mainly concentrated around seamounts. The chronological data from a typical piston core indicate that REY-rich sediments were mainly deposited during the early Oligocene to early Miocene. Therefore, since the opening of the Drake Passage in the early Oligocene, the AABW began to affect the Pacific Ocean. The upwelling of AABW induced fish proliferation around seamounts, simultaneously slowing sediment deposition and trying for enough time for REY entering and enriching in the apatite (fish teeth debris). During the early Oligocene to Miocene, when the AABW intensity was moderate, REY was enriched in the apatite (fish detritus) in the pelagic and zeolite clays. The specific spatial distribution of REY-rich sediments, described in this study, can provide guidance for future searches for REY-rich sediments in the Pacific, as well as in other oceans.

Conclusions
This study indicates that the REY-rich sediments (>1000 µg/g) in the Central and Western Pacific have a clear spatial distribution and are mainly concentrated around seamounts. The chronological data from a typical piston core indicate that REY-rich sediments were mainly deposited during the early Oligocene to early Miocene. Therefore, since the opening of the Drake Passage in the early Oligocene, the AABW began to affect the Pacific Ocean. The upwelling of AABW induced fish proliferation around seamounts, simultaneously slowing sediment deposition and trying for enough time for REY entering and enriching in the apatite (fish teeth debris). During the early Oligocene to Miocene, when the AABW intensity was moderate, REY was enriched in the apatite (fish detritus) in the pelagic and zeolite clays. The specific spatial distribution of REY-rich sediments, described in this study, can provide guidance for future searches for REY-rich sediments in the Pacific, as well as in other oceans.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.