High-Resolution Reconstruction of Oxidation–Reduction Conditions: Raman Spectroscopy and μ-XRF Analysis of Manganese Nodule and Crust on Tabletop of Western Pacific Magellan Seamounts

Abstract

Ferromanganese (Fe-Mn) deposits are widely used in paleoenvironmental reconstructions owing to their mineralogical and geochemical properties. We analyzed Fe-Mn deposits using micro-X-ray fluorescence and Raman spectroscopy to study the paleo-ocean environment. Samples were collected from the OSM-XX seamount in the western Pacific. The Fe-Mn crust was divided into three parts: phosphatized, massive non-phosphatized, and porous non-phosphatized. Vernadite was identified in all layers. Furthermore, in the nodule, high values of Mn, Ni, and Cu were observed near the nucleus, with vernadite and todorokite, and these values decreased outward. A high Mn/Fe ratio near the nucleus indicates early diagenetic processes. Formation of Fe-Mn nodules began around 19–16 Ma, and this period corresponded to a minor phosphatization event and persistent reducing conditions. From 11–10 Ma, the Mn/Fe and Co/Mn ratios decreased due to the formation of a western Pacific warm pool during this period. Subsequently, with the opening of the Indonesian seaway and global cooling, the Mn/Fe and Co/Mn ratios in the Fe-Mn deposits increased again. The comparative analysis of variations in Mn/Fe ratio and vernadite crystallinity in the Fe-Mn deposits confirmed that it is possible to reconstruct paleo-productivity and redox condition changes in the western Pacific Magellan Seamount.

1. Introduction

Ferromanganese (Fe-Mn) deposits are widely distributed in the deep sea. Fe-Mn crusts occur mainly on the flanks of seamounts, whereas Fe-Mn nodules mainly cover the deep-sea floor or tabletops of seamounts. As Fe-Mn deposit minerals are rich in iron and manganese, as well as nickel, copper, zinc, cobalt, and rare metals [1], they have been studied as marine mineral resources [2,3,4].

The mechanisms underlying the Fe-Mn crust formation are widely recognized as being hydrogenetic and hydrothermal [4,5,6,7]. Most Fe-Mn crusts of the western Pacific Magellan seamounts are formed by direct precipitation from seawater. Therefore, marine environmental factors such as the oxygen minimum zone (OMZ) influence the mineralogical and geochemical properties of Fe-Mn crusts [8]. Oxygen-rich deep-sea currents that rise along the flanks of seamounts oxidize dissolved manganese (Mn²⁺) and iron (Fe²⁺) in the OMZ, causing the precipitation of poorly crystalline vernadite (δ-MnO₂). The Fe-Mn crusts in the western Pacific can be broadly divided into four parts based on age and the presence of phosphatization [7,9].

The Fe-Mn nodules are categorized into two types: hydrogenetic and diagenetic. The type of formation depends on the environmental conditions, such as oxidation–reduction conditions and primary productivity [9,10,11,12]. Hydrogenetic formation occurs due to precipitation of Mn oxides and Fe-oxyhydroxides, similar to those of the Fe-Mn crusts, from seawater. Hydrogenetic nodules typically contain high amounts of Fe (with a Mn/Fe ratio of less than 2.5) and Co, but have low amounts of Mn, Ni, and Cu. The mineralogy of hydrogenetic nodules is predominantly Fe-vernadite, characterized by the intergrowth of feroxyhyte (δ-FeOOH) and monodispersed phyllomanganate layers [13,14,15,16,17,18]. In contrast, when the oxic-suboxic front (OSF) is located near surface sediment due to increased primary productivity, diagenetic formation becomes dominant. These nodules obtain elements from redox cycling linked to the early diagenetic reactions of organic matter in the sediment [10,19]. Diagenetic nodules have high amounts of Mn (with a Mn/Fe ratio greater than 2.5), Cu, and Ni [20,21,22,23]. In terms of mineralogical properties, diagenetic nodules are mainly composed of 10 Å manganates, such as buserite and todorokite [24].

Fe-Mn mineral deposits reflect variations in their surrounding environmental conditions through their mineralogical and geochemical properties. As Fe-Mn crusts form and cover a seamount at relatively shallow depths, variations in the OMZ are reflected in the properties of the Fe-Mn crust [25]. Similarly, the Fe-Mn nodules that formed on sediments record changes in the OSF [26]. Although Fe-Mn nodules have been studied widely for paleo-ocean reconstruction, research on nodules located on seamount tabletops remains relatively limited [27,28]. Moreover, a previous study confirmed that mixed-type Fe-Mn nodules from the tabletops of seamounts simultaneously reflect variations in OMZ and OSF [12]. Therefore, we hypothesized that a comparative analysis of the characteristics of these nodules and Fe-Mn crusts would facilitate an integrated reconstruction of paleo-productivity and paleo-redox conditions.

The Magellan Seamount Cluster in the western Pacific facilitates the formation of Fe-Mn deposits due to its low sedimentation rate (0.4–4 mm/ky) and the blockage of bottom currents by the seamounts (Figure 1). This cluster formed south of the equator during the Cretaceous Period, around 120 Ma, and has since shifted northwest due to tectonic activity [2]. Although older Fe-Mn deposits have been recognized, it is commonly noted that the formation of Fe-Mn deposits on Pacific seamounts commenced in the late Oligocene to early Miocene, coinciding with the activation of Antarctic Bottom Water (AABW) in the Pacific [6,29,30,31,32]. Fe-Mn crusts in the western Pacific can be classified by age [7,9], and the growth pattern of Fe-Mn nodules on western Pacific seamounts has been reported to be comparable with that of younger sections. To address this issue, Raman spectroscopy, which offers improved resolution, has recently been applied to elucidate the distribution of Fe-Mn oxide minerals. This technique is particularly advantageous for distinguishing between tunnel- and sheet-type manganese oxides, making it a valuable tool for studying Mn deposit minerals.

In this study, we aimed to reconstruct the paleo oxidation–reduction conditions of the Magellan seamounts in the western Pacific. To this end, quantitative analysis and 2D elemental distribution-mapping analysis were performed on a cross-section of Fe-Mn deposits using μ-XRF. Additionally, a mineralogical analysis of thin sections of Fe-Mn deposits was conducted using micro-laser Raman spectroscopy.

2. Materials and Methods

Samples

The Fe-Mn nodule and crust (OSMXX_EBS01_97, nodule 97; OSMXX_EBS04_20, crust 20) utilized in this study were collected using an epibenthic sledge from OSM-XX (151°51.12′7.2″ E, 16°8.16′9.6″ N; depth: 1557 mbsl) on the Magellan Seamount of the Western Pacific during the R/V Isabu Expedition Hl-21-06 (14 May–11 June 2021) conducted by the Korea Institute of Ocean Science and Technology (KIOST). One Fe-Mn nodule and one Fe-Mn crust were selected for geochemical and mineralogical analyses (Figure 2). The collected samples were halved and their cross-sections were polished using sandpaper for geochemical analysis. For mineralogical sample preparation, the samples were molded with epoxy. To prevent material loss during polishing attributable to differences in physical properties between the nuclei or basaltic substrate and the Fe-Mn deposit mineral layer, the cut surface was re-coated with epoxy before creating the polished samples. Additionally, dry polishing was performed to prevent mineral loss caused by water.

3. Data

3.1. Micro-Laser Raman Spectroscopy

To identify the mineral species comprising the Fe-Mn crust and nodule, micro-Raman spectroscopy was conducted using a micro-laser Raman spectrometer (RAMANtouch, Nano-photon) at Pusan National University, Busan, Republic of Korea. In this study, a 785 nm laser source was employed, and Raman spectra were obtained in the range of ~200–1000 cm⁻¹ with a grating condition of 600 gr/mm. The laser beam size was set to approximately 3 μm using a 50 × objective lens. To prevent damage to the samples, the spectra were acquired with a laser output of 0.3 mW. Each Raman spectrum was obtained as an average of 10 spectra recorded over 10 s, following the experimental conditions established in previous studies. The spectra were cross-referenced against spectral mineral patterns in the laboratory collection and databases available through CrystalSleuth provided by RRUFF for comparison. Before measuring the Raman spectra of the samples, the spectrometer was calibrated using the sharp first-order phonon peak of the silicon (Si) wafer at 520 cm⁻¹. The spectra acquisition and data analyses were accomplished using Fityk software (version 1.3.1) [33]. For the comparison analysis of intensity, the measured spectra were baseline-corrected and fitted with pseudo-Voigt functions [34].

3.2. Micro X-ray Fluorescence (μ-XRF)

The chemical compositions of Mn, Ni, Co, Cu, Fe, Zn, Ca, and P were assessed with an accuracy of ±5% using μ-XRF (Bruker M4 Tornado, Karlsruhe, Germany) at the Korea Institute of Industrial Technology, Wonju, Republic of Korea. The Rh anode tube was operated at 50 kV, with a collection time of 1 s. Geochemical analysis of the Fe-Mn deposits was performed at 0.100 mm intervals, employing a counting time of 1000 ms per point to generate two-dimensional element distribution maps. Quantitative analysis for each element was conducted using the built-in software (Bruker XRF, Karlsruhe, Germany). The quantitative analyses were performed at the same intervals (0.22 mm) and at 24 and 43 points for nodule 97 and crust 20, as shown by the red boxes in Figure 2.

3.3. Growth Rate and Age Dating of Fe-Mn Deposits Using a Co-Chronometer

A cobalt dating method was employed to measure the growth rate and apparent age of the Fe-Mn crust [34]. The amount of cobalt (Co) obtained via μ-XRF (wt.%) was substituted into the following equation to calculate the growth rate [35]:

$$Growthrate={\displaystyle \frac{1.28}{(Co-0.24)}}$$

(1)

To account for the influence of early diagenetic processes, an alternative Co-chronometer empirical equation [34] was utilized to estimate the growth rate of the Fe-Mn nodule. The growth rate was calculated using cobalt (Co), iron (Fe), and manganese (Mn) contents (wt.%) obtained by μ-XRF in the following equation [36]:

$$Growthrate={\displaystyle \frac{0.68}{\left({Co}_n^{1.67}\right)}},where{Co}_n=Co\times {\displaystyle \frac{50}{(Fe+Mn)}}$$

(2)

The Co-chronometer is based on the assumption that the incorporation of Co by the manganese oxides remains consistent over both time and space. This assumption is only applicable to hydrogenetic origin, which raises concerns regarding the reliability of the diagenetic components of Fe-Mn nodules that do not effectively scavenge dissolved cobalt.

4. Results

4.1. Morphology and Texture

4.1.1. Texture and Morphology of Fe-Mn Crust

The Fe-Mn crust used in this study, with a total height of approximately 9.59 cm, was broadly divided into basaltic and Fe-Mn oxide layers. The Fe-Mn oxide layer was further divided into sections in which the secondary phosphate minerals carbonate fluorapatite and CFA were present and absent (Figure 2), respectively, and the basaltic substrate was categorized into a lower reddish section and an upper brownish section. White phosphate deposits were observed in the Fe-Mn oxide layer, where phosphatization had occurred. In areas without phosphate minerals, the lower part exhibited a massive structure, whereas the upper part showed a porous structure with brown sediments.

4.1.2. Texture and Morphology of Fe-Mn Nodule

The Fe-Mn nodule had a spherical shape and botryoidal surface texture (Figure 2). The inner structure of the Fe-Mn nodule was determined using optical observation. An analysis of the cross-sections revealed that nodule 97 contained one large, ivory-colored nucleus. The thickness of the Fe-Mn oxide layer of nodule 97 varied from approximately 0.9 cm to 1.2 cm, and the layer was broadly divided into two smaller layers: a massive layer surrounding the nucleus and a porous layer outside of it.

4.2. Geochemistry

4.2.1. Geochemistry of Fe-Mn Crust

The two-dimensional elemental distribution was measured on the polished cross-section of the Fe-Mn crust (Figure 3). The distribution of Mn and Fe showed a negative correlation, whereas the distribution of Ni, Co, Cu, and Zn followed that of Mn (Figure 4). The positive correlation between Co and Mn reflected the characteristics of hydrogenetic Fe-Mn deposit minerals formed at relatively shallow depths (about 1500 mbsf) [8]. High signal intensities were observed for Ca and P in the phosphatized and basaltic substrates.

4.2.2. Geochemistry of Fe-Mn Nodule

The two-dimensional elemental distribution of nodule 97 was obtained using μ-XRF (Figure 5). This mapping indicated a significant negative correlation between the layers rich in Mn and those rich in Fe. Specifically, Fe was enriched in the outer layers of the nodules, whereas Mn was predominantly located in the inner layers. The color pattern observed in the mapping confirmed that the Mn-rich layers exhibited massive structures, whereas the Fe-rich layers displayed porous botryoidal forms, consistent with the optical observations. This alternation suggests that these nodules were a mixed type [26], similar to those found in the Magellan Seamount Cluster. Notably, the Fe-Mn nodule exhibited high-intensity areas of Ca and P in its nucleus, indicating that it was composed of biogenic compounds (Figure 5).

A quantitative analysis of nodule 97 was also performed using μ-XRF. Figure 6 presents vertical profiles of cross-sections of the Fe-Mn nodule. The results showed higher Mn concentrations near the nucleus than in the outer sections, which contained significant amounts of Fe. The vertical profiles of diagenetic elements, including Ni, Co, Cu, and Zn, revealed that their distribution closely mirrored that of Mn, with higher concentrations near the nucleus. This observation was consistent with the two-dimensional elemental distribution, which displayed a Mn-rich layer located near the nucleus of the Fe-Mn nodule.

Figure 7 illustrates vertical profiles of the Co/Mn and Mn/Fe ratios for the Fe-Mn nodule and crust. The nodule from OSM-XX exhibited Mn/Fe ratios in the Fe-Mn deposit layer ranging from 1.03 to 3.76, with an average value of 2.72. High Mn/Fe ratios were observed in the inner regions of nodule 97 (up to 3.76), with a notably sharp decrease toward the outer sections. After this sharp decline, the Mn/Fe ratios increased slightly as they approached the outer areas. The Co/Mn ratio was deemed reliable for only the hydrogenated portion (where the Mn/Fe ratio was <2.5). In the hydrogenetic regions of the Fe-Mn nodule, the Co/Mn ratio increased after a significant decrease in the Mn/Fe ratio.

4.3. Mineralogy

4.3.1. Mineralogy of Fe-Mn Crust

To confirm the mineralogy of the Fe-Mn crust, Raman spectroscopy was performed following the same analytical pathway as the μ-XRF quantitative analysis. The Raman spectroscopy pathway is shown in the sample image in Figure 8. For subsequent intensity comparisons of vernadite, the analysis was conducted on the upper part, which was not affected by phosphatization. In this study, Raman spectroscopy was conducted by moving the signal acquisition point approximately 2.2 mm from the outermost Fe-Mn oxide layer. The Raman signals are illustrated in Figure 8.

Observation of the Raman signals obtained from the Fe-Mn crust revealed differences in intensity, but consistent overall positions of the bands. The spectra indicated relatively broad bands at approximately ~500, ~590, and ~650 cm⁻¹, similar to the band positions reported for vernadite (δ-MnO₂) in previous studies [37].

4.3.2. Mineralogy of Fe-Mn Nodule

To confirm the mineral characteristics of the Fe-Mn nodule, Raman spectroscopy was performed on selected sections of the nodule, following the same analytical pathway as the μ-XRF quantitative analysis. The Raman spectroscopy pathway is illustrated in the sample diagram in Figure 9. In this study, Raman analysis was conducted by moving the signal acquisition point by approximately 2.5 mm from the starting point. The changes in the Raman signal are illustrated in Figure 9. The results of the horizontal Raman spectroscopy indicated that the distribution patterns of the minerals varied depending on their position within the nodule.

In nodule 97, relatively narrow Raman bands indicating todorokite were apparent at around ~240, ~508, and ~733 cm⁻¹ in the innermost part near the nucleus (Figure 2, red box number 6) [38,39]. Additionally, at the same location, vernadite was observed at ~500, ~590, and ~650 cm⁻¹, indicating the coexistence of both minerals within the laser analysis area (~3 μm).

Moving outward from the nodule, the signal for todorokite (~240, ~508, and ~733 cm⁻¹) disappeared, and only the vernadite signals (~500, ~590, and ~650 cm⁻¹) remained, displaying a pattern similar to that of the Fe-Mn crust. This suggests that the mixed-type Fe-Mn nodule underwent changes in its formation during its development, which aligns with the results of the geochemical analysis.

4.4. Growth Rate and Age Estimated by the Co-Chronometer

The results of the Co-chronometer dating for the Fe-Mn crust indicated that the growth rate ranged from 0.57 to 4.12 mm/My, with an average rate of 1.62 mm/My, resulting in a total age of 51.87 Ma (

Table 1. Estimated ages (Ma) and growth rates (mm/My) of hydrogenetic Fe-Mn crust and nodules from OSM-XX.

Samples	Quantitative Analysis Points	Mean Growth Rate (mm/My)	Age (Ma)	Reference
Nodule 97	1–5	1.06	13.94	[36]
Crust 20	1–30	1.62	51.87	[35]

, Figure 10). Additionally, the dating results for the outer part of the Fe-Mn crust that had not undergone phosphatization (quantitative analysis points 1–17) were confirmed to be 34.75 Ma. Furthermore, among the areas not affected by phosphatization, the boundary between the dense and porous structures (quantitative analysis point 13) was confirmed to have an age of approximately 22.56 Ma.

The growth rate and age of the hydrogenetic sections of the Fe-Mn nodule were calculated using the equations established by Manheim and Lane-Bostwick (1988) [36]. The growth rates of nodule 97 were between 0.85 and 1.47 mm/My (Figure 10). The age of the nodule was confirmed to be 13.94 Ma.

5. Discussion

5.1. Formation of Fe-Mn Deposit Minerals of OSM-XX

To understand the mineralogical and geochemical characteristics of the OSM-XX Fe-Mn crust and nodule, analyses were conducted using μ-XRF and Raman spectroscopy. Analysis of the Fe-Mn crust of OSM-XX confirmed that the crust had a hydrogenetic origin, with the Fe-Mn oxide layer divided into an inner part affected by phosphatization and an outer part not affected by phosphatization. Additionally, the non-phosphatized outer part was classified into dense and porous structures containing sediments. This is consistent with previous studies on the typical structure of Fe-Mn crusts from the western Pacific [28,40,41]. Given the characteristics of Fe-Mn crust formation, which occurs by sequential precipitation over a bedrock substrate, it can be inferred that the lower section formed first. Therefore, according to the Co-chronology data (Table 1), the deposition of the manganese crust began at 51.87 Ma, with evidence of phosphatization occurring at around 34.75 Ma. The dense inner part of the crust exhibited high concentrations of Mn, Ni, Co, Cu, and Zn, as well as a high Mn/Fe ratio. In contrast, the porous outer section displayed high Fe contents and contained significant amounts of detrital sediment in its pores. The boundary between these two sections was dated to approximately 22.56 Ma, indicating that environmental changes occurred around that time.

In the case of the Fe-Mn nodule, the mineralogical and geochemical results reflect transitions in the origin of the Fe-Mn nodule on the seamount tabletop. The tabletops of the seamounts are shallower than the seafloor and thus closer to the OMZ. Consequently, like the Fe-Mn crust, the amounts of Mn, Co, and Ni show a positive correlation [5,42,43]. The Fe-Mn nodule used in this study also exhibited similar correlations among elements with respect to variations in Mn, Ni, and Co (Figure 6). This characteristic is indicative of Fe-Mn oxide minerals formed near the OMZ, suggesting that they did not experience significant water depth changes during their formation [5,42,43]. The (Cu+Ni)–Fe-Mn ternary diagram of the quantitative analysis points (Figure 11) shows the division between the diagenetic origin points and hydrogenetic origin points. This figure shows that the Fe-Mn nodule formed on the OSM-XX summit changed during its formation process. The Co-chronology dating results for the hydrogenetic-origin portion constituting the outer part of the nodule indicate a formation age of around 14–13 Ma (Table 1), suggesting that the nodule formation has a diagenetic origin. This result suggests that Fe-Mn nodules on the tabletops of the Magellan Seamount began forming during a minor global phosphatization event, specifically in the mid-Miocene (19–16 Ma). This observation is consistent with previous findings indicating that the formation of Fe-Mn nodules in western Pacific seamounts also started during this minor phosphatization event [28,40,41].

5.2. Paleo-Ocean Environment Recorded in the Fe-Mn Deposit Minerals of OSM-XX

5.2.1. Period Before Mid-Miocene Minor Phosphatization Event

Generally, Fe-Mn nodules formed on deep-sea sediment provide insights into fluctuations in the oceanic surface environment [26]. The geochemical and mineralogical properties of Fe-Mn crusts formed on the slopes of seamounts preserve variations in the OMZ [25,41]. Particularly in mixed-type Fe-Mn nodules found at seamount summits, variations in the geochemical and mineralogical characteristics of Fe-Mn crusts and nodules enable the reconstruction of redox conditions and productivity [12].

As mentioned earlier, the formation of the Fe-Mn crust on the OSM-XX seamount began around 51.87 Ma, with evidence of phosphatization occurring around 34.78 Ma. The period from 36 Ma to 32 Ma (Eocene/Oligocene boundary) coincided with a global phosphatization event. During this period, global bottom current temperatures decreased by about 4 °C [44], while sea levels rose sharply [45]. Additionally, accelerated bottom current flows intensify upwelling induced by seamounts, leading to increased surface productivity. This increase in surface productivity is preserved in carbon isotope values, which rose by approximately 0.8‰ during this period [44]. Therefore, the rapid increase in sea level would have increased the supply of phosphates from land, while the increased surface productivity likely supplied large amounts of phosphate to the sea [1,5,6].

In the non-phosphatized part of the Fe-Mn crust, the relatively high concentrations of Mn, Ni, and Co in the massive inner part indicate formation at depths close to the OMZ [5,42,43]. However, recent studies have suggested that the depth changes experienced by seamounts over the past 30 Ma have been minimal [7], indicating that the OMZ strengthened at the time of formation of the dense structures. Subsequently, at approximately 22.56 Ma, a porous structure formed along with a decrease in the Mn, Ni, and Co concentrations and an increase in the Fe concentration, suggesting a reduction in the influence of the OMZ and indicating that the manganese nodules formed in an oxidizing environment.

Additionally, hydrogenetic Fe-Mn nodules and crusts exhibit smooth and botryoidal structures influenced by the grain size of nearby sediments [1]. When adjacent sediments are composed of fine particles, a dense structure forms; conversely, if they are formed near coarser sediments, botryoidal layers of iron-manganese oxides form. Therefore, approximately 22.56 Ma, changes occurred in the sedimentary environment near the OSM-XX summit. This period marked the onset of warming after the Mi-1 glacial event at the Oligocene–Miocene boundary, during which global productivity decreased and the OMZ weakened. Consequently, the Mn, Ni, and Co concentrations in the Fe-Mn crust decreased as the Fe concentration increased, resulting in the formation of a porous rather than a dense structure.

5.2.2. Period After Mid-Miocene Minor Phosphatization Event

Analysis of the Fe-Mn nodule indicated that the formation of the OSM-XX Fe-Mn nodules began in the mid-Miocene (19–16 Ma), which is consistent with previous studies on nearby seamounts [12,28,40]. This period coincided with a minor global phosphatization event; however, the Fe-Mn crust and nodule from the OSM-XX study did not contain the phosphate secondary mineral CFA.

The precipitation of CFA occurs because of an increase in phosphorus input. The input of phosphorus into the ocean during sea-level elevations occurs via upwelling and remineralization of inorganic phosphorus from primary productivity [43,44]. During the mid-Miocene (19–16 Ma), when a minor phosphatization event occurred, the seawater temperature gradient between latitudes increased, accelerating ocean current circulation. This intensified upwelling delivered various nutrients, such as phosphates, to the sea surface, resulting in a rapid boost in primary production [46,47].

The coexistence of the todorokite bands (~240, ~508, and ~733 cm⁻¹) and vernadite bands (~500, ~590, and ~650 cm⁻¹) in the innermost part of the nodule 97 (Figure 9), along with a high Mn/Fe ratio and a decrease in vernadite intensity (~590 cm⁻¹) in the Fe-Mn crust (Figure 12), indicate high productivity and redox conditions during that period. Todorokite is a tunnel-structured manganese oxide mineral formed under suboxic or anoxic conditions [48,49,50]. Notably, the coexistence of todorokite and vernadite suggests that the structure of δ-MnO₂ phyllomanganite may have collapsed to form todorokite due to strong reducing conditions during the minor phosphatization event. Additionally, in Fe-Mn crusts formed through hydrogenetic origin without the influence of diagenesis, low crystallinity δ-MnO₂ phyllomanganite was precipitated (Figure 12), further supporting the existence of strong reducing conditions at that time. Further studies using sediment core analysis are required to confirm the absence of CFA on the OSM-XX seamount. However, considering that the primary source of phosphorus is terrigenous, the geographic conditions located farthest from land in the Magellan Seamount Cluster may be a contributing factor.

Subsequently, in the Fe-Mn crust, the intensity of vernadite increased again (Figure 12), and the Mn/Fe ratio decreased (Figure 6). In the Fe-Mn nodule, the todorokite band disappeared (Figure 9), and the Mn/Fe ratio also decreased (Figure 6). This indicates that the environment near the OSM-XX seamount is oxidizing. Around this time, global cooling began (Figure 12), which typically promoted reducing conditions. However, our results indicated oxidizing conditions. Approximately 10 Ma, global sea levels declined, leading to the closure of the Indonesian seaways. This disruption of the Indonesian throughflow resulted in the isolation of warm seawater in the western Pacific, similar to the La Niña effect, and facilitated the formation of a western Pacific warm pool [51]. This situation has led to the strengthening of the thermocline and a decline in productivity, weakening the OMZ, and allowing the oxygen saturation front to create oxidizing conditions. These environmental conditions were recorded in both the Fe-Mn nodules and the crust from OSM-XX.

Subsequently, the intensity of vernadite in the Fe-Mn crust decreased again, the Mn/Fe ratio increased slightly, and the Co/Mn ratio increased in both the Fe-Mn crust and nodule. With respect to the Co/Mn ratio, the cobalt content in hydrogenetic Mn-Fe deposit minerals is elevated within the OMZ [52]. As a result, a high Co/Mn ratio within the extended OMZ suggests that the OMZ plays a crucial role in the formation of Fe-Mn deposit minerals. The gradual rise in the Co/Mn ratio suggests that the influence of the OMZ intensified progressively after approximately 9–8 Ma. This section dates to approximately 9.66 Ma, at which time global sea level rise opened the Indonesian seaway, lowering the temperature of the warm pool [51], all while global cooling continued [53].

Consequently, ocean currents accelerated, productivity increased, and the OMZ was slightly strengthened. In fact, from around 9 Ma to the present, carbon isotope values have risen by 0.3‰ [46]. These environmental changes may be reflected in the chemical properties of the Fe-Mn crust and nodule. However, the fact that the Fe-Mn nodule was formed through hydrogenetic processes suggests that reducing conditions were not sufficient for the nodule to have had a diagenetic origin.

6. Conclusions

We performed geochemical and mineralogical analyses of an Fe-Mn crust and nodule collected from OSM-XX in the western Pacific Magellan Seamount Cluster to reconstruct the paleoceanographic conditions (Figure 13). Based on the findings of the present study and previous studies, the formation of the OSM-XX crust began around 51.87 Ma and underwent phosphatization at around 34.75 Ma, indicating a heightened sea level and strong reducing environment at that time. Subsequently, until approximately 22.56 Ma, there were high levels of Mn, Co, and Ni because of the strong influence of the OMZ. After the end of the Mi-1 glaciation, the values of Mn, Co, and Ni decreased, whereas those of Fe increased.

The Fe-Mn nodule began to form with a diagenetic origin during the mid-Miocene (19–16 Ma), a period marked by a minor phosphatization event. Although CFA was not present in the OSM-XX Fe-Mn crust and nodule, the high productivity and reducing environment at the time were reflected in their geochemical and mineralogical properties. Despite the onset of global cooling, as the western Pacific warm pool developed and primary surface productivity declined, the reducing conditions shifted to oxidizing conditions. This transition facilitated the dominance of hydrogenetic Fe-Mn nodule formation. Finally, with the opening of the Indonesian seaway and global cooling, the bottom currents and the OMZ strengthened; however, the Fe-Mn nodules still exhibited hydrogenetic origins.

These results further confirm that the mixed-type nodules at the tabletop of seamounts are useful tools for reconstructing fluctuations in redox conditions and productivity. A more detailed reconstruction of paleo-redox conditions and paleo-productivity could be achieved by also comparing the characteristics of Fe-Mn crusts that are unaffected by early diagenesis. Further research using sediment cores is needed to investigate the early diagenesis processes that may have affected the Fe-Mn nodule and the changes in sediment grain size related to texture variations in the Fe-Mn crust.